Measuring the Testing Mode in General Chemistry: The Effect of Computer Versus Paper Mode on Test Performance, Cognitive Load, and Scratch Paper

摘要

The purpose of this dissertation was to study the effect of testing mode (giving a test on computer versus paper) in authentic classroom environments. Specifically, this work sought to broaden the pedagogical implications of testing mode by examining variations between computer and paper mode in student test performance, cognitive load, and use of scratch paper. Data were collected from students who were enrolled in undergraduate general chemistry courses. Students took multiple proctored general chemistry tests that included a variety of chemistry topics and three question types: algorithmic, conceptual, and definition. After each test students were provided with correct-answer feedback. The tests were either graded and part of the class (i.e., class quizzes) or served as a practice test for the upcoming test in the course. To measure testing mode, tests were delivered on computer or paper, forming four conditions. The conditions, defined by the mode of the initial and final test, were: Computer - Computer, Computer - Paper, Paper - Computer, and Paper - Paper conditions. A series of studies were conducted and their findings are presented in three separate articles. Article 1 discusses test performance of two groups of students (N = 207 and N = 215) who took two similar practice tests in one session. Article 2 discusses test performance of new students (N = 221) who took two quizzes in class on Day 1 and Day 8 and one practice test on Day 12. Article 3 examines testing mode with students (N = 221) who reported their cognitive load and provided scratch paper. The results revealed that overall there seems to be little difference between online and paper-based testing. No significant differences were found in student test performance nor in the cognitive load measures between the computer and paper modes. However, a significant difference between two modes was detected with use of scratch paper, though the effect was very small (eta2 = .04). Although students used scratch paper more when taking a chemistry test on paper than online, it is not clear whether the difference of approximately one question represents any practical significance. The present dissertation supports the conclusion that online testing is a promising alternative to the traditional paper-and-pencil mode most often used in general chemistry courses and changing from paper to online mode would not impose an additional cognitive load on students.