Mothers' Work and Children's Lives: Low-Income Families after Welfare Reform

Appendix A - Measures

107 Appendix A Measures ChILD ouTCoME MEASuRES The WES survey contains a subset of items from the Behavioral Problems Index (BPI) described in Chase-Lansdale et al. (1991). Unfortunately, the WES did not include the entire 28-item BPI at each wave. The items in these scales ask the mother to report on the child’s behavior as she has observed it over the past three months. Mothers respond whether these behaviors are not true, sometimes true, or often true for their child. The externalizing behavior scale (three items) includes items such as “bullies or is cruel or mean to others” and “breaks things deliberately.” This variable ranges from 3 to 9. Alphas at the first and fifth waves, respectively, are 0.48 and 0.65. The internalizing behavior scale (five items) focuses on sadness (“unhappy , sad”), being withdrawn, and feelings (“feels worthless”). Values range from 5 to 15, and the alphas are 0.66 and 0.75 for waves 1 and 5, respectively. These alphas are consistent with those found in the National Longitudinal Survey of Youth (Baker et al. 1993), which also administered the BPI. Our measure of total behavior problems is a 12-item summary index that combines these two scales and includes four additional items measuring fear/ anxiety in the child. Disruptive behavior in school is coded 1 if the mother reports that the child is sometimes or often either disobedient in school or has trouble getting along with teachers; 0 otherwise. School absenteeism problems are coded 1 if the child regularly missed school at least one or more times a month; 0 otherwise. Repeated a grade or placed in special education is coded 1 if either event occurred since last survey. 108 Johnson, Kalil, and Dunifon CuMuLATIvE MEASuRES oF MoThERS’ WoRk ExPERIEnCES We utilize information on mothers’ job tenure, monthly job/employment history, and information collected from self-reported reasons for job changes to characterize employment patterns and the extent of job stability and job mobility over the seven-year study period (February 1997–2004). Using retrospective questions from each wave, we construct cumulative full- and part-time work experience measures that capture the total number of years of work experience accumulated (as of the relevant wave), and the cumulative number of years in which the mother experienced job stability, voluntary job mobility, and job instability. We use the same job transition pattern definitions as discussed in Chapter 2, but summed over the successive periods since February 1997, when respondents were originally observed on the welfare rolls. We also count the total number of years in which the mother experienced fluctuating work hours on the job. These cumulative employment pattern measures are used in the baseline OLS regression models reported in Table 3.4. FAMILy ChARACTERISTICS uSED In ThE FuLL MoDEL Used paid child care services is coded 1 if yes since last survey. Family-income-to-needs ratio is the monthly income-to-needs ratio net of taxes, CPI-U deflated to 1997 dollars. Maternal earnings is measured at the monthly level and CPI-U deflated to 1997 dollars. In the regressions, this is expressed in thousands of dollars to ease interpretation of effect sizes. Received welfare is coded 1 if respondent received any income from FIP/ TANF in past month. Food insufficiency is coded 1 if respondent answers sometimes or often not enough to eat to the following question: “Which of the following best describes the amount of food your household has to eat: enough to eat, sometimes not enough to eat, or often not enough to eat? Moved is coded 1 for yes if the respondent reported moving since the previous survey and/or the residential address changed since the previous survey wave. Evicted is coded 1 for yes if respondent reports being evicted or experiencing an episode of homelessness since the previous interview. Appendix A 109 Neighborhood problems is a summary scale based on 11 self-report items (each item ranges from 1 to 3, with higher scores indicating higher levels of problems) asking the respondent how big a problem the following issues are in her neighborhood: availability of public transportation, availability of affordable housing, slow/no police response, groups of teenagers hanging about, vandalism, prostitution, sexual assault/rape, muggings, gangs, drug use/dealing , and general safety of neighborhood. Neighborhood poverty rate is the proportion of households in poverty in the census tract in which the family lives. Parental...