Constructing Rubrics for Open-ended Activities ASEE Annual Conference 16 June 2002
Form Groups • This workshop includes two group activities • Form groups of 4 or 5 • Group with people you don’t yet know
Workshop Presenters • Rita Caso, Director of Assessment & Evaluation, Educational Achievement Division, College of Engineering, Texas A&M University • Ann Kenimer,Associate Professor, Department of Agricultural Engineering, Texas A&M University
Rita Caso, Texas A&M • Ph.D. Applied Research and Evaluation in Education, Counseling and Psychology • 20+ years experience in teaching, administration, research, assessment, evaluation, and accreditation-review preparation in K-12, Adult and Higher Education, in Human Services, and National Market Research. • 7 years specific experience assessing and evaluating University Level Engineering programs, and Science, Math, Engineering, and Technology [SMET] programs
Ann Kenimer, Texas A&M • B.S., M.S., Agricultural Engineering, Virginia Tech • Ph.D., Agricultural Engineering, University of Illinois • Teaches engineering design processes, fundamental problem solving, environmental engineering • FC Assessment and Evaluation involvement since 2000
Workshop Agenda • Introduction to the Foundation Coalition • Rubrics: • What is a Rubric? • How Are Rubrics Used? • Examples of Rubrics • Characteristics of a Rubric • Team Activities: • Use & Evaluate a Rubric • Develop a Rubric • Common Problems and Solutions: • Resources • Wrap up
The Foundation Coalition • Six cooperating universities: • Arizona State University • Rose-Hulman Institute of Technology • Texas A&M University • University of Alabama • University of Massachusetts—Dartmouth • University of Wisconsin—Madison • Funded by NSF
The Foundation Coalition Mission • Establish improved curricula and learning environments • Attract and retain a more demographically diverse student body • Graduate engineers who posses those transforming attributes that fully reflect the FC outcomes
Holistic Model for Engineering • Technology enhanced classrooms • Clustered students in common courses • Course integration • Teaming in the classroom • Active/cooperative learning pedagogy • Faculty team teaching • Industry in the classroom • Course modules for EC2000 • Assessment and evaluation
Qualitative Assessment Open-ended data -Content Analysis -Rubric -Check-list -Inter-rater -Intra-rater “Objective” Assessment Closed-ended data -Forced-choice response Terms Used in Workshop Pre-Determined Criteria Reliability Validity -Theoretical -Face -Criterion
What is a Rubric?(”Open-ended” Data) • It is a tool used in the qualitative assessment of “open-ended” data, such as.. • Written or oral narratives, • Diagrams or models • written or oral enumerations • behavioral demonstrations of a student’s knowledge, applied skill, or ability to perform
How Are Rubrics Used? (“Open-ended” Data) • Advantages and Drawbacks of assessing “open ended” data7 Advantages: • Can yield “rich” information (I.e., individual, creative, complex, fine-tuned) Drawbacks: • involves “subjectivity” in interpreting and scoring data (i.e. the judgments of individuals scoring) as contrasted with “objective” tests • problems with reliability (both inter-raters and intra-rater, across time)
How Are Rubrics Used? (“Open-ended” Data) • Other Methods of Qualitative Assessment used with open-ended data • Content analysis and coding10 • Inventory Checklists11 • Rubrics
What is a Rubric? (Pre-Determined Criteria) • Definition of Rubric3,9: • a systematic scoring methodology to make qualitative assessment and evaluation more reliable and objective by applying pre-determined criteria. e.g., Descriptive criteria are developed to serve as guidelines for scorers to assess, rate and judge student performance.
How Are Rubrics Used? (Diagnostic Feedback) • Descriptions of performance standards may serve to communicate to students what is expected of quality performance5 . e.g., Ideal, expected performance described in a rubric can be explicitly compared with individual performance in order to convey what aspects of performance need improvement.
