Paternal and maternal involvement in Malaysian adolescents: test of factor structure, measurement invariance and latent mean differences

• Main Authors: Yap, Siew Ting, Baharudin, Rozumah, Yaacob, Siti Nor, Osman, Syuhaily
• Format: Article
• Published: Springer 2014
• Online Access: http://psasir.upm.edu.my/id/eprint/34834/; http://ezproxy.upm.edu.my:2078/article/10.1007/s12187-013-9209-3

As fathering gains increasing attention in parenting studies, the same instrument is always used to measure fathering and mothering concurrently and comparisons are made without prior establishment of measurement invariance. It is common that parenting scales possess different factor structures across different cultural settings. This study aimed to examine the factor structure, measurement invariance and latent mean differences of the Father/Mother Involvement Scale across adolescents’ perceived paternal and maternal involvement and by adolescent gender. A random sample of 720 Malaysian high school adolescents (Mage = 16 years; SD = 0.16) was used in the current study. Initial confirmatory factor analysis (CFA) indicated that the original factor structure had inadequate fit in the current sample. Subsequent exploratory factor analysis (EFA) yielded a three-factor structure that demonstrated good fit in further CFAs. Multigroup CFA provided support for configural invariance, metric invariance and scalar invariance across adolescents’ perceived paternal and maternal involvement and across adolescent gender. The latent factor mean comparisons showed that mothers were perceived to have higher expressive, instrumental and leisure/companionship involvement. Female adolescents perceived higher paternal expressive and instrumental involvement and maternal expressive, instrumental and leisure/companionship involvement than did male adolescents. These mean differences found at the latent level may reflect true differences. Taken together, the Father/Mother Involvement Scale is a sound measure of paternal and maternal involvement in Asian people, at least in Malaysian adolescents. The differences in factor structure may be attributable to sample variations in cultural background and/or family form.