presentation of real life problems in grasps

Use GRASPS for Real-World Assessment

Innovative educators understand that there is more to learning than processed worksheets and tests.

Innovative educators understand that there is more to learning than processed worksheets and tests. That's why real-world tasks and assessments are finally making it out of just the elite schools and are becoming more prevalent in mainstream education.  At the  Tech & Learning Leadership Summit  experts in the area of technology and education came together to discuss a variety of topics including how technology supports bringing real learning experiences to the classroom. 

G.R.A.S.P.S. Model

One model popular among attendees was one adapted from Grant Wiggins and Jay McTighe.  It is called GRASPS, which is an acronym standing for:

• Provide a statement of the task. Establish the goal, problem, challenge, or obstacle in the task.
• Possible sentence starters:
• Your task is to… The goal is to… The problem or challenge is… The obstacle to overcome is…
• Define the role of the students in the task. State the job of the students for the task.
• You are… You have been asked to… Your job is…

A: Audience

• Identify the target audience within the context of the scenario. Example audiences might include a client or committee. 
• Your clients are… The target audience is… You need to convince…

S: Situation

• Set the context of the scenario. Explain the situation.
• The context you find yourself in is… The challenge involves dealing with…

P: Products or Performances

Tech & Learning Newsletter

Tools and ideas to transform education. Sign up below.

• Clarify what the students will create and why they will create it,
• You will create a … in order to… You need to develop a … so that 

S: Standards

• Provide students with a clear picture of success. Identify specific standards for success. Issue rubrics to the students or develop them with the student.
• Your performance needs to… Your work will be judged by… Your product must meet the following standards… A successful result will… 

Note that it is unnecessary to use all or even any of the sentence starters. You can replace a prompt with your own. These are provided to help the learning designer think about the task. Generally one sentence starter can be used to write 

cross posted at The Innovative Educator  

Lisa Nielsen  ( @InnovativeEdu ) has worked as a public-school educator and administrator since 1997. She is a prolific writer best known for her award-winning blog,  The Innovative Educator . Nielsen is the author of  several books and her writing has been featured in media outlets such as  The New York Times , The Wall Street Journal ,  Tech&Learning , and  T.H.E. Journal .   

Lisa Nielsen  ( @InnovativeEdu ) has worked as a public-school educator and administrator since 1997. She is a prolific writer best known for her award-winning blog, The Innovative Educator . Nielsen is the author of several books  and her writing has been featured in media outlets such as The New York Times,  The Wall Street Journal , and Tech & Learning.  

Disclaimer : The information shared here is strictly that of the author and does not reflect the opinions or endorsement of her employer.

Best Grants for Education

Building Equity Through Next Generation Science Standards

Edtech Show & Tell: April 2024

Most Popular

By Luke Edwards 28 February 2024

By Sascha Zuger 27 February 2024

By Stephanie Smith Budhai, Ph.D. 26 February 2024

By Luke Edwards 23 February 2024

By Luke Edwards 20 February 2024

By Steve Baule 20 February 2024

By Erik Ofgang 15 February 2024

By Stephanie Smith Budhai, Ph.D. 15 February 2024

By Luke Edwards 12 February 2024

By Erik Ofgang 7 February 2024

By Erik Ofgang 2 February 2024

Unveiling the GRASPS Assessment Model: A Holistic Approach to Project-Based Learning

As educators, we strive to create authentic, meaningful, and engaging learning experiences for our students. One powerful framework that can help us achieve this goal is the GRASPS assessment model. Developed as part of the Understanding by Design framework by Grant Wiggins and Jay McTighe, GRASPS is an acronym that represents six key components of a well-designed performance task. In this blog post, we'll explore the GRASPS model in depth and offer real-world examples to inspire your teaching practice.

What is the GRASPS Assessment Model?

GRASPS stands for G oal, R ole, A udience, S ituation, P roduct , and S tandards for Success. Each of these six elements helps teachers design performance tasks that closely mirror real-life situations, promoting deeper understanding and improved learning retention. Let's break down each component and provide examples to illustrate its importance.

The goal is the clearly defined purpose of the task. It establishes what students are expected to accomplish and helps them understand the task's significance. By setting a clear goal, students can better focus their efforts and gauge their progress.

Example: In a biology class, the goal might be to "Design a sustainable ecosystem within a terrarium that supports plant and animal life."

The role defines the perspective or position that students will assume while completing the performance task. By taking on a specific role, students gain insight into different perspectives and develop empathy and understanding.

Example: In a social studies project, students might take on the role of a "United Nations representative tasked with creating a plan to improve access to clean water in a developing country."

The audience refers to the real or simulated group of people the students' work is intended for. Identifying an audience helps students consider the needs, interests, and preferences of others, ultimately resulting in more effective communication and collaboration.

Example: In a persuasive writing assignment, students might be asked to "Write a letter to the local government, advocating for the implementation of green energy initiatives in the community."

The situation provides the context and background information necessary for students to understand the task's relevance and urgency. It often describes a real-world problem, challenge, or opportunity that students must address.

Example: In a math project, students might be presented with the following situation: "Your city is experiencing a rapid increase in population, and the local government needs you to analyze population growth trends and suggest appropriate housing solutions."

The product or performance is the tangible outcome that students will create to demonstrate their understanding and ability to apply their knowledge. It can take various forms, such as written work, presentations, or physical objects.

Example: In an English literature class, students might be asked to "Create a modern adaptation of a classic play, incorporating contemporary themes and issues."

Standards for Success

Standards for success outline the specific criteria that will be used to evaluate the students' work. These criteria should be clear, measurable, and aligned with the learning goals. Providing students with a rubric or checklist can help them self-assess their progress and strive for excellence. You can also use this area to tie the model to your own curriculum standards.

Example: In a science experiment, students might be assessed on "The accuracy of their data collection, the clarity of their written report, and the effectiveness of their presentation."

Why Use the GRASPS Model?

The GRASPS assessment model offers numerous benefits for both teachers and students:

Authenticity: By simulating real-life situations, GRASPS encourages students to apply their knowledge and skills in meaningful ways.

Engagement: The model promotes active learning, as students take ownership of their learning process and work towards a clear goal.

Transferability: The skills and knowledge gained through GRASPS tasks can be more easily transferred to other contexts and situations.

Assessment: Performance tasks designed with GRASPS allow teachers to assess students' understanding and application of knowledge more effectively than traditional tests.

In conclusion, the GRASPS assessment model is a powerful tool for designing engaging and authentic performance tasks that promote deep understanding and long-lasting learning. By incorporating the six components of GRASPS into your instructional design, you can create performance tasks that not only assess content knowledge but also foster the development of essential skills, such as problem-solving, critical thinking, and communication.

All of the content published by Matt in this article is free to use. If you use anything, please show your support and follow Matt on the socials below or treat him to a coffee 😉 Matt Fletcher Creator

Instagram Twitter

Ready for more?

A powerful framework for creating authentic tasks

Select the curriculum relevant for you:

You may opt out anytime. Terms of Service and Privacy Policy .

Revisiting GRASPS: a model for project based learning

A while ago we started using GRASPS models to develop our units in MYP and DP design. I wrote about our initial work with this model a few years ago. Since then, we have used it to develop all our units in design and have noticed some meaningful results and benefits for both students and teachers.

What is GRASPS?

GRASPS is a model for demonstrating performance of understanding using authentic assessments. It is one of many performance of understanding models, but is ideally suited to the kind of project-based inquiries we do in design. GRASPS represent a framework for organizing, delivering, and assessing a project-based assessment. The assessment associated with the inquiry is structured around the following expectations and goals.

Goal : A definition of the problem or goal

Role : Define the role of the student

Audience : Identify the target audience

Situation : This is the context or scenario of the goal

Product : What is created and why it will be created

Standards : Rubrics or success criteria

Benefits of GRASPS

Over the years of organizing and implementing our units this way, we have noticed some benefits for students and teachers. Many of these observations are from the perspective of an MYP or DP classroom, but the underlying ideas would benefit any project-based learning experience.

From the teacher’s perspective, we have noticed:

Develop authentic learning experiences: The overall GRASPS structure allows us to identify more authentic learning experiences that drive our units of inquiry.

Clearer presentation of the purpose and content of a project-based inquiry: Because of the way a GRASPS inquiry is framed, communication of the goals, content, and purpose of the inquiry is clearer. During planning it is easier for teachers to plan and develop more authentic units. This has become particularly important for collaboration between teachers, as most our units are planned to be taught by several people.

Clarify the roles, perspectives, and responsibilities of students: The GRASPS model clarifies these aspects of the inquiry. Teachers can choose resources, learning experiences, and content to support the students’ development in these areas. In particular, the Role has become an important part of how we frame units to students (see below)

Communicate the expectations of the inquiry: The structure allows for clear communication of the rubric, assessment expectations, as well as the approaches to learning that students need to utilize to be successful. This has been particularly important in recent times when some of our teaching and learning has shifted to remote

Guide the selection of learning experiences, content and skills necessary for success: Through planning a unit around the GRASPS framework, teachers can think critically and creatively about the type of learning experiences that are needed to support the inquiry. We have started to look more broadly at the skills that re needed, with a particular focus on the Approaches to Learning (ATLs).

New understandings about GRASPS

Since employing GRASPS to guide our unit development, we have come to some understandings about aspects of the model that helping us strengthen the delivery of our units.

In the past, we often defined the role of the student in a very brief way - almost like a job title. You will be a a designer , engineer , marketer , etc. However we found that this often relied on student’s assumptions of what the role is. The role is very important as it defines the perspective from which the student approaches the task.

Now, we spend some time considering the role of the student in this inquiry, the skills they need, and how this role is closely connected to the Goal, Audience, and Product. For example, in a unit that defines the role as a design researcher , we spend time in class unpacking what this role entails, and how it connected to the goal, audience, and product. We discuss and highlight the skills, perspectives, and approaches that a person in this role might need to draw upon in order to be successful.

Some questions we ask in the planning stages to help us better identify and describe the role include:

What are some authentic roles that are related to the goal or discipline?

How will students understand the scope and expectations of the role?

What prior knowledge about the role will students have?

What skills and knowledge will students need to be successful in this role?

Is there a role model that students can refer to or meet in person?

The audience provides much context to the inquiry. To this end, the audience helps teachers identify, organize, and prioritize the content and skills that students need in order to meet the needs of the audience. This goes beyond just satisfying the immediate needs of the audience, but also includes understanding the audience from a user-centered design perspective and empathizing with their needs in order to develop a more successful design solution. We’ve started to use User-Task-Environment analysis as part of the research approach. In Design, this approach also supports our research goals, and helps students think more broadly about the problem.

Some guiding questions we ask include:

What is the relationship between the audience and the role?

What are the defining characteristics of the audience, and how might these influence the skills and knowledge needed by students to be successful?

