causal study hypothesis

Root out friction in every digital experience, super-charge conversion rates, and optimize digital self-service

Uncover insights from any interaction, deliver AI-powered agent coaching, and reduce cost to serve

Increase revenue and loyalty with real-time insights and recommendations delivered to teams on the ground

Know how your people feel and empower managers to improve employee engagement, productivity, and retention

Take action in the moments that matter most along the employee journey and drive bottom line growth

Whatever they’re are saying, wherever they’re saying it, know exactly what’s going on with your people

Get faster, richer insights with qual and quant tools that make powerful market research available to everyone

Run concept tests, pricing studies, prototyping + more with fast, powerful studies designed by UX research experts

Track your brand performance 24/7 and act quickly to respond to opportunities and challenges in your market

Explore the platform powering Experience Management

• Free Account
• For Digital
• For Customer Care
• For Human Resources
• For Researchers
• Financial Services
• All Industries

Popular Use Cases

• Customer Experience
• Employee Experience
• Employee Exit Interviews
• Net Promoter Score
• Voice of Customer
• Customer Success Hub
• Product Documentation
• Training & Certification
• XM Institute
• Popular Resources
• Customer Stories

Market Research

• Artificial Intelligence
• Partnerships
• Marketplace

The annual gathering of the experience leaders at the world’s iconic brands building breakthrough business results, live in Salt Lake City.

• English/AU & NZ
• Español/Europa
• Español/América Latina
• Português Brasileiro
• REQUEST DEMO
• Experience Management
• Causal Research

Try Qualtrics for free

Causal research: definition, examples and how to use it.

16 min read Causal research enables market researchers to predict hypothetical occurrences & outcomes while improving existing strategies. Discover how this research can decrease employee retention & increase customer success for your business.

What is causal research?

Causal research, also known as explanatory research or causal-comparative research, identifies the extent and nature of cause-and-effect relationships between two or more variables.

It’s often used by companies to determine the impact of changes in products, features, or services process on critical company metrics. Some examples:

• How does rebranding of a product influence intent to purchase?
• How would expansion to a new market segment affect projected sales?
• What would be the impact of a price increase or decrease on customer loyalty?

To maintain the accuracy of causal research, ‘confounding variables’ or influences — e.g. those that could distort the results — are controlled. This is done either by keeping them constant in the creation of data, or by using statistical methods. These variables are identified before the start of the research experiment.

As well as the above, research teams will outline several other variables and principles in causal research:

• Independent variables

The variables that may cause direct changes in another variable. For example, the effect of truancy on a student’s grade point average. The independent variable is therefore class attendance.

• Control variables

These are the components that remain unchanged during the experiment so researchers can better understand what conditions create a cause-and-effect relationship.  

This describes the cause-and-effect relationship. When researchers find causation (or the cause), they’ve conducted all the processes necessary to prove it exists.

• Correlation

Any relationship between two variables in the experiment. It’s important to note that correlation doesn’t automatically mean causation. Researchers will typically establish correlation before proving cause-and-effect.

• Experimental design

Researchers use experimental design to define the parameters of the experiment — e.g. categorizing participants into different groups.

• Dependent variables

These are measurable variables that may change or are influenced by the independent variable. For example, in an experiment about whether or not terrain influences running speed, your dependent variable is the terrain.  

Why is causal research useful?

It’s useful because it enables market researchers to predict hypothetical occurrences and outcomes while improving existing strategies. This allows businesses to create plans that benefit the company. It’s also a great research method because researchers can immediately see how variables affect each other and under what circumstances.

Also, once the first experiment has been completed, researchers can use the learnings from the analysis to repeat the experiment or apply the findings to other scenarios. Because of this, it’s widely used to help understand the impact of changes in internal or commercial strategy to the business bottom line.

Some examples include:

• Understanding how overall training levels are improved by introducing new courses
• Examining which variations in wording make potential customers more interested in buying a product
• Testing a market’s response to a brand-new line of products and/or services

So, how does causal research compare and differ from other research types?

Well, there are a few research types that are used to find answers to some of the examples above:

1. Exploratory research

As its name suggests, exploratory research involves assessing a situation (or situations) where the problem isn’t clear. Through this approach, researchers can test different avenues and ideas to establish facts and gain a better understanding.

Researchers can also use it to first navigate a topic and identify which variables are important. Because no area is off-limits, the research is flexible and adapts to the investigations as it progresses.

Finally, this approach is unstructured and often involves gathering qualitative data, giving the researcher freedom to progress the research according to their thoughts and assessment. However, this may make results susceptible to researcher bias and may limit the extent to which a topic is explored.

2. Descriptive research

Descriptive research is all about describing the characteristics of the population, phenomenon or scenario studied. It focuses more on the “what” of the research subject than the “why”.

For example, a clothing brand wants to understand the fashion purchasing trends amongst buyers in California — so they conduct a demographic survey of the region, gather population data and then run descriptive research. The study will help them to uncover purchasing patterns amongst fashion buyers in California, but not necessarily why those patterns exist.

As the research happens in a natural setting, variables can cross-contaminate other variables, making it harder to isolate cause and effect relationships. Therefore, further research will be required if more causal information is needed.

Get started on your market research journey with CoreXM

How is causal research different from the other two methods above?

Well, causal research looks at what variables are involved in a problem and ‘why’ they act a certain way. As the experiment takes place in a controlled setting (thanks to controlled variables) it’s easier to identify cause-and-effect amongst variables.

Furthermore, researchers can carry out causal research at any stage in the process, though it’s usually carried out in the later stages once more is known about a particular topic or situation.

Finally, compared to the other two methods, causal research is more structured, and researchers can combine it with exploratory and descriptive research to assist with research goals.

Summary of three research types

What are the advantages of causal research?

• Improve experiences

By understanding which variables have positive impacts on target variables (like sales revenue or customer loyalty), businesses can improve their processes, return on investment, and the experiences they offer customers and employees.

• Help companies improve internally

By conducting causal research, management can make informed decisions about improving their employee experience and internal operations. For example, understanding which variables led to an increase in staff turnover.

• Repeat experiments to enhance reliability and accuracy of results

When variables are identified, researchers can replicate cause-and-effect with ease, providing them with reliable data and results to draw insights from.

• Test out new theories or ideas

If causal research is able to pinpoint the exact outcome of mixing together different variables, research teams have the ability to test out ideas in the same way to create viable proof of concepts.

• Fix issues quickly

Once an undesirable effect’s cause is identified, researchers and management can take action to reduce the impact of it or remove it entirely, resulting in better outcomes.

What are the disadvantages of causal research?

• Provides information to competitors

If you plan to publish your research, it provides information about your plans to your competitors. For example, they might use your research outcomes to identify what you are up to and enter the market before you.

• Difficult to administer

Causal research is often difficult to administer because it’s not possible to control the effects of extraneous variables.

• Time and money constraints

Budgetary and time constraints can make this type of research expensive to conduct and repeat. Also, if an initial attempt doesn’t provide a cause and effect relationship, the ROI is wasted and could impact the appetite for future repeat experiments.

• Requires additional research to ensure validity

You can’t rely on just the outcomes of causal research as it’s inaccurate. It’s best to conduct other types of research alongside it to confirm its output.

• Trouble establishing cause and effect

Researchers might identify that two variables are connected, but struggle to determine which is the cause and which variable is the effect.

• Risk of contamination

There’s always the risk that people outside your market or area of study could affect the results of your research. For example, if you’re conducting a retail store study, shoppers outside your ‘test parameters’ shop at your store and skew the results.

How can you use causal research effectively?

To better highlight how you can use causal research across functions or markets, here are a few examples:

Market and advertising research

A company might want to know if their new advertising campaign or marketing campaign is having a positive impact. So, their research team can carry out a causal research project to see which variables cause a positive or negative effect on the campaign.

For example, a cold-weather apparel company in a winter ski-resort town may see an increase in sales generated after a targeted campaign to skiers. To see if one caused the other, the research team could set up a duplicate experiment to see if the same campaign would generate sales from non-skiers. If the results reduce or change, then it’s likely that the campaign had a direct effect on skiers to encourage them to purchase products.

Improving customer experiences and loyalty levels

Customers enjoy shopping with brands that align with their own values, and they’re more likely to buy and present the brand positively to other potential shoppers as a result. So, it’s in your best interest to deliver great experiences and retain your customers.

For example, the Harvard Business Review found that an increase in customer retention rates by 5% increased profits by 25% to 95%. But let’s say you want to increase your own, how can you identify which variables contribute to it?Using causal research, you can test hypotheses about which processes, strategies or changes influence customer retention. For example, is it the streamlined checkout? What about the personalized product suggestions? Or maybe it was a new solution that solved their problem? Causal research will help you find out.

Discover how to use analytics to improve customer retention.

Improving problematic employee turnover rates

If your company has a high attrition rate, causal research can help you narrow down the variables or reasons which have the greatest impact on people leaving. This allows you to prioritize your efforts on tackling the issues in the right order, for the best positive outcomes.

For example, through causal research, you might find that employee dissatisfaction due to a lack of communication and transparency from upper management leads to poor morale, which in turn influences employee retention.

To rectify the problem, you could implement a routine feedback loop or session that enables your people to talk to your company’s C-level executives so that they feel heard and understood.

How to conduct causal research first steps to getting started are:

1. Define the purpose of your research

What questions do you have? What do you expect to come out of your research? Think about which variables you need to test out the theory.

2. Pick a random sampling if participants are needed

Using a technology solution to support your sampling, like a database, can help you define who you want your target audience to be, and how random or representative they should be.

3. Set up the controlled experiment

Once you’ve defined which variables you’d like to measure to see if they interact, think about how best to set up the experiment. This could be in-person or in-house via interviews, or it could be done remotely using online surveys.

4. Carry out the experiment

Make sure to keep all irrelevant variables the same, and only change the causal variable (the one that causes the effect) to gather the correct data. Depending on your method, you could be collecting qualitative or quantitative data, so make sure you note your findings across each regularly.

5. Analyze your findings

Either manually or using technology, analyze your data to see if any trends, patterns or correlations emerge. By looking at the data, you’ll be able to see what changes you might need to do next time, or if there are questions that require further research.

6. Verify your findings

Your first attempt gives you the baseline figures to compare the new results to. You can then run another experiment to verify your findings.

7. Do follow-up or supplemental research

You can supplement your original findings by carrying out research that goes deeper into causes or explores the topic in more detail. One of the best ways to do this is to use a survey. See ‘Use surveys to help your experiment’.

Identifying causal relationships between variables

To verify if a causal relationship exists, you have to satisfy the following criteria:

• Nonspurious association

A clear correlation exists between one cause and the effect. In other words, no ‘third’ that relates to both (cause and effect) should exist.

• Temporal sequence

The cause occurs before the effect. For example, increased ad spend on product marketing would contribute to higher product sales.

• Concomitant variation

The variation between the two variables is systematic. For example, if a company doesn’t change its IT policies and technology stack, then changes in employee productivity were not caused by IT policies or technology.

How surveys help your causal research experiments?

There are some surveys that are perfect for assisting researchers with understanding cause and effect. These include:

• Employee Satisfaction Survey – An introductory employee satisfaction survey that provides you with an overview of your current employee experience.
• Manager Feedback Survey – An introductory manager feedback survey geared toward improving your skills as a leader with valuable feedback from your team.
• Net Promoter Score (NPS) Survey – Measure customer loyalty and understand how your customers feel about your product or service using one of the world’s best-recognized metrics.
• Employee Engagement Survey – An entry-level employee engagement survey that provides you with an overview of your current employee experience.
• Customer Satisfaction Survey – Evaluate how satisfied your customers are with your company, including the products and services you provide and how they are treated when they buy from you.
• Employee Exit Interview Survey – Understand why your employees are leaving and how they’ll speak about your company once they’re gone.
• Product Research Survey – Evaluate your consumers’ reaction to a new product or product feature across every stage of the product development journey.
• Brand Awareness Survey – Track the level of brand awareness in your target market, including current and potential future customers.
• Online Purchase Feedback Survey – Find out how well your online shopping experience performs against customer needs and expectations.

That covers the fundamentals of causal research and should give you a foundation for ongoing studies to assess opportunities, problems, and risks across your market, product, customer, and employee segments.

If you want to transform your research, empower your teams and get insights on tap to get ahead of the competition, maybe it’s time to leverage Qualtrics CoreXM.

Qualtrics CoreXM provides a single platform for data collection and analysis across every part of your business — from customer feedback to product concept testing. What’s more, you can integrate it with your existing tools and services thanks to a flexible API.

Qualtrics CoreXM offers you as much or as little power and complexity as you need, so whether you’re running simple surveys or more advanced forms of research, it can deliver every time.

Related resources

Market intelligence 10 min read, marketing insights 11 min read, ethnographic research 11 min read, qualitative vs quantitative research 13 min read, qualitative research questions 11 min read, qualitative research design 12 min read, primary vs secondary research 14 min read, request demo.

Ready to learn more about Qualtrics?

What is causal research design?

Last updated

14 May 2023

Reviewed by

Examining these relationships gives researchers valuable insights into the mechanisms that drive the phenomena they are investigating.

Organizations primarily use causal research design to identify, determine, and explore the impact of changes within an organization and the market. You can use a causal research design to evaluate the effects of certain changes on existing procedures, norms, and more.

This article explores causal research design, including its elements, advantages, and disadvantages.

Analyze your causal research

Dovetail streamlines causal research analysis to help you uncover and share actionable insights

• Components of causal research

You can demonstrate the existence of cause-and-effect relationships between two factors or variables using specific causal information, allowing you to produce more meaningful results and research implications.

These are the key inputs for causal research:

The timeline of events

Ideally, the cause must occur before the effect. You should review the timeline of two or more separate events to determine the independent variables (cause) from the dependent variables (effect) before developing a hypothesis. 

If the cause occurs before the effect, you can link cause and effect and develop a hypothesis .

For instance, an organization may notice a sales increase. Determining the cause would help them reproduce these results. 

Upon review, the business realizes that the sales boost occurred right after an advertising campaign. The business can leverage this time-based data to determine whether the advertising campaign is the independent variable that caused a change in sales. 

Evaluation of confounding variables

In most cases, you need to pinpoint the variables that comprise a cause-and-effect relationship when using a causal research design. This uncovers a more accurate conclusion. 

Co-variations between a cause and effect must be accurate, and a third factor shouldn’t relate to cause and effect. 

Observing changes

Variation links between two variables must be clear. A quantitative change in effect must happen solely due to a quantitative change in the cause. 

You can test whether the independent variable changes the dependent variable to evaluate the validity of a cause-and-effect relationship. A steady change between the two variables must occur to back up your hypothesis of a genuine causal effect. 

• Why is causal research useful?

Causal research allows market researchers to predict hypothetical occurrences and outcomes while enhancing existing strategies. Organizations can use this concept to develop beneficial plans. 

Causal research is also useful as market researchers can immediately deduce the effect of the variables on each other under real-world conditions. 

Once researchers complete their first experiment, they can use their findings. Applying them to alternative scenarios or repeating the experiment to confirm its validity can produce further insights. 

Businesses widely use causal research to identify and comprehend the effect of strategic changes on their profits. 

• How does causal research compare and differ from other research types?

Other research types that identify relationships between variables include exploratory and descriptive research . 

Here’s how they compare and differ from causal research designs:

Exploratory research

An exploratory research design evaluates situations where a problem or opportunity's boundaries are unclear. You can use this research type to test various hypotheses and assumptions to establish facts and understand a situation more clearly.

You can also use exploratory research design to navigate a topic and discover the relevant variables. This research type allows flexibility and adaptability as the experiment progresses, particularly since no area is off-limits.

It’s worth noting that exploratory research is unstructured and typically involves collecting qualitative data . This provides the freedom to tweak and amend the research approach according to your ongoing thoughts and assessments. 

Unfortunately, this exposes the findings to the risk of bias and may limit the extent to which a researcher can explore a topic. 

This table compares the key characteristics of causal and exploratory research:

Descriptive research

This research design involves capturing and describing the traits of a population, situation, or phenomenon. Descriptive research focuses more on the " what " of the research subject and less on the " why ."

Since descriptive research typically happens in a real-world setting, variables can cross-contaminate others. This increases the challenge of isolating cause-and-effect relationships. 

You may require further research if you need more causal links. 

This table compares the key characteristics of causal and descriptive research.  

Causal research examines a research question’s variables and how they interact. It’s easier to pinpoint cause and effect since the experiment often happens in a controlled setting. 

Researchers can conduct causal research at any stage, but they typically use it once they know more about the topic.

In contrast, causal research tends to be more structured and can be combined with exploratory and descriptive research to help you attain your research goals. 

• How can you use causal research effectively?

Here are common ways that market researchers leverage causal research effectively:

Market and advertising research

Do you want to know if your new marketing campaign is affecting your organization positively? You can use causal research to determine the variables causing negative or positive impacts on your campaign. 

Improving customer experiences and loyalty levels

Consumers generally enjoy purchasing from brands aligned with their values. They’re more likely to purchase from such brands and positively represent them to others. 

You can use causal research to identify the variables contributing to increased or reduced customer acquisition and retention rates. 

Could the cause of increased customer retention rates be streamlined checkout? 

Perhaps you introduced a new solution geared towards directly solving their immediate problem. 

Whatever the reason, causal research can help you identify the cause-and-effect relationship. You can use this to enhance your customer experiences and loyalty levels.

Improving problematic employee turnover rates

Is your organization experiencing skyrocketing attrition rates? 

You can leverage the features and benefits of causal research to narrow down the possible explanations or variables with significant effects on employees quitting. 

This way, you can prioritize interventions, focusing on the highest priority causal influences, and begin to tackle high employee turnover rates. 

• Advantages of causal research

The main benefits of causal research include the following:

Effectively test new ideas

If causal research can pinpoint the precise outcome through combinations of different variables, researchers can test ideas in the same manner to form viable proof of concepts.

Achieve more objective results

Market researchers typically use random sampling techniques to choose experiment participants or subjects in causal research. This reduces the possibility of exterior, sample, or demography-based influences, generating more objective results. 

Improved business processes

Causal research helps businesses understand which variables positively impact target variables, such as customer loyalty or sales revenues. This helps them improve their processes, ROI, and customer and employee experiences.

Guarantee reliable and accurate results

Upon identifying the correct variables, researchers can replicate cause and effect effortlessly. This creates reliable data and results to draw insights from. 

Internal organization improvements

Businesses that conduct causal research can make informed decisions about improving their internal operations and enhancing employee experiences. 

• Disadvantages of causal research

Like any other research method, casual research has its set of drawbacks that include:

Extra research to ensure validity

Researchers can't simply rely on the outcomes of causal research since it isn't always accurate. There may be a need to conduct other research types alongside it to ensure accurate output.

Coincidence

Coincidence tends to be the most significant error in causal research. Researchers often misinterpret a coincidental link between a cause and effect as a direct causal link. 

Administration challenges

Causal research can be challenging to administer since it's impossible to control the impact of extraneous variables . 

Giving away your competitive advantage

If you intend to publish your research, it exposes your information to the competition. 

Competitors may use your research outcomes to identify your plans and strategies to enter the market before you. 

• Causal research examples

Multiple fields can use causal research, so it serves different purposes, such as. 

Customer loyalty research

Organizations and employees can use causal research to determine the best customer attraction and retention approaches. 

They monitor interactions between customers and employees to identify cause-and-effect patterns. That could be a product demonstration technique resulting in higher or lower sales from the same customers. 

Example: Business X introduces a new individual marketing strategy for a small customer group and notices a measurable increase in monthly subscriptions. 

Upon getting identical results from different groups, the business concludes that the individual marketing strategy resulted in the intended causal relationship.

Advertising research

Businesses can also use causal research to implement and assess advertising campaigns. 

Example: Business X notices a 7% increase in sales revenue a few months after a business introduces a new advertisement in a certain region. The business can run the same ad in random regions to compare sales data over the same period. 

This will help the company determine whether the ad caused the sales increase. If sales increase in these randomly selected regions, the business could conclude that advertising campaigns and sales share a cause-and-effect relationship. 

Educational research

Academics, teachers, and learners can use causal research to explore the impact of politics on learners and pinpoint learner behavior trends. 

Example: College X notices that more IT students drop out of their program in their second year, which is 8% higher than any other year. 

The college administration can interview a random group of IT students to identify factors leading to this situation, including personal factors and influences. 

With the help of in-depth statistical analysis, the institution's researchers can uncover the main factors causing dropout. They can create immediate solutions to address the problem.

Is a causal variable dependent or independent?

When two variables have a cause-and-effect relationship, the cause is often called the independent variable. As such, the effect variable is dependent, i.e., it depends on the independent causal variable. An independent variable is only causal under experimental conditions. 

What are the three criteria for causality?

The three conditions for causality are:

Temporality/temporal precedence: The cause must precede the effect.

Rationality: One event predicts the other with an explanation, and the effect must vary in proportion to changes in the cause.

Control for extraneous variables: The covariables must not result from other variables.  

Is causal research experimental?

Causal research is mostly explanatory. Causal studies focus on analyzing a situation to explore and explain the patterns of relationships between variables. 

Further, experiments are the primary data collection methods in studies with causal research design. However, as a research design, causal research isn't entirely experimental.

What is the difference between experimental and causal research design?

One of the main differences between causal and experimental research is that in causal research, the research subjects are already in groups since the event has already happened. 

On the other hand, researchers randomly choose subjects in experimental research before manipulating the variables.

Get started today

Go from raw data to valuable insights with a flexible research platform

Editor’s picks

Last updated: 21 December 2023

Last updated: 16 December 2023

Last updated: 6 October 2023

Last updated: 5 March 2024

Last updated: 25 November 2023

Last updated: 15 February 2024

Last updated: 11 March 2024

Last updated: 12 December 2023

Last updated: 6 March 2024

Last updated: 10 April 2023

Last updated: 20 December 2023

Latest articles

Related topics, log in or sign up.

Get started for free

Causal Research: Definition, Design, Tips, Examples

Appinio Research · 21.02.2024 · 33min read

Ever wondered why certain events lead to specific outcomes? Understanding causality—the relationship between cause and effect—is crucial for unraveling the mysteries of the world around us. In this guide on causal research, we delve into the methods, techniques, and principles behind identifying and establishing cause-and-effect relationships between variables. Whether you're a seasoned researcher or new to the field, this guide will equip you with the knowledge and tools to conduct rigorous causal research and draw meaningful conclusions that can inform decision-making and drive positive change.

What is Causal Research?

Causal research is a methodological approach used in scientific inquiry to investigate cause-and-effect relationships between variables. Unlike correlational or descriptive research, which merely examine associations or describe phenomena, causal research aims to determine whether changes in one variable cause changes in another variable.

Importance of Causal Research

Understanding the importance of causal research is crucial for appreciating its role in advancing knowledge and informing decision-making across various fields. Here are key reasons why causal research is significant:

• Establishing Causality:  Causal research enables researchers to determine whether changes in one variable directly cause changes in another variable. This helps identify effective interventions, predict outcomes, and inform evidence-based practices.
• Guiding Policy and Practice:  By identifying causal relationships, causal research provides empirical evidence to support policy decisions, program interventions, and business strategies. Decision-makers can use causal findings to allocate resources effectively and address societal challenges.
• Informing Predictive Modeling:  Causal research contributes to the development of predictive models by elucidating causal mechanisms underlying observed phenomena. Predictive models based on causal relationships can accurately forecast future outcomes and trends.
• Advancing Scientific Knowledge:  Causal research contributes to the cumulative body of scientific knowledge by testing hypotheses, refining theories, and uncovering underlying mechanisms of phenomena. It fosters a deeper understanding of complex systems and phenomena.
• Mitigating Confounding Factors:  Understanding causal relationships allows researchers to control for confounding variables and reduce bias in their studies. By isolating the effects of specific variables, researchers can draw more valid and reliable conclusions.

Causal Research Distinction from Other Research

Understanding the distinctions between causal research and other types of research methodologies is essential for researchers to choose the most appropriate approach for their study objectives. Let's explore the differences and similarities between causal research and descriptive, exploratory, and correlational research methodologies .

Descriptive vs. Causal Research

Descriptive research  focuses on describing characteristics, behaviors, or phenomena without manipulating variables or establishing causal relationships. It provides a snapshot of the current state of affairs but does not attempt to explain why certain phenomena occur.

Causal research , on the other hand, seeks to identify cause-and-effect relationships between variables by systematically manipulating independent variables and observing their effects on dependent variables. Unlike descriptive research, causal research aims to determine whether changes in one variable directly cause changes in another variable.

Similarities:

• Both descriptive and causal research involve empirical observation and data collection.
• Both types of research contribute to the scientific understanding of phenomena, albeit through different approaches.

Differences:

• Descriptive research focuses on describing phenomena, while causal research aims to explain why phenomena occur by identifying causal relationships.
• Descriptive research typically uses observational methods, while causal research often involves experimental designs or causal inference techniques to establish causality.

Exploratory vs. Causal Research

Exploratory research  aims to explore new topics, generate hypotheses, or gain initial insights into phenomena. It is often conducted when little is known about a subject and seeks to generate ideas for further investigation.

Causal research , on the other hand, is concerned with testing hypotheses and establishing cause-and-effect relationships between variables. It builds on existing knowledge and seeks to confirm or refute causal hypotheses through systematic investigation.

• Both exploratory and causal research contribute to the generation of knowledge and theory development.
• Both types of research involve systematic inquiry and data analysis to answer research questions.
• Exploratory research focuses on generating hypotheses and exploring new areas of inquiry, while causal research aims to test hypotheses and establish causal relationships.
• Exploratory research is more flexible and open-ended, while causal research follows a more structured and hypothesis-driven approach.

Correlational vs. Causal Research

Correlational research  examines the relationship between variables without implying causation. It identifies patterns of association or co-occurrence between variables but does not establish the direction or causality of the relationship.

Causal research , on the other hand, seeks to establish cause-and-effect relationships between variables by systematically manipulating independent variables and observing their effects on dependent variables. It goes beyond mere association to determine whether changes in one variable directly cause changes in another variable.

• Both correlational and causal research involve analyzing relationships between variables.
• Both types of research contribute to understanding the nature of associations between variables.
• Correlational research focuses on identifying patterns of association, while causal research aims to establish causal relationships.
• Correlational research does not manipulate variables, while causal research involves systematically manipulating independent variables to observe their effects on dependent variables.

How to Formulate Causal Research Hypotheses?

Crafting research questions and hypotheses is the foundational step in any research endeavor. Defining your variables clearly and articulating the causal relationship you aim to investigate is essential. Let's explore this process further.

1. Identify Variables

Identifying variables involves recognizing the key factors you will manipulate or measure in your study. These variables can be classified into independent, dependent, and confounding variables.

• Independent Variable (IV):  This is the variable you manipulate or control in your study. It is the presumed cause that you want to test.
• Dependent Variable (DV):  The dependent variable is the outcome or response you measure. It is affected by changes in the independent variable.
• Confounding Variables:  These are extraneous factors that may influence the relationship between the independent and dependent variables, leading to spurious correlations or erroneous causal inferences. Identifying and controlling for confounding variables is crucial for establishing valid causal relationships.

2. Establish Causality

Establishing causality requires meeting specific criteria outlined by scientific methodology. While correlation between variables may suggest a relationship, it does not imply causation. To establish causality, researchers must demonstrate the following:

• Temporal Precedence:  The cause must precede the effect in time. In other words, changes in the independent variable must occur before changes in the dependent variable.
• Covariation of Cause and Effect:  Changes in the independent variable should be accompanied by corresponding changes in the dependent variable. This demonstrates a consistent pattern of association between the two variables.
• Elimination of Alternative Explanations:  Researchers must rule out other possible explanations for the observed relationship between variables. This involves controlling for confounding variables and conducting rigorous experimental designs to isolate the effects of the independent variable.

3. Write Clear and Testable Hypotheses

Hypotheses serve as tentative explanations for the relationship between variables and provide a framework for empirical testing. A well-formulated hypothesis should be:

• Specific:  Clearly state the expected relationship between the independent and dependent variables.
• Testable:  The hypothesis should be capable of being empirically tested through observation or experimentation.
• Falsifiable:  There should be a possibility of proving the hypothesis false through empirical evidence.

For example, a hypothesis in a study examining the effect of exercise on weight loss could be: "Increasing levels of physical activity (IV) will lead to greater weight loss (DV) among participants (compared to those with lower levels of physical activity)."

By formulating clear hypotheses and operationalizing variables, researchers can systematically investigate causal relationships and contribute to the advancement of scientific knowledge.

Causal Research Design

Designing your research study involves making critical decisions about how you will collect and analyze data to investigate causal relationships.

Experimental vs. Observational Designs

One of the first decisions you'll make when designing a study is whether to employ an experimental or observational design. Each approach has its strengths and limitations, and the choice depends on factors such as the research question, feasibility , and ethical considerations.

• Experimental Design: In experimental designs, researchers manipulate the independent variable and observe its effects on the dependent variable while controlling for confounding variables. Random assignment to experimental conditions allows for causal inferences to be drawn. Example: A study testing the effectiveness of a new teaching method on student performance by randomly assigning students to either the experimental group (receiving the new teaching method) or the control group (receiving the traditional method).
• Observational Design: Observational designs involve observing and measuring variables without intervention. Researchers may still examine relationships between variables but cannot establish causality as definitively as in experimental designs. Example: A study observing the association between socioeconomic status and health outcomes by collecting data on income, education level, and health indicators from a sample of participants.

Control and Randomization

Control and randomization are crucial aspects of experimental design that help ensure the validity of causal inferences.

• Control: Controlling for extraneous variables involves holding constant factors that could influence the dependent variable, except for the independent variable under investigation. This helps isolate the effects of the independent variable. Example: In a medication trial, controlling for factors such as age, gender, and pre-existing health conditions ensures that any observed differences in outcomes can be attributed to the medication rather than other variables.
• Randomization: Random assignment of participants to experimental conditions helps distribute potential confounders evenly across groups, reducing the likelihood of systematic biases and allowing for causal conclusions. Example: Randomly assigning patients to treatment and control groups in a clinical trial ensures that both groups are comparable in terms of baseline characteristics, minimizing the influence of extraneous variables on treatment outcomes.

Internal and External Validity

Two key concepts in research design are internal validity and external validity, which relate to the credibility and generalizability of study findings, respectively.

• Internal Validity: Internal validity refers to the extent to which the observed effects can be attributed to the manipulation of the independent variable rather than confounding factors. Experimental designs typically have higher internal validity due to their control over extraneous variables. Example: A study examining the impact of a training program on employee productivity would have high internal validity if it could confidently attribute changes in productivity to the training intervention.
• External Validity: External validity concerns the extent to which study findings can be generalized to other populations, settings, or contexts. While experimental designs prioritize internal validity, they may sacrifice external validity by using highly controlled conditions that do not reflect real-world scenarios. Example: Findings from a laboratory study on memory retention may have limited external validity if the experimental tasks and conditions differ significantly from real-life learning environments.

Types of Experimental Designs

Several types of experimental designs are commonly used in causal research, each with its own strengths and applications.

• Randomized Control Trials (RCTs): RCTs are considered the gold standard for assessing causality in research. Participants are randomly assigned to experimental and control groups, allowing researchers to make causal inferences. Example: A pharmaceutical company testing a new drug's efficacy would use an RCT to compare outcomes between participants receiving the drug and those receiving a placebo.
• Quasi-Experimental Designs: Quasi-experimental designs lack random assignment but still attempt to establish causality by controlling for confounding variables through design or statistical analysis . Example: A study evaluating the effectiveness of a smoking cessation program might compare outcomes between participants who voluntarily enroll in the program and a matched control group of non-enrollees.

By carefully selecting an appropriate research design and addressing considerations such as control, randomization, and validity, researchers can conduct studies that yield credible evidence of causal relationships and contribute valuable insights to their field of inquiry.

Causal Research Data Collection

Collecting data is a critical step in any research study, and the quality of the data directly impacts the validity and reliability of your findings.

Choosing Measurement Instruments

Selecting appropriate measurement instruments is essential for accurately capturing the variables of interest in your study. The choice of measurement instrument depends on factors such as the nature of the variables, the target population , and the research objectives.

• Surveys :  Surveys are commonly used to collect self-reported data on attitudes, opinions, behaviors, and demographics . They can be administered through various methods, including paper-and-pencil surveys, online surveys, and telephone interviews.
• Observations:  Observational methods involve systematically recording behaviors, events, or phenomena as they occur in natural settings. Observations can be structured (following a predetermined checklist) or unstructured (allowing for flexible data collection).
• Psychological Tests:  Psychological tests are standardized instruments designed to measure specific psychological constructs, such as intelligence, personality traits, or emotional functioning. These tests often have established reliability and validity.
• Physiological Measures:  Physiological measures, such as heart rate, blood pressure, or brain activity, provide objective data on bodily processes. They are commonly used in health-related research but require specialized equipment and expertise.
• Existing Databases:  Researchers may also utilize existing datasets, such as government surveys, public health records, or organizational databases, to answer research questions. Secondary data analysis can be cost-effective and time-saving but may be limited by the availability and quality of data.

