If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Although studies suggest that employment promotes mental health, it is unclear whether this pattern extends to low-income urban women with children who are disproportionately employed in unstable jobs and often unable to obtain child care. In this paper, we consider whether becoming employed reduces symptoms of psychological distress among low-income women with children. We also assess whether having trouble securing adequate child care offsets these benefits.
Study Design
We use longitudinal data from the Welfare, Children, and Families project, a probability sample of low-income women with children living in Boston, Chicago, and San Antonio, to test whether becoming employed reduces symptoms of psychological distress over time and whether having trouble securing child care moderates this association.
Results
We find that employment is associated with lower levels of distress among women who have no trouble with child care and higher levels of distress among women who struggle with child care.
Conclusion
Taken together, our results suggest that valuing the benefits of paid work over unpaid work is an oversimplification and that the emphasis on placing poor women with children into paid work could be misguided. Policies that focus on moving low-income women off of government assistance and into paid work could be more effective if greater resources were devoted to increasing access to quality child care.
For more than four decades, numerous studies have shown that being employed or becoming employed is associated with better mental health outcomes, including lower levels of nonspecific psychological distress, depression, anxiety, and somatization (
in: Horwitz A. Scheid T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York1999: 284-294
in: Scheid T. Brown T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York2010: 213-225
). According to previous research, employment supports mental health by reducing chronic stress and by promoting financial security, caring coworker relationships, self-esteem, and a sense of personal control over one's life (
in: Horwitz A. Scheid T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York1999: 284-294
in: Scheid T. Brown T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York2010: 213-225
Although employment has been shown to benefit the mental health of women and men, there is some evidence to suggest that employment is related less consistently to the mental health of women with children. Some researchers suggest that the association between employment and mental health is attenuated for women with children or by the mere presence of young children in the household (
). Others suggest that the association between employment and mental health is not attenuated by the presence of young children per se. The more precise and compelling argument is that the mental health benefits of employment are offset or even eliminated by the stress of work–family role conflict from inadequate child care or an unequal distribution of child care responsibilities within the household (
In this paper, we consider whether becoming employed reduces symptoms of psychological distress among low-income women with children. We also assess whether having trouble securing adequate child care offsets these benefits. To this point, previous studies of employment, child care, and mental health have employed cross-sectional data collected from general probability samples of the U.S. population (
). We extend this seminal work by using longitudinal data collected from a probability sample of low-income women with children living in Boston, Chicago, and San Antonio. Methodologically, our longitudinal design is important for addressing the possibility of health selection into or out of employment (
). Theoretically, our focus on low-income urban women with children is important because this population is especially vulnerable to work–family conflict and inadequate child care (
). When low-income women receive welfare benefits, they are legally required by the 1996 Personal Responsibility Work Opportunity Reconciliation Act (PRWORA) to be gainfully employed as a precondition for financial assistance. The cultural problem is an ideology of “wage-earning over child-rearing” for poor women (
, p. 100). The structural problem is that low-income women tend to live in resource-poor neighborhoods, are disproportionately employed in precarious jobs, and often have trouble securing child care (
). The question to be addressed empirically is whether conflicting child care arrangements condition the mental health benefits of becoming employed for low-income urban women.
Background
Our central argument is that the mental health benefits of being employed or becoming employed should be attenuated for those low-income women who also have trouble securing child care. Following
, p. 129), we expect that “the meaning of employment to emotional well-being may depend on children and on child care.” This expectation primarily rests on the stress process model and previous studies of work–family conflict.
The Stress Process Model
The stress process model provides a general framework for understanding how and under which conditions socially patterned stressors might undermine mental health (
, p. 245). Employment is a role that is never played in isolation from other roles. The problem is that “individuals cannot satisfy the demands and expectations of one of these roles without forsaking those of the other” (
:245). When women with children transition into the worker role, they are often forced to make compromises between roles that unemployed women with children do not have to make. Perhaps the most common compromise for employed women is between work and child care arrangements.
, p. 136) explain that “it is the lack of readily available child care and the lack of shared responsibility for children within the couple that place stress on the mothers and on their families.”
