Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: Fam Relat. 2018 Jun 7;67(4):449–466. doi: 10.1111/fare.12327

Cohort Differences in Mothers’ Perceptions of Neighborhood Quality, Child Well-being, and Parental Strain, 1976–2002

Kei Nomaguchi 1, Marshal Neal Fettro 1
PMCID: PMC6239198  NIHMSID: NIHMS968971  PMID: 30455537

Abstract

Objective

To examine differences in mothers’ subjective experiences of child-rearing between two cohorts who lived in different eras of sociodemographic contexts and parenting norms.

Background

A resource perspective suggests that child-rearing experiences should be easier when mothers have a higher standard of living, higher education, and fewer children, as was the case for mothers in the 2000s compared with mothers in the 1970s. However, a cultural perspective indicates that the intensive mothering ideology, emerged in the 1980s, increased mothers’ anxiety and stress over their children’s safety and proper development.

Method

Data from two national samples of mothers with children between 6 and 12 years of age in the United States collected in 1976 and 2002 (N = 2,465) were analyzed using ordinary least squares regression models.

Results

Mothers in 2002 reported better neighborhood quality and better health of their children than mothers in 1976, even after sociodemographic advantages of mothers in 2002 relative to 1976 were taken into account. Despite these sociodemographic advantages of mothers in 2002, there was little difference in mothers’ reports of their children’s behavioral adjustments between the two cohorts. Furthermore, albeit among Whites only, mothers in 2002 reported more parental strain than did mothers in 1976. There was little variation in mothers’ perceptions by socioeconomic status (as measured by a college degree).

Conclusion

Our results support the idea of a stressed-moms phenomenon in the contemporary United States, particularly among Whites.

Implications

Experts who advise parent education policy and programs should take into account the psychological burden of the current neoliberal emphasis on personal responsibility for raising children on individual mothers.

Keywords: cohort differences, intensive mothering ideology, parental strain, race and ethnicity, sociodemographic changes, subjective perceptions of parenting

Much research has shown that mothers’ parenting behavior in the United States changed markedly during the last quarter of the 20th century. Most notably, mothers’ investments in children increased. Mothers spent more directed time with their children (Bianchi, 2011) and more money on child-related expenses (Kornrich & Furstenberg, 2013) in the 2000s than in the 1970s. Scholars tend to argue that changes in sociodemographic and cultural contexts of motherhood, which took place during that period, contributed to the increases in maternal investment in child-rearing (Sayer, Bianchi, & Robinson, 2004). For example, compared with mothers in the 1970s, mothers in the 2000s were on average more educated and older, had fewer children, and stayed employed more consistently throughout their children’s childhood (Grieco, 2010; Leibowitz & Klerman, 1995; Mathews & Hamilton, 2009). Furthermore, in the mid-1980s parenting norms began to emphasize the necessity of mothers’ involvement throughout their children’s development (Hays, 1996) and children’s need to be constantly supervised to avoid seemingly increased dangers and insecurities (Rutherford, 2011; Stearns, 2003).

Scant quantitative research has examined how these sociodemographic and cultural changes might have been reflected in mothers’ subjective perceptions of child-rearing between the 1970s and the 2000s. Qualitative studies have suggested that mothers in the 2000s were more likely than mothers in the 1970s to feel pressured to protect their children from any potential harm as well as to prepare their children for a competitive workplace (Nelson, 2010). Using anecdotal evidence, social critics have asserted that such cultural pressures affect mothers negatively through greater anxiety and feelings of being rushed (Anderegg, 2003; Warner, 2005). In the past 2 decades, major media outlets have published stories claiming that many mothers in the current parenting culture feel as if child-rearing is more difficult today than in the past (e.g., Kantrowitz & Wingert, 2001; Wallace, 2014). Yet this supposition has not been examined quantitatively to identify differences in mothers’ perceptions of their children’s well-being, their neighborhood quality, or parenting demands between the 1970s and the 2000s. This gap in the literature is in part due to a lack of data; it is difficult to find comparable samples and questions to measure mothers’ subjective experiences of parenting in the 1970s and in more recent years.

Using two national samples of mothers with children 6 to 12 years of age, the 1976 National Survey of Children (NSC 1976) and the 2002 Child Development Supplement of the Panel Study of Income Dynamics (PSID-CDS 2002), the present study was designed to examine differences in mothers’ subjective experiences of parenting between the mid-1970s and the early 2000s. We focus on mothers’ perceptions of (a) neighborhood quality specific to raising children, (b) their children’s general health, (c) their children’s behavioral adjustments, and (d) parental strain, defined as perceptions of difficulty associated with raising children (Pearlin, 1989). Further, because some prior research has suggested that changes in sociodemographic contexts and parenting norms during the period occurred unevenly by socioeconomic status (SES) and race/ethnicity (Lareau, 2003; Levy, 1998), we examine whether patterns of differences in mothers’ subjective experiences of parenting between the two cohorts varied by SES and race/ethnicity. Because subjective experiences of parenting have implications for maternal mental health (Pearlin, 1989), understanding cohort differences in mothers’ subjective experiences is important for practitioners and policymakers.

Literature Review

Between the mid-1970s and the early 2000s, contexts surrounding child-rearing in the United States changed in many ways. We approach these changes from two perspectives. One is what we call a resource perspective, which focuses on changes in overall standards of living and mothers’ sociodemographic characteristics. We call the other perspective a cultural perspective, which focuses on changes in parenting norms. Although recent scholarly and public debates tend to focus on the latter, the former perspective is also important to consider. In this section, we discuss how each perspective may predict differences between the mid-1970s and the early 2000s in mothers’ perceptions of neighborhood quality to raise children, children’s health, children’s behavior, and parental strain.

Changes in Sociodemographic Characteristics

Macro-level changes in standards of living, fertility, women’s educational attainment, and maternal labor force participation rates between the 1970s and the 2000s suggest that mothers in the 2000s had greater access to resources that would help them deal with the demands of parenting, and thus they should have reported better neighborhood quality, better child health and behavior, and less parental strain than mothers in the 1970s. Yet other changes such as the decline in marriage rates and the increase in the proportion of non-Whites during the same period might have resulted in increases in mothers’ reports of poor neighborhood quality, poor child health and behavior, and higher parental strain from 1976 to 2002.

Standards of living

Macro-level standards of living in the United States increased between the mid-1970s and the early 2000s, at least by some indicators. Gross domestic product per capita rose throughout the 20th century (World Bank, 2016). Median household income increased from $54,971 in 1976 to $68,085 in 2002, in 2015 dollars (U.S. Census Bureau, 2016). Several health-related indicators, important measures of standards of living, suggest improvements in children’s health. For example, infant mortality rates declined from 14.56 per 1,000 live births in the 1975–1979 period to 7.53 per 1,000 live births in the 1995–2000 period (Singh & Kogan, 2007). In addition, childhood exposure to lead paint declined rapidly from the mid-1970s to the early 2000s (Levin et al., 2008). Maternal smoking during pregnancy, which is related to poorer child health, declined during the 1990s (Bianchi, Cohen, Raley, & Nomaguchi, 2004; Mathews, 2001). Further, more children gained access to health insurance during this period: More disadvantaged families received health insurance through Medicaid expansions and the launch of the Children’s Health Insurance Program in 1997, which covered children in families with incomes too high to be eligible for Medicaid but who were unable to afford private insurance (Mann, Rowland, & Garfield, 2003). Together, these macro-level trends suggest that mothers in the early 2000s would be more likely than mothers in the mid-1970s to report better neighborhood quality, better child health and behavior, and less parental strain.

Fertility

Women in the early 2000s became mothers at older ages and had fewer children than in the mid-1970s. The median age at first birth increased from 22.1 years in 1970 to 24.6 years in 2000 (Mathews & Hamilton, 2002). Women who become parents at an older age tend to have greater economic, social, and psychological resources, which may be related to better neighborhood quality, better child health, fewer problematic child behaviors, and less parental strain (Lee & Ryff, 2016; Nomaguchi & Brown, 2011). The mean number of children per woman declined from approximately three in the mid-1970s to approximately two in the 2000s (Monte & Ellis, 2014). Having fewer children at home is related to better child health and behavior, perhaps because it allows mothers to provide more resources and attention to each child (Blake, 1981; Datar, 2017). Fewer children create fewer parenting demands and, thus, this characteristic is related to lower parental strain (Nomaguchi & House, 2013).

