A Model to Measure the Complexity of Readability of Accounting Explanatory Disclosures

One of the research topics in the world that domestic researchers have considered in recent years is to study the readability of explanatory accounting disclosures. The reason for the entrance of this kind of research to financial and accounting literature is related to the “management obfuscation hypothesis”. Based on this hypothesis, when there is bad news and negative performance, managers tend to complicate the readability of the company's reports so that investors gradually recognize the bad news and negative performance, and as a result, its impact is gradually visible on the company's shares.Much research has been done to measure the readability of explanatory accounting disclosures and to examine the interrelationships between readability and financial reporting. These studies have used common readability assessment models used in foreign research to measure the readability of the text. Using these models in domestic research has two main limitations of non-compliance with Persian and financial languages. Accordingly, this study aims to achieve a readability measurement model with components and indicators most consistent with the Persian language and financial reports. 

In the first stage, an in-depth review of the related literature was conducted to identify the indicators and influential components of the readability of explanatory accounting disclosures. Then, from semi-structured interviews with experts, the identified components and indicators were modified from a local perspective and with the view of domestic experts. In order to validate and reach a consensus on important components and indicators, fuzzy Delphi analysis has been used. Finally, using a correlational research method, and confirmatory factor analysis method, the coefficient of the importance of each of the components and indicators of the model were extracted. 

The results of this study can be divided into two parts; The results of the first part were obtained from the literature review and interview with the experts, which include extracting and identifying components and indicators that can be used to measure the readability of a text. In this stage, 17 interviews took place with experts in accounting and linguistics. The interviews were recorded and then transcribed into written form. Then, the written forms of interviews were coded and analyzed by thematic analysis. In the second part, important indicators were identified after collecting the components and indicators affecting the readability of the text using the Delphi fuzzy method. Delphi fuzzy method is a more advanced version of the Delphi Method. It utilizes triangulation statistics to determine the distance between the levels of consensus within the expert panel. This research uses triangular fuzzy numbers of the five-Likert scale to analyze the experts' opinions. Delphi fuzzy method was conducted on a questionnaire with 35 questions and 83 experts answered. According to the results of Delphi fuzzy, all readability indicators found in the previous part were confirmed. Next, the confirmatory factor analysis method was employed to weigh the readability indicators. In this stage, two indicators of “passive verbs” and “number of pages” were omitted from the model and five indicators of “sentence length”, “tables”, “charts”, “number of letters”, and “number of syllables” were remained in the model.Based on the results of previous research and studies, four components and fifteen indicators were identified. After consulting with experts and performing fuzzy Delphi, seven indicators remained in the model. Finally, after completing the confirmatory factor analysis, the coefficients of each index were determined and three components, along with five related indicators, formed the final research model.

The results show that the three components of “text length”, “graphic signs” and “complex words”, alongside the five indicators of “sentences length”, “table”, “chart” “number of letters” and “number of syllables”, have the greatest impact on the readability of explanatory accounting disclosures. The results of this study can be used as a basis for determining the transparency rank of companies listed on the Tehran Stock Exchange. Besides, based on the results, researchers suggest using the proposed model of this research in financial and accounting research instead of using past readability assessment models, which are compatible with the local environment and valid. The model of this research can be used by a wide range of actors in the capital market. Measuring the level of readability of financial reports to determine the transparency rating of companies on the Tehran Stock Exchange is an example of the application of the current research model. Another achievement of this research is boosting the readability literature in accounting and finance.