Text readability and fraud detection
Intan Waheedah Othman, Hazlina Hasan, Roszana Tapsir, Norhafizah Abdul Rahman, Indarawati Tarmuji, Suria Majdi, Seri Ayu Masuri, Normah Binti Omar · 2012
Researchers and practitioners have employed various analytical and empirical examinations of financial data in their attempts to detect fraud. Nonetheless, despite the various comprehensive analytical examinations and other supporting techniques, there are still fraud incidences that are not discovered. There have been scarce research works that examine the readability of qualitative text in company annual report, particularly in detecting fraud. There could have been hidden clues in the narrative disclosures that can be indicators of fraud and thus, examining the narrative sections of a company's annual report may lead to discovery of fraud. Management obfuscation theory predicts that the management of a badly performed company tends to hide bad news by decreasing the readability of financial reports. Therefore, there is an imperative need for a reliable technique in measuring text complexity to detect the likelihood of fraud and deception in financial statements. Such fraud detection technique is also crucial to auditors and other stakeholders in their decision-making. This study suggests the application of a readability index tool as a proactive means in identifying fraud. The methodology suggests by this study is developed by using the Coh Metrix version 2 indices - Flesch Reading Ease Score Formula. To the best of our knowledge, no empirical study has employed the Coh Metrix-Flesch Reading index to examine text complexity to detect fraud of Malaysian companies.