A Methodology for Generating Subquestions for the Force Concept Inventory (and Other Research-Based Assessments)

In this paper, we discuss the methodology we have developed and have been using for creating subquestions for the Force Concept Inventory (FCI). The FCI is a research-based assessment that is used internationally to assess student understanding of Newtonian mechanics. The assessment has been investigated from a number of perspectives and many suggestions have been made for its improvement. One challenge that is becoming more and more pervasive as more learning has transitioned to an online environment and more students are completing the FCI without a proctor is that of test security. If the answers to FCI items become easily accessible, then students will be able to provide correct answers despite lacking in understanding of Newtonian mechanics. One approach to mitigate the effects of items leaking into the public sphere and maintaining test security is the creation and administration of subquestions in place of the original FCI questions. Subquestions have an additional benefit of reducing false positives (answering a survey item correctly without correct understanding) and false negatives (answering incorrectly despite correct understanding). In this paper, we will discuss how we created subquestions for four items on the FCI, informed by survey-based interviews with students and the original intended targets of the items.