Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Aug 12.
Published in final edited form as: Soc Work. 2008 Jul;53(3):267–278. doi: 10.1093/sw/53.3.267

Low-Wage Maternal Employment and Parenting Style

Aurora P Jackson 1,*, Peter M Bentler 2, Todd M Franke 3
PMCID: PMC2725923  NIHMSID: NIHMS125778  PMID: 19275122

Abstract

This 3-year longitudinal study investigated whether low-wage employment was associated with improved psychological and parenting outcomes in a sample of 178 single mothers who were employed and nonemployed current and former welfare recipients both before and subsequent to the passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996. Participation in employment predicted fewer depressive symptoms and less negative parenting style over time. Employment at time 1 was associated with a reduced likelihood of receiving welfare in the interim between Time 1 and Time 2, less financial strain at Time 2, and (through these) a decrease in mothers’ depressive symptoms at Time 2. Fewer depressive symptoms at Time 2, in turn, predicted less negative parenting style, net of the mothers’ earlier demographic, mental health, and parenting characteristics. Mothers with higher educational attainment were more likely to be employed (and to earn more) at both time points. Implications of these findings for welfare policies are discussed.

Keywords: maternal employment, single mothers, parenting, depressive symptoms, welfare reform

The 1996 welfare reform law changed the main cash welfare program for poor, mostly single-mother families by mandating stricter work requirements and imposing a 5-year lifetime limit on the receipt of benefits. As a consequence, large numbers of poor single mothers have left the welfare roles and entered the work force. Results from studies of mothers leaving welfare for work suggest that participation in employment can be beneficial if income is adequate (Duncan & Chase-Lansdale, 2001; Morris, Huston, Duncan, Crosby, & Bos, 2001). This study used data gathered over 3 years from a sample of current and former single-mother welfare recipients in New York City to investigate whether increases in employment were associated with improved mental health and parenting outcomes over time. This is important because poverty and poor parental mental health are consistent risk factors for less optimal parenting and childhood disruptive behavior disorders (Garrett, Ng’andu, & Ferron, 1994; McLoyd, 1998; Weissman et al., 2006). Our study is restricted to single black mothers because they are disproportionately represented among the very poor and the welfare dependent (Duncan, 1991; Wilson, 1987, 1997).

Maternal Employment

In general, the research on maternal employment has focused either on middle-class married white families (for example, Desai, Chase-Lansdale, & Michael, 1989; Harvey, 1999) or, more recently, with respect to welfare recipients, on the impact of experimental employment programs and/or welfare reform demonstration projects on families and children (for example, Bos et al., 1999; Duncan et al., 2001; Gennetian & Miller, 2002; Huston et al., 2001). Concerning the former, the findings suggest that maternal employment has positive effects in families, especially when mothers want to be employed (see, also, Hoffman & Youngblade, 1999; Jackson, 1993).

Concerning current and former welfare recipients, the findings suggest that there are positive effects on mothers and young children in programs that increase both employment and income (Bos et al., 1999; Gennetian & Miller, 2002; Huston et al., 2001), although studies comparing poor children in families receiving welfare and those in families not receiving welfare have found no differences (Duncan et al., 2001). Some suggest that leaving welfare is neither beneficial nor harmful for children (Kalil et al., 2001; see, also, Chase-Lansdale et al., 2003). However, other research has found a negative relationship between welfare receipt and children’s outcomes, controlling for income level (Haveman & Wolfe, 1995). Still others have found that the higher family incomes associated with maternal employment—even low-wage employment—can lead to improvements in children’s well-being (Smith et al., 2001; Smith et al., 2000).

Another line of inquiry has examined whether and how the working conditions of former welfare recipients influence maternal and child outcomes. One such study found that among former welfare recipients in an urban Michigan county, neither long working hours, erratic and irregular working schedules, nor non-daytime shifts were associated with negative behavioral outcomes for young children (Dunifon & Kalil, 2005). Others have suggested that negative associations between low-wage employment and maternal and child outcomes may occur when job quality is considered (Raver, 2003). Menaghan and Parcel (1995) used data from the National Longitudinal Survey of Youth to examine this issue. While less than a third of their sample was low-income and single (most were middle class and married), they found that among single mothers, the least adequate parents were those who were either not employed or who became employed in a low-wage job. Tilly and his colleagues (1996; Moss & Tilly, 2001) found that wages often are an indicator of job quality, that undesirable aspects of jobs are on average offset by higher wages, and that low-income single women as a group have to do more paid work (i.e., work longer hours) to sustain their families.

Raver (2003) has suggested, aptly, that findings such as these open debate to broader questions regarding the effect of low-wage employment on parenting among poor families in contexts other than experimental welfare-reform-related research projects (for example, Brody & Flor, 1998; Jackson, Brooks-Gunn, Huang, & Glassman, 2000; Raver, 2003). Moreover, most of the existing experimental studies on the effects of employment on current and former welfare recipients were carried out before the 1996 welfare reforms when many of the income supports were more liberal and there were no time limits (for a discussion of this, see Chase-Lansdale et al., 2003). While a new wave of nonexperimental studies on the consequences for children of mothers’ transitions from welfare to work is growing, much of this research compares mothers of different races and different marital statuses (for example, Chase-Lansdale et al., 2003; Dunifon, Kalil, & Danziger, 2003; Dunifon & Kalil, 2005). The present study—different from other recent research investigations—focuses on individual differences among poor and near-poor single black mothers, a group that is more likely than others to experience stressful events that put them at risk for psychological distress (McLoyd, 1990). In addition, we focus on mothers who were current and former welfare recipients both before and subsequent to the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996. Other studies were carried out either pre- or post-PRWORA (for example, Chase-Lansdale et al., 2003; Dunifon et al., 2003; Dunifon & Kalil, 2005). As such, our data allow us to add to this literature a comparison of mothers who were employed and nonemployed both before and following the 1996 welfare act.