Education

Between the mid-1970s and the early 2000s, women in the United States made large advancements in educational attainment. The percentage of women 25 to 29 years of age who had a bachelor’s degree increased from 17.6% in the mid-1970s to 32.0% in 2002 (U.S. Census Bureau, 1974, 2003). Mothers with higher levels of education are more likely to live in better neighborhoods and have children with no health problems (Barbarin et al., 2006; Singh & Kogan, 2007). Mothers with college degrees experience less parental anxiety regarding the safety and health of their children, in part because their material resources allow them to place their children in better quality child-care programs and schools (Nomaguchi & Brown, 2011). Thus, increases in levels of education among mothers may have been related to mothers’ perceptions of better neighborhood quality, better child health and adjustment, and less parental strain in the early 2000s than in the mid-1970s.

Employment

Mothers in the early 2000s were more likely than mothers in the mid-1970s to be working for pay. Labor force participation among mothers with children 6 to 17 years of age increased from 56.2% in 1976 to 78.6% in 2002 (Mosisa & Hipple, 2006). Paid work brings economic, social, and psychological resources for mothers and thus may be related to their reports of better neighborhood, better child health and behavior, and less parental strain (Johnson, Kalil, & Dunifon, 2012; Nomaguchi & Milkie, 2017). Although balancing work and family responsibilities could be difficult and stressful, past research tends to show that paid work is more likely to be related to less parental strain (Nomaguchi & Brown, 2011).

Romantic partnerships

As cohabitations and marriages became less stable, more mothers experienced single parenthood in the 2000s than in the mid-1970s (Cherlin, 2010). Compared with partnered mothers, single mothers tend to have fewer resources, such as lower levels of family income, which may in turn be related to poorer neighborhood quality, poorer child health and behavior, and more parental strain (Avison, Ali, & Walters, 2007; Bzostek & Beck, 2011).

Race and ethnicity

The proportion of Black mothers in the population changed little from the 1970s to the 2000s (Hobbs & Stoops, 2002). However, due to recent increases in Latino and Asian immigrants, mothers in the early 2000s were more racially and ethnically diverse than mothers in the mid-1970s (Grieco, 2010). Blacks, Hispanics, and Asians are more likely than non-Hispanic Whites to live in houses with poorer quality (e.g., lead paint) and in poorer quality neighborhoods, which have negative consequences for child health (Jacobs, 2011; Walton, 2009). Asian and Hispanic immigrants as well as U.S.-born non-Hispanic Blacks tend to use a method of parenting that emphasizes obedience and diligence in children, and this method of parenting is related to more parental strain than the praise-and-reason method that non-Hispanic Whites tend to use (Nomaguchi & House, 2013). From these research findings, and given that the proportion of Hispanic and Asian mothers increased (Grieco, 2010), it is reasonable to expect that mothers in the early 2000s reported poorer neighborhood quality, poorer child health, and more parenting stress than mothers in the mid-1970s.

In sum, although a few changes may suggest otherwise (i.e., the increases in single mothers and non-White mothers), many more sociodemographic changes, such as the increases in education levels, labor force participation, and age at first birth, as well as the decline in family size, indicate that mothers in the early 2000s had more resources than mothers in the mid-1970s. Thus, we expect that mothers in 2002 would be more likely than mothers in 1976 to report better neighborhood quality, better child health and behavior, and less parental strain.

Changes in Parenting Culture

Cultural views of appropriate parenting methods changed dramatically in the United States between the mid-1970s and the early 2000s. Hays (1996) argued that in the 1980s, child-rearing experts began to emphasize the importance of mothers’ paying close attention to their children’s day-to-day needs to raise physically and emotionally healthy children. In this new parenting method, which Hays labeled intensive mothering, mothers are expected to learn age-appropriate developmental milestones from child development experts while understanding each child’s unique personality to carefully orchestrate their children’s daily life and development. Similarly, Rutherford (2011) found that child-rearing advice in popular magazines in the 1980s began to stress parental, especially maternal, responsibility for constantly keeping tabs on children to shield them from possible dangers. Children began to be viewed as innocent and vulnerable, and mothers are held accountable for protecting their children from any potential harm (Nelson, 2010). In short, “good” mothering in the contemporary United States requires a tremendous amount of physical, mental, and emotional involvement on the part of mothers.

Several explanations have been offered for the emergence of these new parenting norms. Demographers argue that the decline in the number of children per family led to the notion that each child is precious and that parents must invest greatly in each child to ensure his or her successful development (Blake, 1981). Others argue that increases in crime and drug use during the 1980s led to a barrage of negative news in the media and subsequently created a “culture of fear” (Glassner, 1999; Warr & Ellison, 2000). As the media has given greater attention to sex, violence, and illegal drugs as potential threats to children, childhood is no longer regarded as safe (Nelson, 2010). Meanwhile, social cohesion within neighborhoods eroded, leaving parents with fewer trusted extra eyes on which they could rely to help keep their children safe (Coleman, 1988). More recently scholars have maintained that the shifts in the economy since the 1980s, due in part to the automation of production, have escalated the pressure on parents to prepare their children to survive or thrive in the uncertain and competitive world of work (Milkie & Warner, 2014; Nelson, 2010).

Quantitative research has not adequately examined how these changes in parenting culture influenced individual mothers’ subjective experiences of parenting. Drawing on past research in related areas, we speculate the following possibilities. First, mothers’ fears regarding children’s safety in their neighborhoods might have increased. Qualitative research has shown that in the 1970s, children walked or biked to school unaccompanied by their parents and played in the neighborhood without adult supervision (Rutherford, 2011; Tulloch, 2004). In sharp contrast, mothers in the 2000s rarely allowed their children to play or hang out without adult supervision (Nelson, 2010), in part because they were concerned about their children’s safety in their neighborhoods and were often worried that their children might get into trouble with others or become involved with drugs or alcohol (Kurz, 2000). However, quantitative evidence to support these suppositions is lacking.

Second, according to the intensive mothering ideology, mothers are expected to be watchful of children’s behaviors to ensure that their children are following the age-specific social and emotional maturity and physical development trajectories laid out by experts (Hays, 1996; Rutherford, 2011). Thus, compared with mothers in the 1970s, mothers in the 2000s might have been more sensitive to any deviations from expected developmental milestones and quicker to assume physical, behavioral, or emotional problems. Certain behaviors that would have been seen as typical misbehaviors for children in the 1970s were considered to be evidence of problem behaviors in the early 2000s, such as hyperactivity or a lack of self-control (Bruchmüller, Margraf, & Schneider, 2012). These differences suggest that more mothers in the 2000s likely reported their children’s health and behavior to be poorer or more problematic than did mothers in the mid-1970s.

Finally, some researchers have linked contemporary parenting norms to increased guilt, worry, and stress among mothers (Rizzo, Schiffrin, & Liss, 2013; Sutherland, 2010; Wall, 2010). The intensive parenting ideology has pressured mothers to meet impossibly high standards of involvement in their children’s lives (Rizzo et al., 2013; Wall, 2010). In fact, despite their higher labor force participation rates, mothers in the 2000s spent more time actively engaged with their children than those in the 1970s—perhaps by multitasking and eliminating personal leisure time (Bianchi, 2011). Yet a majority of them still felt as if they were not spending enough time with their children, which negatively influenced their well-being (Nomaguchi, Milkie, & Bianchi, 2005). In sum, we expect that mothers in the early 2000s reported greater parental strain than mothers in the mid-1970s.

Variations by Education and Race/Ethnicity

Some of the changes in sociodemographic characteristics discussed in the preceding paragraphs occurred unevenly by SES. In the early 2000s, mothers without a college degree were more likely than their college-educated counterparts to have children at younger ages, more likely to have a larger number of children, and more likely to be a single mother (Bianchi, 2011; Cherlin, 2010). Such differences between those with and without a college degree were in part because of the increase in economic inequality by the education level within the United States. Since the 1980s, the globalization of the economy and the automation in manufacturing have shifted the United States from a manufacturing- to a service-based economy. This shift fostered the polarization of the labor force between those with a college degree (who tend to have high-skilled and high-paying jobs) and those without a college degree (who tend to have low-skilled and low-paying jobs; Levy, 1998). With these unequal changes, inequality in the levels of resources for parenting by college education grew (McLanahan, 2004). Thus, we expect that positive changes in mothers’ perceptions of neighborhood quality, child health and behavioral adjustments, and parental strain, as predicted by the resource perspective, are less pronounced among mothers without a college degree than mothers with a college degree. In contrast, the cultural perspective may predict different patterns of variation by SES. Some researchers have indicated that the intensive parenting ideology is mostly found among parents with a college degree (Lareau, 2003). Thus, we expect that negative changes in mothers’ perceptions of neighborhood quality, child health and behavioral adjustments, and parental strain, as predicted by the cultural perspective, are more prevalent among mothers with a college degree than mothers without a college degree.

