Small-group activities on assessment. 11/09/04

I A What are we measuring?

Why are we assessing student performance? What is/are the goal(s)? Who are the clients for the assessment (students, faculty, employers, parents, friends, ???)? What information do these constituencies want/need?

I B Grades

How is our assessment reflected in the grades that are assigned? Should grades be strictly deterministic (plug in the numbers and get a result) or subjective? Which of these is "fair"? What does "fair" mean? Should grades reflect competence at the end of the course or performance through the term? Should grades be "curved" or reflect some arbitrary standard? What happens if the tests or other requirements this term are significantly easier/harder than previous terms? Should that make a difference?

II Assessing and grading CSE courses.

For each of the following kinds of course, discuss the following:

• What sort of work is appropriate for the learning goals of the course? (Homework exercises, medium-length projects, longer term projects, tests, presentations, reports, demos, portfolios, ???)
• How do you assess the quality of the student's work? What if some of the work is done in groups instead of individually? Does that change anything?
• Does performance on some kinds of work tell you more or less about student mastery of the subject (homework vs projects vs tests, for example)?
• How should these assessments be factored into student grades?

(continued on the next page)

II A. Introductory courses (typically intro programming courses)

II B. Junior-senior level courses involving significant projects (networks, databases, compilers, operating systems, etc.)

II C. Theoretical/analytic courses (discrete math, formal languages & automata, complexity theory, etc.)

II D. Capstone courses (Full-term project courses that involve synthesis across multiple subjects instead of a single identified topic; typically involving building a large project or artifact.)