Family structure and children’s cognitive development

View author details View Less
  • 1 University at Buffalo, , USA
  • | 2 DePaul University, , USA
  • | 3 Wake Forest University, , USA
  • | 4 Central University of Finance and Economics, , China
Full Access
Get eTOC alerts
Rights and permissions Cite this article

Previous research on family structure and child development has largely focused on the disadvantages faced by children who transitioned out of married families. However, we know less about how family structure affects child outcomes for children starting out in single-mother families. In this article, we use the kindergarten cohort of the Early Childhood Longitudinal Study to analyse children’s academic outcomes between kindergarten and eighth grade. We found that living in single-mother or step-families was clearly associated with lower test scores for children starting kindergarten in married biological-parent families, but the same disadvantages associated with living outside a married biological-parent family structure were not found for children starting kindergarten in single-mother families. We also found preliminary evidence of a buffering effect of maternal education in the relationship between family structure and children’s academic outcomes.

Abstract

Previous research on family structure and child development has largely focused on the disadvantages faced by children who transitioned out of married families. However, we know less about how family structure affects child outcomes for children starting out in single-mother families. In this article, we use the kindergarten cohort of the Early Childhood Longitudinal Study to analyse children’s academic outcomes between kindergarten and eighth grade. We found that living in single-mother or step-families was clearly associated with lower test scores for children starting kindergarten in married biological-parent families, but the same disadvantages associated with living outside a married biological-parent family structure were not found for children starting kindergarten in single-mother families. We also found preliminary evidence of a buffering effect of maternal education in the relationship between family structure and children’s academic outcomes.

Introduction

One of the most notable changes in family life over the last several decades is the increasing prevalence of cohabitation and never-married single-parent families, and more broadly, what some have characterised as a ‘retreat’ from marriage. Since the 1960s, the divorce rate has increased, non-marital childbearing has become more common, and cohabitation has become increasingly normative. Indeed, the percentage of people in the US who report having ever cohabited has increased from 49 per cent in 1995 to 73 per cent in 2011–13 (Lamidi and Manning, 2016). The majority of children today are still living with married biological parents (according to the US Census Bureau’s most recent report, approximately 70 per cent in 2018 as compared to 85 per cent in 1968); however, an increasing number of children in the US and UK are living in single-parent, step-family or cohabiting households and experiencing greater family instability (Fomby, 2011; US Census Bureau, 2018). In the US, the proportion of children living in single-parent families has grown from 12 per cent in 1970 to 27 per cent in 2018 (US Census Bureau, 2018), the vast majority of which are mother-only families (Eickmeyer, 2017). Families have become increasingly diverse, and as a result, children’s experiences in families have diversified as well.

Overall, there is evidence in the US and UK that social class and family structure – that is, socioeconomic status (SES) – are closely connected, such that lower-SES children are more often in single-parent or cohabiting households (Kiernan and Mensah, 2011; Smock and Schwartz, 2020). Even when starting off in cohabiting relationships, higher-SES women are more likely to transition to marriage than lower-SES women (Sassler and Lichter, 2020). McLanahan’s (2004) presidential address to the Population Association of America pointed to the ‘diverging destinies’ of American families. She argued that families have come to follow one of two trajectories: one marked by delayed childbearing and increased maternal employment; the other by higher rates of divorce and nonmarital childbearing. Because it is mothers with the most limited resources and economic opportunities who have followed the second trajectory marked by family instability, children in these families are more likely to be disadvantaged in their educational and behavioural outcomes compared with their peers in families following the first trajectory.

Many scholars have investigated how cohabitation, single-parenthood and parenting in the presence of a step-parent are associated with child outcomes (Manning and Lamb, 2003; Kroeger and Smock, 2014). Cross-sectional studies generally find that children in cohabiting and single-parent families fare worse academically and behaviourally in comparison with children in stepfamilies or intact married families (Manning and Lamb, 2003), although some studies suggest that these differences disappear when controlling for SES, parenting practices, and other parent and child characteristics (Brown, 2004; Artis, 2007). While cross-sectional analyses provide detailed comparisons across children living in different family types, these studies suffer from the limitation of not being able to completely control for all of the differentiating factors in the lives of children in these diverse family structures. Several studies of the effects of family structure on children have analysed panel data (for example, Craigie et al, 2012; Amato and Anthony, 2014; Bzostek and Berger, 2017; Fomby and Osborne, 2017),but many of these studies are limited by an inability to account for unmeasured family background characteristics or stable individual child characteristics (for example, in personality, academic aptitude and so on), which may bias estimates of the effects of family structure. Other studies only analyse children starting off in married biological parent-families, leaving out the experiences of children who spend their first years of life in single-parent, cohabiting or step-families.

Using longitudinal data from the kindergarten cohort of the Early Childhood Longitudinal Study, we analyse the association between family structure and children’s academic outcomes between kindergarten and eighth grade. Our primary interest is in understanding the diversity of experiences among children living in different family types. In order to do this, we analyse the effects of family structure on child outcomes longitudinally, with attention to the earlier family contexts in which the children lived. We also examine how a mother’s education, specifically having a college-educated mother, may serve as a protective factor on children’s academic outcomes. We examine academic outcomes because of the importance of early school performance for later-life educational and occupational achievement (Entwisle and Alexander, 1993).

Given the empirical evidence of an academic gap between children in married and in cohabiting or single-parent households (Manning and Lamb, 2003), we set out to investigate the extent to which the association between family structure and academic outcomes varies depending on a child’s family structure earlier in life. We ask: (1) how is a child’s academic performance associated with family structure? (2) To what extent does the association between family structure and children’s academic outcomes vary depending on the child’s family structure earlier in life? (3) Do differences in academic performance by family structure persist even when paternal contact is accounted for? (4) To what extent does maternal education serve as a protective factor in buffering the association between family structure and academic outcomes?

Link between family structure and child outcomes

There is a long-standing finding both in the US and the UK that a family with two married parents can provide a better environment for children than a single-parent or cohabiting family can. Previous studies have found that family structure and instability affect children’s wellbeing in many domains, including physical health, psychological health, academic achievement and social behaviour (Tillman 2007; Fomby, 2011; Schneider, 2016; Bzostek and Berger, 2017; Gray et al, 2018; Panico et al, 2019; Smock and Schwartz, 2020).

One explanation for the link between family structure and children’s academic, health and wellbeing outcomes can be traced back to Coleman’s (1988: S109) idea of social capital in the family, which suggests that parents play an essential role in assisting children’s attainment in life. According to Coleman (1988), children gain access to adult human capital through the social capital within the family, which depends on the physical presence of parents in the family and on the attention children receive from their parents. Amato (1995) further contended that single-parent families are structurally disadvantaged compared with two-parent families due to the limited access to economic and interpersonal resources they can provide for the children living in the household. Later on, McNeal (1999: 120) extended Coleman’s conceptualisation by pointing out that social capital in the family is the existence and degree of resources in various forms, including physical capital, human capital and cultural capital, which parents can invest in their children.

An alternative explanation of the relationship between family structure and children’s wellbeing and academic achievements is family-stress theory. According to this theory, the stress both on children and parents caused by parents’ marital disruption explains the behavioural problems and poorer overall academic performance at school of children in non-marital or married step-family households (see Hill, 1958; McLanahan, 1985). Stable, single-parent families could potentially lead to better outcomes for children than other family structures that result from marital disruption, according to family stress theory (Fomby and Cherlin, 2007; Fomby and Osborne, 2017).

Over the last two decades, empirical research in the US and UK has confirmed the predictions of social capital and family stress theories by showing that family structure affects children’s wellbeing and attainment through two mechanisms: economic resources (Manning and Lamb, 2003; Brown, 2004; Bradbury et al, 2012) and parental behaviours (Manning and Lamb, 2003; Kelly et al, 2011; Augustine, 2014; Fomby and Osborne, 2017; Hernández-Alava and Popli, 2017). Additionally, among different family structures, previous studies have found that a family with married biological parents enhances children’s wellbeing and cognitive and emotional development compared with other family structures, including cohabiting, married step-parent and single-parent families (for example, Manning and Lamb, 2003; Brown, 2004). Our first hypothesis is based on this rich research literature investigating the relationship between family structure and children’s outcomes.

