Skip to content

Propensity score stratification with multilevel models

Research Method · All research methods
Evidence · 1 claim (1 for) · 1 study (1 causal), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

When estimating the achievement effects of school transfer or school choice from non-experimental data, the article recommends combining propensity score stratification with multilevel models. The authors report that this combination "reduces selection bias which arises from confounding variables and improves the accuracy of charter school effect estimations". The multilevel component addresses the clustered structure of students within schools in cross-state samples.

Accounts

  • Use propensity score stratification with multilevel models when estimating school achievement effects from observational data: When estimating the achievement effects of school transfer or school choice from non-experimental data, the article recommends combining propensity score stratification with multilevel models. The authors report that this combination "reduces selection bias which arises from confounding variables and improves the accuracy of charter school effect estimations". The multilevel component addresses the clustered structure of students within schools in cross-state samples. (Beth Tarasawa (2015))
  • Propensity score stratification within multilevel models as a quasi-experimental approach to school achievement effects: The article employs a combined methodological framework: propensity score stratification is used to balance key covariates between treatment and control groups, and multilevel models accommodate the nesting of students within schools in a cross-state sample. The authors state they "employ propensity score stratification and multilevel models to balance key covariates between treatment and control groups of a cross-state sample of students". The framework is presented as suited to estimating school transfer effects when randomization is not possible, providing "a more complex picture of charter school achievement effects in a quasi-experimental context". (Beth Tarasawa (2015))

Claims

-

Key Sources

  • Beth Tarasawa, Yun Xiang. (2015). Propensity score stratification using multilevel models to examine charter school achievement effects. Journal of School Choice, 9(2), 179–196. https://www.tandfonline.com
  • Beth Tarasawa, Yun Xiang. (2015). Propensity score stratification using multilevel models to examine charter school achievement effects. Journal of School Choice, 9(2), 179–196. https://www.nwea.org/research/publication/propensity-score-stratification-using-multilevel-models-to-examine-charter-school-achievement-effects/
Page metadata
Field Value
type research-method
id propensity-score-stratification-with-multilevel-models
title Propensity score stratification with multilevel models
description When estimating the achievement effects of school transfer or school choice from non-experimental data, the article recommends combining propensity score stratification with multilevel models.
status review
moved_from theory/propensity-stratification-multilevel-quasi-experimental-approach
generated by process:settle-candidates on 2026-10-09
trust unverified — no one has confirmed this page's content
sources 1 — beth-tarasawa-2015

Frontmatter is the page's machine-readable record — see the Schema & Guide.