Parenting and Depression

Abstract

Few empirical studies have examined the association between parenthood and psychological well-being. Using NLSY79 data (n = 6,297), we examined how various parental roles, or specific parent–child relationship types, were associated with depressive symptoms in adults. We hypothesized that less traditional and more complex parental roles would be associated with higher depressive symptoms. Ordinary least squares regression results revealed that having a stepchild was associated with higher depressive symptoms, regardless of the stepchild’s residential status. Additionally, certain combinations of parental roles were a risk factor for depressive symptoms, including having a biological child residing in the home and another biological child residing outside the home simultaneously, a biological child and a stepchild residing together (with or without a new biological child), and having more than two combined parental roles in general. Findings suggested certain parental roles are indeed associated with higher depressive symptoms, while others may be null relationships.

Keywords

children depression NLSY79 parenting parent–child relationships psychological well-being

Nearly 7% of adults will experience a major depressive episode in a given year, with about 16% having at least one in their lifetime (Kessler et al., 2003). With such prevalence, it is widely recognized depression is associated with negative outcomes in individuals, families, and society. Some of these include lost work days and wages (Kessler et al., 2006), lower educational attainment and income (Coryell, Endicott, & Keller, 1990), medical problems (Evans et al., 2005), and poor parenting practices (Goodman et al., 2011). The understanding and treatment of depression is given considerable emphasis in mental health research, yet little is known about the relationship between parenting and depression.

Although parenting may be a rewarding experience (Nelson, Kushlev, English, Dunn, & Lyubomirsky, 2013), it seems to have little positive effect on adult mental health. Indeed, it is likely that many parents have lower psychological well-being than their childless counterparts (Evenson & Simon, 2005). Although several studies have addressed the relationship between parenthood and depression, the results are far from conclusive (Davies, Avison, & McAlpine, 1997; Evenson & Simon, 2005; Williams & Dunne-Bryant, 2006). Positive, negative, and null relationships between depression and parenting are all common in the literature (Burton, 1998; Glenn & Mclanahan, 1981; Hughes, 1989; Kandel, Davies, & Raveis, 1985; Umberson & Gove, 1989; Williams & Dunne-Bryant, 2006). The divergent conclusions of so many studies may be an artifact of methodological choices, such as inconsistent comparison groups, various time frames, different study designs, and different age ranges at observation. Ultimately, little is known about how different parental roles are associated with psychological well-being, and few studies have addressed this topic in depth.

In our study, we address the lack of research findings about the association of parenthood and psychological well-being among adults. In particular, we address how various parent–child relationships are associated with depressive symptoms in adults. We use data from the National Longitudinal Survey of Youth, 1979 cohort (NLSY79), a nationally representative longitudinal study of adult participants to address this question. In the subsequent sections we describe relevant research and theory concerning the link between parenting and depression and how various parental roles might be associated with depression.

Parental Roles and Psychological Well-Being

Both micro- and macro-level explanations have been forwarded to explain the link between parenthood and psychological well-being. Micro-level work tends to emphasize what may be thought of as a “cost–benefit analysis” of parenting (e.g., Menaghan, 1989). In other words, parents are more likely to experience high depressive symptoms when the rewards of parenting, such as life-purpose or a sense of accomplishment, are outweighed by the more troubling aspects of parenthood, such as stress, mental and physical demands, or reduced personal time. Macro-level approaches tend to focus on the social rewards associated with parenting and the presence or lack of formal and informal parental supports (Hewlett, Rankin, & West, 2002). This explanation suggests that in the past motherhood and fatherhood were associated with increased social support, social capital, and institutional benefits that served to reward parenting (Huber, 1980; Preston, 1986). However, as childlessness has become increasingly normative and parenthood viewed as one of many life choices, parental supports and benefits have become less common (Hewlett et al., 2002; Thornton & Young-DeMarco, 2004). The lack of pro-parenting assistance can increase the burdens, demands, and stresses associated with parenthood—in turn, increasing depressive symptoms.

Whether parental psychological well-being is more strongly associated with macro-level phenomena, micro-level factors, or is some combination of both, there may be variability in how the internal and external pressures associated with parenting affect individuals (Evenson & Simon, 2005). Specifically, we expect that the relationship between depressive symptoms and parenting is dependent on one’s parental role or roles. The parental role perspective (Scott & Alwin, 1989) suggests that the negative psychological effects of parenting are directly correlated with the childrearing expectations of a given role. In other words, because the demands and stresses of parenting depend on parenting type (e.g., biological parent, stepparent, nonresidential parent), so too will any association between parenthood and psychological well-being. Although the majority of studies have addressed how the different parental roles and childrearing expectations of mothers and fathers can lead to gender differences in mental health outcomes, the parental role perspective can also provide context to why different parental roles (not necessarily associated with gender) can affect psychological well-being. For example, stepparents may experience fewer depressive symptoms than biological parents because the expectations of residential stepparents are significantly different than biological parents or more because of stepparent–stepchild relationships are unique from biological relationships (Stewart, 2007). Similarly, Evenson and Simon (2005) focused on the importance of different parental roles by studying disparate parental psychological well-being by gender, marital status, and the residential status of children. In line with this extension of the parental role perspective, we address how dissimilar childrearing responsibilities and combinations of roles within diverse families can contribute to variability in depressive symptoms.