H1: Regardless of past family structure, children who are living in married biological-parent families will have better academic outcomes compared to when living in other family structures.

Family structure history and child outcomes

Another consideration in analysing the association between family structure and child outcomes is the effects of early experiences on later outcomes from the life course perspective (Giele and Elder, 1998). From this perspective, we cannot analyse the effects of a current status on an outcome without considering the lasting impacts of earlier experiences on the present. Relatedly, the sequence of life events may also have important consequences as well. Given the growing diversity of children’s living arrangements in the US and the UK (Hadfield et al, 2018; Smock and Schwartz, 2020), some scholars have argued that we cannot understand the relationship between the current family structure in which a child is living and their academic and behavioural outcomes without considering how that association might be influenced by earlier family experiences and transitions (Sweeney, 2010; Craigie, et al, 2012; Bzostek and Berger, 2017). Buehler (2020) argues that family research needs greater attention to the importance of time: that is, the difference between early childhood and middle childhood and the changing impact of family processes over time. In addition, Mooney and colleagues (2009) point to the diversity of children’s experiences within any particular family type, noting that differences in child outcomes can be greater within family types than across them. One conclusion that can be drawn from this literature is that the effect of living in a married, two-parent biological family for children’s outcomes is not fixed but rather varies depending on the child’s previous family structures and experiences.

In order to investigate children’s academic achievement outcomes from a life course perspective, longitudinal data are needed to analyse the time-varying relationship between family structure and achievement. More specifically, longitudinal studies make it possible for researchers to compare children before and after the transition from one family structure to another family structure (Amato and Anthony, 2014). Moreover, studies focusing on one point in time tend to exaggerate the negative effect of family structure (McLanahan and Percheski, 2008). Generally, these studies conclude that residing in a ‘non-traditional’ household has a negative effect on child development. In both the US and UK, scholars have found that children who live in married biological-parent families in early life enjoy better outcomes later on (Klausli and Owen, 2009; Crawford et al, 2012).

Our second hypothesis is based on both the life course literature which contextualises family structure in broader trajectories of family change and the predictions of social capital theory. Single-parent, married step-parent and cohabiting families only imply a reduction of access to capital (social, human and economic) and/or an increase in exposure to stress when children have previously lived in a married biological-parent family. For children previously in single-mother families, these family structures represent either no change (that is, for children in single-parent families) or an increase in access to capital (that is, for children in married step-parent and cohabiting families). Based on this logic, we propose the following hypothesis.

H1A: Accounting for past family structure, the positive effects of living in a married biological-parent household on academic outcomes will be found only for children who started kindergarten in a married biological-parent family. For children starting kindergarten in single-parent families, the effects of living in a married step-parent, cohabiting or single-parent family will not be as disadvantageous for academic outcomes.

As an alternative to hypothesis H1A, it is also possible that any change in family structure will result in a negative effect on children’s academic outcomes, based on the predictions of family stress theory. Based on this logic, we propose the following alternative hypothesis.

H1B: Any new family structure will be associated with a negative impact on academic outcomes, when compared with academic outcomes in kindergarten.

We also predict, based on social capital in the family perspectives, that some of the negative effects of living in a single-parent, cohabiting or married step-family on children’s reading and mathematics test scores can be explained by the lower level of contact children have with their biological fathers in these families:

H2: Contact with the child’s biological father will partly explain the negative association between living in a step-family, single-mother or cohabiting family and academic outcomes.

Maternal education and child outcomes across family structures

The benefits of high levels of maternal education for child academic outcomes are widely supported in the literature in the US and the UK (see Stevenson and Baker 1987; Magnuson et al, 2009; Bradbury et al, 2012; Hernández-Alava and Popli, 2017; Jackson et al, 2017). Mothers who are more highly educated provide greater access to social capital (including network connections to obtain information about tutoring and school options), human capital and cultural capital, which leads to the development of a set of behaviours that are valued by teachers and administrators at school (Harding et al, 2015). Highly educated mothers also are more likely to have parenting styles that help children develop academic skills, such as language skills and reading skills. The reason why mothers play an important role in child educational outcomes is that mothers are still the main caregiver in the majority of households (Sayer, 2016). Since mothers’ education has a significant impact on child outcomes, some scholars have extended the literature on family structure and child outcomes by examining the moderating effect of mothers’ education and argued that mothers’ education may (1) protect children from the stress caused by unstable family structure circumstances, and (2) provide access to capital that buffers the effect of family instability. Based on the research literature examining the buffering role played by maternal education (for example, Augustine, 2014; Jackson et al, 2017), we make the following prediction.

H3: Maternal education will moderate the effect of family structure on children’s academic outcomes, with children who have college-educated mothers scoring higher in both mathematics and reading than children whose mothers have less than a college degree, across all family structures.

Data and method

Data

Building on the rich research literature investigating the links between family structure and children’s outcomes in the US and UK, we set out to investigate this association in US families. Data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998–99 (ECLS-K) are used to examine cohabitation, family instability and children’s academic outcomes. The ECLS-K sample consists of nearly 20,000 children and is representative of the roughly four million students enrolled in kindergarten in the US during the fall of 1998 (Tourangeau et al, 2009). These data were collected by the National Center for Education Statistics in order to examine a variety of factors affecting children’s academic performance on entry to school. The ECLS-K data are longitudinal, and data are available from kindergarten to eighth grade. In each wave, children were assessed directly on reading and mathematics. In addition, their parents were interviewed at each point in time, providing more information about family life and context.

Schooling is compulsory for all children in the US, but the age range for which school attendance is required varies from state to state. After pre-kindergarten and kindergarten, there are typically five years in elementary school, and the student will enter middle school (or junior high school) after completing fifth grade, and then high school. Most children begin elementary education at kindergarten (usually when 5 to 6 years old) and finish high school at twelfth grade (usually when 18 years old). We draw on data from the fall and spring of kindergarten, fall and spring of first grade, spring of third grade, spring of fifth grade and spring of eighth grade. To be included in the study sample, the focal child must have resided in a married biological-parent family or single-mother family at wave 1. We estimated all models using sampling weights to adjust for children’s unequal probabilities of selection. Based on these criteria, our sample includes children who resided in a two-parent, married biological family (N=5,716) and those who resided in a single-mother family (N=601) in kindergarten.1

The base year (kindergarten) weighted response rate in the ECLS-K was 65.1 per cent for the child assessment and 62.1 per cent for the parent questionnaire (Tourangeau et al, 2009). As with most longitudinal surveys, there was attrition during subsequent rounds of data collection, which could lead to non-response bias. However, appropriate use of longitudinal weights should minimise bias. Other researchers have found that there are few differences between the characteristics of responders and non-responders in the ECLS-K between kindergarten and first grade (Denton and West, 2002). In our analysis, we found that attrition was not associated with the child’s performance on the reading and mathematics assessments, their gender, disability status, race, family income, or the family trajectory experienced between kindergarten and eighth grade.

Measures

Reading and mathematics scores

Our dependent variables are the reading and mathematics cognitive assessments. For our primary analyses, we use the reading and mathematics achievement test scores from each wave as measures of academic outcomes (see also Amato and Anthony, 2014). These assessments are individually administered two-stage adaptive tests. The specific items in the assessment are proprietary but are developmentally appropriate. The reading assessment focuses on basic skills such as print familiarity, letter recognition, beginning and ending sounds, recognition of common words (sight vocabulary), vocabulary knowledge, and reading comprehension. The mathematics assessment includes questions measuring conceptual knowledge, procedural knowledge, and problem solving.

Child characteristics

Age indicates the child’s age in months and Age-squared is also included in our models to account for the curvilinear relationship found between child age and academic test scores in initial descriptive analyses. Age is centred around age 5, the average age of children in the first wave.

Family structure

We include time-varying measures of family structure in our models. Children are coded as living in a married family with both biological parents, a married family with a biological mother and a step-parent, a cohabiting family (which includes their biological mother and either their biological father or their mother’s partner) or a single-mother family. In models 1, 1A and 1B, married biological-parent families are the reference category. In models 2, 2A and 2B, the reference category is single-mother families. Table 1 reports descriptive statistics for the time-varying independent variables by wave.