Differences in mothers’ perceptions of parenting between the two cohorts may also depend on mothers’ race or ethnicity. The economic shift discussed earlier had a more detrimental impact on Black families than White families (Levy, 1998). In the early 2000s, Black mothers were more likely than White mothers to have more children and be single mothers (Hummer & Hamilton, 2010), and the growth of family income was lower among Black families than White families (Aud, Fox, & KewalRamani, 2010). Thus, we expect that positive changes in mothers’ perceptions of neighborhood quality, child health and behavioral adjustments, and parental strain, as predicted by the resource perspective, are less pronounced among Black mothers than White mothers. For the cultural perspective, whether there were racial or ethnic differences in the degree to which mothers were affected by the intensive mothering ideology has been debated. Some researchers emphasized little difference between White and Black mothers (Hays, 1996; Lareau, 2003), whereas other researchers argued that the intensive mothering was found more among White mothers than non-White mothers (Dow, 2016). Thus, the cultural perspective does not align with particular predictions on racial or ethnic differences between cohorts with regard to mothers’ subjective experiences of child-rearing.

The Present Study

The social contexts in which mothers raised their children were different in many ways between the mid-1970s and the early 2000s. However, scant research has examined how mothers’ subjective experiences of child-rearing differed between these time points. To fill this gap, we used national samples of mothers with 6- to 12-year-old children drawn from the 1976 NSC and the 2002 PSID-CDS, the best available data sets to examine differences in mothers’ perceptions of neighborhood quality, child health, child behavioral problems, and parental strain.

Two perspectives produced contracting predictions. First, the resource perspective contends that changes in overall standards of living and sociodemographic characteristics of mothers suggest that mothers should report better neighborhood quality, better child health, fewer child behavioral problems, and lower parental strain in 2002 than mothers in 1976. To assess this perspective, we examined whether differences in mothers’ perceptions between the two cohorts would disappear once differences in mothers’ sociodemographic characteristics between the two periods were taken into account. Second, the cultural perspective predicts that changes in parenting culture should result in mothers reporting poorer neighborhood quality, poorer child health, more child behavioral problems, and more parental strain in 2002 than in 1976. Although we did not have measures to directly assess this prediction, the year 2002 variable, once we adjusted for key sociodemographic variables, may indicate the possibility that the change in parenting culture could be responsible for changes in mothers’ perceptions between the two cohorts. Finally, we examined whether differences in mothers’ perceptions of parenting experiences between the two cohorts varied according to mothers’ education level and race or ethnicity.

Method

Data

Data were drawn from two national surveys: the 1976 NSC and the 2002 PSID-CDS. The 1976 NSC was sponsored by the Foundation for Child Development and is a sample of children and their families living in households in the contiguous United States (Zill, Furstenberg, Peterson, & Moore, 1990). The children were born between September 1, 1964, and December 31, 1969, and were between 6 and 12 years of age at the time of the interview. The sample size for the 1976 NSC was n = 2,301. We selected cases where the respondents were the study child’s mothers, the majority of whom were biological mothers, although 4.2% were stepmothers or adoptive mothers (n = 2,137). Excluding nonbiological mothers did not change the findings. The present analysis focused on mothers’ subjective experiences of child-rearing, so it did not make sense to have two sets of responses from one mother. Therefore, for the 513 cases where more than one child per household was included in the sample, we randomly selected one child (n = 1,624). The 1976 NSC provided a weight variable to correct for the oversampling of Black children as well as other characteristics such as age, gender, race, and residential location so the sample would resemble families with children between 6 and 12 years of age in the general U.S. population.

The PSID is a national longitudinal study of families the University of Michigan launched in 1968. PSID families who completed the 1997 interview were recruited into the first wave (W1) of the PSID-CDS if they had at least one child less than 13 years of age in 1997 (n = 3,563). The second wave (W2), collected between October 2002 and May 2003, included 82% of children from the first wave (n = 2,907; for more information about the PSID-CDS, see http://psidonline.isr.umich.edu/studies.aspx). Between W1 and W2, 292 children were dropped because either their families were not active in the main PSID data collection in 2001 or the youngest child had turned 13 years of age and their families were therefore no longer eligible, and 364 children were dropped for a variety of reasons, such as refusing to continue their participation, unable to be located, or establishing their own residence (Institute for Social Research, 2010). Then, to be comparable to the 1976 NSC sample, we selected cases in which the children were between 6 and 12 years of age (n = 1,607) and where the respondents were the children’s biological, step, or adoptive mothers (n = 1,453). For the 612 cases in which more than one child per household were included in the sample, we randomly selected one child (n = 841). The PSID oversampled lower income households. Population weights provided by the PSID-CDS were used to adjust for this oversampling, as well as for the probability of selection into the CDS and for attrition.

In each data set, some variables had a small percentage of missing values. To deal with those missing values, we used the multiple imputation method suggested by Allison (2001) with five iterations in SAS statistical software. Combining the 1976 NSC and the 2002 PSID-CDS samples resulted in the sample size of N = 2,465. To check how close each of the two samples was to a nationally representative sample, we compared mothers’ and children’s characteristics in the 1976 NSC and the 2002 PSID-CDS with those in equivalent households (i.e., those living with at least one child between 6 and 12 years of age) in the Current Population Survey (CPS), a nationally representative sample of households in the United States sponsored jointly by the U.S. Census Bureau and the U.S. Bureau of Labor Statistics collected in 1976 and 2002, respectively (for more information about the CPS, see http://www.census.gov/programs-surveys/cps.html). We describe comparisons between the sample of the present study and a nationally representative sample in the Results section.

With the sample of children in the similar age range and common questions, the 1976 NSC and the 2002 PSID-CDS provided an opportunity to investigate cohort differences in mothers’ subjective experiences of parenting. We found no other surveys that would provide comparable measures of mothers’ subjective experiences of parenting as well as comparable sample characteristics that were collected 2 decades apart. For example, the Early Childhood Longitudinal Study, Kindergarten Class of 2010–2011 has the same question asking mothers about their parental strain as the 1976 NSC did, but the age of children in the two surveys differed. The Fragile Families and Child Well-Being Study also asked the same question about maternal feelings of being worn out, but the sample focused on a cohort of children born in 1998–1999 in 77 cities with populations greater than 200,000 residents. The 2003, 2007, and 2011–2012 National Survey of Children’s Health used a nationally representative sample of children under 18 years of age, but measures of parental strain and children’s behavioral adjustments were not comparable with those used in the 1976 NSC. Thus, the 1976 NSC and the 2002 PSID-CDS were the only two surveys we identified that gave us the opportunity to examine cohort differences in mothers’ experiences of parenting with comparable national samples of children and their mothers.

Measures

Dependent variables

Mother’s perception of neighborhood quality for raising children was measured by the question “How is this neighborhood as a place for kids to grow up in?” (1976 NSC) and “How would you rate your neighborhood as a place to raise children? (2002 PSID-CDS). Response options for each question were excellent (scored as 1), very good (2), good (3), fair (4), and poor (5); we coded the variable so that higher values indicated poorer perceived neighborhood quality. Mother’s reports of the study child’s health was measured by the question “How would you describe [child’s] health?” Response options once again ranged from excellent (1) to poor (5), with higher values indicating poorer health. Mother’s report of the study child’s behavioral adjustments was measured as the mean of four items (α = .66 for the 1976 NSC and α = .68 for the 2002 PSID-CDS): “(He/She) has difficulty concentrating,” “(He/She) is easily confused,” “(He/She) is restless or overly active,” and “(He/She) breaks things on purpose or deliberately destroys (his/her) own or another’s things.” Because the number of response categories differed for the 1976 NSC and the 2002 PSID-CDS, each item was standardized before the scale was created. Mother’s parental strain was measured by one question that was asked in slightly different ways between the two surveys. In the 1976 NSC, mothers were asked: “Many parents feel worn out from time to time with the burdens of raising a family. How often do you feel tired, worn out, or exhausted from raising your family?” Response options ranged from never (1) to all the time (5). In the 2002 PSID-CDS, mothers were asked to answer how true the following question was: “I often feel tired, worn out, or exhausted from raising a family,” with response options ranging from not at all true (1) to completely true (5). Although additional indicators of parental strain were included in each survey, only the question we used was asked in both surveys.

Independent variables

Our primary independent variable was year, which was measured as a dichotomous variable where mothers in the 2002 PSID-CDS were assigned 1 and mothers in the 1976 NSC were assigned 0. We also included seven indicators of mothers’ sociodemographic characteristics. College degree was a dichotomous variable where those with 4-year college degrees and above were assigned 1, and others were assigned 0. Age was measured in years. Those with the extreme values were assigned the value of the 95th percentile (i.e., 48 years). The number of children under age 18 in the household was measured as the total number of children under 18 years of age at the time of survey. Employment status included three categories: full-time, part-time, and not employed. Family income was a continuous variable indicating household income during the previous year. In the NSC 1976, family income was measured as 13 categories indicating total family income before taxes in 1975, ranging from less than $3,000 to $35,000 and over. To make this a continuous variable, we calculated the median of each income category. Using the inflation calculator provided by the U.S. Bureau of Labor Statistics (2016), we adjusted the household income in 1975 by multiplying it by 3.29182. Marital status was measured as a dichotomous variable; unmarried respondents were assigned 1 and married respondents were assigned 0. Race was measured by three dummy variables, including White (reference), Black, and another race.

