Abstract:Math proficiency is a strong predictor of postsecondary attainment, yet cross-national comparative assessments show that U.S. students consistently score below many of their peers, especially those in East Asia. Similar math achievement disparities exist across ethnic/racial groups within the U.S., where Asian American students routinely outperform their White and Hispanic counterparts. The stability of math achievement rankings within and across countries suggests that several predictors are transportable and context-invariant, as they forecast math achievement even following striking shifts in contextual characteristics, such as country of residence, language, or school system. We evaluated a multiple mediator model of math achievement and postsecondary attainment that incorporates several of these predictors, using cross-sectional international data (PISA-2012) and longitudinal U.S. data (NELS and ELS). In all samples and data sets, the model explained substantial portions of the variance in educational outcomes. Parents’ educational aspirations for their children were strongly associated with students’ own aspirations which, in turn, were powerful predictors of math achievement and postsecondary attainment—both directly, and through their influence on important mediators such as math self-perception. Longitudinal relations identified in an older national dataset (NELS) were generally cross-validated in ELS. Moreover, in both data sets, construct relations were found to be largely invariant across racial/ethnic groups. Although additional research is needed to clarify the mechanisms underlying group disparities, the outstanding fit of the model (across samples and data sets) provides the empirical foundation for educational interventions that are likely to help all students, regardless of their ethnicity or national origin.

On the Correspondence Between the Latent Growth Curve and Latent Change Score Models

Jenss-Bayley Latent Change Score Model with Individual Ratio of Growth Acceleration in the Framework of Individual Measurement Occasions

Latent change scores models for applied research: A practical guide using Mplus

Modeling Nonlinear Change Via Latent Change and Latent Acceleration Frameworks: Examining Velocity and Acceleration of Growth Trajectories.

Fitting Structural Equation Model Trees and Latent Growth Curve Mixture Models in Longitudinal Designs: The Influence of Model Misspecification

Latent Growth Curve Models

Bayesian Methods and Model Selection for Latent Growth Curve Models with Missing Data

Latent Growth Curve and Repeated Measures ANOVA Contrasts: What the Models are Telling You

Measurement Models, Estimation, and the Study of Change

Bayesian Approach to Non-ignorable Missingness in Latent Growth Models

Understanding Variation in Longitudinal Data Using Latent Growth Mixture Modeling

Modeling construct change over time amidst potential changes in construct measurement: A longitudinal moderated factor analysis approach

Fitting Nonlinear Latent Growth Curve Models With Individually Varying Time Points

Two-step growth mixture model to examine heterogeneity in nonlinear trajectories

Scale-Invariance, Equivariance and Dependency of Structural Equation Models

Longitudinal Mediation Analysis with Latent Growth Curves

Evaluation of A Bayesian Approach to Estimating Nonlinear Mixed-Effects Mixture Models

A Dynamic Model of Demand-Ability Fit Change in Proactive Socialization: A Latent Change Score Approach

Assessment of School Performance Through a Multilevel Latent Markov Rasch Model

A latent growth moderated mediation model of math achievement and postsecondary attainment: Focusing on context-invariant predictors.

Model Selection Criteria for Latent Growth Models Using Bayesian Methods