Table 1:

Weighted means and standard errors for time-varying variables, by wave

W1W4W5W6W7
Fall-KSpring-1stSpring-3rdSpring-5thSpring-8th
N6,3175,7905,5395,4294,847
ReadingM37.2983.12135.52157.12176.58
SE0.290.710.760.710.71
MathsM28.5765.99106.7130.06146.85
SE0.240.480.640.630.54
Child age (centred around age 5, in months)M8.6427.0550.4973.67109.75
SE0.110.110.120.120.12
SingleM0.120.110.120.140.15
CohabitingM00.0010.0020.0070.01
Married – bio-parentsM0.890.880.870.840.81
Married – step-parentM00.0030.010.020.02
Time since transition (months)M63.9215.9937.8856.8889.91
SE0.420.450.490.670.76
Paternal contactM26.3426.0625.7125.32
SE0.180.180.210.23
Number of siblingsM1.441.481.511.511.44
SE0.030.030.030.030.03
Family incomeM51.4854.9761.4667.7377.49
SE1.240.971.071.181.32

Note: Paternal contact was not available at wave 4.

We also include a control for the length of time (measured in months) since the last family structure transition, or time since transition. This measure was coded at wave 1 to measure the time since the last family structure transition prior to kindergarten. For children living in single-mother households at kindergarten, we coded time since transition in wave 1 as the number of months the mother had been single prior to wave 1 or as the child’s age if the child had continuously lived in a single-mother household. Time since transition-squared is included in our models to account for the curvilinear relationship found between time since transition and academic test scores in initial descriptive analyses.

Family characteristics

We include measures of both the number of siblings and amount of paternal contact. Number of siblings is a continuous variable measuring the number of siblings the child has. Siblings potentially may create additional demands on parental resources, reducing the resources available to the target child. Paternal contact measures how often the child saw their biological father in the past four weeks. Children living with their biological father are coded at the maximum (that is, 28 times in the past four weeks) for this variable. This variable is missing at wave 4 but present for the other waves of data.

Socioeconomic resources

Socioeconomic resources were measured by family income and mother’s college education. Family income measures household income, adjusted for inflation and divided by 1,000. The ECLS-K collected various measures related to SES, including broad-range and detailed-range income questions, in the spring-first grade wave of data collection. For the entire base sample, 10.5 per cent of detailed income range measures were missing (N=1,672) in spring of kindergarten. A two-stage procedure was used to impute income. First, if a previous parent interview had been conducted in fall-first grade or kindergarten, missing values for detailed income range variables were replaced with previous values (N=1,593). Second, data still missing after this step were imputed using hot-deck methodology (N=79). The same procedure was used to impute income in later waves. Hot-deck imputation methodology sorts previous rounds of key predictors for income in imputing missing values of income (see Tourangeau et al, 2009: 7–25). More specifically, educational level, occupation and labour force status of both the mother and father, when available, were used to impute the income variable. Mother’s college education indicates that the child’s mother has earned a college degree or higher. In our sample, 2,210 of 6,317 mothers (35 per cent) had completed a college education at wave 1. In fixed-effects models, time-invariant predictors cannot be entered directly as predictors and so maternal education is included in our models only in interaction with the time-varying measure of family structure.

Analytic approach

Our longitudinal study design allows us to compare children with themselves over time. Instead of comparing the academic outcomes of children in different family types, we compare children’s performance on reading and mathematics assessments in a particular family structure with their performance at other points in their life in which they were living in a different family structure. In our sample, children were living in single-mother or married biological-parent families in kindergarten. We chose kindergarten family structure as our measure of earlier family structure because our analysis is of academic outcomes; kindergarten, as the starting point of schooling, serves as a natural reference point.2

Our research objective is to identify how family structure is associated with children’s academic outcomes and to further investigate the claim that two-parent married biological families are the ideal context in which to raise children (Tillman, 2007; Schneider, 2016; Bzostek and Berger, 2017). Estimates of the relationship between children’s average academic outcomes and family structure may be spurious in random-effects models, reflecting other confounding factors that are not controlled for in the model. It is also possible that the association between family structure and academic outcomes may be due at least in part to selection effects. In other words, perhaps observable differences in child academic outcomes are not explained by family structure per se, but instead by the factors that pre-date union formation and childbearing. From this perspective, for example, it may be that people with fewer resources, who struggle with depression, or who have lower levels of commitment to relationships more generally may be less likely to marry and also less likely to foster an environment for children that is conducive to learning. Prior research suggests that selection effects are responsible for part, but not all, of the association between cohabitation and child outcomes (Osborne and McLanahan, 2007).

In order to control for all stable individual characteristics, both observed and unobserved, we estimate fixed-effects models (Allison, 2005). While other scholars have also used child or family fixed-effects models (see, for example, Amato and Anthony, 2014), we use child fixed-effects models in an analysis that stratifies the sample based on family structure in kindergarten to assess how previous family structure shapes the effects of subsequent family structures on child outcomes: first, by estimating our models in a sample of children starting kindergarten in married biological-parent families, and second, in a sample of children starting in single-mother families. In the fixed-effects models, only variation within an individual, over time, is used to estimate the model coefficients. This means that the coefficients for different family structure indicators should be interpreted as the association between a particular family structure and the child’s test score, among those who experienced a change in family structure during the analysis window. Those children who remain in the same family structure throughout the study period do not contribute to the coefficients for family structure in the fixed-effects models. Of the 6,317 kindergarten students in wave 1, 703 (11.13 per cent) experienced a change in family structure sometime between kindergarten and eighth grade.

Results

In model 1, we estimated a fixed-effects model which estimates the association between family structure and children’s academic outcomes, net of all stable individual and family characteristics, in a sample of children starting kindergarten in married families with both biological parents. In this model, children’s test scores were compared within-child over time. The reference category in this model was married biological-parent families, thus the comparisons in this model were between children when living in married biological-parent families versus when living in any other family types. Wald tests indicate that the fixed-effects model is a better fit than a random-effects model in both the models for reading and mathematics test scores (Allison, 2018).

In model 1 (sample restricted to children starting kindergarten in married biological-parent families), living in a married step-family or single-parent family was statistically significantly negatively associated with reading test scores in comparison to living in a married biological-parent family. Children in married step-parent families were disadvantaged by an equivalent of more than five months of learning, and children in single-mother families were disadvantaged by the equivalent of more than three months, compared to when living in married biological-parent families. (The coefficient for cohabiting families did not reach statistical significance.) In the model for mathematics test scores, only the coefficient for single-mother families reached statistical significance. Children in single-mother families were disadvantaged by the equivalent of more than three months of learning compared to when in married biological-parent families. These findings for the sample of children starting kindergarten in married biological-parent families are consistent with many past studies finding that children are advantaged in married biological-parent families.

In model 2, the sample was restricted to children starting kindergarten in single-mother families and we see a different set of associations between family structure and children’s test scores. To maintain consistency with model 1, in which the reference category represents the child’s family structure in kindergarten, the reference category in this model is single-mother families. In the model for reading test scores, none of the measures of family structure reach statistical significance. In the model for mathematics test scores, children’s test scores were statistically significantly higher when living in married step-families or cohabiting families, compared to when living in single-mother families but the association between living in a married biological-parent family and mathematics test scores was statistically insignificant. Children in married step-families were advantaged by the equivalent of approximately four months of learning in mathematics and those in cohabiting families by approximately five months of learning in their mathematics test scores, compared to when living in single-mother families (and controlling for age).

The results from models 1 and 2 provide some support for hypothesis 1A: although children in married biological-parent families were found to be advantaged in their academic outcomes in the sample of children starting kindergarten in married biological-parent families, this was not found in the sample of children starting kindergarten in single-mother families (therefore failing to support hypothesis 1). Hypothesis 1B is also not supported because all new family structures were not associated with a decline in academic achievement.