Ensuring accurate data collection is the cornerstone of any successful research endeavor. With the right tools in place, you can unlock invaluable insights to drive your causal research forward. From surveys to tests, each instrument offers a unique lens through which to explore your variables of interest.

At Appinio , we understand the importance of robust data collection methods in informing impactful decisions. Let us empower your research journey with our intuitive platform, where you can effortlessly gather real-time consumer insights to fuel your next breakthrough.   Ready to take your research to the next level? Book a demo today and see how Appinio can revolutionize your approach to data collection!

Book a Demo

Sampling Techniques

Sampling involves selecting a subset of individuals or units from a larger population to participate in the study. The goal of sampling is to obtain a representative sample that accurately reflects the characteristics of the population of interest.

• Probability Sampling:  Probability sampling methods involve randomly selecting participants from the population, ensuring that each member of the population has an equal chance of being included in the sample. Common probability sampling techniques include simple random sampling , stratified sampling, and cluster sampling .
• Non-Probability Sampling:  Non-probability sampling methods do not involve random selection and may introduce biases into the sample. Examples of non-probability sampling techniques include convenience sampling, purposive sampling, and snowball sampling.

The choice of sampling technique depends on factors such as the research objectives, population characteristics, resources available, and practical constraints. Researchers should strive to minimize sampling bias and maximize the representativeness of the sample to enhance the generalizability of their findings.

Ethical Considerations

Ethical considerations are paramount in research and involve ensuring the rights, dignity, and well-being of research participants. Researchers must adhere to ethical principles and guidelines established by professional associations and institutional review boards (IRBs).

• Informed Consent:  Participants should be fully informed about the nature and purpose of the study, potential risks and benefits, their rights as participants, and any confidentiality measures in place. Informed consent should be obtained voluntarily and without coercion.
• Privacy and Confidentiality:  Researchers should take steps to protect the privacy and confidentiality of participants' personal information. This may involve anonymizing data, securing data storage, and limiting access to identifiable information.
• Minimizing Harm:  Researchers should mitigate any potential physical, psychological, or social harm to participants. This may involve conducting risk assessments, providing appropriate support services, and debriefing participants after the study.
• Respect for Participants:  Researchers should respect participants' autonomy, diversity, and cultural values. They should seek to foster a trusting and respectful relationship with participants throughout the research process.
• Publication and Dissemination:  Researchers have a responsibility to accurately report their findings and acknowledge contributions from participants and collaborators. They should adhere to principles of academic integrity and transparency in disseminating research results.

By addressing ethical considerations in research design and conduct, researchers can uphold the integrity of their work, maintain trust with participants and the broader community, and contribute to the responsible advancement of knowledge in their field.

Causal Research Data Analysis

Once data is collected, it must be analyzed to draw meaningful conclusions and assess causal relationships.

Causal Inference Methods

Causal inference methods are statistical techniques used to identify and quantify causal relationships between variables in observational data. While experimental designs provide the most robust evidence for causality, observational studies often require more sophisticated methods to account for confounding factors.

• Difference-in-Differences (DiD):  DiD compares changes in outcomes before and after an intervention between a treatment group and a control group, controlling for pre-existing trends. It estimates the average treatment effect by differencing the changes in outcomes between the two groups over time.
• Instrumental Variables (IV):  IV analysis relies on instrumental variables—variables that affect the treatment variable but not the outcome—to estimate causal effects in the presence of endogeneity. IVs should be correlated with the treatment but uncorrelated with the error term in the outcome equation.
• Regression Discontinuity (RD):  RD designs exploit naturally occurring thresholds or cutoff points to estimate causal effects near the threshold. Participants just above and below the threshold are compared, assuming that they are similar except for their proximity to the threshold.
• Propensity Score Matching (PSM):  PSM matches individuals or units based on their propensity scores—the likelihood of receiving the treatment—creating comparable groups with similar observed characteristics. Matching reduces selection bias and allows for causal inference in observational studies.

Assessing Causality Strength

Assessing the strength of causality involves determining the magnitude and direction of causal effects between variables. While statistical significance indicates whether an observed relationship is unlikely to occur by chance, it does not necessarily imply a strong or meaningful effect.

• Effect Size:  Effect size measures the magnitude of the relationship between variables, providing information about the practical significance of the results. Standard effect size measures include Cohen's d for mean differences and odds ratios for categorical outcomes.
• Confidence Intervals:  Confidence intervals provide a range of values within which the actual effect size is likely to lie with a certain degree of certainty. Narrow confidence intervals indicate greater precision in estimating the true effect size.
• Practical Significance:  Practical significance considers whether the observed effect is meaningful or relevant in real-world terms. Researchers should interpret results in the context of their field and the implications for stakeholders.

Handling Confounding Variables

Confounding variables are extraneous factors that may distort the observed relationship between the independent and dependent variables, leading to spurious or biased conclusions. Addressing confounding variables is essential for establishing valid causal inferences.

• Statistical Control:  Statistical control involves including confounding variables as covariates in regression models to partially out their effects on the outcome variable. Controlling for confounders reduces bias and strengthens the validity of causal inferences.
• Matching:  Matching participants or units based on observed characteristics helps create comparable groups with similar distributions of confounding variables. Matching reduces selection bias and mimics the randomization process in experimental designs.
• Sensitivity Analysis:  Sensitivity analysis assesses the robustness of study findings to changes in model specifications or assumptions. By varying analytical choices and examining their impact on results, researchers can identify potential sources of bias and evaluate the stability of causal estimates.
• Subgroup Analysis:  Subgroup analysis explores whether the relationship between variables differs across subgroups defined by specific characteristics. Identifying effect modifiers helps understand the conditions under which causal effects may vary.

By employing rigorous causal inference methods, assessing the strength of causality, and addressing confounding variables, researchers can confidently draw valid conclusions about causal relationships in their studies, advancing scientific knowledge and informing evidence-based decision-making.

Causal Research Examples

Examples play a crucial role in understanding the application of causal research methods and their impact across various domains. Let's explore some detailed examples to illustrate how causal research is conducted and its real-world implications:

Example 1: Software as a Service (SaaS) User Retention Analysis

Suppose a SaaS company wants to understand the factors influencing user retention and engagement with their platform. The company conducts a longitudinal observational study, collecting data on user interactions, feature usage, and demographic information over several months.

• Design:  The company employs an observational cohort study design, tracking cohorts of users over time to observe changes in retention and engagement metrics. They use analytics tools to collect data on user behavior , such as logins, feature usage, session duration, and customer support interactions.
• Data Collection:  Data is collected from the company's platform logs, customer relationship management (CRM) system, and user surveys. Key metrics include user churn rates, active user counts, feature adoption rates, and Net Promoter Scores ( NPS ).
• Analysis:  Using statistical techniques like survival analysis and regression modeling, the company identifies factors associated with user retention, such as feature usage patterns, onboarding experiences, customer support interactions, and subscription plan types.
• Findings: The analysis reveals that users who engage with specific features early in their lifecycle have higher retention rates, while those who encounter usability issues or lack personalized onboarding experiences are more likely to churn. The company uses these insights to optimize product features, improve onboarding processes, and enhance customer support strategies to increase user retention and satisfaction.

Example 2: Business Impact of Digital Marketing Campaign

Consider a technology startup launching a digital marketing campaign to promote its new product offering. The company conducts an experimental study to evaluate the effectiveness of different marketing channels in driving website traffic, lead generation, and sales conversions.

• Design:  The company implements an A/B testing design, randomly assigning website visitors to different marketing treatment conditions, such as Google Ads, social media ads, email campaigns, or content marketing efforts. They track user interactions and conversion events using web analytics tools and marketing automation platforms.
• Data Collection:  Data is collected on website traffic, click-through rates, conversion rates, lead generation, and sales revenue. The company also gathers demographic information and user feedback through surveys and customer interviews to understand the impact of marketing messages and campaign creatives .
• Analysis:  Utilizing statistical methods like hypothesis testing and multivariate analysis, the company compares key performance metrics across different marketing channels to assess their effectiveness in driving user engagement and conversion outcomes. They calculate return on investment (ROI) metrics to evaluate the cost-effectiveness of each marketing channel.
• Findings:  The analysis reveals that social media ads outperform other marketing channels in generating website traffic and lead conversions, while email campaigns are more effective in nurturing leads and driving sales conversions. Armed with these insights, the company allocates marketing budgets strategically, focusing on channels that yield the highest ROI and adjusting messaging and targeting strategies to optimize campaign performance.

These examples demonstrate the diverse applications of causal research methods in addressing important questions, informing policy decisions, and improving outcomes in various fields. By carefully designing studies, collecting relevant data, employing appropriate analysis techniques, and interpreting findings rigorously, researchers can generate valuable insights into causal relationships and contribute to positive social change.

How to Interpret Causal Research Results?

Interpreting and reporting research findings is a crucial step in the scientific process, ensuring that results are accurately communicated and understood by stakeholders.

Interpreting Statistical Significance

Statistical significance indicates whether the observed results are unlikely to occur by chance alone, but it does not necessarily imply practical or substantive importance. Interpreting statistical significance involves understanding the meaning of p-values and confidence intervals and considering their implications for the research findings.

• P-values:  A p-value represents the probability of obtaining the observed results (or more extreme results) if the null hypothesis is true. A p-value below a predetermined threshold (typically 0.05) suggests that the observed results are statistically significant, indicating that the null hypothesis can be rejected in favor of the alternative hypothesis.
• Confidence Intervals:  Confidence intervals provide a range of values within which the true population parameter is likely to lie with a certain degree of confidence (e.g., 95%). If the confidence interval does not include the null value, it suggests that the observed effect is statistically significant at the specified confidence level.

Interpreting statistical significance requires considering factors such as sample size, effect size, and the practical relevance of the results rather than relying solely on p-values to draw conclusions.

Discussing Practical Significance

While statistical significance indicates whether an effect exists, practical significance evaluates the magnitude and meaningfulness of the effect in real-world terms. Discussing practical significance involves considering the relevance of the results to stakeholders and assessing their impact on decision-making and practice.

• Effect Size:  Effect size measures the magnitude of the observed effect, providing information about its practical importance. Researchers should interpret effect sizes in the context of their field and the scale of measurement (e.g., small, medium, or large effect sizes).
• Contextual Relevance:  Consider the implications of the results for stakeholders, policymakers, and practitioners. Are the observed effects meaningful in the context of existing knowledge, theory, or practical applications? How do the findings contribute to addressing real-world problems or informing decision-making?

Discussing practical significance helps contextualize research findings and guide their interpretation and application in practice, beyond statistical significance alone.

Addressing Limitations and Assumptions

No study is without limitations, and researchers should transparently acknowledge and address potential biases, constraints, and uncertainties in their research design and findings.

• Methodological Limitations:  Identify any limitations in study design, data collection, or analysis that may affect the validity or generalizability of the results. For example, sampling biases , measurement errors, or confounding variables.
• Assumptions:  Discuss any assumptions made in the research process and their implications for the interpretation of results. Assumptions may relate to statistical models, causal inference methods, or theoretical frameworks underlying the study.
• Alternative Explanations:  Consider alternative explanations for the observed results and discuss their potential impact on the validity of causal inferences. How robust are the findings to different interpretations or competing hypotheses?

Addressing limitations and assumptions demonstrates transparency and rigor in the research process, allowing readers to critically evaluate the validity and reliability of the findings.

Communicating Findings Clearly

Effectively communicating research findings is essential for disseminating knowledge, informing decision-making, and fostering collaboration and dialogue within the scientific community.

• Clarity and Accessibility:  Present findings in a clear, concise, and accessible manner, using plain language and avoiding jargon or technical terminology. Organize information logically and use visual aids (e.g., tables, charts, graphs) to enhance understanding.
• Contextualization:  Provide context for the results by summarizing key findings, highlighting their significance, and relating them to existing literature or theoretical frameworks. Discuss the implications of the findings for theory, practice, and future research directions.
• Transparency:  Be transparent about the research process, including data collection procedures, analytical methods, and any limitations or uncertainties associated with the findings. Clearly state any conflicts of interest or funding sources that may influence interpretation.

By communicating findings clearly and transparently, researchers can facilitate knowledge exchange, foster trust and credibility, and contribute to evidence-based decision-making.

Causal Research Tips

When conducting causal research, it's essential to approach your study with careful planning, attention to detail, and methodological rigor. Here are some tips to help you navigate the complexities of causal research effectively:

• Define Clear Research Questions:  Start by clearly defining your research questions and hypotheses. Articulate the causal relationship you aim to investigate and identify the variables involved.
• Consider Alternative Explanations:  Be mindful of potential confounding variables and alternative explanations for the observed relationships. Take steps to control for confounders and address alternative hypotheses in your analysis.
• Prioritize Internal Validity:  While external validity is important for generalizability, prioritize internal validity in your study design to ensure that observed effects can be attributed to the manipulation of the independent variable.
• Use Randomization When Possible:  If feasible, employ randomization in experimental designs to distribute potential confounders evenly across experimental conditions and enhance the validity of causal inferences.
• Be Transparent About Methods:  Provide detailed descriptions of your research methods, including data collection procedures, analytical techniques, and any assumptions or limitations associated with your study.
• Utilize Multiple Methods:  Consider using a combination of experimental and observational methods to triangulate findings and strengthen the validity of causal inferences.
• Be Mindful of Sample Size:  Ensure that your sample size is adequate to detect meaningful effects and minimize the risk of Type I and Type II errors. Conduct power analyses to determine the sample size needed to achieve sufficient statistical power.
• Validate Measurement Instruments:  Validate your measurement instruments to ensure that they are reliable and valid for assessing the variables of interest in your study. Pilot test your instruments if necessary.
• Seek Feedback from Peers:  Collaborate with colleagues or seek feedback from peer reviewers to solicit constructive criticism and improve the quality of your research design and analysis.

Conclusion for Causal Research

Mastering causal research empowers researchers to unlock the secrets of cause and effect, shedding light on the intricate relationships between variables in diverse fields. By employing rigorous methods such as experimental designs, causal inference techniques, and careful data analysis, you can uncover causal mechanisms, predict outcomes, and inform evidence-based practices. Through the lens of causal research, complex phenomena become more understandable, and interventions become more effective in addressing societal challenges and driving progress. In a world where understanding the reasons behind events is paramount, causal research serves as a beacon of clarity and insight. Armed with the knowledge and techniques outlined in this guide, you can navigate the complexities of causality with confidence, advancing scientific knowledge, guiding policy decisions, and ultimately making meaningful contributions to our understanding of the world.

How to Conduct Causal Research in Minutes?

Introducing Appinio , your gateway to lightning-fast causal research. As a real-time market research platform, we're revolutionizing how companies gain consumer insights to drive data-driven decisions. With Appinio, conducting your own market research is not only easy but also thrilling. Experience the excitement of market research with Appinio, where fast, intuitive, and impactful insights are just a click away.

Here's why you'll love Appinio:

• Instant Insights:  Say goodbye to waiting days for research results. With our platform, you'll go from questions to insights in minutes, empowering you to make decisions at the speed of business.
• User-Friendly Interface:  No need for a research degree here! Our intuitive platform is designed for anyone to use, making complex research tasks simple and accessible.
• Global Reach:  Reach your target audience wherever they are. With access to over 90 countries and the ability to define precise target groups from 1200+ characteristics, you'll gather comprehensive data to inform your decisions.

Get free access to the platform!

Join the loop 💌

Be the first to hear about new updates, product news, and data insights. We'll send it all straight to your inbox.

Get the latest market research news straight to your inbox! 💌

Wait, there's more

17.04.2024 | 25min read

Quota Sampling: Definition, Types, Methods, Examples

15.04.2024 | 34min read

What is Market Share? Definition, Formula, Examples

11.04.2024 | 34min read

What is Data Analysis? Definition, Tools, Examples

Bivariate Analysis: Associations, Hypotheses, and Causal Stories

• Open Access
• First Online: 04 October 2022

Cite this chapter

You have full access to this open access chapter

• Mark Tessler 2  

Part of the book series: SpringerBriefs in Sociology ((BRIEFSSOCY))

2776 Accesses

Every day, we encounter various phenomena that make us question how, why, and with what implications they vary. In responding to these questions, we often begin by considering bivariate relationships, meaning the way that two variables relate to one another. Such relationships are the focus of this chapter.

You have full access to this open access chapter,  Download chapter PDF

3.1 Description, Explanation, and Causal Stories

There are many reasons why we might be interested in the relationship between two variables. Suppose we observe that some of the respondents interviewed in Arab Barometer surveys and other surveys report that they have thought about emigrating, and we are interested in this variable. We may want to know how individuals’ consideration of emigration varies in relation to certain attributes or attitudes. In this case, our goal would be descriptive , sometimes described as the mapping of variance. Our goal may also or instead be explanation , such as when we want to know why individuals have thought about emigrating.

Description

Description means that we seek to increase our knowledge and refine our understanding of a single variable by looking at whether and how it varies in relation to one or more other variables. Descriptive information makes a valuable contribution when the structure and variance of an important phenomenon are not well known, or not well known in relation to other important variables.

Returning to the example about emigration, suppose you notice that among Jordanians interviewed in 2018, 39.5 percent of the 2400 men and women interviewed reported that they have considered the possibility of emigrating.

Our objective may be to discover what these might-be migrants look like and what they are thinking. We do this by mapping the variance of emigration across attributes and orientations that provide some of this descriptive information, with the descriptions themselves each expressed as bivariate relationships. These relationships are also sometimes labeled “associations” or “correlations” since they are not considered causal relationships and are not concerned with explanation.

Of the 39.5 percent of Jordanians who told interviewers that they have considered emigrating, 57.3 percent are men and 42.7 percent are women. With respect to age, 34 percent are age 29 or younger and 19.2 percent are age 50 or older. It might have been expected that a higher percentage of respondents age 29 or younger would have considered emigrating. In fact, however, 56 percent of the 575 men and women in this age category have considered emigrating. And with respect to destination, the Arab country most frequently mentioned by those who have considered emigration is the UAE, named by 17 percent, followed by Qatar at 10 percent and Saudi Arabia at 9.8 percent. Non-Arab destinations were mentioned more frequently, with Turkey named by 18.1 percent, Canada by 21.1 percent, and the U.S. by 24.2 percent.

With the variables sex, age, and prospective destination added to the original variable, which is consideration of emigration, there are clearly more than two variables under consideration. But the variables are described two at a time and so each relationship is bivariate.

These bivariate relationships, between having considered emigration on the one hand and sex, age, and prospective destination on the other, provide descriptive information that is likely to be useful to analysts, policymakers, and others concerned with emigration. They tell, or begin to tell, as noted above, what might-be migrants look like and what they are thinking. Still additional insight may be gained by adding descriptive bivariate relationships for Jordanians interviewed in a different year to those interviewed in 2018. In addition, of course, still more information and possibly a more refined understanding, may be gained by examining the attributes and orientations of prospective emigrants who are citizens of other Arab (and perhaps also non-Arab) countries.

With a focus on description, these bivariate relationships are not constructed to shed light on explanation, that is to contribute to causal stories that seek to account for variance and tell why some individuals but not others have considered the possibility of emigrating. In fact, however, as useful as bivariate relationships that provide descriptive information may be, researchers usually are interested as much if not more in bivariate relationships that express causal stories and purport to provide explanations.

Explanation and Causal Stories

There is a difference in the origins of bivariate relationships that seek to provide descriptive information and those that seek to provide explanatory information. The former can be thought to be responding to what questions: What characterizes potential emigrants? What do they look like? What are their thoughts about this or that subject? If the objective is description, a researcher collects and uses her data to investigate the relationship between two variables without a specific and firm prediction about the relationship between them. Rather, she simply wonders about the “what” questions listed above and believes that finding out the answers will be instructive. In this case, therefore, she selects the bivariate relationships to be considered based on what she thinks it will be useful to know, and not based on assessing the accuracy of a previously articulated causal story that specifies the strength and structure of the effect that one variable has on the other.

A researcher is often interested in causal stories and explanation, however, and this does usually begin with thinking about the relationship between two variables, one of which is the presumed cause and the other the presumed effect. The presumed cause is the independent variable, and the presumed effect is the dependent variable . Offering evidence that there is a strong relationship between two variables is not sufficient to demonstrate that the variables are likely to be causally related, but it is a necessary first step. In this respect it is a point of departure for the fuller, probably multivariate analysis, required to persuasively argue that a relationship is likely to be causal. In addition, as discussed in Chap. 4 , multivariate analysis often not only strengthens the case for inferring that a relationship is causal, but also provides a more elaborate and more instructive causal story. The foundation, however, on which a multivariate analysis aimed at causal inference is built, is a bivariate relationship composed of a presumed independent variable and a presumed dependent variable.

A hypothesis that posits a causal relationship between two variables is not the same as a causal story, although the two are of course closely connected. The former specifies a presumed cause, a presumed determinant of variance on the dependent variable. It probably also specifies the structure of the relationship, such as linear as opposed to non-linear, or positive (also called direct) as opposed to negative (also called inverse).

On the other hand, a causal story describes in more detail what the researcher believes is actually taking place in the relationship between the variables in her hypothesis; and accordingly, why she thinks this involves causality. A causal story provides a fuller account of operative processes, processes that the hypothesis references but does not spell out. These processes may, for example, involve a pathway or a mechanism that tells how it is that the independent variable causes and thus accounts for some of the variance on the dependent variable. Expressed yet another way, the causal story describes the researcher’s understandings, or best guesses, about the real world, understandings that have led her to believe, and then propose for testing, that there is a causal connection between her variables that deserves investigation. The hypothesis itself does not tell this story; it is rather a short formulation that references and calls attention to the existence, or hypothesized existence, of a causal story. Research reports present the causal story as well as the hypothesis, as the hypothesis is often of limited interpretability without the causal story.

A causal story is necessary for causal inference. It enables the researcher to formulate propositions that purport to explain rather than merely describe or predict. There may be a strong relationship between two variables, and if this is the case, it will be possible to predict with relative accuracy the value, or score, of one variable from knowledge of the value, or score, of the other variable. Prediction is not explanation, however. To explain, or attribute causality, there must be a causal story to which a hypothesized causal relationship is calling attention.

An instructive illustration is provided by a recent study of Palestinian participation in protest activities that express opposition to Israeli occupation. Footnote 1 There is plenty of variance on the dependent variable: There are many young Palestinians who take part in these activities, and there are many others who do not take part. Education is one of the independent variables that the researcher thought would be an important determinant of participation, and so she hypothesized that individuals with more education would be more likely to participate in protest activities than individuals with less education.

But why would the researcher think this? The answer is provided by the causal story. To the extent that this as yet untested story is plausible, or preferably, persuasive, at least in the eyes of the investigator, it gives the researcher a reason to believe that education is indeed a determinant of participation in protest activities in Palestine. By spelling out in some detail how and why the hypothesized independent variable, education in this case, very likely impacts a person’s decision about whether or not to protest, the causal story provides a rationale for the researcher’s hypothesis.

In the case of Palestinian participation in protest activities, another investigator offered an insightful causal story about the ways that education pushes toward greater participation, with emphasis on its role in communication and coordination. Footnote 2 Schooling, as the researcher theorizes and subsequently tests, integrates young Palestinians into a broader institutional environment that facilitates mass mobilizations and lowers informational and organizational barriers to collective action. More specifically, she proposes that those individuals who have had at least a middle school education, compared to those who have not finished middle school, have access to better and more reliable sources of information, which, among other things, enables would-be protesters to assess risks. More schooling also makes would-be protesters better able to forge inter-personal relationships and establish networks that share information about needs, opportunities, and risks, and that in this way facilitate engaging in protest activities in groups, rather than on an individual basis. This study offers some additional insights to be discussed later.

The variance motivating the investigation of a causal story may be thought of as the “variable of interest,” and it may be either an independent variable or a dependent variable. It is a variable of interest because the way that it varies poses a question, or puzzle, that a researcher seeks to investigate. It is the dependent variable in a bivariate relationship if the researcher seeks to know why this variable behaves, or varies, as it does, and in pursuit of this objective, she will seek to identify the determinants and drivers that account for this variance. The variable of interest is an independent variable in a particular research project if the researcher seeks to know what difference it makes—on what does its variance have an impact, of what other variable or variables is it a driver or determinant.

The variable in which a researcher is initially interested, that is to say the variable of interest, can also be both a dependent variable and an independent variable. Returning to the variable pertaining to consideration of emigration, but this time with country as the unit of analysis, the variance depicted in Table 3.1 provides an instructive example. The data are based on Arab Barometer surveys conducted in 2018–2019, and the table shows that there is substantial variation across twelve countries. Taking the countries together, the mean percentage of citizens that have thought about relocating to another country is 30.25 percent. But in fact, there is very substantial variation around this mean. Kuwait is an outlier, with only 8 percent having considered emigration. There are also countries in which only 21 percent or 22 percent of the adult population have thought about this, figures that may be high in absolute terms but are low relative to other Arab countries. At the other end of the spectrum are countries in which 45 percent or even 50 percent of the citizens report having considered leaving their country and relocating elsewhere.

The very substantial variance shown in Table 3.1 invites reflection on both the causes and the consequences of this country-level variable, aggregate thinking about emigration. As a dependent variable, the cross-country variance brings the question of why the proportion of citizens that have thought about emigrating is higher in some countries than in others; and the search for an answer begins with the specification of one or more bivariate relationships, each of which links this dependent variable to a possible cause or determinant. As an independent variable, the cross-country variance brings the question of what difference does it make—of what is it a determinant or driver and what are the consequences for a country if more of its citizens, rather than fewer, have thought about moving to another country.

3.2 Hypotheses and Formulating Hypotheses

Hypotheses emerge from the research questions to which a study is devoted. Accordingly, a researcher interested in explanation will have something specific in mind when she decides to hypothesize and then evaluate a bivariate relationship in order to determine whether, and if so how, her variable of interest is related to another variable. For example, if the researcher’s variable of interest is attitude toward gender equality and one of her research questions asks why some people support gender equality and others do not, she might formulate the hypothesis below to see if education provides part of the answer.

Hypothesis 1. Individuals who are better educated are more likely to support gender equality than are individuals who are less well-educated.

The usual case, and the preferred case, is for an investigator to be specific about the research questions she seeks to answer, and then to formulate hypotheses that propose for testing part of the answer to one or more of these questions. Sometimes, however, a researcher will proceed without formulating specific hypotheses based on her research questions. Sometimes she will simply look at whatever relationships between her variable of interest and a second variable her data permit her to identify and examine, and she will then follow up and incorporate into her study any findings that turn out to be significant and potentially instructive. This is sometimes described as allowing the data to “speak.” When this hit or miss strategy of trial and error is used in bivariate and multivariate analysis, findings that are significant and potentially instructive are sometimes described as “grounded theory.” Some researchers also describe the latter process as “inductive” and the former as “deductive.”

Although the inductive, atheoretical approach to data analysis might yield some worthwhile findings that would otherwise have been missed, it can sometimes prove misleading, as you may discover relationships between variables that happened by pure chance and are not instructive about the variable of interest or research question. Data analysis in research aimed at explanation should be, in most cases, preceded by the formulation of one or more hypotheses. In this context, when the focus is on bivariate relationships and the objective is explanation rather than description, each hypothesis will include a dependent variable and an independent variable and make explicit the way the researcher thinks the two are, or probably are, related. As discussed, the dependent variable is the presumed effect; its variance is what a hypothesis seeks to explain. The independent variable is the presumed cause; its impact on the variance of another variable is what the hypothesis seeks to determine.

Hypotheses are usually in the form of if-then, or cause-and-effect, propositions. They posit that if there is variance on the independent variable, the presumed cause, there will then be variance on the dependent variable, the presumed effect. This is because the former impacts the latter and causes it to vary.

An illustration of formulating hypotheses is provided by a study of voting behavior in seven Arab countries: Algeria, Bahrain, Jordan, Lebanon, Morocco, Palestine, and Yemen. Footnote 3 The variable of interest in this individual-level study is electoral turnout, and prominent among the research questions is why some citizens vote and others do not. The dependent variable in the hypotheses proposed in response to this question is whether a person did or did not vote in the country’s most recent parliamentary election. The study initially proposed a number of hypotheses, which include the two listed here and which would later be tested with data from Arab Barometer surveys in the seven countries in 2006–2007. We will return to this illustration later in this chapter.

Hypothesis 1: Individuals who have used clientelist networks in the past are more likely to turn out to vote than are individuals who have not used clientelist networks in the past.

Hypothesis 2: Individuals with a positive evaluation of the economy are more likely to vote than are individuals with a negative evaluation of the economy.

Another example pertaining to voting, which this time is hypothetical but might be instructively tested with Arab Barometer data, considers the relationship between perceived corruption and turning out to vote at the individual level of analysis.

The normal expectation in this case would be that perceptions of corruption influence the likelihood of voting. Even here, however, competing causal relationships are plausible. More perceived corruption might increase the likelihood of voting, presumably to register discontent with those in power. But greater perceived corruption might also actually reduce the likelihood of voting, presumably in this case because the would-be voter sees no chance that her vote will make a difference. But in this hypothetical case, even the direction of the causal connection might be ambiguous. If voting is complicated, cumbersome, and overly bureaucratic, it might be that the experience of voting plays a role in shaping perceptions of corruption. In cases like this, certain variables might be both independent and dependent variables, with causal influence pushing in both directions (often called “endogeneity”), and the researcher will need to carefully think through and be particularly clear about the causal story to which her hypothesis is designed to call attention.

The need to assess the accuracy of these hypotheses, or any others proposed to account for variance on a dependent variable, will guide and shape the researcher’s subsequent decisions about data collection and data analysis. Moreover, in most cases, the finding produced by data analysis is not a statement that the hypothesis is true or that the hypothesis is false. It is rather a statement that the hypothesis is probably true or it is probably false. And more specifically still, when testing a hypothesis with quantitative data, it is often a statement about the odds, or probability, that the researcher will be wrong if she concludes that the hypothesis is correct—if she concludes that the independent variable in the hypothesis is indeed a significant determinant of the variance on the dependent variable. The lower the probability of being wrong, of course, the more confident a researcher can be in concluding, and reporting, that her data and analysis confirm her hypothesis.

Exercise 3.1

Hypotheses emerge from the research questions to which a study is devoted. Thinking about one or more countries with which you are familiar: (a) Identify the independent and dependent variables in each of the example research questions below. (b) Formulate at least one hypothesis for each question. Make sure to include your expectations about the directionality of the relationship between the two variables; is it positive/direct or negative/inverse? (c) In two or three sentences, describe a plausible causal story to which each of your hypotheses might call attention.

Does religiosity affect people’s preference for democracy?

Does preference for democracy affect the likelihood that a person will vote? Footnote 4

Exercise 3.2

Since its establishment in 2006, the Arab Barometer has, as of spring 2022, conducted 68 social and political attitude surveys in the Middle East and North Africa. It has conducted one or more surveys in 16 different Arab countries, and it has recorded the attitudes, values, and preferences of more than 100,000 ordinary citizens.

The Arab Barometer website ( arabbarometer.org ) provides detailed information about the Barometer itself and about the scope, methodology, and conduct of its surveys. Data from the Barometer’s surveys can be downloaded in either SPSS, Stata, or csv format. The website also contains numerous reports, articles, and summaries of findings.

In addition, the Arab Barometer website contains an Online Data Analysis Tool that makes it possible, without downloading any data, to find the distribution of responses to any question asked in any country in any wave. The tool is found in the “Survey Data” menu. After selecting the country and wave of interest, click the “See Results” tab to select the question(s) for which you want to see the response distributions. Click the “Cross by” tab to see the distributions of respondents who differ on one of the available demographic attributes.

The charts below present, in percentages, the response distributions of Jordanians interviewed in 2018 to two questions about gender equality. Below the charts are questions that you are asked to answer. These questions pertain to formulating hypotheses and to the relationship between hypotheses and causal stories.

For each of the two distributions, do you think (hypothesize) that the attitudes of Jordanian women are:

About the same as those of Jordanian men

More favorable toward gender equality than those of Jordanian men

Less favorable toward gender equality than those of Jordanian men

For each of the two distributions, do you think (hypothesize) that the attitudes of younger Jordanians are:

About the same as those of older Jordanians

More favorable toward gender equality than those of older Jordanians

Less favorable toward gender equality than those of older Jordanians

Restate your answers to Questions 1 and 2 as hypotheses.

