Economic Adversity and Depressive Symptoms in Mothers: Do Marital Status and Perceived Social Support Matter?
- First Online:
- Cite this article as:
- Kingston, S. Am J Community Psychol (2013) 52: 359. doi:10.1007/s10464-013-9601-7
- 404 Views
Over the past decade there has been increasing interest in the idea that marriage and perhaps other forms of interpersonal support can buffer the negative effects of poverty. The current study tests the hypothesis that marital status, perceived social support and neighborhood collective efficacy can moderate the effects of economic adversity on depressive symptoms among parents. Hierarchical Linear Modeling was used to analyze data from the Project on Human Development in Chicago Neighborhoods. Participants were 1,957 mothers of minor children. Analysis of main effects revealed associations between neighborhood SES (β = −0.69, SE (0.15), p < .001), family income (β = −0.11, SE (0.05), p = .02) financial strain (β = 0.51, SE (0.18), p = .004), being single (β = 0.63, SE (0.24), p = .009) and perceived social support (β = −0.22, SE (0.03), p < .001) on depressive symptoms. The hypothesis that interpersonal resources can buffer the effects of economic adversity was not supported. There were no significant interactions between marital status and economic adversity. There was a significant interaction between perceived social support and neighborhood level socioeconomic status (β = −0.07, SE (0.03), p = .04) but the effects of social support were weakest in neighborhoods characterized by low socioeconomic status.
KeywordsDepressionParentsSocial supportMarital statusPovertyNeighborhood collective efficacy
Parents of minor children report greater numbers of depressive symptoms than adults without children (Evenson and Simon 2005). The burdens associated with childcare and lack of cultural support for parents have been theorized to account for this finding (McLanahan and Adams 1987) in fact, high levels of self-reported housework and childcare burden are associated with increased depressive symptoms among parents (Cunningham and Knoester 2007). Parents facing economic adversity are particularly at risk for depressive symptoms (Goosby 2007). While associations between economic adversity and depressive symptoms in parents have been established less is known about factors that may moderate these associations. Understanding mechanisms that moderate risk for depression has important implications for promoting well-being among parents at high-risk of depression due to limited financial resources.
Economic adversity is associated with chronic exposure to stressors such as inadequate housing and monetary shortfalls, as well as, increased exposure to discrete major life stressors increasing individuals’ vulnerability to psychological distress (McLoyd and Wilson 1990). Economic adversity, as measured by household income and poverty level has been linked to increased depressive symptoms (Goosby 2007). Another measure of economic adversity, economic strain, commonly defined as the discrepancy between a person’s income and their expenses has been strongly implicated as a predictor of depressive symptoms and other negative mental health outcomes (Zimmerman and Katon 2005).
In addition to the stress associated with family level economic adversity, recent research has uncovered a link between neighborhood level poverty and increased symptoms of depression and psychological strain. After controlling for individual level risk factors, living in a neighborhood characterized by high levels of economic adversity was associated with higher levels of depressive symptoms, psychological distress and a higher risk for the onset of depression following negative life events (Schulz et al. 2006). The relationship between neighborhood-level poverty and depressive symptoms suggests that the risks of economic adversity are not simply associated with resource deprivation at the individual and family level but operate within larger social contexts.
In fact, research suggests that positive social environments decrease an individual’s risk of experiencing depressive symptoms. Social support, especially emotional support is associated with decreased depressive symptoms (Nærde et al. 2000). Receipt of social support from friends and family has a buffering effect on depressive symptoms among individuals experiencing economic adversity (Lincoln et al. 2005). Emotional, practical and instrumental aid can buffer the effects of a stressor in a variety of ways. A parent’s support network can provide a solution to problems created by economic adversity, facilitate effective responses from a parent, bolster a parent’s confidence in their ability to handle problems and reduce the perception that a particular problem will result in serious harm (Cohen and Wills 1985).
