Cheating, Reactions, and Performance in Remotely Proctored Testing: An Exploratory Experimental Study

Abstract

Purpose: We sought to provide empirical insight and develop theory for a new organizational phenomenon: remote proctoring for Internet-based tests. We examined whether this technology is effective at decreasing cheating and whether it has unintended effects on test-taker reactions, performance, or selection procedures.Design/methodology/approach: Participants (582) were randomly assigned to a webcam proctored or honor code condition and completed two (one searchable, one non-searchable) cognitive ability tests online. Complete data were collected from 295 participants. We indirectly determined levels of cheating by examining the pattern of test-score differences across the two conditions. We directly measured dropout rates, test performance, and participants’ perceived tension and invasion of privacy.Findings: The use of remote proctoring was associated with more negative test-taker reactions and decreased cheating. Remote proctoring did not directly affect test performance or interact with individual differences to predict test performance or test-taker reactions.Implications: Technological advances in selection should be accompanied by empirical evidence. Although remote proctoring may be effective at decreasing cheating, it may also have unintended effects on test-taker reactions. By outlining an initial classification of remote proctoring technology, we contribute to the theoretical understanding of technology-enhanced assessment, while providing timely insight into the practice of Internet-based testing.Originality/value: We provide timely insight into the development and evaluation of remotely proctored tests. The current study utilizes a unique randomized experimental design in order to indirectly determine levels of cheating across two conditions. Following the results of the current study, we outline an integrative model for future research on remotely proctored tests.