Conceptual Model

In addressing the role of maternal psychological well-being in linking employment in the low-wage job market to parenting behaviors for poor and near-poor single black mothers, our model is informed theoretically by the works of Conger et al. (1992) and McLoyd et al. (1990; McLoyd, Jayarantne, Ceballo, & Borquez, 1994). Conger and his colleagues postulate that objective economic circumstances affect parents’ experience of economic pressure, which reduces their psychological well-being. Psychological distress, in turn, disrupts effective parenting behaviors (leading, thereby, to less optimal child outcomes). McLoyd and her colleagues postulate that an accumulation of risks is associated with economic hardship in black families (e.g., single-parent status, low parental educational attainment, unemployment). These risks may have different effects on the development of young black children, depending on the presence of protective factors that mediate between economic hardship and child developmental outcomes via more adequate parenting.

In this study, we considered maternal psychological well-being, higher educational attainment, and being employed—even in a low-wage job—to be protective factors. However, it also is possible that working long hours to earn more income might be quite stressful for mothers in low-wage jobs (see, for example, Dunifon & Kalil, 2005). In such circumstances, long working hours might predict greater psychological distress and, thereby, more negative parenting. We reasoned, nevertheless, that if single mothers must do more paid work to sustain their families (Moss & Tilly, 2001; Hofferth et al., 2000), then doing so might be associated with better maternal psychological well-being (and, thereby, more positive parenting) in poor and near-poor black families (see Alvarez, 1985, for a discussion of how mothers’ positive motivation for working might offset negative aspects of employment). In addition, based on the findings of Haverman and Wolfe (1995) suggesting that mothers who do not receive welfare benefits are in better mental health, we reasoned as well that not receiving welfare benefits in the interim between Time 1 and Time 2 would be associated with lower depressive symptoms.

Our conceptual model (Figure 1) tested the effect of maternal employment on family outcomes by controlling for earlier levels of those outcomes (see, for example, Duncan, Yeung, Brooks-Gunn, & Smith, 1998). Figure 1 shows paths from depressive symptoms and parenting style at Time 1 to their counterparts at Time 2 and from welfare receipt at Time 1 to its counterpart in the interim between Time 1 and Time 2. Consistent with the theoretical and research literatures already reviewed, welfare receipt and earnings from employment were selected as objective economic circumstances that might affect parents’ experience of economic pressure or financial strain.

Figure 1.

Figure 1

Open in a new tab

Conceptual model of demographic characteristics, mothers’ employment, welfare receipt, financial strain, depressive symptoms, and parenting style.

In addition, Figure 1 shows direct paths from depressive symptoms and welfare receipt early on, maternal educational attainment, and the number of children in the household to mothers’ participation in employment at Time 1. There is evidence that poor psychological functioning among welfare-dependent mothers might be a barrier to gainful employment (Danziger et al., 2000; Hershey & Pavetti, 1997; Wolfe & Hill, 1995). From the bodies of research examining the relationship between work and welfare (Duncan & Brooks-Gunn, 1997; Harris, 1996), we selected maternal educational attainment as an important predictor of earnings. Lack of education is likely to constrict opportunities for jobs paying higher wages, which might, in turn, be associated with less stable employment and a greater likelihood of receiving welfare benefits in the interim between Time 1 and Time 2. There is evidence, as well, that large family size (number of children) can tax family resources and may constrain single mothers’ employability with respect to working hours and wages (for example, Danziger et al, 2000; Jackson, Tienda, & Huang, 2001). This also would be related to a greater need of welfare benefits in the interim between Time 1 and Time 2. Financial strain at Time 2 in our model is proposed as a key mechanism through which wages influence depressive symptoms and parenting style at Time 2. A number of studies support this likelihood (Conger et al., 1992; Jackson et al., 2000, for example) and Figure 1 shows the corresponding paths from participation in employment at Time 1 to its Time-2 counterpart and from these to financial strain and, in turn, depressive symptoms at Time 2. Finally, our expectation that higher levels of depressive symptoms at Time 2 would be related directly to the quality of parenting at Time 2 (note the corresponding path) was informed by studies that have found that maternal depression is associated with diminished nurturance toward children and less adequate parenting (Downey & Coyne, 1990; McLoyd, 1990; McLoyd et al., 1994).

In sum, Figure 1 hypothesizes that earlier assessments of family background characteristics (mothers’ educational attainment, number of children in the household), psychological characteristics (mothers’ depressive symptoms and parenting style), and welfare receipt at Time 1 are exogenous variables that may possibly predict the extent of mothers’ employment and earnings at Times 1 and 2 and mothers’ psychological and parenting outcomes at Time 2. In addition, our model postulates that employment might influence mothers’ psychological well-being and parenting indirectly through its association with welfare receipt between Time 1 and Time 2 and financial strain at Time 2. This expectation was based on recent reports of the positive impact of employment activities on maternal psychological functioning and parenting (Duncan & Chase-Lansdale, 2001; Huston & Rosenkrantz Aronson, 2005; Jackson et al., 2000; Raver, 2003). It is important to stress, nevertheless, that even though the present data are longitudinal, we are dealing with complicated social relationships and a scientific setting in which precise measurement of these relationships is difficult at best. A two-wave longitudinal study, moreover, does not match an experiment in terms of causal inference (see, for example, Jackson & Scheines, 2005). With these caveats in mind, we turn to a description of the sample and measures. This is followed by the results and a discussion of the findings and implications.

