Design and Fit of Auditors’ Psychological Model and the Disclosure Quality of the Board of Directors’ Activity Report

Abstract

This study examines the design and fit of a psychological model of auditors and the disclosure quality of the Board of Directors’ activity report. The present research is applied in terms of purpose and descriptive-analytical in terms of nature, and it is conducted within the category of survey studies. First, the factors influencing the psychological model of auditors and the disclosure quality of the Board of Directors’ activity report were identified using the grounded theory method. Subsequently, the model was designed through the interpretive structural modeling (ISM) approach, and finally, the model was estimated using structural equation modeling (SEM). The data collection procedures in this research include both library and field methods. For reviewing the literature and research background, library methods were used, while field methods were employed to collect data aimed at confirming or rejecting the study hypotheses. Observation and a structured questionnaire served as the primary data-gathering tools. The statistical population of this research comprises partners, employees, managers, and supervisors of auditing firms. The first step in determining the sample size involved defining an expert according to specific criteria. In this study, the perspectives of 20 experts, audit firm managers, and university faculty members were utilized. In the quantitative section, the opinions of partners, employees, managers, and supervisors of auditing firms were employed to evaluate the model fit and validate it using the partial least squares (PLS) method. Ultimately, 317 questionnaires were selected and analyzed in the quantitative phase. The findings indicated that the measurement instruments demonstrated acceptable levels of validity and reliability. Moreover, convergent validity was confirmed through the evaluation of factor loadings, showing adequate differentiation among latent variables and the absence of overlap. Further, the structural model was analyzed using the Bootstrap method to determine the significance of relationships among latent variables and to verify the accuracy of the research hypotheses. The results showed that all model relationships were significant with appropriate strength, indicating the proposed model’s strong predictive and explanatory power. Additionally, the Variance Inflation Factor (VIF) was calculated for mediating and independent variables, confirming the absence of multicollinearity and ensuring the accuracy of parameter estimations. Overall, the findings revealed that the partial least squares (PLS) analysis technique is not only a powerful tool for evaluating multivariate structural models but also highly capable of analyzing complex and real-world data.