Investigating the Effects of Exam Length on Performance and Cognitive Fatigue

Abstract

This study examined the effects of exam length on student performance and cognitive fatigue in an undergraduate biology classroom. Exams tested higher order thinking skills. To test our hypothesis, we administered standard- and extended-length high-level exams to two populations of non-majors biology students. We gathered exam performance data between conditions as well as performance on the first and second half of exams within conditions. We showed that lengthier exams led to better performance on assessment items shared between conditions, possibly lending support to the spreading activation theory. It also led to greater performance on the final exam, lending support to the testing effect in creative problem solving. Lengthier exams did not result in lower performance due to fatiguing conditions, although students perceived subjective fatigue. Implications of these findings are discussed with respect to assessment practices.

Mixed-format exams in higher education: Assessment of internal consistency reliability

“…Recently, however, Jensen et al (2013) reported α = .66 for the internal consistency reliability for student scores on a single MCQ exam in two introductory biology classes taught by one instructor. Such poor reliability implies that error is relatively high and that decisions based on the results of such MCQ exams could be somewhat compromised.…”

“…Previous studies of reliability are relatively rare in higher education (Cox, 1967;Jensen et al, 2013). Recently, however, Jensen et al (2013) reported α = .66 for the internal consistency reliability for student scores on a single MCQ exam in two introductory biology classes taught by one instructor.…”

“…Nevertheless, these are among the most commonly used types of exams in higher education. Others have reported reliabilities for essay-only exams in psychology (e.g., Dracup, 1997), problem-solving-only exams in engineering (e.g., Hill, 1978), and MCQ exams in biology (e.g., Jensen et al, 2013). Those interested in these other types of exams should consult the references cited.…”

“…The internal consistency reliability of exams in higher education has, somewhat surprisingly, rarely been reported in the literature (Jensen et al, 2013, Cox, 1967. More recently, Jensen et al (2013), in a quasi-experimental study of two introductory biology classes with 155 students in total, reported the internal consistency reliability of their MCQ exam to be quite poor, α = .66.…”

“…More recently, Jensen et al (2013), in a quasi-experimental study of two introductory biology classes with 155 students in total, reported the internal consistency reliability of their MCQ exam to be quite poor, α = .66. Although some professors have resorted to the use of MCQ tests in response to dramatic rises in class sizes, many eschew the use of such exams.…”