Poster #18 - Cumulative effects of poverty stability, duration, and intensity in elementary school on 5th grade academic outcomes

• In Event: Poster Session 20

Abstract

Poverty research focuses on the timing and duration of exposure to poverty, providing evidence for more significant impacts during early childhood and for cumulative effects (Duncan et al., 2015). Student free or reduced-price lunch status (FRL) is the most widely used measure of poverty in education research, and often, children receiving free or reduced-price lunch are grouped together (Harwell & LeBeau, 2010). Such grouping loses important income variance as we know that even small increases in family income are linked to improvements in cognitive abilities (Noble et al., 2015). Another problem is that longitudinal research often assumes that poverty status is stable over time and only reports student FRL status for one year/grade. Little is known about the longitudinal stability of FRL status over time and how much cumulative measures of poverty over time are associated differentially with academic outcomes. We describe stability over time in FRL status for a large and ethnically diverse sample of students from K through 5th grade. We examine how intensity and duration of poverty over time relates to 5th grade academic outcomes.
Data come from a longitudinal project that assessed the school readiness of children attending various ECE programs at age 4 and followed students through elementary school (N = 39,140). Poverty was measured by school FRL status each year (0=not, 1=reduced-price, 2=free meals). Poverty was operationalized in four ways: (1) concurrent 5th grade FRL status, (2) duration – number of years enrolled in free lunch over elementary school, (3) cumulative intensity of poverty across grades, and (4) an ordinal listing of different FRL stability/mobility patterns across grades. Outcome measures were standardized test scores in math and reading and overall teacher-assigned grades (GPA) in 5th grade.
Only moderate stability in FRL receipt across grades was found, with 20% of students changing their poverty status category from grade to grade. Over the course of elementary school, only 58% of the sample was stable in their poverty category, with combinations of movement between all categories observed (Table 1). Multiple regressions revealed significant negative relations between all four dimensions of poverty and 5th-grade outcomes (Table 1). Covariates included cognitive and social skills at school entry, disability status, gender, and ethnicity (Latino, Black, White/other). Cognitive skills at school entry were always the strongest predictor of 5th grade academic outcomes. However, poverty dimensions which weighted both duration and depth of poverty over time were always strong predictors of outcomes, and importantly, stronger than concurrent 5th grade FRL status (Table 1). Figure 1 shows how each dimension of poverty (severity and duration – i.e., dropping to reduced lunch versus free lunch for 1-2 years) matters for 5th grade school outcomes (reading is displayed in the Figure, but it is the same pattern for other outcomes). We show considerable student mobility of poverty status from year to year in elementary school, and that multifaceted metrics of poverty capturing duration and intensity of poverty over time better explain 5th grade academic outcomes than simple concurrent measurements common in the literature.

Authors

• Alison Hundertmark, George Mason University
  Presenting Author

• Adam Winsler, George Mason University