Give the reasons for your answers to Questions 1 and 2. In two or three sentences, make explicit the presumed causal story on which your hypotheses are based.

Using the Arab Barometer’s Online Analysis Tool, check to see whether your answers to Questions 1 and 2 are correct. For those instances in which an answer is incorrect, suggest in a sentence or two a causal story on which the correct relationship might be based.

In which other country surveyed by the Arab Barometer in 2018 do you think the distributions of responses to the questions about gender equality are very similar to the distributions in Jordan? What attributes of Jordan and the other country informed your selection of the other country?

In which other country surveyed by the Arab Barometer in 2018 do you think the distributions of responses to the questions about gender equality are very different from the distributions in Jordan? What attributes of Jordan and the other country informed your selection of the other country?

Using the Arab Barometer’s Online Analysis Tool, check to see whether your answers to Questions 6 and 7 are correct. For those instances in which an answer is incorrect, suggest in a sentence or two a causal story on which the correct relationship might be based.

We will shortly return to and expand the discussion of probabilities and of hypothesis testing more generally. First, however, some additional discussion of hypothesis formulation is in order. Three important topics will be briefly considered. The first concerns the origins of hypotheses; the second concerns the criteria by which the value of a particular hypothesis or set of hypotheses should be evaluated; and the third, requiring a bit more discussion, concerns the structure of the hypothesized relationship between an independent variable and a dependent variable, or between any two variables that are hypothesized to be related.

Origins of Hypotheses

Where do hypotheses come from? How should an investigator identify independent variables that may account for much, or at least some, of the variance on a dependent variable that she has observed and in which she is interested? Or, how should an investigator identify dependent variables whose variance has been determined, presumably only in part, by an independent variable whose impact she deems it important to assess.

Previous research is one place the investigator may look for ideas that will shape her hypotheses and the associated causal stories. This may include previous hypothesis-testing research, and this can be particularly instructive, but it may also include less systematic and structured observations, reports, and testimonies. The point, very simply, is that the investigator almost certainly is not the first person to think about, and offer information and insight about, the topic and questions in which the researcher herself is interested. Accordingly, attention to what is already known will very likely give the researcher some guidance and ideas as she strives for originality and significance in delineating the relationship between the variables in which she is interested.

Consulting previous research will also enable the researcher to determine what her study will add to what is already known—what it will contribute to the collective and cumulative work of researchers and others who seek to reduce uncertainty about a topic in which they share an interest. Perhaps the researcher’s study will fill an important gap in the scientific literature. Perhaps it will challenge and refine, or perhaps even place in doubt, distributions and explanations of variance that have thus far been accepted. Or perhaps her study will produce findings that shed light on the generalizability or scope conditions of previously accepted variable relationships. It need not do any of these things, but that will be for the researcher to decide, and her decision will be informed by knowledge of what is already known and reflection on whether and in what ways her study should seek to add to that body of knowledge.

Personal experience will also inform the researcher’s search for meaningful and informative hypotheses. It is almost certainly the case that a researcher’s interest in a topic in general, and in questions pertaining to this topic in particular, have been shaped by her own experience. The experience itself may involve many different kinds of connections or interactions, some more professional and work-related and some flowing simply and perhaps unintentionally from lived experience. The hypotheses about voting mentioned earlier, for example, might be informed by elections the researcher has witnessed and/or discussions with friends and colleagues about elections, their turnout, and their fairness. Or perhaps the researcher’s experience in her home country has planted questions about the generalizability of what she has witnessed at home.

All of this is to some extent obvious. But the take-away is that an investigator should not endeavor to set aside what she has learned about a topic in the name of objectivity, but rather, she should embrace whatever personal experience has taught her as she selects and refines the puzzles and propositions she will investigate. Should it happen that her experience leads her to incorrect or perhaps distorted understandings, this will be brought to light when her hypotheses are tested. It is in the testing that objectivity is paramount. In hypothesis formation, by contrast, subjectivity is permissible, and, in fact, it may often be unavoidable.

A final arena in which an investigator may look for ideas that will shape her hypotheses overlaps with personal experience and is also to some extent obvious. This is referenced by terms like creativity and originality and is perhaps best captured by the term “sociological imagination.” The take-away here is that hypotheses that deserve attention and, if confirmed, will provide important insights, may not all be somewhere out in the environment waiting to be found, either in the relevant scholarly literature or in recollections about relevant personal experience. They can and sometimes will be the product of imagination and wondering, of discernments that a researcher may come upon during moments of reflection and deliberation.

As in the case of personal experience, the point to be retained is that hypothesis formation may not only be a process of discovery, of finding the previous research that contains the right information. Hypothesis formation may also be a creative process, a process whereby new insights and proposed original understandings are the product of an investigator’s intellect and sociological imagination.

Crafting Valuable Hypotheses

What are the criteria by which the value of a hypothesis or set of hypotheses should be evaluated? What elements define a good hypothesis? Some of the answers to these questions that come immediately to mind pertain to hypothesis testing rather than hypothesis formation. A good hypothesis, it might be argued, is one that is subsequently confirmed. But whether or not a confirmed hypothesis makes a positive contribution depends on the nature of the hypothesis and goals of the research. It is possible that a researcher will learn as much, and possibly even more, from findings that lead to rejection of a hypothesis. In any event, findings, whatever they may be, are valuable only to the extent that the hypothesis being tested is itself worthy of study.

Two important considerations, albeit somewhat obvious ones, are that a hypothesis should be non-trivial and non-obvious. If a proposition is trivial, suggesting a variable relationship with little or no significance, discovering whether and how the variables it brings together are related will not make a meaningful contribution to knowledge about the determinants and/or impact of the variance at the heart of the researcher’s concern. Few will be interested in findings, however rigorously derived, about a trivial proposition. The same is true of an obvious hypothesis, obvious being an attribute that makes a proposition trivial. As stated, these considerations are themselves somewhat obvious, barely deserving mention. Nevertheless, an investigator should self-consciously reflect on these criteria when formulating hypotheses. She should be sure that she is proposing variable relationships that are neither trivial nor obvious.

A third criterion, also somewhat obvious but nonetheless essential, has to do with the significance and salience of the variables being considered. Will findings from research about these variables be important and valuable, and perhaps also useful? If the primary variable of interest is a dependent variable, meaning that the primary goal of the research is to account for variance, then the significance and salience of the dependent variable will determine the value of the research. Similarly, if the primary variable of interest is an independent variable, meaning that the primary goal of the research is to determine and assess impact, then the significance and salience of the independent variable will determine the value of the research.

These three criteria—non-trivial, non-obvious, and variable importance and salience—are not very different from one another. They collectively mean that the researcher must be able to specify why and how the testing of her hypothesis, or hypotheses, will make a contribution of value. Perhaps her propositions are original or innovative; perhaps knowing whether they are true or false makes a difference or will be of practical benefit; perhaps her findings add something specific and identifiable to the body of existing scholarly literature on the subject. While calling attention to these three connected and overlapping criteria might seem unnecessary since they are indeed somewhat obvious, it remains the case that the value of a hypothesis, regardless of whether or not it is eventually confirmed, is itself important to consider, and an investigator should, therefore, know and be able to articulate the reasons and ways that consideration of her hypothesis, or hypotheses, will indeed be of value.

Hypothesizing the Structure of a Relationship

Relevant in the process of hypothesis formation are, as discussed, questions about the origins of hypotheses and the criteria by which the value of any particular hypothesis or set of hypotheses will be evaluated. Relevant, too, is consideration of the structure of a hypothesized variable relationship and the causal story to which that relationship is believed to call attention.

The point of departure in considering the structure of a hypothesized variable relationship is an understanding that such a relationship may or may not be linear. In a direct, or positive, linear relationship, each increase in the independent variable brings a constant increase in the dependent variable. In an inverse, or negative, linear relationship, each increase in the independent variable brings a constant decrease in the dependent variable. But these are only two of the many ways that an independent variable and a dependent variable may be related, or hypothesized to be related. This is easily illustrated by hypotheses in which level of education or age is the independent variable, and this is relevant in hypothesis formation because the investigator must be alert to and consider the possibility that the variables in which she is interested are in fact related in a non-linear way.

Consider, for example, the relationship between age and support for gender equality, the latter measured by an index based on several questions about the rights and behavior of women that are asked in Arab Barometer surveys. A researcher might expect, and might therefore want to hypothesize, that an increase in age brings increased support for, or alternatively increased opposition to, gender equality. But these are not the only possibilities. Likely, perhaps, is the possibility of a curvilinear relationship, in which case increases in age bring increases in support for gender equality until a person reaches a certain age, maybe 40, 45, or 50, after which additional increases in age bring decreases in support for gender equality. Or the researcher might hypothesize that the curve is in the opposite direction, that support for gender equality initially decreases as a function of age until a particular age is reached, after which additional increases in age bring an increase in support.

Of course, there are also other possibilities. In the case of education and gender equality, for example, increased education may initially have no impact on attitudes toward gender equality. Individuals who have not finished primary school, those who have finished primary school, and those who have gone somewhat beyond primary school and completed a middle school program may all have roughly the same attitudes toward gender equality. Thus, increases in education, within a certain range of educational levels, are not expected to bring an increase or a decrease in support for gender equality. But the level of support for gender equality among high school graduates may be higher and among university graduates may be higher still. Accordingly, in this hypothetical illustration, an increase in education does bring increased support for gender equality but only beginning after middle school.

A middle school level of education is a “floor” in this example. Education does not begin to make a difference until this floor is reached, and thereafter it does make a difference, with increases in education beyond middle school bringing increases in support for gender equality. Another possibility might be for middle school to be a “ceiling.” This would mean that increases in education through middle school would bring increases in support for gender equality, but the trend would not continue beyond middle school. In other words, level of education makes a difference and appears to have explanatory power only until, and so not after, this ceiling is reached. This latter pattern was found in the study of education and Palestinian protest activity discussed earlier. Increases in education through middle school brought increases in the likelihood that an individual would participate in demonstrations and protests of Israeli occupation. However, additional education beyond middle school was not associated with greater likelihood of taking part in protest activities.

This discussion of variation in the structure of a hypothesized relationship between two variables is certainly not exhaustive, and the examples themselves are straightforward and not very complicated. The purpose of the discussion is, therefore, to emphasize that an investigator must be open to and think through the possibility and plausibility of different kinds of relationships between her two variables, that is to say, relationships with different structures. Bivariate relationships with several different kinds of structures are depicted visually by the scatter plots in Fig. 3.4 .

These possibilities with respect to structure do not determine the value of a proposed hypothesis. As discussed earlier, the value of a proposed relationship depends first and foremost on the importance and salience of the variable of interest. Accordingly, a researcher should not assume that the value of a hypothesis varies as a function of the degree to which it posits a complicated variable relationship. More complicated hypotheses are not necessarily better or more correct. But while she should not strive for or give preference to variable relationships that are more complicated simply because they are more complicated, she should, again, be alert to the possibility that a more complicated pattern does a better job of describing the causal connection between the two variables in the place and time in which she is interested.

This brings the discussion of formulating hypotheses back to our earlier account of causal stories. In research concerned with explanation and causality, a hypothesis for the most part is a simplified stand-in for a causal story. It represents the causal story, as it were. Expressing this differently, the hypothesis states the causal story’s “bottom line;” it posits that the independent variable is a determinant of variance on the dependent variable, and it identifies the structure of the presumed relationship between the independent variable and the dependent variable. But it does not describe the interaction between the two variables in a way that tells consumers of the study why the researcher believes that the relationship involves causality rather than an association with no causal implications. This is left to the causal story, which will offer a fuller account of the way the presumed cause impacts the presumed effect.

3.3 Describing and Visually Representing Bivariate Relationships

Once a researcher has collected or otherwise obtained data on the variables in a bivariate relationship she wishes to examine, her first step will be to describe the variance on each of the variables using the univariate statistics described in Chap. 2 . She will need to understand the distribution on each variable before she can understand how these variables vary in relation to one another. This is important whether she is interested in description or wishes to explore a bivariate causal story.

Once she has described each one of the variables, she can turn to the relationship between them. She can prepare and present a visual representation of this relationship, which is the subject of the present section. She can also use bivariate statistical tests to assess the strength and significance of the relationship, which is the subject of the next section of this chapter.

Contingency Tables

Contingency tables are used to display the relationship between two categorical variables. They are similar to the univariate frequency distributions described in Chap. 2 , the difference being that they juxtapose the two univariate distributions and display the interaction between them. Also called cross-tabulation tables, the cells of the table may present frequencies, row percentages, column percentages, and/or total percentages. Total frequencies and/or percentages are displayed in a total row and a total column, each one of which is the same as the univariate distribution of one of the variables taken alone.

Table 3.2 , based on Palestinian data from Wave V of the Arab Barometer, crosses gender and the average number of hours watching television each day. Frequencies are presented in the cells of the table. In the cell showing the number of Palestinian men who do not watch television at all, row percentage, column percentage, and total percentage are also presented. Note that total percentage is based on the 10 cells showing the two variables taken together, which are summed in the lower right-hand cell. Thus, total percent for this cell is 342/2488 = 13.7. Only frequencies are given in the other cells of the table; but in a full table, these four figures – frequency, row percent, column percent and total percent – would be given in every cell.

Exercise 3.3

Compute the row percentage, the column percentage, and the total percentage in the cell showing the number of Palestinian women who do not watch television at all.

Describe the relationship between gender and watching television among Palestinians that is shown in the table. Do the television watching habits of Palestinian men and women appear to be generally similar or fairly different? You might find it helpful to convert the frequencies in other cells to row or column percentages.

Stacked Column Charts and Grouped Bar Charts

Stacked column charts and grouped bar charts are used to visually describe how two categorical variables, or one categorical and one continuous variable, relate to one another. Much like contingency tables, they show the percentage or count of each category of one variable within each category of the second variable. This information is presented in columns stacked on each other or next to each other. The charts below show the number of male Palestinians and the number of female Palestinians who watch television for a given number of hours each day. Each chart presents the same information as the other chart and as the contingency table shown above (Fig. 3.1 ).

Stacked column charts and grouped bar charts comparing Palestinian men and Palestinian women on hours watching television

Box Plots and Box and Whisker Plots

Box plots, box and whisker plots, and other types of plots can also be used to show the relationship between one categorical variable and one continuous variable. They are particularly useful for showing how spread out the data are. Box plots show five important numbers in a variable’s distribution: the minimum value; the median; the maximum value; and the first and third quartiles (Q1 and Q2), which represent, respectively, the number below which are 25 percent of the distribution’s values and the number below which are 75 percent of the distribution’s values. The minimum value is sometimes called the lower extreme, the lower bound, or the lower hinge. The maximum value is sometimes called the upper extreme, the upper bound, or the upper hinge. The middle 50 percent of the distribution, the range between Q1 and Q3 that represents the “box,” constitutes the interquartile range (IQR). In box and whisker plots, the “whiskers” are the short perpendicular lines extending outside the upper and lower quartiles. They are included to indicate variability below Q1 and above Q3. Values are usually categorized as outliers if they are less than Q1 − IQR*1.5 or greater than Q3 + IQR*1.5. A visual explanation of a box and whisker plot is shown in Fig. 3.2a and an example of a box plot that uses actual data is shown in Fig. 3.2b .

The box plot in Fig. 3.2b uses Wave V Arab Barometer data from Tunisia and shows the relationship between age, a continuous variable, and interpersonal trust, a dichotomous categorical variable. The line representing the median value is shown in bold. Interpersonal trust, sometimes known as generalized trust, is an important personal value. Previous research has shown that social harmony and prospects for democracy are greater in societies in which most citizens believe that their fellow citizens for the most part are trustworthy. Although the interpersonal trust variable is dichotomous in Fig. 3.2b , the variance in interpersonal trust can also be measured by a set of ordered categories or a scale that yields a continuous measure, the latter not being suitable for presentation by a box plot. Figure 3.2b shows that the median age of Tunisians who are trusting is slightly higher than the median age of Tunisians who are mistrustful of other people. Notice also that the box plot for the mistrustful group has an outlier.

( a ) A box and whisker plot. ( b ) Box plot comparing the ages of trusting and mistrustful Tunisians in 2018

Line plots may be used to visualize the relationship between two continuous variables or a continuous variable and a categorical variable. They are often used when time, or a variable related to time, is one of the two variables. If a researcher wants to show whether and how a variable changes over time for more than one subgroup of the units about which she has data (looking at men and women separately, for example), she can include multiple lines on the same plot, with each line showing the pattern over time for a different subgroup. These lines will generally be distinguished from each other by color or pattern, with a legend provided for readers.

Line plots are a particularly good way to visualize a relationship if an investigator thinks that important events over time may have had a significant impact. The line plot in Fig. 3.3 shows the average support for gender equality among men and among women in Tunisia from 2013 to 2018. Support for gender equality is a scale based on four questions related to gender equality in the three waves of the Arab Barometer. An answer supportive of gender equality on a question adds +.5 to the scale and an answer unfavorable to gender equality adds −.5 to the scale. Accordingly, a scale score of 2 indicates maximum support for gender equality and a scale score of −2 indicates maximum opposition to gender equality.

Line plot showing level of support for gender equality among Tunisian women and men in 2013, 2016, and 2018

Scatter Plots

Scatter plots are used to visualize a bivariate relationship when both variables are numerical. The independent variable is put on the x-axis, the horizontal axis, and the dependent variable is put on the y-axis, the vertical axis. Each data point becomes a dot in the scatter plot’s two-dimensional field, with its precise location being the point at which its value on the x-axis intersects with its value on the y-axis. The scatter plot shows how the variables are related to one another, including with respect to linearity, direction, and other aspects of structure. The scatter plots in Fig. 3.4 illustrate a strong positive linear relationship, a moderately strong negative linear relationship, a strong non-linear relationship, and a pattern showing no relationship. Footnote 5 If the scatter plot displays no visible and clear pattern, as in the lower left hand plot shown in Fig. 3.4 , the scatter plot would indicate that the independent variable, by itself, has no meaningful impact on the dependent variable.

Scatter plots showing bivariate relationships with different structures

Scatter plots are also a good way to identify outliers—data points that do not follow a pattern that characterizes most of the data. These are also called non-scalar types. Figure 3.5 shows a scatter plot with outliers.

Outliers can be informative, making it possible, for example, to identify the attributes of cases for which the measures of one or both variables are unreliable and/or invalid. Nevertheless, the inclusion of outliers may not only distort the assessment of measures, raising unwarranted doubts about measures that are actually reliable and valid for the vast majority of cases, they may also bias bivariate statistics and make relationships seem weaker than they really are for most cases. For this reason, researchers sometimes remove outliers prior to testing a hypothesis. If one does this, it is important to have a clear definition of what is an outlier and to justify the removal of the outlier, both using the definition and perhaps through substantive analysis. There are several mathematical formulas for identifying outliers, and researchers should be aware of these formulas and their pros and cons if they plan to remove outliers.

If there are relatively few outliers, perhaps no more than 5–10 percent of the cases, it may be justifiable to remove them in order to better discern the relationship between the independent variable and the dependent variable. If outliers are much more numerous, however, it is probably because there is not a significant relationship between the two variables being considered. The researcher might in this case find it instructive to introduce a third variable and disaggregate the data. Disaggregation will be discussed in Chap. 4 .

A scatter plot with outliers marked in red

Exercise 3.4 Exploring Hypotheses through Visualizing Data: Exercise with the Arab Barometer Online Analysis Tool

Go to the Arab Barometer Online Analysis Tool ( https://www.arabbarometer.org/survey-data/data-analysis-tool/ )

Select Wave V and a country that interests you

Select “See Results”

Select “Social, Cultural and Religious topics”

Select “Religion: frequency: pray”

Questions: What does the distribution of this variable look like? How would you describe the variance?

Click on “Cross by,” then

Select “Show all variables”

Select “Kind of government preferable” and click

Select “Options,” then “Show % over Row total,” then “Apply”

Questions: Does there seem to be a relationship between religiosity and preference for democracy? If so, what might explain the relationship you observe—what is a plausible causal story? Is it consistent with the hypothesis you wrote for Exercise 3.1?

What other variables could be used to measure religiosity and preference for democracy? Explore your hypothesis using different items from the list of Arab Barometer variables

Do these distributions support the previous results you found? Do you learn anything additional about the relationship between religiosity and preference for democracy?

Now it is your turn to explore variables and variable relationships that interest you!

Pick two variables that interest you from the list of Arab Barometer variables. Are they continuous or categorical? Ordinal or nominal? (Hint: Most Arab Barometer variables are categorical, even if you might be tempted to think of them as continuous. For example, age is divided into the ordinal categories 18–29, 30–49, and 50 and more.)

Do you expect there to be a relationship between the two variables? If so, what do you think will be the structure of that relationship, and why?

Select the wave (year) and the country that interest you

Select one of your two variables of interest

Click on “Cross by,” and then select your second variable of interest.

On the left side of the page, you’ll see a contingency table. On the right side at the top, you’ll see several options to graphically display the relationship between your two variables. Which type of graph best represents the relationship between your two variables of interest?

Do the two variables seem to be independent of each other, or do you think there might be a relationship between them? Is the relationship you see similar to what you had expected

3.4 Probabilities and Type I and Type II Errors

As in visual presentations of bivariate relationships, selecting the appropriate measure of association or bivariate statistical test depends on the types of the two variables. The data on both variables may be categorical; the data on both may be continuous; or the data may be categorical on one variable and continuous on the other variable. These characteristics of the data will guide the way in which our presentation of these measures and tests is organized. Before briefly describing some specific measures of association and bivariate statistical tests, however, it is necessary to lay a foundation by introducing a number of terms and concepts. Relevant here are the distinction between population and sample and the notions of the null hypothesis, of Type I and Type II errors, and of probabilities and confidence intervals. As concepts, or abstractions, these notions may influence the way a researcher thinks about drawing conclusions about a hypothesis from qualitative data, as was discussed in Chap. 2 . In their precise meaning and application, however, these terms and concepts come into play when hypothesis testing involves the statistical analysis of quantitative data.

To begin, it is important to distinguish between, on the one hand, the population of units—individuals, countries, ethnic groups, political movements, or any other unit of analysis—in which the researcher is interested and about which she aspires to advance conclusions and, on the other hand, the units on which she has actually acquired the data to be analyzed. The latter, the units on which she actually has data, is her sample. In cases where the researcher has collected or obtained data on all of the units in which she is interested, there is no difference between the sample and the population, and drawing conclusions about the population based on the sample is straightforward. Most often, however, a researcher does not possess data on all of the units that make up the population in which she is interested, and so the possibility of error when making inferences about the population based on the analysis of data in the sample requires careful and deliberate consideration.

This concern for error is present regardless of the size of the sample and the way it was constructed. The likelihood of error declines as the size of the sample increases and thus comes closer to representing the full population. It also declines if the sample was constructed in accordance with random or other sampling procedures designed to maximize representation. It is useful to keep these criteria in mind when looking at, and perhaps downloading and using, Arab Barometer data. The Barometer’s website gives information about the construction of each sample. But while it is possible to reduce the likelihood of error when characterizing the population from findings based on the sample, it is not possible to eliminate entirely the possibility of erroneous inference. Accordingly, a researcher must endeavor to make the likelihood of this kind of error as small as possible and then decide if it is small enough to advance conclusions that apply to the population as well as the sample.

The null hypothesis, frequently designated as H0, is a statement to the effect that there is no meaningful and significant relationship between the independent variable and the dependent variable in a hypothesis, or indeed between two variables even if the relationship between them has not been formally specified in a hypothesis and does not purport to be causal or explanatory. The null hypothesis may or may not be stated explicitly by an investigator, but it is nonetheless present in her thinking; it stands in opposition to the hypothesized variable relationship. In a point and counterpoint fashion, the hypothesis, H1, posits that the variables are significantly related, and the null hypothesis, H0, replies and says no, they are not significantly related. It further says that they are not related in any meaningful way, neither in the way proposed in H1 nor in any other way that could be proposed.

Based on her analysis, the researcher needs to determine whether her findings permit rejecting the null hypothesis and concluding that there is indeed a significant relationship between the variables in her hypothesis, concluding in effect that the research hypothesis, H1, has been confirmed. This is most relevant and important when the investigator is basing her analysis on some but not all of the units to which her hypothesis purports to apply—when she is analyzing the data in her sample but seeks to advance conclusions that apply to the population in which she is interested. The logic here is that the findings produced by an analysis of some of the data, the data she actually possesses, may be different than the findings her analysis would hypothetically produce were she able to use data from very many more, or ideally even all, of the units that make up her population of interest.

This means, of course, that there will be uncertainty as the researcher adjudicates between H0 and H1 on the basis of her data. An analysis of these data may suggest that there is a strong and significant relationship between the variables in H1. And the stronger the relationship, the more unlikely it is that the researcher’s sample is a subset of a population characterized by H0 and that, therefore, the researcher may consider H1 to have been confirmed. Yet, it remains at least possible that the researcher’s sample, although it provides strong support for H1, is actually a subset of a population characterized by the null hypothesis. This may be unlikely, but it is not impossible, and so, therefore, to consider H1 to have been confirmed is to run the risk, at least a small risk, of what is known as a Type I error. A Type I error is made when a researcher accepts a research hypothesis that is actually false, when she judges to be true a hypothesis that does not characterize the population of which her sample is a subset. Because of the possibility of a Type I error, even if quite unlikely, researchers will often write something like “We can reject the null hypothesis,” rather than “We can confirm our hypothesis.”

Another analysis related to voter turnout provides a ready illustration. In the Arab Barometer Wave V surveys in 12 Arab countries, Footnote 6 13,899 respondents answered a question about voting in the most recent parliamentary election. Of these, 46.6 percent said they had voted, and the remainder, 53.4 percent, said they had not voted in the last parliamentary election. Footnote 7 Seeking to identify some of the determinants of voting—the attitudes and experiences of an individual that increase the likelihood that she will vote, the researcher might hypothesize that a judgment that the country is going in the right direction will push toward voting. More formally:

H1. An individual who believes that her country is going in the right direction is more likely to vote in a national election than is an individual who believes her country is going in the wrong direction.

Arab Barometer surveys provide data with which to test this proposition, and in fact there is a difference associated with views about the direction in which the country is going. Among those who judged that their country is going in the right direction, 52.4 percent voted in the last parliamentary election. By contrast, among those who judged that their country is going in the wrong direction, only 43.8 percent voted in the last parliamentary election.

This illustrates the choice a researcher faces when deciding what to conclude from a study. Does the analysis of her data from a subset of her population of interest confirm or not confirm her hypothesis? In this example, based on Arab Barometer data, the findings are in the direction of her hypothesis, and differences in voting associated with views about the direction the country is going do not appear to be trivial. But are these differences big enough to justify the conclusion that judgements about the country’s path going forward are a determinant of voting, one among others of course, in the population from which her sample was drawn? In other words, although this relationship clearly characterizes the sample, it is unclear whether it characterizes the researcher’s population of interest, the population from which the sample was drawn.

Unless the researcher can gather data on the entire population of eligible voters, or at least almost all of this population, it is not possible to entirely eliminate uncertainty when the researcher makes inferences about the population of voters based on findings from the subset, or sample, of voters on which she has data. She can either conclude that her findings are sufficiently strong and clear to propose that the pattern she has observed characterizes the population as well, and that H1 is therefore confirmed; or she can conclude that her findings are not strong enough to make such an inference about the population, and that H1, therefore, is not confirmed. Either conclusion could be wrong, and so there is a chance of error no matter which conclusion the researcher advances.

The terms Type I error and Type II error are often used to designate the possible error associated with each of these inferences about the population based on the sample. Type I error refers to the rejection of a true null hypothesis. This means, in other words, that the investigator could be wrong if she concludes that her finding of a strong, or at least fairly strong, relationship between her variables characterizes Arab voters in the 12 countries in general, and if she thus judges H1 to have been confirmed when the population from which her sample was drawn is in fact characterized by H0. Type II error refers to acceptance of a false null hypothesis. This means, in other words, that the investigator could be wrong if she concludes that her finding of a somewhat weak relationship, or no relationship at all, between her variables characterizes Arab voters in the 12 countries in general, and that she thus judges H0 to be true when the population from which her sample was drawn is in fact characterized by H1.

In statistical analyses of quantitative data, decisions about whether to risk a Type I error or a Type II error are usually based on probabilities. More specifically, they are based on the probability of a researcher being wrong if she concludes that the variable relationship—or hypothesis in most cases—that characterizes her data, meaning her sample, also characterizes the population on which the researcher hopes her sample and data will shed light. To say this in yet another way, she computes the odds that her sample does not represent the population of which it is a subset; or more specifically still, she computes the odds that from a population that is characterized by the null hypothesis she could have obtained, by chance alone, a subset of the population, her sample, that is not characterized by the null hypothesis. The lower the odds, or probability, the more willing the researcher will be to risk a Type I error.

There are numerous statistical tests that are used to compute such probabilities. The nature of the data and the goals of the analysis will determine the specific test to be used in a particular situation. Most of these tests, frequently called tests of significance or tests of statistical significance, provide output in the form of probabilities, which always range from 0 to 1. The lower the value, meaning the closer to 0, the less likely it is that a researcher has collected and is working with data that produce findings that differ from what she would find were she to somehow have data on the entire population. Another way to think about this is the following:

If the researcher provisionally assumes that the population is characterized by the null hypothesis with respect to the variable relationship under study, what is the probability of obtaining from that population, by chance alone, a subset or sample that is not characterized by the null hypothesis but instead shows a strong relationship between the two variables;

The lower the probability value, meaning the closer to 0, the less likely it is that the researcher’s data, which support H1, have come from a population that is characterized by H0;

The lower the probability that her sample could have come from a population characterized by H0, the lower the possibility that the researcher will be wrong, that she will make a Type I error, if she rejects the null hypothesis and accepts that the population, as well as her sample, is characterized by H1;

When the probability value is low, the chance of actually making a Type I error is small. But while small, the possibility of an error cannot be entirely eliminated.

If it helps you to think about probability and Type I and Type II error, imagine that you will be flipping a coin 100 times and your goal is to determine whether the coin is unbiased, H0, or biased in favor of either heads or tails, H1. How many times more than 50 would heads have to come up before you would be comfortable concluding that the coin is in fact biased in favor of heads? Would 60 be enough? What about 65? To begin to answer these questions, you would want to know the odds of getting 60 or 65 heads from a coin that is actually unbiased, a coin that would come up heads and come up tails roughly the same number of times if it were flipped many more than 100 times, maybe 1000 times, maybe 10,000. With this many flips, would the ratio of heads to tails even out. The lower the odds, the less likely it is that the coin is unbiased. In this analogy, you can think of the mathematical calculations about an unbiased coin’s odds of getting heads as the population, and your actual flips of the coin as the sample.

But exactly how low does the probability of a Type I error have to be for a researcher to run the risk of rejecting H0 and accepting that her variables are indeed related? This depends, of course, on the implications of being wrong. If there are serious and harmful consequences of being wrong, of accepting a research hypothesis that is actually false, the researcher will reject H0 and accept H1 only if the odds of being wrong, of making a Type I error, are very low.

There are some widely used probability values, which define what are known as “confidence intervals,” that help researchers and those who read their reports to think about the likelihood that a Type I error is being made. In the social sciences, rejecting H0 and running the risk of a Type I error is usually thought to require a probability value of less than .05, written as p < .05. The less stringent value of p < .10 is sometimes accepted as sufficient for rejecting H0, although such a conclusion would be advanced with caution and when the consequences of a Type I error are not very harmful. Frequently considered safer, meaning that the likelihood of accepting a false hypothesis is lower, are p < .01 and p < .001. The next section introduces and briefly describes some of the bivariate statistics that may be used to calculate these probabilities.

3.5 Measures of Association and Bivariate Statistical Tests

The following section introduces some of the bivariate statistical tests that can be used to compute probabilities and test hypotheses. The accounts are not very detailed. They will provide only a general overview and refresher for readers who are already fairly familiar with bivariate statistics. Readers without this familiarity are encouraged to consult a statistics textbook, for which the accounts presented here will provide a useful guide. While the account below will emphasize calculating these test statistics by hand, it is also important to remember that they can be calculated with the assistance of statistical software as well. A discussion of statistical software is available in Appendix 4.

Parametric and Nonparametric Statistics

Parametric and nonparametric are two broad classifications of statistical procedures. A parameter in statistics refers to an attribute of a population. For example, the mean of a population is a parameter. Parametric statistical tests make certain assumptions about the shape of the distribution of values in a population from which a sample is drawn, generally that it is normally distributed, and about its parameters, that is to say the means and standard deviations of the assumed distributions. Nonparametric statistical procedures rely on no or very few assumptions about the shape or parameters of the distribution of the population from which the sample was drawn. Chi-squared is the only nonparametric statistical test among the tests described below.

