Early Childhood Family Structure and Mother–Child Interactions:

Early Childhood Family Structure and Mother–Child Interactions: Variation by Race and Ethnicity
Early Childhood Family Structure and Mother–Child Interactions:
Variation by Race and Ethnicity
Christina M. Gibson-Davis and Anna Gassman-Pines
Duke University
With data from the Early Childhood Longitudinal Study–Birth Cohort (n
6,449), a nationally
representative sample of births in 2001, we used hierarchical linear modeling to analyze differences in
observed interactions between married, cohabiting, never-married, and divorced mothers and their
children. In contrast to previous studies, we concentrated on early childhood, a developmentally critical
period that has been understudied in the family structure literature, and relied on objective observational
measures of mother– child interactions, which are unlikely to be biased by maternal perceptions of
interactions with children. Nonmarital family structures were common in the lives of young children, as
32% lived outside of a married, biological parent home. Initial results indicated that married families
were consistently associated with higher quality interactions. Moreover, though it was hypothesized that
the presence of a biological father might be associated with higher quality interactions than single-parent
households, this hypothesis was not confirmed. Additional models suggest that race and ethnicity
moderated the effect of family structure, as non-Hispanic Black and non-Hispanic White participants
showed little significant variation between married and unmarried families. Among Hispanics, mothers
living in cohabiting, divorced, or single families, when compared to married mothers, exhibited more
negative and more intrusive behaviors; cohabiting mothers also scored lower on the measure of cognitive
stimulation. Results suggest that marriage may not be uniformly associated with higher levels of
mother– child interactions and that cohabitation, particularly for Hispanics, may be associated with
adverse outcomes.
Keywords: family structure, early childhood, race and ethnicity
The American familial landscape has increasingly been characterized
by children, particularly those of color, residing in households
that consist of arrangements other than two married biological
parents. The birthrate for unmarried women increased 16%
between 2002 and 2006 and rose to 50.6 births per 1,000 unmarried
women (Martin et al., 2009). Among non-Hispanic Whites,
27% of all births were to unmarried women, as compared to 50%
for Hispanics and 71% for non-Hispanic Blacks (Martin et al.,
2009). Moreover, minority children were twice as likely as non-
Hispanic White children to be born into an unmarried family with
two resident parents, and Black children were more than three
times as likely as White children to be born to a never-married
mother living without a romantic partner (Bumpass & Lu, 2000).
The prevalence of unmarried family arrangements, particularly
among minority households, has prompted concern because of the
potentially negative effects such arrangements may have on parenting
and, by extension, on child well-being. If such negative
effects exist, they may be particularly detrimental to young children,
who are acquiring and developing fundamental socioemotional
skills as they transition from infancy to school age (Belsky,
1999; Bornstein, 1995). The majority of the literature in this area,
however, has concentrated on middle childhood and older, and has
found that parents in never-divorced, married families typically
displayed the highest levels of warm, responsive, and sensitive
caregiving (Hetherington & Clingempeel, 1992; Hetherington &
Jodle, 1994; Thomson, Mosley, Hanson, & McLanahan, 2001).
The few studies of parenting and diverse family arrangements among
young children have not uniformly shown that married mothers exhibit
more optimal parenting but have suggested that relationship
transitions can be harmful (Beck, Cooper, McLanahan, & Brooks-
Gunn, 2009; Cooper, McLanahan, Meadows, & Brooks-Gunn, in
press; Gibson-Davis, 2008).
An additional limitation of the existing literature is that, despite
racial and ethnic differences in family structure patterns, few
studies have examined how the associations between family arrangements
and parenting differ among non-Hispanic Whites, non-
Hispanic Blacks, and Hispanics. The associations between family
structure and parent– child interaction are likely to differ by race
and ethnicity because of differences between groups in wealth and
assets, cultural norms, and informal and kinship support (Lerman,
2002; Oropesa & Landale, 2004). Although race and ethnicity
likely moderate effects of family structure on mother– child interactions,
we are aware of no study that has examined the effect of
diverse family structure on mother– child interactions for young
children of different racial and ethnic groups.
This article seeks to fill that gap in the literature by examining
relationships between family structure and the quality of the interactions
between mothers and young children in families of
different racial/ethnic groups. Using data from the Early Child-
Christina M. Gibson-Davis and Anna Gassman-Pines, Sanford School
of Public Policy, Duke University.
Correspondence concerning this article should be addressed to Christina
M. Gibson-Davis, Sanford School of Public Policy, Duke University, P.O.
Box 90245, Durham, NC 27708. E-mail: cgibson@duke.edu
Developmental Psychology © 2010 American Psychological Association
2010, Vol. 46, No. 1, 151–164 0012-1649/10/$12.00 DOI: 10.1037/a0017410
151
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
hood Longitudinal Study–Birth Cohort, a longitudinal, nationally
representative sample of American children born in 2001, we make
three contributions to the literature with this study. First, as described
earlier, we focused on a heretofore understudied age range,
2- to 4-year-olds. Second, we considered how race and ethnicity
moderated the associations between family structure and the quality
of mother– child interactions. Third, we relied on objective,
independently coded measures of mother– child interactions, with
mothers’ and children’s behaviors and actions coded distinctly.
Because our outcome variables do not rely on maternal report, they
are unlikely to reflect preexisting characteristics of the mother that
could potentially bias her perceptions of her interaction with her
child.
Importance of Parenting for Child Development
There has been a long-standing interest in the ways that parenting
and parent– child interactions shape relationships between parents
and children and, in turn, children’s development. The quality
of parent– child interactions during children’s early years has been
associated with children’s social, emotional, and cognitive functioning.
In particular, caregiving during children’s early years that
is sensitive and responsive, includes appropriate levels of stimulation,
and involves synchrony in interactions between parent and
child has been associated with more secure attachment, which
serves as an important foundation for children’s later development
across domains, such as self-esteem, positive peer interactions, and
literacy development (Belsky, 1999; Berlin & Cassidy, 1999;
Bornstein, 1995; De Wolff & van IJzendoorn, 1997). In contrast,
caregiving that is intrusive, excessively stimulating, controlling, or
unresponsive has been associated with less secure attachment in
young children and with more negative child development, including
negative affect, behavior problems, and difficulties in school
(Bornstein, 1995; Ispa et al., 2004). Although much of the research
on parenting and child development has been correlational, some
experimental studies have provided evidence of a causal link
between parenting and children’s developmental outcomes (Belsky,
1999).
Theoretical Links Between Family Structure and
Mother–Child Interactions
Previous studies of mother– child interactions in diverse family
arrangements have largely concentrated on children in middle
childhood or adolescence (DeGarmo & Forgatch, 1999; Hetherington
& Stanley-Hagan, 1999; McLanahan & Sandefur, 1994).
Only a few studies have studied early childhood (Beck et al., 2009;
Cooper et al., in press; Gibson-Davis, 2008; Shaw, Winslow,
Owens, & Hood, 1998). Results regarding family status have been
mixed: Gibson-Davis (2008) found few effects of family structure
on parenting behavior, whereas Shaw et al. (1998) reported that
mothers outside of never-divorced married families exhibited
higher levels of rejecting parenting behaviors. Results for family
transitions, however, have been more uniform and have suggested
that family instability can have detrimental effects on mother–
child interactions (Beck et al., 2009; Cooper et al., in press).
Despite the paucity of studies that have concentrated on early
childhood, the existing literature on mother– child interactions is
instructive about why family structure may influence that dynamic.
Below we discuss why family structure is likely to affect mother–
child interactions, concentrating on the family types used in this
study: never-divorced, married biological parents; cohabiting unmarried
parents (e.g., two biological parents sharing living quarters);
and single-mother families (both divorced and never married).
A primary reason why mother– child interactions may differ by
family type is because of the high correlation between family
structure and economic status. National estimates from the United
States have found that the poverty rate of single-parent homes was
43.5%, as compared to 21.1% for cohabiting families and 7.6% for
married families (Acs & Nelson, 2002). Single and cohabiting
parents also had fewer assets, less stable employment, and higher
levels of material hardship (Kreider & Fields, 2005; Manning &
Lichter, 1996). Among single parents, never-married mothers may
be even worse off than divorced mothers, because they had lower
levels of human capital and were less likely to receive child
support (Hao, 1996). Though cohabiting couples had more financial
resources than single parents (Lerman, 2002), they derived
fewer economic benefits from living together when compared to
married couples, insofar as they were less likely to pool resources
(Bauman, 1999).
As has been well documented, these differences in financial
resources are likely to have pronounced effects on parenting (Conger
et al., 1992; Conger, Patterson, & Ge, 1995; Whitbeck et al.,
1997). Research has shown that, compared to their higher income
counterparts, low-income parents were more likely to use harsh
discipline, engage in erratic and inconsistent parenting practices,
and provide less cognitive stimulation (Bradley, Corwyn,
McAdoo, & Coll, 2001; Yeung, Linver, & Brooks-Gunn, 2002).
Economic well-being has also been directly linked to parental
