| The major premise of this study is that audit committee effectiveness is a function of the power of audit committees. Broadly defined, power is "a capacity to get things done" (Stinchcombe, 1968). Through an extensive process of review and application of relevant theoretical and empirical works in the audit committee and power literatures, nine major hypotheses are developed. The independent variables used in the study are sanctionary power, legitimate power, referent power, expert power, information power, control over the agenda/awareness, coalition power, diligence, and board dominance. Four measures of perceived audit committee effectiveness are used for the dependent variables: financial reporting, the annual audit and the external auditors, internal controls and the internal auditors, and overall effectiveness.; The study used a five-phase process in exploring these relationships. First, a comprehensive literature review was performed. Second, initial interviews were conducted with people knowledgeable about audit committees. The third phase consisted of constructing a questionnaire and sending it to a sample of key people in 17 corporations. For each company, questionnaires were sent to the chief executive officer, chief financial officer, chief internal auditor, one audit committee member, and the partner in charge of the annual independent audit. Fourth, a revised questionnaire was sent to the key people in 100 corporations. Quantitative analysis was performed on the data collected. Fifth, follow-up interviews were conducted with selected respondents for additional evidence and to help interpret the results of the questionnaire.; Research results indicate that audit committee effectiveness is highly associated with referent power, expert power, information power, coalition power, and the diligence of audit committees. Strong support was also found for subhypotheses regarding sanctionary power over external and internal auditors. Moderate support was found for the hypotheses related to legitimate power and control over agenda/awareness. Finally, there was no support for the board dominance hypothesis. |