It has been hypothesized that social support provided in the context of a marital relationship may be particularly beneficial (Jackson 1992). Married parents report fewer depressive symptoms than single parents (Evenson and Simon 2005). Married parents experience less financial stress, increased perceived social support and lower housework and child care burden and consequently report increased psychological well-being relative to single parents (Cunningham and Knoester 2007).
Beneficial effects have also been found for neighborhood level social resources. Neighborhood collective efficacy, a combined measure of perceived social cohesion and informal social control (Sampson et al. 1997) has been shown to have a direct relationship with decreased rates of depression among urban Latino immigrants (Vega et al. 2011), and men (Vaeth et al. 2010) and an indirect relationship with adolescent internalizing problems as mediated through adolescent self-efficacy (Dupéré et al. 2012).
The objective of the current investigation is to examine the potential buffering effects of interpersonal resources: (1) perceived emotional support from family and friends, (2) marital status and (3) neighborhood collective efficacy on the relationship between economic adversity and depressive symptoms in a diverse group of parents living in large urban city.
This investigation builds on previous research in a number of ways. First, parents who are married, cohabitating and single will be separated into distinct groups when analyzing the effects of marital status. Much of the research on the effects of marital status combines cohabitating parents into the single or married group (e.g. Cairney et al. 2003). Studies that distinguish between married, single and cohabitating parents indicate that cohabitating parents may be a distinct group. For example, cohabitating mothers have been found to have incomes almost one-third higher than single mothers (Hilton and Kopera-Frye 2007) and cohabitating custodial parents of minor children report more depressive symptoms than married parents of minor children (Brown 2000).
Second, economic adversity will be measured using three distinct methods (household income, parent reports of financial strain, and a measure of neighborhood-level poverty) in order to gain insight into how various aspects of economic adversity affect psychological well-being among urban parents. In a review of methods used to define and measure economic adversity, Roosa and colleagues note that numerous measures of economic adversity have been employed and researchers rarely discuss the issues inherent in their choice of method, making comparisons across studies difficult (Roosa et al. 2005).
Household income, perhaps the most commonly used measure of economic resources, provides a continuous, objective measure of economic resources and enables researchers to measure variation among families that would be lost by using stratified measures such as percent above or below the poverty level (Roosa et al. 2005). However, household income may fail to capture aspects of economic adversity related to differential costs of living such as housing or child care across different geographic areas or family situations. For example, a family of four composed of one adult wage earner, one adult homemaker and two minor children may face significantly less financial strain than a family of four consisting of one adult wage earner with childcare expenses for three minor children.
Financial strain scales are generally self-report measures that ask about the ability of an individual or a family to secure basic material goods and services and to meet financial obligations. One criticism of such measures are that they are not objective and may be overly sensitive to discrete economic setbacks or poor money management practices among families that have incomes that are considerably above the poverty line (Roosa et al. 2005). Despite these criticisms, these measures have been described as a way to understand real experiences surrounding economic adversity and give meaning to living with economic constraints (Conger et al. 1999).
Measures of neighborhood-level or collective poverty are used to understand the ecological experience of poverty. Many poor urban families live in neighborhoods with high concentrations of other poor families. Such neighborhoods are often characterized by high crime rates, dilapidated physical conditions, lack of amenities and in some cases, compromised social support networks (Coulton et al. 1995). The stresses and constraints conferred by living in such conditions have the potential to have deleterious effects on parental well-being above and beyond the negative effects conferred by family-level economic adversity. Neighborhood poverty has been associated with increased rates of depression and depressive symptoms (Galea et al. 2007). Measuring the linkages between family and neighborhood-level economic adversity, social support and depressive symptoms will provide a much more nuanced understanding of how individuals interact with their social contexts and the implications of those interactions on emotional well-being.