Developing stronger GRASPS assessments

To support teachers I’ve created a guide to developing a GRASPS assessment and incorporating into MYP and DP units of inquiry.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Using Performance Task-GRASPS to Assess Student Performance in Higher Education Courses

2017, American Journal of Educational Research

This qualitative study explores student ability to integrate the use of knowledge and skills and demonstrates how students utilize skills in real-world situations through performance tasks using the performance task—GRASPS (Goals, Rules, Audience, Situation, Product/Performance, Standards) model. This study was carried out in the Introduction for Education course for 44 students in a teacher qualification program. Interviews, performance task-GRASPS reports, student focus group conversations, student reflections, and student products were used. Grounded theory was employed to analyze qualitative data. Findings show that students explained many educative features, including their views and beliefs toward performance tasks and authentic assessment. Students also understood their abilities through their products and reports about their roles in real-life situations. In addition, students demonstrated what they achieved and developed by themselves, and they felt happy and enjoyed their ...

Related Papers

sciepub.com SciEP

International Journal of Language Teaching and Education

Stefanie Nike Nurtjahyo

In literacy, writing becomes an essential component. It becomes the most difficult skill to be learnt because students need to express their ideas and write them accurately. Interesting learning in writing process is expected by them in term of developing their writing skill. This Classroom Action Research shows the students’ ability in exploring their writing skill in the real-world situation by using G.R.A.S.P.S. (Goals, Rules, Audience, Situation, Product/performance, Standards) strategy. This study was carried out in a private school in Surakarta. One class from social program was taken for the sampling. There were 35 students who contributed to the study. Observation, questionnaire, recording, writing assessment reports were used. Findings show that: (1) G.R.A.S.P.S strategy can motivate the students in expressing their ideas; (2) The interesting teaching-learning process in the classroom makes students feel enjoyed because of their roles in real-life situations. It can low pr...

AsTEN Journal of Teacher Education

Enriqueta D Reston

Within the framework of Constructive Alignment in Outcomes-based Teaching and Learning (OBTL), we examined the connections between teaching practice and student outcomes in the implementation of authentic performance assessment in pre-service teacher education. Using the case study design, we described the interactions of teaching-learning and assessment in the implementation of an authentic assessment task based on a Five-Dimensional Framework. From the lens of critical reflections on teaching practice, we found that authentic assessment provided a holistic approach to classroom assessment where formative assessments are embedded within a summative performance task. Students found the activity a meaningful learning experience as they produced quality products and performances that relate to their future role as professional teachers. We conclude on the importance of authentic assessments in aligning teaching and assessment with student learning outcomes and recommend the use of organizing frameworks in both planning and reflection of teaching and assessment practices

Within the framework of Constructive Alignment in Outcomes-based Teaching and Learning (OBTL), we examined the connections between teaching practice and student outcomes in the implementation of authentic performance assessment in pre-service teacher education. Using the case study design, we described the interactions of teaching-learning and assessment in the implementation of an authentic assessment task based on a Five-Dimensional Framework. From the lens of critical reflections on teaching practice, we found that authentic assessment provided a holistic approach to classroom assessment where formative assessments are embedded within a summative performance task. Students found the activity a meaningful learning experience as they produced quality products and performances that relate to their future role as professional teachers. We conclude on the importance of authentic assessments in aligning teaching and assessment with student learning outcomes and recommend the use of org...

Eddie Fetalvero , Virgil Duad

Understanding by Design (UbD) curriculum framework is now adopted by most secondary schools in the Philippines to improve the quality of education. The design is composed of three stages: identification of desired results; determination of acceptable evidence of understanding; and planning of learning experiences and instruction. This contrasts with the traditional practices by which teacher begins with textbooks, favored lessons and time-honored activities rather than deriving them from targeted goals or standards. This paper presents procedures in developing assessment tasks and rubrics in science subjects for Stage 2 of UbD. It also addresses some important issues in selecting assessment tasks; designing their cognitive structures to ensure that learning can occur while students are being assessed; and developing, validating and testing the reliability of rubrics.

Pete Swanson

مجلة کلیة التربیة. بورسعید

Hager Al-Tonsi

Angelo Chiarle

Within the medical research, actually the present contribution could be considered a kind of “Case Report”, which can be considered the product of nearly sixteen years of teaching both Italian and Latin language and literature just in two Licei Scientifici Statali in the Province of Turin, to students aged 14-19. The result of ten years of experimentations begun attending a first training course on Cooperative Learning given in Turin by Mario Comoglio, the author of the first Italian book on this subject (1996), and carried on in the following years with several other courses (and related books) about Understanding by Design, Authentic Assessment, Student Portfolio, Learning Communities, Differentiated Instruction, Habits of Mind. And more recently, I could enrich my xperience and grasp various insights working as teachers’ trainer within a group of colleagues from all kinds of schools, a group responsible for innovation projects within the Turin district.

Maria Heizel S Agujar

ABSTRACT Name : Salayo-Agujar, Maria Heizel T. Title : Designing Curriculum Plan in Secondary Mathematics IV Using UbD Framework Key concepts : Understanding by Design (UbD), Enduring Ideas, Essential Understandings Essential Questions, Knowledge and Skills, Performance Task, and Learning Plan Degree : Master of Education Major Subject : Mathematics Adviser : Dr. Rene R. Belecina Statement of Purpose This study aimed to design a curriculum plan in teaching Mathematics IV using Understanding by Design. Specifically, the following were the objectives of the study: 1. Develop a curriculum plan in Secondary Mathematics IV using Understanding by Design based on the 2002 BEC Philippine Secondary Learning Competencies. 2. Establish the content validity of the proposed curriculum plan. 3. Evaluate and finalize the curriculum plan in Mathematics IV. Research Design The study utilized the descriptive developmental research method. It established the development of a proposed curriculum plan in secondary mathematics IV using UbD. METHODOLOGY The developed curriculum plan underwent the different stages of development: (a) Planning Stage, (b) Writing Stage, (c) Validation Stage, and (d) Evaluation Stage. Planning Stage The curriculum designer conducted a survey of related literature and existing materials. The gathered information served as guide in identifying the contents of the curriculum plan based on the 2002 BEC Philippine Secondary Learning Competencies. Consultations with several experts in the field were also considered. It was during this stage, that the specific format of the curriculum plan was decided. Writing Stage Based on the chosen format, the curriculum plan was formulated. Drafts of the materials were presented to the adviser for the review of the contents, length of the lessons, clarity of directions, and the suitability of the language used. Consultation from UbD Experts and Mathematics teachers were also done on this stage. Likewise, the rating instrument was developed and validated. There were two research instruments used in this study. The Personal Data of the Expert Form was utilized to determine the profile of the respondents such as name, gender, years of teaching experience, years of teaching mathematics, school affiliation, present position, and educational attainment. On the other hand, The Rating Instrument of the Curriculum Plan is the evaluation checklist developed based on the guidance of the adviser and UbD experts who validated the rating instrument. Scoring ranged from 1 to 5 with 5 as the highest. It aims to validate curriculum plan based on Content, Format and Design, Clarity, and Usefulness. Validation Stage The prepared curriculum plan was developed based primarily on the 2002 BEC Philippine Secondary Learning Competencies and Understanding by Design as a curriculum framework. The comments and suggestions of the UbD consultants and other Mathematics teachers were used to revise the material. Evaluation Stage The revised curriculum plan was further evaluated by nine (9) experts in the field using the rating instrument that was developed on the first stage. Their comments and suggestions were analyzed and served as basis for revision and improvement of the proposed curriculum plan. Statistical Treatment of Data The statistical tool used in the treatment of data was the computation of the means of responses using Microsoft Excel to analyze and interpret the data, and to provide answers to the specified problems. The five-point Likert Scale with was used as a guide for data interpretation. Summary of Findings The data obtained were presented and interpreted in the sequence by which the problems were posted in the study. 1. The UbD curriculum plan is made up of the following parts: (a) Stage 1 – Desired Results; (b) Stage 2 – Assessment Evidences; and (c) Stage 3 – Learning Plan. The UbD Curriculum plan was developed around the 2002 BEC Philippine Secondary Learning Competencies. It has four Units: (a) Unit 1 - Linear and Quadratic Function; (b) Unit 2 – Polynomials, Exponential and Logarithmic Function; (c) Unit 3 - Circular Function and Trigonometry; and (d) Unit 4 – Statistics and Probability. 2. The developed curriculum plan underwent the following stages: (a) Planning Stage; (b) Writing Stage; (c) Validation Stage; and (d) Evaluation Stage. 3. The Mathematics and UbD experts rated the curriculum plan in all aspects of the quality elements, namely: (a) content; (b) format and design; (c) clarity; and (d) usefulness. Mean scores in these aspects were 4.47, 4.45, 4.43, and 4.52 respectively. The overall mean score is 4.47. The verbal interpretations are all Outstanding. Conclusion In light of the findings derived from the study, the following conclusions are drawn. 1. The necessary quality elements of a highly commendable instructional material are present in the developed curriculum plan in Secondary Mathematics IV using Understanding by Design (UbD) based on the experts judgments. 2. The curriculum plan will most likely be useful to teachers who teach Mathematics IV or Advance Algebra and Trigonometry. Recommendation In the light of the findings and conclusions, the following recommendations are given. 1. The developed curriculum plan in this study on Secondary Mathematics IV Using Understanding by Design should be experimentally validated to test its effectiveness. 2. Teachers can try to utilize the UbD as a curriculum plan in teaching Mathematics. 3. Seminars and workshops on UbD should be offered by the department to all teachers, department heads, and principals to have better understanding about the incorporation of UbD curriculum framework in Secondary Mathematics IV. 4. Demonstration teachings using UbD Curriculum Framework in the learning area should be conducted as often as possible. 5. The development of educational materials should be accompanied by the teachers’ trainings, workshops and lectures to uphold the morale of teachers in the field. 6. The UbD curriculum plan prepared by the author be utilized by the mathematics teachers and become a part of their teaching in New Era University. 7. Similar studies should be undertaken in this, and in other areas of Mathematics for high school level.

Steve Ferrara

RELATED PAPERS

Dr. Anthony Bordoh (PhD., M.Phil., B.Ed., Tr. Cert 'A')

Richard Dean Rubrica

Ivana Fratter

Assessment & Evaluation in Higher Education

Carola Bruna Jofré

Research on Humanities and Social Sciences

Isaac Eshun

Sonja Upton

Theory and Practice in Language Studies

Ligia Fernanda Espinosa Cevallos

Sharon Sullivan Rubin

Yeni Agustin

Educational Leadership

Jay McTighe

Choice Reviews Online

rita sanchez

Daniel Light

MejaBundar Channel

Michele Dornisch

Bulletin du Groupe français des argiles

Elen Roaldset

New Trends and Issues Proceedings on Humanities and Social Sciences

Δώρα Παπαγεωργίου

International Journal of Elementary Education

New Trends and Issues Proceedings on Humanities and Social Sciences (PROSOC)

Sharon Friesen

MEd Scholarly Project, Teaching English Language Learners, University of North Dakota

Nick Kasparek

2020 ASEE Virtual Annual Conference Content Access Proceedings

Ala Al-Fuqaha

Journal of Education and Practice

Abdallah Ghaicha

Siri Reynolds

Dr. Bryan R. Drost , Anita C Levine

Gifted Child Today

Melissa Matusevich

Clark Dominic Alipasa

Anushree Priyadarshini

John Kendall

Dr. Isaac Eshun (Ph.D., M.Phil., B.Ed., Tr. Cert 'A')

Academia Lasalliana Journal of Education and Humanities

Trần Khánh Hoa

carol oberg

Connie Hale

Susan Eliason

Journal of Case Studies in Accreditation and Assessment

RELATED TOPICS

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

GRASPS Assessment Design and Student Metacognition

The why of grasps assessment design.