Control variable

Because age of children might be related to mothers’ perceptions of neighborhood quality, child health, child behavioral adjustments, and the strain of raising children (Nomaguchi & Brown, 2011), we controlled for child age, measured in years, to account for differences in the distribution of children by age between the two samples.

Analytic Plan

First we examined descriptive statistics for all variables for the total sample and for two cohorts of mothers separately. We used ordinary least squares (OLS) regression models to examine whether differences in means across cohorts were statistically significant. For multivariate analyses, we used OLS regression models. Note that we also conducted ordered logit models for the dependent variables that were ordered variables (i.e., neighborhood quality, health, parental strain) but do not present the results because the patterns of findings were very similar to those using OLS. Five models were examined for each dependent variable. Model 1 was the baseline model, which showed differences in mothers’ perceptions between 1976 and 2002 at the bivariate level. In Model 2, we added mothers’ education, age, the number of children in the household, employment status, and family income to examine differences in mothers’ perceptions between 1976 and 2002 when sociodemographic changes that indicate more resources in 2002 than in 1976 were controlled for. In Model 3, we added marital status and race/ethnicity to examine differences in mothers’ perceptions between 1976 and 2002 when sociodemographic changes that indicate fewer resources in 2002 than in 1976 were controlled for. In Model 4, we included all sociodemographic variables. These models were conducted to examine whether differences in mothers’ subjective experiences of child-rearing between 1976 and 2002 would remain statistically different once differences in sociodemographic characteristics between the two cohorts were controlled for. In Model 5, we included interaction terms between year and college degree and between year and race/ethnicity to examine whether differences by year varied by SES or race/ethnicity, respectively.

Results

Table 1 presents weighted means for all variables in the present analyses for the total sample and by year. We also presented weighted means of corresponding mothers from the 1976 CPS and the 2002 CPS for comparison purposes. Differences in mothers’ characteristics between the two cohorts in the present sample and those in the CPS sample were very similar except for the following characteristics. There were no statistical differences in the mean age of mothers and the percent of unmarried mothers between the two cohorts in the present sample, whereas mothers in 2002 were slightly older and more likely to be unmarried than mothers in 1976 in the CPS sample. There were a lower proportion of Black mothers in 2002 than in 1976 in the present sample, whereas there were a slightly greater proportion of Black mothers in 2002 than in 1976 in the CPS sample. Finally, the gap in the proportion of mothers of “other” races between the two cohorts was greater in the present sample than in the CPS sample. These characteristics were controlled for in the full models of the multivariate analyses, and as we will see in the Results section, key findings did not seem to be affected by these differences in the present sample from the nationally representative sample. Nonetheless, we discuss limitations of the present sample in the Discussion section.

Table 1.

Weighted Means for Variables in the Analyses for the Study Sample and the Current Population Survey in Equivalent Years for Households With Children Aged 6 to 12 and Their Mothers

Variables Range Present sample
CPS
Total sample
(N = 2,465)		 1976 NSC
(n = 1,624)		 2002 PSID-CDS
(n = 841)		 Difference 1976
(N = 15,720)		 2002
(N = 24,082)		 Difference
M SD M SD M SD p d M SD M SD p d
Mother’s college degree 0–1 .15 .11 .23 <.001 0.13 .09   .25   <.001 0.16
Mother’s age 21–50 37.44 6.22 37.35 6.36 37.61 5.95 .329 0.26 35.34d 6.60 36.94   6.32 <.001 1.60
Number of children 1–12 2.69 1.33 2.87 1.42 2.35 1.04 <.001 −0.52 3.23d 1.57 2.61d 1.17 <.001 0.62
Mother’s employment
 Employed full time 0–1 .36 .28 .52 <.001 0.23 .33   .51   <.001 0.18
 Employed part time 0–1 .20 .14 .29 <.001 0.15 .21b .22c .019 0.01
 Not employed 0–1 .44 .57 .19 <.001 −0.38 .46d .27d <.001 0.19
Family incomea 0–142 52.33 33.20 48.88 27.92 58.74 40.77 <.001 9.86 53.44c 26.46 58.74   36.25 <.001 5.30
Unmarried 0–1 .20 .19 .22 .079 0.03 .12c .21   <.001 0.09
Mother’s race
 White 0–1 .72 .76 .66 <.001 −0.11 .87d .80d <.001 0.07
 Black 0–1 .19 .21 .15 <.001 −0.06 .11b .15   <.001 0.03
 Hispanic or other race 0–1 .09 .03 .19 <.001 0.16 .02   .06d <.001 0.04
Child’s age 6–12 9.19 1.76 9.01 1.57 9.53 2.02 <.001 0.53 9.11   2.02 9.05   1.97 .006 0.06
Mother’s perceptions
 Neighborhood poor quality 1–5 2.33 1.16 2.43 1.18 2.14 1.10 <.001 −0.29
 Child poor health 1–5 1.66 0.84 1.71 0.86 1.56 0.79 <.001 −0.15
 Child problem behavior −1.12 to 4.00 −0.01 0.70 −0.02 0.69 0.01 0.72 .360 0.03
 Mother’s parental strain 1–5 2.56 1.04 2.50 0.87 2.68 1.31 <.001 0.18
Open in a new tab

Note. CPS = Current Population Surveys; NSC = National Survey of Children; PSID-CDS = Child Development Supplement of the Panel Study of Income Dynamics. aIn thousands of dollars, in 2002 values. Differences in means between the 1976 NSC and the 1976 CPS or between the 2002 PSID-CDS and the 2002 CPS were statistically significant at the bp < .05, cp < .01, and dp < .001 levels.

Focusing on the sample of the present analyses (Table 1), differences in mothers’ sociodemographic characteristics between the two cohorts are worthy of note. Mothers in 2002 were more likely than mothers in 1976 to have a college degree (11% in 1976 vs. 23% in 2002). There was little difference in the mean age of mothers between 1976 and 2002, but the mean number of children in the household was lower in 2002 than in 1976 (2.87 in 1976 vs. 2.35 in 2002). Mothers in 2002 were less likely to be out of the labor force than mothers in 1976 (57% in 1976 vs. 19% in 2002). The mean family income increased from $48,880 in 1976 (adjusted to 2002 dollars) to $58,740 in 2002. Because the mean is sensitive to extreme values, we examined differences in median family income and found that the median family income also increased from $44,440 (adjusted to 2002 dollars) in 1976 to $50,000 in 2002 (not shown). The percent of mothers who were unmarried was not statistically different between 1976 and 2002 (19% in 1976 vs. 22% in 2002). There were more Whites and Blacks in the 1976 sample than in the 2002, whereas the share of “other race” increased from 3% in 1976 to 19% in 2002.

Mothers’ subjective experiences of child-rearing differed between the two cohorts in three out of the four measures examined in the present analysis. The mean score of mothers’ perceptions of poor neighborhood quality declined from 2.43 in 1976 to 2.14 in 2002 (with a possible range of 1 to 5). The mean score of mothers’ report of the focal child’s poor health declined from 1.71 in 1976 to 1.56 in 2002. There was little difference in the mean scores of mothers’ report of the focal child’s behavioral problems between the two cohorts. The mean score of mothers’ parental strain increased from 2.50 in 1976 to 2.68 in 2002.

Neighborhood quality

We began by examining mothers’ perception of neighborhood quality for raising children (see Table 2). As a reminder, we coded the neighborhood quality variable in the way that higher values indicated poorer quality. Model 1 provided a baseline model that replicated the bivariate result shown in Table 1; that is, mothers in 2002 were less likely than mothers in 1976 to rate neighborhood quality poor. The magnitude of this difference was f = .25 (i.e., 0.29/1.16), which was classified as a medium effect according to Cohen (1988). Model 2 shows that when mothers’ education, number of children, age, employment status, and family income were controlled for, the coefficient of the year 2002 variable declined about 40% (i.e., from B = −.29 to B = −.17) but remained statistically significant. These results suggest that the increases in mothers’ advantages in these characteristics from 1976 to 2002 contributed to some but not all of the differences in mothers’ rating of neighborhood quality between 1976 and 2002. Supplemental analyses (not shown) suggested that education and family income contributed more than other variables to the change in the year coefficient. When marital status and race-ethnicity were controlled for (Model 3), the coefficient of the year variable increased about 31%. These results suggest that non-White mothers were more likely than White mothers to report poor neighborhood quality and that if the share of non-Whites among mothers in the United States had not increased, the mean score of mothers’ perceptions of poor neighborhood quality would have been even higher—that is, worse—in 2002 than in 1976. Model 4 added all the sociodemographic characteristics to the model. The year coefficient declined 7% from the year coefficient in Model 1. The positive and negative contributions of sociodemographic characteristics to differences in mothers’ neighborhood quality between the two cohorts shown in Models 2 and 3 counterbalanced each other and resulted in a smaller differences between the two cohorts. Finally, Model 5 showed that there was little variation by mothers’ education and race/ethnicity in the mothers’ perceptions of better neighborhood quality in 2002 than in 1976.