Differences in Parental Roles

Expectations around parenting, such as how to provide support to children, a parent’s disciplinary role, and what kinds of care to give children, vary across parental roles. For instance, the expectations of a biological mother or father are different from a stepparent or a nonresidential parent. Parents to a coresidential biological child are largely able to adhere to a set of norms, boundaries, and roles that are, for the most part, clearly defined by social convention (Cherlin, 2004; Cherlin & Furstenberg, 1994). These likely guide parenting behaviors and parent–child interactions and allow the parent to feel a sense of authority in enforcing rules, providing social support, and other parenting practices. Similarly, children tend to recognize biological parents as legitimate authority figures and form strong emotional and supportive bonds with residential biological parents (Nelson et al., 2013). Nevertheless, even in this traditional role, the demands and stresses of parenthood can strain a parent’s mental health—especially for women who tend to take on more housework and childcare responsibilities (Craig & Mullan, 2010; Scott & Alwin, 1989).

Though parenting a biological child is difficult, other parental roles may be more challenging. One such parental role is that of a stepparent. Stepparent–stepchild relationships tend to lack norms and clear expectations, which regulate a stepparent’s role within the family (Cherlin, 1978, 2004). As a result, nearly all aspects of the stepparental role are negotiable in the majority of stepfamilies (Kinniburgh-White, Cartwright, & Seymour, 2010; MacDonald & DeMaris, 2002). To illustrate, Marsiglio (1992) reported that about half of all stepfathers felt their parental role was better described as friend than father. Interestingly, many biological parents (typically mothers) feel differently. Weaver and Coleman (2010) report that many mothers want their spouses (the stepparent) to almost immediately take on a fatherly role so their new families may operate like biological nuclear families. Further complications result from stepchildren feeling that stepparents should be emotionally supportive (Kinniburgh-White et al., 2010), but any attempt to enforce rules or influence family culture is often viewed as off-putting and illegitimate (Hetherington & Kelly, 2002; Kinniburgh-White et al., 2010). In addition to the stress and potential negative psychological consequences associated with the stepparent role, it appears these problems and difficulties can engender difficulties in romantic relationships. For example, Stanley, Markman, and Whitton (2002) found that the most common issue couples in first marriages report arguing about is money, but in stepfamilies that issue is children. Perhaps these parenting challenges partially explain the increased risk of divorce in stepfamilies compared with biological families (Copen, Daniels, Vespa, & Mosher, 2012).

The residential status of children is also important for parental roles and psychological well-being. On the one hand, nonresidential parents may experience more stress and depressive symptoms than parents with coresidential children because, like stepparents, noncustodial parents lack a clear set of norms, expectations, and boundaries that guide their parenting practices (Evenson & Simon, 2005). On the other hand, day-to-day parenting responsibilities rarely fall to nonresidential parents—especially since contact with a nonresidential parent is far from consistent (Bastaits, Ponnet, & Mortelmans, 2012; Cooksey & Craig, 1998). As such, nonresidential parenting may not be associated with increased depressive symptoms. One’s parental role and residential status can also combine to affect psychological well-being. For example, nonresidential biological parents may feel more stress and have higher risk for depressive symptoms than nonresidential stepparents because of differences in parental expectations associated with each parental role.

Combined Parental Roles

With changing family demographics over the past several decades (see Cherlin, 2010), many adults take on multiple parental roles. As a result, it is important to consider how combined roles might be associated with psychological well-being. For instance, having a biological child and a stepchild in the same household might prove difficult for many adults. Parents with such roles may feel caught between their biological children and stepchildren and/or resist sharing full parental responsibilities of their biological child with their new spouse (the child’s stepparent)—meaning they take on the bulk of parental responsibilities (Weaver & Coleman, 2010). Biological parents in stepfamilies are often the only residential biological parent (Stewart, 2007) and may be the only biological parent with whom a child has contact, given the high rate of absent fatherhood among divorced individuals (Sweeney, 2010). Although it might be difficult to parent biological children in this setting, certainly stepparenting challenges play a role as well, as we described earlier. Attempts to balance these roles may be difficult for many parents, increasing stress, and the chances for high depressive symptoms.

Another common combined parental role, particularly in stepfamilies, is for a parent to have a biological child, a stepchild, and shared (new) child with their new partner. These children are often viewed as a way to “cement” the family together (Stewart, 2007). However, there is little evidence to suggest that a shared child will improve bonding in stepfamilies (MacDonald & DeMaris, 1996; Stewart, 2005), and our theoretical framework suggests the complexity of these roles within the same household might contribute to psychological distress in parents. In general, additional children in any family are associated with increased psychological distress for parents (Johnson & Wu, 2002). We are not aware of any studies that have analyzed the interaction of multiple types of parent–child relationships on parental mental health.

The Current Study

In this article, we focus on the role of various parent–child relationships, or parental roles, as stressors to parenting, as measured by adult depressive symptoms. Specifically, we consider how biological children and stepchildren, residing in or out of the home, having a new biological child with a partner at T2, and having no children are associated with depression. Our article builds on earlier studies by using nationally representative longitudinal data over a significant period (1992-2008) with detailed family relationship demographics. We hypothesize nontraditional (i.e., stepchild) and multiple parental roles will be associated with higher levels of depressive symptoms than traditional (i.e., biological child) parental roles or having no parental roles.