Degrees of Freedom

Degrees of freedom (df) is the number of values in the calculation of a statistic that are free to vary. Statistical software programs usually give degrees of freedom in the output, so it is generally unnecessary to know the number of the degrees of freedom in advance. It is nonetheless useful to understand what degrees of freedom represent. Consistent with the definition above, it is the number of values that are not predetermined, and thus are free to vary, within the variables used in a statistical test.

This is illustrated by the contingency tables below, which are constructed to examine the relationship between two categorical variables. The marginal row and column totals are known since these are just the univariate distributions of each variable. df = 1 for Table 3.3a , which is a 4-cell table. You can enter any one value in any one cell, but thereafter the values of all the other three cells are determined. Only one number is not free to vary and thus not predetermined. df = 2 for Table 3.3b , which is a 6-cell table. You can enter any two values in any two cells, but thereafter the values of all the other cells are determined. Only two numbers are free to vary and thus not predetermined. For contingency tables, the formula for calculating df is:

Chi-Squared

Chi-squared, frequently written X 2 , is a statistical test used to determine whether two categorical variables are significantly related. As noted, it is a nonparametric test. The most common version of the chi-squared test is the Pearson chi-squared test, which gives a value for the chi-squared statistic and permits determining as well a probability value, or p-value. The magnitude of the statistic and of the probability value are inversely correlated; the higher the value of the chi-squared statistic, the lower the probability value, and thus the lower the risk of making a Type I error—of rejecting a true null hypothesis—when asserting that the two variables are strongly and significantly related.

The simplicity of the chi-squared statistic permits giving a little more detail in order to illustrate several points that apply to bivariate statistical tests in general. The formula for computing chi-squared is given below, with O being the observed (actual) frequency in each cell of a contingency table for two categorical variables and E being the frequency that would be expected in each cell if the two variables are not related. Put differently, the distribution of E values across the cells of the two-variable table constitutes the null hypothesis, and chi-squared provides a number that expresses the magnitude of the difference between an investigator’s actual observed values and the values of E.

The computation of chi-squared involves the following procedures, which are illustrated using the data in Table 3.4 .

The values of O in the cells of the table are based on the data collected by the investigator. For example, Table 3.4 shows that of the 200 women on whom she collected information, 85 are majoring in social science.

The value of E for each cell is computed by multiplying the marginal total of the column in which the cell is located by the marginal total of the row in which the cell is located divided by N, N being the total number of cases. For the female students majoring in social science in Table 3.4 , this is: 200 * 150/400 = 30,000/400 = 75. For the female students majoring in math and natural science in Table 3.4 , this is: 200 * 100/400 = 20,000/400 = 50.

The difference between the value of O and the value of E is computed for each cell using the formula for chi-squared. For the female students majoring in social science in Table 3.4 , this is: (85–75) 2 /75 = 10 2 /75 = 100/75 = 1.33. For the female students majoring in math and natural science, the value resulting from the application of the chi-squared is: (45–50) 2 /50 = 5 2 /75 = 25/75 = .33.

The values in each cell of the table resulting from the application of the chi-squared formula are summed (Σ). This chi-squared value expresses the magnitude of the difference between a distribution of values indicative of the null hypothesis and what the investigator actually found about the relationship between gender and field of study. In Table 3.4 , the cell for female students majoring in social science adds 1.33 to the sum of the values in the eight cells, the cell for female students majoring in math and natural science adds .33 to the sum, and so forth for the remaining six cells.

A final point to be noted, which applies to many other statistical tests as well, is that the application of chi-squared and other bivariate (and multivariate) statistical tests yields a value with which can be computed the probability that an observed pattern does not differ from the null hypothesis and that a Type I error will be made if the null hypothesis is rejected and the research hypothesis is judged to be true. The lower the probability, of course, the lower the likelihood of an error if the null hypothesis is rejected.

Prior to the advent of computer assisted statistical analysis, the value of the statistic and the number of degrees of freedom were used to find the probability value in a table of probability values in an appendix in most statistics books. At present, however, the probability value, or p-value, and also the degrees of freedom, are routinely given as part of the output when analysis is done by one of the available statistical software packages.

Table 3.5 shows the relationship between economic circumstance and trust in the government among 400 ordinary citizens in a hypothetical country. The observed data were collected to test the hypothesis that greater wealth pushes people toward greater trust and less wealth pushes people toward lesser trust. In the case of all three patterns, the probability that the null hypothesis is true is very low. All three patterns have the same high chi-squared value and low probability value. Thus, the chi-squared and p-values show only that the patterns all differ significantly from what would be expected were the null hypothesis true. They do not show whether the data support the hypothesized variable relationship or any other particular relationship.

As the three patterns in Table 3.5 show, variable relationships with very different structures can yield similar or even identical statistical test and probability values, and thus these tests provide only some of the information a researcher needs to draw conclusions about her hypothesis. To draw the right conclusion, it may also be necessary for the investigator to “look at” her data. For example, as Table 3.5 suggests, looking at a tabular or visual presentation of the data may also be needed to draw the proper conclusion about how two variables are related.

How would you describe the three patterns shown in the table, each of which differs significantly from the null hypothesis? Which pattern is consistent with the research hypothesis? How would you describe the other two patterns? Try to visualize a plot of each pattern.

Pearson Correlation Coefficient

The Pearson correlation coefficient, more formally known as the Pearson product-moment correlation, is a parametric measure of linear association. It gives a numerical representation of the strength and direction of the relationship between two continuous numerical variables. The coefficient, which is commonly represented as r , will have a value between −1 and 1. A value of 1 means that there is a perfect positive, or direct, linear relationship between the two variables; as one variable increases, the other variable consistently increases by some amount. A value of −1 means that there is a perfect negative, or inverse, linear relationship; as one variable increases, the other variable consistently decreases by some amount. A value of 0 means that there is no linear relationship; as one variable increases, the other variable neither consistently increases nor consistently decreases.

It is easy to think of relationships that might be assessed by a Pearson correlation coefficient. Consider, for example, the relationship between age and income and the proposition that as age increases, income consistently increases or consistently decreases as well. The closer a coefficient is to 1 or −1, the greater the likelihood that the data on which it is based are not the subset of a population in which age and income are unrelated, meaning that the population of interest is not characterized by the null hypothesis. Coefficients very close to 1 or −1 are rare; although it depends on the number of units on which the researcher has data and also on the nature of the variables. Coefficients higher than .3 or lower than −.03 are frequently high enough, in absolute terms, to yield a low probability value and justify rejecting the null hypothesis. The relationship in this case would be described as “statistically significant.”

Exercise 3.5

Estimating Correlation Coefficients from scatter plots

Look at the scatter plots in Fig. 3.4 and estimate the correlation coefficient that the bivariate relationship shown in each scatter plot would yield.

Explain the basis for each of your estimates of the correlation coefficient.

Spearman’s Rank-Order Correlation Coefficient

The Spearman’s rank-order correlation coefficient is a nonparametric version of the Pearson product-moment correlation . Spearman’s correlation coefficient, (ρ, also signified by r s ) measures the strength and direction of the association between two ranked variables.

Bivariate Regression

Bivariate regression is a parametric measure of association that, like correlation analysis, assesses the strength and direction of the relationship between two variables. Also, like correlation analysis, regression assumes linearity. It may give misleading results if used with variable relationships that are not linear.

Regression is a powerful statistic that is widely used in multivariate analyses. This includes ordinary least squares (OLS) regression, which requires that the dependent variable be continuous and assumes linearity; binary logistic regression, which may be used when the dependent variable is dichotomous; and ordinal logistic regression, which is used with ordinal dependent variables. The use of regression in multivariate analysis will be discussed in the next chapter. In bivariate analysis, regression analysis yields coefficients that indicate the strength and direction of the relationship between two variables. Researchers may opt to “standardize” these coefficients. Standardized coefficients from a bivariate regression are the same as the coefficients produced by Pearson product-moment correlation analysis.

The t-test, also sometimes called a “difference of means” test, is a parametric statistical test that compares the means of two variables and determines whether they are different enough from each other to reject the null hypothesis and risk a Type I error. The dependent variable in a t-test must be continuous or ordinal—otherwise the investigator cannot calculate a mean. The independent variable must be categorical since t-tests are used to compare two groups.

An example, drawing again on Arab Barometer data, tests the relationship between voting and support for democracy. The hypothesis might be that men and women who voted in the last parliamentary election are more likely than men and women who did not vote to believe that democracy is suitable for their country. Whether a person did or did not vote would be the categorical independent variable, and the dependent variable would be the response to a question like, “To what extent do you think democracy is suitable for your country?” The question about democracy asked respondents to situate their views on a 11-point scale, with 0 indicating completely unsuitable and 10 indicating completely suitable.

Focusing on Tunisia in 2018, Arab Barometer Wave V data show that the mean response on the 11-point suitability question is 5.11 for those who voted and 4.77 for those who did not vote. Is this difference of .34 large enough to be statistically significant? A t-test will determine the probability of getting a difference of this magnitude from a population of interest, most likely all Tunisians of voting age, in which there is no difference between voters and non-voters in views about the suitability of democracy for Tunisia. In this example, the t-test showed p < .086. With this p-value, which is higher than the generally accepted standard of .05, a researcher cannot with confidence reject the null hypotheses, and she is unable, therefore, to assert that the proposed relationship has been confirmed.

This question can also be explored at the country level of analysis with, for example, regime type as the independent variable. In this illustration, the hypothesis is that citizens of monarchies are more likely than citizens of republics to believe that democracy is suitable for their country. Of course, a researcher proposing this hypothesis would also advance an associated causal story that provides the rationale for the hypothesis and specifies what is really being tested. To test this proposition, an investigator might merge data from surveys in, say, three monarchies, perhaps Morocco, Jordan, and Kuwait, and then also merge data from surveys in three republics, perhaps Algeria, Egypt, and Iraq. A t-test would then be used to compare the means of people in republics and people in monarchies and give the p-value.

A similar test, the Wilcoxon-Mann-Whitney test, is a nonparametric test that does not require that the dependent variable be normally distributed.

Analysis of variance, or ANOVA, is closely related to the t-test. It may be used when the dependent variable is continuous and the independent variable is categorical. A one-way ANOVA compares the mean and variance values of a continuous dependent variable in two or more categories of a categorical independent variable in order to determine if the latter affects the former.

ANOVA calculates the F-ratio based on the variance between the groups and the variance within each group. The F-ratio can then be used to calculate a p-value. However, if there are more than two categories of the independent variable, the ANOVA test will not indicate which pairs of categories differ enough to be statistically significant, making it necessary, again, to look at the data in order to draw correct conclusions about the structure of the bivariate relationships. Two-way ANOVA is used when an investigator has more than two variables.

Table 3.6 presents a summary list of the visual representations and bivariate statistical tests that have been discussed. It reminds readers of the procedures that can be used when both variables are categorical, when both variables are numerical/continuous, and when one variable is categorical and one variable is numerical/continuous.

Bivariate Statistics and Causal Inference

It is important to remember that bivariate statistical tests only assess the association or correlation between two variables. The tests described above can help a researcher estimate how much confidence her hypothesis deserves and, more specifically, the probability that any significant variable relationships she has found characterize the larger population from which her data were drawn and about which she seeks to offer information and insight.

The finding that two variables in a hypothesized relationship are related to a statistically significant degree is not evidence that the relationship is causal, only that the independent variable is related to the dependent variable. The finding is consistent with the causal story that the hypothesis represents, and to that extent, it offers support for this story. Nevertheless, there are many reasons why an observed statistically significant relationship might be spurious. The correlation might, for example, reflect the influence of one or more other and uncontrolled variables. This will be discussed more fully in the next chapter. The point here is simply that bivariate statistics do not, by themselves, address the question of whether a statistically significant relationship between two variables is or is not a causal relationship.

Only an Introductory Overview

As has been emphasized throughout, this chapter seeks only to offer an introductory overview of the bivariate statistical tests that may be employed when an investigator seeks to assess the relationship between two variables. Additional information will be presented in Chap. 4 . The focus in Chap. 4 will be on multivariate analysis, on analyses involving three or more variables. In this case again, however, the chapter will provide only an introductory overview. The overviews in the present chapter and the next provide a foundation for understanding social statistics, for understanding what statistical analyses involve and what they seek to accomplish. This is important and valuable in and of itself. Nevertheless, researchers and would-be researchers who intend to incorporate statistical analyses into their investigations, perhaps to test hypotheses and decide whether to risk a Type I error or a Type II error, will need to build on this foundation and become familiar with the contents of texts on social statistics. If this guide offers a bird’s eye view, researchers who implement these techniques will also need to expose themselves to the view of the worm at least once.

Chapter 2 makes clear that the concept of variance is central and foundational for much and probably most data-based and quantitative social science research. Bivariate relationships, which are the focus of the present chapter, are building blocks that rest on this foundation. The goal of this kind of research is very often the discovery of causal relationships, relationships that explain rather than merely describe or predict. Such relationships are also frequently described as accounting for variance. This is the focus of Chap. 4 , and it means that there will be, first, a dependent variable, a variable that expresses and captures the variance to be explained, and then, second, an independent variable, and possibly more than one independent variable, that impacts the dependent variable and causes it to vary.

Bivariate relationships are at the center of this enterprise, establishing the empirical pathway leading from the variance discussed in Chap. 2 to the causality discussed in Chap. 4 . Finding that there is a significant relationship between two variables, a statistically significant relationship, is not sufficient to establish causality, to conclude with confidence that one of the variables impacts the other and causes it to vary. But such a finding is necessary.

The goal of social science inquiry that investigates the relationship between two variables is not always explanation. It might be simply to describe and map the way two variables interact with one another. And there is no reason to question the value of such research. But the goal of data-based social science research is very often explanation; and while the inter-relationships between more than two variables will almost always be needed to establish that a relationship is very likely to be causal, these inter-relationships can only be examined by empirics that begin with consideration of a bivariate relationship, a relationship with one variable that is a presumed cause and one variable that is a presumed effect.

Against this background, with the importance of two-variable relationships in mind, the present chapter offers a comprehensive overview of bivariate relationships, including but not only those that are hypothesized to be causally related. The chapter considers the origin and nature of hypotheses that posit a particular relationship between two variables, a causal relationship if the larger goal of the research is explanation and the delineation of a causal story to which the hypothesis calls attention. This chapter then considers how a bivariate relationship might be described and visually represented, and thereafter it discusses how to think about and determine whether the two variables actually are related.

Presenting tables and graphs to show how two variables are related and using bivariate statistics to assess the likelihood that an observed relationship differs significantly from the null hypothesis, the hypothesis of no relationship, will be sufficient if the goal of the research is to learn as much as possible about whether and how two variables are related. And there is plenty of excellent research that has this kind of description as its primary objective, that makes use for purposes of description of the concepts and procedures introduced in this chapter. But there is also plenty of research that seeks to explain, to account for variance, and for this research, use of these concepts and procedures is necessary but not sufficient. For this research, consideration of a two-variable relationship, the focus of the present chapter, is a necessary intermediate step on a pathway that leads from the observation of variance to explaining how and why that variance looks and behaves as it does.

Dana El Kurd. 2019. “Who Protests in Palestine? Mobilization Across Class Under the Palestinian Authority.” In Alaa Tartir and Timothy Seidel, eds. Palestine and Rule of Power: Local Dissent vs. International Governance . New York: Palgrave Macmillan.

Yael Zeira. 2019. The Revolution Within: State Institutions and Unarmed Resistance in Palestine . New York: Cambridge University Press.

Carolina de Miguel, Amaney A. Jamal, and Mark Tessler. 2015. “Elections in the Arab World: Why do citizens turn out?” Comparative Political Studies 48, (11): 1355–1388.

Question 1: Independent variable is religiosity; dependent variable is preference for democracy. Example of hypothesis for Question 1: H1. More religious individuals are more likely than less religious individuals to prefer democracy to other political systems. Question 2: Independent variable is preference for democracy; dependent variable is turning out to vote. Example of hypothesis for Question 2: H2. Individuals who prefer democracy to other political systems are more likely than individuals who do not prefer democracy to other political systems to turn out to vote.

Mike Yi. “A complete Guide to Scatter Plots,” posted October 16, 2019 and seen at https://chartio.com/learn/charts/what-is-a-scatter-plot/

The countries are Algeria, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Palestine, Sudan, Tunisia, and Yemen. The Wave V surveys were conducted in 2018–2019.

Not considered in this illustration are the substantial cross-country differences in voter turnout. For example, 63.6 of the Lebanese respondents reported voting, whereas in Algeria the proportion who reported voting was only 20.3 percent. In addition to testing hypotheses about voting in which the individual is the unit of analysis, country could also be the unit of analysis, and hypotheses seeking to account for country-level variance in voting could be formulated and tested.

Author information

Authors and affiliations.

Department of Political Science, University of Michigan, Ann Arbor, MI, USA

Mark Tessler

You can also search for this author in PubMed   Google Scholar

Rights and permissions

Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

Reprints and permissions

Copyright information

© 2023 The Author(s)

About this chapter

Tessler, M. (2023). Bivariate Analysis: Associations, Hypotheses, and Causal Stories. In: Social Science Research in the Arab World and Beyond. SpringerBriefs in Sociology. Springer, Cham. https://doi.org/10.1007/978-3-031-13838-6_3

Download citation

DOI : https://doi.org/10.1007/978-3-031-13838-6_3

Published : 04 October 2022

Publisher Name : Springer, Cham

Print ISBN : 978-3-031-13837-9

Online ISBN : 978-3-031-13838-6

eBook Packages : Social Sciences Social Sciences (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

4.2 Causality

Learning objectives.

• Define and provide an example of idiographic and nomothetic causal explanations
• Describe the role of causality in quantitative research as compared to qualitative research
• Identify, define, and describe each of the main criteria for nomothetic causal explanations
• Describe the difference between and provide examples of independent, dependent, and control variables
• Define hypothesis, be able to state a clear hypothesis, and discuss the respective roles of quantitative and qualitative research when it comes to hypotheses

Most social scientific studies attempt to provide some kind of causal explanation.  In other words, it is about cause and effect. A study on an intervention to prevent child abuse is trying to draw a connection between the intervention and changes in child abuse. Causality refers to the idea that one event, behavior, or belief will result in the occurrence of another, subsequent event, behavior, or belief.  It seems simple, but you may be surprised to learn there is more than one way to explain how one thing causes another. How can that be? How could there be many ways to understand causality?

Think back to our chapter on paradigms, which were analytic lenses comprised of assumptions about the world. You’ll remember the positivist paradigm as the one that believes in objectivity and social constructionist paradigm as the one that believes in subjectivity. Both paradigms are correct, though incomplete, viewpoints on the social world and social science.

A researcher operating in the social constructionist paradigm would view truth as subjective. In causality, that means that in order to try to understand what caused what, we would need to report what people tell us. Well, that seems pretty straightforward, right? Well, what if two different people saw the same event from the exact same viewpoint and came up with two totally different explanations about what caused what? A social constructionist might say that both people are correct. There is not one singular truth that is true for everyone, but many truths created and shared by people.

When social constructionists engage in science, they are trying to establish one type of causality—idiographic causality.  The word idiographic comes from the root word “idio” which means peculiar to one, personal, and distinct. An idiographic causal explanation means that you will attempt to explain or describe your phenomenon exhaustively, based on the subjective understandings of your participants. Idiographic causal explanations are intended to explain one particular context or phenomenon.  These explanations are bound with the narratives people create about their lives and experience, and are embedded in a cultural, historical, and environmental context. Idiographic causal explanations are so powerful because they convey a deep understanding of a phenomenon and its context. From a social constructionist perspective, the truth is messy. Idiographic research involves finding patterns and themes in the causal themes established by your research participants.

If that doesn’t sound like what you normally think of as “science,” you’re not alone. Although the ideas behind idiographic research are quite old in philosophy, they were only applied to the sciences at the start of the last century. If we think of famous Western scientists like Newton or Darwin, they never saw truth as subjective. They operated with the understanding there were objectively true laws of science that were applicable in all situations. In their time, another paradigm–the positivist paradigm–was dominant and continues its dominance today. When positivists try to establish causality, they are like Newton and Darwin, trying to come up with a broad, sweeping explanation that is universally true for all people. This is the hallmark of a nomothetic causal explanation .  The word nomothetic is derived from the root word “nomo” which means related to a law or legislative, and “thetic” which means something that establishes.  Put the root words together and it means something that is establishing a law, or in our case, a universal explanation.

Nomothetic causal explanations are incredibly powerful. They allow scientists to make predictions about what will happen in the future, with a certain margin of error. Moreover, they allow scientists to generalize —that is, make claims about a large population based on a smaller sample of people or items. Generalizing is important. We clearly do not have time to ask everyone their opinion on a topic, nor do we have the ability to look at every interaction in the social world. We need a type of causal explanation that helps us predict and estimate truth in all situations.

If these still seem like obscure philosophy terms, let’s consider an example. Imagine you are working for a community-based non-profit agency serving people with disabilities. You are putting together a report to help lobby the state government for additional funding for community support programs, and you need to support your argument for additional funding at your agency. If you looked at nomothetic research, you might learn how previous studies have shown that, in general, community-based programs like yours are linked with better health and employment outcomes for people with disabilities. Nomothetic research seeks to explain that community-based programs are better for everyone with disabilities. If you looked at idiographic research, you would get stories and experiences of people in community-based programs. These individual stories are full of detail about the lived experience of being in a community-based program. Using idiographic research, you can understand what it’s like to be a person with a disability and then communicate that to the state government. For example, a person might say “I feel at home when I’m at this agency because they treat me like a family member” or “this is the agency that helped me get my first paycheck.”

Neither kind of causal explanation is better than the other. A decision to conduct idiographic research means that you will attempt to explain or describe your phenomenon exhaustively, attending to cultural context and subjective interpretations. A decision to conduct nomothetic research, on the other hand, means that you will try to explain what is true for everyone and predict what will be true in the future. In short, idiographic explanations have greater depth, and nomothetic explanations have greater breadth. More importantly, social workers understand the value of both approaches to understanding the social world. A social worker helping a client with substance abuse issues seeks idiographic knowledge when they ask about that client’s life story, investigate their unique physical environment, or probe how they understand their addiction. At the same time, a social worker also uses nomothetic knowledge to guide their interventions. Nomothetic research may help guide them to minimize risk factors and maximize protective factors or use an evidence-based therapy, relying on knowledge about what in general helps people with substance abuse issues.

Nomothetic causal explanations

If you are trying to generalize about causality, or create a nomothetic causal explanation, then the rest of these statements are likely to be true: you will use quantitative methods, reason deductively, and engage in explanatory research. How can we make that prediction? Let’s take it part by part.

Because nomothetic causal explanations try to generalize, they must be able to reduce phenomena to a universal language, mathematics. Mathematics allows us to precisely measure, in universal terms, phenomena in the social world. Because explanatory researchers want a clean “x causes y” explanation, they need to use the universal language of mathematics to achieve their goal. That’s why nomothetic causal explanations use quantitative methods.  It’s helpful to note that not all quantitative studies are explanatory. For example, a descriptive study could reveal the number of people without homes in your county, though it won’t tell you why they are homeless. But nearly all explanatory studies are quantitative.

What we’ve been talking about here is an association between variables. When one variable precedes or predicts another, we have what researchers call independent and dependent variables. Two variables can be associated without having a causal relationship.  However, when certain conditions are met (which we describe later in this chapter), the independent variable is considered as a “ cause ” of the dependent variable.  For our example on spanking and aggressive behavior, spanking would be the independent variable and aggressive behavior addiction would be the dependent variable.  In causal explanations, the  independent variable is the cause, and the dependent variable is the effect.  Dependent variables depend on independent variables. If all of that gets confusing, just remember this graphical depiction:

The strength of the association between the independent variable and dependent variable is another important factor to take into consideration when attempting to make causal claims when your research approach is nomothetic.  In this context, strength refers to statistical significance . When the  association between two variables is shown to be statistically significant, we can have greater confidence that the data from our sample reflect a true association between those variables in the target population. Statistical significance is usually represented in statistics as the p- value .  Generally a p -value of .05 or less indicates the association between the two variables is statistically significant.

A hypothesis is a statement describing a researcher’s expectation regarding the research findings. Hypotheses in quantitative research are nomothetic causal explanations that the researcher expects to demonstrate. Hypotheses are written to describe the expected association between the independent and dependent variables. Your prediction should be taken from a theory or model of the social world. For example, you may hypothesize that treating clinical clients with warmth and positive regard is likely to help them achieve their therapeutic goals. That hypothesis would be using the humanistic theories of Carl Rogers. Using previous theories to generate hypotheses is an example of deductive research. If Rogers’ theory of unconditional positive regard is accurate, your hypothesis should be true.

Let’s consider a couple of examples. In research on sexual harassment (Uggen & Blackstone, 2004), one might hypothesize, based on feminist theories of sexual harassment, that more females than males will experience specific sexually harassing behaviors. What is the causal explanation being predicted here? Which is the independent and which is the dependent variable? In this case, we hypothesized that a person’s gender (independent variable) would predict their likelihood to experience sexual harassment (dependent variable).

Sometimes researchers will hypothesize that an association will take a specific direction. As a result, an increase or decrease in one area might be said to cause an increase or decrease in another. For example, you might choose to study the association between age and support for legalization of marijuana. Perhaps you’ve taken a sociology class and, based on the theories you’ve read, you hypothesize that age is negatively related to support for marijuana legalization. In fact, there are empirical data that support this hypothesis. Gallup has conducted research on this very question since the 1960s (Carroll, 2005). What have you just hypothesized? You have hypothesized that as people get older, the likelihood of their supporting marijuana legalization decreases. Thus, as age (your independent variable) moves in one direction (up), support for marijuana legalization (your dependent variable) moves in another direction (down). So, positive associations involve two variables going in the same direction and negative associations involve two variables going in opposite directions. If writing hypotheses feels tricky, it is sometimes helpful to draw them out and depict each of the two hypotheses we have just discussed.

It’s important to note that once a study starts, it is unethical to change your hypothesis to match the data that you found. For example, what happens if you conduct a study to test the hypothesis from Figure 4.3 on support for marijuana legalization, but you find no association between age and support for legalization? It means that your hypothesis was wrong, but that’s still valuable information. It would challenge what the existing literature says on your topic, demonstrating that more research needs to be done to figure out the factors that impact support for marijuana legalization. Don’t be embarrassed by negative results, and definitely don’t change your hypothesis to make it appear correct all along!

Establishing causality in nomothetic research

Let’s say you conduct your study and you find evidence that supports your hypothesis, as age increases, support for marijuana legalization decreases. Success! Causal explanation complete, right? Not quite. You’ve only established one of the criteria for causality. The main criteria for causality have to do with covariation, plausibility, temporality, and spuriousness. In our example from Figure 4.3, we have established only one criteria—covariation. When variables covary , they vary together. Both age and support for marijuana legalization vary in our study. Our sample contains people of varying ages and varying levels of support for marijuana legalization and they vary together in a patterned way–when age increases, support for legalization decreases.

Just because there might be some correlation between two variables does not mean that a causal explanation between the two is really plausible. Plausibility means that in order to make the claim that one event, behavior, or belief causes another, the claim has to make sense. It makes sense that people from previous generations would have different attitudes towards marijuana than younger generations. People who grew up in the time of Reefer Madness or the hippies may hold different views than those raised in an era of legalized medicinal and recreational use of marijuana.

Once we’ve established that there is a plausible association between the two variables, we also need to establish that the cause happened before the effect, the criterion of temporality . A person’s age is a quality that appears long before any opinions on drug policy, so temporally the cause comes before the effect. It wouldn’t make any sense to say that support for marijuana legalization makes a person’s age increase. Even if you could predict someone’s age based on their support for marijuana legalization, you couldn’t say someone’s age was caused by their support for legalization.

Finally, scientists must establish nonspuriousness. A spurious association is one in which an association between two variables appears to be causal but can in fact be explained by some third variable. For example, we could point to the fact that older cohorts are less likely to have used marijuana. Maybe it is actually use of marijuana that leads people to be more open to legalization, not their age. This is often referred to as the third variable problem, where a seemingly true causal explanation is actually caused by a third variable not in the hypothesis. In this example, the association between age and support for legalization could be more about having tried marijuana than the age of the person.

Quantitative researchers are sensitive to the effects of potentially spurious associations. They are an important form of critique of scientific work. As a result, they will often measure these third variables in their study, so they can control for their effects. These are called control variables , and they refer to variables whose effects are controlled for mathematically in the data analysis process. Control variables can be a bit confusing, but think about it as an argument between you, the researcher, and a critic.

Researcher: “The older a person is, the less likely they are to support marijuana legalization.” Critic: “Actually, it’s more about whether a person has used marijuana before. That is what truly determines whether someone supports marijuana legalization.” Researcher: “Well, I measured previous marijuana use in my study and mathematically controlled for its effects in my analysis. The association between age and support for marijuana legalization is still statistically significant and is the most important association here.”

Let’s consider a few additional, real-world examples of spuriousness. Did you know, for example, that high rates of ice cream sales have been shown to cause drowning? Of course, that’s not really true, but there is a positive association between the two. In this case, the third variable that causes both high ice cream sales and increased deaths by drowning is time of year, as the summer season sees increases in both (Babbie, 2010). Here’s another good one: it is true that as the salaries of Presbyterian ministers in Massachusetts rise, so too does the price of rum in Havana, Cuba. Well, duh, you might be saying to yourself. Everyone knows how much ministers in Massachusetts love their rum, right? Not so fast. Both salaries and rum prices have increased, true, but so has the price of just about everything else (Huff & Geis, 1993).

Finally, research shows that the more firefighters present at a fire, the more damage is done at the scene. What this statement leaves out, of course, is that as the size of a fire increases so too does the amount of damage caused as does the number of firefighters called on to help (Frankfort-Nachmias & Leon-Guerrero, 2011). In each of these examples, it is the presence of a third variable that explains the apparent association between the two original variables.

In sum, the following criteria must be met for a correlation to be considered causal:

• The two variables must vary together.
• The association must be plausible.
• The cause must precede the effect in time.
• The association must be nonspurious (not due to a third variable).

Once these criteria are met, there is a nomothetic causal explanation, one that is objectively true. However, this is difficult for researchers to achieve. You will almost never hear researchers say that they have proven their hypotheses. A statement that bold implies that a association has been shown to exist with absolute certainty and that there is no chance that there are conditions under which the hypothesis would not be true. Instead, researchers tend to say that their hypotheses have been supported (or not). This more cautious way of discussing findings allows for the possibility that new evidence or new ways of examining an association will be discovered. Researchers may also discuss a null hypothesis. The null hypothesis is one that predicts no association between the variables being studied. If a researcher fails to accept the null hypothesis, she is saying that the variables in question are likely to be related to one another.

Idiographic causal explanations

If you not trying to generalize, but instead are trying to establish an idiographic causal explanation, then you are likely going to use qualitative methods, reason inductively, and engage in exploratory or descriptive research. We can understand these assumptions by walking through them, one by one.

Researchers seeking idiographic causal explanation are not trying to generalize, so they have no need to reduce phenomena to mathematics. In fact, using the language of mathematics to reduce the social world down is a bad thing, as it robs the causality of its meaning and context. Idiographic causal explanations are bound within people’s stories and interpretations. Usually, these are expressed through words. Not all qualitative studies analyze words, as some can use interpretations of visual or performance art, but the vast majority of social science studies do.

But wait, we predicted that an idiographic causal explanation would use descriptive or exploratory research. How can we build causality if we are just describing or exploring a topic? Wouldn’t we need to do explanatory research to build any kind of causal explanation?  To clarify, explanatory research attempts to establish nomothetic causal explanations—an independent variable is demonstrated to cause changes a dependent variable. Exploratory and descriptive qualitative research are actually descriptions of the causal explanations established by the participants in your study. Instead of saying “x causes y,” your participants will describe their experiences with “x,” which they will tell you was caused by and influenced a variety of other factors, depending on time, environment, and subjective experience. As stated before, idiographic causal explanations are messy. The job of a social science researcher is to accurately identify patterns in what participants describe.

Let’s consider an example. What would you say if you were asked why you decided to become a social worker?  If we interviewed many social workers about their decisions to become social workers, we might begin to notice patterns. We might find out that many social workers begin their careers based on a variety of factors, such as: personal experience with a disability or social injustice, positive experiences with social workers, or a desire to help others. No one factor is the “most important factor,” like with nomothetic causal explanations. Instead, a complex web of factors, contingent on context, emerge in the dataset when you interpret what people have said.