psychological health, because financial strain increases stress, anxiety,
and depression, further compromising parenting (McLoyd,
1990, 1998).
In addition to differences in economic resources, unmarried
parents may have less access to emotional support (Carlson &
McLanahan, 2006). Parents without a partner may be less likely to
meet the demands and challenges of raising children and may be
more likely to become overwhelmed and frustrated by the task
(Belsky, 1990; Cox, Owen, Lewis, & Henderson, 1989). In turn, a
depletion of parental emotional energy can lead to inconsistent and
harsh parenting (Thomson et al., 2001). Divorced parents may be
at particular risk for harsh parenting practices (Hetherington,
1999), as they seek to regain equilibrium after a marriage dissolves.
Even among two-parent families, emotional support may be
lower in cohabiting as compared to married households. Studies of
cohabiting couples have found that cohabiters received less emotional
and instrumental support from their partners, had higher
levels of conflict, and exhibited lower levels of commitment
(Brown & Booth, 1996; Carlson & McLanahan, 2006; Nock,
1995). The quality of the mother–father relationship has been
found to have direct effects on parenting ability (Belsky, 1984;
Erel & Burman, 1995; Voydanoff & Donnelly, 1998), because
parents in unloving and unsupportive relationships engaged in
harsher and less sensitive parenting (Voydanoff & Donnelly,
1998).
A third and final consideration is the stability of parental relationships.
Family instability can affect child well-being, beyond
152 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
any effects associated with a particular family type, because it not
only influences a family’s financial resources but also parenting
roles and family dynamics (Amato, 2005). Relationship transitions
have been associated with more authoritarian, erratic, and inattentive
parenting and with less warmth and nurturing as families
adjust to the new roles and responsibilities associated with family
reconfigurations (Hetherington, 1999; Stewart, 2007). Children
who experience multiple transitions are particularly vulnerable, in
part because of increased levels of maternal conflict and depression
(Capaldi & Patterson, 1991; DeGarmo & Forgatch, 1999).
Not surprisingly, children born to single or cohabiting mothers are
more likely to experience family instability (Manning, Smock, &
Majumdar, 2004); one birth cohort study found that the rate of
transitions was eight times higher for single and three times higher
for cohabiting families when compared to married-parent families
(Osborne & McLanahan, 2007).
In summary, family structure is thought to affect mother– child
interactions through its associations with economic status, emotional
support, and family stability. Economic status is of particular
interest, because its key determinants—namely, age, employment,
household structure, and educational status (of both the
mother and her parents)—vary greatly between married and unmarried
mothers (Demo & Cox, 2000). It is therefore necessary to
control for these characteristics to avoid attributing to family
structure what is really caused by sociodemographic factors.
The literature reviewed here suggests that mothers in neverdivorced,
biological married families exhibit more optimal parenting
practices than mothers in any other arrangement and that their
children exhibit the most positive behaviors toward their mothers.
Marriage by no means guarantees good mother– child interactions—
marital conflict, for example, can have a particularly detrimental
effect (Cummings & Davies, 2002)—but it does increase
the likelihood that, all else being equal, married mothers will have
sufficient economic and emotional resources to exhibit more favorable
maternal behaviors. Mothers living in cohabiting households,
by virtue of having higher economic resources than singleparent
households as well as an extra adult, may exhibit more
positive mother– child interactions than either divorced or nevermarried
mothers. Despite the strains faced by divorced mothers,
never-married families may have the most negative mother– child
interactions, because of their low levels of income and high rates
of relationship instability.
How Race and Ethnicity Affect
Mother–Child Interactions
Race and ethnicity likely moderate the associations between
family structure and mother– child interactions. Notably, marriage
has been found to convey fewer economic benefits to minority
couples, because Black and Hispanic married parents have been
found to have lower per capita family income than White married
parents (Thomas & Sawhill, 2005). Moreover, even though poverty
rates among Blacks and Hispanics were higher across all
family types, the relative gap between married and unmarried
families was greatest in White families. Among Whites, femaleheaded
households were seven times more likely to be in poverty
than married-couple households. For Blacks, female-headed
households were three times more likely to be poor than married
households; the same number for Hispanic female-headed households
was one and a half (Federal Interagency Forum on Child and
Family Statistics, 2007). Insofar as economic well-being directly
influences parenting outcomes (McLoyd, 1990), the difference
between White married and unmarried parents may be larger than
that observed for their minority counterparts.
Nonmarital fertility and cohabitation are more common among
Blacks and Hispanics, and women of color are less likely to marry
in response to a birth (Graefe & Lichter, 1999). Therefore, minorities
may be more likely to view nonmarital parenting situations as
normative and may more comfortably adapt than Whites to a
nonmarital birth. Blacks and Hispanics may also be more likely
than Whites to be able to draw on an extended kinship network, a
historically vital source of resources and strength in minority
communities (Hogan, Eggebeen, & Clogg, 1993; Oropesa &
Landale, 2004). These differences in family and community context
suggest fewer adverse associations for a child of color in an
unmarried family structure, as norms, community support, and
kinship networks may buffer the adverse effects.
Variations by race and ethnicity in economic resources and
social norms suggest that the most pronounced differences in
parenting outcomes between married and unmarried parents are
found for Whites. Counterbalancing these effects, however, is
evidence that Whites exhibit more stability in their relationships
(Popenoe & Whitehead, 2006). For example, Whites are more
likely to marry following a nonmarital birth and are less likely to
dissolve a cohabiting relationship (Graefe & Lichter, 2002; Manning
et al., 2004). Minority children also experience a higher
number of maternal relationship transitions over the first 3 years of
life (Osborne & McLanahan, 2007).
Despite the higher rate of relationship stability among Whites,
we believe that the lack of economic benefits associated with
marriage for minority couples as well as the cultural context of the
Black and Hispanic communities will lessen the relative gap
between married and unmarried parents in parenting outcomes for
those groups. We therefore expect to see the largest differences
among Whites; within that group, we expect that married mothers
would exhibit more optimal parenting practices than cohabiting
mothers, followed by divorced and never-married mothers. Differences
should be less pronounced among Black and Hispanic parents,
but the same pattern among married, cohabiting, and divorced
and never-married mothers should still hold.
The Present Study
In this study, we sought to answer the following research questions.
First, what are the patterns of family structure among children
between the ages of 24 and 48 months, and do they differ by
race and ethnicity? Because national estimates of family structure
among young children have not been previously available, we first
investigated the prevalence of different household arrangements.
Second, what is the association between family structure and the
quality of mother– child interactions, and how does that association
vary by race and ethnicity? Married parents are likely to have
the highest levels of economic resources, emotional support, and
relationship stability, suggesting that the quality of mother– child
interactions is highest in married families. The differing economic
and cultural contexts of marriage for families of color, however,
suggest that differences in mother– child interactions by family
structure are most pronounced in White families. One limitation in
FAMILY STRUCTURE 153
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
addressing these questions is that our research design does not
allow for causal inference; our findings represent associations
between family structure and mother– child interactions but do not
represent causal effects.
Method
Data: The Early Childhood Longitudinal Study–Birth
Cohort (ECLS-B)
The ECLS-B is a nationally representative survey of nearly
11,000 births that occurred in 2001. The ECLS-B was designed to
provide a comprehensive source of information about young children
and collected data on a wide range of parent and child
outcomes. Interviews occurred at 9, 24, and 48 months (families
were also interviewed in kindergarten; data were not yet available
at the time of our study). To construct the sample, the ECLS-B
relied on birth certificate data provided by the National Center for
Health Statistics. The sampling schema was designed to generate
a nationally representative sample, with counties or county groups
serving as the primary sampling units. The ECLS-B did not include
children who died, those who were adopted prior to the first
home visit, or those whose birth mothers were younger than 15.
Additional details on sample selection are available elsewhere
(Snow et al., 2007). Response rates were 74% at 9 months, 69% at
24 months, and 63% at 48 months.
The ECLS-B collected information from the child’s caregiver
through both a computer-assisted questionnaire and a selfadministered
questionnaire. In more than 95% of cases, the child’s
caregiver was the child’s mother or female guardian. The
computer-assisted questionnaire covered domains such as household
structure, economic resources, and social support. The selfadministered
questionnaire asked about more sensitive topics, including
parental relationship quality and children residing outside
the home. Information on parent– child interactions was collected
through videotaped and independently rated assessments with the
Two Bags Task, described in more detail later.
Data used here come from the second and the third waves of the
study and were collected when the children were approximately 24
and 48 months old (the first wave did not contain videotaped
parent– child interactions). Our sample consists of the 6,742 children