Method

Data

We used the National Longitudinal Survey of Youth, 1979 cohort (NLSY79), a nationally representative sample of 12,686 men and women born between 1957 and 1965. NLSY was collected annually between 1979 and 1994 and biennially since. The data include a full marital history, sociodemographic information on spouses and children, and various other measures. Because NLSY did not include questions on depressive symptoms until 1992, we do not include data from 1979 to 1991. This limitation resulted in 3,670 respondents being excluded from our data, 2,700 of which were a result of dropped oversamples. Some respondents refused to answer or did not receive the depression questions in 1992, but did so in 1994. We use these 1994 values as the baseline for these respondents. Statistical tests indicated no difference in responses. Depression was reassessed at age 40 or 50 in the health questionnaires of the NLSY, which raises issues of data attrition. By 2008, 7,757 respondents remained in the data, meaning that more than 46% (1,259 respondents) of the remaining respondents were lost due to data attrition. The remaining 54% (1,460 respondents) of respondents missing from our data refused to answer any depression question or validly skipped the questions for various reasons. In total, of the 9,016 eligible respondents, 6,297 remain in our analytic sample. Sample characteristics are reported in Table 1.

Table 1.

Sample Characteristics and Mean Depression Score by Characteristic, NLSY79^a.

	Mean depression score	n in category	% of respondents in category
All respondents	3.54	6,297
Residential biological child^b	3.21*	3,544	56.3
Only	2.89*	1,786	50.4
and nonresidential biological child	4.27*	197	5.6
and residential stepchild	3.52	782	22.1
and nonresidential stepchild	4.29	46	1.3
and new child	2.98*	435	12.3
More than two roles	4.00	298	8.4
Residential stepchild^b	3.67	1,407	22.3
Only	3.82	259	18.4
and residential biological child	3.52	782	55.6
and nonresidential biological child	2.60	30	2.1
and nonresidential stepchild	4.33	86	6.1
and new child	3.74	43	3.1
and residential biological child and new child	4.04	207	14.7
Nonresidential biological child	4.27*	197	3.1
Nonresidential stepchild	3.99*	433	6.9
New biological child	3.35	673	10.7
No children	3.62	1,352	21.5
Female	4.01*	3,180	50.5
First married	2.77*	2,551	40.5
Never married	4.29*	939	14.9
Divorced, no relationship	4.48*	1,230	19.5
Cohabiting	3.80	514	8.2
Remarried	3.60	1,021	16.2
Full-time employed	2.90*	4,120	65.4
Not a high school graduate	5.04*	706	11.2
High school graduate	3.77*	2,726	43.3
Some college	3.38	1,465	23.3
College degree or more	2.53*	1,397	22.2
Southern residence	3.72*	2,535	40.3
Urban residence	3.53	4,982	79.1
Non-Hispanic Black	3.91*	1,908	30.3
Non-Hispanic White	3.37*	3,154	50.1
Hispanic	3.43	1,235	19.6
Catholic	3.18*	2,217	35.2
Conservative Protestant	3.97*	1,866	29.6
Mainline Protestant	3.61	1,284	20.4
Other religious affiliation	3.75	676	10.7
No religious affiliation	3.98	254	4.0
Frequently attends church	3.98*	1,070	17.0
Two-parent household at age 14	3.35*	4,224	67.1
Stepparent present at age 14	3.99*	541	8.6
Single parent and other arrangements at 14	3.74	1,532	24.3

Source. National Longitudinal Survey of Youth, 1979 cohort (NLSY79).

Numbers exceed 6,297 and 100% because respondents may have multiple parental roles.

Figures for subgroups are for respondents within category.

Indicates statistically significant difference between the overall mean and the characteristic mean for depression via a t test at p < .05.

Depressive Symptoms

Our dependent variable, self-reported depressive symptoms, was assessed using a 7-item Center for Epidemiologic Studies Depression Scale (CES-D scale), which is reliable across gender, race, and age (Radloff, 1977). The measure includes items which assess the frequency respondents felt depressed or sad, had problems eating or sleeping, could not focus on tasks, and required extra effort to complete tasks in the past week. Responses were given on a 4-point Likert-type scale (0-3) with scaled scores ranging from 0 to 21 and higher scores indicating more depressive symptoms. Baseline depression was measured in 1992 or 1994, when respondents were 27 to 37 years old, while T2 depression was measured at age 40 or 50 (approximately 58% of the sample had T2 depression measured at age 50). T2 depression comes from the NLSY health evaluation, which began in 1998. Because using the 40-year-old evaluation would have produced a very small time gap between T1 and T2 for the oldest divorced men and women, we used depression scores at age 50 when possible. Most T2 evaluations took place between 2000 and 2006, a difference of 8 to 12 years from baseline. Cronbach’s alpha at baseline was excellent (α = .97) and good at age 40 (α = .86). Individual items on the scale were not available for 50-year-olds in an effort to protect respondent anonymity since relatively few respondents had reached age 50 in 2008 and the reliability coefficient was not provided by NLSY. Nevertheless, screening efficacy and the reliability of the CES-D scale has been established for 50-year-olds in other samples (e.g., Lewinsohn, Seeley, Roberts, & Allen, 1997).