In models 1a and 2a, we add a measure of contact with the child’s biological father as a test of hypothesis 2. We hypothesised that some of the differences in test scores by family structure found in models 1 and 2 would likely be explained by the amount of contact children have with their fathers. Instead, in the models for children starting kindergarten in married biological-parent families, paternal contact is not statistically significant and the negative and statistically significant coefficients for living in a married step-family or a single-parent family remain even after the addition of the paternal contact variable. In the models for children starting kindergarten in a single-mother family, paternal contact is also not statistically significant. However, the positive effect of living in a married step-family or a cohabiting family on mathematics test scores disappears once contact with father is controlled for. Additional tests of mediation using KHB analysis in Stata 16 (Karlson et al, 2011) did not show that paternal contact statistically significantly mediates the effects of living in a married step-family or cohabiting family on mathematics test scores among children starting kindergarten in single-mother families. Overall, these results fail to support the hypothesis (H2) that controlling for paternal contact would reduce the negative effects of living in a single-parent, cohabiting or married step-parent family on test scores.

Models 1, 1a, 2 and 2a present the average associations between different family structures and children’s academic outcomes. To further explore heterogeneity in these family forms, we added an interaction between family structure and maternal education in models 1b and 2b. In model 1b for reading test scores, the direct effect of living in a cohabiting, single-parent, or married step-parent family was negative and statistically significant. However, the interaction between living in a cohabiting family and maternal education was positive and statistically significant. The other interaction terms with maternal education did not reach statistical significance. This suggests that maternal education buffers the negative effect of living in a cohabiting family on children’s test scores. In the model for mathematics test scores, there was a negative direct effect of living in a single-mother family, representing the effect of living with a single mother for children with mothers with less than a college degree. However, the interaction between living with a single mother and having a college-educated mother was statistically significant and positive. This suggests that the effects of living in a single-mother family vary depending on the level of maternal education, with maternal education again serving as a buffer on the negative effects of living in a single-mother family on mathematics test scores. Other interactions between family structure and maternal education did not reach statistical significance.

In model 2b for reading test scores, none of the direct effects of living in a cohabiting, married biological-parent or married step-parent family (having previously lived in a single-mother family) were statistically significant but the interaction between living with married biological parents and maternal education was positive and statistically significant. Among children starting kindergarten in single-mother families, children whose mothers have at least a college degree were particularly advantaged in their reading test scores when subsequently living in married biological-parent families, compared to when living in single-mother families. In the model for mathematics test scores, the direct effect of subsequently living in a cohabiting family was again (as in model 2) statistically significant and positive, while the direct effect of subsequently living in a married biological-parent family was statistically significant and negative. The direct effect of living in a married step-family did not reach statistical significance. This means that among children whose mothers have less than a college degree, there is a statistically significant positive association between living in a cohabiting family and mathematics test scores and a statistically significant negative association between living in a married biological-parent family and mathematics test scores (compared to when living in a single-mother family). However, the interaction between living in a married biological-parent family and maternal education was positive and statistically significant, suggesting that the effects of living in a married biological-parent family vary depending on the level of maternal education, with only those children with college educated mothers experiencing an advantage in their mathematics test scores when living in a married biological-parent family, compared to when they were living in a single-mother family. The interactions between living in a cohabiting or married step-family and maternal education did not reach statistical significance. Given that maternal education moderates some, but not all, family structures in models 1b and 2b, these results provide qualified support for hypothesis 3.

The control variables in all models were associated with both mathematics and reading test scores in largely the same ways as in previous studies and can be found in Tables 2 and 3.

Table 2:

Estimated coefficients for fixed-effects models of mathematics and reading test scores (among children starting kindergarten in married biological-parent families)

Model 1Model 1aModel 1b
ReadingMathsReadingMathsReadingMaths
CoefS.E.CoefS.E.CoefS.E.CoefS.E.CoefS.E.CoefS.E.
Intercept17.91***(1.29)9.44***(0.88)28.04***(5.41)9.36**(3.54)29.92***(5.33)10.56**(3.38)
Level-1 variables
Child age2.68***(0.04)2.24***(0.03)2.52***(0.04)2.15***(0.03)2.52***(0.04)2.15***(0.03)
Child age-squared–0.01***(0.0004)–0.01***(0.0003)–0.01***(0.0005)–0.01***(0.0003)–0.01***(0.0005)–0.01***(0.0003)
Time since transition0.27***(0.04)0.22***(0.03)–0.08(0.06)0.15***(0.04)–0.08(0.06)0.14***(0.04)
Time since trans.-squared–0.005***(0.0005)–0.003***(0.0003)–0.003***(0.001)–0.003***(0.0004)–0.003***(0.001)–0.003***(0.0004)
Family structure
Married step-parents–14.31*(6.23)–4.63(4.22)–15.95*(7.64)–3.42(5.03)–16.92*(8.50)–3.88(5.49)
Cohabiting–3.79(6.00)1.00(3.45)–8.33(6.80)1.28(3.10)–18.79**(5.64)–1.91(3.44)
Single–9.85***(1.94)–7.07***(1.46)–17.86***(3.95)–7.62**(2.61)–20.55***(4.13)–11.22***(2.54)
Family income0.01(0.01)0.02**(0.01)0.01(0.01)0.02*(0.01)0.01(0.01)0.02*(0.01)
Number siblings0.68(0.59)0.70(0.38)0.74(0.65)0.68(0.44)0.75(0.64)0.69(0.44)
Contact with father0.22(0.18)0.20(0.11)0.17(0.18)0.17(0.11)
Interactions between mothers’ education (ME) and family structure
ME X Married Step-parents–1.18(11.34)–3.12(7.15)
ME X Cohabiting30.34***(6.92)8.76(4.76)
ME X Single3.80(3.74)7.26*(2.87)
(N=5716)(N=5715)(N=5716)(N=5713)(N=5716)(N=5713)

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

Table 3:

Estimated coefficients for fixed-effects models of mathematics and reading test scores (among children starting kindergarten in single-parent families)

Model 2Model 2aModel 2b
ReadingMathsReadingMathsReadingMaths
CoefS.E.CoefS.E.CoefS.E.CoefS.E.CoefS.E.CoefS.E.
Intercept5.69(4.18)3.26(2.70)5.07(6.57)0.27(3.57)7.34(6.59)1.56(3.58)
Level-1 variables
Child age2.51***(0.11)2.02***(0.06)2.49***(0.16)2.04***(0.09)2.50***(0.16)2.04***(0.09)
Child age-squared–0.01***(0.001)–0.01***(0.0004)–0.01***(0.001)–0.01***(0.001)–0.01***(0.001)–0.01***(0.001)
Time since transition0.08(0.08)0.12**(0.04)0.03(0.11)0.11(0.07)0.05(0.11)0.13*(0.06)
Time since trans.-squared0.0003(0.00)–0.0004(0.0003)0.002*(0.00)–0.0001(0.0004)0.001*(0.001)–0.0001(0.0004)
Family structure
Married step-parents13.24(6.88)8.19*(3.85)17.40(10.20)8.48(5.96)16.19(10.42)8.53(6.15)
Cohabiting10.69(7.92)10.69*(4.32)14.87(12.73)12.63(6.90)18.86(13.27)16.37*(7.28)
Married bio-parents–6.52(5.49)–1.25(4.42)–6.78(8.90)–8.73(5.34)–10.65(8.36)–11.46*(4.91)
Family income0.11*(0.05)0.05(0.03)0.14*(0.07)0.10(0.05)0.07(0.07)0.06(0.05)
Number siblings1.31(2.03)0.52(1.22)1.49(3.06)1.33(1.52)0.67(3.17)1.01(1.62)
Contact with father0.11(0.28)0.12(0.14)0.06(0.28)0.08(0.14)
Interactions between mothers’ education (ME) and family structure
ME X Married step-parents12.64(7.76)5.02(5.71)
ME X Cohabiting12.39(10.74)–6.82(8.19)
ME X Married bio-parents30.65**(10.21)20.05*(8.07)
(N=601)(N=601)(N=501)(N=501)(N=501)(N=501)

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

Conclusion and discussion

In this article, we set out to investigate family structure differences in children’s academic outcomes. In particular, we were interested in testing the argument that children in two-parent biological married families benefit, in comparison with children in other family structures, from greater resources and lower exposure to stressful events (Coleman, 1988; Amato, 1995). We first estimated a fixed-effects model in which the effects of family structure were assessed in a sample of children starting kindergarten in married biological-parent families. Although children in married biological-parent families were found to be advantaged in their academic outcomes, this was not found in the sample of children starting kindergarten in single-mother families, providing some support for hypothesis 1A and challenging hypothesis 1. At the same time, these models did not support the predictions of hypothesis 1B in that some changes in family structure were associated with an improvement in test scores (namely, the effect of living in married step-families or cohabiting families after living in a single-mother family was associated with better mathematics scores).

