A social contract model of ‘disintegrity’ within the dual-process paradigm of moral psychology: reducing the scope of the ‘belief-behavior incongruity’

Brad Barnhardt

PhD thesis, conferred 2015

Explaining why students cheat when it violates their moral beliefs, also called the ‘belief-behavior incongruity’ (BBI), is a difficult challenge most often overcome by referring to neutralization techniques, first described by Sykes and Matza (1957), whereby individuals deceive themselves with specious justifications for ignoring the moral imperative to follow rules. An underlying assumption of the neutralization view, that individuals’ abstract moral beliefs apply automatically to all contexts, is critiqued in the present work. The account of academic dishonesty developed herein is centered on the hypothesis that adolescent students’ felt moral obligation is informed by an intuitive sense of reciprocity between themselves and their learning contexts, which resembles a social contract, or ‘psychological teaching- learning contract’ (PTLC). Students who regard a class context or teacher more negatively are thus expected to feel less moral obligation to follow rules, and to cheat more as a result. The hypothesized PTLC model, which included key variables related to (A) self-concept, (B) achievement goal structure, (C) learning strategies, (D) moral obligation, and (E) social comparison theory, was tested with data from a diverse sample of secondary students in fifteen international schools across Asia, Europe, and Africa. A pilot study (N = 96) of the construct validity of psychometric measures was conducted prior to the Main Study, which included a Time 1 sample of N = 493, a Time 2 sample of N = 297 (spaced by approximately one year), and a longitudinal matched sample of N = 225. Structural equation modeling techniques were used to test the validity and invariance of the measurement model, as well as the structural relations hypothesized between variables. A small degree of gender non-invariance prompted separate analyses of gender-specific models. Results supported the PTLC hypothesis. Moral obligation overwhelmingly mediated the effects of perceived class quality on academic integrity, indicating that students felt morally obliged to be honest in a given class, as a function of their regard for its quality.

Supervisor: Dr Paul Ginns