Parental Roles

Our key independent variable is a measure of parental roles, which comes from the respondent’s relationship to their own children, their partner’s children (where applicable), and the residential status of the children. We include a set of several dichotomous variables for the various parental roles of respondents. These measures include the presence of a residential biological child, a nonresidential biological child, a residential stepchild, a nonresidential stepchild, a new biological child with the T2 partner, and no children. Each of these variables is included in the model so that comparisons are in reference to individuals who do not have that particular parental role (e.g., residential stepchild vs. no residential stepchild). Of course, respondents can have various parental roles and can be in multiple categories. We account for this possibility as well, with models that include variables for combined roles.

Control Variables

We include numerous control variables associated with depression and parental roles in our analyses. We control for respondent sex because women are at higher risk for depressive symptoms than men and often take on more significant parental roles than men (Evenson & Simon, 2005). Some parental roles are more common in particular relationship types than others, such as the relatively high likelihood of having stepchildren if one is remarried (Teachman, 2008). Similarly, relationship status is also associated with depression levels. For example, divorced men and women tend to have higher levels of depression than the continuously married (Amato, 2010). Time could be responsible for any association between parental roles and psychological well-being. As a result, we included a dichotomous variable for T2 depression measured at age 40 (compared with age 50) and a continuous variable for duration (in years) of the current relationship at T2. The value of this variable was zero if the respondent was not currently in a romantic union.

We also controlled for various personal characteristics associated with relationship status (Cherlin, 2010), childbearing (Stewart, 2002), and psychological well-being (Evenson & Simon, 2005). We include dichotomous variables for race (Hispanic, non-Hispanic Black, and non-Hispanic White), along with three measures of socioeconomic status. Employment status was measured by a set of time-varying control variables indicating if the respondent had full-time employment, which we operationalized as averaging 30 or more hours of work per week. Family income was total household income adjusted to 2008 dollars and logged to adjust for positive skew. Measures of educational attainment at baseline were dichotomous variables for less than 12 years of schooling (less than high school graduate), 12 years (high school graduate), 13 to 15 years (some college), and 16 or more years (college graduate). Similarly, we included variables for Southern and urban residence because stepfamilies and divorce are more common in these locations (U.S. Census Bureau, 2008).

Since religious involvement is associated with somewhat fewer depressive symptoms and acts as protective factor among individuals undergoing psychosocial stress (Smith, McCullough, & Poll, 2003), religious affiliation was measured by dichotomous variables for Mainline Protestant, Conservative Protestant, Catholic, and Other religious affiliation. We also included a dichotomous variable for frequent religious attendance, which was operationalized as attending church at least once a week. Finally, because family structure can be transmitted intergenerationally (Teachman, 2002) and family-of-origin experiences may help individuals acclimate in less normative family structures, we included variables for the family structure of the respondent while an adolescent.

Analytic Method

We used ordinary least squares (OLS) regression, a standard method in studies of parenting and depression (e.g., Evenson & Simon, 2005; Frech & Williams, 2007). We also considered the use of logistic regression analysis, where we dichotomized high depressive and low depressive symptoms via a clinically acceptable definition–a cutoff for high depressive symptoms of 6 on a 21 point scale (e.g., Comstock & Helsing, 1976; Cseh, 2008). These models produced substantively similar results to the results presented here. However, we present the OLS models because psychological well-being may be better conceived as being on a continuum than as an either/or proposition. There was a small amount (<5%) of missing data on our control variables—with data most commonly missing for family income. No missing data were present on either our dependent variable or our key independent variables. The missing data appeared to be missing at random or missing completely at random, which permitted us to use multiple imputation procedures (Enders, 2010). We conducted 10 imputations using the mi command in Stata 12.0. The results from our imputed models were similar to those from nonimputed data.

Our analysis proceeds in four stages. First, we present a model that includes only parental roles without the presence of any control variables. In Model 2, we add control variables to our measures of parental roles. Because of our interest in combined parental roles, we also include models for combined roles among respondents with a residential biological child and respondents with a residential stepchild. We focus on these two roles in particular because more than 73% of respondents and 90% of parents in our sample have at least one of these two parental roles. In the first model, we focus on respondents who had at least a residential biological child but could have had additional parental roles. Similarly, the second model focuses on respondents with a residential stepchild, but also includes roles combined with the stepparent role.

Results

Descriptive Statistics

In addition to sample characteristics, Table 1 also provides mean depression score by characteristic. The mean CES-D depression score for all respondents was 3.21. Considering the importance of parental roles, respondents with a residential biological child have lower than average depression scores, whereas those with nonresidential biological or stepchildren have a higher than average depression score. Fifty-six percent of respondents reported having a biological child, with 50% of that group having only a biological child. Twenty-two percent reported having a stepchild and residential biological child and 12% reported a new child. Among adults with residential stepchildren (22% of sample), 18% had only stepchildren, whereas nearly 56% had a biological child and residential stepchild in the same home. An additional 15% had a residential biological child, stepchild, and new child with their current partner.

Respondents with higher than average depression scores included women, the never-married and divorced, those with a high school degree or lower, Southerners, Blacks, Conservative Protestants, frequent church attendees, and those who grew up in stepfamilies. First married, full-time employed, college educated, White, and Catholic respondents had lower than average depression scores. Furthermore, those raised in two-parent households had better psychological well-being, as well.

