Visual analytics and visual audit

Nowadays, increased use of audit data analytics (ADA) is likely to be important for maintaining and enhancing the relevance and value of the financial statement audit in the new “Big Data” era (AICPA, 2017). This dissertation suggests Visual analytics (VA) as an ADA tool to promote ADA adoption in auditing and proposes a new ADA methodology, named Visual Audit, to systematically integrate VA with current practice. The first essay gives comprehensive definitions of the VA-based ADA and the mature VA tool to help the auditing industry establish a consensus on what VA is and what VA tools can be used to facilitate VA-based ADA. The essay also provides a set of factors that should be considered when formulating strategies for VA initiatives in auditing firms. Those factors, including potential and perceived challenges, are organized, presented, and discussed. The second essay develops a new ADA methodology, named Visual Audit, to explain how VA can be integrated with current audit practices. In addition, a descriptive model, named Visual Audit Model, is also created to schematically summarize the Visual Audit methodology. The model demonstrates how exploratory and confirmatory analyses are performed, how actionable insights, hypotheses, and audit evidence are obtained, how auditors can provide a desired level of assurance through human-computer interactions, and how VA systems can be utilized to communicate information accurately, efficiently, and effectively to auditors via visualization. The third essay demonstrates how the Visual Audit methodology can be applied to a real audit case. It exhibits the effectiveness of Visual Audit in exposing and communicating the essential facts and potential anomalies hidden in a large hospital dataset using the popular VA tool, Tableau. After reviewing the data set, we identify three visual audit tasks and then illustrate these tasks by working through each visual audit loop composing the job. This dissertation attempts to contribute to the auditing literature by identifying the value that VA can add to audits. It informs practitioners that, under what circumstances, VA could represent a significant change regarding what VA is, how VA-based audit is performed, and the challenges that need to be addressed before adopting VA in auditing. The proposed Visual Audit methodology and its descriptive model can have many applications. The qualitative experience presented in the paper gives implications for those interested in adopting VA in the audit practice.Ph.D.Includes bibliographical reference