The PACES-2 Study
The Modified Theory of Planned Behavior
They applied a modified version of Ajzen’s Theory of Planned Behavior that includes the variables of the original model (attitude toward behavior, subjective norm, perceived behavioral control, and intention) as well as measures of past behavior, demographics, behavioral context, moral obligation, and moral judgment (described by Kohlberg as the process by which an individual reasons about moral issues when presented with a moral dilemma).
PACES-2 Instruments and Sample
To validate the model, the E3 Team designed a two-part survey instrument. The first part of the instrument, the PACES-2 Survey, consists of appropriate demographic questions, items to assess the variables of the original Theory of Planned Behavior model, and items about self-reported college cheating. The survey also includes questions to address moral obligation, frequency of high school cheating, social desirability bias (measured by the Balanced Inventory of Desirable Responding (BIDR)). Because previous work by the authors highlighted the importance of context, all questions (except demographic ones) are posed in two separate contexts, exams and homework.
The second part of the instrument, the DIT-2, is a multiple-choice test that presents five moral dilemmas was originally developed by Rest. It is based on Kohlberg’s Theory of Moral Development and provides a measure of how an individual reasons when faced with a moral dilemma. Respondents with higher scores on this measure expand their understanding of justice from egocentric to societal to principled, where what is considered to be fair or morally right serves larger communities, including strangers. Respondents are asked to identify concepts important in resolving each of five dilemmas representing modern social problems. A report is generated for every respondent that includes an individual moral judgment score for each of five dilemmas as well as an average moral judgment score. The DIT-2 has been shown to have good internal and test-retest reliability and has shown discriminate validity.
The instrument was pilot tested to develop reliable, internally-consistent scales from the PACES-2 Survey and to identify how the scales relate to indices generated by the DIT-2. Subsequently, the instrument was administered to 527 engineering undergraduates at three different institutions. Because research has shown that students in humanities tend to self-report cheating at rates lower than students in other disciplines, the E3 Team has chosen to use humanities students as a reference group against which to compare engineering students. Students from the engineering and humanities disciplines who were first-year or senior-level undergraduate students participated in the study.
Findings of the PACES-2 Study
- There are differences in cheating rates between engineering and humanities students, and these differences are independent of the number of opportunities an individual student has to cheat. In the sample, the percentage of engineering students who reported cheating on a test “at least a few times they took tests during the previous term” is about twice that of humanities students (33% versus 18%). Similarly, when queried about cheating on homework, the percentage of engineering students who reporting cheating “at least a few times they worked on an assignment” is about twice that of humanities students (60% versus 36%).
- Differences in cheating rates between engineering and humanities students exist only in college, not in high school. Both groups of students reported cheating in high school at statistically identical rates. This implies that the historically higher rates of cheating reported by engineering students are more likely a result of the engineering curricula or academic environment than any inherent difference between engineering students and students from other disciplines.
- The modified Theory of Planned Behavior is appropriate for understanding why students cheat. The model accurately predicted an individual’s intention to cheat in the near future based on his/her sense of moral obligation to avoid cheating, their attitude toward cheating, their perception of norms with respect to cheating, and their moral reasoning levels.