To further examine the predictions of social capital in the family explanations, we added measures of the child’s contact with their biological father. Paternal contact was not a statistically significant predictor of test scores in any of the models and we did not find evidence that paternal contact significantly explained some of the negative association between living in a single-parent, cohabiting or married step-family and test scores. These results fail to support hypothesis 2.

To further explore heterogeneity in these family forms, and to examine a key source of capital that may buffer the negative effect of living in a single-mother, cohabiting or step-family, we added an interaction between family structure and maternal education to our models. These models provide qualified support for hypothesis 3. In the models for children starting kindergarten in married biological-parent families, the negative direct effect of living in a cohabiting family on reading test scores was positively moderated by maternal education and the negative direct effect of living with a single mother on mathematics test scores was buffered by having a college-educated mother. In the models for children starting kindergarten in single-mother families, children were particularly advantaged in their mathematics and reading test scores when they had a college-educated mother and were living in a married biological-parent family, compared to when they were living in single-mother families.

These results contribute an understanding of the effects of family structure on academic achievement, net of confounding individual and family characteristics. This goes beyond other studies that analyse panel data using growth curve models because growth curve analyses cannot account for the effects of unmeasured individual and family background characteristics on student achievement. This study makes a contribution compared to other fixed-effects analyses of panel data by including children who start off their schooling in single-parent families, rather than restricting the sample to children living in married biological-parent families. When studies limit the sample to children starting off in married biological-parent families, this reinforces the dominance of the Standard North American Family and overlooks the experiences of children in other family types (Smith, 1993).

Our results present some wrinkles to family stress theory and nuance to social capital theories. First, only children who transitioned from married biological-parent families to other family structures experienced a negative impact on academic outcomes. Children who had previously lived in single-parent families but moved to other family structures (specifically married step-parent and cohabiting families) experienced an increase in mathematics test scores relative to when they were living in single-mother households. Family stress theory predicts that all family transitions are somewhat stressful due to the accompanying changes in family resources (Craigie et al, 2012); however, our analyses suggest that some family transitions may be associated with an increase in academic performance, rather than a decline.

Second, although the departure of a parent from the home was generally associated with a decline in academic outcomes in model 1, this negative effect was buffered in some cases by the presence of a college-educated mother (model 1b). This provides evidence that not all children in families that depart from the Standard North American Family structure are similarly disadvantaged, but rather that there is a diversity of experience depending on maternal education in particular. Additionally, the effects of living in a married biological-parent family (after starting kindergarten in a single-mother family) on academic performance varied depending on the presence of a college-educated mother (model 2b). Although children living in married biological-parent families after living in single-mother families had better test scores when they had college-educated mothers, this was not the case for children with mothers with lower levels of education. Based on social capital theory alone, we would expect that living in a married biological-parent family would be associated with more positive academic outcomes, so this finding that, among children starting kindergarten in a single-mother family, it is only children with a college-educated mother who achieve better academic outcomes adds a more nuanced understanding of the association between family structure and child academic performance.

These findings are consistent with the claim that college-educated mothers provide access to additional social networks and social support, as well as perhaps cultural capital, and this helps children navigate institutional settings. In some cases, maternal education buffers the disadvantage associated with the absence of a biological parent from the home, and in other cases, it may serve as an indicator of different experiences in different family structures. It seems that while the marriage of biological parents during early or middle childhood might not be associated with any advantages for children whose mothers have less education, these marriages are associated with better academic outcomes for children whose mothers have a college degree.

Our results are limited to the analysis of children’s academic outcomes in different family structures. Future research should extend the analysis to examine how other child outcomes (for example, behavioural outcomes, wellbeing outcomes) are associated with different family structures. Moreover, our results represent the effects of family structure on children’s academic test scores for children who experienced a change in family structure during the analysis window. Future research should investigate the effects of family structure among children who do not experience family instability. We also think future analyses should investigate the effects of other time-varying family variables on child academic outcomes (for example, measures of the health of family members, other major life changes such as residential changes). Finally, our results on the buffering effects of maternal education should be treated as preliminary and further investigated in a larger sample of children in cohabiting and single-mother families. However, additional sensitivity tests in which all family structure changes were combined into one category and interacted with maternal education provided the same substantive results as the results presented here, lending further support for the conclusions we have drawn.

Our results contribute to our understanding of how children’s academic performance is contextualised by family structure. We contribute an analysis of family structure and child academic outcomes that accounts for the stable, unobserved heterogeneity that exists among children in different family structures. We also contribute by drawing attention to the diversity of experiences that exist within family structure categories. We show that the academic advantages of living in a married biological-parent family exist for children who started kindergarten in married biological-parent families but not for children starting in single-mother families. We contribute an understanding of how paternal contact matters in shaping academic outcomes and how maternal education can serve as a buffer in moderating the association between family structure and test scores, with mothers’ resources potentially compensating in some cases for the missing resources of a biological parent who is absent from the home. Overall, these results challenge the argument that two married biological parents are always the preferred context for rearing children by highlighting how the association between family structure and child academic outcomes varies depending on both previous family structure and maternal education.

Notes

1

Models that include paternal contact have a reduced sample size due to missing data on the paternal contact variable. This is one reason why we estimate models with and without paternal contact included.

2

We also tested the effects of family structure at birth by interacting our measure of family structure at birth with our measures of family structure in grades K through 8. None of these interaction terms reached statistical significance and so we instead decided to make family structure in kindergarten (as the start of schooling) the earlier-life reference point for later family structure effects in investigating children’s academic outcomes.

Funding

This work was supported by the Spencer Foundation under Grant no. 201800093.

Conflict of interest

The authors declare that there is no conflict of interest.

References

  • Allison, P.D. (2005) Fixed Effects Regression Methods for Longitudinal Data Using Sas, Cary, NC: Sas Institute.

  • Allison, P.D. (2018) Longitudinal Data Analysis Using Stata, Ardmore, PA: Statistical Horizons.

  • Amato, P.R. (1995) Single-parent households as settings for children’s development, well-being, and attainment: a social network/resources perspective, in N. Mandell and A.-M. Ambert (eds) Sociological Studies of Children, Greenwich, CT: JAI Press, pp 1947.

    • Search Google Scholar
    • Export Citation
  • Amato, P.R. and Anthony, C.J. (2014) Estimating the effects of parental divorce and death with fixed effects models, Journal of Marriage and Family, 76(2): 37086. doi: 10.1111/jomf.12100

    • Search Google Scholar
    • Export Citation
  • Artis, J.E. (2007) Maternal cohabitation and child well-being among kindergarten children, Journal of Marriage and Family, 69(1): 22236. doi: 10.1111/j.1741-3737.2006.00355.x

    • Search Google Scholar
    • Export Citation
  • Augustine, J.M. (2014) Maternal education and the unequal significance of family structure for children’s early achievement, Social Forces, 93(2): 687718. doi: 10.1093/sf/sou072

    • Search Google Scholar
    • Export Citation
  • Bradbury, B., Corak, M., Waldfogel, J. and Washbrook, E. (2012) Inequality in early childhood outcomes, in J. Ermisch, M. Jäntti and T. Smeeding (eds) From Parents to Children, The Intergenerational Transmission of Advantage, New York, NY: Russell Sage Foundation, pp 87119.