Work–Family Conflict
The many compromises that employed women make can be psychologically distressing. In fact, numerous studies show that work-to-family conflict and family-to-work conflict are associated with poorer mental health outcomes, including higher levels of nonspecific psychological distress, depression, anxiety, guilt, anger, somatization, and substance abuse (
). According to previous research, the competing demands of work–family conflict undermine mental health by creating social conditions that are conducive to negative self-evaluations (“I am no good at parenting”), stress (“I can't do this”), and dissatisfaction (“This isn't what I want”) with work and family life (
Although work and family roles may conflict in many ways, problems associated with inadequate child care and the unequal distribution of child care responsibilities within the household may be especially detrimental to the mental health of employed women (
). Ross and Mirowsky (1988, p. 129) explain that, “Many women are faced with conflicts between their roles of mother and of employee; the institutional support necessary to fulfill both roles often does not exist. Possibly the major lack of institutional support is found in child care.”
Our review of the empirical literature resulted in two studies focused specifically on employment, child care, and mental health (
show that employment is associated with lower levels of anxiety (but not lower levels of depression) among women whose husbands share in child care. Referring to women whose husbands do not share in child care,
find increased levels of depression among employed wives with young children who have 1) difficulty arranging child care and 2) husbands who do not share in child care. Interestingly, these authors also report that difficulty arranging child care and sharing in child care are unrelated to depression among husbands of employed wives with young children.
, p. 135) conclude that,For employed wives with young children, difficulty in arranging child care and husbands' participation in child care have major effects on depression. Child care may be the overwhelming concern for employed mothers; it is not children per se who create stress for employed mothers, but the absence of supportive arrangements.
The Case of Low-income Women
In this paper, we focus specifically on low-income women with children because this population may be especially vulnerable to work–family conflict and inadequate child care (
). One reason is the unique cultural pressure for low-income women with children to be gainfully employed. Whereas upper middle-class women with children may be expected to stay at home and to engage in intensive mothering (
‘Some of us are excellent at babies’: Paid work, mothering, and the construction of need in a welfare-to-work program.
in: Piven F. Acker J. Hallock M. Morgen S. Work, welfare and politics: Confronting poverty in the wake of welfare reform. University of Oregon Press,
Eugene2002: 81-94
in: Avison W. Aneshensel C. Schieman S. Wheaton B. Advances in the conceptualization of the stress process: Essays in honor of Leonard I. Pearlin. Springer,
New York2010: 131-145
‘Some of us are excellent at babies’: Paid work, mothering, and the construction of need in a welfare-to-work program.
in: Piven F. Acker J. Hallock M. Morgen S. Work, welfare and politics: Confronting poverty in the wake of welfare reform. University of Oregon Press,
Eugene2002: 81-94
, p. 82) points out that in worker-focused welfare rhetoric, “unemployed low-income women are characterized as neglectful or even harmful to their children, as failing to instill a work ethic in them, and as creating the next generation of the underclass through this failure.” Indeed, as politicians debated over welfare reform in the 1980s and 1990s, child-rearing and welfare reliance were framed as detrimental to women and their children, and paid work was framed as a pathway to enhanced well-being for women and their families. President
believed that paid work would “encourage self-respect, build strength of character, and develop to the fullest each individual's potential for a productive, meaningful life.” On the anniversary of the PRWORA,
Clinton, W. J. (1998). Remarks by the President on Welfare Reform, August 4, 1998. Online by G. Peters and J. Woolley, The American Presidency Project. Available: http://www.presidency.ucsb.edu/ws/index.php?pid=54734. Accessed: November 14, 2013.
offered the following: “Work is more than a punch card, more than a paycheck. It provides structure to a day, link to a society, dignity for a family. It can build self-confidence and self-esteem.”
All of these sentiments and legal requirements are of course built on the “assumption that paid employment of any kind is preferable to not having paid work” (Kalleberg, 2011, p. 207). In reality, low-income women with children are typically obligated to work in low-wage jobs that fail to provide stable hours, benefits, affordable child care, or any opportunities for advancement (
). In this way, the cultural constraints and work requirements help to create work–family conflict in the lives of low-income women with children. The idea is that welfare reform's codification of the wage earner role nearly ensures that low-income women with children will be forced to experience the conflicting expectations of worker and caregiver (
Although all women with children are likely to experience some degree of work–family conflict or difficulty in securing child care, low-income women are unique in that their structural positions often create conflicts and tend to limit the resources needed to cope with the strain of competing roles (
in: Avison W. Aneshensel C. Schieman S. Wheaton B. Advances in the conceptualization of the stress process: Essays in honor of Leonard I. Pearlin. Springer,
New York2010: 131-145
). The fundamental problem is that low-income families are frequently unable to afford basic necessities like food, clothing, and child care. Child care is especially problematic for low-income women because it is often too expensive, unreliable, or of low quality (
). There is also some evidence to suggest that families living under the conditions of economic hardship tend to exhibit higher levels of work–family conflict (
). Another important issue is that low-income families tend to live in resource-poor neighborhoods. The problem is that residents of disadvantaged neighborhoods often have less access to child care facilities and greater exposure to work–family conflict (
). Given all of these structural constraints, low-income women with children may be especially vulnerable to work–family conflict and inadequate child care.
