Breaking rules for moral reasons: Development and validation of the Prosocial and Antisocial Rule-Breaking (PARB) scale.

Abstract

To determine whether prosocial rule-breaking exists as a separate construct from antisocial rule-breaking and to develop a valid rule-breaking scale with prosocial and antisocial subscales. We hypothesized that (a) rule-breaking would have prosocial and antisocial subfactors; (b) the prosocial rule-breaking subscale would positively associate with prosocial intentions, empathy, moral identity, and guilt proneness, whereas the antisocial rule-breaking subscale would negatively associate with these same factors; and (c) the two subscales would predict prosocial and antisocial cheating behaviors, respectively. We developed the Prosocial and Antisocial Rule-Breaking (PARB) scale using a sample of 497 undergraduates (Study 1) and 257 Amazon Mechanical Turk workers (Study 2). Participants completed all surveys (Studies 1 and 2) and took part in a between-subjects experiment (Study 2) in which cheating behavior was measured in two conditions-when cheating helps others (prosocial) or oneself (antisocial). The final PARB scale demonstrated the expected factor structure (comparative fit index = .96, Tucker-Lewis index = .93, root-mean-square error of approximation = .064; χ² = 177, df = 88, p < .001), with the prosocial (α = .81) and antisocial (α = .93) subscales showing good reliability. Prosocial rule-breaking was positively associated with prosocial intentions, empathy, and guilt proneness, whereas antisocial rule-breaking was negatively associated with these same factors. Each additional point in prosocial rule-breaking PARB score predicted a 37% increased likelihood of participating in protest behavior in an exploratory investigation (p = .025) and predicted a 268% increase in actual prosocial cheating behavior (p < .001) but did not predict antisocial cheating behavior (p = .293). Conversely, each additional point in antisocial rule-breaking PARB score did not predict protest participation (p = .410) but did predict a 69% increase in actual antisocial cheating behavior (p = .025). These findings suggest that our current understanding of rule-breaking is limited, as many types of rule-breaking are prosocially motivated and are not necessarily antisocial. (PsycInfo Database Record (c) 2022 APA, all rights reserved).