    • Search Google Scholar
    • Export Citation
  • Brown, S.L. (2004) Family structure and child well-being: the significance of parental cohabitation, Journal of Marriage and Family, 66(2): 35167. doi: 10.1111/j.1741-3737.2004.00025.x

    • Search Google Scholar
    • Export Citation
  • Buehler, C. (2020) Family processes and children’s and adolescents’ well-being, Journal of Marriage and Family, 82(1): 14574. doi: 10.1111/jomf.12637

    • Search Google Scholar
    • Export Citation
  • Bzostek, S.H. and Berger, L.M. (2017) Family structure experiences and child socioemotional development during the first nine years of life: examining heterogeneity by family structure at birth, Demography, 54(2): 51340. doi: 10.1007/s13524-017-0563-5

    • Search Google Scholar
    • Export Citation
  • Coleman, J.S. (1988) Social capital in the creation of human capital, American Journal of Sociology, 94(3): S95S120. doi: 10.1086/229033

    • Search Google Scholar
    • Export Citation
  • Craigie, T.A.L., Brooks-Gunn, J. and Waldfogel, J. (2012) Family structure, family stability and outcomes of five-year-old children, Families, Relationships and Societies, 1(1): 4361. doi: 10.1332/204674312X633153

    • Search Google Scholar
    • Export Citation
  • Crawford, C., Goodman, A., Greaves, E. and Joyce, R. (2012) Cohabitation, marriage and child outcomes: an empirical analysis of the relationship between marital status and child outcomes in the UK using the Millennium Cohort Study, Child & Family Law Quarterly, 4(2): 176198.

    • Search Google Scholar
    • Export Citation
  • Denton, K. and West, J. (2002) Children’s Reading and Mathematics Achievement in Kindergarten and First Grade, Washington, DC: National Center for Education Statistics, Report No. NCES-2002–125.

    • Search Google Scholar
    • Export Citation
  • Entwisle, D.R. and Alexander, K.L. (1993) Entry into school: the beginning school transition and educational stratification in the United States, Annual Review of Sociology, 19(1): 40123. doi: 10.1146/annurev.so.19.080193.002153

    • Search Google Scholar
    • Export Citation
  • Eickmeyer, K.J. (2017) FP-17-15 American Children’s Family Structure: Two Biological Parent Families, Bowling Green, OH: National Center for Family and Marriage Research Family Profiles. 185. https://scholarworks.bgsu.edu/ncfmr_family_profiles/185

    • Search Google Scholar
    • Export Citation
  • Fomby, P. (2011) Family instability and school readiness in the United Kingdom, Family Science, 2(3): 17185. doi: 10.1080/19424620.2011.633274

    • Search Google Scholar
    • Export Citation
  • Fomby, P. and Cherlin, A.J. (2007) Family instability and child well-being, American Sociological Review, 72(2): 181204. doi: 10.1177/000312240707200203

    • Search Google Scholar
    • Export Citation
  • Fomby, P. and Osborne, C. (2017) Family instability, multipartner fertility, and behavior in middle childhood, Journal of Marriage and Family, 79(1): 7593. doi: 10.1111/jomf.12349

    • Search Google Scholar
    • Export Citation
  • Giele, J.Z. and Elder, G.H. Jr. (1998) Methods of Life Course Research: Qualitative and Quantitative Approaches, Thousand Oaks, CA: Sage.

    • Search Google Scholar
    • Export Citation
  • Gray, L.A., Hernández-Alava, M., Kelly, M.P. and Campbell, M.J. (2018) Family lifestyle dynamics and childhood obesity: evidence from the Millennium Cohort Study, BMC Public Health, 18(1): 500. doi: 10.1186/s12889-018-5398-5

    • Search Google Scholar
    • Export Citation
  • Hadfield, K., Amos, M., Ungar, M., Gosselin, J. and Ganong, L. (2018) Do changes to family structure affect child and family outcomes? A systematic review of the instability hypothesis, Journal of Family Theory & Review, 10(1): 87110. doi: 10.1111/jftr.12243

    • Search Google Scholar
    • Export Citation
  • Harding, J.F., Morris, P.A. and Hughes, D. (2015) The relationship between maternal education and children’s academic outcomes: a theoretical framework, Journal of Marriage and Family, 77(1): 6076. doi: 10.1111/jomf.12156

    • Search Google Scholar
    • Export Citation
  • Hernández-Alava, M. and Popli, G. (2017) Children’s development and parental input: evidence from the UK Millennium Cohort Study, Demography, 54(2): 485511.

    • Search Google Scholar
    • Export Citation
  • Hill, R. (1958) 1. Generic features of families under stress, Social Casework, 39(2–3): 13950. doi: 10.1177/1044389458039002-318

  • Jackson, M.I., Kiernan, K. and McLanahan, S. (2017) Maternal education, changing family circumstances, and children’s skill development in the United States and UK, The ANNALS of the American Academy of Political and Social Science, 674(1): 5984. doi: 10.1177/0002716217729471

    • Search Google Scholar
    • Export Citation
  • Karlson, K.B., Holm, A. and Breen, R. (2011) Comparing regression coefficients between same-sample nested models using logit and probit: a new method, Sociological Methodology, 42(1): 286313. doi: 10.1177/0081175012444861

    • Search Google Scholar
    • Export Citation
  • Kelly, Y., Sacker, A., Del Bono, E., Francesconi, M. and Marmot, M. (2011) What role for the home learning environment and parenting in reducing the socioeconomic gradient in child development? Findings from the Millennium Cohort Study, Archives of Disease in Childhood, 96(9): 83237. doi: 10.1136/adc.2010.195917

    • Search Google Scholar
    • Export Citation
  • Kiernan, K.E. and Mensah, F.K. (2011) Poverty, family resources and children’s early educational attainment: the mediating role of parenting, British Educational Research Journal, 37(2): 31736, doi: doi: 10.1080/01411921003596911.

    • Search Google Scholar
    • Export Citation
  • Klausli, J.F. and Owen, M.T. (2009) Stable maternal cohabitation, couple relationship quality, and characteristics of the home environment in the child’s first two years, Journal of Family Psychology, 23(1): 103. doi: 10.1037/a0014588

    • Search Google Scholar
    • Export Citation
  • Kroeger, R.A. and Smock, P. (2014) Cohabitation: recent research and implications, in J. Treas, J. Scott and M. Richards (eds) The Wiley Blackwell Companion to the Sociology of Families, Hoboken, NJ: Wiley.

    • Search Google Scholar
    • Export Citation
  • Lamidi, E. and Manning, W.D. (2016) FP-16-17 Marriage and Cohabitation Experiences Among Young Adults, Bowling Green, OH: National Center for Family and Marriage Research Family Profiles. 60. https://scholarworks.bgsu.edu/ncfmr_family_profiles/60.

    • Search Google Scholar
    • Export Citation
  • Manning, W.D. and Lamb, K.A. (2003) Adolescent well-being in cohabiting, married, and single-parent families, Journal of Marriage and Family, 65(4): 87693. doi: 10.1111/j.1741-3737.2003.00876.x

    • Search Google Scholar
    • Export Citation
  • Magnuson, K., Sexton, H., Davis-Kean, P., and Huston, A. (2009) The effects of increases in maternal education on young children’s language skills, Merrill Palmer Quarterly, 55(3): 319349.

    • Search Google Scholar
    • Export Citation
  • McLanahan, S. (1985) Family structure and the reproduction of poverty, American Journal of Sociology, 90(4): 873901. doi: 10.1086/228148

    • Search Google Scholar
    • Export Citation
  • McLanahan, S. (2004) Diverging destinies: how children are faring under the second demographic transition, Demography, 41(4): 60727. doi: 10.1353/dem.2004.0033

    • Search Google Scholar
    • Export Citation
  • McLanahan, S. and Percheski, C. (2008) Family structure and the reproduction of inequalities, Annu. Rev. Sociol, 34(1): 25776. doi: 10.1146/annurev.soc.34.040507.134549

    • Search Google Scholar
    • Export Citation
  • McNeal, Jr, R.B. (1999) Parental involvement as social capital: differential effectiveness on science achievement, truancy, and dropping out, Social Forces, 78(1): 11744. doi: 10.1093/sf/78.1.117

    • Search Google Scholar
    • Export Citation
  • Mooney, A., Oliver, C. and Smith, M. (2009) Impact of family breakdown on children’s well-being: evidence review, London: Institute of Education.

