Difference-in-Differences (DiD) Analysis of Data Driven Accounting Systems and its Impact on Financial Error Rates in SMEs

Abstract

This study examined the impact of data-driven accounting systems on financial error rates in small and medium-sized enterprises, addressing the persistent problem of manual processing mistakes, weak reconciliation practices, fragmented records, delayed reporting, and limited internal control in SME accounting operations. The purpose was to determine whether SMEs using cloud-based and enterprise data-driven accounting systems achieve stronger financial error reduction than SMEs using manual, traditional, or basic accounting software. A quantitative, cross-sectional, case-based design was adopted, using 220 valid responses from SME owners, accountants, finance officers, bookkeepers, internal auditors, and financial managers across retail, manufacturing, logistics, services, and other sectors. The study focused on key variables including data-driven accounting system capability, accounting automation, real-time reporting, data analytics capability, system integration, internal control integration, and financial error reduction. Data were analyzed using descriptive statistics, reliability testing, correlation analysis, multiple regression, adoption-timing comparison, error-type sensitivity analysis, and Difference-in-Differences estimation. The findings showed a high level of data-driven accounting system capability among SMEs, with an overall mean of 4.08 and financial error reduction mean of 4.13. Reliability was strong, with the overall questionnaire producing Cronbach’s alpha of 0.89. Correlation results showed that data-driven accounting system capability had a strong positive relationship with financial error reduction, r = 0.68, p < 0.001. Regression findings indicated that the model explained 58.4% of the variance in financial error reduction, with internal control integration emerging as the strongest predictor, β = 0.31, p < 0.001, followed by accounting automation, β = 0.27, p < 0.001. Difference-in-Differences results showed that adopters reduced perceived financial error rates from 3.42 to 2.01, compared with non-adopters from 3.35 to 3.08, producing a significant DiD effect of -1.14, p < 0.001. The study implies that SMEs should treat digital accounting adoption as a financial accuracy, governance, and risk-control strategy, not only as a bookkeeping modernization effort.