Effect of Audit Committee Characteristics on Fraud Detection and Prevention- a Study of County Governments in Western Region

Abstract

There have been many high profile mega scandals in both the public and the private sector with regards to corruption as highlighted through print and electronic media some being brought before court of competent jurisdiction raising eyebrows on the effectiveness of the board audit committees in ensuring policies and guidelines are followed. Counties in western region have hit the headlines for several fraud cases. The main aim of this research study was to establish how audit committee characteristics influence fraud detection and prevention among counties in the western region in Kenya. The research adopted agency theory, stakeholder theory and fraud triangle theory. The study utilized a descriptive survey design. The study population was the 7 counties in the western region in Kenya. The unit of observation was the 7 departments in each of the 7 counties giving a total of 49 respondents. Data from 48 of the 49 respondents (98% response) was obtained. The study used primary data obtained from the original sources using questionnaires. The questionnaires issued via email and drop and pick method. Data obtained using questionnaires was converted from simple responsive into a quantitative form to be useful in the analysis that was done using SPSS. This process generated descriptive statistics which included frequencies and percentages and inferential statistics. A multiple linear regression model was used to show how the variables relate. Regression results produced an R-square value of 0.552, meaning that 55.2 percent of changes in fraud detection and prevention among western region county governments was the result of the four independent variables while 44.8 percent of changes was the result of other factors which are not highlighted. This research showed independent variables had a strong association with fraud detection and prevention (R=0.743). ANOVA results showed a substantial F statistic at 5% level with p=0.000. This implies that the overall regression was appropriate in explaining the influence of the independent variables on fraud detection and prevention. Findings also showed that audit committee composition, independence and expertise have a positive and significant influence on fraud detection and prevention while audit committee meetings exhibited positive but not significant influence on fraud detection and prevention. The study recommends the need for having audit committee that are more independent, diverse and with expertise as this will significant contribute towards fraud detection and prevention. The findings of this study will help western region county governments to focus on critical factors that will be useful to detect and prevent fraud. The main limitation of the study was that it covered only three aspects of audit committee characteristics and so there are other areas that require research.