Saturday, 15 January 2005 - 12:00 PM

This presentation is part of: Poster Session II

The Capability Approach to Intergenerational Poverty: Accounting for the Differences in Adult Outcomes of Low-income Children

Purpose This study tests a capability model of the intergenerational transmission of economic status. This model is the result of an application of Amartya Sen’s Capabilities approach to intergenerational poverty. The research questions addressed follow: 1. What are the differences in backgrounds between the poor and non-poor adults who were all low-income children? 2. Is there an association between the parental capabilities, measured by parent education, parent satisfaction with self, and family assets, with the adult economic status of low-income children while controlling for parental marital status and number of children in childhood family unit?

Methods The Panel Study of Income Dynamics (1968 to 2001) is the source of data. The study sample consists of the cohort of low-income children, aged 0 to 10 in 1970, and 31 to 41 in 2001. The dependent variable, adulthood economic status, is defined using the family income-to-needs ratio in 2001 (poor=78.14%, non-poor=21.86%). The independent variables include parent’s education, parent’s satisfaction with self, and childhood family asset holdings. Control variables include parent’s marital status and number of children in childhood family unit. Both independent variables and control variable are measured using 1970 variables.

Descriptive statistics are generated on the backgrounds of the sampled respondents. The main statistical method used is the logistic regression since the dependent variable is dichotomous. The model in this study follows:

Y = a + &beta1X1 + &beta2X2 + &beta3X3 + &beta4Z1 + &beta5Z2 + e, (1)

where Y is the log odds of adulthood economic status (ln(p/(1-p))), X1 is the independent variable Parent’s education, X2 is the independent variable parent’s satisfaction with self, X3 is the independent variable childhood family asset holdings, Z1 is the control variable parental marital status, and Z2 is the control variable number of children in childhood family unit.

Results Discernable differences were found in the respondents’ backgrounds. Logistic regression analysis reveals an overall significant model (Wald &chi=41.93, p<0.0001). Results suggest that parental satisfaction with self (b=-0.1582, Wald &chi=10.3082, p<0.01) is significantly related to the probability of a low-income child being non-poor as an adult. In terms of odds, for each unit increment in the satisfaction-with-self variable (corresponding to lowered satisfaction), the odds of a respondent being non-poor decrease by 14.6%. The control variables, parental marital status (b=0.3999, Wald &chi=5.52, p<0.0188) and the number of children in childhood family unit (b=0.1580, Wald &chi=20.90, p<0.0001) were both significant. Results indicate that the odds of a child of married parents being non-poor are 1.5 times the odds of a child of unmarried parents being non-poor. Furthermore, for each additional child in a family unit, the odds of a child being non-poor in adulthood increase by 17%. Implications for Policy and Practice The significance of the capability model indicates that seeking capability expansion promises to be an optimal social policy intervention. This study only provides preliminary evidence for further exploration of the notion of capabilities/functional development. There is a methodological implication as time series analysis would be necessary to fully tap the functional development component of human capabilities.

See more of Poster Session II
See more of Oral and Poster

See more of Celebrating a Decade of SSWR (January 13 - 16, 2005)