GRASPS is a model advocated for by Grant Wiggins and Jay McTighe to guide teachers in designing authentic performance-based assessment. It’s a form of assessment that engages learners to employ their thinking skills and demonstrate application of essential knowledge, conceptual understanding, and skills acquired throughout a unit of learning. 

Wiggins defined authentic assessment as “…Engaging and worthy problems or questions of importance, in which students must use knowledge to fashion performances effectively and creatively. The tasks are either replica of or analogous to the kinds of problems faced by adult citizens and consumers or professionals in the field.” (1993, qtd. by Jon Mueller).   

The main takeaway for me is that teachers can use the GRASPS assessment model to:

• engage students through contextualized learning;
• provide simulations of real-world situations or challenges that adults might encounter;
• create opportunities for students to practice transfer of learning; 
• foster curiosity and building experiences of students; 
• develop  project management skills  of students . 

The  WHAT  of GRASPS assessment model 

To help educators construct authentic assessment, Wiggins and McTighe’s came up with GRASPS model. GRASPS is an acronym for teachers to: 

• G oal: establish the challenge, issue or problem to solve; 
• R ole: give students a role that they might be taking in a familiar real-life situation; 
• A udience: identify the target audience whom students are solving the problem for or creating the product for; 
• S ituation: create the scenario or explain the context of the situation; 
• P roduct/ P erformance and  P urpose: paint a clear picture of the  WHAT  and  WHY  of the product creation or the performance;   
• S tandards & Criteria for Success: inform students how their work will be assessed by the assumed audience.    

Is the GRASPS assessment model misunderstood?​

A set of sentence stems has been provided to help teachers construct a performance task and often is introduced in IB workshops. It might be because limited time was allotted for teachers to explore thoroughly the designing principles of using the GRASPS assessment model; therefore, the summative task is sometimes described in the format of a GRASPS performance task but fails to illustrate an actual real-world problem or issue that can inspire students to take authentic or simulated action on.

An example might be: 

• G oal: Your goal is to write a short story.  
• R ole: You are a middle school student. 
• A udience: Your target audience is your teacher, and students and parents in our school community. 
• S ituation: You have been asked by your school community to write a short story. (This section is sometimes omitted by teachers as a clear situation is not identified.) 
• P roduct/ P erformance and  P urpose: write a 800 word short story to entertain others. 
• S tandards & Criteria for Success: You will be assessed against criteria B, C and D.​

It might look good at first by framing the assessment through GRASPS model, but it’s like déjà vu all over again. It is definitely a step-up when teachers begin to use this model when creating a summative assessment task. However, this is still very much like a traditional assessment task. First of all, the range of the target audience is too big. The way an author writes to entertain young children, teens, or adults is very different. The situation described above is unlikely to happen as it is vague and more details are needed.  

Another issue that needs to be addressed is that MYP teachers often inform students that they will be assessed against criteria B, C, and D. But what do criteria B, C and D mean? If we want students to organize, produce text, and use language (MYP Language and Literature criteria BCD), wouldn’t it be more effective for teachers to clearly specify the criteria and engage students in understanding the assessment objectives and strands? We can’t expect students to develop assessment capabilities without explicitly involving them in developing assessment literacy.

Develop student metacognition through GRASPS 

In my humble opinion, through the use of the GRASPS assessment model, we can also create opportunities for students to develop their metacognition. In order to create a product or solve a problem effectively and efficiently, students first need to clarify the task, identify their strengths and weakness, set appropriate, challenging goals, analyze the context, chunk the big task into small subtasks within the timeline, seek feedback for improvement, and self-evaluate their work against the success criteria before the final submission. During the process of product creation, teachers provide both explicit and implicit feedback and guide students to monitor their progress. Frequent check-ins are essential. It should never be the case that teachers give students a big project and only find out that students have not addressed a requirement one or two days before the due date.

Refocus GRASPS implementation 

Teachers and students can both benefit from the use of the GRASPS assessment model. In this poster design, I refocused the use of the GRASPS assessment model and created essential questions respectively to guide teachers in designing the GRASPS authentic assessment, and students in developing their metacognition through conducting the GRASPS assessment. 

As mentioned previously, teachers do not always set up a clear situation for the assessment task. In the MYP framework, when illustrating the situation or creating the scenario for the task, teachers can refer back to the MYP global context exploration predetermined. It is also through the careful design of the scenario or situation, students can be challenged to think about intercultural communication and thus develop international-mindedness.   

This poster can be downloaded as a PDF file by simply clicking on the image below.    

Mueller, Jon. “What Is Authentic Assessment?”  Authentic Assessment Toolbox ,  jfmueller.faculty.noctrl.edu/toolbox/whatisit.htm#definitions .

Spencer, John. “Five Structures for Helping Students Learn Project Management.”  John Spencer , 20 Aug. 2019,  www.spencerauthor.com/project-management/ .

Quigley , Alex, et al. “ Metacognition and Self-Regulated Learning .”  Education Endowment Foundation , (Education Endowment Foundation), 27 Apr. 2018, educationendowmentfoundation.org.uk/tools/guidance-reports/metacognition-and-self-regulated-learning/.

​基于整体表现的GRASPS评估设计与学生元认知

如果希望閱讀本篇文章的中文版，請拜訪 www.sohu.com/a/403764639_120362876 ​謝謝IB教學研的翻譯。

Related Posts

Use Twitter for Professional Development

6 Reflection sticky notes

MYP and DP language acquisition alignment

Develop student assessment literacy

Making assessment criteria comprehensible to increase student agency

Inquiry Learning through Wonder Day Project

3 thoughts on “grasps assessment design and student metacognition”.

Really interesting !!!

Hi Alison, Thank you for this comprehensive summary of GRASPS, very helpful when reflecting on assessment practices. I noticed that the graphic model posted above is actually linked to your Sanity-Saving Feedback Strategies. Would you please share the correct link with me? I world really appreciate it. Thank you, Karli Lomax, [email protected]

Dear Karli, Thank you for your message. 😀 I just fixed the link. If you click the image, the correct PDF will be downloaded. Best, Alison

Leave a Comment Cancel Reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

Privacy Overview

• Special Issues
• Conferences
• Turkish Journal of Analysis and Number Theory Home
• Current Issue
• Browse Articles
• Editorial Board
• Abstracting and Indexing
• Aims and Scope
• American Journal of Educational Research
• International Journal of Celiac Disease
• American Journal of Medical Case Reports
• American Journal of Public Health Research
• World Journal of Agricultural Research
• Turkish Journal of Analysis and Number Theory
• Social Science
• Medicine & Healthcare
• Earth & Environmental
• Agriculture & Food Sciences
• Business, Management & Economics
• Biomedical & Life Science
• Mathematics & Physics
• Engineering & Technology
• Materials Science & Metallurgy
• Quick Submission
• Apply for Editorial Position
• Propose a special issue
• Launch a new journal
• Authors & Referees
• Advertisers
• Open Access

• Full-Text PDF
• Full-Text HTML
• Full-Text Epub
• Full-Text XML
• Nuha Iter. Using Performance Task-GRASPS to Assess Student Performance in Higher Education Courses. American Journal of Educational Research . Vol. 5, No. 5, 2017, pp 552-558. http://pubs.sciepub.com/education/5/5/12 ">Normal Style
• Iter, Nuha. 'Using Performance Task-GRASPS to Assess Student Performance in Higher Education Courses.' American Journal of Educational Research 5.5 (2017): 552-558. ">MLA Style
• Iter, N. (2017). Using Performance Task-GRASPS to Assess Student Performance in Higher Education Courses. American Journal of Educational Research , 5 (5), 552-558. ">APA Style
• Iter, Nuha. 'Using Performance Task-GRASPS to Assess Student Performance in Higher Education Courses.' American Journal of Educational Research 5, no. 5 (2017): 552-558. ">Chicago Style

Using Performance Task-GRASPS to Assess Student Performance in Higher Education Courses

This qualitative study explores student ability to integrate the use of knowledge and skills and demonstrates how students utilize skills in real-world situations through performance tasks using the performance task — GRASP S (Goals, Rules, Audience, Situation, Product/Performance, Standards) model. This study was carried out in the Introduction for Education course for 44 students in a teacher qualification program. Interviews, performance task-GRASPS reports, student focus group conversations, student reflections, and student products were used. Grounded theory was employed to analyze qualitative data. Findings show that students explained many educative features, including their views and beliefs toward performance tasks and authentic assessment. Students also understood their abilities through their products and reports about their roles in real-life situations. In addition, students demonstrated what they achieved and developed by themselves, and they felt happy and enjoyed their roles in real-life situations. The students reported that the evaluation method improved their self-confidence. Diversity was observed among the products and performances; students addressed the same challenges differently. This method develops the performance of university professors in authentic assessment by establishing performance tasks and using various rubrics to assess various products. These findings indicate that teacher educators must use authentic assessments and performance tasks to make students interactive in courses and utilize rubrics in evaluation that provide students real description of their performances.

