View in EDS

School-Level Poverty in Education Survey Research: Can Free or Reduced-Price Lunch Be Substituted as a Weighting Variable by Other, More Reliable Measures? Working Paper. WR-A4691-1

Saved in:
Bibliographic Details
Title: School-Level Poverty in Education Survey Research: Can Free or Reduced-Price Lunch Be Substituted as a Weighting Variable by Other, More Reliable Measures? Working Paper. WR-A4691-1
Language: English
Authors: Dorothy Seaman, Joshua Eagan, Claude Messan Setodji, RAND Education, Employment, and Infrastructure
Source: RAND Corporation. 2026.
Availability: RAND Corporation. P.O. Box 2138, Santa Monica, CA 90407-2138. Tel: 877-584-8642; Tel: 310-451-7002; Fax: 412-802-4981; e-mail: order@rand.org; Web site: http://www.rand.org
Peer Reviewed: Y
Page Count: 26
Publication Date: 2026
Document Type: Reports - Research
Education Level: Elementary Secondary Education
Descriptors: Poverty, Low Income Students, Educational Research, Surveys, Measures (Individuals), Sampling, Statistical Inference, Elementary Secondary Education
Geographic Terms: Mississippi, New Mexico, Nevada, South Carolina
DOI: 10.7249/WRA4691-1
Abstract: Changes to state and federal policy regarding how free or reduced-price lunch (FRPL) eligibility is determined have added to increasing concern among education researchers about the suitability of these data as a proxy for poverty. The purpose of this exploratory study was to examine the potential utility of school-level poverty measures other than FRPL for population representativeness in survey weighting. We assessed the impact of three alternative school-level poverty measures on survey weight calibration using surveys conducted on the American Educator Panels (AEP). Comparisons of survey weights derived using alternative poverty measures revealed strong, positive, linear correlations with weights derived using FRPL. Results indicated that the choice of poverty measure did not meaningfully affect inferences about teachers and principals drawn from analyses of survey response data. Overall, while survey weights are fundamental for valid inference, the effect of weights on carefully designed survey-based studies is generally slight rather than substantial. [Funding for this report was provided by American Educator Panels.]
Abstractor: As Provided
Entry Date: 2026
Accession Number: ED680498
Database: ERIC
Description
Abstract:Changes to state and federal policy regarding how free or reduced-price lunch (FRPL) eligibility is determined have added to increasing concern among education researchers about the suitability of these data as a proxy for poverty. The purpose of this exploratory study was to examine the potential utility of school-level poverty measures other than FRPL for population representativeness in survey weighting. We assessed the impact of three alternative school-level poverty measures on survey weight calibration using surveys conducted on the American Educator Panels (AEP). Comparisons of survey weights derived using alternative poverty measures revealed strong, positive, linear correlations with weights derived using FRPL. Results indicated that the choice of poverty measure did not meaningfully affect inferences about teachers and principals drawn from analyses of survey response data. Overall, while survey weights are fundamental for valid inference, the effect of weights on carefully designed survey-based studies is generally slight rather than substantial. [Funding for this report was provided by American Educator Panels.]
DOI:10.7249/WRA4691-1