How Are Rubrics Used? (Rubric Types) Rubrics may be used “holistically” or “analytically”… • “Holistic” Rubric5: • The entire response is evaluated and scored as a single performance category • “Analytical” Rubric5: • The response is evaluated with multiple descriptive criteria for multiple performance categories
How Are Rubrics Used? (Rubric Types: Example) • “Holistic”Rubric for Open-Ended Math Problems 11 • Criteria for Demonstrated Competence: (6 points) Description of Exemplary Response: • Gives a complete response with a clear, coherent,unambiguous, and elegant explanation; includes a clear and simplified diagram; communicates effectively to the identified audience; shows understanding of the problem’s mathematical ideas and processes; identifies all the important elements of the problem; may include examples and counter-examples; presents strong supporting arguments.
How Are Rubrics Used? (Rubric Types: Example..cont) • “Holistic”Rubric for Open-Ended Math Problems.. • Criteria for Inadequate Response : (2 points) Description of a Response which Begins, but Fails to Complete Problem: • Explanation is not understandable; diagram may be unclear; shows no understanding of the problem situation; may make major computational errors.
How Are Rubrics Used? (Rubric Types: Example ..cont) • “Analytical” Rubric for TIDEE Design Knowledge Test 5, 5Design Process question subcategories: Information Gathering; Problem Definition;Idea Generation; Evaluation & Decision Making; Implementation; Process Development ScoreSubcategory: Information Gathering 1……… No information gathered specifically to support design 2 3……… Information gathered primarily once or from single source; aware that information varies in quality. 4 5……… Varied sources used to obtain information; some judgment of information quality; information gathered multiple times.
How Are Rubrics Used? (Rubric Types: Example..cont) • “Analytical” Rubric for TIDEE Design Knowledge Test , Design Process question subcategories: Information Gathering; Problem Definition;Idea Generation; Evaluation & Decision Making; Implementation; Process Development Score Subcategory:Implementation 1…….. No deliverables produced or they fail to meet requirements 2 3…….. Design decisions converted to deliverables; design products meet primary requirements 4 5…….. Decisions integrated to yield design products that satisfy system requirements; products delivered on time and within allowed resources
Characteristics of a Rubric (Reliability) A good rubric must posses “reliability” • Definition of Reliability4: • the extent to which the measuring instrument yields responses that are consistent and stable across time (intra-rater) and between different scorers (inter-rater).*
Characteristics of a Rubric (Validity) A good rubric must posses “validity” • Definition of Validity1: • the extent to which what is being measured by an instrument is actually what is intended. Are the test and rubric actually measuring the desired performance-outcomes? (Construct, Criterion and Face Validity)
Team Activity I Evaluate a Rubric
Team-Based Design Communication Knowledge AssessmentTIDEE ENTERING-JUNIOR DESIGN KNOWLEDGE ASSESSMENT Instrument Short-Answer Test Item on Communication5 OBJECTIVE:Demonstrate your knowledge of key elements in the engineering design process, teamwork, and communication associated with team-based engineering design. ASSIGNMENT: Respond to the following questions/statements. You have 5 minutes. In team-based design, documentation and exchange of design information are important. Describe communication qualities and how communication occurred among team members in a Design project assignment.
Team-Based Design Communication Knowledge Assessment (Criteria) Criteria Elements: Knowledge of Effective Communication Five specific elements to be articulated by students. • Structure (i.e. organization, highly understandable, flow of thoughts) • Content (i.e. details, key points, clarity of ideas, complete and accurate information) • Relevance to audience (i.e. communicated well and understandable to audience) • Team attitude (i.e. co-operation, listening) • Involvement (i.e. planning meetings, effective interaction between members)
Team-Based Design Communication Knowledge Assessment(Scoring Rubric) 5 Points--- Student’s response shows detailed knowledge of the listed elements of effective communication if it includes 5 of 5 elements. 4 Points --- Student’s response shows above moderate knowledge, if it includes 4 elements of the effective communication. 3 Points---Student’s response shows moderate knowledge of the subject if it includes 3 elements of the effective communication . 2 Points---Student’s response shows little knowledge of the subject if it includes 2 elements of the effective communication criteria.
