School Choice Advocacy Report Plays Regression Modeling Games to Make Its Case Against Public Schools - REVISED

Share Article

Key New NEPC Review Takeaway: Report is of little use to policymakers and others with an interest in understanding parent satisfaction associated with school choice.

*We are resending this newsletter because the earlier version misidentified the report's publisher. Although the report's author identified himself using his affiliations with the Reason Foundation and Cato Institute, we should not have listed those organizations as the publishers of the report. The report's publisher is now correctly listed as the Annenberg Institute for School Reform. We apologize for our mistake.*

A recent report published by the Annenberg Institute for School Reform at Brown University argues that private and charter schools have a strong positive effect on parents’ reported satisfaction with their children’s education. argues that private and charter schools have a strong positive effect on parents’ reported satisfaction with their children’s education.

Steven V. Miller of Clemson University reviewed School Sector and Satisfaction: Evidence from a Nationally Representative Sample, and found critical errors that limit its value to those weighing its policy proposals.

The report presents regression analyses purporting to show that choice parents are more satisfied with their schools than are parents whose children attend their local public schools. It then explains this increased parental satisfaction by pointing to competitive pressures and the importance of reducing the monopoly power of the traditional public school system. Its analyses are based on a nationally representative sample from the National Household Education Surveys Program (NHES).

However, Professor Miller explains, the report suffers from two major flaws. First, it saturates the important analyses with over 230 covariates. This amounts to a “garbage can” regression modeling approach that obscures more than it illuminates; variables are misspecified and results are sensitive to the oversaturation of the regression model. A reader versed in statistical modeling will have no confidence in the substance of the findings.

Second, and even more importantly, the report’s decision to focus on just the “very satisfied” overstates the comparative effect of private and charter schools on parent satisfaction. Almost 90% of public school parents are satisfied with their child’s education, and the report’s decision to focus on just the “very satisfied” appears to be a deliberate modeling choice to overstate the purported effects of private and charter schools on parent satisfaction. This appearance is confirmed by the report’s selective use of past research literature, again suggesting an interest in findings that will support advocacy for curtailing the supply of public education.

For each of these reasons, the report is of little or no use to policymakers and others with an interest in understanding parent satisfaction associated with school choice.

Find the review, by Steven V. Miller, at:
http://nepc.colorado.edu/thinktank/school-satisfaction

Find School Sector and Satisfaction: Evidence from a Nationally Representative Sample, written by Corey A. DeAngelis and published by the Annenberg Institute for School Reform at Brown University, at:
https://edworkingpapers.com/sites/default/files/ai19-147.pdf

NEPC Reviews (http://thinktankreview.org) provide the public, policymakers, and the press with timely, academically sound reviews of selected publications. NEPC Reviews are made possible in part by support provided by the Great Lakes Center for Education Research and Practice: http://www.greatlakescenter.org

The National Education Policy Center (NEPC), a university research center housed at the University of Colorado Boulder School of Education, produces and disseminates high-quality, peer-reviewed research to inform education policy discussions. Visit us at: https://nepc.colorado.edu

Share article on social media or email:

View article via:

Pdf Print

Contact Author

William J. Mathis

Steven V. Miller
Clemson University
(864) 656-3233
Email >
@NEPCtweet
since: 09/2010
Follow >