Secondary data analysis of the longitudinal cohort study of the Project on Human Development in Chicago Neighborhoods (PHDCN) was performed. The PHDCN sampled participants from 7 age groups referred to as cohorts in 80 neighborhoods in Chicago and collected data on participants and their primary caregivers across 3 time points. A full description of the PHDCN study can be found at http://www.icpsr.umich.edu/icpsrweb/PHDCN/about.jsp. Data provided from primary caregiver interviews from the 3rd and final wave of this longitudinal study were used. Mothers of children in 5 of the 7 cohorts: 00, 03, 06, 09 and 12 were included. The oldest two cohorts were excluded because the members of these cohorts were young adults at the final wave of data collection. In order to include data on the socio-economic conditions of mothers’ neighborhoods in the analysis, only mothers who still lived in one of the original 80 Chicago neighborhoods were included in the analysis.
Participants were 1,957 mothers of children enrolled in the Project on Human Development in Chicago Neighborhoods. Thirty-eight percent of mothers reported that they were Latina with 67 % of Latinas reporting that they were Mexican or Mexican–American and 20 % of Latinas reported that they were Puerto Rican. Forty percent of mothers reported their race as African–American, 20 % as European–American, 6 % as Native American, 2 % as Asian or Pacific Islander and 32 % as “Other”. Twenty-three percent reported less than a high school education, 48 % reported completing High School or a GED, 12 % had Associates Degrees, 8 % had Bachelor’s Degrees and 4 % had completed Masters or Doctoral Degrees. Sixty-six percent of mothers reported that they were employed full or part time, 24 % reported that they were homemakers, 2 % were students, 1 % were unable to work and 5 % were unemployed.
Ninety-nine percent of mothers reported that they were the target child’s birth mother and 1 % were adoptive or foster mothers. The target children ranged from 3 to 19 years of age at the time of data collection with a mean age of 10, SD = 4.
Mothers were asked to report their marital status according to the following categories: (1) single, (2) separated, (3) divorced (4) widowed, (5) married or (6) living with a partner. For the purpose of the current investigation marital status was recoded into three categories: (1) married, (2) cohabitating (living with a partner) and (3) single (single, divorced, separated, widowed). Sixty-one percent of mothers were married, 33 % were coded as single, 7 % were coded as cohabitating and 1 % did not provide information on marital status.
Annual household income from all sources was recorded on an 11 point scale ranging from less than $5,000 to more than $90,000. Twenty-eight percent reported annual household incomes below $20,000, 43 % reported annual incomes between $20,000 and $49,999, 21 % reported incomes between $50,000 and $89,999 and 8 % reported incomes of $90,000 or above.
Financial strain was measured using a six-item scale which asked mothers to answer yes or no to questions asking if they had been unable to afford to necessities such as food, utilities, rent or medical care within the last 6 months. Items were scored 0 if participants answered no and 1 if they answered yes. Total scores ranged from 0 to 6, with a mean score of 0.38, SD = 0.85.
Neighborhood Socioeconomic Status
Neighborhoods in the PHDCN study were coded as having low, medium or high socioeconomic status based on census measures by the primary investigators of the PHDCN study. Thirty-seven percent of neighborhoods were coded as low, 38 % as medium and 26 % as high.
Perceived Social Support
Social support was measured using the Provision of Social Relations Scale (Turner et al. 1983), a 13 item scale measuring perceived social support from family and friends. Mothers rated statements such as “People in my family help me find solutions to my problems” and “I feel very close to some of my friends” as “very true”, “somewhat true” or “not true”. Total scores ranged from 16 to 39 with a mean of 35.03 SD = 3.96, higher scores indicated higher levels of perceived social support.
Collective efficacy was measured using a 9 item scale that measured mothers’ ratings of neighborhood cohesion, (“People around here are willing to help their neighbors”) and ratings of the likelihood that their neighbors would exert informal social control to address problems in the neighborhood, (“If some children were spray-painting graffiti on a local building, how likely is it that your neighbors would do something about it?”) on a five point likert scale (Sampson et al. 1997). Total scores for collective efficacy ranged from 9 to 45 with a mean of 32.12, SD = 7.30, higher scores indicated higher levels of perceived collective efficacy.