METHOD

Sample

First interviewed between February 1996 and January 1997, participants in this study consisted of 188 current and former single-mother welfare recipients (93 employed, 95 nonemployed) and their preschool children at Time 1. The mothers resided in three communities in New York City—Bedford-Stuyvesant in Brooklyn, Harlem in Manhattan, and Jamaica in Queens—with large numbers of low-income black families. Recruited through the Office of Employment Services of the New York City Human Resources Administration, the sample consisted of 266 randomly selected mothers with a 3- or 4-year-old child. For the initial interview, a 71% response rate was achieved (see, for example, Jackson et al., 2000). For the final interview (between July 1998 and December 1999), the sample consisted of 178 mothers (130 employed, 48 nonemployed) and their early school-age children; 95% of those first interviewed. For each of the two interviews, mothers and focal children were visited in their homes for 1½ to 2 hours. During each visit, mothers completed a questionnaire focusing on individual and family characteristics and received $50.00 in total for their time.

Measures

Corresponding to the model delineated in Figure 1, description of the measures proceeds across constructs from left to right. Except for single-item measures, all variables included in the analyses are scales whose values represent the mean. Alpha coefficients were obtained for scales with three or more items. When calculating the mean value on scales, items were reversed as necessary so that, with the exception of parenting style, a higher score indicates more of the attribute named in the label. For parenting style, a higher score indicates more negative parenting, as explained below.

The Center for Epidemiological Studies Depression (CES-D) scale (20 items, alpha = .88 at Time 1 and .89 at Time 2) was used to measure depressive symptoms. Mothers were asked to indicate on a four-point scale (0 = less than once a day to 3 = most or all of the time) how often during the past week they felt depressed, lonely, sad, unusually bothered by things, or that they could not get going. This scale is not intended as a measure of clinical depression, but groups with scores of 16 or above are considered to be at risk for depression (Radloff, 1977). Negative parenting was measured using the Parenting Stress Scale (Abidin, 1990; 6 items, alpha = .72 at Time 1 and .64 at Time 2). Mothers were asked to indicate on a six-point scale (0 = not at all to 5 = completely) how true each of the following statements was for them: “I know I should always enforce my rules, but if I’m sad or tired, sometimes I let things go and other times I lose my temper;” “It is sometimes necessary to discipline a child with a good, hard spanking;” “Sometimes I lose patience with my child’s questions and demands, and I just don’t listen to him or her anymore;” “I often feel angry with my child;” “When a parent asks a child to do something, the child should just do it without having to be told why;” “By the end of a long day, I find it hard to be warm and loving toward my child.” A higher score on this scale indicates more negative parenting.

Mothers’ educational attainment was indicated on a five-point scale (1 = grade school to 5 = BA/BS degree) that asked mothers to give the highest level of education they had completed. Number of children in the household was constructed from mothers’ answers to questions about the make-up of their household. Welfare receipt was indicated by whether the mother received welfare benefits at Time 1 and/or in the interim between Time 1 and Time 2. In these analyses, we focused on receipt of cash welfare benefits and these data were coded 0 if no and 1 if yes. Financial strain (2 items) was measured by asking mothers to indicate on a four-point scale (developed by McLoyd et al., 1994) how often they had decided not to buy something they really needed for themselves or their children because they couldn’t afford it (1 = not at all to 4 = a lot), and lately how difficult they had found it to pay bills (1 = not at all difficult to 4 = very difficult). These are dimensions of objective circumstances and subjective feelings of economic hardship. Participation in employment was constructed from mothers’ answers to questions about their weekly working hours and hourly wages at Time 1 and Time 2. For example, mothers were asked how many hours, on average, they worked each week. Then, they were asked how much they earned hourly, weekly, or monthly, before taxes (see, Jackson et al., 2000, for a similar measure of income). Women who were not currently working at the time of each interview were assigned a value of $0 for earnings. A variable was constructed designating hourly pay at each wave. Informed by the work of Moss and Tilly (2001), we considered employed mothers’ wages to be an indicator of job quality, especially for single black mothers transitioning from welfare to work.

RESULTS

Overview of Analyses

First, descriptive statistics are provided on the economic, psychological, parenting, and demographic variables in our model. Then, simultaneous SEM-fitting techniques were used to test hypotheses regarding the effect of employment on mothers’ depressive symptoms and parenting style at Time 2, net of prior parenting and mental health, welfare receipt, and a set of demographic characteristics. Structural equation models were estimated using maximum likelihood with robust statistics in EQS (Bentler, in press) to examine the goodness of fit for a model that included the three exogenous variables of mothers’ depressive symptoms, parenting style, and welfare receipt, as well as the two exogenous demographic characteristics of mothers’ educational attainment and number of children in the household. These five variables were modeled as predictors of the endogenous variables of depressive symptoms and parenting style at Time 2. As illustrated in Figure 1, welfare receipt between Time 1 and Time 2, financial strain at Time 2, and two factors representing, respectively, work extent and pay at Time 1 and work extent and pay at Time 2 were hypothesized to function as mediating endogenous variables. We created factors representing work extent and pay (F1 and F2) because these variables correlated above .8 and thus clearly mirror a common factor of income.

