This paper evaluates the effectiveness of two cross-sectional discretionary accruals models, the Cross-Sectional Jones Model and the Cross-Sectional Modified Jones Model, in detecting earnings management compared to their time-series counterparts. The study aims to address the limitations of prior research, which often used time-series models that may suffer from survivorship bias and smaller sample sizes. The authors use a matched-pair design to control for potential confounding variables and perform multiple logistic regression tests to assess the robustness of the findings. The results show that, while all models except the DeAngelo Model successfully detect earnings management, only the cross-sectional models consistently perform well after controlling for variables such as book-to-market ratios, financial leverage, and earnings performance. The study concludes that the cross-sectional models are superior for future earnings management research due to their larger sample sizes and ability to handle firms with shorter histories. Additionally, the study highlights the importance of controlling for correlated omitted variables in accruals management studies.This paper evaluates the effectiveness of two cross-sectional discretionary accruals models, the Cross-Sectional Jones Model and the Cross-Sectional Modified Jones Model, in detecting earnings management compared to their time-series counterparts. The study aims to address the limitations of prior research, which often used time-series models that may suffer from survivorship bias and smaller sample sizes. The authors use a matched-pair design to control for potential confounding variables and perform multiple logistic regression tests to assess the robustness of the findings. The results show that, while all models except the DeAngelo Model successfully detect earnings management, only the cross-sectional models consistently perform well after controlling for variables such as book-to-market ratios, financial leverage, and earnings performance. The study concludes that the cross-sectional models are superior for future earnings management research due to their larger sample sizes and ability to handle firms with shorter histories. Additionally, the study highlights the importance of controlling for correlated omitted variables in accruals management studies.