MULTIVARIATE MULTILEVEL MODELLING TO ASSESS FACTORS AFFECTING THE QUALITY OF VOCATIONAL HIGH SCHOOLS IN SOUTH SULAWESI PROVINCE

This study analyzes the quality of Vocational High Schools (VHS), which have a hierarchical data structure and more than one response variable. Data gathered for this is from Basic Education (DAPODIK) in form raw variables several that characterize VHS other independent South Sulawesi four years (2018 to 2021) Ministry Finance Republic Indonesia (KEMENKEU), Statistics (BPS). The explanatory variable at regency level consists 2021), multi-year high-dimensional structure. Therefore, Principal Component Analysis (PCA) used overcome this. modelling done by using multivariate multilevel (MVMM) on one-level two-level structures. aims model average National Examination Accreditation scores School MVMM considers regency/city identifies factors influence scores. results showed with random intercept as component was better based minor Deviance Information Criterion (DIC) value. Simultaneously, 5% significance, contribute significantly Province are produced. significant effect school ratio number students/pupils per group, percentage certified teachers teachers, toilets, laboratory availability, availability supporting rooms. Meanwhile, level, it found poverty Gross Regional Domestic Product (GRDP) had Schools.