Description of the Sample

The final sample consisted of 178 mothers and children. On average, the mothers were 31.7 years of age at Time 2; the focal children were 6.6 years old (range was 5 to 8). Close to a third of the mothers (31.5%) had completed high school and about half (52.8%) had some education beyond high school. Although we considered any education or training after high school education beyond, about 4% of the sample had a bachelor’s degree. At Time 1, the mothers worked, on average, 34.8 hours a week (SD = 12.5) and earned $4.34 an hour (SD = 4.83); at Time 2, the corresponding figures were 37.7 weekly working hours (SD = 12.5) at $7.33 an hour (SD = 6.36). The mothers had somewhat higher levels of depressive symptoms at Time 1; the means, at Times 1 and 2, respectively, were 15.5 (SD = 10.3) and 15.1 (SD = 10.3). They had, on average, 2 children (mean = 2.3; SD = 1.3). At Time 1, 59.0% of the mothers received cash welfare benefits; in the interim between Time 1 and Time 2, 61.2% did so.

Descriptive Analyses

Correlational analyses (Table 1) revealed that higher maternal educational attainment was associated with working more hours weekly and with higher pay at both time points and these, in turn, were associated with less financial strain at Time 2. Significant positive associations also were obtained between financial strain at Time 2 and depressive symptoms at Time 2. Fewer depressive symptoms at Time 2 were associated with working more hours, earning more income, not receiving cash welfare benefits, and less negative parenting style at Time 2.

Table 1.

Low-Income Single Mothers’ Employment, Psychological Well-Being, and Parenting: Correlations and Descriptive Statistics (N = 178)

Variables 1 2 3 4 5 6 7 8 9 10 11 12 13
1. Educational attainment --
2. Number of children −.22** --
3. Weekly working hours T1 .17* −.17* --
4. Weekly working hours T2 .36*** −.12 .33*** --
5. Hourly pay T1 .26*** −.13 .83*** .27*** --
6. Hourly pay T2 .42*** −.08 .34*** .80*** .37*** --
7. Welfare receipt T1 −.15* .00 −.43*** −.16* −.47*** −.23** --
8. Welfare between T1 & T2 −.27*** .18* −.44*** −.29*** −.52*** −36*** .47*** --
9. Financial strain T2 −.03 .04 −.17* −.16* −.18* −.15* .11 .14 --
10. Depressive symptoms T1 −.22** .01 −.17* −.16* −.22** −.23** .14 .23** .27*** --
11. Depressive symptoms T2 −.22** −.06 16* −.19* −.15* −.22** .19* .27*** .30*** .55*** --
12. Negative parenting T1 −.19* .10 −.14 −.12 −.19* −.19* .11 .14 .22** .31*** .29*** --
13. Negative parenting T2 −.18* .04 −.06 −.13 −.11 −.16* .11 .13 .16* .27*** .35*** .68*** --
Mean 4.4 2.3 34.8 37.7 4.34 7.33 .60 .61 2.6 15.5 15.1 1.5 1.7
SD 1.4 1.3 12.5 12.5 4.83 6.36 .49 .49 1.0 10.3 10.3 1.0 1.0
Open in a new tab

Note. Dummy codes for welfare receipt T1 and welfare receipt between T1 & T2: 0 = no, 1 = yes.

*

p < .05.

**

p < .01.

***

p < .001

Model Estimation

Although we employed maximum likelihood to examine the empirical credibility of the proposed conceptual model (Figure1), we anticipated that some of the variables might not be normally distributed. There was a somewhat large Mardia’s normalized coefficient of multivariate kurtosis (3.1), indicating the need for robust corrections to standard errors and Satorra-Bentler test statistics (Mardia, 1974; Satorra & Bentler, 1994). A number of covariances and paths were nonsignificant (and hence were dropped), and a correlated error term for residuals in hours worked and pay was added to the final model, which is presented in Figure 2. All remaining covariances and paths in Figure 2 were statistically significant (p < .05). The model, with 55 degrees of freedom, produced a maximum likelihood chi-square of 67.8 (p = .11), as well as a comparative fit index (CFI) of .98, and a root mean square of error of approximation (RMSEA) of .04, all indicating excellent fit to the data.1 The robust statistics yielded similar conclusions, with a Satorra-Bentler chi-square of 72.6 (p = .06), robust CFI of .98, and robust RMSEA of .04.

Figure 2.

Figure 2

Open in a new tab

Final Structural Equation Model. N = 177 (1 case with missing data); chi-square = 67.83, p = .11, CFI = .98, RMSEA = .04. The data are nonnormal and the robust Satorra-Bentler chi-square = 72.6, p=.06. Standardized parameter estimates shown (*p<.05). Note: F1 = factor representing extent of employment and pay Time 1; F2 = factor representing extent of employment and pay Time 2. Residuals not shown.

Figure 2 provides the standardized parameter estimates, which are easier to interpret than unstandardized coefficients, as they represent correlations (two-way arrows) or beta weights (one-way arrows). The path between mothers’ depressive symptoms at Time 1 and depressive symptoms at Time 2 is (Beta =) .51, indicating that earlier levels of depressive symptoms predicted later levels, which in turn exhibited the expected positive relationship to negative parenting style at Time 2 (Beta = .16). Consistent with the theoretical expectation, moreover, depressive symptoms (Beta = −.15) and welfare receipt at Time 1 (Beta = −.46), as well as mothers’ educational attainment (Beta = .13) and the number of children in the household (Beta = −.12), were associated directly with the extent of employment and pay at Time 1 (F1), which in turn predicted working and earning more at Time 2 (F2) (Beta = .31), reduced welfare receipt in the interim between Times 1 and 2 (Beta = −.40), and less financial strain at Time 2 (Beta = −.20). In other words, higher educational attainment, lower levels of depressive symptoms and welfare receipt, and fewer children early on were associated directly with a greater likelihood of mothers’ participating in employment at Time 1. Educational attainment also had a further direct (and positive) effect on employment and pay at Time 2 (F2) (Beta = .35). Not only were depressive symptoms and welfare receipt at Time 1 associated with reduced employment and pay at Time 1 (F1), both depressive symptoms (Beta = −.05) and welfare receipt at Time 1 (Beta = −.14) additionally had significant negative indirect effects on employment and pay at Time 2 (F2) at p < .05.