Parental Roles Model

The results of our OLS regression focusing on individual parental roles can be found in Table 2. Because the parental roles categories are not mutually exclusive, respondents with a particular parental role (e.g., has a residential biological child) is compared with respondents who do not have that role (i.e., no biological child present in home) in this part of the analyses. Model 1 shows that men and women with nonresidential biological children have depression scores 0.798 points higher than respondents without this role. Stepparents also have worse psychological well-being. Adults with residential stepchildren have 0.586 points higher depression scores and those with nonresidential stepchildren 1.527 points higher scores, relative to respondents without such roles. Those with no children score 0.631 points higher on the depression scale than respondents with any parental role.

Table 2.

OLS Regression of CES-D Composite Depression Score at Age 40 or 50 on Parental Roles and Control Variables.

	Model 1	Model 2
Parental roles^a
Residential biological child	−0.058 (0.261)	0.092 (0.263)
Nonresidential biological child	0.798*** (0.227)	0.320 (0.211)
Residential stepchild	0.586*** (0.155)	0.545** (0.177)
Nonresidential stepchild	1.527*** (0.307)	0.806* (0.343)
New biological child since T1	0.249 (0.239)	0.137 (0.224)
No children	0.631* (0.300)	0.095 (0.304)
Female		0.716*** (0.104)
Relationship status^b
Never married		0.790*** (0.182)
Divorced, no relationship		0.738*** (0.169)
Cohabiting		0.129 (0.215)
Remarried		0.206 (0.162)
Years in current relationship state		−0.004 (0.063)
Depression status
Depression score at T1		0.282*** (0.013)
Depression measured at age 50		0.030 (0.026)
Full-time employed		−0.876*** (0.119)
Income, logged		−0.126*** (0.021)
Educational attainment^c
Not a high school graduate		1.099*** (0.198)
High school graduate		0.455*** (0.137)
Some college education		0.266 (0.151)
Southern residence		0.127 (0.109)
Urban residence		−0.061 (0.125)
Racial/ethnic identification^d
Non-Hispanic Black		0.355* (0.141)
Hispanic		−0.254 (0.154)
Religious affiliation^e
Catholic		−0.227 (0.155)
Conservative Protestant		0.184 (0.158)
Other religious affiliation		0.125 (0.189)
No religious affiliation		0.128 (0.275)
Frequently attends religious services		−0.035 (0.031)
Family Structure at age 14^f
Stepparent present		0.196 (0.180)
Single parent and other arrangements		−0.084 (0.137)
Constant	2.989	−57.133
N	6,297	6,297
R ²	0.02	0.16

Note. OLS = ordinary least squares; CES-D = Center for Epidemiologic Studies Depression Scale.

Source. National Longitudinal Survey of Youth, 1979 cohort (NLSY79).

This is not a multinomial variable, each comparison is between having that parental role and not having that parental role.

Reference is first marriage.

Reference is college graduate or higher.

Reference is non-Hispanic White.

Reference is Mainline Protestant.

Reference is two biological parents.

p < .05. **p < .01. ***p < .001.

Control variables are added in Model 2. As a result, the effects of having nonresidential biological children and having no children were no longer statistically significant. However, stepparents remained at risk for worse psychological well-being than nonstepparents. Those with residential stepchildren scored 0.545 points higher on the depression scale than respondents without residential stepchildren, while having nonresidential stepchildren was associated with depression scores 0.806 higher than individuals without nonresidential stepchildren. Turning to the control variables, women had substantially higher depression scores than men, as did the never-married and divorced when compared with first married respondents. Higher depression at T1 was associated with increased depression at T2, although we found no differences in whether depression was measured at age 40 or 50. Less educated and Black respondents were also at risk for more depressive symptoms than their respective better-educated and White counterparts. Two protective factors were identified, however. Full-time employment and income were both associated with a reduction in depressive symptoms.

Combined Parental Roles

OLS regressions addressing combined parental roles are reported in Table 3. Although we do not present the results of control variables in this table, the full controls highlighted in Model 2 of Table 2 were included in both models shown in Table 3. Model 1 focuses on respondents who reported a residential biological child (n = 3,534). We compared respondents with only residential biological children with respondents with residential biological and nonresidential biological children, with residential biological and residential stepchildren, and with more than two parental roles have worse psychological well-being. Having a nonresidential biological child increases depression scores by 0.590 points, whereas the presence of a stepchild increases scores by 0.543 points. Multiple roles appear to be particularly harmful, increasing depression scores by 1.041 points.

Table 3.

OLS Regression of CES-D Composite Depression Score at Age 40 or 50 on Combined Parental Roles.

	Model 1^a	Model 2^a
Residential biological child^b
Residential biological and nonresidential biological	0.590* (0.280)
Residential biological and residential stepchild	0.543** (0.207)
Residential biological and nonresidential stepchild	0.139 (0.557)
Residential biological and new biological child	0.046 (0.199)
More than two roles	1.041*** (0.264)
Residential stepchild^c
Residential stepchild only		0.028 (0.426)
Residential stepchild and new biological child		0.305 (0.703)
Residential stepchild and nonresidential stepchild		0.805 (0.541)
Residential stepchild, residential biological child, and new biological child		0.869* (0.251)
Constant	−65.292	−73.002
N	3,534	1,094
R ²	0.15	0.14

Note. OLS = ordinary least squares; CES-D = Center for Epidemiologic Studies Depression Scale.

