Privacy-Preserving Aggregated Dataset for Data-Driven Management Information System: A Cross-Demographic Empirical Framework and Explainable Intervention Prediction

This repository provides a privacy-preserving aggregated dataset supporting the study entitled “Data-Driven Management Information System: A Cross-Demographic Empirical Framework and Explainable Intervention Prediction.” The study uses administrative demographic data derived from the Integrated Service Management Information System (MSIPT) operated by the Protestant Church of Maluku (GPM), Indonesia. Two contrasting population contexts were selected for cross-demographic analysis: the GPM Lateri congregation, representing an urban population, and the GPM Manusa congregation, representing a rural/mountainous population. The study includes 4,797 individuals from Lateri and a final analytical cohort of 650 individuals from Manusa. The publicly deposited dataset contains privacy-preserving aggregated information on age groups, gender, education, occupation, and family status. It also includes cohort-level documentation and quality-control information supporting interpretation of the dataset. The original administrative datasets contain personally identifiable and potentially re-identifying information, including individual names, residential information, congregation sector and unit information, age, family status, education, and occupation. Therefore, the original individual-level datasets are not included in this public repository. No names, addresses, personal identification numbers, row-level pseudonyms, or individual-level administrative records are included in the deposited dataset. To reduce the risk of re-identification, only congregation-level marginal aggregates are publicly released. Rare categorical values with fewer than five observations are combined into a general category and are not released separately. In the Manusa source data, the occupation export contained 655 records, whereas the corresponding age and education datasets contained 650 records. Following consistency checking, the public analytical cohort was restricted to 650 matched records. Five unmatched occupation records were excluded from the public analytical dataset, and no identifying information related to those records is disclosed. This dataset is made available to support research transparency, verification of demographic distributions, and reproducibility of source-data preparation while preserving the privacy and confidentiality of congregation members. The complete individual-level administrative data remain under institutional data governance and are not publicly available. Additional non-public research materials may be considered upon reasonable request to the corresponding author, subject to approval by the responsible institutional data custodian and applicable confidentiality requirements. This repository represents a privacy-preserving research dataset and should not be interpreted as a public release of the original MSIPT administrative database.