Can Online Academic Integrity Instruction Affect University Students’ Perceptions of and Engagement in Academic Dishonesty? Results From a Natural Experiment in New Zealand

The problem of academic dishonesty is as old as it is widespread – dating back millennia and perpetrated by the majority of students. Attempts to promote academic integrity, by comparison, are relatively new and rare – stretching back only a few hundred years and implemented by a small fraction of schools and universities. However, the past decade has seen an increase in efforts among universities to promote academic integrity among students, particularly through the use of online courses or tutorials. Previous research has found this type of instruction to be effective in increasing students’ knowledge of academic integrity and reducing their engagement in academic dishonesty. The present study contributes to this literature with a natural experiment on the effects of the Academic Integrity Course (AIC) at The University of Auckland, which became mandatory for all students in 2015. In 2012, a convenience sample of students (n= 780) had been asked to complete a survey on their perceptions of the University’s academic integrity polices and their engagement in several forms of academic dishonesty over the past year. In 2017, the same procedures and survey were used to collect data from second sample of students (n= 608). After establishing measurement invariance across the two samples on all latent factors, analysis of variance revealed mixed support for the studies hypotheses. Unexpectedly, students who completed the AIC (i.e., the 2017 sample) reported: (1) significantly lower (not higher) levels of understanding, support, and effectiveness with respect to the University’s academic integrity policies; (2) statistically equivalent (not higher) levels of peer disapproval of academic misconduct, and; (3) significantly higher (not lower) levels of peer engagement in academic misconduct. However, results related to participants’ personal engagement in academic misconduct offered partial support for hypotheses – those who completed the AIC reported significantly lower rates of engagement on three of the eight behaviors included in the study. The implications and limitations of these findings are discussed as well as possible future directions for research.