Comparison of Student Performance on Open- and Closed-Response Versions of a Research-Based Assessment

Closed response assessments can be scored by computer and administered online, thus increasing the ease with which they can be given to large populations of students. We have translated the Colorado Classical Mechanics and Math Methods Instrument (CCMI) from an open-response format to a closed-response one. To determine if scores on this new, closed-form CCMI can be compared to the old, open-response CCMI, we performed a comparison of students' responses to both versions of the instrument. Preliminary findings suggest that many of the items show comparable responses in both formats, but a small number show significant differences. In this talk, we will both present preliminary results comparing the two formats and discuss characteristics of the items that translated well and those that did not. We believe that these findings will be found valuable by the assessment design community in understanding what kinds of assessment items can easily make the transition to closed- from open-response.