    • Search Google Scholar
    • Export Citation
  • Osborne, C. and McLanahan, S. (2007) Partnership instability and child well-being, Journal of Marriage and Family, 69(4): 106583. doi: 10.1111/j.1741-3737.2007.00431.x

    • Search Google Scholar
    • Export Citation
  • Panico, L., Bartley, M., Kelly, Y.J., McMunn, A. and Sacker, A. (2019) Family structure trajectories and early child health in the UK: pathways to health, Social Science & Medicine, 232 July 2019: 22029.

    • Search Google Scholar
    • Export Citation
  • Sassler, S. and Lichter, D.T. (2020) Cohabitation and marriage: complexity and diversity in union-formation patterns, Journal of Marriage and Family, 82(1): 3561. doi: 10.1111/jomf.12617

    • Search Google Scholar
    • Export Citation
  • Sayer, L.C. (2016) Trends in women’s and men’s time use, 1965–2012: back to the future?, in Gender and Couple Relationships, Springer, pp 4377.

    • Search Google Scholar
    • Export Citation
  • Schneider, W. (2016) Relationship transitions and the risk for child maltreatment, Demography, 53(6): 1771800. doi: 10.1007/s13524-016-0514-6

    • Search Google Scholar
    • Export Citation
  • Smith, D.E. (1993) The Standard North American Family: SNAF as an ideological code, Journal of Family Issues, 14(1): 5065. doi: 10.1177/0192513X93014001005

    • Search Google Scholar
    • Export Citation
  • Smock, P.J. and Schwartz, C.R. (2020) The demography of families: a review of patterns and change, Journal of Marriage and Family, 82(1): 934, doi: doi: 10.1111/jomf.12612.

    • Search Google Scholar
    • Export Citation
  • Stevenson, D. and Baker, D. (1987) The family-school relation and the child’s school performance, Child Development, 58(5): 13481357. doi: 10.2307/1130626.

    • Search Google Scholar
    • Export Citation
  • Sweeney, M.M. (2010) Remarriage and stepfamilies: strategic sites for family scholarship in the 21st century, Journal of Marriage and Family, 72(3): 66784. doi: 10.1111/j.1741-3737.2010.00724.x

    • Search Google Scholar
    • Export Citation
  • Tillman, K.H. (2007) Family structure pathways and academic disadvantage among adolescents in stepfamilies, Sociological Inquiry, 77(3): 383424. doi: 10.1111/j.1475-682X.2007.00198.x

    • Search Google Scholar
    • Export Citation
  • Tourangeau, K., Nord, C., , T., Sorongon, A.G. and Najarian, M. (2009) Early Childhood Longitudinal Study, kindergarten class of 1998–99 (ECLS-K): combined user’s manual for the ECLS-K eighth-grade and K-8 full sample data files and electronic codebooks. NCES 2009–004, Washington, DC: National Center for Education Statistics.

    • Search Google Scholar
    • Export Citation
  • US Census Bureau (2018) Living arrangements of children under 18 years old: 1960 to present, www.census.gov/content/dam/Census/library/visualizations/time-series/demo/families-and-households/ch-1.pdf.

    • Search Google Scholar
    • Export Citation
  • Allison, P.D. (2005) Fixed Effects Regression Methods for Longitudinal Data Using Sas, Cary, NC: Sas Institute.

  • Allison, P.D. (2018) Longitudinal Data Analysis Using Stata, Ardmore, PA: Statistical Horizons.

  • Amato, P.R. (1995) Single-parent households as settings for children’s development, well-being, and attainment: a social network/resources perspective, in N. Mandell and A.-M. Ambert (eds) Sociological Studies of Children, Greenwich, CT: JAI Press, pp 1947.

    • Search Google Scholar
    • Export Citation
  • Amato, P.R. and Anthony, C.J. (2014) Estimating the effects of parental divorce and death with fixed effects models, Journal of Marriage and Family, 76(2): 37086. doi: 10.1111/jomf.12100

    • Search Google Scholar
    • Export Citation
  • Artis, J.E. (2007) Maternal cohabitation and child well-being among kindergarten children, Journal of Marriage and Family, 69(1): 22236. doi: 10.1111/j.1741-3737.2006.00355.x

    • Search Google Scholar
    • Export Citation
  • Augustine, J.M. (2014) Maternal education and the unequal significance of family structure for children’s early achievement, Social Forces, 93(2): 687718. doi: 10.1093/sf/sou072

    • Search Google Scholar
    • Export Citation
  • Bradbury, B., Corak, M., Waldfogel, J. and Washbrook, E. (2012) Inequality in early childhood outcomes, in J. Ermisch, M. Jäntti and T. Smeeding (eds) From Parents to Children, The Intergenerational Transmission of Advantage, New York, NY: Russell Sage Foundation, pp 87119.

    • Search Google Scholar
    • Export Citation
  • Brown, S.L. (2004) Family structure and child well-being: the significance of parental cohabitation, Journal of Marriage and Family, 66(2): 35167. doi: 10.1111/j.1741-3737.2004.00025.x

    • Search Google Scholar
    • Export Citation
  • Buehler, C. (2020) Family processes and children’s and adolescents’ well-being, Journal of Marriage and Family, 82(1): 14574. doi: 10.1111/jomf.12637

    • Search Google Scholar
    • Export Citation
  • Bzostek, S.H. and Berger, L.M. (2017) Family structure experiences and child socioemotional development during the first nine years of life: examining heterogeneity by family structure at birth, Demography, 54(2): 51340. doi: 10.1007/s13524-017-0563-5

    • Search Google Scholar
    • Export Citation
  • Coleman, J.S. (1988) Social capital in the creation of human capital, American Journal of Sociology, 94(3): S95S120. doi: 10.1086/229033

    • Search Google Scholar
    • Export Citation
  • Craigie, T.A.L., Brooks-Gunn, J. and Waldfogel, J. (2012) Family structure, family stability and outcomes of five-year-old children, Families, Relationships and Societies, 1(1): 4361. doi: 10.1332/204674312X633153

    • Search Google Scholar
    • Export Citation
  • Crawford, C., Goodman, A., Greaves, E. and Joyce, R. (2012) Cohabitation, marriage and child outcomes: an empirical analysis of the relationship between marital status and child outcomes in the UK using the Millennium Cohort Study, Child & Family Law Quarterly, 4(2): 176198.

    • Search Google Scholar
    • Export Citation
  • Denton, K. and West, J. (2002) Children’s Reading and Mathematics Achievement in Kindergarten and First Grade, Washington, DC: National Center for Education Statistics, Report No. NCES-2002–125.

    • Search Google Scholar
    • Export Citation
  • Entwisle, D.R. and Alexander, K.L. (1993) Entry into school: the beginning school transition and educational stratification in the United States, Annual Review of Sociology, 19(1): 40123. doi: 10.1146/annurev.so.19.080193.002153

    • Search Google Scholar
    • Export Citation
  • Eickmeyer, K.J. (2017) FP-17-15 American Children’s Family Structure: Two Biological Parent Families, Bowling Green, OH: National Center for Family and Marriage Research Family Profiles. 185. https://scholarworks.bgsu.edu/ncfmr_family_profiles/185

    • Search Google Scholar
    • Export Citation
  • Fomby, P. (2011) Family instability and school readiness in the United Kingdom, Family Science, 2(3): 17185. doi: 10.1080/19424620.2011.633274

    • Search Google Scholar
    • Export Citation
  • Fomby, P. and Cherlin, A.J. (2007) Family instability and child well-being, American Sociological Review, 72(2): 181204. doi: 10.1177/000312240707200203

    • Search Google Scholar
    • Export Citation
  • Fomby, P. and Osborne, C. (2017) Family instability, multipartner fertility, and behavior in middle childhood, Journal of Marriage and Family, 79(1): 7593. doi: 10.1111/jomf.12349

    • Search Google Scholar
    • Export Citation
  • Giele, J.Z. and Elder, G.H. Jr. (1998) Methods of Life Course Research: Qualitative and Quantitative Approaches, Thousand Oaks, CA: Sage.

