Date of Conferral

2018

Degree

Ph.D.

School

Education

Advisor

Deborah Y. Bauder

Abstract

The credibility of unsupervised exams, one of the biggest challenges of e-learning, is currently maintained by proctoring. However, little has been done to determine whether expensive and inconvenient proctoring is necessary. The purpose of this quantitative study was to determine whether the use of security mechanisms, based on the taxonomy of cheating reduction techniques rooted in the fraud triangle theory, can be an effective alternative to proctoring. A quasi-experimental 1 group sequential design was used to answer the research questions whether the format, proctored versus unproctored, order in which the exams are administered, course delivery mode, and instructor make a difference in student performance. The archival scores of 850 Californian community college students on 2 sets of equivalent proctored and unproctored web-based exams in face-to-face, hybrid, and online introductory statistics courses taught by 7 instructors were compared. The format effect was tested with repeated-measures ANOVA; the order, course delivery mode and instructor effects were tested with mixed ANOVA. No significant difference in scores in Set 1, and significantly lower scores on unproctored exams in Set 2 indicated that the used security mechanisms allowed for maintaining the credibility of the exams without proctoring. There was no significant difference in scores across the course delivery modes in both sets and instructors in Set 2, but significant order effect was observed. Further research on order effect was recommended. With the use of the utilized security mechanisms education will get an inexpensive and convenient way to increase the credibility of unsupervised web-based exams, and the society will gain more online college graduates with credentials that reflect their knowledge.

Share

 
COinS