1. Introduction

Students should be knowledge producers and not only knowledge consumers or keepers. The question is how this can be achieved. Allowing students to experience challenges when facing real-life situations and solve these problems enables them to produce solutions, manage situations, and develop different perspectives. These approaches integrate knowledge and skills in various ways. Thus, methodologies must be developed to assess well-being from the personal perspective of children and young individuals that can contribute useful, relevant, and reliable data; these data can then be a basis for policy formulation, implementation, and evaluation (O’Toole, & Kropf, 2012). Authentic assessment recognizes human rights, because this type of assessment is suitable to evaluate inner diversity, intelligences, abilities, and learning styles. Motivational theories such as Vroom’s expectancy theory (1964) states that individuals are motivated by three beliefs. First, individuals must feel that their level of effort will lead to a corresponding level of performance. Self-efficacy or a person’s belief in their own ability to achieve their desired goal (Gist & Mitchell, 1992) is the most critical component of this model. Individuals without a high level of self-efficacy will not be motivated. The second component of expectancy theory links motivation and outcome. The most critical type of learning ensures that performance is measured fairly and accurately, minimizing bias. Expectancy theory also suggests that individuals must value the reward. Whether or not a student actually values a high grade may be out of the educator’s control; nevertheless, this insight may help diagnose a particularly low level of motivation 11 . Adam’s equity theory (1965) posits that people maintain a fair relationship between performance and rewards in comparison with others. Adams (1965) states that views of justice are related to inputs, outputs, and social comparison. Inputs are contributions that are used to obtain a certain type of return on a personal investment. Contributions can involve time, effort, skills, and determination 13 . Assessment is a rich source of feedback for students. Authentic assessment is effective, because it allows an educator to provide positive feedback in a more motivational form than the usual numerical grade in a test (Litchfield, Mata, & Gray 2007). For most university instructors, authentic assessment is a radical paradigm shift from teacher-centered to student-centered teaching and learning. The majority of instructors continue to teach the way they were taught, namely, via lectures and objective tests. Any one form of teaching or assessment is insufficient to adequately teach a subject or measure learning progress and student performance. Most university courses consist primarily of three components: lectures, traditional assessments, and assignments. Assignments lag behind and are often few during a semester given that the standard objective testing is often used. A course is usually composed of a midterm and a final exam 9 . Wiggins and McTighe (1998) said, “evaluation, assessment, performance tasks, and other acceptable evidence” are used for evaluative purposes, and their linear model goes directly from desired results to determine acceptable evidence. Determining acceptable evidence involves completing sets of assessment methods, such as (a) performance tasks or projects, (b) quizzes, tests, academic prompts, (c) informal observations, discussions, and (d) student self-assessments 4 . Backward design theory is more likely to be the same as an assessment model, in which the politics of those in areas of measurement and psychometrics are prioritized and thus legitimated. Backward assessment criticizes traditional paper-pencil tests, including standardized tests. Backward theory posits that classroom teachers should be aware of the potential for student engagement as part of their design consideration 4 . The hallmark of backward curriculum theory is its great emphasis on assessment and its focus on how student learning increased. In this backward planning model, the entity of assessment is prioritized in which teachers are seen as assessors as opposed to developers (Wiggins & McTighe, 1998). Performance task is aligned with one or more desired results, which will yield appropriate evidence of the identified understanding. Involving complex and real-world (i.e., “authentic”) applications of the identified knowledge, skills, and understanding written in the goal, role, audience, situation, and product (GRASPS) form allows students to demonstrate understanding with some options in the performances and/or products. The performance task is meant to assess and requires one or more of the six facets of understandings. The scoring rubrics includes distinct traits of understanding and successful performance. The scoring rubric highlight what is appropriate, given the evidence needs suggested by the desired results 18 . Performance tasks using GRASPS are outlined below. The culminating activities that the students produce are the products that are based on the goal of a performance task. Each task contains between five and eight products that represent cross curricular topics.

Performance task-GRASPS is a design tool to develop a performance task with an emphasis on context and role playing. The acronym stands for the steps in the process, which include goals, roles, audience, situation, product-performance-purpose, and standards, which are the criteria developed for success. The GRASPS design tool includes a stem statement that a teacher can construct in a scenario for a performance task (Mayes, & Myers, 2015). McTighe 12 said that through the defined STEM performance task editor a teacher can edit a task, remove or add products, or up-load other pertinent information to the task. Rubrics are designed for each task for each type of product. The GRASPS frame includes real-world goals, meaningful roles of students, authentic or simulated real-world audience, and a contextualized situation that involves real-world applications. Students generated culminating products and performance, and consensus- driven performance standards (criteria) are used to determine success. Performance tasks with these features provide meaningful learning targets for learners, worthy performance goals for teaching, and the kind of evidence needed to assess true understanding (Tomlinson, & McTighe, 2006), Biggs and Tang 1 are presented standards model of assessment “that designed to assess changes in performance as a result of learning, for the purpose of seeing what, and how well, something has been learned. Such assessment is criterion-referenced (CRA), that is, the results of assessment are reported in terms of how well an individual meets the criteria of learning that have been set”. In this study these tasks show how a student uses math and science in a real-life situation rather than just providing information on a student’s theoretical knowledge. A performance task rates a student’s learning process, and assessing both product and process provides an accurate profile of a student’s ability and makes them value their work processes and products. This paradigm of assessment gives them an opportunity to apply self-monitoring, self-reflection, and self- evaluation using rubrics and reflection journals. Similarly, Frey et al. 7 have a review for significant studies of assessment indicates the following characteristics that promote learning- oriented assessment and employability:

1. Tasks should be challenging, demanding higher order learning and integration of learning and integration of learning from both the university and other contexts such as work-based settings;

2. Learning and assessment should be integrated, assessment should not come at the end of learning but should be part of the learning process

3. Assessment should encourage metacognition, promoting thinking about the learning process not just the learning outcome.

4. Tasks should be involve the active engagement of students developing the capacity to find things out for themselves and learn independently.

5. Tasks should be authentic, worthwhile, relevant and offering students some level of control over their work;

6. Tasks are fit for purpose and align with important learning outcomes.

Assessment refers to the act of determining the extent to which the desired results are achieved and to what extent they have been achieved. Assessment is the umbrella term for the deliberate use of various methods of gathering evidence of meeting desired results, whether these results are state-content standards or local curricular objectives. The collected evidence we seek may include observations and dialogues, traditional quizzes and tests, performance tasks and projects as well as students “self-assessments” gathered over time 19 , 20 . Thus, assessment is a more learning-focused term than evaluation, and the two concepts should not be viewed as synonymous. Assessment is the process of providing and using feedback against standards for improvement and to meet goals. By contrast, evaluation is more summative and credential-related than assessment. We need not give a grade—an evaluation—to everything we give feedback to. A central premise of our argument is that understanding can be developed and evoked only through multiple methods of ongoing assessment with far greater attention paid to formative (and performance) assessment than typical 18 , 19 , 20 . An extended performance task may develop into a project. A project adapted from Wiggins and McTighe (1999, 2004) is described as follows: “A project is an extended and complex performance task, usually occurring over a period of time. Projects usually involve extensive student inquiry culminating in pupil products and performances which are assessed using a variety of assessment tools.” Tomlinson and McTighe show in their book that educators find understanding by design addresses their need “a model that acknowledges the centrality of standards but also ensures that students truly understand content and can apply it in meaningful ways”; they find it increasingly difficult to ignore the diversity of the learners in their classrooms. For many educators, differentiated instruction offers a framework to address learner variance as a critical component of instructional planning (Tomlinson, & McTighe, 2006).

2. Research Problem

Students in higher education suffer from the way they are assessed, as described by Tomlinson and McTighe (2006), who said that both teaching and learning were redirected in ways that are potentially impoverishing for those who teach and those who learn. The traditional model is usually used, which includes a first exam, second exam, and final exam to assess the achievement of their students. These exams aim to know what students know and how much they know. This type of assessment does not consider that students have different ideas about one point, and exams do not provide ideas about their mistakes. Educators struggle in assessing different students in various ways to measure their in-depth understanding on the concepts and to make students utilize their skills. The research problem concerns changing the assessment paradigm from the traditional one to one that explores students to understand the content and apply it in meaningful ways as well as finding a frame of assessment that considers the diversity of students. “Because they allow students to construct or perform an original response rather than just recognizing a potentially right answer out of a list provided, performance assessments can measure students’ cognitive thinking and reasoning skills and their ability to apply knowledge to solve realistic, meaningful problems” 5 , a new paradigm of assessment in higher education, such as performance task-GRASPS, is used in this study to address the main problem of this research.

“Performance assessments are common in high-achieving countries, which have long relied on open-ended items and tasks that require students to analyze, apply knowledge, and write extensively” 5 , then students need to explore the different ways to enhance their effectiveness in response to any task given to them. Students in this course are future teachers, and, thus, they must develop good assessment practices and determine the basis of formative assessments. The Introduction for Education course can support this approach by using a new paradigm in formative assessment and by integrating contextually situated evaluation activities that help professors and students develop and improve their assessment practices. Professor will be learning with their students, whose productivity and engagement in the new paradigm in the assessment process will improve.

3. Purpose of the Study

Assessment method must mirror that which is considered to be important, this is when students “learn for life” at the same time as they focus on passing the course, assessment methods govern what student learn, also govern how the students study(summative, or formative assessment), and teachers plan the assessment will also affected when student study 6 . Biggs & Tang 1 said that backwash can be a positive force if only the assessment method is constructively aligned with the learning outcomes and teaching and learning activates. Accordingly the qualitative study examines how students reflect and interact with the new paradigm in assessment based on performance tasks. A qualitative design was used to examine how performance tasks are used to assess student knowledge and skills in teacher qualification program courses and determine how (44) students implement their performance tasks in real-life situations while recognizing the diversity in student products. The research specifically examined students’ interaction with and reflections on the use of backward assessment. Qualitative research approach was used to generate detailed and in-depth data to answer the research questions. This method allowed the researcher to create an interpretive analysis for the students, interactions, and reflections with the performance tasks.

All these aspects are crucial in the authentic assessment of performance based-tasks. Therefore, the research question is about how students accept and react to the new paradigm of assessment and towards their individual performance task?

3. Research Approach and Design

A qualitative design was used to examine how students accept and react to the new paradigm of assessment and towards their individual performance task. The empirical data in this paper were obtained from a case study, which develops evaluation processes in a teacher qualification program in Palestinian Technical University in the Introduction to Education course. The case study was conducted when a teacher qualification program was introduced for the first time in academic year 2014–2015. Teacher qualification program offers certification of teaching to individuals who completed all requirements, obtained a bachelor’s degree, and completed a teacher preparation program.

Many courses merely include a midterm and a final exam 9 . However, the prospective teachers in this program want to practice different approaches of assessing learning. Thus, various assessment processes had to changed or improved to meet the needs of student teachers when they eventually become teachers in their classrooms. The question here is what types of authentic assessment is respectful and meaningful. The shift in the assessment from traditional assessment paradigm to authentic assessment paradigm (Figure 1) is described below.

• Fig ure 1 . Assessment paradigms components

In this case, the instructor designed performance tasks using the three following simple steps:

Step 1: Identify the goals (integrate between knowledge and skills) that the pupils are expected to reach in each teaching unit.

Step 2: Set the tasks that will demonstrate the language knowledge and skills that were developed.

Step 3: Develop explicit performance criteria and expected performance levels measuring pupils’ mastery of skills and knowledge (rubrics). A rubric is a scoring tool that outlines the required criteria for a piece of work or the important aspect to assess. It also indicates the weight for each criterion based on its relative importance to the overall task and describes what the performance would look like at different quality levels 18 .