who were interviewed in both the second and third waves and
for whom Two Bags Task data were available (an additional 6% of
children participated in the Two Bags Task, but their tapes were
not coded because of technical difficulties). The sample was further
restricted to the 6,449 (96%) of children who lived in household
with biological parents (either married or cohabiting) or with
a biological divorced or never-married mother at Wave 2. This
sample restriction was imposed because there were very few
examples of adopted (n
8), step-parent married (n
60),
step-parent cohabiting (n
95), widowed (n
11), single-father
(n
29), or other (n
90) arrangements. Such households are
likely to be qualitatively different from those under consideration
here (McLanahan & Sandefur, 1994; Miller, Fan, Christensen,
Grotevant, & van Dulmen, 2000).
Though the survey had relatively high rates of attrition, we used
weights that corrected for sampling and nonreponse bias (Snow et
al., 2007). To adjust for attrition, certain cases were overweighted
so that the remaining sample was representative of the original
sample. As compared to those observed at 24 months but not
observed at 48 months, those observed at both waves did not differ
significantly in most of the mother– child interactions or in their
race and ethnicity. The one exception was that families observed at
both waves had slightly higher scores on the measure of maternal
negative regard compared to those observed only at 24 months. All
estimates were also corrected for the ECLS-B’s complex sampling
design. There were virtually no missing data on the covariates.
Measures
Dependent variables: Mother–child interactions. Mother–
child interactions were assessed through administration of the Two
Bags Task, a 10-min videotaped interaction between parent and
child that measured parenting behaviors and child responsiveness.
The Two Bags Task was a modification of the Three Bags Task
that was used in the Early Head Start Research and Evaluation
Project (Love et al., 2005) and in the National Institute of Child
Health and Human Development’s (1999) Study of Early Child
Care. In this semistructured interaction, caregivers and their children
were given two bags of toys: The first bag contained a book,
and the second bag contained Play-Doh, cookie cutters, and a
rolling pin. To elicit naturally occurring parent– child interactions,
parents were not given any specific instructions, except that they
had to play first with the book and then with the Play-Doh and
cookie cutters. Videotapes were sent to a professional survey
organization, where trained coders rated the interactions on two
child scales and four parent scales. The coders were unaware of
information about the family. Each scale corresponded to a behavioral
domain central to child development. The two child scales
measured negativity and engagement, and the four parent scales
measured negativity, intrusiveness, cognitive stimulation, and detachment.
We utilized all available scales that were measured at
both waves. For each scale, coders rated the videotaped interaction
on a 1 (very low) to 7 (very high) scale. Coders underwent
extensive training at each wave, and various procedures were used
to maintain interrater reliability and prevent coder drift (Andreassen
& Fletcher, 2007). At the 24-month wave, interrater reliability
for each scale ranged from .94 to .99 (Andreassen & Fletcher,
2007). At the 48-month wave, interrater reliability for each scale
ranged from .87 to 1.00 (G. Mulligan, personal communication,
November 6, 2008). The descriptions given next are drawn from
Snow et al. (2007). We use the word parent to describe the
measures, but readers should note that the sample used here
consists of only mothers. Table 1 presents the means and standard
deviations of these measures at 24 and 48 months.
Although child and parent scales were rated during the same
interactions, the child and parent scales appear to represent distinct
constructs. Within each wave, correlations between the scales
ranged from .09 to .49. The correlations across waves ranged from
.10 to .28.
Child engagement of parent. This scale measured the degree
of interaction originated and sustained by the child and the amount
of positive affect shown toward the parent. Children at the high
end of the scale would frequently interact with the parent and
would exhibit positive behaviors (e.g., frequently laughing or
smiling). Children at the low end of the scale would have little to
no interaction with the parent and would express no positive affect.
154 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Child negativity toward parent. This scale measured the extent
of antipathy the child had for the parent, as measured by
expressions of anger or hostility (e.g., pushing the parent away,
throwing a toy, screaming “No!”). Coders were instructed to rate
only those behaviors that were directed specifically toward the
parent, rather than hostile actions more generally.
Parental negative regard. This scale measured the amount of
dissatisfaction, displeasure, and rejection the parent exhibited toward
the child. Parents exhibited these behaviors by not listening
to, ignoring, or condemning the child or the child’s actions.
Parental intrusiveness. This scale measured the extent to
which the parent tried to manipulate and control the child, rather
than recognizing and respecting the child’s autonomy. Highly
intrusive parents imposed their own agenda and adopted a parentcentered,
rather than child-centered, perspective.
Parental stimulation of cognitive development. This scale
measured a parent’s efforts to stimulate a child’s cognitive and
language skills and the degree to which those efforts sought to help
the child reach the next developmental level. Parents were rated
lower on this scale if their efforts were poorly matched to their
child’s developmental status.
Parental detachment. This scale measured a parent’s indifference
or inattention to the child, as measured by both the extent and
the quality of the parent– child interaction. Highly detached parents
were those who were consistently inattentive, inconsistently
attentive, or had perfunctory interactions with the child.
Independent variables: Family structure. Family structure
was coded at each wave by combining information on parents’
current marital status with information on the people residing in
the household with the child. Parents were asked to indicate if they
were currently married, separated, divorced, widowed, or had
never been married. Parents also listed the people who lived in the
home and stated that person’s relationship to the child, including
father/male guardian. If a parent listed a father or father figure
living in the household, the parent was further asked to indicate the
type of father: birth, adoptive, step, or foster or guardian.
On the basis of these responses, mothers were coded into four
mutually exclusive categories: married (a biological mother married
to a biological father), cohabiting (a biological mother residing
with a biological father), never married (to anyone), and
divorced (from either the child’s father or another man). Because
the sample was limited to these four family structures and because
some children transitioned into another family structure between
the two waves, an other arrangement category was created. This
category included children who were living with a step-parent
(either married or unmarried), with a widowed mother, with a
single father (either never married, widowed, or divorced), or in a
household without biological parents.
Mediators.
Transitions. Two variables measured transitions: if the child
transitioned from one family structure to another between 9 and 24
months and if the child transitioned between 24 and 48 months. In
preliminary models, we included a variable that measured whether
the biological parents were married when the child was born. This
measure was nearly collinear with the cohabiting and nevermarried
categories and was not included in our final analyses.
Relationship quality. The ECLS-B also measured relationship
quality for married and cohabiting families, through measures of
relationship happiness and conflict. The measure of happiness with
the relationship asked individuals to indicate how happy their
marriage or partnership was, using a 1 (not too happy) to 3 (very
happy) scale. Relationship conflict asked respondents, using a 1
(never) to 4 (often) scale, to indicate how often they argued with
their spouse or partner about 10 items, including children, money,
sex, religion, drinking, and leisure time. The final score was a
mean of the 10 responses (
.83).
Socioeconomic status (SES). SES was a composite variable
constructed by the ECLS-B that measured maternal and paternal
education, maternal and paternal occupational prestige scores, and
household income. Occupational prestige was coded according to
standards used by the U.S. Office of Budget and Management.
Income was measured by 13 categories, ranging from less than
$5,000 to $200,001 or more (ranges were not equally divided but
were in $5,000 increments until $40,000 and then varied in dollar
interval). Each of the five components of SES was standardized,
and the final score was a weighted average. For households that
had only one parent, the score was averaged over the available
components. The SES scale ranged from 4.75 to 2.75, with
higher scores indicating a higher level of SES. More details on the
construction of the composite SES variable are available from the
ECLS-B (Snow et al., 2007). Even though the SES composite
contained parental education, fit statistics of the hierarchical linear
modeling (HLM) improved when education was included as a
separate component, so all models include parental education as
well as SES.
Moderators: Child race and ethnicity. Child race and ethnicity
was based on maternal report. The categories are non-
Hispanic White (the omitted category), non-Hispanic Black, Hispanic,
Asian, and other race or ethnicity.
Other control variables. Child controls included child age (in
months), child sex (1
female), and whether a child is a twin.
Family controls included maternal age (in years), number of adults
and number of children under the age of 18 in the household,
maternal age at first birth, parents’ education, grandparents’ education,
maternal household structure in her family of origin, and
maternal employment. Parents’ and grandparents’ education consisted
of four dichotomous indicators: no high school diploma, a
high school diploma or general equivalency diploma, some college
but no bachelor’s degree, and a bachelor’s degree or higher (omitted
category). The parental educational attainment variables were
coded to reflect the highest level of education obtained in the
household (e.g., if the household had two parents, parental educa-

