A Structural Model of the Relationships Between Teacher, Peer, and Parental Support, Behavioural Engagement, Academic Efficacy and Cognitive Engagement of Secondary School Adolescents

The present investigation examined the hypothesized structural relationships between teacher, parental, and peer academic and emotional support on adolescents' cognitive engagement directly and indirectly through academic efficacy and behavioural engagement. A cross-sectional study utilizing survey method, the present study examined perceived academic and emotional support from teachers, classmates and parents via instruments, which were adapted from a number of established measures. A total of 450 adolescents (selected via multistage cluster sampling) from eleven secondary schools in the state of
Kedah in Malaysia took part in the survey. The psychometric properties of the instruments were assessed through reliability analysis, exploratory and confirmatory factor analysis. Through the various analysis conducted, the
instrument used in the present study is valid to be used in the local educational settings. Utilizing Structural Equation Modeling via AMOS version 16.0, the structural model was tested against the data for its fitness. Results indicated that academic efficacy and behavioural engagement did serve as plausible mediators in the relationships between support factors and cognitive engagement. Specifically, teacher academic support and parental emotional support emerged as two salient predictors in the indirect relationships between support factors and cognitive engagement. Peer academic support was the only
variable that had direct association with cognitive engagement. In addition to the preceding findings, results also indicated that the model for male and female differed significantly in that there were limited significant paths for the male model whereas there were a number of significant paths in the female model. Apparently, parental emotional support became the only indirect predictor on cognitive engagement for male adolescents whereas teacher
academic and emotional support as well as parental academic and emotional support both contributed towards cognitive engagement indirectly through academic efficacy and behavioural engagement for female adolescents. Female
adolescents also perceived significantly higher peer academic and emotional support and their level of cognitive engagement were significantly higher compared with their male counterparts. Additionally, younger adolescents
received significantly higher parental academic support compared with older adolescents. On the other hand, older adolescents received significantly higher teacher emotional support compared with younger adolescents. The findings
support the assumptions of Social Cognitive theory through its Triadic Determinism Principle whilst at the same time lends support to the Bronfenbrenner Ecological approach, Person Environment Fit theory as well as Social Capital theory. Lastly, theoretical as well as practical implications of the study are also discussed.