Finding patterns in data, as you’ll remember from Chapter 2, is what inductive reasoning is all about. A qualitative researcher collects data, usually words, and notices patterns. Those patterns inform the theories we use in social work. In many ways, the idiographic causal explanations created in qualitative research are like the social theories we reviewed in Chapter 2  and other theories you use in your practice and theory courses. Theories are explanations about how different concepts are associated with each other how that network of associations works in the real world. While you can think of theories like Systems Theory as Theory (with a capital “T”), inductive causality is like theory with a small “t.” It may apply only to the participants, environment, and moment in time in which the data were gathered. Nevertheless, it contributes important information to the body of knowledge on the topic studied.

Unlike nomothetic causal explanations, there are no formal criteria (e.g., covariation) for establishing causality in idiographic causal explanations. In fact, some criteria like temporality and nonspuriousness may be violated. For example, if an adolescent client says, “It’s hard for me to tell whether my depression began before my drinking, but both got worse when I was expelled from my first high school,” they are recognizing that oftentimes it’s not so simple that one thing causes another. Sometimes, there is a reciprocal association where one variable (depression) impacts another (alcohol abuse), which then feeds back into the first variable (depression) and also into other variables (school). Other criteria, such as covariation and plausibility still make sense, as the associations you highlight as part of your idiographic causal explanation should still be plausibly true and it elements should vary together.

Similarly, idiographic causal explanations differ in terms of hypotheses. If you recall from the last section, hypotheses in nomothetic causal explanations are testable predictions based on previous theory. In idiographic research, instead of predicting that “x will decrease y,” researchers will use previous literature to figure out what concepts might be important to participants and how they believe participants might respond during the study. Based on an analysis of the literature a researcher may formulate a few tentative hypotheses about what they expect to find in their qualitative study. Unlike nomothetic hypotheses, these are likely to change during the research process. As the researcher learns more from their participants, they might introduce new concepts that participants talk about. Because the participants are the experts in idiographic causal explanation, a researcher should be open to emerging topics and shift their research questions and hypotheses accordingly.

Complementary approaches to causality

Over time, as more qualitative studies are done and patterns emerge across different studies and locations, more sophisticated theories emerge that explain phenomena across multiple contexts. In this way, qualitative researchers use idiographic causal explanations for theory building or the creation of new theories based on inductive reasoning. Quantitative researchers, on the other hand, use nomothetic causal explanations for theory testing , wherein a hypothesis is created from existing theory (big T or small t) and tested mathematically (i.e., deductive reasoning).  Once a theory is developed from qualitative data, a quantitative researcher can seek to test that theory. In this way, qualitatively-derived theory can inspire a hypothesis for a quantitative research project.

Two different baskets

Idiographic and nomothetic causal explanations form the “two baskets” of research design elements pictured in Figure 4.4 below. Later on, they will also determine the sampling approach, measures, and data analysis in your study.

In most cases, mixing components from one basket with the other would not make sense. If you are using quantitative methods with an idiographic question, you wouldn’t get the deep understanding you need to answer an idiographic question. Knowing, for example, that someone scores 20/35 on a numerical index of depression symptoms does not tell you what depression means to that person. Similarly, qualitative methods are not often used to deductive reasoning because qualitative methods usually seek to understand a participant’s perspective, rather than test what existing theory says about a concept.

However, these are not hard-and-fast rules. There are plenty of qualitative studies that attempt to test a theory. There are fewer social constructionist studies with quantitative methods, though studies will sometimes include quantitative information about participants. Researchers in the critical paradigm can fit into either bucket, depending on their research question, as they focus on the liberation of people from oppressive internal (subjective) or external (objective) forces.

We will explore later on in this chapter how researchers can use both buckets simultaneously in mixed methods research. For now, it’s important that you understand the logic that connects the ideas in each bucket. Not only is this fundamental to how knowledge is created and tested in social work, it speaks to the very assumptions and foundations upon which all theories of the social world are built!

Key Takeaways

• Idiographic research focuses on subjectivity, context, and meaning.
• Nomothetic research focuses on objectivity, prediction, and generalizing.
• In qualitative studies, the goal is generally to understand the multitude of causes that account for the specific instance the researcher is investigating.
• In quantitative studies, the goal may be to understand the more general causes of some phenomenon rather than the idiosyncrasies of one particular instance.
• For nomothetic causal explanations, an association must be plausible and nonspurious, and the cause must precede the effect in time.
• In a nomothetic causal explanations, the independent variable causes changes in a dependent variable.
• Hypotheses are statements, drawn from theory, which describe a researcher’s expectation about an association between two or more variables.
• Qualitative research may create theories that can be tested quantitatively.
• The choice of idiographic or nomothetic causal explanation requires a consideration of methods, paradigm, and reasoning.
• Depending on whether you seek a nomothetic or idiographic causal explanation, you are likely to employ specific research design components.
• Causality-the idea that one event, behavior, or belief will result in the occurrence of another, subsequent event, behavior, or belief
• Control variables- potential “third variables” effects are controlled for mathematically in the data analysis process to highlight the relationship between the independent and dependent variable
• Covariation- the degree to which two variables vary together
• Dependent variable- a variable that depends on changes in the independent variable
• Generalize- to make claims about a larger population based on an examination of a smaller sample
• Hypothesis- a statement describing a researcher’s expectation regarding what she anticipates finding
• Idiographic research- attempts to explain or describe your phenomenon exhaustively, based on the subjective understandings of your participants
• Independent variable- causes a change in the dependent variable
• Nomothetic research- provides a more general, sweeping explanation that is universally true for all people
• Plausibility- in order to make the claim that one event, behavior, or belief causes another, the claim has to make sense
• Spurious relationship- an association between two variables appears to be causal but can in fact be explained by some third variable
• Statistical significance- confidence researchers have in a mathematical relationship
• Temporality- whatever cause you identify must happen before the effect
• Theory building- the creation of new theories based on inductive reasoning
• Theory testing- when a hypothesis is created from existing theory and tested mathematically

Image attributions

Mikado by 3dman_eu CC-0

Weather TV Forecast by mohamed_hassan CC-0

Figures 4.2 and 4.3 were copied from Blackstone, A. (2012) Principles of sociological inquiry: Qualitative and quantitative methods. Saylor Foundation. Retrieved from: https://saylordotorg.github.io/text_principles-of-sociological-inquiry-qualitative-and-quantitative-methods/ Shared under CC-BY-NC-SA 3.0 License

Beatrice Birra Storytelling at African Art Museum by Anthony Cross public domain

Foundations of Social Work Research Copyright © 2020 by Rebecca L. Mauldin is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License , except where otherwise noted.

Share This Book

Causal Research: The Complete Guide

Published: February 22, 2023

As we grow up, all humans learn about cause and effect. While it’s not quite as nuanced as causal research, the concept is something our brains begin to comprehend as young as 18 months old. That understanding continues to develop throughout our lives.

In the marketing world, data collection and market research are invaluable. That’s where causal research, the study of cause and effect, comes in.

First-party data can help you learn more about the impact of your marketing campaigns, improve business metrics like customer loyalty, and conduct research on employee productivity. In this guide, we’ll review what causal research is, how it can improve your marketing efforts, and how to conduct your research.

Table of Contents

What is causal research?

The benefits of causal research, causal research examples, how to conduct causal research.

Causal research is a type of study that evaluates whether two variables (one independent, one dependent) have a cause-and-effect relationship. Experiments are designed to collect statistical evidence that infers there is cause and effect between two situations. Marketers can use causal research to see the effect of product changes, rebranding efforts, and more.

Once your team has conducted causal research, your marketers will develop theories on why the relationship developed. Here, your team can study how the variables interact and determine what strategies to apply to future business needs.

Companies can learn how rebranding a product influences sales, how expansion into new markets will affect revenue, and the impact of pricing changes on customer loyalty. Keep in mind that causality is only probable, rather than proven.

Don't forget to share this post!

Related articles.

Market Research: A How-To Guide and Template

SWOT Analysis: How To Do One [With Template & Examples]

20+ Tools & Resources for Conducting Market Research

What's a Competitive Analysis & How Do You Conduct One?

TAM SAM SOM: What Do They Mean & How Do You Calculate Them?

How to Run a Competitor Analysis [Free Guide]

5 Challenges Marketers Face in Understanding Audiences [New Data + Market Researcher Tips]

Total Addressable Market (TAM): What It Is & How You Can Calculate It

What Is Market Share & How Do You Calculate It?

3 Ways Data Privacy Changes Benefit Marketers [New Data]

Free Guide & Templates to Help Your Market Research

Marketing software that helps you drive revenue, save time and resources, and measure and optimize your investments — all on one easy-to-use platform

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• BMC Med Res Methodol

Case study research and causal inference

Judith green.

1 Wellcome Centre for Cultures & Environments of Health, University of Exeter, Exeter, UK

Benjamin Hanckel

2 Institute for Culture and Society, Western Sydney University, Sydney, Australia

Mark Petticrew

3 Department of Public Health, Environments & Society, London School of Hygiene & Tropical Medicine, London, UK

Sara Paparini

4 Wolfson Institute of Population Health, Queen Mary University of London, London, UK

5 Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK

Associated Data

Not applicable; no new data generated in this study.

For the purpose of open access, the author has applied a ‘Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising.

Case study methodology is widely used in health research, but has had a marginal role in evaluative studies, given it is often assumed that case studies offer little for making causal inferences. We undertook a narrative review of examples of case study research from public health and health services evaluations, with a focus on interventions addressing health inequalities. We identified five types of contribution these case studies made to evidence for causal relationships. These contributions relate to: (1) evidence about system actors’ own theories of causality; (2) demonstrative examples of causal relationships; (3) evidence about causal mechanisms; (4) evidence about the conditions under which causal mechanisms operate; and (5) inference about causality in complex systems. Case studies can and do contribute to understanding causal relationships. More transparency in the reporting of case studies would enhance their discoverability, and aid the development of a robust and pluralistic evidence base for public health and health services interventions. To strengthen the contribution that case studies make to that evidence base, researchers could: draw on wider methods from the political and social sciences, in particular on methods for robust analysis; carefully consider what population their case is a case ‘of’; and explicate the rationale used for making causal inferences.

Case study research is widely used in studies of context in public health and health services research to make sense of implementation and service delivery as enacted across complex systems. A recent meta-narrative review identified four broad, overlapping traditions in this body of work: developing and testing complex interventions; analysing change in organisations; undertaking realist evaluations; and studying complex change naturalistically [ 1 ]. Case studies can provide essential thick description of interventions, context and systems; qualitative understanding of the mechanisms of interventions; and evidence of how interventions are adapted in the ‘real’ world [ 2 , 3 ].

However, in evaluative health research, case study designs remain relegated to a minor, supporting role [ 4 , 5 ], typically at the bottom of evidence hierarchies. This relegation is largely due to assumptions that they offer little for making the kinds of causal claims that are essential to evaluating the effects of interventions. The strengths of deep, thick studies of specific cases are conventionally set against the benefits of ‘variable-based’ designs, with the former positioned as descriptive, exploratory or illustrative, and the latter as providing the strongest evidence for making causal claims about the links between interventions and outcomes. In conventional hierarchies of evidence, the primary evidence for making causal claims comes from randomised controlled trials (RCTs), in which the linear relationship between a change in one phenomenon and a later change in another can be delineated from other causal factors. The classic account of causality drawn on in epidemiology requires identifying that the relationship between two phenomena is characterised by co-variation; time order; a plausible relationship; and a lack of competing explanations [ 6 ]. The theoretical and pragmatic limitations of RCT designs for robust and generalizable evaluation of interventions in complex systems are now well-rehearsed [ 2 , 7 – 10 ]. In theory, though, random selection from a population to intervention exposure maximises ability to make causal claims: randomisation minimises risks of confounding, and enables both an unbiased estimate of the effect size of the intervention and extrapolation to the larger population [ 6 ]. Guidance for evaluations in which the intervention cannot be manipulated, such as in natural experiments, therefore typically focuses on methods for addressing threats to validity from non-random allocation in order to strengthen the credibility of probabilistic causal effect estimates [ 4 , 11 ].

This is, however, not the only kind of causal logic. Case study research typically draws on other logics for understanding causation and making causal inferences. We illustrate some of the contributions made by case studies, drawing on a narrative review of research relating to one particularly enduring and complex problem: inequalities in health. The causal chains linking interventions to equity outcomes are long and complex, with recognised limitations in the evidence base for ‘what works’ [ 12 ]. Case study research, we argue, has a critical role to play in making claims about whether, how and why interventions reduce, mitigate, or exacerbate inequalities. Our examples are drawn from a broader review of case study research [ 1 ] and supporting literature reviews [ 5 ], from which we focused on cases which had an explanatory aim, and which shed light on how interventions in public health or health services might reduce, create or sustain inequality. In this paper, we: i) outline some different kinds of evidence relevant to causal relationships that can be  derived from case study research; ii) outline what is needed for case study research to contribute to explanatory, as well as exploratory claims; and iii) advocate for greater clarity in reporting case study research to foster discoverability.

Cases and causes

There are considerable challenges in defining case study designs or approaches in ways that adequately delineate them from other research designs. Yin [ 13 ], for instance, one of the most highly cited source texts on case studies in health research [ 1 ], resists providing a definition, instead suggesting case study research is more a strategy for doing empirical research. Gerring [ 14 ] defines case study research as: “ an intensive study of a single unit for the purpose of understanding a larger class of (similar) units ” (p342, emphasis in original). This definition is useful in suggesting the basis for the inferences drawn from cases, and the need to consider the relationships between the ‘case’ (and phenomena observed within it) and the population from which it is drawn. Gerring notes that studies of single cases may have a greater “affinity” for descriptive aims, but that they can furnish “evidence for causal propositions” ( [ 14 ], p347). Case studies are, he suggests, more likely to be useful in elucidating deterministic causes: those conditions that are necessary and/or sufficient for an outcome, whereas variable based designs have advantages for demonstrating probabilistic causation, where the aim is to estimate the likelihood of two phenomena being causally related. Case studies provide evidence for the mechanisms of causal relationships (e.g. through process tracing, through observing two variables interacting in the real world) and corroboration of causal relationships (for instance, through pattern matching).

Gerring’s argument, drawing on political science examples, is that there is nothing epistemologically distinct about research using the case study: rather, it has particular affinities with certain styles of causal modelling. We take this as a point of departure to consider not whether case studies can furnish evidence to help with causal inference in health research, but rather how they have done this. From our examples on case study research on inequalities in health, we identify the kinds of claims that relate to causality that were made. We note that some relate to (1) Actors’ accounts of causality : that is, the theories of those studied about if, how and why interventions work. Other types of claim use various kinds of comparative analytic logic to elucidate evidence of causal relationships between phenomena. These claims include: (2) Demonstrations of causal relationships – in which evidence from one case is sufficient for identifying a plausible causal relationship; (3) Mechanisms – evidence of the mechanisms through which causal relationships work; (4) Conditions —evidence of the conditions under which such mechanisms operate; and (5) Complex causality —evidence for outcomes that arise from complex causality within a system. This list is neither mutually exclusive, nor exhaustive: many case studies aim to do several of these (and some more). It is also a pragmatic rather than theoretical list, focusing on the kinds of evidence claimed by researchers rather than the formal methodological underpinnings of causal claims (for a discussion of the latter, see Rohlfing [ 15 ]).

What kinds of causal evidence do case studies provide?

Actors’ accounts of causality.

This is perhaps the most common kind of evidence provided by case study research. Case studies, through in-depth research on the actors within systems, can generate evidence about how those actors themselves account for causal relationships between interventions and outcomes. This is an overt aim of many realist evaluation studies, which focus on real forces or processes that exist in the world that can provide insight into causal mechanisms for change.

Ford and colleagues [ 16 ], for example, used a series of five case studies of local health systems to explore socio-economic inequalities in unplanned hospital admission. Cases were selected on the basis of either narrowing or widening inequalities in admission, with a realist evaluation focused on delineating the context-mechanisms-outcome (CMO) configurations in each setting, to develop a broader theory of change for addressing inequalities. The case study approach used a mix of methods, including drawing on documentary data to assess the credibility of mechanisms proposed by health providers. The authors identified 17 distinct CMO configurations; and five factors that were related to trends for inequalities in emergency admissions, including health service factors (primary care workforce challenges, case finding and proactive case management) and those external to the health service (e.g., financial constraints on public services, residential gentrification). Ford and colleagues noted that none of the CMO configurations were clearly associated with improved or worsening trends in inequalities in admission.

Clearly, actors’ accounts of causality are not in themselves evidence of causality. Ford and colleagues noted that they interrogated accounts for plausibility (e.g. that interventions mentioned were prior to effects claimed) and triangulated these accounts with other sources of data, but that inability to empirically corroborate the hypothesized CMO links limited their ability to make claims about causal inference. This is crucial: actors in a system may be aware of the forces and processes shaping change but unaware of counterfactuals, and they are unlikely to have any privileged insight into whether factors are causal or simply co-occurring (see, for instance, Milton et. al. [ 17 ] on how commonly cited ‘barriers’ in accounts of not doing evaluations are also evident in actors’ accounts of doing successful evaluations). Over-interpretation of qualitative accounts of insiders’ claims about causal relationships as if they provide conclusive evidence of causal relationships is poor methodology.

This does not mean that actors’ accounts are not of value. First, in realist evaluation, as in Ford and colleagues’ study [ 16 ], these accounts provide the initial theories of change for thinking about the potential causal pathways in logic models of interventions. Second, insiders’ accounts of causality are part of the system that is being explained. An example comes from Mead and colleagues [ 18 ], who used a case study drawing largely on qualitative interviews to explore “how local actors from public health, and the wider workforce, make sense of and work on social inequalities in health” ( [ 18 ] p168). This used a case study of a partnership in northwest England to address an enduring challenge in inequalities policy: the tendency for policies that address upstream health determinants to transform, in practice, to focus more on behavioural and individual level factors . Local public health actors in the partnership recognised the structural causes of unequal health outcomes, yet discourses of policy action tended to focus only on the downstream, more individualising levels of health, and on personal choice and agency as targets for intervention. Professionals conceptualised action on inequality as relating only to the health of the poorest, rather than as a problem of a gradient in health outcomes across society. There was a geographical localism in their approach, which framed particular places as constellations of health and social problems. Drawing on theory from figurational sociology, Mead and colleagues note that actors’ own accounts are the starting point of an analysis, which then puts those accounts into play with theory about how such discourses are reproduced. The researchers suggest that partnership working itself exacerbated the individualising frameworks used to orient action, as it became a hegemonic framing, reducing the possibilities for partnerships to transform health inequalities. Here, then, a case study approach is used to shed light on the causes of a common failure in policies addressing inequalities. The authors take seriously the divergence of actors’ own accounts of causality and those of other sources, and analyse these as part of the system.

Finally, insider accounts should be taken seriously as contributing to evidence about causal inference through shedding light on the complex looping effects of theoretical models of causality and public accounts. For instance, Smith and Anderson [ 19 ], drawing on a meta-ethnographic literature review of ‘lay theorising’ about health inequalities, note that, counter to common assumptions, public understanding of the structural causes of health inequalities is sophisticated: but that it may be disavowed to avoid stigma and shame and to reassert some agency. This is an important finding for informing knowledge exchange, suggesting that further ‘awareness raising’ may be unnecessary for policy change, and counter-productive in needlessly increasing stigma and shame.

Demonstrations of causal relationships

When strategically sampled, and rooted in a sound theoretical framework, studies of single cases can provide evidence for generalizable causal inferences. The strongest examples are perhaps those that operate as ‘black swans’ for deterministic claims, in that one case may be all that is needed to show that a commonly held assumption is not generalizable. That is, a case study can demonstrate unequivocally that one phenomenon is not inevitably related to another. These can come from cases sampled because they are extreme or unusual. Prior’s [ 20 ] study of a single man in a psychiatric institution in Northern Ireland, for instance, showed that, counter to Goffman’s [ 21 ] original theory of how ‘total institutions’ lead to stigmatisation and depersonalisation, the effects of institutionalisation depended on context—in this case, how the institution related to the local community and the availability of alternative sources of self-worth available to residents.

Strategically sampled typical cases can also provide demonstrative evidence of causal relationships. To take the enduring health services challenge of inequalities in self-referral to emergency care, Hudgins and Rising’s [ 22 ] case study of a single patient is used to debunk a common assumption that high use of emergency care is related to inappropriate care-seeking by low-income patients. They look in detail at the case of “a 51-year-old low-income, recently insured, African American man in Philadelphia (USA) who had two recent ED [emergency department] visits for evaluation of frequent headaches and described fear of being at risk for a stroke.” ( [ 22 ] p50). Drawing on theories of structural violence and patient subjectivity, they use this single case to shed light on why emergency department use may appear inappropriate to providers. They analyse the interplay of gender roles, employment, and insurance status in generating competing drivers of health seeking, and point to the ways in which current policies deterring self-referral do not align well with micro- and macro-level determinants of service use. The study authors also note that because their methods generate data on ‘why’ as well ‘what’ people do, they can “lay the groundwork” ( [ 22 ], p54] for developing future interventions. Here, again, a single case is sufficient. In understanding the causal pathways that led to this patient’s use of emergency care, it is clear why policies addressing inequalities through deterring low-income users would be unlikely to work.

Mechanisms: how causal relationships operate

A strength of case study approaches compared with variable-based designs is furnishing evidence of how causal relationships operate, deriving from both direct observations of causal processes and from analysis of comparisons within and between cases. All cases contain multiple observations; variations can be observed over time and space, across or within cases [ 14 ]. Observing regularities, co-variation and deviant or surprising findings, and then using processes of analytic induction [ 23 ] or abductive logic [ 24 ] to derive, develop and test causal theories using observations from the case, can build a picture of causal pathways.

Process tracing is one formal qualitative methodology for doing this. Widely used in political and policy studies, but less in health evaluations [ 25 ], process tracing links outcomes with their causes, focusing on the mechanisms that link events on causal pathways, and on the strength of evidence for making connections on that causal chain. This requires sound theoretical knowledge (such that credible hypotheses can be developed), well described cases (ideally at different time points), observed causal processes (the activities that transfer causes to effects), and careful assessment of evidence against tests of varying strength for the necessity and sufficiency for accepting or rejecting a candidate hypothesis [ 26 , 27 ]. In health policy, process tracing methods have been combined to good effect with quantitative measures to examine casual processes leading to outcomes of interest. Campbell et. al. [ 28 ], for instance, used process tracing to look at four case studies of countries that had made progress towards universal health coverage (measured through routine data on maternal and neonatal health indicators), to identify key causal factors related to health care workforce.

An example of the use of process tracing in evaluation comes from Lohmann and colleagues’ [ 25 ] case study of a single country, Burkina Faso, to examine why performance based financing (PBF) fails to improve equity. PBF, coupled with interventions to improve health care take up among the poor, aims to improve health equity in low and middle-income countries, yet impact evaluations suggest that these benefits are typically not realised. This case study drew on data from the quantitative impact assessment; programme documentation; the intervention process evaluation; and primary qualitative research for the process tracing, in the light of the theory of change of the intervention. Lohmann and colleagues [ 25 ] identified that a number of conditions that would have been necessary for the intervention to work had not been met (such as eligible patients not receiving the card needed to access health care or providers not receiving timely reimbursement). A key finding was that although implementation challenges were a partial cause of policy failure, other causal conditions were external to the intervention, such as lack of attention to the non-health care costs incurred by the poorest to access care. Again, a single case, if there are good grounds for extrapolating to similar contexts (i.e., those in which transport is required to access health care), is enough to demonstrate a necessary part of the causal pathway between PBF and intended equity outcomes.

Conditions under which causal mechanisms operate

The example of ‘transport access’ as a necessary condition for PBF interventions to ‘work’ also illustrates a fourth type of causal evidence: that relating to the transferability of interventions. Transferable causal claims are essential for useful evidence: “(f)or policy and practice we do not need to know ‘it works somewhere’. We need evidence for ‘it-will-work-for-us’ claims: the treatment will produce the desired outcome in our situation as implemented there” ( [ 8 ] p1401). Some causal mechanisms operate widely (using a parachute will reduce injury from jumping from a plane; taking aspirin will relieve pain); others less so. In the context of health services and public health research, few interventions are likely to be widely generalizable, as the mechanisms will operate differently across contexts [ 7 ]. This context dependency is at the heart of realist evaluations, with the assumption that underlying causal mechanisms require particular contexts in order to operate, hence the focus on ‘how, where, and for whom’ interventions work [ 29 ]. Making useful claims therefore requires other kinds of evidence, relating to what Cartwright and Munro [ 30 ] call the ‘capacities’ of the intervention: what power it has to work reliably, what stops it working, what other conditions are needed for it to work. This evidence is critical for assessing whether an intervention is likely to work in a given context and to assess the intended and unintended consequences of intervention adoption and implementation. Cartwright and Munro’s recommendation is therefore to study causal powers rather than causes. That is, as well as interrogating whether the intervention ‘causes’ a particular outcome, it is also necessary to address the potential for and stability of that causal effect. To do that entails addressing a broader range of questions about the causal relationship, such as how the intervention operates in order to bring about changes in outcomes; what other conditions need to be present; what might constrain this effect; what other factors within the system also promote or constrain those effects; and what happens when different capacities interact? [ 30 ]. Case study research can be vital in providing this kind of evidence on the capacities of interventions [ 31 ].

One example is from Gibson and colleagues [ 32 ], who use within-case comparisons to shed light on why a ‘social prescribing’ intervention may have different effects across socioeconomic classes. These interventions, typically entailing link workers who connect people with complex health care needs to local services and resources, are often framed as a way to address enduring health inequalities. Drawing on sociological theory on how social class is reproduced through socially structured and unequal distribution of resources (‘capitals’), and through how these shape people’s practices and dispositions, Gibson and colleagues [ 32 ] explicate how capitals and dispositions shaped encounters with the intervention. Their analysis of similarities and differences within their case (of different clients) in the context of theory enables them to abstract inferences from the case. Drawing out the ways in which more advantaged clients mobilised capital in their pursuit of health, with dispositions more closely aligned to the intervention, they unravel classed differences in ability to benefit from the intervention, with less advantaged clients inevitably having ‘shorter horizons’ focused on day to day challenges: “This challenges the claim that social prescribing can reduce inequalities, instead suggesting it has the potential to exacerbate existing inequalities” ( [ 32 ], p6).

Case studies can shed light on the capacities of interventions to improve or exacerbate inequalities, including identifying unforeseen consequences. Hanckel and colleagues [ 33 , 34 ], for example, used a case study approach to explore implementation of a physical health intervention involving whole classes of children running for 15 min each day in the playground in schools in south London, UK. This documented considerable adaption of the intervention at the level of school, class and pupil, and identified different pathways through which the intervention might impact on inequalities. In terms of access, the intervention appeared to be equitable, in that there was no evidence of disproportionate roll out to schools with more affluent pupils or to those with fewer minority ethnic pupils [ 33 ]. However, identifying the ‘capacities’ of the intervention also identified other pathways through which it could have negative equity effects. The authors found that in practice, the intervention emphasised body weight rather than physical activity, and intervention roll-out reinforced class and ethnicity-based stigmatising discourses about lower income neighbourhoods [ 34 ].

Complex causality

There is increasing recognition that the systems that reproduce unequal health outcomes are complex: that is, that they consist of multiple interacting components that cannot be studied in isolation, and that change is likely to be non-linear, characterised by, for instance, phase shifts or feedback loops [ 35 ]. This has two rather different implications. First, case study designs can be particularly beneficial for taking a system perspective on interventions. Case studies enable a focus on aspects that are not well explicated through other designs, such as how context interacts with interventions within systems [ 7 ], or on how multiple conditional pathways might link interventions and outcomes [ 36 ]. Second, when causation is not linear, but ‘emergent’, in that it is not reducible to the accumulated changes at lower levels, evaluation designs focused on only one outcome at one level (such as weight loss in individuals) may fail to identify important effects. Case studies have an invaluable role here in unpacking and surfacing these effects at different levels within the systems within which interventions and services are delivered. One example is transport systems, which have been the focus of considerable public health interest to encourage more ‘active’ modes, in which more of the population walk or cycle, and fewer drive. However, more simplistic evaluations looking at one part of a causal chain (such as that between traffic calming interventions and local mode shift) may fail to appreciate how systems are dynamic, and that causation might be emergent. This is evident in a case study of transport policy impacts from Sheller [ 37 ], who takes the case of Philadelphia, USA, to reveal how this post-car trend has racialized effects that can exacerbate inequality. Weaving in data from participant observations, historical documentary sources and statistical evidence of declining car use, Sheller documents the racialized impacts of transport policies which may have reduced car use and encouraged active modes overall, but which have largely prioritised ‘young white’ mobility in the context of local gentrification and neglect of public transit.

One approach to synthesising evidence from multiple case studies to make claims about complex causation is Qualitative Comparative Analysis (QCA), which combines quantitative methods (based on Boolean algebra) with detailed qualitative understanding of a small to medium N sample of cases. This has strengths for identifying multiple pathways to outcomes, asymmetrical sets of conditions which lead to success or failure, or ‘conjunctural causation’, whereby some conditions are only causally linked to outcomes in relation to others [ 38 ]. There is growing interest in using these approaches in evaluative health studies [ 39 ]. One example relating to the effectiveness of interventions addressing inequalities in health comes from Blackman and colleagues [ 36 ], who explored configurations of conditions which did or did not lead to narrowing inequalities in teenage conception rates across a series of local areas as cases. This identified some surprising findings, including that ‘basic’ rather than good or exemplary standards of commissioning were associated with narrowing the equity gap, and that the proportion of minority ethnic people in the population was a key condition.

Not all case study research aims to contribute to causal inference, and neither should it [ 1 , 5 , 40 ]. However, it can. We have identified five ways in which case study evidence has contributed to causal explanations in relation to a particularly intractable challenge: inequalities in health. It is therefore time to stop claiming that case study designs have only a supporting role to play in evaluative health research. To develop a theoretical evidence base on ‘what works’, and how, in health services and public health, particularly around complex issues such as addressing unequal health outcomes, we need to draw on a greater range of evidential resources for informing decisions than is currently used. Best explanations are unlikely to be made from single studies based on one kind of causality, but instead will demand some kind of evidential pluralism [ 41 ]. That is, one single study, of any design, is unlikely to generate evidence for all links in complex causal chains between an intervention and health outcomes. We need a bricolage of evidence from a diverse range of designs [ 42 ] to make robust and credible cases for what will improve health and health equity. This will include evidence from case studies, both from single and small N studies, and from syntheses of findings from multiple cases.

Our focus on case studies that shed light on interventions for health inequalities identified the critical role that case studies can play in theorising, illuminating and making sense of: system actors’ own causal reasoning; whether there are causal links between intervention and outcome; what mechanism(s) might link them; when, where and for whom these causal relationships operate; and how unequal outcomes can be generated from the operation of complex systems. These examples draw on a range of different theoretical and methodological approaches, often from the wider political and social sciences. The approaches illustrated are rooted in very different, even incompatible, philosophical traditions: what researchers understand by ‘causality’ is diverse [ 43 ]. However, there are two commonalities across this diversity that suggest some conditions for producing good case studies that can generate evidence to support causal inferences. The first is the need for theoretically informed and comparative analysis. As Gerring [ 14 ] notes, causal inferences rely on comparisons – across units or time within a case, or between cases. It is comparison that drives the ability to make claims about the potential of interventions to produce change in outcomes of interest, and under what conditions. There are a range of approaches to qualitative data analysis, and choice of method has to be appropriate for the kinds of causal logics being explicated, and the availability of data on particular phenomena within the case. Typically, though, this will require analysis that goes beyond descriptive thematic analysis [ 31 ]. Approaches such as process tracing or analytic induction require both fine-grained and rigorous comparative analysis, and a sound theoretical underpinning that provides a framework for making credible inferences about the relationships between phenomena within the case and to the wider population from which the case is selected.

This leads to the second commonality: the need to clarify what the case is a case ‘of’, and how it relates to other candidate cases. What constitutes a ‘case’ is inevitably study specific. The examples we have drawn on include: PBF in a country [ 25 ], transport systems in a city [ 37 ], and a social prescribing intervention in primary care [ 32 ]. Clearly, in other contexts, each of these ‘cases’ could be sampling units within variable based studies (of financing systems, or countries; of infrastructures systems, or cities in a state; of particular kinds of service intervention, or primary care systems). Conversely, these cases could be populations within which lower level phenomena (districts, neighbourhoods, patients) are studied. What leads to appropriate generalisations about causal claims is a sound theorisation of the similarities and particularities of the case compared with other candidate cases: how Burkina Faso has commonalities with, or differences from, other settings in which PBF has failed to improve equity; or the contexts of gentrification and residential churn that make Philadelphia similar to other cities in the US; or the ways in which class-based dispositions and practices intersect with similar types of service provisions.