Table 1
Descriptive Statistics of Mother–Child Interaction Variables by
Wave (N
6,449)
Variable measured
24 months 48 months
M SD M SD
Child
Negativity 1.35 1.12 1.34 1.03
Engagement 4.56 1.70 4.48 1.25
Parent
Negativity 1.11 0.64 1.18 0.69
Intrusiveness 1.17 0.78 1.52 1.22
Cognitive stimulation 4.12 1.60 4.19 1.37
Detached 1.05 0.49 1.30 0.95
FAMILY STRUCTURE 155
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
tional attainment would reflect the highest level obtained by either
parent). Grandparents’ education likewise refers to the highest
educational level obtained by either parent’s parents. A dichotomous
variable assessed whether the mother lived with both biological
parents until the age of 16. Maternal employment was
divided into four mutually exclusive categories: working full-time
(more than 35 hr per week; omitted category), working part-time
(fewer than 35 hr per week), out of the labor force, or unemployed.
Analytic Approach
Because repeated measures were collected for each family,
analyses needed to account for nonindependence of observations.
To address nonindependence, we used HLM through the PROC
MIXED command in SAS. HLM models are regression based,
assuming linear relationships between variables, but they use a
maximum likelihood estimation strategy.
The relationship between family structure and mother– child
interactions was modeled at two levels of analysis: within family
(Level 1) and between family (Level 2). The following equations
summarize this approach: For the within-family (Level 1) model,
MCINTij
0j  1jFAMSTRUCij  2j(COVAij)  rij,
where each mother– child interaction outcome (MCINT) at time (i)
for a particular family (j) was modeled as a function of that
family’s intercept (0j), the family structure at that wave (1j), a
vector of covariates (2j), and an error term (rij). Family structure
(FAMSTRUCij) was represented by a set of indicator variables
(with married-parents family as the reference group).
For the between-family (Level 2) models,
0j
00  u0j,
1j
10  u1j,
2j
20,
where 00 is the overall intercept (grand mean), u0j is the deviation
of family j’s intercept from the overall intercept, 10 is the overall
average association of family structure and mother– child interactions,
u1j is the deviation of family j’s coefficient from the overall
coefficient, and 20 is the overall average association of the covariates
and mother– child interactions.
Because the focus of this study is on associations between
family structure and mother– child interactions, only the parameter
10, which represents the average within-family association between
family structure and mother– child interaction outcomes, is
reported. The 10 coefficients examine average levels across the
two waves but do not capture change over time. The coefficients
presented later represent the effects of family structure, holding
constant changes in family structure.
We also used this basic model to test for differences by our
moderating variables (i.e., race and ethnicity). We ran the model
separately for each racial and ethnic group as described earlier and
then conducted a set of follow-up joint hypothesis tests. First, we
compared the associations between all the family structure categories
and each mother– child interaction outcome by race and
ethnicity. Second, we compared, also by race and ethnicity, the
associations between specific family structure categories (e.g.,
cohabiting) and each mother– child outcome.
Each family was surveyed at two waves, generating withinfamily
variability in the key constructs. The HLM approach described
earlier models within-family change and adjusts standard
errors for nonindependence of observations. Another potential
approach would be to use fixed effects, which adjust for all
measured and unmeasured static family characteristics. Although
our approach could produce results that are subject to selection
bias, we have chosen to use it rather than fixed effects for the
following reasons. First, fixed effects have much lower levels of
power than HLM models, because they rely only on within-family
variation and do not utilize between-family variation in generating
estimates. Second, HLM models allow for the estimation of both
within-family and between-family factors.
Results
Descriptive Statistics by Relationship Status at
24 and 48 Months
Table 2 presents descriptive statistics for each of the outcome
measures, by relationship status, at both 24 and 48 months. Means
that did not differ significantly at the 5% level or lower between
the married family category and the other categories are denoted
by a superscript a. For all of the outcomes, the scores were more
favorable for married families (e.g., mothers in married families
had lower scores on intrusiveness than did parents in other arrangements,
and children in married families scored lower on
negativity than children in other arrangements). All outcomes
differed significantly between married families and the other
groups, with the exception of cohabiting mothers’ negativity score
at 48 months.
Table 3 presents the descriptive statistics for the control variables,
by relationship status, at 24 months. Most of these variables
are time invariant and are the same at 24 and 48 months; means for
variables that are time variant (i.e., child and maternal age, SES,
maternal employment,) are not presented for 48 months but are
available on request (results at 48 months are very similar to those
at 24 months). We also include the percentage of families who
experienced a transition between 9 and 24 months and between 24
and 48 months. Similar to Table 2, outcomes that did not differ
significantly between the married-parent category and the other
relationship categories at the 5% level are denoted with a superscript
a.
Married parents exhibited a very different sociodemographic
profile than parents in other arrangements, because they had more
education, they were older, and their children were more likely to
be non-Hispanic White. Additionally, only 3% of children living in
married families at 24 months experienced a marital transition
during the subsequent 2 years, as compared to 16% of children
living in a cohabiting family and 54% of children in a divorced
family. Differences in SES were particularly pronounced, because
married households had an SES score (.24) that was significantly
higher than that of cohabiting families (.57), never-married families
(.69), and divorced families (.46).
Relationship Status by Year
Table 4 presents children’s living arrangements by survey wave
and by race and ethnicity. Across all racial/ethnic groups, 72% of
156 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
children were living in married-parent families at age 24 months.
Over 18% of children were living in female-headed households
(13% with never-married mothers and 5% with divorced mothers),
and 10% of children were living with cohabiting parents. By age
48 months, the percentage of children living in married-parent
families had decreased slightly to 68%. The percentage of children
living with a divorced mother increased to 8%, and the percentage
of children living in another arrangement, most commonly a stepparent
arrangement, increased to 6%.
When looking separately by racial/ethnic group, clear group
differences emerged. At age 24 months, 86% of White children
lived with married parents, whereas only 34% of Black children
and 62% of Hispanic children lived with married parents. Hispanic
children had the highest rate of living with cohabiting parents
(21%). Black children had the highest rate of living with nevermarried
mothers (47%). Although initial status differed by race
and ethnicity, changes between the two waves were similar across
groups. By age 48 months, the percentage of children living with
married or cohabiting parents or with never-married mothers had
decreased slightly for all groups, whereas the percentage of children
living with a divorced mother or in another arrangement had
increased for all groups.
HLM Results
The results from the HLM models predicting mother– child
interaction outcomes from family structure are presented in
Table 5. The comparison category in all models is biological,
married families. All models include the control variables as
discussed earlier. Models are presented for all children across
racial/ethnic groups and separately by racial/ethnic group. To ease
comparisons, the outcome variables have been standardized to
have a mean of 0 and a standard deviation of 1.
Results for married mothers as reference category.
Across all racial and ethnic groups. The top of Table 5
displays results for all families across race and ethnicity. Families
falling into each of the three unmarried categories— cohabiting,
never married, and divorced—compared unfavorably to marriedparent
families on at least half of the outcomes. Children in
cohabiting families, for example, had worse scores on the two
measures of child behavior, because they were rated as more
negative (
.130, p  .001, 95% CI
.059, .202) and less
engaged (
.146, p  .001, 95% CI
.076, .216) than
children in married families. Cohabiting mothers were also rated as
higher on intrusiveness and lower on cognitive stimulation. Nevermarried
mothers were likewise rated as higher on negativity (

