Reliability and validity of the short Hong Kong Chinese Self-Regulation of Learning Self-Report Scale (SRL-SRS-C)

Abstract

Objectives: This study translated and shortened the original English, six-factor, Self-Regulated Learning Self-Report Scale (SRL-SRS). This version was subsequently assessed for its reliability and validity, for use in a Hong Kong Chinese adolescent population. Design and methods: The SRL-SRS was translated into Chinese following the back-translation procedure. Exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and test–retest reliability analysis, were conducted on a sample of 314 Hong Kong Chinese adolescents aged 12–17 years (mean age = 13.2 years, SD = .99; male = 155) from one Hong Kong secondary school. Cross-validation was conducted on a second sample of adolescents (n = 477; mean age = 14.92 years; male = 283) from three Hong Kong secondary schools. Results: EFA results supported the original six-factor model. Subsequent CFA results supported the short version with sufficient goodness-of-fit statistics (comparative fit index; CFI = 0.93; Tucker–Lewis index (TLI) = 0.92; root square mean error of approximation (RMSEA; 95% CI) = 0.0047 (0.041 – 0.053); SRMR = 0.044). Further support was also found by cross-validation in the second sample (CFI = 0.915; TLI = 0.906; RMSEA = 0.052; SRMR = 0.043). Internal consistency analysis revealed that Cronbach's α ranged from.72 to .89 for all subscales (sample 1); test–retest reliability analysis (sample 1) produced Intra-class Correlation Coefficient (ICC) values ranging from .69 to .80 for all subscales. Conclusions: The short SRL-SRS-C is a sufficiently reliable and valid instrument to measure self-regulation of learning in a Chinese adolescent population. In addition, the SRL-SRS-C has demonstrated sufficient content, construct validity, and test–retest reliability.