Considering the Stress Universe
Before we assess the psychological costs of work–family conflict among low-income urban women, it is important to consider the unique stress universe of this population. The primary concern is whether psychological distress is the result of inadequate child care per se or the consequence of women living in urban poverty. Because stressful conditions are often rooted in people's locations within systems of social stratification, especially those based on race, gender, and socioeconomic status (
in: Horwitz A. Scheid T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York1999: 131-145
), low-income urban women rarely, if ever, experience stressful circumstances in isolation. Low-income, urban women are regularly confronted with difficulties associated with financial instability, living in poor neighborhood environments and substandard housing, and intimate partner violence (
, p. 112) explains that, “In the elaborated stress universe, no one form or source of stress is sufficient to capture, even remotely, all that is stressful.” To more accurately gauge the unique emotional impacts of work–family conflict and inadequate child care, it is important to consider the broader stress universe by taking into account a range of adversities that are common in the lives of low-income urban women.
Methods
Hypotheses
Drawing from previous studies of employment, work–family conflict, and mental health, we developed the following three hypotheses:
H1: Respondents who acquire employment will report fewer symptoms of psychological distress from baseline to follow-up than continuously unemployed respondents.
H2: Respondents who acquire employment without child care conflict will report fewer symptoms of psychological distress from baseline to follow-up than continuously unemployed respondents.
H3: Respondents who acquire employment with child care conflict and respondents who remain continuously unemployed will report similar changes in symptoms of psychological distress over the study period.
Data
To formally test our hypotheses, we use data from the Welfare, Children, and Families (WCF) project (available: www.jhu.edu/∼welfare/). The WCF project is a household-based, stratified random sample of 2,402 low-income women living in low-income neighborhoods in Boston, Chicago, and San Antonio. The WCF first sampled census blocks (or neighborhoods) with at least 20% of residents below the federal poverty line based on the 1990 census. Within these neighborhoods, households under 200% of the poverty line were sampled, with an oversampling of households below 100% of the poverty line. Because one of the goals of the WCF project is to assess the impact of welfare policy and work on children, households were screened for the presence of children. Households with at least one infant or child (aged 0–4) or young adolescent (aged 10–14) were sampled. The children's caregivers, all women, were interviewed face to face. We refer to the caregivers as “women with children” instead of “mothers” because some caretakers did not identify themselves as the child's parent. The data were collected in 1999 with a follow-up in 2001. The baseline response rate was 75%, and 89% of the original sample was re-interviewed. The overall respondent-level response rate is 75%, with city-specific response rates of 74% (Boston), 71% (Chicago), and 79% (San Antonio).
The WCF data are well-suited to the study of employment and mental health for two key reasons. First, since the 1996 passage of PRWORA, low-income women with children have been legally required to work to receive benefits. These women do not have the resources upper- and middle-class women with children might have to cope with conflicting role expectations, yet they are required to participate in the workforce. Second, there is considerable volatility in the employment status of low-income, urban women. This volatility allows for the assessment of changes in employment status and changes in mental health.
Measures
Mental Health
We use the Brief Symptom Inventory (BSI-18) to assess mental health or symptoms of psychological distress (
). Subsequent multivariate analyses include four specifications of distress: 1) nonspecific psychological distress (the total BSI index) and the subscales for 2) anxiety, 3) depression, and 4) somatization. Anxiety is measured as the mean response to six items (α = 0.84). Respondents were asked to indicate how much in the past 7 days they were distressed or bothered by a) feeling tense or keyed up, b) feeling nervous or shaky inside, c) suddenly feeling scared for no reason, d) feeling so restless they could not sit still, e) spells of terror or panic, and f) feeling fearful. Depression is measured as the mean response to six items (α = 0.84). Respondents were asked to indicate how much in the past 7 days they were distressed or bothered by a) feeling no interest in things, b) feeling lonely, c) feeling blue, d) feelings of worthlessness, e) feeling hopeless about the future, and f) thoughts of ending their life. Somatization, the physiological manifestation of distress, is measured as the mean response to six items (α = 0.80). Respondents were asked to indicate how much in the past 7 days they were distressed or bothered by a) faintness or dizziness, b) pains in their heart or chest, c) nausea or upset stomach, d) trouble getting their breath, e) weakness, and f) numbness or tingling in parts of their body. Finally, total BSI is measured as the mean response to all 18 items (α = 0.92). Response categories for each item ranged from 1 (not at all) to 5 (extremely). These items consistently demonstrate adequate validity, reliability, and principal components in previous research (
Psychological assessment of patients with temporomandibular disorders: Confirmatory analysis of the dimensional structure of the Brief Symptoms Inventory 18.