.185, p  .001, 95% CI
.123, .247) and intrusiveness (
.096,
p  .01, 95% CI
.032, .160). Divorced mothers had worse
outcomes on all four measures of parenting behavior, exhibiting
more negativity, intrusiveness, and detachment and less cognitive
stimulation. Effect sizes were generally small but nevertheless
exhibited a clear pattern of more negative child–parent interactions
for unmarried mothers.
Moderation by race/ethnicity. To begin our examination of
moderation by race/ethnicity, we ran each HLM model separately
within each racial and ethnic group. Results for the three racial and
ethnic groups (the bottom of Table 5) indicated that Hispanics had
the most consistent associations between family structure and the
mother– child interaction outcomes. Though there were scattered
Table 2
Descriptive Statistics of Mother–Child Interaction and Transition Variables by Relationship Status at 24 and 48 Months of
Age (N
6,449)
Variable measured
Married Cohabiting Never married Divorced Other relationship
24 months 48 months 24 months 48 months 24 months 48 months 24 months 48 months 24 months 48 months
Child
Negativity
M 1.30 1.30 1.48 1.45 1.52 1.44 1.41 1.44 n/a 1.34
SD 1.33 1.26 1.66 1.97 1.76 1.63 1.55 1.41 1.22
Engagement
M 4.68 4.54 4.12 4.24 4.33 4.37 4.36 4.42 n/a 4.38
SD 2.15 1.67 2.15 1.85 2.15 1.68 2.43 1.44 1.48
Parent
Negativity
M 1.07 1.14 1.13 1.16a 1.28 1.37 1.19 1.28 n/a 1.21
SD 0.61 0.76 0.73 1.03 1.49 1.40 1.14 0.95 0.87
Intrusiveness
M 1.13 1.45 1.29 1.55 1.33 1.73 1.20 1.69 n/a 1.54
SD 0.84 1.49 1.21 1.95 1.47 2.03 1.08 1.59 1.56
Cognitive stimulation
M 4.25 4.32 3.68 3.66 3.85 3.93 3.90 4.03 n/a 4.05
SD 2.07 1.78 1.77 1.97 1.85 1.93 2.25 1.50 1.57
Detached
M 1.04 1.25 1.10 1.36 1.06 1.48 1.12 1.36 n/a 1.32
SD 0.53 1.11 0.75 1.76 0.83 1.68 1.19 1.16 1.24
Sample size 4,659 4,586 602 357 885 728 303 392 0 386
a When compared to the married family category, the other categories had significantly different means on all outcomes at the .05 level or lower, with the
exception of parent negativity at 48 months for cohabiting parents.
FAMILY STRUCTURE 157
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
effects for both Whites and Blacks—White never-married and
divorced mothers were both rated as higher in negativity than
White married mothers, and all three Black unmarried groups
compared unfavorably to Black married mothers on cognitive
stimulation—associations between family structure and mother–
child interactions were primarily concentrated in Hispanic families.
Effects were particularly notable for cohabiting and divorced
Hispanic children and mothers, because cohabiting and divorced
families compared negatively to married families on five of the six
Table 3
Descriptive Statistics at 24 Months by Relationship Status (N
6,449)
Descriptive statistic Married Cohabiting Never married Divorced
Transition
9–24 months .03 .12 .15 .40
24–48 months .03 .16 .15a .54
Demographics
Child’s race and ethnicity
Non-Hispanic White .66 .26 .21 .51
Non-Hispanic Black .06 .17a .48 .19a
Hispanic .21 .52 .23a .20a
Non-Hispanic Asian .03 .01 .01 .01
Other race/ethnicityb .04 .05a .07a .08a
Child’s age (months) 24.3 24.4a 24.4a 24.6a
Mother’s age (years) 31.2 25.3 24.2 28.6a
Child is girl .49 .41a .47a .48a
Child is twin .03 .02a .02a .03a
No. of children in household 2.3 2.4a 2.5 2.6a
No. of adults in household 2.3 2.9 2.4 2.2a
Mother’s age at first birth 25.4 20.4 19.9 21.5
Mother lived with both parents at age 16 .65 .48 .37 .42
Socioeconomic statusc .24 .57 .69 .46
Education
Parentsd
No high school diploma .06 .23 .28 .16a
High school diploma .19 .46 .47 .44
Some college .30 .25a .23a .32a
College graduate .46 .07 .03 .08
Grandparentsd
No high school diploma .20 .44 .32a .29a
High school diploma .25 .27a .34a .28a
Some college .21 .14 .21a .21a
College graduate .34 .15 .13 .23
Maternal employment
Unemployed .03 .10a .19 .11
Out of work force .41 .37 .26 .27a
Employed part-time .22 .19a .20a .13a
Employed full-time .34 .34a .35a .26
Sample size 4,659 602 885 303
a No difference between this family structure type and the married family category. b Includes children who are of
more than one race or ethnicity. c Refers to a composite index that measures parental educational level, occupational
prestige, and household income. Scores range from 4.75 to 2.75, with higher scores indicating a higher level of
socioeconomic status. d Refers to the highest level of educational attainment by either parent (or either grandparent).
Table 4
Living Arrangements by Wave, Race, and Ethnicity
Living arrangement
All races/ethnicities White Black Hispanic
24 months 48 months 24 months 48 months 24 months 48 months 24 months 48 months
Married 72.1 68.0 85.9 80.7 33.9 31.4 62.3 61.8
Cohabiting 9.8 7.0 4.6 2.7 12.4 9.0 21.1 15.3
Never married 13.4 11.8 5.0 4.2 47.2 41.6 12.6 11.2
Divorced 4.7 7.7 4.4 6.5 6.6 11.1 4.0 7.9
Other arrangement 0.0 5.5 0.0 5.9 0.0 6.9 0.0 3.8
Sample sizea 6,449 2,907 985 1,239
a Sample size for all races and ethnicities exceeds that for Whites, Blacks, and Hispanics, because this category includes children of another race or ethnicity.
158 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
outcomes. Effect sizes were modest but were much larger than
those observed for the White or Black samples. Compared to those
living in a married family, Hispanic children living in a cohabiting
family had negativity ratings that were a statistically significant
.215 SD higher ( p  .001, 95% CI
.097, .333) as compared to
a nonsignificant .057 SD for Whites ( p  .10, 95% CI
.085,
.199) and a nonsignificant .024 SD for Blacks ( p  .10, 95% CI

