Unformatted text preview: st of managers in the SG setting (although I cannot disentangle the
relative inﬂuences of the two forces). The null and competing alternatives are stated below.
H3 (Null): Audit levels will be the same in the WG and SG settings.
H3 (Group Effect): Auditors select the Nash or the defensive audit more frequently
in the SG settings than in the WG settings.
H4: Comparing Audit Levels across the Resolution and Regular Play Phases
Each auditor learns of his corresponding manager’s fraud levels at the end of the regular
play phase, but does not learn of his own consequent penalties. Hypothesis 4 investigates
the effect of this information on auditor choices during the resolution phase (see Table 2).
The null hypothesis predicts no difference between audit levels across the regular play
rounds and those from the resolution rounds, suggesting that the information about the
manager’s fraud level does not change the auditor’s strategy. That is, audits are in ‘‘steady
state’’ during the regular play phase (see Bloomﬁeld  for a discussion of equilibrium
View Full Document