A meta‐analysis of the determinants of modified audit opinion decisions

Purpose – The purpose of this paper is to provide a meta‐analysis of the effect of: auditor and audit‐related variables; and firm‐specific variables on auditors' propensity to issue modified audit opinions. Auditor and audit‐related variables include Big N affiliation, audit firm industry specialization, audit firm and audit partner tenure, provision of non audit services and audit report lag. Some of the important firm‐specific variables include firm size, leverage, and profitability. Design/methodology/approach – The Stouffer combined test is employed as the meta‐analysis technique for this paper. The test produces a z ‐statistic that can be used to test the direction and significance of the effect of the hypothesized variables on the propensity of auditors to issue modified audit opinions. A total of 73 published studies are aggregated from 1982 to 2011. Findings – Meta‐analysis result reveals that the effect of audit and auditor‐related variables on audit opinion decisions is far from conclusive. Big N affiliation and audit report lag variables are found to be positively related while the association between non‐audit fees and modified audit opinion decisions is negative. However, the significant effect of non‐audit fee variable is found only in non US studies. Evidence on the effect of firm‐specific variables on auditors' propensity to issue modified audit opinions is broadly consistent with hypotheses formulated in the published studies. Practical implications – Meta‐analysis statistically aggregates results across individual studies and corrects for statistical artefacts like sampling and measurement error and, thereby, provides much greater precision with respect to the findings, compared with narrative reviews. The findings should be relevant for the current project on audit reporting initiated by the International Auditing and Assurance Standards Board (IAASB). Originality/value – Audit opinion formulation is a complex procedure that culminates in the issuance of appropriate audit opinions. This paper adds value to the strand of audit opinion formulation research by documenting that some of the variables are more significant in explaining auditors' modified audit opinion decisions compared to other variables.