.170, .218).
To test for moderation, we ran additional models in which we
tested the joint significance of the family structure variables’
associations with the mother– child interaction outcomes by race
and ethnicity (results available on request). These tests indicated
that the associations between family structure and mother– child
interactions varied significantly between Hispanics and the other
two racial and ethnic groups on three of the six outcomes: child
negativity, F(3, 12895)
3.29, p
.020; child engagement, F(3,
12895)
3.24, p
.021; and parental detachment, F(3, 12,895)

3.89, p
.009. In addition to testing interactions by race and
ethnicity across all family structure categories, we also tested for
differences by race and ethnicity for cohabiting and divorced
families. We found that the cohabitation coefficient for Hispanics
differed significantly from that of both Whites and Blacks on child
negativity, child engagement, parental cognitive stimulation, and
parental detachment but that the divorced coefficient for Hispanics
differed only on child negativity (test statistics are not presented
but are available on request). Although the association between
divorce and mother– child interactions did vary by race and ethnicity,
these differences were not statistically significant.
Potential mediating effects. Looking next at family stability,
the results indicated that when all races and ethnicities were
combined, experiencing a transition in family structure (from 9 to
24 months or from 24 to 48 months) was generally not associated
with any of the mother– child interaction outcomes. One exception
was a small increase associated with parental cognitive stimulation
for the 9 to 24 month transition measure (
.073, p  .05, 95%
CI
.003, .143). The same pattern of generally null findings was
found within each racial and ethnic group. Notably, though, White
mothers who transitioned between 9 and 24 months displayed
higher parental cognitive stimulation scores, yet White mothers
who transitioned between 24 and 48 months displayed lower
parental cognitive stimulation scores, compared to those who did
not transition. Unfortunately, our data do not allow us to investi-
Table 5
Associations Between Family Structure and Mother–Child Interaction Variables, Hierarchical Linear Modeling Analyses
Family structure
Child outcome Parenting outcome
Negativity Engagement Negativity Intrusiveness
Cognitive
stimulation Detached
 SE  SE  SE  SE  SE  SE
All races/ethnicities
Cohabiting .130


