Purpose
This paper aims to assess internal audit quality (IAQ) by using automated textual analysis of disclosures of internal audit mechanisms in annual reports.
Design/methodology/approach
This paper uses seven text mining techniques to construct classification models that predict whether companies listed on the Athens Stock Exchange are audited by a Big 4 firm, an auditor selection that prior research finds is associated with higher IAQ. The classification accuracy of the models is compared to predictions based on financial indicators.
Findings
The results show that classification models developed using text analysis can be a promising alternative proxy in assessing IAQ. Terms, N-Grams and financial indicators of a company, as they are presented in the annual reports, can provide information on the IAQ.
Practical implications
This study offers a novel approach to assessing the IAQ by applying textual analysis techniques. These findings are important for those who oversee internal audit activities, assess internal audit performance or want to improve or evaluate internal audit systems, such as managers or audit committees. Practitioners, regulators and investors may also extract useful information on internal audit and internal auditors by using textual analysis. The insights are also relevant for external auditors who are required to consider various aspects of corporate governance, including IAQ.
Originality/value
IAQ has been the subject of thorough examination. However, this study is the first attempt, to the authors’ knowledge, to introduce an innovative text mining approach utilizing unstructured textual disclosure from annual reports to develop a proxy for IAQ. It contributes to the internal audit field literature by further exploring concerns relevant to IAQ.
Data Mining methods can be used in order to facilitate auditors to issue their opinions. Numerous of these methods have not yet been tested on the purpose of discriminating cases of qualified opinions. In this study, we employ three Data Mining classification techniques to develop models capable of identifying qualified auditors' reports. The techniques used are C4.5 Decision Tree, Multilayer Perceptron Neural Network, and Bayesian Belief Network. The sample contains 450 publicly listed, nonfinancial U.K. and Irish firms. The input vector is composed of one qualitative and several quantitative variables. The three developed models are compared in terms of their performance. Additionally, variables that are associated with qualified reports and can be used as indicators are also revealed. The results of this study can be useful to internal and external auditors and company decision-makers.
The selection of a proper auditor is driven by several factors. Here, we use three data mining classification techniques to predict the auditor choice. The methods used are Decision Trees, Neural Networks and Support Vector Machines. The developed models are compared in term of their performances. The wrapper feature selection technique is used for the Decision Tree model. Two models reveal that the level of debt is a factor that influences the auditor choice decision. This study has implications for auditors, investors, company decision makers and researchers.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.