Source. National Longitudinal Survey of Youth, 1979 cohort (NLSY79).

All models include full controls found in Model 2 of Table 2.

This model includes only individuals who report a residential biological child with individuals with only a residential biological child as the reference.

This model includes only individuals who report residential stepchildren with respondents with a residential stepchild and residential biological child as the reference category.

p < .05. **p < .01. ***p < .001.

Among respondents with residential stepchildren, as reported in Model 2, only one statistically significant effect was found. In comparison with respondents with a residential biological child and residential stepchild (the most common role among respondents with stepchildren), having a stepchild, biological child, and a shared child with the new partner increases depression scores by 0.869 points. Thus, multiple parental roles are harmful to the psychological well-being of respondents with a residential stepchild, as well.

Discussion

In this study, we analyzed depressive symptoms in adults across numerous parent–child relationships. We hypothesized that less traditional parental roles, such as stepparenting or having nonresidential biological children, and multiple parental roles would be associated with higher levels of depressive symptoms than more traditional parental roles, such as having residential biological children or no parental roles. We found evidence to support our hypothesis since having a stepparent–stepchild relationship, regardless of residential status, was associated with higher depressive symptoms compared with those without these respective relationships. This is in contrast to findings by Evenson and Simon (2005), who found a null relationship for minor residential stepchildren but a negative relationship with adult nonresidential stepchildren. We also found some association between depressive symptoms and multiple roles—but not in all cases.

Having more than two roles, such as having a biological child, a stepchild, and a new child, seems to be problematic for parents and is associated with higher depression scores, on average. This finding is in agreement with the argument that a “cement” child might not improve parental well-being and family bonding (MacDonald & DeMaris, 1996; Stewart, 2005). Having a residential biological child and a nonresidential biological child seemed to put parents at risk of depressive symptoms, perhaps because of stresses and worries associated with not having custody of one of the children. Having a residential biological child and a residential stepchild was also problematic, which seems to confirm previous research on difficulties associated with stepfamilies (see Stewart, 2007, for a full discussion). Yet having a biological residential child and nonresidential stepchild, a biological residential child and a new biological child, a residential stepchild and a new biological child, or a residential stepchild and a nonresidential biological child each had no association with depression. To speculate why we observe such patterns, much of this may be because of proximity to parental role conflict. By this we mean that multiple parental roles may only be problematic for mental health when they are experienced in day-to-day interactions. Multiple parental roles where a (step)child is nonresidential appears to have no association with well-being because daily interaction with multiple parental roles is unlikely. In contrast, the presence of a biological child and stepchild in the home is associated with depressive symptoms. Furthermore, it appears that feeling caught between different parental roles (Weaver & Coleman, 2010) may also prove problematic for psychological well-being. For example, the presence of a biological child and stepchild or a biological child, stepchild, and new child is associated with higher depression scores, but a biological child and new child or stepchild and new child is not.

Limitations

Of course, we are merely speculating why different parental roles are associated with poorer psychological well-being. Like most research on this question, the mechanisms that link parenting to depression (e.g., Evenson & Simon, 2005) are not well-established. One issue is the lack of questions, particularly in nationally representative data sets, that address specific problems, frustrations, and stresses within the household. Such items would allow researchers to better link these issues with outcomes, such as depression. Although we have good theoretical reasons for why we see the specific associations we observe in our data, this argument would have been bolstered by the presence of such variables.

Our study has other limitations, as well. Depression was only measured at two time points—age 27 to 37 and age 40 or 50. It would have been beneficial if NLSY measured symptoms more frequently. Also, our data have CES-D scores for only one partner, rather than both. It is unfortunate we cannot control for depressive symptoms of respondent’s partner, if any. Other child characteristics beyond the nature of the parent–child relationship, such as child sex, were not included in our models—mostly because of the lack of sufficient data on nonbiological children and the substantial complexity of some households in the NLSY sample. Despite these limitations, our study possesses strengths. We used nationally representative longitudinal data with considerably detailed family demographics. We included men and women in our sample, whereas the majority of studies on parenting only include women in their respective samples. We tested the association of numerous parent–child relationships and depressive symptoms with a degree of detail, which to the best of our knowledge has not been undertaken in the literature.

Importantly, more research on the relationship between parenting and psychological well-being is needed. Future studies should consider including fathers, when possible. For instance, even though depression in fathers is probably less strongly associated with internalizing and externalizing problems in children than mothers’ depression, there is still a negative association (Connell & Goodman, 2002). Studies should also identify mediators and moderators in the association between parenting and mental health and utilize alternative constructs of psychological well-being (e.g., anxiety). When possible, measures using more frequent time points in psychological well-being outcomes would yield important findings.