Journal of Psychosomatic Research.2006; 60: 365-370
We assess employment status with one item. Respondents were asked if they worked for pay in the last week. Responses were dummy coded (0) no and (1) yes. To assess if there was conflict between work and child care arrangements, we use a question that asked, “While you've been working since [last interview], have you had any of the following problems in relation to your job?” One of several possible answers was, “Problems getting child care if you worked.” Responses were dummy coded (1) if the respondent reported problems getting child care and (0) otherwise. “Problems getting child care if you worked” was the most commonly selected problem, and prior research shows that finding quality child care arrangements is a significant barrier for low-income working women (
Our assessment of the stress universe includes measures of neighborhood disorder household disrepair, and intimate partner violence. Neighborhood disorder refers to a range of potentially stressful conditions in the neighborhood and is measured as the mean response to 10 items (α = 0.89). For example, respondents were asked to rate their neighborhood environment in terms of “assaults and muggings,” “abandoned houses,” and “high unemployment.” Response categories for these items are coded not a problem (1), somewhat of a problem (2), and a big problem (3).
Household disrepair refers to potentially harmful housing conditions and is measured as a summed response to eight items. For example, respondents were asked to indicate whether any of the following conditions were present in their households: “a leaky roof, a toilet, hot water, or other plumbing that doesn't work,” “broken windows,” and “rats, mice, cockroaches, or other pests.” Responses to these items are dummy coded no (0) and yes (1).
Data on intimate partner violence in the past year are taken from the Revised Conflict Tactics Scale (
, we assess four major types of violence, including psychological aggression, minor physical assault, severe physical assault, and sexual coercion. Psychological aggression captures nonphysical or verbal acts of violence. Respondents were asked to indicate how often in the past 12 months a romantic partner had threatened to a) hit them, b) use a weapon on them, and c) hurt their child or take (him/her) away. Minor physical assault assesses non-severe or “common couple” acts of physical violence. Respondents were asked to indicate how often in the past 12 months a romantic partner had d) thrown something at them and e) pushed, grabbed, or shoved them. Severe physical assault measures some of the more brutal and vicious acts of physical violence. Respondents were asked to indicate how often in the past 12 months a romantic partner had f) slapped, kicked, bit, or punched them, g) beaten them, and h) choked or burned them. Finally, sexual coercion is intended to capture forced sexual acts of violence. Respondents were asked to indicate i) how often in the past 12 months a romantic partner had forced them into any sexual activity against their will. Original response categories for all past-year violence items were coded as never (0), once or twice (1), several times (2), and often (3). Owing to limited incidence rates, we coded measures of relationship violence into dummy variables. Respondents were assigned a value of (1) if they reported a particular act of violence in the past 12 months and (0) otherwise. To account for substantial comorbidity among the violence items, our final measure of relationship violence is dummy coded (1) for one or more experiences with relationship violence in the past year and (0) for no experiences of violence in the past year.
Background Variables
Subsequent multivariate analyses control for a range of background variables that are known correlates of employment status and psychological distress (e.g.,
). These factors include age (in years), race and ethnicity (four dummy variables capturing non-Hispanic Whites, Mexicans, other Hispanics, and African American—the reference category), education (in years), household income (logged), marital status (1 = currently married/spouse in house, 0 = otherwise), number of children (1–6 or more, top-coded continuous variable), the presence of a child under age 5, and city of residence (three dummy variables capturing Boston residence, Chicago residence, and San Antonio—the reference category).
