Residential Mobility and Attempted Suicide Among Adolescents: An Individual-Level Analysis

Abstract

Building on the Durkheimian legacy's emphasis on social integration as a determinant of suicidal behavior, many macrolevel studies have observed an association between aggregate rates of geographical mobility and suicide, but little research has explored this connection at the individual level. We use data from 9,594 respondents who participated in two waves of the National Longitudinal Study of Adolescent Health to examine the effect of recent residential mobility on attempted suicide one year later and to explore the mechanisms that potentially transmit this effect. We find that among adolescent girls, recent movers are about 60 percent more likely than nonmovers to report having attempted suicide during the following year and that this difference cannot be readily explained by mover versus nonmover differences in preexisting demographic and family background characteristics. Some of the apparent effect of residential mobility on females' risk of attempting suicide operates through higher rates of victimization and delinquency, lower levels of school attachment, higher rates of social isolation, and a tendency for movers to associate with peers who exhibit delinquent behaviors and who themselves have attempted suicide. In contrast, we find no evidence that mobile female adolescents' deficit of parental social capital or lower levels of school engagement can account for the difference in attempted suicide risk between movers and nonmovers. We also find that residential mobility is not significantly associated with suicide attempts among adolescent males.

ACKNOWLEDGMENTS

This research was supported by grants to the first two authors from the National Science Foundation (SBR-0131876) and by grants to the University at Albany Center for Social and Demographic Analysis from the National Institute of Child Health and Human Development (P30 HD32041) and the National Science Foundation (SBR-9512290). This research uses data from the Add Health project, a program project designed by J. Richard Udry (PI) and Peter Bearman, and funded by grant P01-HD31921 from the National Institute of Child Health and Human Development to the Carolina Population Center, University of North Carolina at Chapel Hill, with cooperative funding from 17 other agencies. Persons interested in obtaining data files from the National Longitudinal Study of Adolescent Health should contact Add Health, Carolina Population Center, 123 West Franklin Street, Chapel Hill, NC 27516-2524 (http://www.cpc.unc.edu/addhealth).

NOTES

Notes

¹ With grounding in the classic work of the Chicago School (Park 1928; Short 1971), the disruptive and destabilizing impact of residential mobility and rapid population change is a seminal theme in community and urban sociology (Kasarda and Janowitz 1974; also see Sampson, Morenoff, and Gannon-Rowley 2002). Lacking traditional controls exercised by families, neighborhoods, and other community institutions, migrants are relieved of the constraints that otherwise induce conformity and are thus free to engage in a variety of nonnormative and deviant behaviors.

² To avoid losing observations to missing data on the control or mediating variables, we use an imputation procedure to retain these cases.

³ The Add Health data do not allow for an examination of the predictors of successful suicides (i.e., those adolescents who succeeded in killing themselves). However, our use of a measure of suicide attempts is consistent with prior literature (e.g., Bearman and Moody 2004). Although few suicide attempts result in death, more than one-third of suicide attempts result in serious injuries necessitating professional treatment (Bearman and Moody 2004). This suggests that identification of the mechanisms that transmit the potentially detrimental effect of mobility on suicide attempts can generate both theoretical and policy implications.

⁴ Ideally, our measure of adolescent residential mobility would distinguish long-distance from short-distance moves since the former are more likely to sever relationship ties in the community of origin and potentially cause more psychological distress among adolescents. Unfortunately, the Add Health data do not allow us to make this distinction. However, Tucker, Marx, and Long (1998) show that short-distance moves have at least as strong an effect as long-distance moves on children's problem behavior. Moreover, since short-distance moves are less likely to alter adolescents' networks than long-distance moves, their inclusion in the category of movers would result in a conservative observed effect of residential mobility on adolescent violence.

⁵ Johnson, Crosnoe, and Elder (2001) also include a measure of truancy in their index of scale of school engagement. We omit this item from our measure because it is included in our index of delinquency.

⁶ On average, respondents have ties to 8.5 friends in their school networks.

⁷ In supplementary analyses, we examined the effect of extracurricular activities on suicide attempts more closely and found that it is not a robust effect. Among females, the bivariate association between participation in extracurricular activities and suicide attempts is not statistically significant, nor is the relationship between these variables significant when only the control variables are included. Although we do not wish to overinterpret or overstate the importance of this finding, one possible explanation for the net positive association between girls' participation in extracurricular activities and suicide attempts involves “role strain” (Goode 1960), in which adolescent girls who are overextended through their participation in many activities become overwhelmed by these pressures to the point of attempting suicide.

⁸ In supplementary analyses, we explored whether the impact of residential mobility on attempted suicide varies by other respondent characteristics, including age, race, family structure, and parent–child relationship quality. However, we found no evidence that these variables significantly moderate the effect of residential mobility on attempted suicide.