A critical question remains: How can well-conducted case study evidence be better integrated into the evidence base? Calls for greater recognition for case study designs within health research are hardly new: Flyvberg’s advocacy for a greater role for case studies in the social sciences [ 44 ] has now been cited around 20,000 times, and calls for methodological pluralism in health research go back decades [ 42 , 45 , 46 ]. Yet, case studies remain somewhat neglected, with ongoing misconceptions about their limited role, despite calls for evidence based medicine to incorporate evidence for mechanisms as complementary to evidence of correlation, rather than as inferior [ 47 ]. Even where the value of case studies for contributing to causal inference is recognised, searching for good evidence is not straightforward. Case studies are neither consistently defined nor necessarily well reported. Some of the examples in this paper do not use the term ‘case study’ in the title or abstract, although they meet our definition. Conversely, many small scale qualitative studies describe themselves as ‘case studies’, but focus on thick description rather than generalisability, and are not aiming to contribute to evaluative evidence. It is therefore challenging, currently, to undertake a more systematic review of empirical material. Forthcoming guidance on reporting case studies of context in complex systems aims to aid discoverability and transparency of reporting (Shaw S, et al: TRIPLE C Reporting Principles for Case study evaluations of the role of Context in Complex interventions, under review). This recommends including ‘case study’ in the title, clarifying how terms are used, and explicating the philosophical base of the study. To further advance the usefulness of case study evidence, we suggest that where an aim is to contribute to causal explanations, researchers should, in addition, specify their rationales for making causal inferences, and identify what broader class of phenomena their case is a case ‘of’.

Conclusions

Case study research can and does contribute to evidence for causal inferences. On challenging issues such as addressing health inequalities, we have shown how case studies provide more than detailed description of context or process. Contributions include: describing actors’ accounts of causal relationships; demonstrating theoretically plausible causal relationships; identifying mechanisms which link cause and effect; identifying the conditions under which causal relationships hold; and researching complex causation.

Acknowledgements

The research underpinning this paper was conducted as part of the Triple C study. We gratefully acknowledge the input of the wider study team, and that of the participants at a workshop held to discuss forthcoming guidance on reporting case study research.

Abbreviations

Authors’ contributions.

BH, JG and MP drafted the first version of the paper, which was revised with theoretical input from SS and SP. All authors contributed to the paper and have reviewed and approved the final manuscript.

The research was funded by the Medical Research Council (MR/S014632/1). JG is supported with funding from the Wellcome Trust (WT203109/Z/16/Z). Additional funding for SP and SS salaries over the course of the study was provided by the UK National Institute for Health Research Oxford Biomedical Research Centre (BRC-1215–20008), Wellcome Trust (WT104830MA; 221457/Z/20/Z) and the University of Oxford's Higher Education Innovation Fund.

The views and opinions expressed herein are those of the authors. Funding bodies had no input to the design of the study and collection, analysis, and interpretation of data or preparation of this paper.

Availability of data and materials

Declarations.

Not applicable.

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Causal Hypothesis

In scientific research, understanding causality is key to unraveling the intricacies of various phenomena. A causal hypothesis is a statement that predicts a cause-and-effect relationship between variables in a study. It serves as a guide to study design, data collection, and interpretation of results. This thesis statement segment aims to provide you with clear examples of causal hypotheses across diverse fields, along with a step-by-step guide and useful tips for formulating your own. Let’s delve into the essential components of constructing a compelling causal hypothesis.

What is Causal Hypothesis?

A causal hypothesis is a predictive statement that suggests a potential cause-and-effect relationship between two or more variables. It posits that a change in one variable (the independent or cause variable) will result in a change in another variable (the dependent or effect variable). The primary goal of a causal hypothesis is to determine whether one event or factor directly influences another. This type of Simple hypothesis is commonly tested through experiments where one variable can be manipulated to observe the effect on another variable.

What is an example of a Causal Hypothesis Statement?

Example 1: If a person increases their physical activity (cause), then their overall health will improve (effect).

Explanation: Here, the independent variable is the “increase in physical activity,” while the dependent variable is the “improvement in overall health.” The hypothesis suggests that by manipulating the level of physical activity (e.g., by exercising more), there will be a direct effect on the individual’s health.

Other examples can range from the impact of a change in diet on weight loss, the influence of class size on student performance, or the effect of a new training method on employee productivity. The key element in all causal hypotheses is the proposed direct relationship between cause and effect.

100 Causal Hypothesis Statement Examples

Size: 185 KB

Causal hypotheses predict cause-and-effect relationships, aiming to understand the influence one variable has on another. Rooted in experimental setups, they’re essential for deriving actionable insights in many fields. Delve into these 100 illustrative examples to understand the essence of causal relationships.

• Dietary Sugar & Weight Gain: Increased sugar intake leads to weight gain.
• Exercise & Mental Health: Regular exercise improves mental well-being.
• Sleep & Productivity: Lack of adequate sleep reduces work productivity.
• Class Size & Learning: Smaller class sizes enhance student understanding.
• Smoking & Lung Disease: Regular smoking causes lung diseases.
• Pesticides & Bee Decline: Use of certain pesticides leads to bee population decline.
• Stress & Hair Loss: Chronic stress accelerates hair loss.
• Music & Plant Growth: Plants grow better when exposed to classical music.
• UV Rays & Skin Aging: Excessive exposure to UV rays speeds up skin aging.
• Reading & Vocabulary: Regular reading improves vocabulary breadth.
• Video Games & Reflexes: Playing video games frequently enhances reflex actions.
• Air Pollution & Respiratory Issues: High levels of air pollution increase respiratory diseases.
• Green Spaces & Happiness: Living near green spaces improves overall happiness.
• Yoga & Blood Pressure: Regular yoga practices lower blood pressure.
• Meditation & Stress Reduction: Daily meditation reduces stress levels.
• Social Media & Anxiety: Excessive social media use increases anxiety in teenagers.
• Alcohol & Liver Damage: Regular heavy drinking leads to liver damage.
• Training & Job Efficiency: Intensive training improves job performance.
• Seat Belts & Accident Survival: Using seat belts increases chances of surviving car accidents.
• Soft Drinks & Bone Density: High consumption of soft drinks decreases bone density.
• Homework & Academic Performance: Regular homework completion improves academic scores.
• Organic Food & Health Benefits: Consuming organic food improves overall health.
• Fiber Intake & Digestion: Increased dietary fiber enhances digestion.
• Therapy & Depression Recovery: Regular therapy sessions improve depression recovery rates.
• Financial Education & Savings: Financial literacy education increases personal saving rates.
• Brushing & Dental Health: Brushing teeth twice a day reduces dental issues.
• Carbon Emission & Global Warming: Higher carbon emissions accelerate global warming.
• Afforestation & Climate Stability: Planting trees stabilizes local climates.
• Ad Exposure & Sales: Increased product advertisement boosts sales.
• Parental Involvement & Academic Success: Higher parental involvement enhances student academic performance.
• Hydration & Skin Health: Regular water intake improves skin elasticity and health.
• Caffeine & Alertness: Consuming caffeine increases alertness levels.
• Antibiotics & Bacterial Resistance: Overuse of antibiotics leads to increased antibiotic-resistant bacteria.
• Pet Ownership & Loneliness: Having pets reduces feelings of loneliness.
• Fish Oil & Cognitive Function: Regular consumption of fish oil improves cognitive functions.
• Noise Pollution & Sleep Quality: High levels of noise pollution degrade sleep quality.
• Exercise & Bone Density: Weight-bearing exercises increase bone density.
• Vaccination & Disease Prevention: Proper vaccination reduces the incidence of related diseases.
• Laughter & Immune System: Regular laughter boosts the immune system.
• Gardening & Stress Reduction: Engaging in gardening activities reduces stress levels.
• Travel & Cultural Awareness: Frequent travel increases cultural awareness and tolerance.
• High Heels & Back Pain: Prolonged wearing of high heels leads to increased back pain.
• Junk Food & Heart Disease: Excessive junk food consumption increases the risk of heart diseases.
• Mindfulness & Anxiety Reduction: Practicing mindfulness lowers anxiety levels.
• Online Learning & Flexibility: Online education offers greater flexibility to learners.
• Urbanization & Wildlife Displacement: Rapid urbanization leads to displacement of local wildlife.
• Vitamin C & Cold Recovery: High doses of vitamin C speed up cold recovery.
• Team Building Activities & Work Cohesion: Regular team-building activities improve workplace cohesion.
• Multitasking & Productivity: Multitasking reduces individual task efficiency.
• Protein Intake & Muscle Growth: Increased protein consumption boosts muscle growth in individuals engaged in strength training.
• Mentoring & Career Progression: Having a mentor accelerates career progression.
• Fast Food & Obesity Rates: High consumption of fast food leads to increased obesity rates.
• Deforestation & Biodiversity Loss: Accelerated deforestation results in significant biodiversity loss.
• Language Learning & Cognitive Flexibility: Learning a second language enhances cognitive flexibility.
• Red Wine & Heart Health: Moderate red wine consumption may benefit heart health.
• Public Speaking Practice & Confidence: Regular public speaking practice boosts confidence.
• Fasting & Metabolism: Intermittent fasting can rev up metabolism.
• Plastic Usage & Ocean Pollution: Excessive use of plastics leads to increased ocean pollution.
• Peer Tutoring & Academic Retention: Peer tutoring improves academic retention rates.
• Mobile Usage & Sleep Patterns: Excessive mobile phone use before bed disrupts sleep patterns.
• Green Spaces & Mental Well-being: Living near green spaces enhances mental well-being.
• Organic Foods & Health Outcomes: Consuming organic foods leads to better health outcomes.
• Art Exposure & Creativity: Regular exposure to art boosts creativity.
• Gaming & Hand-Eye Coordination: Engaging in video games improves hand-eye coordination.
• Prenatal Music & Baby’s Development: Exposing babies to music in the womb enhances their auditory development.
• Dark Chocolate & Mood Enhancement: Consuming dark chocolate can elevate mood.
• Urban Farms & Community Engagement: Establishing urban farms promotes community engagement.
• Reading Fiction & Empathy Levels: Reading fiction regularly increases empathy.
• Aerobic Exercise & Memory: Engaging in aerobic exercises sharpens memory.
• Meditation & Blood Pressure: Regular meditation can reduce blood pressure.
• Classical Music & Plant Growth: Plants exposed to classical music show improved growth.
• Pollution & Respiratory Diseases: Higher pollution levels increase respiratory diseases’ incidence.
• Parental Involvement & Child’s Academic Success: Direct parental involvement in schooling enhances children’s academic success.
• Sugar Intake & Tooth Decay: High sugar intake is directly proportional to tooth decay.
• Physical Books & Reading Comprehension: Reading physical books improves comprehension better than digital mediums.
• Daily Journaling & Self-awareness: Maintaining a daily journal enhances self-awareness.
• Robotics Learning & Problem-solving Skills: Engaging in robotics learning fosters problem-solving skills in students.
• Forest Bathing & Stress Relief: Immersion in forest environments (forest bathing) reduces stress levels.
• Reusable Bags & Environmental Impact: Using reusable bags reduces environmental pollution.
• Affirmations & Self-esteem: Regularly reciting positive affirmations enhances self-esteem.
• Local Produce Consumption & Community Economy: Buying and consuming local produce boosts the local economy.
• Sunlight Exposure & Vitamin D Levels: Regular sunlight exposure enhances Vitamin D levels in the body.
• Group Study & Learning Enhancement: Group studies can enhance learning compared to individual studies.
• Active Commuting & Fitness Levels: Commuting by walking or cycling improves overall fitness.
• Foreign Film Watching & Cultural Understanding: Watching foreign films increases understanding and appreciation of different cultures.
• Craft Activities & Fine Motor Skills: Engaging in craft activities enhances fine motor skills.
• Listening to Podcasts & Knowledge Expansion: Regularly listening to educational podcasts broadens one’s knowledge base.
• Outdoor Play & Child’s Physical Development: Encouraging outdoor play accelerates physical development in children.
• Thrift Shopping & Sustainable Living: Choosing thrift shopping promotes sustainable consumption habits.
• Nature Retreats & Burnout Recovery: Taking nature retreats aids in burnout recovery.
• Virtual Reality Training & Skill Acquisition: Using virtual reality for training accelerates skill acquisition in medical students.
• Pet Ownership & Loneliness Reduction: Owning a pet significantly reduces feelings of loneliness among elderly individuals.
• Intermittent Fasting & Metabolism Boost: Practicing intermittent fasting can lead to an increase in metabolic rate.
• Bilingual Education & Cognitive Flexibility: Being educated in a bilingual environment improves cognitive flexibility in children.
• Urbanization & Loss of Biodiversity: Rapid urbanization contributes to a loss of biodiversity in the surrounding environment.
• Recycled Materials & Carbon Footprint Reduction: Utilizing recycled materials in production processes reduces a company’s overall carbon footprint.
• Artificial Sweeteners & Appetite Increase: Consuming artificial sweeteners might lead to an increase in appetite.
• Green Roofs & Urban Temperature Regulation: Implementing green roofs in urban buildings contributes to moderating city temperatures.
• Remote Work & Employee Productivity: Adopting a remote work model can boost employee productivity and job satisfaction.
• Sensory Play & Child Development: Incorporating sensory play in early childhood education supports holistic child development.

Causal Hypothesis Statement Examples in Research

Research hypothesis often delves into understanding the cause-and-effect relationships between different variables. These causal hypotheses attempt to predict a specific effect if a particular cause is present, making them vital for experimental designs.

• Artificial Intelligence & Job Market: Implementation of artificial intelligence in industries causes a decline in manual jobs.
• Online Learning Platforms & Traditional Classroom Efficiency: The introduction of online learning platforms reduces the efficacy of traditional classroom teaching methods.
• Nano-technology & Medical Treatment Efficacy: Using nano-technology in drug delivery enhances the effectiveness of medical treatments.
• Genetic Editing & Lifespan: Advancements in genetic editing techniques directly influence the lifespan of organisms.
• Quantum Computing & Data Security: The rise of quantum computing threatens the security of traditional encryption methods.
• Space Tourism & Aerospace Advancements: The demand for space tourism propels advancements in aerospace engineering.
• E-commerce & Retail Business Model: The surge in e-commerce platforms leads to a decline in the traditional retail business model.
• VR in Real Estate & Buyer Decisions: Using virtual reality in real estate presentations influences buyer decisions more than traditional methods.
• Biofuels & Greenhouse Gas Emissions: Increasing biofuel production directly reduces greenhouse gas emissions.
• Crowdfunding & Entrepreneurial Success: The availability of crowdfunding platforms boosts the success rate of start-up enterprises.

Causal Hypothesis Statement Examples in Epidemiology

Epidemiology is a study of how and why certain diseases occur in particular populations. Causal hypotheses in this field aim to uncover relationships between health interventions, behaviors, and health outcomes.

• Vaccine Introduction & Disease Eradication: The introduction of new vaccines directly leads to the reduction or eradication of specific diseases.
• Urbanization & Rise in Respiratory Diseases: Increased urbanization causes a surge in respiratory diseases due to pollution.
• Processed Foods & Obesity Epidemic: The consumption of processed foods is directly linked to the rising obesity epidemic.
• Sanitation Measures & Cholera Outbreaks: Implementing proper sanitation measures reduces the incidence of cholera outbreaks.
• Tobacco Consumption & Lung Cancer: Prolonged tobacco consumption is the primary cause of lung cancer among adults.
• Antibiotic Misuse & Antibiotic-Resistant Strains: Misuse of antibiotics leads to the evolution of antibiotic-resistant bacterial strains.
• Alcohol Consumption & Liver Diseases: Excessive and regular alcohol consumption is a leading cause of liver diseases.
• Vitamin D & Rickets in Children: A deficiency in vitamin D is the primary cause of rickets in children.
• Airborne Pollutants & Asthma Attacks: Exposure to airborne pollutants directly triggers asthma attacks in susceptible individuals.
• Sedentary Lifestyle & Cardiovascular Diseases: Leading a sedentary lifestyle is a significant risk factor for cardiovascular diseases.

Causal Hypothesis Statement Examples in Psychology

In psychology, causal hypotheses explore how certain behaviors, conditions, or interventions might influence mental and emotional outcomes. These hypotheses help in deciphering the intricate web of human behavior and cognition.

• Childhood Trauma & Personality Disorders: Experiencing trauma during childhood increases the risk of developing personality disorders in adulthood.
• Positive Reinforcement & Skill Acquisition: The use of positive reinforcement accelerates skill acquisition in children.
• Sleep Deprivation & Cognitive Performance: Lack of adequate sleep impairs cognitive performance in adults.
• Social Isolation & Depression: Prolonged social isolation is a significant cause of depression among teenagers.
• Mindfulness Meditation & Stress Reduction: Regular practice of mindfulness meditation reduces symptoms of stress and anxiety.
• Peer Pressure & Adolescent Risk Taking: Peer pressure significantly increases risk-taking behaviors among adolescents.
• Parenting Styles & Child’s Self-esteem: Authoritarian parenting styles negatively impact a child’s self-esteem.
• Multitasking & Attention Span: Engaging in multitasking frequently leads to a reduced attention span.
• Childhood Bullying & Adult PTSD: Individuals bullied during childhood have a higher likelihood of developing PTSD as adults.
• Digital Screen Time & Child Development: Excessive digital screen time impairs cognitive and social development in children.

Causal Inference Hypothesis Statement Examples

Causal inference is about deducing the cause-effect relationship between two variables after considering potential confounders. These hypotheses aim to find direct relationships even when other influencing factors are present.

• Dietary Habits & Chronic Illnesses: Even when considering genetic factors, unhealthy dietary habits increase the chances of chronic illnesses.
• Exercise & Mental Well-being: When accounting for daily stressors, regular exercise improves mental well-being.
• Job Satisfaction & Employee Turnover: Even when considering market conditions, job satisfaction inversely relates to employee turnover.
• Financial Literacy & Savings Behavior: When considering income levels, financial literacy is directly linked to better savings behavior.
• Online Reviews & Product Sales: Even accounting for advertising spends, positive online reviews boost product sales.
• Prenatal Care & Child Health Outcomes: When considering genetic factors, adequate prenatal care ensures better health outcomes for children.
• Teacher Qualifications & Student Performance: Accounting for socio-economic factors, teacher qualifications directly influence student performance.
• Community Engagement & Crime Rates: When considering economic conditions, higher community engagement leads to lower crime rates.
• Eco-friendly Practices & Brand Loyalty: Accounting for product quality, eco-friendly business practices boost brand loyalty.
• Mental Health Support & Workplace Productivity: Even when considering workload, providing mental health support enhances workplace productivity.

What are the Characteristics of Causal Hypothesis

Causal hypotheses are foundational in many research disciplines, as they predict a cause-and-effect relationship between variables. Their unique characteristics include:

• Cause-and-Effect Relationship: The core of a causal hypothesis is to establish a direct relationship, indicating that one variable (the cause) will bring about a change in another variable (the effect).
• Testability: They are formulated in a manner that allows them to be empirically tested using appropriate experimental or observational methods.
• Specificity: Causal hypotheses should be specific, delineating clear cause and effect variables.
• Directionality: They typically demonstrate a clear direction in which the cause leads to the effect.
• Operational Definitions: They often use operational definitions, which specify the procedures used to measure or manipulate variables.
• Temporal Precedence: The cause (independent variable) always precedes the effect (dependent variable) in time.

What is a causal hypothesis in research?

In research, a causal hypothesis is a statement about the expected relationship between variables, or explanation of an occurrence, that is clear, specific, testable, and falsifiable. It suggests a relationship in which a change in one variable is the direct cause of a change in another variable. For instance, “A higher intake of Vitamin C reduces the risk of common cold.” Here, Vitamin C intake is the independent variable, and the risk of common cold is the dependent variable.

What is the difference between causal and descriptive hypothesis?

• Causal Hypothesis: Predicts a cause-and-effect relationship between two or more variables.
• Descriptive Hypothesis: Describes an occurrence, detailing the characteristics or form of a particular phenomenon.
• Causal: Consuming too much sugar can lead to diabetes.
• Descriptive: 60% of adults in the city exercise at least thrice a week.
• Causal: To establish a causal connection between variables.
• Descriptive: To give an accurate portrayal of the situation or fact.
• Causal: Often involves experiments.
• Descriptive: Often involves surveys or observational studies.

How do you write a Causal Hypothesis? – A Step by Step Guide

• Identify Your Variables: Pinpoint the cause (independent variable) and the effect (dependent variable). For instance, in studying the relationship between smoking and lung health, smoking is the independent variable while lung health is the dependent variable.
• State the Relationship: Clearly define how one variable affects another. Does an increase in the independent variable lead to an increase or decrease in the dependent variable?
• Be Specific: Avoid vague terms. Instead of saying “improved health,” specify the type of improvement like “reduced risk of cardiovascular diseases.”
• Use Operational Definitions: Clearly define any terms or variables in your hypothesis. For instance, define what you mean by “regular exercise” or “high sugar intake.”
• Ensure It’s Testable: Your hypothesis should be structured so that it can be disproven or supported by data.
• Review Existing Literature: Check previous research to ensure that your hypothesis hasn’t already been tested, and to ensure it’s plausible based on existing knowledge.
• Draft Your Hypothesis: Combine all the above steps to write a clear, concise hypothesis. For instance: “Regular exercise (defined as 150 minutes of moderate exercise per week) decreases the risk of cardiovascular diseases.”

Tips for Writing Causal Hypothesis

• Simplicity is Key: The clearer and more concise your hypothesis, the easier it will be to test.
• Avoid Absolutes: Using words like “all” or “always” can be problematic. Few things are universally true.
• Seek Feedback: Before finalizing your hypothesis, get feedback from peers or mentors.
• Stay Objective: Base your hypothesis on existing literature and knowledge, not on personal beliefs or biases.
• Revise as Needed: As you delve deeper into your research, you may find the need to refine your hypothesis for clarity or specificity.
• Falsifiability: Always ensure your hypothesis can be proven wrong. If it can’t be disproven, it can’t be validated either.
• Avoid Circular Reasoning: Ensure that your hypothesis doesn’t assume what it’s trying to prove. For example, “People who are happy have a positive outlook on life” is a circular statement.
• Specify Direction: In causal hypotheses, indicating the direction of the relationship can be beneficial, such as “increases,” “decreases,” or “leads to.”

AI Generator

Text prompt

• Instructive
• Professional

10 Examples of Public speaking

20 Examples of Gas lighting

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base

Methodology

• Correlation vs. Causation | Difference, Designs & Examples

Correlation vs. Causation | Difference, Designs & Examples

Published on July 12, 2021 by Pritha Bhandari . Revised on June 22, 2023.

Correlation means there is a statistical association between variables. Causation means that a change in one variable causes a change in another variable.

In research, you might have come across the phrase “correlation doesn’t imply causation.” Correlation and causation are two related ideas, but understanding their differences will help you critically evaluate sources and interpret scientific research.

Table of contents

What’s the difference, why doesn’t correlation mean causation, correlational research, third variable problem, regression to the mean, spurious correlations, directionality problem, causal research, other interesting articles, frequently asked questions about correlation and causation.

Correlation describes an association between types of variables : when one variable changes, so does the other. A correlation is a statistical indicator of the relationship between variables. These variables change together: they covary. But this covariation isn’t necessarily due to a direct or indirect causal link.

Causation means that changes in one variable brings about changes in the other; there is a cause-and-effect relationship between variables. The two variables are correlated with each other and there is also a causal link between them.

Prevent plagiarism. Run a free check.

There are two main reasons why correlation isn’t causation. These problems are important to identify for drawing sound scientific conclusions from research.

The third variable problem means that a confounding variable affects both variables to make them seem causally related when they are not. For example, ice cream sales and violent crime rates are closely correlated, but they are not causally linked with each other. Instead, hot temperatures, a third variable, affects both variables separately. Failing to account for third variables can lead research biases to creep into your work.

The directionality problem occurs when two variables correlate and might actually have a causal relationship, but it’s impossible to conclude which variable causes changes in the other. For example, vitamin D levels are correlated with depression, but it’s not clear whether low vitamin D causes depression, or whether depression causes reduced vitamin D intake.

You’ll need to use an appropriate research design to distinguish between correlational and causal relationships:

• Correlational research designs can only demonstrate correlational links between variables.
• Experimental designs can test causation.

In a correlational research design, you collect data on your variables without manipulating them.

Correlational research is usually high in external validity , so you can generalize your findings to real life settings. But these studies are low in internal validity , which makes it difficult to causally connect changes in one variable to changes in the other.

These research designs are commonly used when it’s unethical, too costly, or too difficult to perform controlled experiments. They are also used to study relationships that aren’t expected to be causal.

Without controlled experiments, it’s hard to say whether it was the variable you’re interested in that caused changes in another variable. Extraneous variables are any third variable or omitted variable other than your variables of interest that could affect your results.

Limited control in correlational research means that extraneous or confounding variables serve as alternative explanations for the results. Confounding variables can make it seem as though a correlational relationship is causal when it isn’t.

When two variables are correlated, all you can say is that changes in one variable occur alongside changes in the other.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

Regression to the mean is observed when variables that are extremely higher or extremely lower than average on the first measurement move closer to the average on the second measurement. Particularly in research that intentionally focuses on the most extreme cases or events, RTM should always be considered as a possible cause of an observed change.

Players or teams featured on the cover of SI have earned their place by performing exceptionally well. But athletic success is a mix of skill and luck, and even the best players don’t always win.

Chances are that good luck will not continue indefinitely, and neither can exceptional success.

A spurious correlation is when two variables appear to be related through hidden third variables or simply by coincidence.

The Theory of the Stork draws a simple causal link between the variables to argue that storks physically deliver babies. This satirical study shows why you can’t conclude causation from correlational research alone.

When you analyze correlations in a large dataset with many variables, the chances of finding at least one statistically significant result are high. In this case, you’re more likely to make a type I error . This means erroneously concluding there is a true correlation between variables in the population based on skewed sample data.

To demonstrate causation, you need to show a directional relationship with no alternative explanations. This relationship can be unidirectional, with one variable impacting the other, or bidirectional, where both variables impact each other.

A correlational design won’t be able to distinguish between any of these possibilities, but an experimental design can test each possible direction, one at a time.

• Physical activity may affect self esteem
• Self esteem may affect physical activity
• Physical activity and self esteem may both affect each other

In correlational research, the directionality of a relationship is unclear because there is limited researcher control. You might risk concluding reverse causality, the wrong direction of the relationship.

Causal links between variables can only be truly demonstrated with controlled experiments . Experiments test formal predictions, called hypotheses , to establish causality in one direction at a time.

Experiments are high in internal validity , so cause-and-effect relationships can be demonstrated with reasonable confidence.

You can establish directionality in one direction because you manipulate an independent variable before measuring the change in a dependent variable.

In a controlled experiment, you can also eliminate the influence of third variables by using random assignment and control groups.

Random assignment helps distribute participant characteristics evenly between groups so that they’re similar and comparable. A control group lets you compare the experimental manipulation to a similar treatment or no treatment (or a placebo, to control for the placebo effect ).

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Chi square test of independence
• Statistical power
• Descriptive statistics
• Degrees of freedom
• Pearson correlation
• Null hypothesis
• Double-blind study
• Case-control study
• Research ethics
• Data collection
• Hypothesis testing
• Structured interviews

Research bias

• Hawthorne effect
• Unconscious bias
• Recall bias
• Halo effect
• Self-serving bias
• Information bias

A correlation reflects the strength and/or direction of the association between two or more variables.

• A positive correlation means that both variables change in the same direction.
• A negative correlation means that the variables change in opposite directions.
• A zero correlation means there’s no relationship between the variables.

Correlation describes an association between variables : when one variable changes, so does the other. A correlation is a statistical indicator of the relationship between variables.

Causation means that changes in one variable brings about changes in the other (i.e., there is a cause-and-effect relationship between variables). The two variables are correlated with each other, and there’s also a causal link between them.

While causation and correlation can exist simultaneously, correlation does not imply causation. In other words, correlation is simply a relationship where A relates to B—but A doesn’t necessarily cause B to happen (or vice versa). Mistaking correlation for causation is a common error and can lead to false cause fallacy .

The third variable and directionality problems are two main reasons why correlation isn’t causation .

The third variable problem means that a confounding variable affects both variables to make them seem causally related when they are not.

The directionality problem is when two variables correlate and might actually have a causal relationship, but it’s impossible to conclude which variable causes changes in the other.

Controlled experiments establish causality, whereas correlational studies only show associations between variables.

• In an experimental design , you manipulate an independent variable and measure its effect on a dependent variable. Other variables are controlled so they can’t impact the results.
• In a correlational design , you measure variables without manipulating any of them. You can test whether your variables change together, but you can’t be sure that one variable caused a change in another.

In general, correlational research is high in external validity while experimental research is high in internal validity .

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

Bhandari, P. (2023, June 22). Correlation vs. Causation | Difference, Designs & Examples. Scribbr. Retrieved April 15, 2024, from https://www.scribbr.com/methodology/correlation-vs-causation/

Is this article helpful?

Pritha Bhandari

Other students also liked, correlational research | when & how to use, guide to experimental design | overview, steps, & examples, confounding variables | definition, examples & controls, what is your plagiarism score.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 17 April 2024

Refining the impact of genetic evidence on clinical success

• Eric Vallabh Minikel   ORCID: orcid.org/0000-0003-2206-1608 1 ,
• Jeffery L. Painter   ORCID: orcid.org/0000-0001-9651-9904 2   nAff5 ,
• Coco Chengliang Dong 3 &
• Matthew R. Nelson   ORCID: orcid.org/0000-0001-5089-5867 3 , 4  

Nature ( 2024 ) Cite this article

Metrics details

• Drug development
• Genetic predisposition to disease
• Genome-wide association studies
• Target validation

The cost of drug discovery and development is driven primarily by failure 1 , with only about 10% of clinical programmes eventually receiving approval 2 , 3 , 4 . We previously estimated that human genetic evidence doubles the success rate from clinical development to approval 5 . In this study we leverage the growth in genetic evidence over the past decade to better understand the characteristics that distinguish clinical success and failure. We estimate the probability of success for drug mechanisms with genetic support is 2.6 times greater than those without. This relative success varies among therapy areas and development phases, and improves with increasing confidence in the causal gene, but is largely unaffected by genetic effect size, minor allele frequency or year of discovery. These results indicate we are far from reaching peak genetic insights to aid the discovery of targets for more effective drugs.

Similar content being viewed by others

Improving the odds of drug development success through human genomics: modelling study

Aroon D. Hingorani, Valerie Kuan, … Juan Pablo Casas

From target discovery to clinical drug development with human genetics

Katerina Trajanoska, Claude Bhérer, … Vincent Mooser

Advancing the use of genome-wide association studies for drug repurposing

William R. Reay & Murray J. Cairns

Human genetics is one of the only forms of scientific evidence that can demonstrate the causal role of genes in human disease. It provides a crucial tool for identifying and prioritizing potential drug targets, providing insights into the expected effect (or lack thereof 6 ) of pharmacological engagement, dose–response relationships 7 , 8 , 9 , 10 and safety risks 6 , 11 , 12 , 13 . Nonetheless, many questions remain about the application of human genetics in drug discovery. Genome-wide association studies (GWASs) of common, complex traits, including many diseases, generally identify variants of small effect. This contributed to early scepticism of the value of GWASs 14 . Anecdotally, such variants can point to highly successful drug targets 7 , 8 , 9 , and yet, genetic support from GWASs is somewhat less predictive of drug target advancement than support from Mendelian diseases 5 , 15 .

In this paper we investigate several open questions regarding the use of genetic evidence for prioritizing drug discovery. We explore the characteristics of genetic associations that are more likely to differentiate successful from unsuccessful drug mechanisms, exploring how they differ across therapy areas and among discovery and development phases. We also investigate how close we may be to saturating the insights we can gain from genetic studies for drug discovery and how much of the genetically supported drug discovery space remains clinically unexplored.