Depressive symptoms were measured using the short form of the Center for Epidemiologic Studies-Depression Scale (Carpenter et al. 1998). The CES-D short form is an 11 item scale that asks respondents to report how often they experienced symptoms of depression during the past week on a 3 point scale (“hardly ever or never”, “some of the time”, “much or most of the time”). Examples of items included on the CES-D are “I did not feel like eating; my appetite was poor” and “I felt sad”. Total scores on the CES-D ranged from 11 to 33 with a mean of 14.55, SD = 3.73.
Hierarchical Linear Models of main effects and interactions
95 % confidence interval
95 % confidence interval
Perceived social support
Level 1 interactions
Family income × perceived social support
Family income × single
Family income × cohabitating
Financial strain × perceived social support
Financial strain × single
Financial strain × cohabitating
Level 2 interactions
Neighborhood SES × perceived social support
Neighborhood SES × single
Neighborhood SES × cohabitating
A second model that tested interactions between the measures of economic adversity and the measures of interpersonal resources was run (Table 1 Model 2). In order to simplify the model only predictors with significant main effects were included in Model 2.
The main effects of financial adversity, interpersonal resources and caregiver ratings of neighborhood collective efficacy are summarized in Table 1, Model 1. As predicted all three measures of economic adversity were associated with depressive symptoms among caregivers such that lower household income, higher financial strain and lower neighborhood SES were associated with higher numbers of depressive symptoms. Marital status was significantly related to depressive symptoms. Single parents reported higher numbers of depressive symptoms relative to married parents as predicted. Cohabitating was not significantly related to depressive symptoms for this sample of parents. Perceived social support was negatively associated with depressive symptoms. Mothers’ ratings of neighborhood collective efficacy were not associated with depressive symptoms.
Interactions Between Economic Adversity and Interpersonal and Environmental Resources
All three measure of economic adversity were related to depressive symptoms in the expected directions. Mothers of minor children who reported lower household incomes, higher levels of financial strain and who lived in neighborhoods characterized by low SES reported more depressive symptoms relative to more affluent mothers. These results are consistent with prior literature linking economic adversity measured by a variety of methods to increased depressive symptoms among parents (Dearing et al. 2004).
The inclusion of three distinct measures allows us to more fully understand the relationship between economic adversity and depressive symptoms among mothers. Financial strain measures, which focus on the inability to afford necessities, are particularly appropriate for research on families given that the amount of income needed to meet basic needs can differ substantially based on family constellation, local cost of living and family needs for childcare, transportation and healthcare. In addition, they measure the inability of parents to successfully provide needed resources to their children, experiences that are likely to be particularly stressful to mothers. Likewise, neighborhood SES is also likely to be associated with conditions that may be acutely stressful for mothers such as high levels of crime and disorder and limited or poor quality recreation, school and childcare options.
The results of the current study did not support the hypothesis that interpersonal resources can attenuate the negative effects of economic adversity on depressive symptoms. Being married was associated with fewer depressive symptoms than being single but marital status did not moderate the relationship between any of the measures of economic adversity and depressive symptoms. These results are consistent with prior research that found that being married lowered mothers’ risk of depression but failed to moderate the relationship between low family income and depression (Cairney et al. 2003).
Perhaps the relationship between economic adversity and marital status is best considered in light of the deleterious effects of economic adversity on martial quality. Recent research on marital satisfaction among African Americans suggests that high levels of financial strain decrease marital satisfaction (Lincoln and Chae 2010) and that marital satisfaction offered less protection against psychological distress among couples facing high levels of financial strain (Cutrona et al. 2003). These results call into question the rationale of anti-poverty programs such as marriage support programs, designed to address the personal lives of poor families rather than policies that directly address economic conditions (McLanahan et al. 2010).