While neither of the Time-1 and Time-2 employment and pay factors was related directly to depressive symptoms at Time 2, the negative effects of working and earning more at Time 1 (F1) on receipt of welfare in the interim between Time 1 and Time 2 and on financial strain at Time 2, and the positive effects of these (Beta = .13 and .16, respectively, for welfare receipt in the interim and financial strain) on depressive symptoms at Time 2 combined to yield a significant (p < .05) indirect effect of F1 on Time-2 depressive symptoms (indirect effect = −.08). Stated differently, working and earning more at Time 1 (F1) predicted reduced depressive symptoms at Time 2 via two mechanisms: the mediation effect of welfare receipt in the interim between Time 1 and Time 2 and that of financial strain at Time 2; that is, by the reduced likelihood of receiving welfare (between Time 1 and Time 2) and diminished financial strain subsequently. It is worth noting, in fact, that working and earning more at Time 1 also predicted reduced negative parenting at Time 2 (p < .05), although the standardized indirect effect is very small (Beta = −.01).2

Finally, an additional structural equation model (not shown; available upon request) was estimated for the subsample of mothers who were not employed at Time 1 (N = 95). For these mothers, becoming employed by Time 2 was related directly and significantly to less negative parenting at Time 2. Taken together, these data suggest that employment at Time 1 (pre-PRWORA) was more beneficial with respect to the psychological and parenting outcomes in this study than was employment at Time 2 (post-PRWORA), except for mothers who were not employed at Time 1 and who became employed by Time 2 (Jackson et al., in press).

DISCUSSION

Using longitudinal data gathered over 3 years, this study investigated whether increases in low-wage employment were associated with improvements in depressive symptoms and negative parenting style in a sample comprised of single mothers with young children who were current and former welfare recipients both before and subsequent to the passage of PRWORA, the 1996 welfare reform act. This is important because economic hardship and poor parental mental health are consistent risk factors for less optimal parenting and poor developmental outcomes for poor and near-poor children (Garrett et al., 1994; McLoyd, 1998; Weissman et al., 2006). We tested a conceptual model hypothesizing that objective economic circumstances affect parents’ experience of economic pressure, which reduces their psychological well-being. Psychological distress (or reduced psychological well-being), in turn, disrupts effective parenting behaviors (Conger et al., 1992; McLoyd et al., 1990, 1994). We hypothesized further that being employed (an objective economic circumstance) would influence mothers’ psychological wellbeing and parenting style (at Time 2) indirectly through its associations with welfare receipt in the interim between Time 1 and Time 2 and financial strain (economic pressure) at Time 2. We employed maximum likelihood structural equation modeling to test the empirical credibility of our conceptual model.

The results were generally consistent with our expectations. Earlier assessments of family background, psychological, and objective economic characteristics predicted the extent of mothers’ employment and earnings at Time 1 and Time 2 and mothers’ psychological and parenting outcomes at Time 2, as expected. More explicitly, higher educational attainment, fewer children in the household, not receiving welfare at Time 1, and fewer depressive symptoms also at Time 1 were associated with mothers’ greater likelihood of being employed at Time 1. Educational attainment also was a predictor of employment at Time 2; i.e., more highly educated mothers were more likely to be employed and earn more not only at Time 1 but also at Time 2. However, even though mothers earned more at Time 2 ($7.33 an hour versus $4.34), it was employment (and pay) at Time 1—not at Time 2—that predicted fewer depressive symptoms and, through the latter, less negative parenting style at Time 2.

One explanation for this finding is that mothers in this study who were employed at Time 1 were already working when the PRWORA legislation was enacted. In comparison with their nonemployed counterparts, they also were significantly lower in depressive symptoms (mean = 13.84 versus 17.22 on the CES-Depression scale) and better educated (mean = 4.63 versus 4.12 vis-à-vis educational attainment) at the outset. Recall that groups with scores of 16 or above on the CES-D are considered to be at risk for depression (Radloff, 1977). Previous research has demonstrated positive associations between single employed black mothers’ educational attainment and preference for employment, as well as among higher educational attainment, consistency between actual and preferred employment status, and fewer depressive symptoms (Jackson, 1993). It is possible that mothers who were employed at Time 1 were working because they chose to do so (preference), whereas a share of those who became employed by Time 2 did so because of sanctions imposed by PRWORA. Perhaps for some of these mothers, being employed was not their preferred status (Jackson, 2003-a). It also is possible that the conditions of their employment were less satisfactory than were those of their counterparts who began the study employed, even though becoming employed by Time 2—in separate analyses of the subsample of mothers who were nonemployed at Time 1—was related directly to less negative parenting.