Performance task designee:

The researcher is the teacher educator for the course in the teacher qualification program. She establishes a performance task to assess student understanding for education, teaching, and learning in schools and to reform their perceptions towards global teachers. The sample consisted of 44 introduction for education course students. All of them are female, (30) of them pre-service science teachers, the other’s (14) pre-service math teachers, and they participated voluntarily in this research, though they had no prior experience with this type of tasks. The following steps are followed to implement this study:

1. Modified the scale of the course grades and used 40% of the grade as the performance tasks

2. Followed the steps of building these tasks to face one real-life challenge in their real life, such developing high-quality teachers who are able to change schools to meet global demands and 21st century learning or other real-life challenges

3. Table 1 lists the performance task demonstrated to the students.

Table 1. Performance task-GRASPS scenario

• Download as PowerPoint Slide Tables index View option Full Size Next Table

4. The challenge, which is a real-life problem, is given to the students and discussed with them for 60 minutes.

5. The performance assessment task steps and template are given to students and discussed with them by allowing them to share their ideas about each step.

6. The rubric related to the standards is distributed to the students for them to learn while building their product.

7. The task is implemented individually and in groups.

8. Each student chose the role suite based on his or her personal, knowledge, skills, and attitudes.

9. All students start working on their tasks with a high degree of responsibility.

10.(44) Students who attend the introduction for educational course in the first semester 2014/2015 finalized their tasks individually or in dyads.

11.One month is spent to finalize their tasks.

12.Rubrics are used to evaluate the products which differ from each other.

13.The students are given feedback according to the rubric by stating their weak and strong points.

4. Data Collection

The products, which were sent by all the students before the deadline, were classified based on the type of products, performance, and purpose of their task using two types of rubrics. The first, which is commonly used, is presented in Table 2 . The second rubric suits the content of the products. The students are allowed to present their products, and all of them exhibited high positive attitudes towards their work. The students know what they do and the purpose of their work. They use passion statements to say how much they appreciate their work, because they chose their role and the type of products for the first time and they finalized their actual work. Individual interviews were conducted to explore their experiences in this task, and the students were asked to write their statements. Questions include the following: How does this task differ from other tasks? Are you happy with this work? Why? How do you distinguish your work from others? What do you recommend? Data were collected from three types of resources, namely, students’ products, notes about their presentation, and their statements from the interviews.

Table 2. Task standards and rubric

• Download as PowerPoint Slide Tables index View option Full Size Previous Table

5. Data Coding and Analyzing

The data were initially coded from different sources using words, phrases, and descriptive codes. To improve the validity and the reliability of the research, a meeting was initiated between one faculty member in the teacher qualification program and a group of students to discuss the coding tables and the draft of the findings. Through the meeting, the final interpretation of the results was confirmed and agreed on. The students’ interviews and reflections were analyzed using grounded theory following these procedures: organization, familiarization, coding, and categorization for their words and the products of (44) students (pre-service teachers). The following statements are sample from student’s reflections:

• Tasneem said, “I’m very happy with this task because this is the first time I collected information by myself and expressed about my point of view.”

• Israa noted, “This is the first time I explored that I am able to write more paragraphs by myself.”

• Anwar said, “I’m happy because I practiced my role in a real life situation.”

• According to Mays, “This is the first time I wrote everything from my analysis and observations without using Google.”

• Layl said, “This is the first time that I feel that I understand everything, and I can discuss with others.”

• Jamila and Mariam stated, “It is amazing to write everything by ourselves, and this is the first time that we depended on our ability. We’ll repeat this experience.”

• Hadeel said, “I will not forget this challenge because I dealt with all the processes.”

• Duaa stated, “I’m interested and happy with my products even though it is not high quality, because it is the first time I worked by myself.”

• According to Muna, “This task enhanced my confidence.”

The following statements indicate the findings.

• (18) Students practice their role in real-life situations, and (26) students practice their role in role playing.

• The student products include poster (1), classroom sessions in different specializations (6), research reports (18), workshop reports (3), university media lecture report (1), formal meeting minutes (1), university president decision (simulation by one student) (1), and action research reports (6).

• Their grades are distributed as shown in the following chart:

The figure above shows that the grades of the students are high between (28–38) degrees, and the majority of the students (124) obtained (30) degrees.

• Most of the feedback given to most of students are as follows:

- Errors of grammar and usage make the meaning unclear.

- Language style and word choice are ineffective and /or inappropriate.

- Introductions, transitions, and other connecting materials may be lacking or unsuccessful.

- Any abrupt transitions do not interfere with the intended meaning.

- Details are lacking.

- Information may include some inaccuracies.

• Most skills were achieved by students within the following tasks:

- Implementing interviews with teachers, deans, principals, parents, students, student teachers, and supervisors.

- Implementing classroom observation

- Simulating characters such as educational minister, PTUK president, dean, and trainer

- Analyzing qualitative collecting data: points of views, opinions, and observations

- Making conclusions.

- Building future visions and recommendations.

- Organizing reports.

• The attitudes of students changed; the students’ opinions about this task and the challenge were positive by the time of completion.

7. Conclusion

1. Various rubrics can be used to assess different products.

2. Students said that they can make products without depending on the internet.

3. Students said that they feel happy and enjoy their roles in real-life situations.

4. Students said that the evaluation approach improved their self-confidence.

5. Students take different roles to face a challenge and created a report about their experiment.

6. Diverse products and performances were achieved.

8. Recommendation

1. GRASPS should be utilized to evaluate performance tasks in all courses.

2. Student recommendations must be considered in reforming the course content.

3. Their results can be used as data to reform policies and practices related to the qualification program.

4. Teacher educators use authentic assessments and performance tasks to make students interactive in the courses.

5. Using rubrics in evaluation provide real descriptions to students about their performances.

Cite this article:

Normal style, chicago style.

• Google-plus

• View in article Full Size Figure
• View in article Full Size

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
• Data Descriptor
• Open access
• Published: 29 May 2018

Human grasping database for activities of daily living with depth, color and kinematic data streams

• Artur Saudabayev 1 ,
• Zhanibek Rysbek 1 ,
• Raykhan Khassenova 1 &
• Huseyin Atakan Varol 1  

Scientific Data volume  5 , Article number:  180101 ( 2018 ) Cite this article

7806 Accesses

32 Citations

2 Altmetric

Metrics details

• Biomedical engineering
• Electrical and electronic engineering

An Erratum to this article was published on 24 July 2018

This paper presents a grasping database collected from multiple human subjects for activities of daily living in unstructured environments. The main strength of this database is the use of three different sensing modalities: color images from a head-mounted action camera, distance data from a depth sensor on the dominant arm and upper body kinematic data acquired from an inertial motion capture suit. 3826 grasps were identified in the data collected during 9-hours of experiments. The grasps were grouped according to a hierarchical taxonomy into 35 different grasp types. The database contains information related to each grasp and associated sensor data acquired from the three sensor modalities. We also provide our data annotation software written in Matlab as an open-source tool. The size of the database is 172 GB. We believe this database can be used as a stepping stone to develop big data and machine learning techniques for grasping and manipulation with potential applications in rehabilitation robotics and intelligent automation.

Machine-accessible metadata file describing the reported data (ISA-Tab format)

Similar content being viewed by others

A large calibrated database of hand movements and grasps kinematics

Néstor J. Jarque-Bou, Manfredo Atzori & Henning Müller

HANDdata – first-person dataset including proximity and kinematics measurements from reach-to-grasp actions

Enzo Mastinu, Anna Coletti, … Christian Cipriani

A kinematic, imaging and electromyography dataset for human muscular manipulability index prediction

Óscar G. Hernández, Jose M. Lopez-Castellanos, … Francisco Gomez-Donoso

Background and Summary

Understanding and replicating human hand grasping and manipulation have been long-running objectives of researchers from multiple disciplines. Qualitative improvements in both will trigger rapid advances in a large number of domains including industrial automation, humanoid robotics, medical rehabilitation and prosthetics. Contemporary anthropomorphic multi-fingered robot hands have achieved levels of dexterity, agility and object manipulation comparable to that of human hands 1 , 2 . Successful in narrow task-specific applications, these artificial systems, however, still fail to deliver the full functional range of a human hand. While novel sensors 3 , actuators and batteries have been continuously driving the progress, control of artificial hands remains non-trivial and challenging. Control systems link the sensing and actuation components of a robot hand. In prosthetics, control inputs originate in a user, and one common way of acquiring the intention is based on electromyography 4 , where electric signals from the arm musculature are classified into one of set of discrete grasp patterns. Implementation of a grasp is then executed by hardware-specific controllers which generate trajectories for different robot hand segments. Rather limited in their versatility and generalization capabilities, current control systems only enable a limited set of operations, such as opening, closing and several programmable actions enabled by straightforward sequential controllers 5 .

Grasp-synthesis models depend on understanding the mechanistic complexity of the human hand, its sensory-motor control loop, and the relation of grasp taxonomy to object geometry. Emergence of data-driven motion planners emphasized the need for multi-modal high-fidelity data enabling thorough analysis of human grasping. Among data acquisition techniques, glove-based systems deliver information straight from a human hand with high spatial resolution 6 . The widespread use of RGB-Depth cameras facilitated fine-grained activity analysis including hand tracking and hand-object interaction 7 . Conjunction of multiple sensing modalities, e.g. augmentation of RGB-Depth data into optical hand motion tracking system 8 proved superior to single mode sensory data. Observation of human hand activity is not limited only to limb tracking. The human body consists of connected segments, and the kinematic configuration of the human body was used to predict a dominant hand location 9 . Ye et al. 10 also demonstrated that hand motion can be predicted from upper body inertial measurement data.

Generation and annotation of a versatile database for grasp planning is an arduous task. Researchers usually seek balance in data representation and focus on one of the components of the grasp planning, e.g. target object shape information. The Columbia Grasp Dataset 11 exemplifies this tendency, where Goldfeder et al . introduced a large labeled dataset of scaled 3D object models and baseline grasps feasible for each of them. They then classified unknown objects to the most similar ones in terms of geometric shape. Another common approach is video recording of human activity and subsequent annotation into discrete actions. Bullock et al . 12 used a head-mounted video camera to record regular work activities of two housekeepers and two machinists. The authors presented the 27.7 hours-long Yale Human Grasping dataset with annotated object, grasp and task types. The dataset was instrumental for revealing the frequencies of common grasps in household and machine shop settings 13 . Recent projects on human hand utilization use novel sensors and expand the knowledge application domain. Kleinhans et al. reported on an ongoing project which acquires RGB-Depth images along with geometric description of objects to create a database of successful and failed grasps 14 . To benefit from crowdsourcing in data generation, Kent and Chernova introduced a user-friendly and intuitive web-based interface for grasp learning by demonstration 15 .

In this work, we introduce a multisensor hand grasping dataset which expands the scope of the observations and human activity range compared to the prior works. The main contribution of our paper is in the dataset generated using three sensing modalities, namely RGB camera, depth sensor and upper body inertial motion capture suit. We conducted experimental sessions for data gathering with multiple human subjects executing activities of daily living (ADLs). These activities included food preparation, housekeeping, folding clothes and ironing. Subsequently, the data streams were synchronized and grasps were annotated. Based on our survey of related works, we state that the presented dataset is structurally correct and useful for multi-faceted analysis and application in grasp planning, prediction and evaluation. Moreover, it can be employed to test object recognition and classification algorithms in machine learning and computer vision.

