Social and economic welfare reflects the quality of life in a region and is influenced by local social, economic, and environmental factors. East Java, as the second most populous province in Indonesia, faces challenges in improving the welfare of its residents, particularly due to varying regional characteristics such as employment, education, and population demographics. To understand the patterns of interrelationships among factors affecting welfare, this study conducted a klaster analysis to group regions based on similar characteristics. The klaster analysis employed both hierarchical (complete linkage) and non-hierarchical (K-means) approaches to determine the optimal number of klasters. The results revealed that the level of diversity across regions in East Java tends to be homogeneous in social and economic aspects, with average values exceeding standard deviations. Assumption tests for the klaster analysis confirmed that the data met the assumptions of multivariate normal distribution and dependency.Through hierarchical (complete linkage) and non-hierarchical (K-means) klaster analysis, two main klasters were formed, dividing districts/cities in East Java based on welfare characteristics. Using the complete linkage method, 27 regions were grouped into klaster 1, and 11 regions into klaster 2, while K-means grouped 26 regions into klaster 1 and 12 regions into klaster 2. Out of the six variables used, one variable was found to be insignificant in influencing the klastering results. Based on the mapping results, the grouping aligns with similar criteria, where urban areas predominantly fall into one klaster, and the other klaster is dominated by rural areas.

Improving the quality of Human Resources (HR) is a major challenge in facing global competition. Education as the main means of improving the quality of HR in Indonesia is still faced with the problem of inequality of access and quality between regions. This inequality causes disparities in educational development between urban and remote areas. This study focuses on grouping provinces in Indonesia based on aspects of educational development in 2023, using cluster analysis. Secondary data from the Central Statistics Agency (BPS) is used as the basis for analysis, including variables of average length of schooling, Gross Participation Rate (APK), Pure Participation Rate (APM), number of senior high schools, and community literacy development index. This study uses hierarchical and non-hierarchical cluster analysis methods to group provinces in Indonesia. The results of the hierarchical cluster analysis using the average linkage method show the most optimal cluster with the formation of three clusters. The first cluster consists of 31 provinces, the second cluster consists of 2 provinces, and the third cluster consists of 1 province. Data characteristics show large variations in the number of senior high schools and relative homogeneity in the average length of schooling between provinces.