Factors Affecting Academic Achievement: Meta-Analytic Structural Equation Modeling

Academic achievement remains one of the most studied outcomes in educational research, yet the relative contributions of psychological, instructional, and environmental factors are rarely examined simultaneously within a unified empirical framework. This study synthesized findings from 53 peer-reviewed studies, drawn from an initial pool of 10,249 records, using Two-Stage Structural Equation Modeling (TSSEM) to examine how attitude toward learning, achievement motivation, prior knowledge, teaching quality, classroom climate, and self-concept jointly influence academic achievement. Publication bias was assessed via Egger's test (z = −1.674, p = .094), confirming an unbiased sample. Significant heterogeneity was detected across all variable pairs (I² ranging from 81.27% to 98.67%), necessitating a random-effects estimation approach. The pooled correlation matrix revealed that all 21 variable pairs were positively and significantly correlated at the .01 level, with prior knowledge demonstrating the strongest association with academic achievement (r = .573). The final structural model fit the data well (χ² = 0.062, df = 3, P-value = 0.996, CFI = 1.000, TLI = 1.000, GFI = 1.000, AGFI = 1.000, NFI = 1.000, IFI = 1.000, RMSEA = 0.000, SRMR = 0.000) and explained 38.5% of the variance in academic achievement. Prior knowledge emerged as the dominant direct predictor (β = .448), while achievement motivation exerted the largest total effect through both direct (β = .088) and indirect (β = .153) pathways. These findings carry substantial implications for curriculum design, instructional policy, and student support frameworks.