Team-Based Design Communication Knowledge Assessment(Scoring Rubric) 1 Point---Student’s response shows little knowledge of effective communication if only one element is indicated. 0 Points---Student’s response shows no knowledge about effective communication in a team-based design project Note: The body of the rubric provides the scale of 0-5 points with benchmarks at 0,1,2,3,4 and 5. However, the students were also scored at half values (I.e.,2.5,3.5) to provide more sensitive distinction between performance levels. A score of 5+ was given to very well articulated, comprehensive type responses
Team-Based Design Communication Knowledge Assessment(Strong Answer Example) The Test Question: In team-based design, documentation and exchange of design information are important. Describe communication qualities and how communication occurred among team members in a Design project assignment. • Each member must be able to speak and explain things clearly so that the other members should understand information well. • Ability to speak and write concisely and accurately, a member must have knowledge to convey a subject clearly. • Focus on the work on hand during team meetings and not on other things. Speak what is of interest for team members
Team-Based Design Communication Knowledge Assessment(Strong Answer Example..cont) • Open discussions and open-minded team members, willingness to compromise, listen to other ideas, be patience, and allow all members their opinions. • Equal input of ideas by each member, ask questions to clarify problems and set up meetings. SCORE= 5 Student’s response shows all the (5) elements of effective communication, thus shows detailed knowledge of the subject
Team-Based Design Communication Knowledge Assessment(Weak Answer Example) The Test Question: In team-based design, documentation and exchange of design information are important. Describe communication qualities and how communication occurred among team members in a Design project assignment • Communication should be often. Other team members should understand your point. • Common design formats should be used. SCORE= 1 Content, Relevance to Audience, Team Attitude or Involvement are not mentioned. However student has made an effort to touch upon one element of effective Communication (Structure) without detail, hence a score of 1 is given.
Team Activity I • Discussion • What did you like about the sample rubric? • What would you change?
Constructing a Rubric Note: there are two components involved in this assessment and evaluation methodology: • the test instrument given to the students • the scoring rubric used by the evaluators
Constructing a Rubric3,6,9 • Develop appropriate performance goals and objectives 2. Select the assessment tasks that reflect and demonstrate the performance goals 3. Differentiate between performance levels and assign relative values to each of the levels [establish “expert”level; establish target students’ developmental level]
Constructing a Rubric 4. Develop descriptive criteria for each level of performance which correspond with local norms . [holistic or analytical]. • Train scorers in application of rubric • Pilot both test and scoring rubric [for inter-rater & intra-rater consistency, apply cross checking methods] • Modify test items and scoring rubric based upon scoring results & content analysis of responses
Team Activity II • Develop a rubric for: • Laboratory report • Engineering design project
Team Activity II • Discussion • What does your rubric contain? • How might you apply this activity to your courses?
Validity Transferability of assessment question interpretation Transferability of specifications for expected performance Changes in Curriculum or instruction Changes in Performance standards Changes in Students’ prior knowledge Common Problems (Transferability & Repeatability ) Transferability and Repeatability of Test Questions and Rubric Criteria • Across similar or different courses • Over time, or across locales • Across populations • Across scorers
Reliability(interacts with validity) Inter-rater Intra-rater(tends to be more validity sensitive) Different Scorers Changes in Scorers’ knowledge Common Problems(Transferability & Repeatability..cont) Transferability and Repeatability of Test Questions and Rubric Criteria • Across similar or different courses • Over time, or across locales • Across populations • Across scorers
Solutions to Common Problems(Transferability & Repeatability..cont) Validity • Address.. -”Theoretical” validity2-- Review literature & other resources for precedents -”Criterion” validity2– Ask sample of experts, novices (if appropriate) and target population to respond -”Face” validity12-- Ask relevant sample of “local” users to respond and critique • Content Analyze responses & compare target population to “local” users, to experts, to novices ( if appropriate), and to rubric criteria
Solutions to Common Problems (Transferability & Repeatability..cont) Validity…cont. • Modify test questions, if necessary, as indicated by discrepancies between response content analysis results of target population and/or local users, and the rubric • Modify rubric criteria or scoring standards, to align with expert content and performance levels; or with local user content and performance levels if these differ from expert results
Solutions to Common Problems (Transferability & Repeatability..cont) Reliability--Train and manage scorers for intra-rater consistency • By having them take the test, then score their own and another scorer’s test, then justify their scoring to a third party • By having them re-view and re-score the 1st test they scored after they have completed scoring their 5th test, and • By having them review and re-score the first 5 tests scored after having completed scoring 10 tests, and continue pattern.
