ABSTRACT
In this paper, we use the Canadian National Longitudinal Survey of Children and Youth data to examine the links between child well-being and neighbourhood 'quality.' This study adds to the literature by (i) investigating the relationship between neighbourhood quality and child health, (ii) by utilizing subjective assessments by individuals familiar with the neighbourhood (i.e., the survey respondent and interviewer), and (iii) by utilizing multiple assessments of neighbourhood quality, and (iv) by investigating several measures of health. Other work has found that controlling for family level characteristics reduces or eliminates the apparent association between neighbourhood quality and health. We find, measuring both child well-being and neighbourhood quality multi-dimensionally, that even after controlling for family level characteristics neighbourhood quality has strong associations with child well-being.
Subject(s)
Child Welfare/statistics & numerical data , Health Status , Residence Characteristics/classification , Social Environment , Adult , Canada/epidemiology , Child , Child Behavior Disorders/epidemiology , Child, Preschool , Family Characteristics , Female , Housing , Humans , Infant , Infant, Newborn , Interviews as Topic , Longitudinal Studies , Male , Multivariate Analysis , Residence Characteristics/statistics & numerical data , Socioeconomic Factors , Sociology, MedicalABSTRACT
We use data from the Ontario Child Health Study (OCHS) to provide the first Canadian estimates of how the empirical association between child health and both low-income and family status (lone-mother versus two-parent) changes when we re-estimate the model with pooled data. Two waves of data provide a better indication of the family's long-run level of economic resources than does one wave. Our measures of health status include categorical indicators and the health utility score derived from the Health Utilities Index Mark 2 (HUI2) system. Consistent with findings from other countries, we find that most outcomes are more strongly related to low-average income (in 1982 and 1986) than to low-current income in either year. Unlike some previous research, we find the quantitative impact of low-income on child health to be modest to large. Lone-mother status is negatively associated with most outcomes, but the lone-mother coefficients did not change significantly when we switched from low-current income to low-average income. This implies that the lone-mother coefficient in single cross-sections is not just a proxy for low-permanent income.
