Examination of School Climate and Distal Outcomes Using the Three-Step Method

• In Event: Methodological Advances in Longitudinal Mixture Modeling

Abstract

Latent profile analysis (LPA) is not a new technique. However, fresh implementations of LPA have emerged in the literature which aim to detect unobserved heterogeneity in a population. While this is one goal of LPA, another is the examination of latent profiles with respect to distal outcomes. While primary and secondary schools strive to meet accountability standards set by the Every Student Succeeds Act (ESSA), the relationship between their ability to meet these benchmarks and distal outcomes for students who exit these institutions should also be evaluated. In other words, we should evaluate the potential impact that latent profile membership has on future, or longitudinal, outcomes for students.
In South Carolina (SC), school climate surveys have historically been administered to gauge public schools’ yearly progress and meet requirements of the state’s accountability legislation. While schools can be classified as having a positive or poor school climate, schools can also be evaluated at a deeper level to determine if latent subclasses of schools exist and whether they share key characteristics that can affect important student outcomes.
To date, SC school climate student data has not been examined using multilevel latent profile analysis (MLPA) which is used to account for the natural nesting of data (e.g., students within schools). Very few other studies have used a nonparametric MLPA to identify both student and school profiles of school climate. The current study will apply a nonparametric MLPA using the manual BCH 3-step method (Bakk et al., 2014; Bolck et al., 2004; Vermunt, 2010) with both covariates and distal outcomes on a state-wide sample of students in elementary, middle, and high school. The inclusion of a multilevel structure will both aid in the correct classification of individuals and provide further insight into school climate differences across schools by allowing for the formation of school-level latent profiles which has yet to be studied on the SC school climate survey for students.
Student school climate data collected in 2018 and 2019 by the South Carolina State Department of Education (SCDE) will be analyzed. A measure of poverty will serve as the covariate. Distal outcomes will include career/college readiness and measures of academic achievement. On balance, the proposed study aims to identify the latent profile assignments of students and schools in SC and the extent to which such assignments are impacted by poverty. Further, this study will determine the extent to which temporal distal outcomes vary across student and school latent profiles. This connection between student and school profiles and temporal distal outcomes can help policymakers and school administrators better understand and address the challenges our students and schools face on a daily basis and help customize district and state services.
References:
Bakk, Z., Oberski, D.L., & Vermunt, J.K. (2014). Relating latent class assignments to external variables: Standard errors for correct inference. Political Analysis, 22, 520-540.
Bolck, A., Croon, M., & Hagenaars, J. (2004). Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political Analysis, 12, 3-27.
Vermunt, J.K. (2010). Latent class modeling with covariates: Two improved three-step approaches. Political Analysis, 18, 450-469.

Author

• Kathleen McGrath, Charlotte-Mecklenburg Schools