Citation

von der Heyde A, Brandhorst S, Kluge A. Safety Sci. 2015; 74: 160-171.

Copyright

(Copyright © 2015, Elsevier Publishing)

DOI

10.1016/j.ssci.2014.12.004

PMID

unavailable

Abstract

The present investigation was conducted in order to gain deeper insights into the determinants of safety-related rule violations. Thus, the aims were to replicate the framing of the production outcome investigated in previous studies; to ascertain the impact of the accuracy of information about audit probabilities; and to determine whether the amount of violations increases just after an audit has been experienced (bomb crater effect). The research questions were experimentally investigated using a 2 × 3 design with the factors framing (gain, loss) and accuracy of information about audit probabilities (no, vague or precise information). The participants (n = 148) were required to put themselves in the role of a control room operator of a waste water treatment plant (WaTrSim-Annual). They had to choose whether to start up the plant using the prescribed safe but non-profitable procedure (compliance) or to apply the profitable but unsafe and therefore forbidden start-up procedure (rule violation). Participants violated the rule significantly more frequently when the production outcome was loss-framed. Furthermore, it was shown that precise information about audit probabilities led to significantly more rule violations than only vague or no information about audit probabilities. The data analysis additionally revealed that participants tend to violate a safety-related rule significantly more often if they have just experienced a safety audit (bomb crater effect). To prevent rule violations, it should be avoided that people (1) think that they are not reaching their objectives (as is suggested by a loss-framed depiction of the production outcomes), and (2) receive precise information about audit probabilities. Furthermore, in order to reduce the violation rate directly after an audit, it is recommendable to occasionally conduct two consecutive audits to demonstrate that audits may occur in quick succession.