Association for Public Policy Analysis & Management

Residential mobility is associated with a range of adverse children’s outcomes, including detriments in health and academic achievement. However, the factors that can exacerbate or buffer the effects of moving are less well understood. For example, little is known as to whether particular reasons for moving are associated with outcomes. In addition, little is known about the co-occurrence of instability across contexts (residential mobility, household instability) and whether the accumulation of instability is the least desirable for child well-being. Finally, does whenmoves occur matter (early childhood, middle childhood, or adolescence) for children’s development? This work seeks to understand the circumstances under which moving matters for children’s health and school behavior. Specifically, we test whether specific reasons for moving and household instability moderate the relationship between moving and child health and school absence. Follow-up analyses will test whether the developmental period in which moves occur moderates relationships.

We answer these questions using data from Making Connections, a neighborhood transformation initiative implemented in low-income neighborhoods in 10 US cities. This study uses data from waves 2 (2005-2007) and 3 (2008-2011) and focuses on families with children between the ages of 2-17 at wave 3 (N=2,311). At wave 3 parents were asked the date they moved to their current address and an indicator of whether families had moved between the two waves was created. Parents were also asked the primary reason for leaving their previous residence and selected their response from options in five domains:  family, job, housing, neighborhood, and health. Household instability was operationalized as a change in household composition occurring between waves. Parents reported on children’s health and school absence at waves 2 and 3. All models adjust for wave 2 child outcomes and a set of demographic covariates. Sample weights were applied to all models. 

 Half of the sample experienced a move between waves. Moving was adversely related to child health (b=-.03, SE=.01, p<.05) and school absence (b=.28, SE=.04, p<.01). The most commonly reported reasons for moving were housing- (47%), neighborhood- (23%), and family-related reasons (22%). Children in families who moved for a family (b=.05, SE=.01, p<.01) related reason had better health than children who did not move while children in families who moved for housing related reasons (b=-.04, SE=.01, p<.01) had worse health compared to non-movers. Similarly, children in families who moved for a family related reason (b=-.24, SE=.05, p<.01) had fewer absences than children who did not. Child outcomes were also predicted using the interaction between mobility and household instability. The interaction was significantly related to both health (b=-.11, SE=.02, p<.01) and school absence (b=-.51, SE=.08, p<.01). Children in families with no change in household composition but moved had the lowest levels of health. In contrast, children in families with no change in household composition and did not experience a move had the lowest levels of school absence. Additional analyses will examine moderation by the age when move occurred. Results will be discussed in terms of implications for housing policy and services to families.