Table 2.

Ordinary Least Squares Regression Models Predicting Differences in Mothers’ Perceptions of Poor Neighborhood Quality Between 1976 and 2002 (N = 2,465)

Model 1
(R2 = .014, p = .014)
Model 2
(R2 = .120, p < .001)
Model 3
(R2 = .122, p < .001)
Model 4
(R2 = .164, p < .001)
Model 5
(R2 = .165, p < .001)
B SE p B SE p B SE p B SE p B SE p
2002 (year) −0.29 .05 <.001 −0.17 .05 <.001 −0.38 .05 <.001 −0.27 .05 <.001 −0.24 .07 <.001
2002 × Black −0.20 .13   .135
2002 × other race −0.07 .19   .700
2002 × college   0.06 .13   .655
College degree −0.19 .07   .010 −0.21 .07   .003 −0.25 .09   .006
Age −0.01 .00   .162 −0.01 .00   .036 −0.01 .00   .028
N of children   0.03 .02   .111   0.00 .02   .974 −0.00 .02   .897
Employed part time −0.05 .06   .439   0.02 .06   .721   0.02 .06   .784
Not employed −0.06 .05   .275   0.02 .05   .686   0.02 .05   .730
Family income −0.01 .00 <.001 −0.01 .00 <.001 −0.01 .00 <.001
Unmarried   0.48 .06 <.001   0.23 .07 <.001   0.23 .00 <.001
Black   0.59 .06 <.001   0.50 .06 <.001   0.55 .07 <.001
Other race   0.70 .08 <.001   0.56 .09 <.001   0.60 .15 <.001
Child’s age   0.02 .01   .157   0.00 .01   .909   0.01 .01   .342   0.01 .01   .316
Intercept   2.43 .03 <.001   2.93 .17 <.001   2.18 .12 <.001   2.76 .17 <.001   2.78 .17 <.001
Open in a new tab

Note. Omitted reference groups are year 1976, year 1976 × White, no college degree, employed full time, married, and White.

Children’s health

Next, we examined mothers’ perceptions of children’s general health (see Table 3). Mothers in 2002 were less likely than mothers in 1976 to report that their children’s general health was poor (Model 1). The effect size was small (f = .18). Model 2 shows that when mothers’ age, education, employment status, family income, and the number of children in the household were controlled for, the coefficient declined markedly and became nonsignificant. This suggests that the increases in mothers’ advantages in these characteristics, especially higher education, higher family income, and a fewer number of children, were related to mothers’ perceptions of better child health in 2002 than in 1976. In Model 3, we found that when marital status and race/ethnicity were controlled for, the coefficient of the year variable increased about 28%, suggesting that if the racial/ethnic composition of mothers in the United States had not changed, the difference in mean scores of mothers’ perception of child health would have been even larger due to a higher 2002 mean score. Model 4 controlled for all the sociodemographic characteristics. The year coefficient was statistically significant and negative, suggesting that these sociodemographic characteristics examined in the analysis did not fully explain the changes in mothers’ report of child health between 1976 and 2002. Model 5 showed that the interaction between year 2002 and “other race” was statistically significant and positive. To interpret this interaction, we calculated the predicted mean of mothers’ report of poor child health in 1976 and 2002 for White, Black, and “other race” mothers separately (data not shown). The mean score for poor child health was lower in 2002 than in 1976 among White mothers (1.62 in 1976 vs. 1.47 in 2002), whereas it was higher in 2002 than in 1976 among mothers of other races (1.50 in 1976 vs. 1.81 in 2002); there was no statistical difference among Black mothers (1.94 in 1976 vs. 1.92 in 2002).

Table 3.

Ordinary Least-Squares Regression Models Predicting Differences in Children’s Poor Health Between 1976 and 2002 (N = 2,465)

Model 1
(R2 = .007, p < .000)
Model 2
(R2 = .087, p < .001)
Model 3
(R2 = .072, p < .001)
Model 4
(R2 = .109, p < .001)
Model 5
(R2 = .114, p <.001)
B SE p B SE p B SE p B SE p B SE p
2002 (year) −0.15 .04 <.001 −0.06 .04   .100 −0.19 .04 <.001 −0.09 .04   .020 −0.15 .05   .002
2002 × Black −0.02 .11   .830
2002 × other race   0.45 .14   .001
2002 × college   0.15 .10   .128
College degree −0.14 .05   .009 −0.15 .05   .004 −0.22 .07 <.001
Age   0.01 .00   .090   0.00 .00   .184   0.00 .00   .268
N of children   0.04 .01 <.001   0.03 .01   .038   0.02 .01   .071
Employed part-time   0.03 .05   .528   0.06 .05   .159   0.06 .05   .191
Not employed −0.01 .04   .843   0.03 .04   .397   0.03 .04   .433
Family income −0.01 .00 <.001 −0.00 .00 <.001 −0.00 .00 <.001
Unmarried   0.22 .05 <.001   0.05 .05   .303   0.07 .05   .212
Black   0.39 .05 <.001   0.32 .05 <.001   0.32 .05 <.001
Other race   0.32 .06 <.001   0.22 .06 <.001 −0.11 .12   .328
Child’s age   0.01 .01   .266   0.01 .01   .297   0.01 .01   .467   0.01 .01   .453
Intercept   1.71 .02 <.001   1.62 .12 <.001   1.49 .09 <.001   1.55 .13 <.001   1.60 .13 <.001
Open in a new tab

Note. Omitted reference groups are year 1976, year 1976 × White, no college degree, employed full time, married, and White.

Children’s behavior

For mothers’ perceptions of their children’s behavioral adjustments (see Table 4), little difference was found between 1976 and 2002 at the bivariate level (Model 1). When mothers’ age, education, employment status, family income, and the number of children in the household were controlled for, the coefficient increased markedly and became statistically significant (Model 2). This suggests that if these sociodemographic conditions had not improved from 1976 to 2002, the difference in mean scores of mothers’ perceptions of child behavioral problems would have been even larger due to a higher in 2002 mean score. Model 3 suggests that controlling for marital status and race/ethnicity did not contribute much to differences in mothers’ perceptions of their children’s behavioral adjustments between 1976 and 2002. There was no difference by college degree or race/ethnicity (Model 5).

Table 4.

Ordinary Least Squares Regression Models Predicting Differences in Child Behavioral Problems Between 1976 and 2002 (N = 2,465)

Model 1
(R2 = .000, p = .360)
Model 2
(R2 = .046, p < .001)
Model 3
(R2 = .012, p < .001)
Model 4
(R2 = .047, p < .001)
Model 5
(R2 = .048, p < .001)
B SE p B SE p B SE p B SE p B SE p
2002 (year)   0.03 .03 <.001   0.08 .03   .016   0.02 .03   .581   0.07 .03   .027   0.09 .04   .038
2002 × Black −0.10 .09   .268
2002 × other race   0.01 .13   .925
2002 × college   0.02 .08   .789
College degree −0.13 .05   .005 −0.13 .05   .004 −0.15 .06   .012
Age −0.01 .00 <.001 −0.01 .00 <.001 −0.01 .00 <.001
N of children   0.03 .01   .010   0.03 .01   .013   0.03 .01   .019
Employed part time −0.01 .04   .792 −0.01 .04   .901 −0.01 .04   .848
Not employed −0.02 .03   .495 −0.02 .04   .628 −0.02 .04   .606
Family income −0.00 .00 <.001 −0.00 .00 <.001 −0.00 .00 <.001
Unmarried   0.15 .04 <.001   0.05 .04   .243   0.05 .04   .220
Black   0.07 .04   .111   0.01 .04   .804   0.04 .05   .430
Other race   0.07 .06   .291   0.00 .07   .959 −0.01 .10   .895
Child’s age   0.01 .01   .415 −0.01 .01   .540   0.01 .01   .444   0.01 .01   .425
Intercept −0.02 .02 <.001   0.37 .11 <.001 −0.02 .08   .815   0.35 .11   .001   0.35 .11   .001
Open in a new tab

Note. Omitted reference groups are year 1976, year 1976 × White, no college degree, employed full time, married, and White.

