Exploring Proxies for Student Socioeconomic Status: The Context of School and Home Resources

• In Event: Measuring Socioeconomic Status to Contextualize Educational Achievement: A Discussion on Methods, Applications, and Challenges

Abstract

Socioeconomic status has long been considered a crucial variable in explaining student achievement variance and associated with long-term health, cognitive, and socioemotional outcomes (Bradley & Corwyn, 2002; Sirin, 2005). Key student-level educational research variables that are typically used as SES measures are parental educational attainment (Shera, 2014), parental occupational status (Quinn et al., 2016), family income (Tienken et al., 2017), and home resources (Caponera & Losito, 2016). Student eligibility for a free or reduced-price lunch of the National School Lunch Program (NSLP) has also been a widely used student SES measure in research (Harwell & LeBeau, 2010). Information about home resources has been frequently used to measure family income, but may go beyond as proxies for family income (Sirin, 2005). There are also school-level variables accounting for school resources such as teacher experiences and qualifications (Gimbert, Bol, & Wallace, 2007), the proportion of teachers and school staff (Pribesh, Gavigan, & Dickinson, 2011), availability of educational resources in the classroom (Chetty et al., 2011), and school funding (Muijs, et al., 2009).
However, concerning data accuracy, some of these student-level measures can be problematic. For example, information collected about parental occupational status tends to be less accurate amongst younger students because these students may not have a clear understanding of their parents’ occupations (APA 2007; Cowan et al., 2012). Although some national surveys do include questions that directly ask parents about their educational levels and occupational status, there are validity concerns related to nonresponse bias, social desirability, and over- or under-estimation (Malthiowetz, Brown, & Bould, 2001). Due to changes in educational policies, the measure of student eligibility of NSLP has become less valid and reliable over time (Cowen et al., 2012). Moreover, direct measures of family income are intrusive, and students may not be able to provide accurate responses to income questions about their parents (APA 2007; Cowan et al., 2012).

The purpose of this paper is to explore the viability of the school and home resources variables as proxies for student SES by using nationally representative data from the 2019 National Assessment of Educational Progress (NAEP) and associated school, teacher and student contextual questionnaire data at grades 4 and 8. The present paper analyzes the NAEP performance scores, student demographic variables (i.e., gender, race/ethnicity), and home resources variables (i.e., number of books at home, access to computer and internet) from the student questionnaire. The present paper also analyzes the relations between school resources and home resources to student performance on the NAEP assessments. Variables related to school resources include teacher experiences (i.e., years teaching the subject), teacher preparation (i.e., degree earned in the subject taught, certifications), and various educational resources indicated on the school questionnaire.

Findings demonstrate the viability of some school and home resources as potential proxies for student SES. The validity of these potential proxies is evaluated. Multiple regression models on student achievement are conducted to examine the applicability of the school and home resources proxies for student SES. Discussions of variance explained in student achievement by various school and home resources proxies for SES are provided.

Authors

• Judy H. Tang, Westat

• Ning Rui, Amazon