Study Participants

The announcement inviting participants for our study was distributed throughout the Nazarbayev University campus. 13 participants (five females and eight males) expressed their intention to participate in the experiments. Their ages ranged from 19 to 42. At the moment of the experiments, all participants reported no known hand or arm injury, or other health issues which could affect their performance. All 13 participants were right-handed. Before engaging in experiments, each subject was comprehensively briefed about the procedure, introduced to the experimental hardware and informed of any potential risks. Additionally, we provided a written description of the experiments and required participants to sign an informed consent form. The study and experiments were carried out in accordance with principles of the Declaration of Helsinki 16 , and approved by the Institutional Research Ethics Committee of Nazarbayev University, Kazakhstan. The study participant wearing the data acquisition setup in Fig. 1a consented to have her picture included in this publication.

( a ) User wearing the hardware setup with major components (Participant consented to the use of her photo in this paper). ( b ) Example snapshots from RGB and Depth sensors. ( c ) Schematic depiction of IMU distribution in upper body XSENS suit configuration. ( d ) Example capture of signals acquired from Right Shoulder segment.

“Activities of Daily Living” is a commonly used term in rehabilitation and occupational therapy, referring to set of everyday tasks critical for unassisted living. Recently, the term gained wider usage amongst the robotics community with similar connotations–to enable evaluation of an artificial system’s performance for daily tasks. ADLs were then sub-categorized to accommodate different application domains. Domestic activities of daily living (DADL) encompass tasks regularly performed in human living environments, e.g. housekeeping and food preparation. Extradomestic activities of daily living (EADL) cover tasks systematically performed outside of home 17 , such as driving or shopping.

Our experimental procedure was comprised of three sessions. Each was designed to encompass a wide range of common activities performed at a household, e.g. cleaning routines, dealing with cutlery, food items, clothes and special equipment such as an iron. Each experiment hence included DADL tasks, which were generalized to food preparation, housekeeping and ironing/clothes folding activities. The exact sequence of DADL experiments is as follows:

cooking breakfast and cleaning the involved kitchen areas

housework activities, e.g. wiping the dust and vacuum cleaning

clothes folding and ironing

Experiments took place in a two-room apartment. All activities were performed in a large room which combines the living room and the kitchen. Before the experiment, subjects were introduced to the general outline of the three experiments, provided with high-level task characterization and thorough instructions about the hardware involved. The duration of each of the experiments and approach to its accomplishment were decided by participants (e.g. subjects were asked to cook breakfast, but decided themselves on what the meal will be and which cutlery to use). Each participant then provided age, gender and a dominant hand information, signed a consent form and were given time to feel themselves comfortable with the setup. Participants were advised to stop the experiment whenever they feel uncomfortable, or with the first sign of fatigue. Two subjects decided to conclude the procedure before starting the laundry activities, with the other eleven finishing the full DADL routine.

Data acquisition and data quality

It took each subject between 30–50 min to finish data acquisition. Total duration of experimental data obtained is around 9 h.

During data acquisition, we encountered instances of dropped depth frames and connection loss with the Xsens motion capture suit. Both occurred sporadically, however, missing data in the two data streams caused mismatch and time shift in data frames, which convoluted the synchronization of sensor channels. The problem aggravated with the increased duration of an experiment recorded in a single-shot, i.e. the number of frame shifts and data mismatches increased and were harder to systemically identify. To tackle this issue, we decided to record experiments in time-limited chunks (up to 10 min sessions). This allowed us to reduce the number of dropped depth frames and broken connections with the motion capture setup (by approximately 90%). Hence, the discrepancy between matching events in RGB and depth data was reduced. Only in two 10-minute chunks, the connection with the motion capture suit was lost causing missing IMU data (subject 7 – Data Segment #5 and subject 9 – Data Segment #1), these instances were denoted in the dataset summary spreadsheet ( Data Citation 1 ) submitted along with the manuscript. Importantly, annotating discrete and compact data chunks proved more convenient to manage.

Before the beginning of an experimental session, the subjects were notified that data acquisition would be conducted in approximately 10 min chunks and they would be notified by the experimenter prior to this time instant, such that when they are notified by the experimenter, they are instructed to conclude their current activity but not to start a new one. During the experiments, the 10-minute constraint was regulated by the experimenter. Towards the 10-minute mark, the subjects would be notified and asked to halt their current activity so to prevent mid-action cuts. With the new recording, subjects commence their current action and proceed. Additionally, operators could terminate the recording intentionally, prior to the expiration of 10 min. This happened when subjects were caught in extended idle states – for example waiting for food in the pan to reach a certain condition.

Data Acquisition Setup

During the experiments, we performed simultaneous acquisition of: (i) Video material from GoPro Hero 4 camera, (ii) Depth and confidence images from SoftKinetic DS325 RGB-Depth camera, and (iii) Upper-body inertial motion data from Xsens MVN motion capture suit (see Fig. 1 ).

A GoPro Hero 4 Silver action camera was attached on top of the head of a participant facing forward slightly inclined downward to capture subject’s hand operational area. The weight of the camera is 83 grams. It is 59 mm in length and 41 mm in height. The data was acquired at 30 frames per second with 1280×720 pixels resolution, and recorded to the camera’s internal storage.

Depth images were acquired using RGB-Depth camera DS325 (SoftKinetic). We located the depth camera to the middle of the ventral-side of the right forearm of a subject. Depending on the forearm size of a subject, the distance between the radiocarpal joint and the camera was in the range of 11–15 cm making the distance between the depth camera and the center of the palm 16 to 22 cm. The camera attached to a 3D-printed forearm brace has a 6 cm offset from the arm surface to the camera sensor. The setup is aligned with the arm frontal plane, however the anatomical variance in forearm thickness results in a slight inclination towards the hand workspace region (around 5 degrees with the arm frontal plane). This configuration enabled us to capture most of the grasping activity with the camera field of view of 74°×58°×87° (H×V×D) and the nominal working range of 0.15 – 1 m.

Additionally, the DS325 camera provides a confidence image along with each depth frame. Confidence images express the magnitude of infrared light inflicting on the detector, hence expressing the level of “reliability” of depth measurement in the corresponding pixel of a depth map. Both depth and corresponding confidence images were recorded with 320×240 pixels resolution at 30 frames per second frame. The camera was connected to the HP EliteBook 8560W laptop (Windows 7 operating system, Intel Core i7 processor), which was used for both acquisition and storage of range data.

Inertial motion data

Upper-body inertial motion data was recorded using an Xsens MVN suit 18 . The suit is comprised of 17 sensors dividing the body into 23 segments and 22 joints. In this work we only used the upper body configuration of the Xsens MVN, which includes 11 sensors and omits segments and joints related to legs, feet and toes. Each sensor has a 3-axis gyroscope, accelerometer and magnetometer. The sensors are placed on a user’s body as indicated in Fig. 1c . The processed raw data from these sensors enables derivation of position, angular velocity and quaternion data for each body segment. With position and angular velocity expressed as triplets with respect to x- , y- and z -axis, and four quaternion components for segment orientation, there are ten records per each of the upper body segments. Data from the inertial measurement unit sensors was acquired at 120 frames per second and transmitted to the HP EliteBook 8560W laptop (Windows 7 operating system, Intel Core i7 processor) via a USB connection.

Data Annotation

Grasp definition.

After acquisition, the experimental data was visually analyzed for annotation and mapping by human experts. To avoid conceptual ambiguity, we adopted the definition of the grasp as established by Feix et al . 19 .

“A grasp is every static hand posture with which an object can be held securely with one hand.”

This definition excludes any intrinsic in-hand motion and object manipulation. We further identified discrete grasp actions, similar to the concept of elementary grasp actions employed by Vergara et al . 20 the starting point of each grasp action was established when a human annotator detects contact of the subject’s hand with an object. The end point of the action was established when the subject released an object or performed another grasp.

The rare cases when a grasp is followed by another grasp without the release of the initially grasped object are handled using the notion of grasp ‘sequence number’. These sequence numbers are included in the annotation data provided as an Excel spreadsheet (seqNum column) along with the dataset. We record every grasp and assign to it a sequence number. The end of the grasp is identified at the moment when an object is released or in-hand repositioning occurs and a new static hand posture is formed.

As an example, consider the case when a grasp is engaged (normally using fingers) while the object acquired during the previous grasp is still in hand (e.g. fixed in palm). At such instances, we ended the preceding grasp (let’s refer to it as grasp 1 ) and denoted the start frame of a new grasp with the new sequence number (grasp 2 ). The initially grasped object in all observed cases was repositioned in hand or manipulated. If the subject finishes the grasp 2 and re-grasps the object from the grasp 1 after manipulation, it would result in a new record grasp 3 .

Annotation and synchronization

Grasp annotation was performed by observing RGB video from the action camera stream. The grasp start frame was identified as the moment of contact of subject hand with an object. In case of occlusion (present in less than five percent of the grasps), the start frame is selected by estimation based on the observation of the depth frames. Grasp end frame is recorded when the subject releases an object, or transits to a different static grasp (which also becomes the start frame of the subsequent grasp). Video, range and inertial motion data were recorded asynchronously, and segments of each data stream started at different time instants. Manual synchronization of RGB, inertial motion and depth data was performed using a custom written software. Specifically, grasp events identified in the action camera videos were mapped to the corresponding frames in the depth and inertial motion data. The start and end frames of each grasp were then recorded, respectively. Additionally, we provide a graphical user interface which allows annotation and visualization of all three streams and the means to access specific frames in each of the channels. The depth stream was visualized using the Jet color map of Matlab (see Fig. 1b ). Only the right-hand grasps (dominant hand of each participant) were considered.

Grasp annotation and synchronization were performed by two researchers with engineering backgrounds. The data was split into two roughly equal parts and annotated separately. Subsequently, the annotators reviewed the results together with a third expert to resolve ambiguous grasps with majority voting. Timeline outside of the grasps was left untagged. Total number of grasp actions annotated in this study is 3826.

Grasp taxonomy and assumptions

To organize and encompass the complexity of the high number of grasping patterns, it is common to arrange them into a taxonomy 21 . This facilitates structural and functional relation analysis of grasps, and provides researchers with powerful analytical tools. As an example of the latter, Feix et al. 19 performed hierarchical structuring of 33 grasps acquired from the literature and reduced them to 17. Quantitative narrowing of the grasp space while retaining its functional gamut might reduce the complexity of artificial limbs with minimal functionality loss.

