Now that comparable data for 2004 are available from SIPP, analyses for 2004 can be compared with data for 1997. The two primary purposes of this paper are (1) to see if child well-being as measured in the previous research has improved since 1997 in each of the four work-poverty categories of children; and (2) to see if the patterns observed for children across the four categories of work and poverty status have persisted over the seven-year interval. In addition, in 2004 only, we explore the extent to which the differences across the four work-poverty categories persist when we control for two sets of variables using multivariate analysis. The first set of controls is basic socio-demographics:
- Mothers race/ethnicity (non-Hispanic white, non-Hispanic black, non-Hispanic Asian, non-Hispanic other, and Hispanic);
- Family structure (married, single mother, single father, cohabiting);
- Educational attainment of better educated parent (continuous); and
- Mothers age (continuous).
These controls address the extent to which differences across the four work-poverty categories of children may be due to differences in the demographic and social composition of the four groups.
The second set of variables control for the work pattern of the parent who worked the most hours during the year, that is, the parent with the most hours worked. They are:
- Full-time/part-time status (usually worked at least 35 hours per week, usually worked less than 35 hours per week); and
- Full-year/part-year status (worked at least 50 weeks during year, worked less than 50 weeks during year).
These controls reveal the extent to which differences in the well-being measures across the four work-poverty categories of children may be due to differences in the work pattern of the more employed parent.
Next we provide definitions of the child well-being measures used in the study.