MEASUREMENT INVARIANCE OF AN ENGLISH LANGUAGE PROFICIENCY SCALE ACROSS GENDER AND SCHOOL LOCATION

This study investigated the measurement invariance of the English Language Proficiency Scale (ELPS) on the basis of gender and school location among secondary school students in southwestern Nigeria. This study was conducted to verify the hypothesized factor structure of the ELPS and examine whether the instrument possessed configural, metric, scalar, and strict invariance with respect to gender and school location. A quantitative instrumentation research design was used. Samples of 1,300 secondary school students were obtained from three randomly chosen states in southwestern Nigeria using a multi-stage sampling technique. Data were collected using the English Language Proficiency Scale, a 28-item scale that measures listening, speaking, reading, and writing proficiency levels. Data analyses were conducted using confirmatory factor and multi-group confirmatory factor analysis. The baseline four-factor model showed a good fit with the data (CFI = 0.951; TLI = 0.944; RMSEA = 0.030; SRMR = 0. 036). Adequate internal consistency coefficients were between 0.87 and 0.90, with an overall alpha coefficient of 0.94. Convergent validity was established, with AVE values within the range of 0.50–0.60. Configural, metric, scalar, and strict invariance were confirmed in relation to gender and school location. The ELPS is a reliable and valid instrument to measure English language proficiency