To characterize the drug development pipeline, we filtered Citeline Pharmaprojects for monotherapy programmes added since 2000 annotated with a highest phase reached and assigned both a human gene target (usually the gene encoding the drug target protein) and an indication defined in Medical Subject Headings (MeSH) ontology. This resulted in 29,476 target–indication (T–I) pairs for analysis (Extended Data Fig. 1a ). Multiple sources of human genetic associations totalled 81,939 unique gene–trait (G–T) pairs, with traits also mapped to MeSH terms. Intersection of these datasets yielded an overlap of 2,166 T–I and G–T pairs (7.3%) for which the indication and the trait MeSH terms had a similarity ≥0.8; we defined these T–I pairs as possessing genetic support (Extended Data Figs. 1b and 2a and  Methods ). The probability of having genetic support, or P(G), was higher for launched T–I pairs than those in historical or active clinical development (Fig. 1a ). In each phase, P(G) was higher than previously reported 5 , 15 , owing, as expected 15 , 16 , more to new G–T discoveries than to changes in drug pipeline composition (Extended Data Fig. 3a–f ). For ensuing analyses, we considered both historical and active programmes. We defined success at each phase as a T–I pair transitioning to the next development phase (for example, from phase I to II), and we also considered overall success—advancing from phase I to a launched drug. We defined relative success (RS) as the ratio of the probability of success, P(S), with genetic support to the probability of success without genetic support ( Methods ). We tested the sensitivity of RS to various characteristics of genetic evidence. RS was sensitive to the indication–trait similarity threshold (Extended Data Fig. 2a ), which we set to 0.8 for all analyses herein. RS was >2 for all sources of human genetic evidence examined (Fig. 1b ). RS was highest for Online Mendelian Inheritance in Man (OMIM) (RS = 3.7), in agreement with previous reports 5 , 15 ; this was not the result of a higher success rate for orphan drug programmes (Extended Data Fig. 2b ), a designation commonly acquired for rare diseases. Rather, it may owe partly to the difference in confidence in causal gene assignment between Mendelian conditions and GWASs, supported by the observation that the RS for Open Targets Genetics (OTG) associations was sensitive to the confidence in variant-to-gene mapping as reflected in the minimum share of locus-to-gene (L2G) score (Fig. 1c ). The differences common and rare disease programmes face in regulatory and reimbursement environments 4 and differing proportions of drug modalities 9 probably contribute as well. OMIM and GWAS support were synergistic with one another (Supplementary Fig. 2b ). Somatic evidence from IntOGen had an RS of 2.3 in oncology (Extended Data Fig. 2c ), similar to GWASs, but analyses below are limited to germline genetic evidence unless otherwise noted.

a , Proportion of T–I pairs with genetic support, P(G), as a function of highest phase reached. n at right: denominator, number of T–I pairs per phase; numerator, number that are genetically supported. b , Sensitivity of phase I–launch RS to source of human genetic association. GWAS Catalog, Neale UKBB and FinnGen are subsets of OTG. n at right: denominator, number of T–I pairs with genetic support from each source; numerator, number of those launched. Note that RS is calculated from a 2 × 2 contingency table ( Methods ). Total n  = 13,022 T–I pairs. c , Sensitivity of RS to L2G share threshold among OTG associations. Minimum L2G share threshold is varied from 0.1 to 1.0 in increments of 0.05 (labels); RS ( y axis) is plotted against the number of clinical (phase I+) programmes with genetic support from OTG ( x axis). d , Sensitivity of RS for OTG GWAS-supported T–I pairs to binned variables: (1) year that T–I pair first acquired human genetic support from GWASs, excluding replications and excluding T–I pairs otherwise supported by OMIM; (2) number of genes exhibiting genetic association to the same trait; (3) quartile of effect size (beta) for quantitative traits; (4) quartile of effect size (odds ratio, OR) for case/control traits standardized to be >1 (that is, 1/OR if <1); (5) order of magnitude of minor allele frequency bins. n at right as in b . Total n  = 13,022 T–I pairs. e , Count of indications ever developed in Pharmaprojects ( y axis) by the number of genes associated with traits similar to those indications ( x axis). Throughout, error bars or shaded areas represent 95% CIs (Wilson for P(G) and Katz for RS) whereas centres represent point estimates. See Supplementary Fig. 1 for the same analyses restricted to drugs with a single known target.

Source Data

As sample sizes grow ever larger with a corresponding increase in the number of unique G–T associations, some expect 17 the value of GWAS genetic findings to become less useful for the purpose of drug target selection. We explored this in several ways. We investigated the year that genetic support for a T–I pair was first discovered, under the expectation that more common and larger effects are discovered earlier. Although there was a slightly higher RS for discoveries from 2007–2010 that was largely driven by early lipid and cardiovascular-related associations, the effect of year was overall non-significant ( P  = 0.46; Fig. 1d ). Results were similar when replicate associations or OMIM discoveries were included (Extended Data Fig. 2d–f ). We next divided up GWAS-supported drug programmes by the number of unique traits associated to each gene. RS nominally increased with the number of associated genes, by 0.048 per gene ( P  = 0.024; Fig. 1d ). The reason is probably not that successful genetically supported programmes inspire other programmes, because most genetic support was discovered retrospectively (Extended Data Fig. 2g ); the few examples of drug programmes prospectively motivated by genetic evidence were primarily for Mendelian diseases 9 . There were no statistically significant associations with estimated effect sizes ( P  = 0.90 and 0.57, for quantitative and binary traits, respectively; Fig. 1d and Extended Data Fig. 2h ) or minor allele frequency ( P  = 0.26; Fig. 1d ). That ever larger GWASs can continue to uncover support for successful targets is also illustrated by two recent large GWASs in type 2 diabetes (T2D) 18 , 19 (Extended Data Fig. 4 ).

Previously 5 , we observed significant heterogeneity among therapy areas in the fraction of approved drug mechanisms with genetic support, but did not investigate the impact on probability of success 5 . Here, our estimates of RS from phase I to launch showed significant heterogeneity ( P  < 1.0 × 10 −15 ), with nearly all therapy areas having estimates greater than 1; 11 of 17 were >2, and haematology, metabolic, respiratory and endocrine >3 (Fig. 2a–e ). In most therapy areas, the impact of genetic evidence was most pronounced in phases II and III and least impactful in phase I, corresponding to capacity to demonstrate clinical efficacy in later development phases. Accordingly, therapy areas differed in P(G) and in whether P(G) increased throughout clinical development or only at launch (Extended Data Fig. 5 ); data source and other properties of genetic evidence including year of discovery and effect size also differed (Extended Data Fig. 6 ). We also found that genetic evidence differentiated likelihood to progress from preclinical to clinical development for metabolic diseases (RS = 1.38; 95% confidence interval (95% CI), 1.25 to 1.54), which may reflect preclinical models that are more predictive of clinical outcomes. P(G) by therapy area was correlated with P(S) ( ρ  = 0.59, P  = 0.013) and with RS ( ρ  = 0.72, P  = 0.0011; Extended Data Fig. 7 ), which led us to explore how the sheer quantity of genetic evidence available within therapy areas (Fig. 2f and Extended Data Fig. 8a ) may influence this. We found that therapy areas with more possible gene–indication (G–I) pairs supported by genetic evidence had significantly higher RS ( ρ  = 0.71, P  = 0.0010; Fig. 2g ), although respiratory and endocrine were notable outliers with high RS despite fewer associations.

a – e , RS by therapy area and phase transitions: preclinical to phase I ( a ), phase I to II ( b ), phase II to III ( c ), phase III to launch ( d ) and phase I to launch ( e ). n at right: denominator, T–I pairs with genetic support; numerator, number of those that succeeded in the phase transition indicated at the top of the panel. For ‘all’, total n  = 22,638 preclinical, 13,022 reaching at least phase I, 7,223 reaching at least phase II and 2,184 reaching at least phase III. Total n for each therapy area is provided in Supplementary Table 27 . f , Cumulative number of possible genetically supported G–I pairs in each therapy ( y axis) as genetic discoveries have accrued over time ( x axis). g , RS ( y axis) by number of possible supported G–I pairs ( x axis) across therapy areas, with dots coloured as in panels a – e and sized according to number of genetically supported T–I pairs in at least phase I. h , Number of launched indications versus similarity of those indications, by approved drug target. i , Proportion of launched T–I pairs with genetic support, P(G), binned by quintile of the number of launched indications per target (top panel) or by mean similarity among launched indications (bottom panel). Targets with exactly 1 launched indication (6.2% of launched T–I pairs) are considered to have mean similarity of 1.0. n at right: denominator, total number of launched T–I pairs in each bin; numerator, number of those with genetic support. j , RS ( y axis) versus mean similarity among launched indications per target ( x axis) by therapy area. k , RS ( y axis) versus mean count of launched indications per target ( x axis). Throughout, error bars or shaded areas represent 95% CIs (Wilson for P(G) and Katz for RS) whereas centres represent point estimates. See Supplementary Fig. 2 for the same analyses restricted to drugs with a single known target.

We hypothesized that genetic support might be most pronounced for drug mechanisms with disease-modifying effects, as opposed to those that manage symptoms, and that the proportions of such drugs differ by therapy area 20 , 21 . We were unable to find data with these descriptions available for a sufficient number of drug mechanisms to analyse, but we reasoned that targets of disease-modifying drugs are more likely to be specific to a disease, whereas targets of symptom-managing drugs are more likely to be applied across many indications. We therefore examined the number and diversity of all-time launched indications per target. Launched T–I pairs are heavily skewed towards a few targets (Fig. 2h ). Of 450 launched targets, the 42 with ≥10 launched indications comprise 713 (39%) of 1,806 launched T–I pairs (Fig. 2h ). Many of these are used across diverse indications for management of symptoms such as inflammatory and immune responses ( NR3C1 , IFNAR2 ), pain ( PTGS2 , OPRM1 ), mood ( SLC6A4 ) or parasympathetic response ( CHRM3 ). The count of launched indications was inversely correlated with the mean similarity of those indications ( ρ  = −0.72, P  = 4.4 × 10 −84 ; Fig. 2h ). Among T–I pairs, the probability of having genetic support increased as the number of launched indications decreased ( P  = 6.3 × 10 −7 ) and as the similarity of a target’s launched indications increased ( P  = 1.8 × 10 −5 ; Fig. 2i ). We observed a corresponding impact on RS, increasing in therapy areas for which the similarity among launched indications increased, and decreasing with increasing indications per target ( ρ  = 0.74, P  = 0.0010, and ρ  = −0.62, P  = 0.0080, respectively; Fig. 2j,k ).

Only 4.8% (284 of 5,968) of T–I pairs active in phases I–III possess human germline genetic support (Fig. 1a ), similar to T–I pairs no longer in development (4.2%, 560 of 13,355), a difference that was not statistically significant ( P  = 0.080). We estimated ( Methods ) that only 1.1% of all genetically supported G–I relationships have been explored clinically (Fig. 3a ), or 2.1% when restricting to the most similar indication. Given that the vast majority of proteins are classically ‘undruggable’, we explored the proportion of genetically supported G–I pairs that had been developed to at least phase I, as a function of therapy area across several classes of tractability and relevant protein families 22 (Fig. 3a ). Within therapy areas, oncology kinases with germline evidence were the most saturated: 109 of 250 (44%) of all genetically supported G–I pairs had reached at least phase I; GPCRs for psychiatric indications were also notable (14 of 53, 26%). Grouping by target rather than G–I pair, 3.6% of genetically supported targets have been pursued for any genetically supported indication (Extended Data Fig. 8 ). Of possible genetically supported G–I pairs, most (68%) arose from OTG associations, mostly in the past 5 years (Fig. 2f ). Such low use is partly due to recent emergence of most genetic evidence (Extended Data Figs. 2f,g and 7a ), as drug programmes prospectively supported by human genetics have had a mean lag time from genetic association of 13 years to first trial 21 and 21 years to approval 9 . Because some types of targets may be more readily tractable by antagonists than agonists, we also grouped by target and examined human genetic evidence by direction of effect for tumour suppressors versus oncogenes (Fig. 3b ), identifying a few substrata for which a majority of genetically supported targets had been pursued to at least phase I for at least one genetically supported indication. Oncogene kinases received the most attention, with 19 of 25 (76%) reaching phase I.

a , Heatmap of proportion of genetically supported T–I pairs that have been developed to at least phase I, by therapy area ( y axis) and gene list ( x axis). b , As panel a , but for genetic support from IntOGen rather than germline sources and grouped by the direction of effect of the gene according to IntOGen ( y axis), and also grouped by target rather than T–I pair. Thus, the denominator for each cell is the number of targets with at least one genetically supported indication, and each target counts towards the numerator if at least one genetically supported indication has reached phase I. c , Of targets that have reached phase I for any indication, and have at least one genetically supported indication, the mean count ( x axis) of genetically supported (left) and unsupported (right) indications pursued, binned by the number of possible genetically supported indications ( y axis). The centre is the mean and bars are Wilson 95% CIs. n  = 1,147 targets. d , Proportion of D–I pairs with genetic support, P(G) ( x axis), as a function of each D–I pair’s phase reached (inner y -axis grouping) and the drug’s highest phase reached for any indication (outer y -axis grouping). The centre is the exact proportion and bars are Wilson 95% CIs. The n is indicated at the right, for which the denominator is the total number of D–I pairs in each bin, and the numerator is the number of those that are genetically supported. See Supplementary Fig. 3 for the same analyses restricted to drugs with a single known target. Ab, antibody; SM, small molecule.

To focus on demonstrably druggable proteins, we further restricted the analysis to targets with both (1) any programme reaching phase I, and (2) ≥1 genetically supported indications. Of 1,147 qualifying targets, only 373 (33%) had been pursued for one or more supported indications (Fig. 3c ), and most (307, 27%) of these targets were pursued for indications both with and without genetic support. Overall, an overwhelming majority of development effort has been for unsupported indications, at a 17:1 ratio. Within this subset of targets, we asked whether genetic support was predictive of which indications would advance the furthest. Grouping active and historical programmes by drug–indication (D–I) pair, we found that the odds of advancing to a later stage in the pipeline are 82% higher for indications with genetic support ( P  = 8.6 × 10 −73 ; Fig. 3d ).

Although there has been anecdotal support—such as the HMGCR example—to argue that genetic effect size may not matter in prioritizing drug targets, here we provide systematic evidence that small effect size, recent year of discovery, increasing number of genes identified or higher associated allele frequency do not diminish the value of GWAS evidence to differentiate clinical success rates. One reason for this is probably because genetic effect size on a phenotype rarely accounts for the magnitude of genetic effect on gene expression, protein function or some other molecular intermediate. In some circumstances, genetic effect sizes can yield insights into anticipated drug effects. This is best illustrated for cardiovascular disease therapies, for which genetic effects on cholesterol and disease risk and treatment outcomes are correlated 23 . A limitation is that, other than Genebass, we did not include whole exome or whole genome sequencing association studies, which may be more likely to pinpoint causal variants. Moreover, all of our analyses are naive to direction of genetic effect (gain versus loss of gene function) as this is unknown or unannotated in most datasets used here.

Our results argue for continuing investment to expand GWAS-like evidence, particularly for many complex diseases with treatment options that fail to modify disease. Although genetic evidence has value across most therapy areas, its benefit is more pronounced in some areas than others. Furthermore, it is possible that the therapy areas for which genetic evidence had a lower impact have seen more focus on symptom management. If so, we would predict that for drugs aimed at disease modification, human genetics should ultimately prove highly valuable across therapy areas.

The focus of this work has been on the RS of drug programmes with and without genetic evidence, limited to drug mechanisms that have entered clinical development. This metric does not address the probability that a gene associated with a disease, if targeted, will yield a successful drug. At the early stage of target selection, is evidence of a large loss-of-function effect in one gene usually a better choice than a small non-coding single nucleotide polymorphism (SNP) effect on the same phenotype in another? We explored this question for T2D studies referenced above. When these GWASs quadrupled the number of T2D-associated genes from 217 to 862, new genetic support was identified for 7 of 95 mechanisms in clinical development whereas the number supported increased from 5 to 7 of 12 launched drug mechanisms. Thus, RS has remained high in light of new GWAS data. One can also, however, consider the proportion of genetic associations that are successful drug targets. Of the 7 targets of launched drugs with genetic evidence, 4 had Mendelian evidence (in addition to pre-2020 GWAS evidence), out of a total of 19 Mendelian genes related to T2D (21%). One launched T2D target had only GWAS (and no Mendelian) evidence among 217 GWAS-associated genes before 2020 (0.46%), whereas 2 launched targets were among 645 new GWAS associations since 2020 (0.31%). At least in this example, the ‘yield’ of genetic evidence for successful drug mechanisms was greatest for genes with Mendelian effects, but similar between earlier and later GWASs. Clearly, just because genetic associations differentiate clinical stage drug targets from launched ones, does not mean that a large fraction of associations will be fruitful. Moreover, genetically supported targets may be more likely to require upregulation, to be druggable only by more challenging modalities 4 , 9 or to enjoy narrower use across indications. More work is required to better understand the challenges of target identification and prioritization given the genetic evidence precondition.

The utility of human genetic evidence in drug discovery has had firm theoretical and empirical footing for several years 5 , 7 , 15 . If the benefit of this evidence were cancelled out by competitive crowding 24 , then currently active clinical phases should have higher rates of genetic support than their corresponding historical phases, and might look similar to, or even higher than, launched pairs. Instead, we find that active programmes possess genetic support only slightly more often than historical programmes and remain less enriched for genetic support than launched drugs. Meanwhile, only a tiny fraction of classically druggable genetically supported G–I pairs have been pursued even among targets with clinical development reported. Human genetics thus represents a growing opportunity for novel target selection and improving indication selection for existing drugs and drug candidates. Increasing emphasis on drug mechanisms with supporting genetic evidence is expected to increase success rates and lower the cost of drug discovery and development.

Definition of metrics

Except where otherwise noted, we define genetic support of a drug mechanism (that is, a T–I pair) as a genetic association mapped to the corresponding target gene for a trait that is ≥0.8 similar to the indication (see MeSH term similarity below). We defined P(G) as the proportion of drug mechanisms satisfying the above definition of genetic support. P(S) is the proportion of programmes in one phase that advance to a subsequent phase (for instance, phase I to phase II). Overall P(S) from phase I to launched is the product of P(S) at each individual phase. RS is the ratio of P(S) for programmes with genetic support to P(S) for programmes lacking genetic support, which is equivalent to a relative risk or risk ratio. Thus, if N denotes the total number of programmes that have reached the reference phase, and X denotes the number of those that advance to a later phase of interest, and the subscripts G and!G indicate the presence or absence of genetic support, then P(G) =  N G /( N G  +  N !G ); P(S) = ( X G  +  X !G )/( N G  +  N !G ); RS = ( X G / N G )/( X !G / N !G ). RS from phase I to launched is the product of RS at each individual phase. The count of ‘programs’ for X and N is T–I pairs throughout, except for Fig. 3d , which uses D–I pairs to specifically interrogate P(G) for which the same drug has been developed for different indications. For clarity, we note that whereas other recent studies 22 , 25 have examined the fold enrichment and overlap between genes with a human genetic support and genes encoding a drug target, without regard to similarity, herein all of our analyses are conditioned on the similarity between the drug’s indication and the genetically associated trait.

Drug development pipeline

Citeline Pharmaprojects 26 is a curated database of drug development programmes including preclinical, all clinical phases and launched (approved and marketed) drugs. It was queried via API (22 December 2022) to obtain information on drugs, targets, indications, phases reached and current development status. T–I pair was the unit of analysis throughout, except where otherwise indicated in the text (D–I pairs were examined in Fig. 3d ). Current development status was defined as ‘active’ if the T–I pair had at least one drug still in active development, and ‘historical’ if development of all drugs for the T–I pair had ceased. Targets were defined as genes; as most drugs do not directly target DNA, this usually refers to the gene encoding the protein target that is bound or modulated by the drug. We removed combination therapies, diagnostic indication and programmes with no human target or no indication assigned. For most analyses, only programmes added to the database since 2000 were included, whereas for the count and similarity of launched indications per target, we used all launches for all time. Indications were considered to possess ‘genetic insight’—meaning the human genetics of this trait or similar traits have been successfully studied—if they had ≥0.8 similarity to (1) an OMIM or IntOGen disease, or (2) a GWAS trait with at least 3 independently associated loci, on the basis of lead SNP positions rounded to the nearest 1 megabase. For calculating RS, we used the number of T–I pairs with genetic insight as the denominator. The rationale for this choice is to focus on indications for which there exists the opportunity for human genetic evidence, consistent with the filter applied previously 5 . However, we observe that our findings are not especially sensitive to the presence of this filter, with RS decreasing by just 0.17 when the filter is removed (Extended Data Fig. 3g,h ). Note that the criteria for determining genetic insight are distinct from, and much looser than, the criteria for mapping GWAS hits to genes (see L2G scores under OTG below). Many drugs had more than one target assigned, in which case all targets were retained for T–I pair analyses. As a sensitivity test, running our analyses restricted to only drugs with exactly one target assigned yielded very similar results ( Supplementary Figures ).

OMIM is a curated database of Mendelian gene–disease associations. The OMIM Gene Map (downloaded 21 September 2023) contained 8,671 unique gene–phenotype links. We restricted to entries with phenotype mapping code 3 (‘the molecular basis for the disorder is known; a mutation has been found in the gene’), removed phenotypes with no MIM number or no gene symbol assigned, and removed duplicate combinations of gene MIM and phenotype MIM. We used regular expression matching to further filter out phenotypes containing the terms ‘somatic’, ‘susceptibility’ or ‘response’ (drug response associations) and those flagged as questionable (‘?’), or representing non-disease phenotypes (‘[’). A set of OMIM phenotypes are flagged as denoting susceptibility rather than causation (‘{’); this category includes low-penetrance or high allele frequency association assertions that we wished to exclude, but also germline heterozygous loss-of-function mutations in tumour suppressor genes, for which the underlying mechanism of disease initiation is loss of heterozygosity, which we wished to include. We therefore also filtered out phenotypes containing ‘{’ except for those that did contain the terms ‘cancer’, ‘neoplasm’, ‘tumor’ or ‘malignant’ and did not contain the term ‘somatic’. Remaining entries present in OMIM as of 2021 were further evaluated for validity by two curators, and gene–disease combinations for which a disease association was deemed not to have been established were excluded from all analyses. All of the above filters left 5,670 unique G–T links. MeSH terms for OMIM phenotypes were then mapped using the EFO OWL database using an approach previously described 27 , with further mappings from Orphanet, full text matches to the full MeSH vocabulary and, finally, manual curation, for a cumulative mapping rate of 93% (5,297 of 5,670). Because sometimes distinct phenotype MIM numbers mapped to the same MeSH term, this yielded 4,510 unique gene–MeSH links.

OTG is a database of GWAS hits from published studies and biobanks. OTG version 8 (12 October 2022) variant-to-disease, L2G, variant index and study index data were downloaded from EBI. Traits with multiple EFO IDs were excluded as these generally represent conditional, epistasis or other complex phenotypes that would lack mappings in the MeSH vocabulary. Of the top 100 traits with the greatest number of genes mapped, we excluded 76 as having no clear disease relevance (for example, ‘red cell distribution width’) or no obvious marginal value (for example, excluded ‘trunk predicted mass’ because ‘body mass index’ was already included). Remaining traits were mapped to MeSH using the EFO OWL database, full text queries to the MeSH API, mappings already manually curated in PICCOLO (see below) or new manual curation. In total, 25,124 of 49,599 unique traits (51%) were successfully mapped to a MeSH ID. We included associations with P  < 5 × 10 −8 . OTG L2G scores used for gene mapping are based on a machine learning model trained on gold standard causal genes 28 ; inputs to that model include distance, functional annotations, expression quantitative trait loci (eQTLs) and chromatin interactions. Note that we do not use Mendelian randomization 29 to map causal genes, and even gene mappings with high L2G scores are necessarily imperfect. OTG provides an L2G score for the triplet of each study or trait with each hit and each possible causal gene. We defined L2G share as the proportion of the total L2G score assigned each gene among all potentially causal genes for that trait–hit combination. In sensitivity analyses we considered L2G share thresholds from 10% to 100% (Fig. 1b and Extended Data Fig. 3a ), but main analyses used only genes with ≥50% L2G share (which are also the top-ranked genes for their respective associations). OTG links were parsed to determine the source of each OTG data point: the EBI GWAS catalog 30 ( n  = 136,503 hits with L2G share ≥0.5), Neale UK Biobank ( http://www.nealelab.is/uk-biobank ; n  = 19,139), FinnGen R6 (ref.  31 ) ( n  = 2,338) or SAIGE ( n  = 1,229).

PICCOLO 32 is a database of GWAS hits with gene mapping based on tests for colocalization without full summary statistics by using Probabilistic Identification of Causal SNPs (PICS) and a reference dataset of SNP linkage disequilibrium values. As described 32 , gene mapping uses quantitative trait locus (QTL) data from GTEx ( n  = 7,162) and a variety of other published sources ( n  = 6,552). We included hits with GWAS P  < 5 × 10 −8 , and with eQTL P  < 1 × 10 −5 , and posterior probability H4 ≥ 0.9, as these thresholds were determined empirically 32 to strongly predict colocalization results.

Genebass 33 is a database of genetic associations based on exome sequencing. Genebass data from 394,841 UK Biobank participants (the ‘500K’ release) were queried using Hail (19 October 2023). We used hits from four models: pLoF (predicted loss-of-function) or missense|LC (missense and low confidence LoF), each with sequencing kernel association test (SKAT) or burden tests, filtering for P  < 1 × 10 −5 . Because the traits in Genebass are from UK Biobank, which is included in OTG, we used the OTG MeSH mappings established above.

IntOGen is a database of enrichments of somatic genetic mutations within cancer types. We used the driver genes and cohort information tables (31 May 2023). IntOGen assigns each gene a mechanism in each tumour type; occasionally, a gene will be classified as a tumour suppressor in one type and an oncogene in another. We grouped by gene and assigned each gene its modal classification across cancers. MeSH mappings were curated manually.

MeSH term similarity

MeSH terms in either Pharmaprojects or the genetic associations datasets that were Supplementary Concept Records (IDs beginning in ‘C’) were mapped to their respective preferred main headings (IDs beginning in ‘D’). A matrix of all possible combinations of drug indication MeSH IDs and genetic association MeSH IDs was constructed. MeSH term Lin and Resnik similarities were computed for each pair as described 34 , 35 . Similarities of −1, indicating infinite distance between two concepts, were assigned as 0. The two scores were regressed against each other across all term pairs, and the Resnik scores were adjusted by a multiplier such that both scores had a range from 0 to 1 and their regression had a slope of 1. The two scores were then averaged to obtain a combined similarity score. Similarity scores were successfully calculated for 1,006 of 1,013 (99.3%) unique MeSH terms for Pharmaprojects indications, corresponding to 99.67% of Pharmaprojects T–I pairs, and for 2,260 of 2,262 (99.9%) unique MeSH terms for genetic associations, corresponding to >99.9% of associations.

Therapeutic areas

MeSH terms for Pharmaprojects indications were mapped onto 16 top-level headings under the Diseases [C] and Psychiatry and Psychology [F] branches of the MeSH tree ( https://meshb.nlm.nih.gov/treeView ), plus an ‘other’. The signs/symptoms area corresponds to C23 Pathological Conditions, Signs and Symptoms and contains entries such as inflammation and pain. Many MeSH terms map to >1 tree positions; these multiples were retained and counted towards each therapy area, except for the following conditions: for terms mapped to oncology, we deleted their mappings to all other areas; and ‘other’ was used only for terms that mapped to no other areas.

Analysis of T2D GWASs

We included 19 genes from OMIM linked to Mendelian forms of diabetes or syndromes with diabetic features. For Vujkovic et al. 18 , we considered as novel any genes with a novel nearest gene, novel coding variant or a novel lead SNP colocalized with an eQTL with H4 ≥ 0.9. Non-novel nearest genes, coding variants and colocalized lead SNPs were considered established variants. For Suzuki et al. 19 , we used the available L2G scores that OTG had assigned for the same lead SNPs in previously reported GWASs for other phenotypes, yielding mapped genes with L2G share >0.5 for 27% of loci. Genes were considered novel if absent from the Vujkovic analysis. Together, these approaches identified 217 established GWAS genes and 645 novel ones (469 from Vujkovic and 176 from Suzuki). We identified 347 unique drug targets in Pharmaprojects reported with a T2D or diabetes mellitus indication, including 25 approved. We reviewed the list of approved drugs and eliminated those for which there were questions around the relevance of the drug or target to T2D ( AKR1B1 , AR , DRD1 , HMGCR , IGF1R , LPL , SLC5A1 ). Because Pharmaprojects ordinarily specifies the receptor as target for protein or peptide replacement therapies, we also remapped the minority of programmes for which the ligand, rather than receptor, had been listed as target (changing INS to INSR , GCG to GCGR ). To assess the proportion of programmes with genetic support, we first grouped by drug and selected just one target, preferring the target with the earliest genetic support (OMIM, then established GWASs, then novel GWASs, then none). Next we grouped by target and selected its highest phase reached. Finally, we grouped by highest phase reached and counted the number of unique targets.

Universe of possible genetically supported G–I pairs

In all of our analyses, targets are defined as human gene symbols, but we use the term G–I pair to refer to possible genes that one might attempt to target with a drug, and T–I pair to refer to genes that are the targets of actual drug candidates in development. To enumerate the space of possible G–I pairs, we multiplied the n  = 769 Pharmaprojects indications considered here by the ‘universe’ of n  = 19,338 protein-coding genes, yielding a space of n  = 14,870,922 possible G–I pairs. Of these, n  = 101,954 (0.69%) qualify as having genetic support per our criteria. A total of 16,808 T–I pairs have reached at least phase I in an active or historical programme, of which 1,155 (6.9%) are genetically supported. This represents an enrichment compared with random chance (OR = 11.0, P  < 1.0 × 10 −15 , Fisher’s exact test), but in absolute terms, only 1.1% of genetically supported G–I pairs have been pursued. A genetically supported G–I pair may be less likely to attract drug development interest if the indication already has many other potential targets, and/or if the indication is but the second-most similar to the gene’s associated trait. Removing associations with many GWAS hits and restricting to the single most similar indication left a space of 34,190 possible genetically supported G–I pairs, 719 (2.1%) of which had been pursued. This small percentage might yet be perceived to reflect competitive saturation, if the vast majority of indications are undevelopable and/or the vast majority of targets are undruggable. We therefore asked what proportion of genetically supported G–I pairs had been developed to at least phase I, as a function of therapy area cross-tabulated against Open Targets predicted tractability status or membership in canonically ‘druggable’ protein families, using families from ref. 22 as well as UniProt pkinfam for kinases 36 . We also grouped at the level of gene, rather than G–I pair (Extended Data Fig. 8 ).

Druggability and protein families

Antibody and small molecule druggability status was taken from Open Targets 37 . For antibody tractability, Clinical Precedence, Predicted Tractable–High Confidence and Predicted Tractable–Medium to Low Confidence were included. For small molecules, Clinical Precedence, Discovery Precedence and Predicted Tractable were included. Protein families were from sources described previously 22 , plus the pkinfam kinase list from UniProt 36 . To make these lists non-overlapping, genes that were both kinases and also enzymes, ion channels or nuclear receptors were considered to be kinases only.

Analyses were conducted in R 4.2.0. For binomial proportions P(G) and P(S), error bars are Wilson 95% CIs, except for P(S) for phase I–launch for which the Wald method is used to compute the confidence intervals on the product of the individual probabilities of success at each phase. RS uses Katz 95% CIs, with the phase I launch RS based on the number of programs entering phase I and succeeding in phase III. Effects of continuous variables on probability of launch were assessed using logistic regression. Differences in RS between therapy areas were tested using the Cochran–Mantel–Haenszel chi-squared test (cmh.test from the R lawstat package, v.3.4). Pipeline progression of D–I pairs conditioned on the highest phase reached by a drug was modelled using an ordinal logit model (polr with Hess = TRUE from the R MASS package, v.7.3-56). Correlations across therapy areas were tested by weighted Pearson’s correlation (wtd.cor from the R weights package, v.1.0.4); to control for the amount of data available in each therapy area, the number of genetically supported T–I pairs having reached at least phase I was used as the weight. Enrichments of T–I pairs in the utilization analysis were tested using Fisher’s exact test. All statistical tests were two-sided.

Reporting summary

Further information on research design is available in the  Nature Portfolio Reporting Summary linked to this article.

Data availability

An analytical dataset is provided at GitHub at https://github.com/ericminikel/genetic_support/ (ref. 38 ) and is sufficient to reproduce all figures and statistics herein. This repository is permanently archived at Zenodo at https://doi.org/10.5281/zenodo.10783210 (ref. 39 ). Source data are provided with this paper.

Code availability

Source code is provided at GitHub at https://github.com/ericminikel/genetic_support/ (ref. 38 ) and is sufficient to reproduce all figures and statistics herein. This code is permanently archived at the Zenodo repository at https://doi.org/10.5281/zenodo.10783210 (ref. 39 ).

