The impact of safety audit timing and framing of the production outcomes on safety-related rule violations in a simulated production environment

In an experimental study, participants (n = 79) assumed the role of control room operator of a simulated plant. For a total of 36 times, they had to decide whether to violate a safety-related rule to maximise their own salary, or to comply with it at the cost of a lower income. The experimental variation was the timing of audits in the simulated production year (“early” = audits accomplished within weeks 13-30 versus “late” = audits accomplished within weeks 31-48) and framing of production outcomes and salary (gain versus loss). Results show that early-audited participants committed significantly fewer rule violations in the early period than those who had not yet been audited. Later on, the early-audited participants began to violate the rule, whereas the late-audited participants continued violating even though they had now been audited. Participants who experienced a loss-framed production outcome committed significantly more violations than participants in the gain-framed condition.