Analytic Strategy
The analysis begins by presenting descriptive statistics of the study, which includes variable ranges, means, and standard deviations for all variables in the study (Table 1). We then model changes in psychological distress over the 2-year (1999 and 2001) study period (Table 2). In lieu of lagged endogenous dependent variable models, we employ change score models to assess 2-year changes in distress. A comparison of two-wave panel designs concluded that change score models are generally preferable to lagged endogenous dependent variable models (
). We computed change scores by subtracting baseline (1999) distress scores from follow-up (2001) distress scores. Change scores are continuous variables that range from some negative number to some positive number, where negative numbers indicate a decrease in distress from 1999 to 2001, and positive numbers indicate an increase in distress from 1999 to 2001. A change score of zero indicates no change in distress across waves. Because change scores are continuous variables, we use ordinary least squares (OLS) regression to predict changes in distress. The unstandardized OLS coefficients describe the difference in the expected change in psychological distress for every 1-unit change in an independent variable.
N = 1,187. Asterisks indicate significant differences between women who were employed and unemployed during Wave 2 (2001): ∗p < .05, ∗∗p < .01, ∗∗∗p < .001. Sample consists of women who were unemployed at wave 1 (1999).
Our change score analyses proceed in two steps. First, we compare the mental health changes of continuously unemployed respondents (those unemployed at both waves) with those who have acquired employment, while controlling for baseline mental health, stressors, and background variables. To account for the possibility of other changing characteristics, we controlled for changes in household income, marital status, the number of children, the presence of young children, neighborhood disorder, and household disrepair. Next, we compare the mental health changes of continuously unemployed respondents with those who have acquired employment a) with and b) without child care conflict, while still controlling for baseline mental health and background variables.
Missing Data
We addressed the problem of missing data in two ways. We first estimated a binary logistic regression model predicting the log odds of sample attrition (results not shown). The dependent variable in this case is a dichotomous variable coded such that respondents who completed questionnaires for both waves were coded as 0 and those who completed the wave 1 questionnaire but were missing for wave 2 were coded as 1. The independent variables include employment status, psychological distress, and all background variables. Approximately 11% of the original sample (n = 276) was lost to follow-up. The logistic regression results show little evidence of bias owing to sample attrition. Only the coefficients for number of children (odds ratio, 1.34; p < .01) and education (OR, 0.84; p < .01) were significant at conventional levels. These results suggest that the odds of attrition were higher for respondents with more children, and lower for respondents who were more educated. We then used multiple imputation by chained equations to replace missing values on the independent and dependent variables (
). Pooled estimates derived from 10 imputations were substantively identical to our results using listwise deletion. Given that our results using listwise deletion are not substantively biased owing to sample attrition or others forms missing data, all subsequent multivariate findings use original data and listwise deletion.
Results
Descriptive Statistics
Table 1 provides weighted descriptive statistics for all study variables. The first column presents descriptive statistics for the entire sample. The next two columns stratify the sample by employment status. Approximately 62% of respondents remained unemployed over the study period, and nearly 37% acquired employment. Roughly 7% of respondents acquired employment with child care conflict. More than 31% of respondents acquired employment without child care conflict. At baseline, the average respondent reported low levels of psychological distress. In the total sample, the change scores for psychological distress indicate slight improvements in distress symptoms over the study period. When we examine the stratified change scores for distress, we see that these improvements are primarily evident in those respondents who acquired employment. Table 1 also presents t tests to formally assess mean differences across employment strata. These bivariate tests indicate that respondents who acquired employment exhibited greater reductions in distress symptoms (nonspecific psychological distress, anxiety, and depression) than respondents who remained unemployed. Changes in somatization symptoms were comparable for those respondents who acquired employed and those who remained unemployed.
Multivariate Analyses
Table 2, Table 3 present our longitudinal change score analyses. Table 2 examines the gross association between employment status and changes in psychological distress, while Table 3 specifies employment status with and without child care conflict. Table 2 shows that respondents who acquired employment over the study period tended to report greater reductions in distress symptoms than respondents who remained continuously unemployed. We observe this general pattern for nonspecific psychological distress and the subscales for anxiety, depression, and somatization.
Table 3OLS Regression of Psychological Distress on Employment and Child Care
Table 3 indicates that the apparent mental health benefits of acquiring employment varies depending on the presence or absence of child care conflict. When there is no child care conflict, acquiring employment (as opposed to remaining unemployed) is associated with significant reductions in nonspecific psychological distress and the subscales for anxiety, depression, and somatization. However, when child care conflicts are reported, we find that acquiring employment is associated with significant increases in nonspecific psychological distress and the subscales for anxiety, depression, and somatization. In other words, under the conditions of child care conflict, acquiring a job undermines mental health, whereas remaining unemployed promotes mental health. These general patterns persist with adjustments for a range of stressful conditions and background factors.