While there is some evidence that employed women—and, in particular, mothers who spend more time at work—are more positive toward their young children, provide higher quality home environments, and are better off economically (Crockenberg & Litman, 1991; Huston & Rosenkrantz Aronson, 2005; Moss & Tilly, 2001; Zaslow et al., 1985), the “work first” emphasis of the PRWORA legislation presumes that most mothers who receive welfare benefits are employable and that the jobs they are able to get will pay a living wage. However, personal problems and high levels of stress can diminish the ability of some to secure meaningful employment (Danziger et al., 2000; Hershey & Pavetti, 1997; Wolfe & Hill, 1995). Dunifon and Kalil (2005) have reported, nevertheless, that among those who do become employed, long working hours are not necessarily associated with negative outcomes. In the present study, the average mother worked full time both at Time 1 and at Time 2. Longer working hours meant higher pay and less financial strain; the latter, especially for those who were employed at the outset, predicted better psychological and parenting outcomes. Yet, although most of the mothers in this study had some education beyond high school (but not college) and worked full time, they were still relatively poor. The positive associations between educational attainment (beyond high school), employment, and parenting suggest that an investment in early and post-high school educational resources (including college) for poor and near-poor single mothers is well worth consideration by policy makers (see, for example, Jackson, 2003-a). This would be an investment in better employment opportunities and pay for such mothers and, quite likely, better developmental outcomes for their children; social work professionals and NASW could advocate for policies allowing welfare recipients to be supported in getting as much education and training as they need to pursue the jobs they want. This would be consistent with research reports vis-à-vis positive motivation and pursuing preferences (see, for example, Alvarez, 1985; Hoffman & Youngblade, 1999; Jackson, 1993). Future research employing experimental methods could test this conjecture.

In addition, if these findings are valid, further implications for social welfare policies and clinical practice might include making clinical services delivered by social workers available to welfare-dependent mothers within welfare departments and through partnerships with family service agencies in the private sector. For example, welfare offices—in partnership with social work agencies—could establish in-house (or agency based) supportive services units with a cadre of masters level social workers who might focus engagement, assessment, and intervention efforts on nonemployed single mothers with personal problems and high levels of stress that diminish their ability to secure meaningful employment. Such units would extend the “work first” emphasis of PRWORA by making supportive and clinical services readily available when needed. To our knowledge, very few clinical social work services are delivered in (or in collaboration with) welfare offices. This could be done much like social work units in hospitals and some agency-school collaborations. Again, future research could test the efficacy of such a partnership with respect to employment, psychological well-being, and parenting outcomes for special populations of welfare recipients, as well as behavioral and school readiness outcomes for their young children.

Conclusion

We have previously acknowledged several caveats with respect to the present data (Jackson, Bentler, & Franke, 2006). First, the sample is relatively small and the mothers were residing in New York City. Further research with additional samples from other cities is needed to explore whether family background and psychological characteristics predict the extent of employment and earnings among special populations of current and former single-mother welfare recipients with young children and whether the latter are predictive of better mental health and parenting outcomes over time. Second, we measured employment extent and pay at only two points (i.e., Time 1 and Time 2), and used only one indicator of income at each point (i.e., hourly pay). Hence, future research might examine whether the present findings could be generalized across populations of current and former welfare recipients with different employment histories and different levels of family income. Third, although the present study used longitudinal data and controlled for earlier levels of the psychological and parenting outcomes, causal inferences about the relations among maternal employment, depressive symptoms, and parenting styles would be inappropriate. As stated earlier, a two-wave study does not match an experiment with respect to presumptions about causality. Fourth, the data were self-report and we do not know the extent to which they correspond to actual behavior. The data also were collected during the economic boom of the 1990s. As such, our findings must be considered preliminary and in need of replication.

Still, our data are unique in that they were collected before and after the passage of the PRWORA legislation, the sample was drawn randomly, and we examined individual differences among poor and near-poor single black mothers, a group that is more likely than others to experience stressful events that put them at risk for psychological distress. There are few such data.

Acknowledgments

This research was assisted by grants to the first author from the William T. Grant Foundation, the Russell Sage Foundation, and the National Institute of Mental Health (#1 RO3 MH56063-01)

Footnotes

1

A single respondent had 2 of 13 variable scores missing, and this respondent was eliminated in the main analysis. An analysis using this case was carried out using the full information maximum likelihood method. It yielded essentially identical parameter estimates (maximum difference was .01), identical parameter test conclusions, and the very similar chi-square statistic of 68.5 (p = .10). The Yuan-Bentler statistic (Yuan & Bentler, 2000) that corrects for nonnormality was 71.9 (p = .06), virtually identical to the Satorra-Bentler statistic for the complete data.

2

Although the standardized indirect effect is very small (−.01), its significance can be verified by tracing the two paths from F1 to parenting style at Time 2.

By path tracing:
  • −.2 * .16 * .16 + (−.4) * .13 * .16
  • = −.00512 − .00832 = .01344.

Contributor Information

Aurora P. Jackson, Professor of Social Welfare, University of California, Los Angeles.

Peter M. Bentler, Professor of Psychology and Statistics, University of California, Los Angeles

Todd M. Franke, Associate Professor of Social Welfare, University of California, Los Angeles

