Residential Mobility during Adolescence: Even “Upward” Neighborhood Mobility Predicts High School Dropout

Families at all points on the socioeconomic continuum experience residential transitions. The reverberating effects of moves in adolescence have been associated with adverse educational outcomes, particularly low high school graduation rates (Wood et al., 1994). However, social capital theory suggests that moving to more affluent neighborhoods may enhance adolescent learning opportunities by exposing youth to additional supports for educational attainment (Coleman, 1988; Sampson, 2012). In fact, federal housing policy and programs assume ‘upward mobility’ promotes educational and employment opportunities, and thus, provides incentives for low-income families to move to better neighborhoods. The present study empirically investigates this assumption. Specifically, we test whether moving to ‘better’ neighborhoods in adolescence predicts higher rates of high school graduation by early adulthood.

The data for this study come from the National Longitudinal Study of Adolescent Health (AddHealth), a nationally representative study following adolescents into early adulthood. The present study uses data from Wave 1 (participants in grades 7-12 in 1994-1995), Wave 2 (grades 8-12 in 1996), and Wave 4 (ages 24-32 in 2007-2008). Youth report of whether they graduated from high school at Wave 4 represents the binary dependent variable. Logistic regressions condition graduation on youth reported number of residential changes between Wave 1 and Wave 2, as well as an array of 23 Wave 1 covariates associated with both educational attainment and propensity to move at the level of the individual (e.g. age, gender, race), family (e.g. parent educational attainment), and neighborhood (i.e. youth reported neighborhood cohesion; parent-reported neighborhood disorder; and Census data on neighborhood risk). Neighborhood mobility from Wave 1 to Wave 2 is captured by comparing change in an index of the four Census neighborhood risk (income, unemployment, poverty, and high school dropout). Youth were identified as remaining in similar neighborhoods, moving up (.20 standard deviation increase in neighborhood risk), or moving down (.20 standard deviation decrease in neighborhood risk).

Findings suggest residential changes in adolescence predict decreased odds of high school graduation a decade later. These effects emerge after controlling for sociodemographic characteristics, and remain statistically significant for moving once (odds ratio =.53, p<.01) or two or more times (OR=.37, p<.01). Importantly, moving to better neighborhoods (OR=.51, p<.01) is equally disruptive as moving to similar (OR=.49, p < .01) and worse neighborhoods (OR=.33, p<.01), as these effect sizes are statistically similar. Implications for housing and educational policy are discussed.