The Management Administration Information System (MAIS) is a form of digitizing administrative processes within public organizations that aims to improve efficiency, data integration, and ease of access to information. MAIS plays a role in simplifying bureaucratic processes and accelerating data-driven decision-making (Desouza & Dawson, 2024). Effective MAIS implementation can minimize redundancy, improve information accuracy, and create a foundation for organizational transparency (Huda et al., 2021).

Digitalization enables operational efficiency through process automation and reduced manual errors. Data efficiency encompasses aspects of access speed, information accuracy, and data consistency across units. According to Al-Shbail & Abo-Ghazala (2024), high data accuracy strengthens the basis for valid and accountable decision-making in public sector organizations.

Transparency in the context of digital government reflects the openness of public information and the system's ability to provide a clear audit trail. A digitized audit trail allows every process within the system to be recorded and traced, increasing public trust and the effectiveness of internal oversight (Siregar, 2020; INTOSAI, 2025). This digital trail also strengthens management accountability because all actions are documented.

Accountability in the public sector emphasizes management's obligation to explain and be accountable for policies, decisions, and resource use. Mardiasmo (2018) emphasized that high accountability results from system transparency and sound information management, and is a determinant of the quality of governance.

Good governance encompasses the principles of transparency, participation, responsiveness, accountability, and effectiveness in government administration. According to UNESCAP (2021), good governance is a prerequisite for realizing a government that is responsive to technological changes and public demands.

Smart governance is an advanced form of government administration that utilizes digital technology to improve the quality of decision-making, public services, and public participation (Pereira & Botelho, 2018). This concept emphasizes collaboration between government, society, and technology to achieve a more adaptive, efficient, and innovative government.

Based on the above framework, the following hypotheses are proposed:

This study uses a quantitative approach with an explanatory research approach. This approach was chosen because the study aims to explain the causal relationship between several independent variables, namely the Management Administration Information System (MAIS), data efficiency and accuracy, audit transparency, and management accountability, with the dependent variable smart governance, with good governance as a mediating variable. This explanatory approach aligns with the theoretical model framework developed based on previous studies, thus enabling hypothesis testing through robust and measurable inferential statistical methods.

The population in this study is all state civil servants (ASN) who work in the local government environment, especially in agencies that have implemented technology-based administrative information systems and are involved in the process of policy making, planning, reporting, and evaluating public programs.

The research sample consisted of 110 respondents selected using purposive sampling, a sampling technique based on specific criteria determined by the researcher. The criteria used included: (1) respondents were structural or functional employees who understood information system-based governance processes; (2) had worked for at least 2 years; and (3) were involved in reporting, planning, or evaluating their work unit's activities. This number of respondents was considered sufficient for analysis using SEM-PLS, because according to Hair et al. (2021), the minimum sample size for a complex structural model is 10 times the number of the broadest indicators in a construct.

The data analysis technique used was Structural Equation Modeling - Partial Least Squares (SEM-PLS). SEM-PLS was chosen because this method is very suitable for predictive and exploratory models, and can handle complex models with many indicators and latent variables even with a relatively small sample size. The analysis was conducted using SmartPLS software version 4.0, which allows simultaneous testing of the measurement model (outer model) and the structural model (inner model). The steps in the analysis include: testing the validity and reliability of the instrument, testing the quality of the measurement model, and testing the hypothesis through bootstrapping.

Data collection was carried out through the distribution of electronic questionnaires, consisting of 6 main constructs:

Each construct is measured through several indicators based on adaptations of previous instruments and relevant theories. The measurement scale uses a 5- point Likert scale (1 = strongly disagree to 5 = strongly agree).

Validity and reliability tests were conducted in two stages:

1. Outer Model Test:

2. Discriminant Validity:

Through the cross-loading value between constructs and the Fornell-Larcker Criterion.

The analysis results show that SIAM digitalization significantly impacts data efficiency and accuracy (coefficient = 0.741, p < 0.001, f² = 0.28). This indicates that SIAM implementation can improve administrative processes to be faster, more precise, and more integrated, thereby minimizing input errors and accelerating information access. These findings support the research of Huda et al. (2021) which states that modern information systems such as SIAM reduce process redundancy and strengthen data management.

Data efficiency and accuracy significantly impact transparency and audit trails (coefficient = 0.683, p < 0.001, f² = 0.23). Efficient and accurate data provides a strong foundation for information transparency and enables the smooth operation of digital reporting and audit systems. This reinforces Siregar's (2020) finding that data quality plays a crucial role in establishing a digitally traceable transparency system.

Transparency and audit trails also significantly influence managerial accountability (coefficient = 0.615, p < 0.001, f² = 0.18). The more transparent an agency is in conveying its data and processes, the higher the social and moral pressure on management to act responsibly. With a digitalized audit trail system, public managers become more careful and systematic in making decisions, in accordance with the principles of public accountability (Mardiasmo, 2018).

Managerial accountability has been shown to have a significant influence on good governance (coefficient = 0.662, p < 0.001, f² = 0.28). This indicates that responsibility in public management is the main foundation for realizing good governance. The principle of accountability strengthens the dimensions of transparency, effectiveness, and participation, which are at the core of good governance as defined by UNESCAP (2021).

Good governance has the largest influence on smart governance in this model (coefficient = 0.707, p < 0.001, f² = 0.35). These results confirm that digital transformation towards smart governance is not sufficient with the use of technology alone but must be supported by a foundation of good governance. Aspects such as integrity, transparency, and public participation are the main drivers of the success of smart governance.

Bootstrapping results show that good governance significantly mediates the effect of managerial accountability on smart governance (indirect coefficient

= 0.468, t = 7.62, p < 0.001). In other words, although accountability has an indirect impact on smart governance, this impact will be maximized if managed through established good governance mechanisms.

This study concludes that the digitalization of the Management Information Administration System (MIAS) plays a crucial role in building smart governance. By improving data efficiency and accuracy, MIAS digitization significantly strengthens transparency and audit trails, which in turn enhances managerial accountability. This accountability has proven to be the foundation for good governance, acting as a crucial mediator in driving the successful implementation of smart governance.

All influence paths between constructs in this research model were statistically significant, with a relatively high coefficient of determination (R²). These findings confirm that public sector digital transformation must be underpinned by good governance principles to maximize the impact on the quality of public services and the effectiveness of government management.