Figure 1 shows the adjusted mean changes in symptoms of psychological distress for those respondents who remained unemployed, those who acquired employment with child care conflict, and those who acquired employment without child care conflict. The mean changes in distress are adjusted for baseline mental health, stressors, and control variables. The results presented in Table 3 are even more striking when displayed in Figure 1. The bars for those respondents who remained unemployed or who acquired employment with child care conflict tend to extend above zero. Bars extending above zero indicate average increases in distress symptoms over the study period. The bars for those respondents who acquired employment without child care conflict tend to extend below zero. Bars extending below zero indicate average reductions in distress symptoms over the study period.
Figure 1Adjusted mean changes in psychological distress. Note: All means are based on models 1–4 of Table 3. All means are adjusted for baseline mental health, number of children, the presence of a child under age 5, respondent's age, race, ethnicity, education, household income, city of residence, marital status, intimate partner violence, household disrepair, and neighborhood disorder.
Although studies suggest that employment promotes mental health, it is unclear whether this pattern extends to low-income urban women with children who are disproportionately employed in unstable jobs and often unable to obtain child care. Building on previous research, we considered whether acquiring employment is beneficial to the mental health of low-income urban women. We also assessed whether this benefit is contingent upon the presence or absence of child care problems.
Our first hypothesis stated that respondents who acquire employment would report fewer symptoms of psychological distress from baseline to follow-up than continuously unemployed respondents. In support of this hypothesis, our results showed that respondents who acquired employment over the study period tended to report greater reductions in distress symptoms than respondents who remained continuously unemployed. We replicated this general pattern across several outcomes, including nonspecific psychological distress, anxiety, depression, and somatization. These results are consistent with several previous studies of more general populations (e.g.,
). To the best of our knowledge, we are the first to establish a longitudinal association between employment status and mental health among low-income urban women with children.
Our second hypothesis stated that respondents who acquire employment without child care conflict would report fewer symptoms of psychological distress from baseline to follow-up than continuously unemployed respondents. Consistent with this hypothesis, we found that when there is no child care conflict, acquiring employment (as opposed to remaining unemployed) is associated with significant reductions in nonspecific psychological distress, anxiety, depression, and somatization. Our anxiety results support earlier research by
. Once again, to our knowledge, we are the first to establish the effects of employment on nonspecific psychological distress and somatization among low-income women without child care conflicts.
Our third hypothesis stated that respondents who acquire employment with child care conflict and respondents who remain continuously unemployed would report similar changes in symptoms of psychological distress over the study period. Following
, we expected that child care conflicts would merely produce a null association between employment and psychological distress. Contrary to our expectation, we found that when child care conflicts are present, acquiring employment is associated with significant increases in nonspecific psychological distress and the subscales for anxiety, depression, and somatization. In other words, under the conditions of child care conflict, acquiring a job actually undermines mental health. Although we are the first to establish the effects of employment on nonspecific psychological distress, anxiety, and somatization among low-income women with child care conflicts, our results are generally consistent with research showing that difficulty arranging child care is associated with higher levels of depression among employed wives with young children (
Although proponents of workfare would argue that any job is preferable to unemployment, these results suggest that when new jobs create role conflict, they are in fact detrimental to metal health. How can we explain these findings? Gaining employment is a role transition that is not experienced in isolation from other roles. Individuals who transition into the role of worker make compromises they would not have to make if they remained unemployed. For low-income women with children who are legally obligated to be wage earners before caregivers, one of these compromises is finding child care as a precondition to work. The problem is that low-skill jobs are often characterized by low pay, low benefits, low security, and low mobility. Under these employment conditions, child care problems are likely to arise. Eventually, the role conflicts and chronic strains associated with providing stable child care with limited resources contribute to the development of psychological distress. Low-income women with children may be especially vulnerable to the psychological consequences of work–family conflict because these women tend to have fewer coping resources than more affluent women (e.g.,
This paper is limited in several respects. Our most important shortcoming is our observational design. Although we use a longitudinal design and control for baseline mental health and an array of relevant covariates, we cannot establish causality or dismiss the possibility of omitted variable bias. We suggest that employment promotes mental health, whereas child care conflicts undermine mental health. However, preexisting mental health issues or unobserved heterogeneity could at least theoretically contribute to unemployment and family conflicts (e.g.,
). We must also acknowledge the limited external validity of our findings. Because our data are restricted to low-income urban women with children living in Boston, Chicago, and San Antonio, we cannot generalize our results beyond these specific population parameters. Another potential weakness is that the data are fairly old, with the second wave being conducted in 2001. Because our results are generally consistent with other studies using older data sources and more recent data sources, and given that there have been no overwhelming changes to welfare or workfare requirements (since 1996), we have no reason to believe that our findings would be time sensitive. Finally, our assessment of the stress universe is based on a rather narrow range of stressors. The veracity of our key results will be contingent upon replications with more comprehensive adjustments.
