Re-examination of the internet financial reporting determinants

Abstract

PurposeThe purpose of this paper is to shed light on the relationship between the Internet Financial Reporting (IFR) levels and corporate characteristics. It is assumed that the relationship between the disclosure level and its determinants is known. Nevertheless, the results of the empirical studies confirm that it is a naive assumption. As a result, the author suggests refusing the conventional methods of econometric analysis.Design/methodology/approachThe research methodology consisted of four stages: First, the author tried to select the “best” model using the Akaike Information Criterion (AIC). Second, the author checked out the stability of the relationship between corporate disclosure level and its determinants. Third, the regression analysis was used. Finally, the author proposed a “genetic-fuzzy system” for studying the determinants of corporate disclosure. The firms' yearly data collected consisted of a random sample of 152 Tunisian companies' websites.FindingsThe results show that the variables that should be used to explain the level of IFR are firm size, ownership concentration, firm performance and liquidity. The Chow forecast test shows that there is a significant and large difference between the actual and the predicted values. Consequently, the author suggests using non-parametric methods, particularly a methodology based on fuzzy logic concepts and genetic algorithms. This technique would allow the author to discover the true form of the relationship between the disclosure level and its determinants. Regarding the hypotheses of this study, the findings of the “genetic-fuzzy system” validate all the hypotheses. Indeed, the arguments of the agency theory, the signaling theory, and the political cost hypothesis were supported using the “genetic-fuzzy system.”Originality/valueThe originality of the paper lies in providing a new research methodology based on several statistical tools for dealing with an important research topic in accounting and finance, i.e. the determinants of IFR. The results of this study can be considered as a starting point to develop a unified methodology.