Social support provided by friends and family was associated with fewer depressive symptoms but did not moderate the association between family income and financial strain. There was a significant interaction between social support and neighborhood SES but not in the predicted direction. Perceived social support decreased depressive symptoms for residents living in all neighborhoods but the effects of social support were weakest in neighborhoods characterized by low socioeconomic status. This finding is consistent with research that found that social integration was related to depressive symptoms only in neighborhoods characterized by high SES (Elliot 2000). Once again, the results suggest that directly addressing economic conditions may be more effective than efforts to change parents’ interpersonal context.
Neighborhood collective efficacy was not related to fewer depressive symptoms. This finding contradicts a prospective study which found that higher levels of social cohesion, one component of neighborhood collective efficacy, did predict fewer depressive symptoms among middle aged and older adults (Mair et al. 2009). The findings are consistent with a study that measured the effects of social cohesion and social control, both of the components included in neighborhood collective efficacy, on psychological distress. Neither social cohesion nor social control was associated with psychological distress after controlling for ratings of neighborhood problems (Steptoe and Feldman 2001). Perhaps other aspects of neighborhood climate such as social capital and sense of community, are more important to psychological well-being among parents (Mulvaney and Kendrick 2005).
The results of the current investigation support the idea that cohabitating mothers should be viewed as a distinct group from single or married mothers. Single mothers reported significantly more depressive symptoms than married mothers consistent with numerous prior studies (e.g. Cunningham and Knoester 2007). Cohabitation was positively related to depressive symptoms but this relationship did not reach significance perhaps because the sample contained relatively few cohabitating mothers (7 %) and there was not enough power to detect an effect. More research is needed on the effects of cohabitation on parental well-being given increases in the number of parents who report that they are not married but living with a partner (Manning and Lichter 1996).
The study had a number of limitations. The study used cross-sectional data. It is possible that the relationships between economic adversity and interpersonal resources would change across time, for example, neighborhood collective efficacy has been shown to predict fewer depressive symptoms in prospective studies (Mair et al. 2009). In addition, the analysis of the relationship between marital status and depressive symptoms was hampered by the relatively low numbers of cohabitating mothers and the lack of measures of relationship quality. Future research should investigate the relationships between intimate partner relationship quality and parental well-being, marital status, and SES. It is quite likely that there are interactions between relationship quality and these factors that will have implications for intervention and policy related to low-income urban families.
Future studies should also investigate possible differences in responses to economic adversity and interpersonal resources among different ethnic and cultural groups. Research has revealed differences between African Americans, European Americans and Latinos on both risk of exposure to stressful and traumatic events and responses to these events (Davidson et al. 2013; Hatch and Dohrenwend 2007; Turner and Avison 2003). It is possible that the model tested in the current study would have produced different results with ethnically homogeneous samples. The sample utilized in the current study was large enough to test the main effects and interactions of economic adversity and interpersonal resources on depressive symptoms with 1,957 individuals nested within 80 neighborhoods and 84 % of neighborhoods containing a minimum of 10 individuals (Raudenbush and Liu 2000; Snijders and Bosker 1993). Unfortunately, the sample was not large enough to run the model separately for African Americans, European Americans and Latinas, the three largest racial/ethnic groups in the sample. Crucially, splitting the sample by race/ethnicity would result in racially homogeneous samples with 56–88 % of neighborhoods with less than 10 individuals.
The results of the current study suggest that interpersonal resources have limited ability to buffer the effects of economic adversity on depressive symptoms among mothers. Risks related to economic adversity are not likely to be mitigated by efforts to bolster interpersonal support such as marriage support programs targeted to low-income parents. Interventions that directly target economic conditions at the family or neighborhood level may be more likely to have positive effects on maternal well-being. The results also call attention to the negative impact of economic adversity on mothers. Research on the impacts of conditions in low-income urban neighborhoods often focuses on children and adolescents but community psychologists interested in cultivating healthy urban environments must not forget the needs the adults raising children in these communities.