Analyzing Determinants of Tax Management Effectiveness in Transportation Construction Enterprises in Vietnam: A Hybrid SEM–ANN Approach

Ngo Thi Huong; Vuong Thi Bach Tuyet; Hoang Thi Hong Le; Quang Hung Do

doi:10.32479/ijefi.20562

Authors

Ngo Thi Huong University of Transport Technology, Hanoi, Vietnam
Vuong Thi Bach Tuyet University of Transport Technology, Hanoi, Vietnam
Hoang Thi Hong Le University of Transport Technology, Hanoi, Vietnam
Quang Hung Do Fintech Lab, Posts and Telecommunications Institute of Technology (PTIT), Hanoi, Vietnam

DOI:

https://doi.org/10.32479/ijefi.20562

Keywords:

Tax Management Effectiveness, Transportation Construction, Structural Equation Modeling, Artificial Neural Network, Tax Policy, Public Finance

Abstract

This study investigates the key determinants influencing tax management effectiveness among transportation construction enterprises in Vietnam. Drawing upon the Institutional Theory, Theory of Planned Behavior, and Technology Acceptance Model, a conceptual framework was developed incorporating six factors: Tax Policy, Tax Officers, Infrastructure for Tax Administration, Corporate Tax Compliance Awareness, Digital Technology, and Tax Administration Functions. Data were collected from 385 firms using structured questionnaires. A hybrid methodological approach combining Structural Equation Modeling (SEM) and Artificial Neural Networks (ANN) was employed to capture both linear and nonlinear relationships among constructs. SEM results revealed that all six factors significantly influence tax management effectiveness, with Tax Policy and Tax Officers being the most impactful. The ANN analysis further confirmed these findings and ranked variable importance, highlighting Tax Policy as the strongest predictor. The study offers practical implications for policymakers aiming to enhance tax governance in project-based, capital-intensive sectors and demonstrates the value of hybrid modeling in public finance research.