Grasp taxonomy reduction offers a tradeoff between simplification and representation capability specific to the application domain and/or use case. In our work, we wanted to provide a wider selection of grasps, and the corresponding sensory and statistical data. This way, we intended to make the dataset useful for researchers from multiple domains. Especially, we considered the field of prosthetics, and development of controllers for artificial hands. It is essential for this domain to have major non-prehensile actions to be included into the taxonomy. Non-prehensile hand movements do not result in force or form closure. They might utilize the fingers or the hand in general, however, they do not result in an object acquisition or capturing by the hand 22 . Therefore, any hand movement which aimed at pushing (button press, moving objects on a surface by pushing them with an open palm, pushing a door to close it, etc.) or lifting (holding an item on a palm without achieving a form closure) were considered as push and lift grasp types, correspondingly.

We followed the original non-reduced taxonomy provided by Feix et al . 19 In addition to these 33 grasps, we added two common non-prehensile configurations – push and lift , obtaining 35 grasps in total (see Fig. 2 ). Furthermore, we employed two assumptions from their work 19 for the grasp annotation. Specifically, we omitted bimanual tasks, which are possible only with the application of two hands (e.g. two-hand bedsheet folding), and we applied the notion of grasp sequence number to deal with sequence of grasps transitioning through in-hand manipulation.

Taxonomy of grasps classified and depicted according to thumb position, hand and finger configuration.

Code availability

The Matlab code used for annotation of the experiments and assigning grasp labels is available at https://github.com/zhanibekrysbek/Annotation-software-of-Human-Grasping-Database . Furthermore, we are providing the Matlab code allowing visualization of data from the three data streams – RGB, depth and inertial motion capture suit. Visualization software is available at https://github.com/zhanibekrysbek/visualization_software_human_grasping_database . Both of the repositories are accompanied with detailed usage instructions.

Data Records

Data records introduced in this paper along with the dataset summary and the grasp annotation file are available through the Figshare repository ( Data Citation 1 ). Dataset summary spreadsheet provides information on human subjects’ gender and age, as well as it locates relevant files for each participant in the dataset. The overall size of the data is 172 GB. It was arranged into five main directories and archived: three archived files for depth data (Depth_1.zip, Depth_2.zip and Depth_3.zip), RGB (GoPro.zip) and inertial motion data (Xsens.zip). Each of the five directories consists of human subject data of the corresponding nature (i.e. depth, RGB and inertial motion) acquired during experiments. We de-identified human subjects, and assigned unique identifiers for the directory names. For instance, a folder named ‘1. Subject 1’ in the Depth_1 directory will contain all depth stream data acquired from subject 1 during experiments. ‘Subject 4’ folder in the GoPro directory will contain all video material collected with GoPro action camera from subject 4 during experiments. As it was mentioned in the Procedure section, there were examples of intentional termination of experimental activities. For example, subject 7 stopped the experiment by their own will before commencing laundry activities, hence the corresponding subject files only contain data for food preparation and housekeeping exercises.

We selected an example single episode from the entire dataset – an excerpt of approximately 10 min duration with RGB, depth and motion data records. The episode is archived and submitted along with the dataset as ZIP archive ( sample_record.zip ). The size of the sample record is around 3.2 GB.

Missing data

There are instances of missing inertial motion capture data. Due to communication errors, some packages were dropped during transmission from the Xsens motion capture suit, making an entire record segment inconsistent. These segments were not included in the dataset, and the corresponding fields in the Data Summary spreadsheet ( Data Citation 1 ) are denoted as ‘Missing Data’. Additionally, depth stream also experiences missing data. For example, the system would acquire 28 frames instead of 30 frames per second. The direct implication of such behavior is the shift in frame numbers for grasp intervals with RGB and inertial motion streams. The magnitude of such shift was minimized by limiting the experiments interval to less than 10 min, and then neutralized by synchronization and manual mapping. In the annotation file, the difference between start and end frames for grasps in RGB and depth modes can be observed. In general, this missing data does not incur any significant information loss and is not a limiting factor.

Five main data directories contain action camera videos (RGB), depth and confidence videos and inertial motion data. Action camera videos are stored as Audio Video Interleaved (AVI) files, depth and confidence images in the Motion JPEG 200 (MJ2) format, and inertial motion data as comma-separated values (CSV) files. As mentioned in the Methods section, the length of the experiments varied across participants and acquisition was performed in segments of no longer than ten-minute duration. This way, for every subject, there are different number of files available.

Data acquired from the upper-body configuration of the Xsens motion capture suit was arranged into 227 data columns and stored as a CSV file. The first 150 columns contain the data from 15 segments including Pelvis, Right Hand, Right Forearm, Right Upper Arm, Right Shoulder, Neck, Head, etc. Each consecutive ten fields out of 150 represent three position, four orientation quaternion and three angular velocity values of each segment. Subsequently, there are 77 columns, which contain three gyroscope and four quaternion values for each of the eleven sensors of the upper body suit such as Pelvis, Head, Right Hand, Right Upper Arm, etc.

Annotation records

The annotation information is provided as a CSV file in the root of the Figshare dataset ( Data Citation 1 ). There are 12 columns in the annotation file. The file indicates the ID of the grasp and the participant to whom it belongs. In addition, the file identifies the grasp type as identified from the taxonomy, task type as annotated by human experts and start and end frames of the episode to locate them in the GoPro video files. Finally, the annotation describes the grasp properties including opposition type, power level and thumb position. From first to last, the columns represent the values for Grasp ID, Participant ID, Grasp Type, ADL, Video File Name, Start Frame GoPro, End Frame Gopro, Opposition Type, Power level, Virtual Fingers, Thumb configuration, Duration(in sec), Start Frame Depth, End Frame Depth, Sequence Number .

Technical Validation

In this section, we provide statistical evaluation of the data acquired from 13 participants during almost nine hours of experiments annotated by human experts. We first analyze the frequency and duration data of the grasp dataset. Subsequently, we compare subjects’ experimental data during specific ADLs, and outline average grasp statistics for each of the DADLs. In order to validate the acquired data, we compare our results with the prior literature (Vergara et al. 20 and Bullock et al. 23 ), where authors evaluated statistical usage of grasps in different activities. While their grasp taxonomy and engaged ADLs differ from ours to some extent, there are many commonalities which can be drawn to bolster the validity of our data.

Out of the 3826 grasps in our dataset, 2922, 564 and 340 grasps were performed during food preparation, housekeeping and laundry activities, respectively. Table 1 illustrates the breakdown of frequency and duration of all grasps given in the context of the ADLs. Out of 35 grasps defined in the taxonomy, three grasps were not employed by the participants and not recorded during the experimental session. These are the distal type, tripod variation and ring. Therefore, Table 1 contains 32 grasps with nonzero frequencies. Sorted in the descending order by total duration, Table 1 reveals that the duration of the top ten grasps accounts for 74% of the total duration, which agrees with Bullock et al . 23 , where the duration of the top ten grasps is around 80% of total duration for a similar taxonomy. The top two grasps in our dataset are index finger extension (IndFE) and adducted thumb, with around 17 and 12% of the total grasp duration, respectively.

Distribution of power, intermediate and precision grasp types

The Feix taxonomy enables wider categorical analysis of grasps. Figure 3 shows the distribution of power, intermediate and precision grasps for specific DADLs. It can be seen that food preparation and laundry activities contrast in the proportion of precision and power grasps. For example, 56% of total grasp time in the laundry activities belongs to the power category revealing the extensive usage of iron. Oppositely, more than half of the grasps in food preparation are precision grasps such as pinch and prismatic configurations. The wide use of precision grasps for food preparation follows a similar trend to Vergara et al . 20 , where authors demonstrated that pinch grasp covered 54% of total food preparation duration. Finally, the distribution of grasps for housekeeping closely matches the results of Bullock 23 (average of 55% for power and average of 33% for precision among two housekeepers in their results compared to 52 and 33% for power and precision types, respectively, in our work). Furthermore, 69 and 27% of the total number of grasps were performed with abducted and adducted thumb, respectively. Finally, 26, 18 and 52% of grasps fell into palm, side and pad opposition types; the remaining 4% of grasps belong to the non-prehensile group (lift and push), which cannot be classified by thumb position and opposition type according the Feix taxonomy.

Distribution of Power, Intermediate and Precision grasp types for each of the three types of ADLs.

The use of top grasps by duration and frequency

The frequent use of IndFE, adducted thumb and medium wrap grasp types in our experiments is explained by the common use of power grasps engaging palm and five fingers with different thumb and finger configurations. This way, a wide set of objects of different sizes and shapes such as a knife, a vacuum cleaner holder, clothes, iron, etc. can be grasped. Similar patterns are observed in other studies. Specifically, medium wrap grasp constitutes around 11% of the total duration and around 6.5% of the total grasp count of our dataset. In the work of Bullock et al . 23 , the medium wrap is the most frequently used grasp with the duration and frequency proportions of approximately 23 and 14%, correspondingly. On the other hand, Vergara et al . 20 employ the notion of the cylindrical grasp, which functionally resembles the medium wrap used in our work. In their work, cylindrical grasp is the fourth most used grasp with the total duration proportion of around 9%.

The dissimilarity in values is explained by the different taxonomies employed in the three studies, and the difference in the experimental activities and their durations. Different ADLs can inherently result in the dominant usage of either power or precision grasps. A suitable example is the work of a machinist, who systemically handles tools. On the contrary, food preparation tasks require more precise actions, hence the frequent use of precision grasp is expected. Bullock et al . 23 conducted four experiments, comprised of housekeeping and machinist activities (two subjects for each category). Vergara et al . 20 introduced a wider range of ADLs (eight types in total, where the tasks of our food preparation class fall into feeding, housekeeping and food preparation types).

Our data for housekeeping and laundry ADLs shows the dominance of the medium wrap grasp among all participants. The high duration and frequency proportion in use of the medium wrap in housekeeping (around 45% of the housekeeping duration and 20% of the frequency) and laundry (around 38% of the laundry duration and 30% of the frequency) activities is explained by the recurrent use of this grasp for holding an iron, clothes and a vacuum cleaner handle. These actions comprise a major duration proportion of these ADLs. In Bullock et al . 23 , average results for housekeepers’ use of medium wrap is 17.5% duration and 12.5% frequency proportions. Laundry activities in our study involve handling clothes and working with an iron, which implies high number of form closure cylindrical grasps (i.e. medium wrap grasps) and this is demonstrated by even higher frequency of medium wrap usage (around 30%) as compared to housekeeping activities. Results of our study and Bullock et al. 23 clearly demonstrate the dominance of the form closure power grasps involving the palm and five fingers in housekeeping and laundry ADLs.

Food preparation ADL comparison

Food preparation experiments generated three quarters of the total number of grasps in our dataset (see Table 1 ). The two highest duration grasps in food preparation are IndFE and writing tripod (around 19% and 14% of total duration, respectively). The outline and distribution of grasp duration and frequency by the two randomly selected subjects 1 and 3 look mostly similar with the exception of few cases. Subject 1 uses inferior pincer grasp frequently compared to the frequent use of lift by subject 3. The grasp with the highest duration in both subjects is the writing tripod. Writing tripod, a variation of the pinch, is a precision grasp employing three fingers. IndFE, on the other hand, is a power grasp, which engages palm with an adducted thumb. In Vergara et al . 20 , we observe the prevalence of pinch and intermediate power-precision (IntPP), with more than 40% and 19% of the total food preparation duration, respectively. IntPP in their taxonomy is described as the grasp which engages the palm and uses the thumb and index finger to stabilize an object, matching the definition of the IndFE in our work.

