University Students' Achievement Goals and Cheating on Exams: A Person-Centered Approach

• In Event: The "Dark Side" of Motivation: Unpacking Malleable Personal Factors Associated With Academic Dishonesty

Abstract

Purpose and Framework
Academic cheating, including cheating on exams, which university administrators, faculty, and students all consider a serious concern (Barnhardt, 2016), has been extensively studied in relation to achievement goals. In general, cheating has been negatively associated with mastery-approach goals (Anderman & Koenka, 2017; Murdock & Anderman, 2006). Although performance goals are associated with competence demonstration or the avoidance of negative judgment from others and thereby assumed to be strongly predictive of cheating, the evidence regarding these relations has been inconsistent (Daumiller & Janke, 2019). A multiple goals perspective (Barron & Harackiewicz, 2001; Pintrich, 2000) which takes into consideration more than one goal simultaneously may resolve the inconsistent findings. The present study used a person-centered approach to identify university students’ multiple goals profiles and test their association with cheating on exams. The complete 2x2 achievement goals model which differentiates mastery goals to approach and avoidance was used (e.g., Elliot & Murayama, 2008) in order to provide more detailed evidence about the association of mastery goals and exam cheating.

Method
A total of 277 undergraduate Greek students (66% female) representing different university departments and achievement levels participated in the study via an online platform. A multi-method approach to measure cheating on exams was adopted including (i) a self-report scale measuring the use of exam cheating, (ii) six questions following a vignette measuring the intention to cheat on an important exam where cheating is possible, and (iii) three hypothetical vignettes measuring the use of other students’ cheating behavior. Participants also completed the Achievement Goal Questionnaire-Revised (Elliot & Murayama, 2008).

Results
Latent Class Analysis (LCA) was applied to identify distinct achievement goal clusters which were associated with cheating outcomes. The underlying latent clusters were identified using the four goals (mastery-approach/avoidance, performance-approach/avoidance) as input variables. Individuals were then allocated to latent classes using the modal assignment approach (Bakk et al., 2013). The three-class solution was indicated as the most parsimonious model in terms of BIC values (see Table 1 and Figure 1). Cluster 1 (45%) included students with all goals relatively high, Cluster 2 (34%) included students with high mastery-approach and all three other goals relatively low, and Cluster 3 (21%) included students with all four goals low. Compared to other clusters, students within the Cluster 1 were more likely to endorse greater cheating. Students in Cluster 2 were more likely to report lower levels of cheating and students in Cluster 3 were more likely to be academically dishonest (Table 2).

Significance
The above results indicate that: (i) high mastery-approach goals prevent exam cheating only when they are coupled with low performance goals, and (ii) high performance goals are significantly associated with cheating even with high mastery goals. By using a multiple-goals approach, the present study contributes to better understanding how mastery and performance goals are associated with exam cheating, indicating the profile with both high mastery and low performance goals as the most adaptive cheating profile for university students.

Authors

• Stuart A. Karabenick, University of Michigan

• Sofia Eleftheria N. Gonida, Aristotle University of Thessaloniki

• Dimitrios Stamovlasis, Aristotle University of Thessaloniki

• Maria Tzasta, Aristotle University of Thessaloniki