DiMasi, J. A., Grabowski, H. G. & Hansen, R. W. Innovation in the pharmaceutical industry: new estimates of R&D costs. J. Health Econ. 47 , 20–33 (2016).

Article   PubMed   Google Scholar  

Hay, M., Thomas, D. W., Craighead, J. L., Economides, C. & Rosenthal, J. Clinical development success rates for investigational drugs. Nat. Biotechnol. 32 , 40–51 (2014).

Article   CAS   PubMed   Google Scholar  

Wong, C. H., Siah, K. W. & Lo, A. W. Estimation of clinical trial success rates and related parameters. Biostatistics 20 , 273–286 (2019).

Article   MathSciNet   PubMed   Google Scholar  

Thomas D. et al. Clinical Development Success Rates and Contributing Factors 2011–2020 (Biotechnology Innovation Organization, 2021); https://go.bio.org/rs/490-EHZ-999/images/ClinicalDevelopmentSuccessRates2011_2020.pdf

Nelson, M. R. et al. The support of human genetic evidence for approved drug indications. Nat. Genet. 47 , 856–860 (2015).

Diogo, D. et al. Phenome-wide association studies across large population cohorts support drug target validation. Nat. Commun. 9 , 4285 (2018).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Plenge, R. M., Scolnick, E. M. & Altshuler, D. Validating therapeutic targets through human genetics. Nat. Rev. Drug Discov. 12 , 581–594 (2013).

Musunuru, K. & Kathiresan, S. Genetics of common, complex coronary artery disease. Cell 177 , 132–145 (2019).

Trajanoska, K. et al. From target discovery to clinical drug development with human genetics. Nature 620 , 737–745 (2023).

Article   ADS   CAS   PubMed   Google Scholar  

Burgess, S. et al. Using genetic association data to guide drug discovery and development: review of methods and applications. Am. J. Hum. Genet. 110 , 195–214 (2023).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Carss, K. J. et al. Using human genetics to improve safety assessment of therapeutics. Nat. Rev. Drug Discov. 22 , 145–162 (2023).

Nguyen, P. A., Born, D. A., Deaton, A. M., Nioi, P. & Ward, L. D. Phenotypes associated with genes encoding drug targets are predictive of clinical trial side effects. Nat. Commun. 10 , 1579 (2019).

Minikel, E. V., Nelson, M. R. Human genetic evidence enriched for side effects of approved drugs. Preprint at medRxiv https://doi.org/10.1101/2023.12.12.23299869 (2023).

Visscher, P. M., Brown, M. A., McCarthy, M. I. & Yang, J. Five years of GWAS discovery. Am. J. Hum. Genet. 90 , 7–24 (2012).

King, E. A., Davis, J. W. & Degner, J. F. Are drug targets with genetic support twice as likely to be approved? Revised estimates of the impact of genetic support for drug mechanisms on the probability of drug approval. PLoS Genet. 15 , e1008489 (2019).

Article   PubMed   PubMed Central   Google Scholar  

Hingorani, A. D. et al. Improving the odds of drug development success through human genomics: modelling study. Sci. Rep. 9 , 18911 (2019).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Reay, W. R. & Cairns, M. J. Advancing the use of genome-wide association studies for drug repurposing. Nat. Rev. Genet. 22 , 658–671 (2021).

Vujkovic M. et al. Discovery of 318 new risk loci for type 2 diabetes and related vascular outcomes among 1.4 million participants in a multi-ancestry meta-analysis. Nat. Genet. 52 , 680–691 (2020).

Suzuki K. et al. Genetic drivers of heterogeneity in type 2 diabetes pathophysiology. Nature 627 , 347–357 (2024).

Lommatzsch, M. et al. Disease-modifying anti-asthmatic drugs. Lancet 399 , 1664–1668 (2022).

Mortberg, M. A., Vallabh, S. M. & Minikel, E. V. Disease stages and therapeutic hypotheses in two decades of neurodegenerative disease clinical trials. Sci. Rep. 12 , 17708 (2022).

Minikel, E. V. et al. Evaluating drug targets through human loss-of-function genetic variation. Nature 581 , 459–464 (2020).

Ference, B. A. et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur. Heart J. 38 , 2459–2472 (2017).

Scannell, J. W. et al. Predictive validity in drug discovery: what it is, why it matters and how to improve it. Nat. Rev. Drug Discov. 21 , 915–931 (2022).

Sun, B. B. et al. Genetic associations of protein-coding variants in human disease. Nature 603 , 95–102 (2022).

Pharmaprojects (Citeline, accessed 30 August 2023); https://web.archive.org/web/20230830135309/https://www.citeline.com/en/products-services/clinical/pharmaprojects

Painter, J. L. Toward automating an inference model on unstructured terminologies: OXMIS case study. Adv. Exp. Med. Biol. 680 , 645–651 (2010).

Mountjoy, E. et al. An open approach to systematically prioritize causal variants and genes at all published human GWAS trait-associated loci. Nat. Genet. 53 , 1527–1533 (2021).

Zheng, J. et al. Phenome-wide Mendelian randomization mapping the influence of the plasma proteome on complex diseases. Nat. Genet. 52 , 1122–1131 (2020).

Sollis, E. et al. The NHGRI-EBI GWAS Catalog: knowledgebase and deposition resource. Nucleic Acids Res. 51 , D977–D985 (2023).

Kurki, M. I. et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613 , 508–518 (2023).

Guo C. et al. Identification of putative effector genes across the GWAS Catalog using molecular quantitative trait loci from 68 tissues and cell types. Preprint at bioRxiv https://doi.org/10.1101/808444 (2019).

Karczewski, K. J. et al. Systematic single-variant and gene-based association testing of thousands of phenotypes in 394,841 UK Biobank exomes. Cell Genomics. 2 , 100168 (2022).

Lin D. An information-theoretic definition of similarity. In Proc. 15th International Conference on Machine Learning (ICML) (ed. Shavlik, J. W.) 296–304 (Morgan Kaufmann Publishers Inc., 1998).

Resnik P. Semantic similarity in a taxonomy: an information-based measure and its application to problems of ambiguity in natural language. J. Artif. Intell. Res. 11 , 95–130 (1999).

The UniProt Consortium. UniProt: the universal protein knowledgebase. Nucleic Acids Res. 45 , D158–D169 (2017).

Article   Google Scholar  

Ochoa, D. et al. The next-generation Open Targets Platform: reimagined, redesigned, rebuilt. Nucleic Acids Res. 51 , D1353–D1359 (2023).

Minikel, E. et al. GitHub https://github.com/ericminikel/genetic_support/ (2024).

Minikel, E. et al. Refining the impact of genetic evidence on clinical success. Zenodo https://doi.org/10.5281/zenodo.10783210 (2024).

Download references

Acknowledgements

This study was funded by Deerfield.

Author information

Jeffery L. Painter

Present address: GlaxoSmithKline, Research Triangle Park, NC, USA

Authors and Affiliations

Stanley Center for Psychiatric Research, Broad Institute, Cambridge, MA, USA

Eric Vallabh Minikel

JiveCast, Raleigh, NC, USA

Deerfield Management Company LP, New York, NY, USA

Coco Chengliang Dong & Matthew R. Nelson

Genscience LLC, New York, NY, USA

Matthew R. Nelson

You can also search for this author in PubMed   Google Scholar

Contributions

M.R.N. and E.V.M. conceived and designed the study. E.V.M., J.L.P., C.C.D. and M.R.N. performed analyses. M.R.N. supervised the research. M.R.N. and E.V.M. drafted the manuscript. E.V.M., J.L.P., C.C.D. and M.R.N. reviewed and approved the final manuscript.

Corresponding author

Correspondence to Matthew R. Nelson .

Ethics declarations

Competing interests.

M.R.N. is an employee of Deerfield and Genscience. C.C.D. is an employee of Deerfield. E.V.M. and J.L.P. are consultants to Deerfield. Unrelated to the current work, E.V.M. acknowledges speaking fees from Eli Lilly, consulting fees from Alnylam and research support from Ionis, Gate, Sangamo and Eli Lilly.

Peer review

Peer review information.

Nature thanks Joanna Howson, Heiko Runz and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended data fig. 1 data processing schematic..

A ) Dataset size, filters, and join process for Pharmaprojects and human genetic evidence. Note that a drug can be assigned multiple targets, and can be approved for multiple indications. The entire analysis described herein has also been run restricted to only those drugs with exactly one target annotated (Figs. S1 – S11 ). B ) Illustration of the definition of genetic support. A table of drug development programs with one row per target-indication pair (left) is joined to a table of human genetic associations based on the identity of the gene encoding the drug target and the similarity between the drug indication MeSH term and the genetically associated trait MeSH term being ≥ 0.8. Drug program rows with a joined row in the genetic associations table are considered to have genetic support.

Extended Data Fig. 2 Further analysis of influence of characteristics of genetic associations on relative success.

A ) Sensitivity of RS to the similarity threshold between the MeSH ID for the genetically associated trait and the MeSH ID for the clinically developed indication. The threshold is varied by units of 0.05 (labels) and the results are plotted as RS (y axis) versus number of genetically supported T-I pairs (x axis). B ) Breakdown of OTG and OMIM RS values by whether any drug for each T-I pair has had orphan status assigned. The N of genetically supported T-I pairs (denominator) and, of those, launched T-I pairs (numerator) is shown at right. Values for the full 2×2 contingency table including the non-supported pairs, used to calculate RS, are provided in Table S12 . Total N = 13,022 T-I pairs, of which 3,149 are orphan. The center is the RS point estimate and error bars are Katz 95% confidence intervals. C ) RS for somatic genetic evidence from IntOGen versus germline genetic evidence, for oncology and non-oncology indications. Note that the approved/supported proportions displayed for the top two rows are identical because all IntOGen genetic support is for oncology indications, yet the RS is different because the number of non-supported approved and non-supported clinical stage programs is different. In other words, in the “All indications” row, there is a Simpson’s paradox that diminishes the apparent RS of IntOGen — IntOGen support improves success rate (see 2 nd row) but also selects for oncology, an area with low baseline success rate (as shown in Extended Data Fig. 6a ). N is displayed at right as in (B), with full contingency tables in Table S13 . Total N = 13,022 T-I pairs, of which 6,842 non-oncology, 6,180 oncology, 1,287 targeting IntOGen oncogenes, 284 targeting tumor suppressors, and 176 targeting IntOGen genes of unknown mechanism. The center is the RS point estimate and error bars are Katz 95% confidence intervals. D ) As for top panel of Fig. 1d , but without removing replications or OMIM-supported T-I pairs. N is displayed as in (B), with full contingency tables in Table S14 . Total N = 13,022 T-I pairs. The center is the RS point estimate and error bars are Katz 95% confidence intervals. E ) As for top panel of Fig. 1d , removing replications but not removing OMIM-supported T-I pairs. N is displayed as in (B), with full contingency tables in Table S15 . Total N = 13,022 T-I pairs. The center is the RS point estimate and error bars are Katz 95% confidence intervals. F ) Proportion of T-I pairs supported by a GWAS Catalog association that are launched (versus phase I-III) as a function of the year of first genetic association. G ) Launched T-I pairs genetically supported by OTG GWAS, shown by year of launch (y axis) and year of first genetic association (x axis). Gene symbols are labeled for first approvals of targets with at least 5 years between association and launch. Of 104 OTG-supported launched T-I pairs (Fig. 1d ), year of drug launch was available for N = 38 shown here, of which 18 (47%) acquired genetic support only in or after the year of launch. The true proportion of launched T-I whose GWAS support is retrospective may be larger if the T-I with a missing launch year are more often older drug approvals less well annotated in Pharmaprojects. H ) Lack of impact of GWAS Catalog lead SNP odds ratio (OR) on RS when using the same OR breaks as used by King et al. 15 . N is displayed as in (B), with full contingency tables in Table S18 . Total N = 13,022 T-I pairs. The center is the RS point estimate and error bars are Katz 95% confidence intervals. See Fig. S4 for the same analyses restricted to drugs with a single known target.

Extended Data Fig. 3 Sensitivity to changes in genetic data and drug pipeline over the past decade and to the ‘genetic insight’ filter.

“2013” here indicates the data freezes from Nelson et al. 5 (that study’s supplementary dataset 2 for genetics and supplementary dataset 3 for drug pipeline); “2023” indicates the data freezes in the present study. All datasets were processed using the current MeSH similarity matrix, and because “genetic insight” changes over time (more traits have been studied genetically now than in 2013), all panels are unfiltered for genetic insight (hence numbers in panel D differ from those in Fig. 1a ). Every panel shows the proportion of combined (both historical and active) target-indication pairs with genetic support, or P(G), by development phase. A ) 2013 drug pipeline and 2013 genetics. B ) 2013 drug pipeline and 2023 genetics. C ) 2023 drug pipeline and 2013 genetics. D ) 2023 drug pipeline and 2023 genetics. E ) 2023 drug pipeline with only OTG GWAS hits through 2013 and no other sources of genetic evidence. F ) 2023 drug pipeline with only OTG GWAS hits for all years, no other sources of genetic evidence. We note that the increase in P(G) over the past decade 5 is almost entirely attributable to new genetic evidence (e.g. contrast B vs. A, D vs. C, F vs. E) rather than changes in the drug pipeline (e.g. compare A vs. C, B vs. D). In contrast, the increase in RS is due mostly to changes in the drug pipeline (compare C, D, E, F vs. A, B), in line with theoretical expectations outlined by Hingorani et al. 16 and consistent with the findings of King et al. 15 We note that both the contrasts in this figure, and the fact that genetic support is so often retrospective (Extended Data Fig. 2g ) suggest that P(G) will continue to rise in coming years. For 2013 drug pipeline, N = 8,624 T-I pairs (1,605 preclinical, 1,772 phase I, 2,779 phase II, 636 phase III, and 1,832 launched); for 2023 drug pipeline, N = 29,464 T-I pairs (N = 12,653 preclinical, 4,946 phase I, 8,268 phase II, 1,781 phase III, and 1,816 launched). Details including numerator and denominator for P(G) and full continency tables for RS are provided in Tables S19 - S20 . In A-F, the center is exact proportion and error bars are Wilson binomial 95% confidence intervals. Because all panels here are unfiltered for genetic insight, we also show the difference in RS across G ) sources of genetic evidence and H ) therapy areas when this filter is removed. In general, removing this filter decreases RS by 0.17; this varies only slightly between sources and areas. The largest impact is seen in Infection, where removing the filter drops the RS from 2.73 to 2.03. The relatively minor impact of removing the genetic insight filter is consistent with the findings of King et al. 15 , who varied the minimum number of genetic associations required for an indication to be included, and found that risk ratio for progression (i.e. RS) was slightly diminished when the threshold was reduced. See Fig. S5 for the same analyses restricted to drugs with a single known target.

Extended Data Fig. 4 Proportion of type 2 diabetes drug targets with human genetic support by highest phase reached.

A) OMIM, B) established (2019 and earlier) GWAS genes, C) novel (new in Vujkovic 2020 or Suzuki 2023) GWAS genes, or D) any of the above. See  Methods for details on GWAS dataset processing. N is indicated at right of each panel, with denominator being the number of T2D targets at each stage and the numerator being the number of those that are genetically supported. Total N = 284 targets. The center is the exact proportion and error bars are Wilson binomial 95% confidence intervals.

Extended Data Fig. 5 P(G) by phase versus therapy area.

Each panel represents one therapy area, and shows the proportion of target-indication pairs in that area with genetic support, or P(G), by development phase. The genetically supported and total number of T-I pairs at each phase in each therapy area are provided in Table S33 . Total number of T-I pairs in any area: N = 10,839 preclinical, N = 4,421 phase I, N = 7,383 phase II, N = 1,551 phase III, N = 1,519 launched. The center is the exact proportion and error bars are Wilson binomial 95% confidence intervals. See Fig. S6 for the same analyses restricted to drugs with a single known target.

Extended Data Fig. 6 Confounding between therapy areas and properties of supporting genetic evidence.

In panels A-E, each point represents one GWAS Catalog-supported T-I pair in phase I through launched, and boxes represent medians and interquartile ranges (25 th , 50 th , and 75 th percentile). Each panel A-E represents the cross-tabulation of therapy areas versus the properties examined in Fig. 1d . Kruskal-Wallis tests treat each variable as continuous, while chi-squared tests are applied to the discrete bins used in Fig. 1d . A ) Year of discovery, Kruskal-Wallis P = 1.1e-11, chi-squared P = 2.9e-16, N = 686 target-indication-area (T-I-A) triplets; B ) gene count, Kruskal-Wallis P = 6.2e-35, chi-squared P = 7.1e-47, N = 770 T-I-A triplets; C ) absolute beta, Kruskal-Wallis P = 1.2e-5, chi-squared P = 1.7e-7, N = 461 T-I-A triplets; D ) absolute odds ratio, Kruskal-Wallis P = 2.5e-5, chi-squared P = 4.3e-6, N = 305 T-I-A triplets; E ) minor allele frequency, Kruskal-Wallis P = 5.7e-4, chi-squared P = 4.3e-3, N = 584 T-I-A triplets; F ) Barplot of therapy areas of genetically supported T-I by source of GWAS data within OTG, chi-squared P = 2.4e-7. See Fig. S7 for the same analyses restricted to drugs with a single known target.

Extended Data Fig. 7 Further analyses of differences in relative success among therapy areas.

A ) Probability of success, P(S), by therapy area, with Wilson 95% confidence intervals. The N shown at right indicates the number of launched T-I pairs (numerator) and number of T-I pairs reaching at least phase I (denominator). The center is the exact proportion and error bars are Wilson binomial 95% confidence intervals. B ) Probability of genetic support, P(G), by therapy area, with Wilson 95% confidence intervals. The N shown at right indicates the number of genetically supported T-I pairs reaching at least phase I (numerator) and total number of T-I pairs reaching at least phase I (denominator). The center is the exact proportion and error bars are Wilson binomial 95% confidence intervals. C ) P(S) vs. P(G), D ) RS s. P(S), and E ) RS vs. P(G) across therapy areas, with centers indicating point estimates and crosshairs representing 95% confidence intervals on both dimensions — Katz for RS and Wilson for P(G) and P(S). For A-E, total N = 13,022 unique T-I pairs, but because some indications belong to > 1 therapy area, N = 16,900 target-indication-area (T-I-A) triples. For exact N and full contingency tables, see Table S28 . F ) Re-analysis of RS (x axis) broken down by therapy area using data from supplementary table  6 of Nelson et al. 5 . G ) Confusion matrix showing the categorization of unique drug indications into therapy areas in Nelson et al. 5 versus current. Note that the current categorization is based on each indication’s position in the MeSH ontological tree and one indication can appear in > 1 area, see  Methods for details. Marginals along the top edge are the number of drug indications in each current therapy area that were absent from the 2015 dataset. Marginals along the right edge are the number of drug indications in each 2015 therapy area that are absent from the current dataset. See Fig. S8 for the same analyses restricted to drugs with a single known target.

Extended Data Fig. 8 Level of utilization of genetic support among targets.

As for Fig. 3 , but grouped by target instead of T-I pair. Thus, the denominator for each cell is the number of targets with at least one genetically supported indication, and each target counts towards the numerator if at least one genetically supported indication has reached phase I. See Fig. S9 for the same analyses restricted to drugs with a single known target.

Supplementary information

Supplementary figures.

Supplementary Figs. 1–9, corresponding to the three main and six extended data figures restricted to drugs with one target only.

Reporting Summary

Peer review file, supplementary data.

Supplementary Tables 1–50, including information on all target-indication pairs, source data for all graphs and additional analyses.

Source data

Source data fig. 1, source data fig. 2, source data fig. 3, source data extended data fig. 2, source data extended data fig. 3, source data extended data fig. 4, source data extended data fig. 5, source data extended data fig. 6, source data extended data fig. 7, source data extended data fig. 8, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Minikel, E.V., Painter, J.L., Dong, C.C. et al. Refining the impact of genetic evidence on clinical success. Nature (2024). https://doi.org/10.1038/s41586-024-07316-0

Download citation

Received : 05 July 2023

Accepted : 14 March 2024

Published : 17 April 2024

DOI : https://doi.org/10.1038/s41586-024-07316-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

ORIGINAL RESEARCH article

Assessment of causal associations between obesity and peripheral artery disease: a bidirectional mendelian randomization study provisionally accepted.

• 1 Guangzhou University of Chinese Medicine, China

The final, formatted version of the article will be published soon.

Background: Several observational studies have documented a potential link between obesity and peripheral artery disease (PAD), although conflicting findings exist. The causal relationship between obesity and PAD continues to be a subject of ongoing debate in the medical community. Objectives: In this study, we employed a bidirectional Mendelian randomization (MR) analysis to explore the potential causal relationship between obesity and the risk of PAD. Methods: To investigate these causal relationships, we conducted bidirectional MR analysis using publicly available genome-wide association study (GWAS) data. Effect estimates were calculated using the random-effects inverse variance-weighted (IVW) method. Results: We identified eight independent single nucleotide polymorphisms (SNPs) associated with obesity in 218,735 samples involving 16,380,465 SNPs, all of which met the genome-wide significance threshold (p<5×10⁻⁸). The IVW analysis indicates a significant positive association between genetic obesity and multiple datasets with PAD as the outcome: Queue-1 (GWAS ID: finn-b-I9_PAD) (OR=1.138, 95%CI: 1.027-1.261, p=0.013), Queue-2 (GWAS ID: bbj-a-144) (OR=1.190, 95%CI: 1.019-1.390, p=0.028), Queue-3 (GWAS ID: ebi-a-GCST90018670) (OR=1.174, 95%CI: 1.014-1.360, p=0.032), and Queue-4 (GWAS ID: ebi-a-GCST90018890) (OR=1.194, 95%CI: 1.099-1.296, p<0.001). However, we did not observe a significant genetic-level association between obesity and PAD for Queue-5 (GWAS ID: ukb-d-I9_PAD) (OR=1.001, 95%CI: 1.000-1.002, p=0.071). Furthermore, we conducted a reverse causal MR analysis to explore the potential reverse causal relationship between obesity and PAD. This comprehensive analysis did not provide evidence of a reverse causal association between these two factors. Conclusions: In summary, our study offers genetic evidence suggesting a possible causal link between obesity and PAD. While we did not find evidence supporting the "obesity paradox", prudent weight management remains crucial, as lower weight does not necessarily guarantee better outcomes. As with any study, caution is required in interpreting the findings. Further research is essential to assess the clinical relevance of weight in preventing PAD, which could inform the development of more precise intervention strategies.

Keywords: Mendelian randomization, Obesity, peripheral artery disease, Atherosclerotic cardiovascular disease, Metabolic Diseases

Received: 03 Nov 2023; Accepted: 17 Apr 2024.

Copyright: © 2024 Huang, Pang, Zhang and Huang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Dr. Xi-wei Huang, Guangzhou University of Chinese Medicine, Guangzhou, China Dr. Chuang-Wei Huang, Guangzhou University of Chinese Medicine, Guangzhou, China

People also looked at

• Skip to main content
• Keyboard shortcuts for audio player

Your Health

• Treatments & Tests
• Health Inc.
• Public Health

Gay people often have older brothers. Why? And does it matter?

Selena Simmons-Duffin

Credit: Lily Padula for NPR

The Science of Siblings is a new series exploring the ways our siblings can influence us, from our money and our mental health all the way down to our very molecules. We'll be sharing these stories over the next several weeks.

This is something I learned years ago through gay bar chatter: Gay people are often the youngest kids in their families. I liked the idea right away — as a gay youngest sibling, it made me feel like there was a statistical order to things and I fit neatly into that order.

When I started to report on the science behind it, I learned it's true: There is a well-documented correlation between having older siblings (older brothers, specifically) and a person's chance of being gay. But parts of the story also struck me as strange and dark. I thought of We the Animals , Justin Torres' haunting semi-autobiographical novel about three brothers — the youngest of whom is queer — growing up in New York state. So I called Torres to get his take on the idea.

The Science of Siblings

Torres' first reaction was to find it considerably less appealing than I did. This makes sense — his latest novel, Blackouts , won a National Book Award last year, and it grapples with the sinister history of how scientists have studied sexuality. "My novel is interested in the pre-Kinsey sexology studies, specifically this one called Sex Variants ," he told me. "It's really informed by eugenics. They were looking for the cause of homosexuality in the body in order to treat it or cure it or get rid of it."

That's why, when he saw my inquiry about a statistical finding that connects sexuality and birth order, he was wary. "To be frank, I find these kinds of studies that're looking for something rooted in the body to explain sexuality to be kind of bunk. I think they rely on a really binary understanding of sexuality itself," he said.

"That's fair," I conceded. But this connection between queerness and older brothers has been found so many times in so many places that one researcher told me it's "a kind of truth" in the science of sexuality.

Rooted in a dark past

The first research on this topic did indeed begin in the 1940s and '50s, during that era of investigations into what causes homosexuality, to be able to cure it. At the time, the queer people whom scientists were studying were living in a world where this facet of their identity was dangerous. Plus, the studies themselves didn't find much, says Jan Kabátek , a senior research fellow at the University of Melbourne.

"Most of it fell flat," he told me. "But there is an exception to this, and that is the finding that men, specifically, who exhibit attraction to the same sex are likely to have more older brothers than other types of siblings."

In the 1990s, this was dubbed the "fraternal birth order effect." In the years since, it has been found again and again, all over the world.

"This pattern has been documented around Canada and the United States, but it goes well beyond that," says Scott Semenyna , a psychology professor at Stetson University. "There's been now many confirmations that this pattern exists in countries like Samoa. It exists in southern Mexico. It exists in places like Turkey and Brazil."

Huge study, consistent findings

An impressive recent study established that this pattern held up in an analysis of a huge sample — over 9 million people from the Netherlands. It confirmed all those earlier studies and added a twist.

"Interestingly enough — and this is quite different from what has been done before — we also showed that the same association manifests for women," explains Kabátek, one of the study's authors. Women who were in same-sex marriages were also more likely to have older brothers than other types of siblings.

At baseline, the chance that someone will be gay is pretty small. "Somewhere around 2 to 3% — we can call it 2% just for the sake of simplicity," Semenyna says. "The fraternal birth order effect shows that you're going to run into about a 33% increase in the probability of, like, male same-sex attraction for every older brother that you have."

The effect is cumulative: The more older brothers someone has, the bigger it is. If you have one older brother, your probability of being gay nudges up to about 2.6%. "And then that probability would increase another 33% if there was a second older brother, to about 3.5%," Semenyna says.

If you have five older brothers, your chance of being gay is about 8% — so, four times the baseline probability.

The author, Selena Simmons-Duffin, at age 3, with her brother, David Simmons-Duffin, at age 5. The Simmons-Duffin family hide caption

The author, Selena Simmons-Duffin, at age 3, with her brother, David Simmons-Duffin, at age 5.

Still, even 8% is pretty small. "The vast majority of people who have a lot of older brothers are still going to come out opposite-sex attracted," Semenyna says. Also, plenty of gay people have no brothers at all, or they're the oldest in their families. Having older brothers is definitely not the only influence on a person's sexuality.

"But just the fact that we are observing effects that are so strong, relatively speaking, implies that there's a good chance that there is, at least partially, some biological mechanism that is driving these associations," Kabátek says.

A hypothesis, but no definitive mechanism

For decades, the leading candidate for that biological mechanism has been the "maternal immune hypothesis," Semenyna explains. "The basic version of this hypothesis is that when a male fetus is developing, the Y chromosome of the male produces proteins that are going to be recognized as foreign by the mother's immune system and it forms somewhat of an immune response to those proteins."

That immune response has some effect on the development of subsequent male fetuses, Semenyna says. The plausibility of this hypothesis was bolstered by a 2017 study that found "that mothers of gay sons have more of these antibodies that target these male-specific proteins than mothers of sons who are not gay or mothers who have no sons whatsoever," he says.

But now that Kabátek's study of the Dutch population has found that this pattern was present among women in same-sex marriages as well, there are new questions about whether this hypothesis is correct.

"One option is that the immune hypothesis works for both men and women," Kabátek says. "Of course, there can be also other explanations. It's for prospective research to make this clearer."

Fun to think about, but concerning too

In a way, I tell Justin Torres, this effect seems simple and fun to me. It's a concrete statistical finding, documented all over the world, and there's an intriguing hypothesis about why it may happen biologically. But darker undercurrents in all of it worry me, like raising a dangerous idea that becoming gay in the womb is the only version of gayness that is real — or a repackaged version of the old idea that mothers are to "blame."

"It is the undercurrents that worry me immensely," he responds. "I remember when I was a kid — I have this memory of watching daytime television. I must have been staying home from school sick in the late '80s or early '90s. The host polled the audience and said, 'If there was a test [during pregnancy] and you could know if your child was gay, would you abort?' I remember being so horrified and disturbed watching all those hands go up in the audience — just feeling so hated. At that young age, I knew this thing about myself, even if I wasn't ready to admit it."

Even if tolerance for queer people in American society has grown a lot since then, he says, "I think that tolerance waxes and wanes, and I worry about that line of thinking."

At the same time, he agrees that the idea of a connection with gay people being the youngest kids in their families is kind of hilarious. "One thing that pops into my mind is, like, maybe if you're just surrounded by a lot of men, you either choose or don't choose men, right?" he laughs.

Essentially, in his view, it's fun to think about, but probably not deeper than that.

"As a humanist, I just don't know why we need to look for explanations for something as complex and joyous and weird as sexuality," Torres says.

Then again, scientists are unlikely to be able to resist that mysterious, weird complexity. Even if the joy and self-expression and community and so many other parts of queerness and sexuality will always be more than statistics can explain.

More from the Science of Siblings series:

• A gunman stole his twin from him. This is what he's learned about grieving a sibling
• In the womb, a brother's hormones can shape a sister's future
• These identical twins both grew up with autism, but took very different paths
• Science of Siblings
• queer community
• homosexuality

• The Graduate School >
• Graduate News >

UB researcher awarded Hypothesis Fund to explore if RNA droplets helped originate life on Earth

research news

UB faculty member Priya Banerjee has received a Hypothesis Fund seed grant to explore RNA molecules and their role in the origin of life on Earth. Photo: Douglas Levere

By TOM DINKI

Published April 17, 2024

The RNA world theory suggests that life on Earth began with RNA molecules that copied themselves. It’s believed this self-replication eventually gave rise over millions of years to DNA and protein, which then formed with RNA to create cells. 

Yet RNA would seem ill-suited to serve such an important role in the harsh environment of the prebiotic world — it’s known to destabilize under high temperature and pressure.

UB faculty member Priya R. Banerjee believes the key to solving this puzzle may be RNA’s intrinsic ability to form liquid-like droplets at high temperatures, which may have protected it from harsh conditions and compartmentalized its functions.

Banerjee, associate professor of physics, has now received a seed grant from the Hypothesis Fund to better study these RNA droplets and their potential role in the origin of life on Earth. 

The project, “Liquid RNA Condensates as Programmable Scaffolds for Compartmentalization and Catalysis,” was selected for the boldness of the science, as well as Banerjee’s willingness to take risks and go after a big idea, according to the Hypothesis Fund, which announced the award this week. 

Hypothesis Fund seed grants fund innovation research at its earliest stages, typically before there is any preliminary data, with the goal of supporting high-risk, high-reward ideas that may not be funded or pursued otherwise. 

“Dr. Banerjee’s project brings unique insights into the origin of life by understanding the biophysical properties and self-organization principles encoded into RNA molecules. His hypothesis is bold and innovative, and has the potential to answer conundrums in how life may have arisen with RNA, while also bringing insight to the development of more effective RNA-based interventions,” says Hypothesis Fund Scout Taekjip Ha, professor of pediatrics at Harvard Medical School and cellular and molecular medicine at Boston Children’s Hospital. 

According to RNA world theory, RNA served functions in the primordial soup later done by DNA and protein — encoding genetic material and catalyzing chemical reactions. 

However, the theory is hotly debated. Key objections include thermal instability of RNAs and a lack of mechanistic understanding of how RNA-driven compartmentalization was achieved in the prebiotic world. 

Banerjee, who is also director of graduate studies in the Department of Physics, has recently reported an unexpected discovery of  RNA phase separation into droplets, or condensates, when exposed to high temperatures . He is now studying how these droplets, which are also formed by DNA and protein, impact cellular function and disease processes. 

“We posit that temperature-controlled reversible and dynamic droplet formation by RNA molecules can address this key knowledge gap,” Banerjee says. “Our working hypothesis is that tiny RNA liquid droplets are programmable microscale compartments for RNA biology.”

By shedding light on the molecular origin of RNAs’ thermo-responsive droplet formation, the project could determine the role of the droplet state of RNAs in diverse biological functions, Banerjee says. 

The project is expected to last for 18 months.