Despite being the top two grasps in food preparation, there is a notable discrepancy in the percentage of the usage of the grasps (19 and 14% for IndFE and writing tripod in our work as compared to the corresponding IntPP and pinch grasps with 19 and 40%, respectively). This discrepancy is due to the differences in experimental activities undertaken by subjects and the taxonomy. Specifically, food preparation consisted of selecting ingredients, preparing them to be cooked and cooking them 20 . In addition to these activities, our food preparation session contained food consumption and cleaning (e.g. wiping the table and washing the dishes), which notably diversified the set of grasps applied by the subjects. It is also important to note the narrower selection of grasps (9 in Vergara et al . 20 versus 35 in our set), which would potentially lead to the generalization of similar grasps (e.g. variations of precision grasps).

Importantly, our work and Vergara et al . 20 revealed frequent use of analogous grasp types in food preparation ADL, which supports the validity of our results.

Housekeeping ADL comparison

Total duration of the experiments in subjects 4 and 9 differ significantly: 2978 and 1861 seconds, respectively. With longer experiment time, subject 4 applies a wider range of grasps. The difference can be observed in subject 4’s predominant use of the prismatic 4-finger and the light tool grasps with 30 and 10%, respectively, of the total grasp frequency in housekeeping ADL. Subject 9 rarely applied the aforementioned grasps (see Fig. 4 housekeeping subplot). Contrarily, subject 9 used ventral and parallel extension grasps with their count comprising around 20 and 10% of total grasp frequency count in this activity, respectively. The most distinct commonality between the two subjects’ grasp patterns is the strong prevalence of adducted thumb grasp with more than 60% of total duration for each participant. It can be also observed that subjects have similar results for usage statistics of precision 4-finger, stick, small diameter and push grasps.

Relative distribution of grasp types durations and frequencies for two randomly selected human subjects in each type of activities of daily living.

Laundry ADL comparison

Subjects 6 and 10 went through all ADLs with 2449 and 2184 seconds duration, respectively. Subject 10 finished the experiment about 4 min earlier, and their record contains less grasps. Figure 4 indicates several differences in the participants’ data. Specifically, subject 6 employed prismatic 2-finger, stick and palmar grasps for a notable duration, while subject 10 did not use these grasps for their tasks. On the contrary, lateral tripod, inferior pincer and IndFE were observed several times in subject 10, and rarely appear in subject 6. Additionally, subject 6 used the lateral grasp around four times more than subject 10 (around 21 and 5%, respectively). The variation can be explained by the difference in the set of performed tasks and the method of their execution. For instance, IndFE grasp is used by subject 10 more than 20% of the total grasp duration in laundry activity for operating an iron. The same operation was accomplished using adducted thumb grasp by subject 6 (more than 30% of total grasp duration in laundry). Usage of medium wrap, prismatic 4-finger and prismatic 3-finger grasps are nearly the same.

Grasp frequency comparison between DADLs

We also averaged total proportional durations and frequencies of grasps for each of the three DADLs across all subjects (see Fig. 5 ). The figure illustrates more frequent use of medium wrap and parallel extension in housekeeping and laundry DADLs, while precision grasps account for more than half of the food preparation grasps (around 55%). Distribution of grasp durations in housekeeping and laundry are more similar by the nature of tasks accomplished (e.g. operating appliances and manipulation with cloth).

Duration and frequency proportions for all grasps illustrated for food preparation, housekeeping and laundry tasks.

Additional information

How to cite this article: Saudabayev, A. et al. Human grasping database for activities of daily living with depth, color and kinematic data streams. Sci. Data 5:180101 doi: 10.1038/sdata.2018.101 (2018).

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

Grebenstein, M. et al. The hand of the DLR hand arm system: Designed for interaction. The International Journal of Robotics Research 31 , 1531–1555 (2012).

Article   Google Scholar  

Palli, G. et al. The DEXMART hand: Mechatronic design and experimental evaluation of synergy-based control for human-like grasping. The International Journal of Robotics Research 33 , 799–824 (2014).

Saudabayev, A. & Varol, H. A. Sensors for robotic hands: A survey of state of the art. IEEE Access 3 , 1765–1782 (2015).

Farina, D. et al. The extraction of neural information from the surface EMG for the control of upper-limb prostheses: Emerging avenues and challenges. IEEE Transactions on Neural Systems and Rehabilitation Engineering 22 , 797–809 (2014).

Atzori, M. et al. Electromyography data for non-invasive naturally-controlled robotic hand prostheses. Scientific Data 1 , 140053 (2014).

Dipietro, L., Sabatini, A. M. & Dario, P. A survey of glove-based systems and their applications. IEEE Transactions on Systems, Man, and Cybernetics 38 , 461–482 (2008).

Lei, J., Ren, X. & Fox, D. Fine-grained kitchen activity recognition using RGB-D. Proceedings of the ACM Conference on Ubiquitous Computing , 208–211 (2012).

Zhao, W., Chai, J. & Xu, Y.-Q. Combining marker-based mocap and RGB-D camera for acquiring high-fidelity hand motion data. Proceedings of the ACM SIGGRAPH/Eurographics Symposium on Computer Animation , 33–42, http://dx.doi.org/10.2312/SCA/SCA12/033-042 (2012).

Kemp, C. C. A wearable system that learns a kinematic model and finds structure in everyday manipulation by using absolute orientation sensors and a camera. PhD thesis, Massachusetts Institute of Technology (2005).

Ye, Y. & Liu, C. K. Synthesis of detailed hand manipulations using contact sampling. ACM Transactions on Graphics 31 , 41 (2012).

Goldfeder, C., Ciocarlie, M., Dang, H. & Allen, P. K. The Columbia grasp database. Proceedings of the IEEE International Conference on Robotics and Automation , 1710–1716 (2009).

Bullock, I. M., Feix, T. & Dollar, A. M. The Yale human grasping dataset: Grasp, object, and task data in household and machine shop environments. The International Journal of Robotics Research 34 , 251–255 (2015).

Zheng, J. Z., De La Rosa, S. & Dollar, A. M. An investigation of grasp type and frequency in daily household and machine shop tasks. Proceedings of the IEEE International Conference on Robotics and Automation, 4169–4175 (2011).

Kleinhans, A., Rosman, B. S., Michalik, M., Tripp, B. & Detry, R. G3DB: A database of successful and failed grasps with RGB-D images, point clouds, mesh models and gripper parameters. Workshop on Robotic Hands, Grasping, and Manipulation, at the IEEE International Conference on Robotics and Automation. (2015).

Kent, D. & Chernova, S. Construction of an object manipulation database from grasp demonstrations. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 3347–3352 (2014).

World Medical Association Declaration of Helsinki. Ethical principles for medical research involving human subjects. Bulletin of the World Health Organization 79 , 373 (2001).

Google Scholar  

Matheus, K. & Dollar, A. M. Benchmarking grasping and manipulation: Properties of the objects of daily living. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 5020–5027 (2010).

Roetenberg, D., Luinge, H. & Slycke, P. XSENS MVN: Full 6DOF human motion tracking using miniature inertial sensors. Xsens Motion Technologies Technical Report 1 , 1–9 (2009).

Feix, T., Romero, J., Schmiedmayer, H.-B., Dollar, A. M. & Kragic, D. The grasp taxonomy of human grasp types. IEEE Transactions on Human-Machine Systems 46 , 66–77 (2016).

Vergara, M., Sancho-Bru, J. L., Gracia-Ibáñez, V. & Pérez-González, A. An introductory study of common grasps used by adults during performance of activities of daily living. Journal of Hand Therapy 27 , 225–234 (2014).

Cutkosky, M. R. On grasp choice, grasp models, and the design of hands for manufacturing tasks. IEEE Transactions on Robotics and Automation 5 , 269–279 (1989).

Cooper, C. Fundamentals of hand therapy: Clinical reasoning and treatment guidelines for common diagnoses of the upper extremity (Elsevier Health Sciences, 2013).

Bullock, I. M., Zheng, J. Z., De La Rosa, S., Guertler, C. & Dollar, A. M. Grasp frequency and usage in daily household and machine shop tasks. IEEE Transactions on Haptics 6 , 296–308 (2013).

Data Citations

Saudabayev, A., Rysbek, Z., Khassenova, R., & Varol, A. H. Figshare https://doi.org/10.6084/m9.figshare.c.3858397 (2017)

Download references

Acknowledgements

The authors would like to thank all subjects for their participation in the data collection experiments.

Author information

Authors and affiliations.

Nazarbayev University, School of Science and Technology, Astana, Z05H0P9, Kazakhstan

Artur Saudabayev, Zhanibek Rysbek, Raykhan Khassenova & Huseyin Atakan Varol

You can also search for this author in PubMed   Google Scholar

Contributions

A.S. created the experimental protocol, checked the data annotation, performed data analysis and wrote the paper. Z.R. created the experimental protocol, conducted the experiments, created data acquisition and annotation software packages and annotated the data. R.K. recruited the subjects, conducted experiments and annotated the data. H.A.V. ideated, planned and supervised the project and reviewed the paper.

Corresponding author

Correspondence to Huseyin Atakan Varol .

Ethics declarations

Competing interests.

The authors declare no competing interests.

ISA-Tab metadata

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 license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’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. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ The Creative Commons Public Domain Dedication waiver http://creativecommons.org/publicdomain/zero/1.0/ applies to the metadata files made available in this article.

Reprints and permissions

About this article

Cite this article.

Saudabayev, A., Rysbek, Z., Khassenova, R. et al. Human grasping database for activities of daily living with depth, color and kinematic data streams. Sci Data 5 , 180101 (2018). https://doi.org/10.1038/sdata.2018.101

Download citation

Received : 08 September 2017

Accepted : 23 March 2018

Published : 29 May 2018

DOI : https://doi.org/10.1038/sdata.2018.101

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

This article is cited by

The ha4m dataset: multi-modal monitoring of an assembly task for human action recognition in manufacturing.

• Grazia Cicirelli
• Roberto Marani
• Tiziana D’Orazio

Scientific Data (2022)

Human hand kinematic data during feeding and cooking tasks

• Alba Roda-Sales
• Margarita Vergara
• Néstor J. Jarque-Bou

Scientific Data (2019)

Loss of haptic feedback impairs control of hand posture: a study in chronically deafferented individuals when grasping and lifting objects

• R. Chris Miall
• Orna Rosenthal
• Fabrice R. Sarlegna

Experimental Brain Research (2019)

Quick links

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

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.