Implications

Our findings have several implications. Although there are effective methods available to identify adults with depression, most screening programs do not assess whether the adult is a parent (National Research Council and Institute of Medicine, 2009). This is unfortunate given depression’s negative association with positive parenting practices and child outcomes, particularly regarding mothers (Goodman et al., 2011; Lovejoy, Graczyk, O’Hare, & Neuman, 2000). Screening programs for depression need to consider whether the adult is a parent as well as relevant parental roles. Our study provides evidence for assessing for not only whether someone is a parent as a potential risk factor for poor psychological well-being but also whether she/he is a stepparent, regardless of whether the stepchild resides in or out of the home. There also seems to be value in assessing for whether the parent has more than two parental roles. Particularly, having a new child in the context of already having a biological child and stepchild in the same household seems to be a risk factor. If any of these risk factors are indicated by an individual during the administration of a screening tool, we recommend professionals enquire further about the influence of these factors on their mental health, as well as how their mental health might be influencing their parenting practices. Ultimately, it is up to administrators, practitioners, and researchers to determine specific items to add to screening tools; nevertheless, our study provides empirical basis for a few basic risk factors, which should be feasible to include.

While parenting may be a rewarding experience, certain parental roles seem problematic for adult psychological well-being. Specifically, being a stepparent or having certain combinations of multiple parental roles can put parents at risk for higher depressive symptoms. Our research has demonstrated the utility of utilizing detailed family demographics and addressing some of the complexities of postmodern parenthood in explaining parental mental health. We concur with Evenson and Simon (2005) that parenthood is accompanied by diverse experiences and circumstances among individuals and partners, and the trends we have found suggest differences across a number of parental roles.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Amato

P. R.

(2010). Research on divorce: Continuing trends and new developments. Journal of Marriage and Family, 72, 650-666.

Bastaits

Ponnet

Mortelmans

(2012). Parenting of divorced fathers and the association with children’s self-esteem. Journal of Youth and Adolescence, 41, 1643-1656.

Burton

R. P. D.

(1998). Global integrative meaning as a mediating factor in the relationship between social roles and psychological distress. Journal of Health and Social Behavior, 39, 201-215.

Cherlin

(1978). Remarriage as an incomplete institution. American Journal of Sociology, 84, 634-650.

Cherlin

A. J.

(2004). The deinstitutionalization of American marriage. Journal of Marriage and Family, 66, 848-861.

Cherlin

A. J.

(2010). Demographic trends in the United States: A review of research in the 2000s. Journal of Marriage and Family, 72, 403-419.

Cherlin

Furstenberg

F. F.

(1994). Stepfamilies in the United-States: A reconsideration. Annual Review of Sociology, 20, 359-381.

Comstock

G. W.

Helsing

K. J.

(1977). Symptoms of depression in two communities. Psychological Medicine, 6, 551-563.

Connell

A. M.

Goodman

S. H.

(2002). The association between psychopathology in fathers versus mothers and children’s internalizing and externalizing behavior problems: A meta-analysis. Psychological bulletin, 128, 746-773.

10.

Cooksey

E. C.

Craig

P. H.

(1998). Parenting from a distance: The effects of paternal characteristics on contact between nonresidential fathers and their children. Demography, 35, 187-200.

11.

Copen

C. E.

Daniels

Vespa

Mosher

W. D.

(2012). First marriage in the United States: Data from the 2006-2010 National Survey of Family Growth. Atlanta, GA: U.S. Department of Health and Human Services.

12.

Coryell

Endicott

Keller

(1990). Outcome of patients with chronic affective disorder: A five-year follow-up. American Journal of Psychiatry, 147, 1627-1633.

13.

Craig

Mullan

(2010). Parenthood, gender and work-family time in the United States, Australia, Italy, France, and Denmark. Journal of Marriage and Family, 72, 1344-1361.

14.

Cseh

(2008). The effects of depressive symptoms on earnings. Southern Economic Journal, 75, 383-409.

15.

Davies

Avison

W. R.

McAlpine

D. D.

(1997). Significant life experiences and depression among single and married mothers. Journal of Marriage and Family, 59, 294-308.

16.

Enders

C. K.

(2010). Applied missing data analysis. New York, NY: Guilford Press.

17.

Evans

D. L.

Charney

D. S.

Lewis

Golden

R. N.

Gorman

J. M.

Krishnan

K. R. R.

. . . Valvo

W. J.

(2005). Mood disorders in the medically ill: Scientific review and recommendations. Biological Psychiatry, 58, 175-189.

18.

Evenson

R. J.

Simon

R. W.

(2005). Clarifying the relationship between parenthood and depression. Journal of Health and Social Behavior, 46, 341-358.

19.

Frech

Williams

(2007). Depression and the psychological benefits of entering marriage. Journal of Health and Social Behavior, 48, 149-163.

20.

Glenn

N. D.

Mclanahan

(1981). The effects of offspring on the psychological well-being of older adults. Journal of Marriage and the Family, 43, 409-421.

21.

Goodman

S. H.

Rouse

M. H.

Connell

A. M.

Broth

M. R.

Hall

C. M.

Heyward

(2011). Maternal depression and child psychopathology: A meta-analytic review. Clinical Child and Family Psychology Review, 14(1), 1-27.

22.

Hetherington

E. M.

Kelly

(2002). For better or for worse: Divorce reconsidered. New York, NY: Norton.

23.

Hewlett

S. A.

Rankin

West

(2002). Taking parenting public: The case for a new social movement. Lanham, MD: Rowman & Littlefield.

24.

Huber

(1980). Will US fertility decline toward zero? Sociological Quarterly, 21, 481-492.

25.

Hughes

(1989). Parenthood and psychological well-being among the formerly married: Are children the primary source of psychological distress? Journal of Family Issues, 10, 463-481.