Parental strain

Finally, the results for mothers’ parental strain showed another pattern (see Table 5). Mothers in 2002 reported more parental strain than mothers in 1976 at the bivariate level (Model 1). When mothers’ age, education, employment status, family income, and the number of children in the household were controlled for, the coefficient increased (Model 2), suggesting that if there were no increases in mothers’ advantages in these characteristics, the difference in mean scores of mothers’ parental strain would have been even larger due to a higher 2002 mean score. Model 3 showed that, unlike the prediction, Black and “other race” mothers reported lower levels of parental strain than White mothers. Controlling for race/ethnicity, as well as marital status, the year coefficient increased, suggesting that if there was no decrease in the proportion of Whites from 1976 to 2002, the mean level of mothers’ parental strain would have been even higher. Further, Model 5 showed that the interactions between year 2002 and Black and between year 2002 and “other race” were statistically significant. To interpret these findings, we calculated the predicted mean of mothers’ parental strain in 1976 and 2002 for White, Black, and “other race” mothers (data not shown). The mean score of parental strain was higher in 2002 than in 1976 among White mothers (2.54 in 1976 vs. 2.95 in 2002), whereas it was lower in 2002 than in 1976 among mothers of other race (2.41 in 1976 vs. 2.29 in 2002); there was no statistical difference among Black mothers (2.31 in 1976 vs. 2.44 in 2002). We conducted the same OLS regression models for these three groups separately. The effect sizes of the year coefficient for both White mothers (f = .37) and “other race” mothers (f =.25) were medium in size (Cohen, 1988).

Table 5.

Ordinary Least Squares Regression Models Predicting Differences in Parental Strain Between 1976 and 2002 (N = 2,465)

Model 1
(R2 = .000, p = .007)
Model 2
(R2 = .027, p < .001)
Model 3
(R2 = .033, p < .001)
Model 4
(R2 = .059, p < .001)
Model 5
(R2 = .064, p < .001)
B SE p B SE p B SE p B SE p B SE p
2002 (year) 0.18 .04 <.001   0.22 .05 <.001   0.25 .05 <.001   0.28 .05 <.001   0.41 .06 <.001
2002 × Black −0.28 .13   .029
2002 × other race −0.53 .17   .002
2002 × college −0.21 .13   .110
College degree   0.06 .07   .420   0.05 .07   .479   0.14 .09   .091
Age −0.01 .00   .038 −0.01 .00   .031 −0.01 .00   .040
N of children   0.11 .02 <.001   0.13 .02 <.001   0.13 .02 <.001
Employed part time −0.03 .06   .675 −0.03 .06   .560 −0.04 .06   .541
Not employed −0.00 .05   .972 −0.01 .05   .899 −0.01 .05   .923
Family income   0.00 .00   .066   0.00 .00   .037   0.00 .00   .042
Unmarried   0.23 .06 <.001   0.32 .07 <.001   0.31 .07 <.001
Black −0.25 .06 <.001 −0.31 .06 <.001 −0.24 .07 <.001
Other race −0.49 .08 <.001 −0.50 .08 <.001 −0.13 .15   .372
Child’s age −0.01 .01   .447 −0.02 .01   .208 −0.01 .01   .496 −0.01 .01   .503
Intercept 2.50 .03 <.001   2.50 .16 <.001   2.66 .11 <.001   2.44 .16 <.001   2.39 .16 <.001
Open in a new tab

Note. Omitted reference groups are year 1976, year 1976 × White, no college degree, employed full-time, married, and White.

Discussion

In this article, we examined cohort differences in mothers’ subjective experiences of child-rearing, focusing on perceived quality of neighborhood to raise children, their children’s health and behavioral adjustments, and parental strain between 1976 and 2002. We drew on two perspectives. First, researchers and critics have contended that contemporary parenting culture in the United States, called the intensive mothering ideology (Hays, 1996), has increased mothers’ anxiety and stress over their children’s safety and proper development (Nelson, 2010; Warner, 2005). We called it the cultural perspective. Second, less often featured in the literature and public debates are the overall increases in standards of living and changes in mothers’ sociodemographic characteristics, such as having more education, higher family incomes, and fewer children (Cherlin, 2010). This perspective, which we called the resource perspective, suggests that mothers in 2002 tended to have more resources to use in the parenting role than did mothers in 1976.

Consistent with the resource perspective, we found that mothers in 2002 scored higher in their perceptions of neighborhood quality to raise children and their children’s general health than did mothers in 1976. After controlling for mothers’ sociodemographic characteristics, differences in mothers’ scores between 1976 and 2002 were smaller but remained statistically significant. Although we were unable to examine them directly, we speculate on possible reasons for these differences in mothers’ reports between the two cohorts. Several macro-level statistics suggest improvements in housing conditions and public health measures from the mid-1970s to the early 2000s, such as the decrease in child exposure to lead (Levin et al., 2008) and the expansion of public health insurance for children (Mann et al., 2003). Mothers’ health-related behaviors, which have implications for child health, also improved during the period with a decline in tobacco use and an increase in exercise (Bianchi et al., 2004). These changes in standards of living may have contributed to better neighborhood quality and better child health overall. Future research is warranted to link macro-level indicators to individual parents’ daily child-rearing experiences.

Results for mothers’ perceptions of their children’s behavioral adjustments and parental strain suggest a different story of cohort changes. Despite the sociodemographic advantages of mothers in 2002 such as fewer children, higher educational attainment, and higher family income, there was little difference in the mean scores of mothers’ perceptions of their children’s behavioral adjustments between 1976 and 2002 at the descriptive level. Findings of the multivariate analyses suggest that if mothers’ sociodemographic characteristics did not improve, mothers in 2002 would have reported higher levels of their children’s behavioral problems than mothers in 1976. Furthermore, mothers in 2002 reported more parental strain than mothers in 1976 before or after controlling for mothers’ sociodemographic characteristics. These findings are perplexing from the resource perspective, which focuses on levels of resources as key determinants of parental strain. However, they may make sense from the cultural perspective, which suggests that the intensive mothering ideology, emerged during the 1980s, created unreasonably high standards of maternal involvement in children’s daily lives and maternal responsibility for how children will turn out (Hays, 1996; Nelson, 2010).

There has been a debate as to whether SES differences in parenting experiences have widened in recent decades. Some scholars have emphasized increasing economic disadvantages of lower SES parents (e.g., McLanahan, 2004); other scholars have argued that a sense of insecurity has increased among higher SES parents (e.g., Milkie & Warner, 2014). For parenting norms too, scholars have disagreed with one another as to whether lower SES parents are subject to the psychological burden of the intensive parenting ideology to the same degree as higher SES parents (Nomaguchi, Milkie, & Denny, 2016). We found little SES variation in cohort differences in mothers’ perceptions of neighborhood quality, child health, and behavioral problems of their children, or parental strain. It is possible that reasons for the increase in parental strain from 1976 to 2002 differed by SES. We were unable to examine this possibility, however.

We found racial/ethnic differences: The higher level of parental strain among mothers in 2002 than mothers in 1976 was found only among White mothers. Black mothers had similar patterns as White mothers did in cohort differences in their perceptions of neighborhood quality, child health, and behavioral problems of their children, but, unlike their White counterparts, they showed no increase in parental strain. One possible explanation is racial differences in norms of the ideal motherhood. Using in-depth interviews, Dow (2016) found that Black mothers held what she called the integrated mothering ideology, which, in sharp contrast of the intensive mothering ideology, does not vest the full responsibility of raising a child to his or her mother alone. Or, as McQuillan, Greil, Shreffler, and Tichenor (2008) found in their study on racial differences in perceived importance of motherhood among women in the United States, Blacks might be less susceptible to the mainstream ideology of motherhood than Whites. We are cautious about making a strong conclusion concerning our findings for mothers of “other races” because the sample size for this group of mothers in the 1976 NSC was small.

All in all, our findings suggest some evidence that mothers—especially White mothers—in recent cohorts are more likely than mothers in the 1970s to report feeling overwhelmed with the parenting role, even though they perceive better neighborhood quality and better child health than mothers in the 1970s did. Although the media tend to focus on stories of middle-class mothers, our findings suggest that the contemporary stressed-moms phenomenon is evident regardless of SES. Practitioners and policymakers should pay more attention to the psychological burden on individual mothers of the current neoliberal emphasis on personal responsibility for child-rearing. As found in qualitative research (e.g., Elliott & Bowen, 2018; Hays, 1996), mothers in the current intensive mothering era have been living with the constant fear of being labeled as bad mothers, contrasting their practices with those that child-rearing experts recommend. It is important for practitioners to consider ways to address the negative consequences of the exclusively child-centered parenting advice that experts tend to give for the well-being of individual mothers.