.036 .146


.035 .034 .033 .086
.034 .174


.034 .027 .034
Never married .135


.033 .038 .033 .185


.031 .096

.032 .007 .032 .040 .031
Divorced .106

.040 .049 .039 .144


.038 .107

.039 .068 .038 .106
.038
Other arrangement .053 .061 .027 .059 .012 .057 .023 .059 .022 .057 .019 .058
Transition: 9–24 months .043 .036 .067 .035 .014 .034 .011 .034 .073
.035 .036 .034
Transition: 24–48 months .039 .030 .027 .029 .001 .028 .005 .029 .032 .029 .012 .028
White
Cohabiting .057 .071 .087 .072 .040 .056 .100 .065 .084 .072 .064 .064
Never married .101 .065 .099 .066 .135

.051 .036 .060 .075 .067 .036 .058
Divorced .149
.063 .014 .063 .202


.050 .048 .058 .039 .063 .024 .057
Other arrangement .131 .088 .083 .087 .048 .070 .074 .082 .067 .087 .089 .081
Transition: 9–24 months .044 .058 .059 .060 .054 .046 .089 .053 .137
.060 .009 .052
Transition: 24–48 months .008 .048 .081
.049 .057 .038 .034 .044 .117
.049 .028 .043
Black
Cohabiting .024 .097 .033 .087 .165 .141 .017 .116 .196
.083 .014 .106
Never married .018 .072 .028 .064 .168 .104 .098 .085 .157
.062 .067 .077
Divorced .127 .098 .044 .088 .077 .141 .140 .116 .222

.084 .239
.106
Other arrangement .346
.150 .045 .135 .248 .215 .282 .178 .062 .128 .328
.166
Transition: 9–24 months .125 .081 .070 .073 .170 .118 .075 .096 .129 .071 .076 .087
Transition: 24–48 months .049 .068 .124
.060 .108 .099 .005 .081 .061 .058 .084 .072
Hispanic
Cohabiting .215


.059 .246


.056 .005 .045 .131
.052 .234


.052 .117
.056
Never married .249


.071 .000 .067 .218


.054 .109 .062 .047 .062 .033 .067
Divorced .192
.094 .178
.088 .117 .071 .228

.081 .047 .082 .305


.088
Other arrangement .349
.164 .220 .152 .285
.121 .278
.138 .141 .139 .193 .148
Transition: 9–24 months .099 .081 .017 .077 .018 .062 .005 .072 .079 .072 .083 .078
Transition: 24–48 months .007 .067 .122 .063 .025 .051 .007 .059 .095 .059 .017 .064
Note. For the full sample N
6,449; for the White sample, N
2,907; for the Black sample, N
985; for the Hispanic sample, N
1,239. All models
control for child’s age, mother’s age, child gender, if child is a twin, number of children and adults in household, mother’s age at first birth, mother’s
household structure in her family of origin, socioeconomic status, parental education, grandparents’ education, and maternal employment. Models for all
races and ethnicities include additional controls for race and ethnicity.
 p  .10.
p  .05.

p  .01.


p  .001.
FAMILY STRUCTURE 159
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
gate the disparate effect of an early versus later transition, but we
note that this finding was confined to one racial and ethnic subgroup
and for only one outcome. Overall, a strong pattern of
associations between the transition variables and the outcome
variables did not emerge.
We also tested whether the transition measures mediated the
effect of family structure by comparing results between models
that included the transition variables with models that did not. If
family stability is a primary pathway through which family structure
affects maternal– child interactions, then we would expect the
coefficients associated with family structure to be significantly
larger in magnitude and to have lower p values when the transition
variables were excluded from the models. Our results indicated
that the family structure coefficients did not change significantly
when the transition variables was excluded. For example, among
the full sample, in models that excluded the transition variables,
the coefficient associated with cohabiting families for child negativity
was .132 ( p  .001, 95% CI
.058, .201). The same
coefficient, including the transition measures, was .130 ( p  .001,
95% CI
.059, .202). We used a similar strategy within each
racial and ethnic group and found that the transition variables had
little effect on the family structure covariates.
We also tested for mediating effects of SES and relationship
quality (results are not presented but are available on request). Our
results indicated that the presence of the SES or relationship
quality variables did not significantly change the family structure
coefficients, either for the sample as a whole or for each racial and
ethnic group. We did find, however, that both SES and relationship
quality operated separately from family structure in the expected
directions. Mothers with higher levels of SES and those reporting
higher quality relationships were rated as having more positive
mother– child interactions.
Results for other covariates. The results for the other covariates
are not shown but were as anticipated on the basis of the
existing literature. In general, mothers who had higher levels of
socioeconomic advantage (as indicated by educational status,
smaller household sizes, and older ages at first birth) were associated
with more positive scores on the mother– child interaction
variables. As noted earlier, these socioeconomic advantages may
account for the observed differences between married and unmarried
mothers. We therefore tested our models to see how sensitive
they were to the inclusion of the control variables. In models that
included only the family structure variables and no covariates, all
family types, when compared to married families, displayed significantly
worse mother– child interactions on all of the outcomes
(results are available on request). Once our control variables were
included, however, the magnitude of the differences in family
structure decreased significantly. For example, in models that
included only the family structure variables, the coefficient on
child negativity for cohabiting mother families was .409 ( p 
.01, 95% CI
.343, .475). The same coefficient, once the
control variables were included, was .161 ( p  .01, 95% CI