26.

Johnson

D. R.

(2002). An empirical test of crisis, social selection, and role explanations of the relationship between marital disruption and psychological distress: A pooled time-series analysis of four-wave panel data. Journal of Marriage and Family, 64, 211-224.

27.

Kandel

D. B.

Davies

Raveis

V. H.

(1985). The stressfulness of daily social roles for women: Marital, occupational and household roles. Journal of Health and Social Behavior, 26(1), 64-78.

28.

Kessler

R. C.

Akiskal

H. S.

Ames

Birnbaum

Greenberg

Hirschfeld

R. M.

. . . Wang

P. S.

(2006). The prevalence and effects of mood disorders on work performance in a nationally representative sample of US workers. American Journal of Psychiatry, 163, 1561-1568.

29.

Kessler

R. C.

Berglund

Demler

Jin

Koretz

Merikangas

K. R.

. . . Wang

P. S.

(2003). The epidemiology of major depressive disorder. Journal of the American Medical Association, 289, 3095-3105.

30.

Kinniburgh-White

Cartwright

Seymour

(2010). Young adults’ narratives of relational development with stepfathers. Journal of Social and Personal Relationships, 27, 890-907.

31.

Lewinsohn

P. M.

Seeley

J. R.

Roberts

R. E.

Allen

N. B.

(1997). Center for Epidemiologic Studies Depression Scale (CES-D) as a screening instrument for depression among community-residing older adults. Psychology and Aging, 12, 277-287.

32.

Lovejoy

M. C.

Graczyk

P. A.

O’Hare

Neuman

(2000). Maternal depression and parenting behavior: A meta-analytic review. Clinical Psychology Review, 20, 561-592.

33.

MacDonald

W. L.

DeMaris

(1996). Parenting stepchildren and biological children: The effects of stepparent’s gender and new biological children. Journal of Family Issues, 17, 5-25.

34.

MacDonald

W. L.

DeMaris

(2002). Stepfather-stepchild relationship quality: The stepfather’s demand for conformity and the biological father’s involvement. Journal of Family Issues, 23, 121-137.

35.

Marsiglio

(1992). Stepfathers with minor children living at home: Parenting perceptions and relationship quality. Journal of Family Issues, 13, 195-214.

36.

Menaghan

E. G.

(1989). Role changes and psychological well-being: Variations in effects by gender and role repertoire. Social Forces, 67, 693-714.

37.

National Research Council and Institute of Medicine. (2009). Depression in parents, parenting, and children: Opportunities to improve identification, treatment, and prevention. Washington, DC: National Academies Press.

38.

Nelson

S. K.

Kushlev

English

Dunn

E. W.

Lyubomirsky

(2013). In defense of parenthood: Children are associated with more joy than misery. Psychological Science, 24(1), 3-10.

39.

Preston

S. H.

(1986). Changing values and falling birth rates. Population and Development Review, 12(Suppl.), 196-200.

40.

Radloff

L. S.

(1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.

41.

Scott

Alwin

D. F.

(1989). Gender differences in parental strain: Parental role or gender role? Journal of Family Issues, 10, 482-503.

42.

Smith

T. B.

McCullough

M. E.

Poll

(2003). Religiousness and depression: Evidence for a main effect and the moderating influence of stressful life events. Psychological Bulletin, 129, 614-636.

43.

Stanley

S. M.

Markman

H. J.

Whitton

S. W.

(2002). Communication, conflict, and commitment: Insights on the foundations of relationship success from a national survey. Family Process, 41, 659-675.

44.

Stewart

S. D.

(2002). The effect of stepchildren on childbearing intentions and births. Demography, 39, 181-197.

45.

Stewart

S. D.

(2005). How the birth of a child affects involvement with stepchildren. Journal of Marriage and Family, 67, 461-473.

46.

Stewart

S. D.

(2007). Brave new stepfamilies: Diverse paths toward stepfamily living. Thousand Oaks, CA: Sage.

47.

Sweeney

M. M.

(2010). Remarriage and stepfamilies: Strategic sites for family scholarship in the 21st century. Journal of Marriage and the Family, 72, 667-684.

48.

Teachman

J. D.

(2002). Childhood living arrangements and the intergenerational transmission of divorce. Journal of Marriage and Family, 64, 717-729.

49.

Teachman

(2008). Complex life course patterns and the risk of divorce in second marriages. Journal of Marriage and Family, 70, 294-305.

50.

Thornton

Young-DeMarco

(2004). Four decades of trends in attitudes toward family issues in the United States: The 1960s through the 1990s. Journal of Marriage and Family, 63, 1009-1037.

51.

Umberson

Gove

W. R.

(1989). Parenthood and psychological well-being: Theory, measurement, and stage in the family-life course. Journal of Family Issues, 10, 440-462.

52.

U.S. States Census Bureau. (2008). Statistical abstract of the United States, 2007. Washington, DC: Department of the Treasury.

53.

Weaver

S. E.

Coleman

(2010). Caught in the middle: Mothers in stepfamilies. Journal of Social and Personal Relationships, 27, 305-326.

54.

Williams

Dunne-Bryant

(2006). Divorce and adult psychological well-being: Clarifying the role of gender and child age. Journal of Marriage and Family, 68, 1178-1196.