Solutions to Common Problems (Transferability & Repeatability..cont) Reliability--Train and manage scorers for inter-rater consistency • By duplicating a sampling of all tests and having all scorers evaluate and score each test • By having all scorers re-view each other’s scoring of this common set of test, having them discuss discrepancies, arrive at consensus on interpretation and application of rubric criteria and having them jointly re-score discrepant tests. • By having all scorers periodically and repeatedly review, each other’s scored tests, individually re-score them, then discuss, and jointly re-score two tests.
Solutions to Common Problems (Transferability & Repeatability..cont) Reliability – Controls • Halfway through the scoring job, have an outsider sample each scorer’s scored tests, and have each scorer justify his/her scoring of the same items across several tests. • Report both intra-rater inconsistencies and inter-rater inconsistencies noted to scorers for their correction • Repeat process near end of scoring job • Also calculate and examine inter-rater and intra-rater consistency rates by test subject, and by test item; as well as inter- item correlations 8
ResourcesCitation References • Bergeson, Dr. Terry. Office of Superintendent of Public Instruction web page. “Scoring the WASL Open-Ended Items” 1998. 1 May 2002 <http://www.k12.wa.us/assessment/assessproginfo/subdocuments/TechReports/g4part4.pdf> • Cronbach, Lee J., Meehl, Paul E. “Construct Validity in Psychological Tests.” Psychological Bulletin (1955). 11 June 2002. http://psychclassics.yorku.ca/Cronbach/fl • Ebert-May, Diane. “Classroom Assessment Techniques: Scoring Rubrics.” Field-tested Learning Assessment Guide (FLAG) web site 1999. 11 June 2002 <http://www.flaguide.org/cat/rubrics/rubrics1.htm> • Graduate School of Education & Information Studies. CRESST. UCLA <http://www/Rubrics/CRESSTUCLAassementglossary.html>
ResourcesCitation References • Davis D.C., Gentili K.L., Calkins D.E., Trevisan M.S. ‘Transferable Integrated Design Engineering Education (TIDEE) Project." October 1998. 29 May 2002. http://www.cea.wsu.edu/TIDEE/monograph.html • Moskal, Barbara M. “Scoring rubrics: what, when and how?” Practical Assessment, Research & Evaluation. (2000). 1 May 2002. <http://ericae.net/pare/getvn.asp?v=7&n=3> • Rowntree, Derek. Home Page. “Designing an assessment” June 2000. 11 June 2002 <http://iet.open.ac.uk/pp/D.G.F.Rowntree/derek.html> • Rudner, Lawrence M. “Reducing Errors due to the Use of Judges.” ED355254 ERIC/TM Digest (1992). 11 June 2002 <http://ericae.net/db/edo/ED355254.htm>
ResourcesCitation References • Seattle School District. “What is a rubric” (2000). 1 May 2002. <http://ttt.ssd.k12.wa.us/dwighth/rubricclass.htm> • Stemler, Steve. “An overview of content analysis.” Practical Assessment, Research, & Evaluation (2001). 11 June 2002. <http://ericae.net/pare/getvn.asp> • Summer Technology Institute at Western Washington University. “Rubric for Open-Ended Math Problems.” California CAP Math Report (1989). 11 June 2002. <http://ttt.ssd.k12.wa.us/dwighth/rubricclass.htm> • Trochim, William M.K. “Measurement Validity Types.” William M.K. Trochim Cornell University Home Page (2002). 11 June 2002. http://trochim.human.cornell.edu/kb
Wrap Up • Please complete the workshop evaluation forms • Thank you!