Limitations

We acknowledge that the present analysis has limitations. First, three of the four dependent variables were measured based on one question. This is mostly because of limitations of the 1976 NSC. Still, the 1976 NSC was the only available national data set collected in the 1970s that could be used for the present analysis. Second, the two cohorts in the present sample had slightly more disadvantaged mothers—more Black mothers in the 1976 NSC and more mothers of other races in the 2002 PSID-CDC—than those in a nationally representative sample. The key findings discussed here do not seem to be merely artifacts of these gaps in the present sample from the representative sample. Nonetheless, using nationally representative samples would be ideal to compare cohort differences in mothers’ experiences. Third, our sample focused on mothers with children between 6 and 12 years of age. It is possible that the degree in which mothers are concerned about the safety and the well-being of children as well as overall levels of parental strain may depend on the age of children. Fourth, our study focused on mothers due to data limitations; it would be desirable to know differences in fathers’ experiences across cohorts as well. Lastly, some past evidence has shown that economic and cultural climates surrounding child-rearing in the United States have changed further in the 2000s toward more economic challenges and a further emphasis on personal responsibility of parents for the well-being of children (Brooks-Gunn, Schneider, & Waldfogel, 2013; Elliott & Bowen, 2018). Thus, we encourage survey researchers to examine cohort differences in mothers’ and fathers’ perceptions of child-rearing experiences by including comparable questions in future national data collections.

Conclusion

Changes in social and cultural contexts of motherhood and their consequences for the well-being of individual mothers have been debated within and outside of academia. Using the best available data sets, we show that mothers’ subjective experiences of child-rearing differed between 1976 and 2002. In part reflecting sociodemographic advantages of mothers in 2002 relative to 1976, mothers in 2002 perceived better neighborhood quality and better child health than mothers in 1976. Yet, albeit Whites only, mothers in 2002 reported more parental strain than mothers in 1976. These results are consistent with the media portrayal of the stressed-moms phenomenon in the contemporary United States in that motherhood is experienced as more stressful today than it was in the pre–intensive mothering era.

Acknowledgments

Author Note

This research was supported by the Center for Family and Demographic Research, Bowling Green State University, which has core funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (P2CHD050959).