    • Search Google Scholar
    • Export Citation
  • Gray, L.A., Hernández-Alava, M., Kelly, M.P. and Campbell, M.J. (2018) Family lifestyle dynamics and childhood obesity: evidence from the Millennium Cohort Study, BMC Public Health, 18(1): 500. doi: 10.1186/s12889-018-5398-5

    • Search Google Scholar
    • Export Citation
  • Hadfield, K., Amos, M., Ungar, M., Gosselin, J. and Ganong, L. (2018) Do changes to family structure affect child and family outcomes? A systematic review of the instability hypothesis, Journal of Family Theory & Review, 10(1): 87110. doi: 10.1111/jftr.12243

    • Search Google Scholar
    • Export Citation
  • Harding, J.F., Morris, P.A. and Hughes, D. (2015) The relationship between maternal education and children’s academic outcomes: a theoretical framework, Journal of Marriage and Family, 77(1): 6076. doi: 10.1111/jomf.12156

    • Search Google Scholar
    • Export Citation
  • Hernández-Alava, M. and Popli, G. (2017) Children’s development and parental input: evidence from the UK Millennium Cohort Study, Demography, 54(2): 485511.

    • Search Google Scholar
    • Export Citation
  • Hill, R. (1958) 1. Generic features of families under stress, Social Casework, 39(2–3): 13950. doi: 10.1177/1044389458039002-318

  • Jackson, M.I., Kiernan, K. and McLanahan, S. (2017) Maternal education, changing family circumstances, and children’s skill development in the United States and UK, The ANNALS of the American Academy of Political and Social Science, 674(1): 5984. doi: 10.1177/0002716217729471

    • Search Google Scholar
    • Export Citation
  • Karlson, K.B., Holm, A. and Breen, R. (2011) Comparing regression coefficients between same-sample nested models using logit and probit: a new method, Sociological Methodology, 42(1): 286313. doi: 10.1177/0081175012444861

    • Search Google Scholar
    • Export Citation
  • Kelly, Y., Sacker, A., Del Bono, E., Francesconi, M. and Marmot, M. (2011) What role for the home learning environment and parenting in reducing the socioeconomic gradient in child development? Findings from the Millennium Cohort Study, Archives of Disease in Childhood, 96(9): 83237. doi: 10.1136/adc.2010.195917

    • Search Google Scholar
    • Export Citation
  • Kiernan, K.E. and Mensah, F.K. (2011) Poverty, family resources and children’s early educational attainment: the mediating role of parenting, British Educational Research Journal, 37(2): 31736, doi: doi: 10.1080/01411921003596911.

    • Search Google Scholar
    • Export Citation
  • Klausli, J.F. and Owen, M.T. (2009) Stable maternal cohabitation, couple relationship quality, and characteristics of the home environment in the child’s first two years, Journal of Family Psychology, 23(1): 103. doi: 10.1037/a0014588

    • Search Google Scholar
    • Export Citation
  • Kroeger, R.A. and Smock, P. (2014) Cohabitation: recent research and implications, in J. Treas, J. Scott and M. Richards (eds) The Wiley Blackwell Companion to the Sociology of Families, Hoboken, NJ: Wiley.

    • Search Google Scholar
    • Export Citation
  • Lamidi, E. and Manning, W.D. (2016) FP-16-17 Marriage and Cohabitation Experiences Among Young Adults, Bowling Green, OH: National Center for Family and Marriage Research Family Profiles. 60. https://scholarworks.bgsu.edu/ncfmr_family_profiles/60.

    • Search Google Scholar
    • Export Citation
  • Manning, W.D. and Lamb, K.A. (2003) Adolescent well-being in cohabiting, married, and single-parent families, Journal of Marriage and Family, 65(4): 87693. doi: 10.1111/j.1741-3737.2003.00876.x

    • Search Google Scholar
    • Export Citation
  • Magnuson, K., Sexton, H., Davis-Kean, P., and Huston, A. (2009) The effects of increases in maternal education on young children’s language skills, Merrill Palmer Quarterly, 55(3): 319349.

    • Search Google Scholar
    • Export Citation
  • McLanahan, S. (1985) Family structure and the reproduction of poverty, American Journal of Sociology, 90(4): 873901. doi: 10.1086/228148

    • Search Google Scholar
    • Export Citation
  • McLanahan, S. (2004) Diverging destinies: how children are faring under the second demographic transition, Demography, 41(4): 60727. doi: 10.1353/dem.2004.0033

    • Search Google Scholar
    • Export Citation
  • McLanahan, S. and Percheski, C. (2008) Family structure and the reproduction of inequalities, Annu. Rev. Sociol, 34(1): 25776. doi: 10.1146/annurev.soc.34.040507.134549

    • Search Google Scholar
    • Export Citation
  • McNeal, Jr, R.B. (1999) Parental involvement as social capital: differential effectiveness on science achievement, truancy, and dropping out, Social Forces, 78(1): 11744. doi: 10.1093/sf/78.1.117

    • Search Google Scholar
    • Export Citation
  • Mooney, A., Oliver, C. and Smith, M. (2009) Impact of family breakdown on children’s well-being: evidence review, London: Institute of Education.

    • Search Google Scholar
    • Export Citation
  • Osborne, C. and McLanahan, S. (2007) Partnership instability and child well-being, Journal of Marriage and Family, 69(4): 106583. doi: 10.1111/j.1741-3737.2007.00431.x

    • Search Google Scholar
    • Export Citation
  • Panico, L., Bartley, M., Kelly, Y.J., McMunn, A. and Sacker, A. (2019) Family structure trajectories and early child health in the UK: pathways to health, Social Science & Medicine, 232 July 2019: 22029.

    • Search Google Scholar
    • Export Citation
  • Sassler, S. and Lichter, D.T. (2020) Cohabitation and marriage: complexity and diversity in union-formation patterns, Journal of Marriage and Family, 82(1): 3561. doi: 10.1111/jomf.12617

    • Search Google Scholar
    • Export Citation
  • Sayer, L.C. (2016) Trends in women’s and men’s time use, 1965–2012: back to the future?, in Gender and Couple Relationships, Springer, pp 4377.

    • Search Google Scholar
    • Export Citation
  • Schneider, W. (2016) Relationship transitions and the risk for child maltreatment, Demography, 53(6): 1771800. doi: 10.1007/s13524-016-0514-6

    • Search Google Scholar
    • Export Citation
  • Smith, D.E. (1993) The Standard North American Family: SNAF as an ideological code, Journal of Family Issues, 14(1): 5065. doi: 10.1177/0192513X93014001005

    • Search Google Scholar
    • Export Citation
  • Smock, P.J. and Schwartz, C.R. (2020) The demography of families: a review of patterns and change, Journal of Marriage and Family, 82(1): 934, doi: doi: 10.1111/jomf.12612.

    • Search Google Scholar
    • Export Citation
  • Stevenson, D. and Baker, D. (1987) The family-school relation and the child’s school performance, Child Development, 58(5): 13481357. doi: 10.2307/1130626.

    • Search Google Scholar
    • Export Citation
  • Sweeney, M.M. (2010) Remarriage and stepfamilies: strategic sites for family scholarship in the 21st century, Journal of Marriage and Family, 72(3): 66784. doi: 10.1111/j.1741-3737.2010.00724.x

    • Search Google Scholar
    • Export Citation
  • Tillman, K.H. (2007) Family structure pathways and academic disadvantage among adolescents in stepfamilies, Sociological Inquiry, 77(3): 383424. doi: 10.1111/j.1475-682X.2007.00198.x

    • Search Google Scholar
    • Export Citation
  • Tourangeau, K., Nord, C., , T., Sorongon, A.G. and Najarian, M. (2009) Early Childhood Longitudinal Study, kindergarten class of 1998–99 (ECLS-K): combined user’s manual for the ECLS-K eighth-grade and K-8 full sample data files and electronic codebooks. NCES 2009–004, Washington, DC: National Center for Education Statistics.

    • Search Google Scholar
    • Export Citation
  • US Census Bureau (2018) Living arrangements of children under 18 years old: 1960 to present, www.census.gov/content/dam/Census/library/visualizations/time-series/demo/families-and-households/ch-1.pdf.

    • Search Google Scholar
    • Export Citation
  • 1 University at Buffalo, , USA
  • | 2 DePaul University, , USA
  • | 3 Wake Forest University, , USA
  • | 4 Central University of Finance and Economics, , China

Content Metrics

May 2022 onwards Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 5 5 5
PDF Downloads 5 5 5

Altmetrics

Dimensions