University of North Texas
Selection Bias and Sensitivity as Moderators of Prekindergarten Age-Cutoff Regression Discontinuity Study Effects: A Meta-Analysis
Abstract
dc:descriptionThe age-cutoff regression discontinuity design (RDD) has emerged as one of the most rigorous quasi-experimental approaches to determining program effects of prekindergarten on literacy and numeracy outcomes for children at kindergarten entry. However, few pre-K meta-analyses have focused attention on validity threats. The current random-effects meta-regression tests the moderating effects of prominent threats to validity, selection bias and sensitivity, on impact estimates generated from age-cutoff regression discontinuity studies from large-scale programs. Results from averaging dependent standardized mean difference effects suggested small positive moderating effects of total attrition and robust 3-month bandwidths on reading effects, but not on math. However, these results were not statistically significant. In contrast, results generated from robust variance estimation yielded a small statistically significant association between total attrition and math effects. These mixed results may warrant further research on prekindergarten evaluation methodology, evaluation estimation methods, and the totality of evidence used to inform policy.
Degree
thesis:*- Grantor dc:publisher
- University of North Texas
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Stewart, Genea K.
- Contributors dc:contributor
-
- Hull, Darrell Magness
- Lee, Jihyun
- Henson, Robin K.
- Middlemiss, Wendy
Subjects
dc:subject × 8Rights
dc:rights- Statement dc:rights
-
- Public
- Stewart, Genea K.
- Copyright
- Copyright is held by the author, unless otherwise noted. All rights Reserved.
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
-
local-cont-no: submission_3489
https://digital.library.unt.edu/ark:/67531/metadc2179287/
ark: ark:/67531/metadc2179287