References

  1. Allison P. Missing data. Thousand Oaks, CA: Sage; 2001. [Google Scholar]
  2. Anderegg D. Worried all the time: Rediscovering the joy in parenthood in an age of anxiety. New York, NY: Free Press; 2003. [Google Scholar]
  3. Aud S, Fox MA, KewalRamani A. Status and trends in the education of racial and ethnic groups (NCES 2010-015) Washington, DC: U.S. Government Printing Office; 2010. [Google Scholar]
  4. Avison WR, Ali J, Walters D. Family structure, stress, and psychological distress: A demonstration of the impact of differential exposure. Journal of Health and Social Behavior. 2007;48:301–317. doi: 10.1177/002214650704800307. [DOI] [PubMed] [Google Scholar]
  5. Barbarin O, Bryant D, McCandies T, Burchinal M, Early D, Clifford R, Howes C. Children enrolled in public pre-K: The relation of family life, neighborhood quality, and socioeconomic resources to early competence. American Journal of Orthopsychiatry. 2006;76:265–276. doi: 10.1037/0002-9432.76.2.265. [DOI] [PubMed] [Google Scholar]
  6. Bianchi SM. Family change and time allocation in American families. The Annals of the American Academy of Political and Social Science. 2011;638:21–44. doi: 10.1177/0002716211413731. [DOI] [Google Scholar]
  7. Bianchi SM, Cohen PN, Raley S, Nomaguchi K. Inequality in parental investment in childrearing: Time, expenditures, and health. In: Necherman K, editor. Social inequality. New York, NY: Russell Sage; 2004. [Google Scholar]
  8. Blake J. Family size and the quality of children. Demography. 1981;18:421–442. doi: 10.2307/2060941. [DOI] [PubMed] [Google Scholar]
  9. Brooks-Gunn J, Schneider W, Waldfogel J. The Great Recession and the risk for child maltreatment. Child Abuse & Neglect. 2013;37:721–729. doi: 10.1016/j.chiabu.2013.08.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bruchmüller K, Margraf J, Schneider S. Is ADHD diagnosed in accord with diagnostic criteria? Overdiagnosis and influence of client gender on diagnosis. Journal of Consulting and Clinical Psychology. 2012;80:128–138. doi: 10.1037/a0026582. [DOI] [PubMed] [Google Scholar]
  11. Bzostek SH, Beck AN. Familial instability and young children’s physical health. Social Science & Medicine. 2011;73:282–292. doi: 10.1016/j.socscimed.2011.04.014. [DOI] [PubMed] [Google Scholar]
  12. Cherlin AJ. Demographic trends in the United States: A review of research in the 2000s. Journal of Marriage and Family. 2010;72:403–419. doi: 10.1111/j.1741-3737.2010.00710.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cohen J. Statistical power analysis for the behavioral sciences. 2nd. Hillsdale, NJ: Lawrence Erlbaum; 1988. [Google Scholar]
  14. Coleman JS. Social capital in the creation of human capital. American Journal of Sociology. 1988;94:S95–S120. doi: 10.1086/228943. [DOI] [Google Scholar]
  15. Datar A. The more the heavier? Family size and childhood obesity in the US. Social Science & Medicine. 2017;180:143–151. doi: 10.1016/j.socscimed.2017.03.035. [DOI] [PubMed] [Google Scholar]
  16. Dow DM. Integrated motherhood: Beyond hegemonic ideologies of motherhood. Journal of Marriage and Family. 2016;78:180–196. doi: 10.1111/jomf.12264. [DOI] [Google Scholar]
  17. Elliott S, Bowen S. Defending motherhood: Morality, responsibility, and double binds in feeding children. Journal of Marriage and Family. 2018;80:499–520. doi: 10.1111/jomf.12465. [DOI] [Google Scholar]
  18. Glassner B. The culture of fear Why Americans are afraid of the wrong things: Crime, drugs, minorities, teen moms, killer kids, mutant microbes, plane crashes, road rage, & so much more. New York, NY: Basic Books; 1999. [Google Scholar]
  19. Grieco EM. Race and Hispanic origin of the foreign-born population in the United States: 2007. Washington, DC: U.S. Census Bureau; 2010. (American Community Survey Reports, ACS-11). Retrieved from https://www.census.gov/prod/2010pubs/acs-11.pdf. [Google Scholar]
  20. Hays S. The cultural contradictions of motherhood. New Haven, CT: Yale University Press; 1996. [Google Scholar]
  21. Hobbs F, Stoops N. Demographic trends in the 20th century. U.S. Census Bureau; Washington, DC: U.S. Census Bureau; 2002. (Census 2000 Special Reports, CENSR-4). Retrieved from https://www.census.gov/prod/2002pubs/censr-4.pdf. [Google Scholar]
  22. Hummer RA, Hamilton ER. Race and ethnicity in Fragile Families. The Future of Children. 2010;20(2):113–131. doi: 10.1353/foc.2010.0003. [DOI] [PubMed] [Google Scholar]
  23. Institute for Social Research. The panel study of income dynamics child development supplement: User guide for CDS–II. Ann Arbor: Institute for Social Research, University of Michigan; 2010. Retrieved from http://psidonline.isr.umich.edu/CDS/cdsii_userGd.pdf. [Google Scholar]
  24. Jacobs DE. Environmental health disparities in housing. American Journal of Public Health. 2011;101:S115–S122. doi: 10.2105/AJPH.2010.300058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Johnson RC, Kalil A, Dunifon RE. Employment patterns of less-skilled workers: Links to children’s behavior and academic progress. Demography. 2012;49:747–772. doi: 10.1007/s13524-011-0086-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kantrowitz B, Wingert P. The parent trap. Newsweek. 2001 Jan 29;137:48–53. [PubMed] [Google Scholar]
  27. Kornrich S, Furstenberg F. Investing in children: Changes in parental spending on children, 1972–2007. Demography. 2013;50:1–23. doi: 10.1007/s13524-012-0146-4. [DOI] [PubMed] [Google Scholar]
  28. Kurz D. Work-family issues of mothers of teenage children. Qualitative Sociology. 2000;24:435–451. doi: 10.1023/A:1005526808830. [DOI] [Google Scholar]
  29. Lareau A. Unequal childhoods: Class, race, and family life. Berkeley: University of California Press; 2003. [Google Scholar]
  30. Lee C, Ryff CD. Early parenthood as a link between childhood disadvantage and adult heart problems: A gender-based approach. Social Science & Medicine. 2016;171:58–66. doi: 10.1016/j.socscimed.2016.10.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Leibowitz A, Klerman JA. Explaining changes in married mothers’ employment over time. Demography. 1995;32:365–378. doi: 10.2307/2061686. [DOI] [PubMed] [Google Scholar]
  32. Levin R, Brown MJ, Kashtock ME, Jacobs DE, Whelan EA, Rodman J, Sinks T. Lead exposures in US children, 2008: Implications for prevention. Environmental Health Perspectives. 2008;116:1285–1293. doi: 10.1289/ehp.11241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Levy F. The new dollars and dreams: American incomes and economic change. New York, NY: Russell Sage; 1998. [Google Scholar]
  34. Mann C, Rowland D, Garfield R. Historical overview of children’s health care coverage. The Future of Children. 2003;13(1):31–53. doi: 10.2307/1602638. [DOI] [PubMed] [Google Scholar]
  35. Mathews TJ. Smoking during pregnancy in the 1990s. National Vital Statistics Report. 2001;49(7):1–14. [PubMed] [Google Scholar]
  36. Mathews TJ, Hamilton BE. National vital statistics reports. Vol. 51. Hyattsville, MD: National Center for Health Statistics; 2002. Mean age of mother, 1970–2000; p. 1. Retrieved from https://www.cdc.gov/nchs/data/nvsr/nvsr51/nvsr51_01.pdf. [PubMed] [Google Scholar]
  37. Mathews TJ, Hamilton BE. Delayed childbearing: More women are having their first child later in life. Hyattsville, MD: National Center for Health Statistics; 2009. (NCHS Data Brief No. 21). Retrieved from https://www.cdc.gov/nchs/data/databriefs/db21.pdf. [PubMed] [Google Scholar]
  38. McLanahan S. Diverging destinies: How children are faring under the second demographic transition. Demography. 2004;41:607–627. doi: 10.1353/dem.2004.0033. [DOI] [PubMed] [Google Scholar]
  39. McQuillan J, Greil AL, Shreffler KM, Tichenor V. The importance of motherhood among women in the contemporary United States. Gender & Society. 2008;22:477–496. doi: 10.1177/0891243208319359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Milkie MA, Warner CH. Status safeguarding: Mothers’ work to secure children’s place in the status hierarchy. In: Ennis L, editor. Intensive mothering: The cultural contradictions of modern motherhood. Bradford, Canada: Demeter Press; 2014. pp. 66–85. [Google Scholar]
  41. Monte LM, Ellis RR. Fertility of women in the United States: 2012. Washington, DC: U.S. Census Bureau; 2014. (Current Population Reports, P20-575). Retrieved from http://www.census.gov/content/dam/Census/library/publications/2014/demo/p20-575.pdf. [Google Scholar]
  42. Mosisa A, Hipple S. Trends in labor force participation in the United States. Monthly Labor Review. 2006;129:35–57. [Google Scholar]
  43. Nelson MK. Parenting out of control: Anxious parents in uncertain times. New York, NY: New York University Press; 2010. [Google Scholar]
  44. Nomaguchi KM, Brown SL. Parental strains and rewards among mothers: The role of education. Journal of Marriage and Family. 2011;73:621–636. doi: 10.1111/j.1741-3737.2011.00835.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Nomaguchi K, House AN. Racial–ethnic disparities in maternal parenting stress the role of structural disadvantages and parenting values. Journal of Health and Social Behavior. 2013;54:386–404. doi: 10.1177/0022146513498511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Nomaguchi K, Milkie MA. “You must work hard”: Changes in US adults’ values for children, 1986–2014; Paper presented at the 2016 Annual Meeting of the American Sociological Association; Seattle, WA. 2016. Aug, [Google Scholar]
  47. Nomaguchi K, Milkie MA. Sociological perspectives on parenting stress: How social structure and culture shape parental strain and the well-being of parents and children. In: Deater-Deckard K, Panneton R, editors. Parenting stress and early child development: Adaptive and maladaptive outcomes. 2017. pp. 47–73. [DOI] [Google Scholar]
  48. Nomaguchi KM, Milkie MA, Bianchi SM. Time strains and psychological well-being: Do dual-earner mothers and fathers differ? Journal of Family Issues. 2005;26:756–792. doi: 10.1177/0192513x05277524. [DOI] [Google Scholar]
  49. Nomaguchi K, Milkie MA, Denny KE. Quantity of maternal time and child and adolescent development: Response to Kalil and Mayer (2016) and to Waldfogel (2016) Journal of Marriage and Family. 2016;78:270–275. doi: 10.1111/jomf.12258. [DOI] [Google Scholar]
  50. Pearlin LI. The sociological study of stress. Journal of Health and Social Behavior. 1989;30:241–256. doi: 10.2307/2136956. [DOI] [PubMed] [Google Scholar]
  51. Rizzo KM, Schiffrin HH, Liss M. Insight into the parenthood paradox: Mental health outcomes of intensive mothering. Journal of Child and Family Studies. 2013;22:614–620. doi: 10.1007/s10826-012-9615-z. [DOI] [Google Scholar]
  52. Rutherford MB. Adult supervision required: Private freedom and public constraints for parents and children. New Brunswick, NJ: Rutgers University Press; 2011. [Google Scholar]
  53. Sayer LC, Bianchi SM, Robinson JP. Are parents investing less in children? Trends in mothers’ and fathers’ time with children. American Journal of Sociology. 2004;110:1–43. doi: 10.1086/386270. [DOI] [Google Scholar]
  54. Singh GK, Kogan MD. Persistent socioeconomic disparities in infant, neonatal, and postneonatal mortality rates in the United States, 1969–2001. Pediatrics. 2007;119:e928–e939. doi: 10.1542/peds.2005-2181. [DOI] [PubMed] [Google Scholar]
  55. Stearns PN. Anxious parents: A history of modern childrearing in America. New York, NY: New York University Press; 2003. [Google Scholar]
  56. Sutherland JA. Mothering, guilt and shame. Sociology Compass. 2010;4:310–321. doi: 10.1111/j.1751-9020.2010.00283.x. [DOI] [Google Scholar]
  57. Tulloch M. Parental fear of crime: A discursive analysis. Journal of Sociology. 2004;40:363–377. doi: 10.1177/1440783304048380. [DOI] [Google Scholar]
  58. U.S. Bureau of Labor Statistics. Inflation calculator. 2016 Retrieved from http://www.bls.gov/data/inflation_calculator.htm.
  59. U.S. Census Bureau. Table 1. Years of school completed by persons 14 years old and over, by age, race, Spanish origin, and sex: March 1974 and March 1973. 1974 Retrieved from https://www.census.gov/hhes/socdemo/education/data/cps/1974/tab-01.pdf.
  60. U.S. Census Bureau. Table 1: Educational attainment of the population 15 years and over, by age, sex, race, and Hispanic origin: March 2002. 2003 Retrieved from http://www.census.gov/population/socdemo/education/ppl-169/tab01.pdf.
  61. U.S. Census Bureau. Table F-9. Presence and number of related children under 18 years old—families by median and mean income. Historical income tables: families. 2016 Retrieved from http://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-families.html.
  62. Wall G. Mothers’ experiences with intensive parenting and brain development discourse. Women’s Studies International Forum. 2010;33:253–263. doi: 10.1016/j.wsif.2010.02.019. [DOI] [Google Scholar]
  63. Wallace K. Longing for the carefree parenting style of yesterday? 2014 Aug 28; Retrieved from http://www.cnn.com/2014/05/09/living/motherhood-now-vs-then-parents.
  64. Walton E. Residential segregation and birth weight among racial and ethnic minorities in the United States. Journal of Health and Social Behavior. 2009;50:427–442. doi: 10.1177/002214650905000404. [DOI] [PubMed] [Google Scholar]
  65. Warner J. Perfect madness: Motherhood in the age of anxiety. New York, NY: Riverhead Books; 2005. [Google Scholar]
  66. Warr M, Ellison CG. Rethinking social reactions to crime: Personal and altruistic fear in family households. American Journal of Sociology. 2000;106:551–578. doi: 10.1086/318964. [DOI] [Google Scholar]
  67. World Bank. GDP (current US$) 2016 Retrieved from http://data.worldbank.org/indicator/NY.GDP.MKTP.CD?locations=US.
  68. Zill N, Furstenberg F, Peterson J, Moore K. National Survey of Children: Wave I, 1976, Wave II, 1981, and Wave III, 1987 (ICPSR08670-v3) Ann Arbor, MI: Inter-University Consortium for Political and Social Research; 1990. [Google Scholar]

RESOURCES