Presenters: James J. Thompson, Professor, Louisiana State Univ. Health Sciences Ctr., Dept. Microbiology, and Sheila W. Chauvin, Professor, LSUHSC Office of Medical Education R&D
Title: Analysis of Examination Time Data at the Question- and Person-Specific Level with Perception
Synopsis:
Time impacts most human activities. In professional education, multifactorial competencies are increasing being utilized for evaluations. If competency can be viewed as defining a certain minimum level of expertise in a given domain, then a large body of literature exists on the elements of expertise. In general, experts are more facile than novices in task performance. This facility is typically associated with shorter performance times and higher quality. Therefore, we believe time is an important component in of any “authentic” competence assessment. Accordingly, it is important to understand the nature and extent of individual diversity in the time domain. There is surprisingly little peer-reviewed literature on individual time variations. Because Questionmark Perception software archives person question response times, we have initiated basic research into this topic in the simplest case (non-speeded exams).
Our initial findings have been published (Applied Measurement in Education, 22:272-289, 2009) We are trying to publish additional longitudinal analyses and speed/accuracy trade-off papers (with difficulty!). In any case, time analysis adds an important novel dimension in the interpretation of person performance. Perception software enables these analyses. With this knowledge, the Questionmark community may be able to devise new approaches/metrics that would facilitate all of our evaluation imperatives with increased authenticity.
Learning Objectives:
- Learn how to access multiple choice question time data
- Consider issues related to processing time data at the question level
- Consider the effects of covariates on time data and their significance
Recommended Experience Level: Any
