Exploring the Construction of a Proxy Socioeconomic Status Index for NAEP Grade 4 Assessments

• In Event: 41.066 - The Development and Applications of Alternative Student Socioeconomic Status Measures

Abstract

This paper builds upon the findings of the research effort on the construction and validation of a NAEP SES index using 2013 NAEP grade 12 mathematics assessment data (presented in paper 2). That NAEP SES index is comprised of five SES proxy variables from the student contextual and school questionnaires including student NSLP eligibility status; student-reported household possessions, parental education, and books at home; and percent of students eligible for NSLP at the school level. Unlike in grades 8 and 12 NAEP assessments, student-reported parental education is not collected at grade 4. However, in spite of the lack of a measure of parental education, it is possible to make use of the other four variables to build a slightly different proxy SES index. To explore the feasibility of constructing a NAEP SES index without parental education in grade 4 NAEP, an alternative SES measure for the 2013 NAEP grade 12 mathematics assessment is constructed without including parental education as a component. The results showed that the alternative SES measure has a similar effect in explaining the White-Hispanic and White-Black achievement gaps and a similar magnitude of power in explaining the variance in student performance (see table 1). Generalizing from this result suggests that it might be possible to construct a useful SES index for NAEP grade 4 assessments without parental education.
The SES index for the NAEP grade 4 assessment is comprised of four components: student NSLP eligibility status; student-reported household possessions and books at home; and percent of students eligible for NSLP at the school level. The NAEP SES index is constructed using NAEP grade 4 mathematics and reading data for 2013 and 2015, respectively. By using data from different subjects and different years, this paper shows that, in spite of the lack of student-reported parental education, the grade 4 NAEP SES index may still be an effective proxy measure to sufficiently account for the impact of socioeconomic factors on student performance.
The criteria to assess the effectiveness of the NAEP SES index include: (1) the amount of variance explained by the SES index controlling for gender and race/ethnicity; and (2) the comparisons of White-Hispanic and White-Black achievement gaps before and after controlling for SES, using NSLP eligibility status and the NAEP SES index, respectively.
The results from the models with 2013 mathematics show that the differences between the observed White-Black and White-Hispanic achievement gaps in the baseline model and the respective achievement gaps after controlling for SES are almost twice as large as those after controlling for NSLP eligibility status alone. Moreover, the index helps to better account for the variance in student performance than the variable for NSLP eligibility status.

Authors

• Qingshu Xie, MacroSys LLC

• Markus Broer, American Institutes for Research