References

  1. Abidin RR. The parenting stress index short form. Charlottesville, VA: Pediatric Psychology Press; 1990. [Google Scholar]
  2. Alvarez WF. The meaning of maternal employment for mothers and their perceptions of their three-year-old children. Child Development. 1985;56:350–360. [PubMed] [Google Scholar]
  3. Bentler PM. EQS 6 structural equations program manual. Encino, CA: Multivariate Software, Inc.; (in press) ( www.mvsoft.com). [Google Scholar]
  4. Bos J, Huston A, Granger R, Duncan G, Brock R, McLoyd V. New hope for people with low-incomes: Two-year results of a program to reduce poverty and reform welfare. New York: Manpower Demonstration Research Corporation; 1999. [Google Scholar]
  5. Brody GH, Flor DL. Maternal resources, parenting practices, and child competence in rural single-parent African American families. Child Development. 1998;69:803–816. [PubMed] [Google Scholar]
  6. Chase-Lansdale PL, Moffitt R, Lohman B, Cherlin A, Coley R, Pittman L, Roff J, Votruba-Drazal E. Mothers’ transitions from welfare to work and the well-being of preschoolers and adolescents. Science. 2003;299:1548–1552. doi: 10.1126/science.1076921. [DOI] [PubMed] [Google Scholar]
  7. Conger RD, Conger KJ, Elder GH, Lorenz FO, Simons RL, Whitbeck LB. A family process model of economic hardship and adjustment of early adolescent boys. Child Development. 1992;63:526–541. doi: 10.1111/j.1467-8624.1992.tb01644.x. [DOI] [PubMed] [Google Scholar]
  8. Crockenberg S, Litman C. Effects of maternal employment on maternal and two-year-old child behavior. Child Development. 1991;62:930–953. [Google Scholar]
  9. Danziger S, Corcoran M, Danziger S, Heflin C, Kalil C, Kalil A, Levine J, Rosen D, Seefeldt K, Siefert K, Tolman R. Barriers to the employment of welfare recipients. In: Cherry R, Rodgers WM, editors. Prosperity for all? The economic boom and African Americans. New York: Russell Sage Foundation; 2000. pp. 245–278. [Google Scholar]
  10. Desai S, Chase-Lansdale PL, Michael RT. Mother or market? Effects of maternal employment on the intellectual ability of four-year-old children. Demography. 1989;26:545–561. [PubMed] [Google Scholar]
  11. Downey G, Coyne J. Children of depressed parents: An integrative review. Psychological Bulletin. 1990;108:50–79. doi: 10.1037/0033-2909.108.1.50. [DOI] [PubMed] [Google Scholar]
  12. Duncan GJ. The economic environment of childhood. In: Huston A, editor. Children in poverty: Child development and public policy. New York: Cambridge University Press; 1991. pp. 23–50. [Google Scholar]
  13. Duncan G, Brooks-Gunn J. Consequences of growing up poor. New York: Russell Sage Foundation; 1997. [Google Scholar]
  14. Duncan GJ, Chase-Lansdale PL. Welfare reform and child well-being. In: Blank RB, Haskins RT, editors. The new world of welfare. Washington, DC: Brookings Institution Press; 2001. pp. 391–417. [Google Scholar]
  15. Duncan GJ, Dunifon RE, Doran MBW, Yeung WJ. How different are welfare and working families? And do these differences matter for children’s achievement? In: Duncan GJ, Chase-Lansdale PL, editors. For better and for worse: Welfare reform and the well-being of children and families. New York, NY: Russell Sage Foundation; 2001. pp. 103–131. [Google Scholar]
  16. Duncan GJ, Yeung W, Brooks-Gunn J, Smith JR. Does poverty affect the life chances of children? American Sociological Review. 1998;63:406–423. [Google Scholar]
  17. Dunifon R, Kalil A. Maternal working conditions and child well-being in welfare-leaving families. National Poverty Center Working Paper Series (#05-5) 2005. [DOI] [PubMed] [Google Scholar]
  18. Dunifon R, Kalil A, Danziger SK. Maternal work behavior under welfare reform: How does the transition from welfare to work affect child development? Children and Youth Services Review. 2003;25:55–82. [Google Scholar]
  19. Garrett P, Ng’andu N, Ferron J. Poverty experiences of young children and the quality of their home environments. Child Development. 1994;65:331–345. [PubMed] [Google Scholar]
  20. Gennetian L, Miller C. Children and welfare reform: A view from an experimental welfare program in Minnesota. Child Development. 2002;73:601–620. doi: 10.1111/1467-8624.00426. [DOI] [PubMed] [Google Scholar]
  21. Harris KM. Life after welfare: Women, work, and repeat dependency. American Sociological Review. 1996;61:407–426. [Google Scholar]
  22. Harvey E. Short-term and long-term effects of early parental employment on children of the National Longitudinal survey of Youth. Developmental Psychology. 1999;35:445–459. doi: 10.1037//0012-1649.35.2.445. [DOI] [PubMed] [Google Scholar]
  23. Haverman R, Wolfe B. The determinants of children’s attainments. Journal of Economic Literature. 1995;33:1829–1878. [Google Scholar]
  24. Hepworth DH, Rooney RH, Rooney GD, Strom-Gottfried K, Larsen JA. Direct social work practice: Theory and skills. 7th ed. Belmont CA: Thompson Brooks/Cole; 2006. [Google Scholar]
  25. Hershey AM, Pavetti LA. Turning job finders into job keepers: The challenge of sustaining employment. The Future of Children. 1997;7:407–426. [PubMed] [Google Scholar]
  26. Hofferth SL, Smith J, McLoyd VC, Finkelstein J. Achievement and behavior among children of welfare recipients, welfare leavers, and low-income single mothers. Journal of Social Issues. 2000;5:747–774. [Google Scholar]
  27. Hoffman LW, Youngblade LM. Mothers and work: Effects on children’s well-being. New York: Cambridge University Press; 1999. [Google Scholar]
  28. Huston AC, Duncan GJ, Granger R, Bos J, McLoyd VC, Mistry R, Crosby D, Gibson C, Magnuson K, Romich J, Ventura A. Work-based anti-poverty programs for parents can enhance the school performance and social behavior of children. Child Development. 2001;72:318–336. doi: 10.1111/1467-8624.00281. [DOI] [PubMed] [Google Scholar]