Implications for Practice and/or Policy
In this paper, we tested whether gaining employment promotes the mental health of low-income women with children, and whether child care arrangements condition this relationship. We found that although employment with no child care conflict tends to promote mental health, employment with child care conflict tends to undermine mental health. Taken together, our results suggest that valuing the benefits of paid work over unpaid work is an oversimplification and that the emphasis on placing poor women with children into paid work could be misguided. Policies that focus on moving low-income women off of government assistance and into paid work could be more effective if greater resources were devoted to increasing access to quality child care. Reducing role conflict and, by extension, symptoms of psychological distress, could help to negate negative feedback processes like poorer mental health contributing to poorer work and family outcomes. There are numerous avenues for future studies in this area. Our results should be replicated with other data sources, including more recent data sources with broader claims to external validity. It is also important to consider additional indicators of work–family conflict and more direct indicators of stress or strain. Finally, we would like for subsequent research to consider other indicators of mental health, including, for example, substance abuse and chemical dependency.
Acknowledgments
We would like to thank C. André Christie-Mizell, Richard N. Pitt, and our anonymous reviewers for their helpful feedback on earlier versions of this paper.
References
Ali J.
Avison W.R.
Employment transitions and psychological distress: The contrasting experiences of single and married mothers.
Journal of Health and Social Behavior.1997; 38: 345-362
Clinton, W. J. (1998). Remarks by the President on Welfare Reform, August 4, 1998. Online by G. Peters and J. Woolley, The American Presidency Project. Available: http://www.presidency.ucsb.edu/ws/index.php?pid=54734. Accessed: November 14, 2013.
Psychological assessment of patients with temporomandibular disorders: Confirmatory analysis of the dimensional structure of the Brief Symptoms Inventory 18.
Journal of Psychosomatic Research.2006; 60: 365-370
in: Horwitz A. Scheid T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York1999: 284-294
in: Scheid T. Brown T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York2010: 213-225
‘Some of us are excellent at babies’: Paid work, mothering, and the construction of need in a welfare-to-work program.
in: Piven F. Acker J. Hallock M. Morgen S. Work, welfare and politics: Confronting poverty in the wake of welfare reform. University of Oregon Press,
Eugene2002: 81-94
in: Avison W. Aneshensel C. Schieman S. Wheaton B. Advances in the conceptualization of the stress process: Essays in honor of Leonard I. Pearlin. Springer,
New York2010: 131-145
in: Horwitz A. Scheid T. A handbook for the study of mental health: Social contexts, theories, and systems. Cambridge University Press,
New York1999: 131-145
Anna W. Jacobs, MS, MA, is a doctoral student, Department of Sociology, Vanderbilt University. She focuses on work and labor movements—detrimental effect of nonstandard work schedules, impact of heterogeneity within the labor movement, and effect of media and cultural discourse on strike success.
Terrence D. Hill, PhD, is Associate Professor, School of Sociology, University of Arizona. His research examines the social distribution of health and health-relevant behaviors. He is especially interested in the effects of religious involvement, neighborhood context, social relationships, and socioeconomic status.
Daniel Tope, PhD, is an Associate Professor of Sociology at Florida State University. He studies politics, race, and work. His research focuses on the role of racial attitudes in politics as well as the determinants of state variations in social and economic policy.
Laureen K. O'Brien, MA, is a doctoral student, School of Sociology, University of Arizona. She focuses on work and organizations—the expansion of nonstandard employment, emergence of third-party organizational brokers in temporary employment, and impact of large-scale employment brokerage on interorganizational relationships.
30038-X&id=gr1.jpg){kind=link}