A Survey-Based Multiple Regression Analysis of Recruitment, Selection, and Employee Performance at Al-Muhajirin Foundation Purwakarta

Abstract

Human resource quality is a critical issue for education organizations because teaching, administration, and institutional service quality depend on placing employees in roles aligned with their competencies. This survey-based quantitative study examines statistical associations among recruitment, selection, and employee performance in an Islamic education foundation in Purwakarta, Indonesia. The study used a descriptive-verificative design and multiple linear regression to examine two staffing predictors and one performance outcome. Data were collected via a structured questionnaire administered to 151 employees selected from a population of 491 using proportional stratified random sampling. Respondents were proportionally allocated across education/work-unit strata and randomly selected from the eligible employee list within each stratum. Recruitment, selection, and employee performance were measured with Likert-type indicators adapted from human resource management and performance literature. Instrument testing showed acceptable item validity and internal consistency, although the reliability values for selection and employee performance were marginal and were interpreted cautiously. Descriptive results indicate that recruitment was generally good but weakest in transparency; selection was generally good but weakest in objectivity; and employee performance was generally good but weakest in work quantity. Regression results indicate that recruitment was positively but not significantly associated with employee performance (B = 0.145; p = 0.088), whereas selection was positively and significantly associated with employee performance (B = 0.349; p < 0.001). Recruitment and selection jointly explained 29.7% of the variance in employee performance. The findings are interpreted as statistical associations rather than causal effects because the design was cross-sectional and self-reported.