Discriminant Analysis of Going Concern Audit Opinion Based on Liquidity, Leverage and Financial Distress

Background. Aim. Through this research, it is hoped that it can be seen to what extent the variables of liquidity, leverage, and financial distress can be used as predictors in the going concern audit opinion classification. The results of this study are expected to contribute to the development of auditing literature and become a consideration for auditors in considering the audit opinion to be given, while supporting the transparency and legitimacy of the company in the eyes of the public. Methods. This research is quantitative research with an explanatory approach, which is to test the relationship between the independent variables (liquidity, leverage, and financial distress) on the dependent variable (going concern audit opinion). Researchers not only want to know if there is a relationship, but also want to explain how and to what extent these variables can distinguish between companies that receive going concern opinions and those that do not, using the analysis method used is Discriminant Analysis, with the aim of classifying and separating groups of companies that receive going concern opinions and those that do not. The author chose a research place in companies listed on the IDXESGL listed on the Indonesia Stock Exchange (IDX) 2019-2021. Result. The discriminant model is able to distinguish companies based on going concern audit opinion. The analysis results show that the discriminant model built with three variables of Financial Distress, Liquidity and Leverage is statistically significant in distinguishing companies that receive going concern opinions and those that do not. Conclusion. Based on the test results, it is found that the discriminant model is able to distinguish companies based on going concern audit opinion. The analysis results show that the discriminant model built with three variables Financial Distress, Liquidity and Leverage is statistically significant in distinguishing companies that receive going concern opinion and those that do not (Wilks' Lambda = 0.824, Sig = 0.001). Two variables are significant in distinguishing groups: Based on the Equality of Group Means test, two variables viz: Financial Distress (Sig. = 0.027), Liquidity (Sig. = 0.026) have a significant effect in distinguishing going concern opinion. While Leverage (Sig. = 0.530) is not significant. The accuracy of the model is quite high, original classification accuracy: 86.2%, cross-validation accuracy: 85.1% This shows that the discriminant function is quite reliable in classifying companies. The model is more accurate in predicting companies that do not receive a going concern opinion (88.8% accuracy) than those that receive a going concern opinion (57.1% accuracy). This indicates an imbalance in the amount of data between groups that could affect the classification results. Impelementation. The results of this study provide several practical implications that can be applied in the world of accounting, auditing, and corporate financial management, especially in the context of predicting going concern audit opinion. The finding that Financial Distress and Liquidity significantly differentiate between companies that receive a going concern opinion and those that do not, provides a basis for external auditors to use this discriminant model as an analytical tool