  29. Huston AC, Rosenkrantz Arronson S. Mothers’ time with infant and time in employment as predictors of mother-child relationships and children’s early development. Child Development. 2005;76:467–482. doi: 10.1111/j.1467-8624.2005.00857.x. [DOI] [PubMed] [Google Scholar]
  30. Jackson AP. Black single work mothers’ perceptions of preschool-aged children. Social Work. 1993;38:26–34. [Google Scholar]
  31. Jackson AP. Mothers’ employment and poor and near-poor African-American children’s development: A longitudinal study. Social Service Review. 2003a;77:93–109. [Google Scholar]
  32. Jackson AP. The effects of family and neighborhood characteristics on the behavioral and cognitive development of poor black children: A longitudinal study. American Journal of Community Psychology. 2003b;32:175–186. doi: 10.1023/a:1025615427939. [DOI] [PubMed] [Google Scholar]
  33. Jackson AP, Bentler PM, Franke TM. Employment and parenting among current and former welfare recipients. Journal of Social Service Research. 2006;33:13–26. [Google Scholar]
  34. Jackson AP, Brooks-Gunn J, Huang C, Glassman M. Single mothers in low-wage jobs: Financial strain, parenting and preschoolers’ outcomes. Child Development. 2000;71:1409–1423. doi: 10.1111/1467-8624.00236. [DOI] [PubMed] [Google Scholar]
  35. Jackson AP, Tienda M, Huang C. Capabilities and employability of unwed mothers. Children and Youth Services Review. 2001;23:327–352. [Google Scholar]
  36. Jackson AP, Scheines R. Single mothers’ self-efficacy, parenting in the home environment, and children’s development in a two-wave study. Social Work Research. 2005;29:7–20. [Google Scholar]
  37. Kalil A, Dunifon RE, Danziger SK. Does maternal employment mandated by welfare reform affect parenting behavior? In: Duncan GJ, Chase-Lansdale L, editors. For better or worse: Welfare reform and the wellbeing of children and families. New York: Russell Sage Foundation; 2001. pp. 154–178. [Google Scholar]
  38. Mardia KV. Applications of some measures of multivariate skewness and kurtosis in testing normality and robustness studies. Sankhya B. 1974:115–128. [Google Scholar]
  39. McLoyd VC. The impact of economic hardship on Black families and children: Psychological distress, parenting, and socioemotional development. Child Development. 1990;61:311–346. doi: 10.1111/j.1467-8624.1990.tb02781.x. [DOI] [PubMed] [Google Scholar]
  40. McLoyd VC. Socioeconomic disadvantage and child development. American Psychologist. 1998;53:185–204. doi: 10.1037//0003-066x.53.2.185. [DOI] [PubMed] [Google Scholar]
  41. McLoyd VC, Jayarantne TE, Ceballo R, Borquez J. Unemployment and work interruption among African American single mothers: Effects on parenting and adolescent socioemotional functioning. Child Development. 1994;65:562–589. [PubMed] [Google Scholar]
  42. Menaghan E, Parcel T. Social sources of change in children’s home environments: The effects of parental occupational experiences and family conditions. Journal of Marriage and the Family. 1995;57:699–684. [Google Scholar]
  43. Morris P, Huston A, Duncan G, Cosby D, Bos H. How welfare and work policies affect children: A systhesis of research. New York: Manpower Demonstration Research Corporation; 2001. [Google Scholar]
  44. Moss P, Tilly C. Stories employers tell: Race, skill, and hiring in America. New York: Russell Sage Foundation; 2001. [Google Scholar]
  45. Radloff L. The CES-D scale: A self-report depression scale for research in the general population. Journal of Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  46. Raver CC. Does work pay psychologically as well as economically? The role of employment in predicting depressive symptoms and parenting among low-income families. Child Development. 2003;74:1720–1736. doi: 10.1046/j.1467-8624.2003.00634.x. [DOI] [PubMed] [Google Scholar]
  47. Satorra A, Bentler PM. Corrections to test statistics and standard errors in covariance structure analysis. In: von Eye A, Clogg CC, editors. Latent variables analysis: Applications for developmental research. Thousand Oaks, CA: Sage; 1994. pp. 399–419. [Google Scholar]
  48. Smith JR, Brooks-Gunn J, Kohen D, McCarton C. Transitions on and off AFDC: Implications for parenting and children's cognitive development. Child Development. 2001;72:1512–1533. doi: 10.1111/1467-8624.00363. [DOI] [PubMed] [Google Scholar]
  49. Smith J, Brooks-Gunn J, Klebanov P, Lee K. Welfare and work: Complementary strategies for low-income women? Journal of Marriage and the Family. 2000;62:808–822. [Google Scholar]
  50. Tilly C. The good, the bad, and the ugly: Good and bad jobs in the Unites States at the millennium (Russell Sage Foundation: June 1996 [ http://epn.org/sage/rstil-jo.html]) 1996
  51. Weissman MM, Pilowsky DJ, Wickramaratne PJ, Ralati A, Sisniewski SR, Fava M, Hughes ClW, Garber J, Malloy E, King CA, Cerda G, Sood AB, Alpert JE, Trivedi MH, Rush AJ. Remissions in maternal depression and child psychopathology. JAMA. 2006;295:1389–1398. doi: 10.1001/jama.295.12.1389. [DOI] [PubMed] [Google Scholar]
  52. Wilson WJ. The truly disadvantaged: The inner city, the underclass, and public policy. Chicago: University of Chicago Press; 1987. [Google Scholar]
  53. Wilson WJ. When work disappears: The world of the new urban poor. New York: Knopf; 1997. [Google Scholar]
  54. Wolfe B, Hill S. The effect of health on the work effort of single mothers. Journal of Human Resources. 1995;30:42–62. [Google Scholar]
  55. Zaslow MH, Pederson FA, Suwalsky JTD, Cain RL, Fivel M. The early resumption of employment by mothers: Implications for parent-infant interaction. Journal of Applied Developmental Psychology. 1985;6:1–16. [Google Scholar]

RESOURCES