.091, .230). We conclude that sociodemographic characteristics
account for some of the differences initially observed between
married and unmarried family types.
Results comparing other family structure categories. In
addition to comparing each family structure variable to married
families, we also compared the other family structures to each
other (results are available on request). Few differences among the
cohabiting, divorced, and never-married categories were found.
The one exception was for White cohabiting mothers, who were
rated as significantly lower on parental negativity and cognitive
stimulation than either White never-married or divorced mothers.
Similarly, White children in cohabiting families were rated as
significantly lower on engagement than were White children in
never-married families.
Results for Mexicans and non-Mexican Hispanics. The
consistent negative associations found for Hispanic unmarried
families relative to married families was contrary to our expectations,
because we had expected that the extended social networks
typical of Hispanic families would lessen the potentially negative
effects of nonstandard family structures. However, Hispanics are
by no means a monolithic group, reflecting instead the diverse
experiences, attitudes, and cultural attitudes of a wide variety of
ethnicities (Garcia-Coll & Pachter, 2002). Furthermore, evidence
suggests that Mexicans, in contrast to Hispanics of other ethnicities,
may exhibit fewer positive maternal– child interactions
(Cabrera, Shannon, West, & Brooks-Gunn, 2005). To further explore
results for Hispanics, we ran additional models in which we
divided the Hispanic sample by Mexican origin (small sample
sizes precluded us from considering other Hispanic groups separately).
Approximately two thirds of the 1,239 Hispanics in the
sample were of Mexican origin. The results of these models are
presented in Table 6. The models controlled for the same variables
as described earlier, including the transition variables, but for
comparison’s sake, we present only the results for the three unmarried
family structures (cohabiting, never married, and divorced)
as compared to married, the omitted family structure
category.
The results do not indicate that the negative effects of family
structure were concentrated in either the Mexican or non-Mexican
group. For example, among both Hispanic subgroups, compared to
those in married families, children in cohabiting families displayed
significantly lower levels of engagement, and mothers in cohabiting
families displayed significantly less cognitive stimulation.
Among divorced families, significant negative associations were
found for three of the outcomes in the non-Mexican group (child
engagement, parental intrusiveness, and detached parenting) and
for one of the outcomes for the Mexican group (detached parenting).
Additionally, the signs of the coefficients were generally
consistent across the two groups (one exception was child negativity,
because Mexican children in never-married families displayed
higher levels of negativity than those in married families,
but non-Mexican children in the same family type displayed lower
levels of negativity). In general, effect sizes were markedly larger
in the non-Mexican group, but this may have been driven by the
very small sample sizes. Models with interaction terms (results are
available on request) indicated very few statistically significant
differences between groups. As a final step, we considered
whether the overall results for Hispanics differed by nativity, but
we found few differences in the pattern of effects for Hispanics
born inside the United States as compared to those born elsewhere.
Discussion
This study used a nationally representative birth cohort study to
demonstrate the pervasiveness of differing family structures in the
lives of young children. Consistent with other estimates of diverse
160 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
family structures among children ages 0 to 18 (Fields, 2004;
Kreider & Fields, 2005), we found that by the age of 4, nearly one
third of children were living in a household that did not consist of
two married biological parents. For Black children, married-parent
households were the exception rather than the norm, with only one
third of Black families consisting of a married mother and father.
Hispanic families had the highest rates of cohabitation, with 21%
of Hispanic children living with an unmarried biological mother
and father at age 24 months.
How did these family arrangements affect the mother– child
relationship? Researchers in this area have had difficulty answering
this question, given that there are likely to be unobserved or
unmeasured characteristics that could confound associations between
family structure and mother– child interactions. This is a
particular problem when relying on maternal reports, because such
reports may be biased by the characteristics of the mother. In this
study, we have sought to minimize bias arising from the nonrandom
nature of family structure by utilizing objectively coded
measures of mother– child interactions.
Overall, our findings suggested that both mothers and children
living in unmarried families displayed lower quality interactions
than mothers and children living in biological, married-parent
families. Compared to mothers and children living in married
families, the behavior of mothers and children living in each of the
unmarried arrangements—cohabiting, divorced, and never married—
was consistently rated as displaying more negative qualities
and fewer positive qualities. Children living in all three unmarried
arrangements were rated as higher in negativity, and mothers were
rated as more intrusive. Both cohabiting and divorced mothers also
exhibited lower levels of cognitive stimulation. Never-married and
divorced mothers were also higher on negativity, and divorced
mothers were higher on detachment. Effects of family structure,
however, were moderated significantly by race and ethnicity, a
point we return to later.
Though we had hypothesized that the interactions of mothers
and children living in cohabiting arrangements would be characterized
as higher in quality than the interactions of those living in
divorced or never-married households, our hypothesis was not
confirmed. We found relatively few significant differences in
mother– child interactions between the three unmarried categories.
No clear hierarchy emerged such that mothers and children living
in one unmarried arrangement exhibited consistently higher quality
interactions than those living in the other two. Thus, our
findings do not support the idea that living in a cohabiting family
(vs. a divorced or never-married family) is universally associated
with better outcomes. Given few differences between biological
cohabiting and single-mother (i.e., divorced or never-married)
households, these results do not suggest that lower quality mother–
child interactions arise simply because a family lacks a biological
father. The presence of two parents can ease the emotional strain
of parenting (Belsky, 1984) and may result in higher quality
mother– child interactions. Yet, we did not find differences between
two-parent and one-parent households, and a much clearer
distinction emerged between married and unmarried families. That
distinction does not appear to have been due to differences in
economic standing, relationship quality, or family instability.
We had theorized that relationship transitions may negatively
affect mother– child interactions, as relationship transitions can
strain family dynamics and disrupt typical patterns of parenting
(Amato, 2005; DeGarmo & Forgatch, 1999; Hetherington &
Clingempeel, 1992). Consistent with previous work (Osborne &
McLanahan, 2007), we found that unmarried households were
more likely to experience a relationship transition than were married
households. In our results, however, the transition coefficients
predicting mother– child interactions were generally insignificant.
These results run contrary to previous studies which has found
negative effects of relationship transitions on maternal parenting
(Beck et al., 2009; Cooper et al., in press). These previous studies
Table 6
Associations Between Family Structure and Child–Parent Interaction Variables in the Hierarchical Linear Modeling Analyses
Family structure
Child outcome Parenting outcome
Negativity Engagement Negativity Intrusiveness
Cognitive
stimulation Detached
 SE  SE  SE  SE  SE  SE
Mexican (N
830)
Cohabiting .246


.066 .171

.065 .031 .044 .156

.058 .174

.060 .081 .068
Never married .175
.088 .050 .087 .187

.058 .090 .077 .097 .079 .037 .090
Divorced .131 .125 .108 .122 .106 .083 .093 .109 .043 .112 .340

.129
Other arrangement .232 .200 .035 .191 .047 .134 .007 .169 .144 .175 .010 .199
Transition: 9–24 months .007 .104 .006 .103 .033 .069 .172 .093 .118 .095 .145 .108
Transition: 24–48 months .110 .081 .014 .080 .006 .053 .125 .072 .013 .074 .104 .084
Not Mexican (N
409)
Cohabiting .129 .132 .414


.114 .082 .120 .077 .110 .437


.112 .193 .103
Never married .323
.126 .190 .109 .290
.114 .056 .105 .133 .107 .150 .098
Divorced .126 .149 .278
.128 .030 .130 .355

.123 .086 .122 .258
.113
Other arrangement .432 .285 .491
.249 .714

.235 .616

.235 .655

.229 .529
.208
Transition: 9–24 months .268
.131 .031 .111 .011 .123 .248
.108 .027 .112 .038 .105
Transition: 24–48 months .231 .119 .303

.101 .033 .110 .204
.098 .200
.101 .274

.094
Note. All models control for child’s age, mother’s age, child gender, if child is a twin, number of children and adults in household, mother’s age at first
birth, mother’s household structure in her family of origin, socioeconomic status, parental education, grandparents’ education, and maternal employment.
 p  .10.
p  .05.

p  .01.


p  .001.
FAMILY STRUCTURE 161
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
relied on maternal reports of parent– child interaction, and it is
possible that the differences in reporting methods (self-report
versus interviewer observation) account for the differences in
findings. Additionally, also in contrast to previous work, we had
only two waves of data, and additional waves may be needed to
fully understand the effect of family structure transitions on the
quality of mother– child interactions.
We had further hypothesized that race and ethnicity would
moderate the effects of nonstandard family arrangements on the
quality of the mother– child interactions, given differing economic
contexts and cultural norms surrounding marriage. Our belief was
that differences by family structures would be most evident in
White households, as minority households derive fewer economic
benefits from marriage and have extensive kinship networks that
may buffer the negative effects of unmarried households (Hogan et
al., 1993; Hogan, Hao, & Parish, 1990; Oropesa & Landale, 2004;
Thomas & Sawhill, 2005). Contrary to our hypothesis, we found
the strongest and most consistent associations between family
structure and mother– child interactions for Hispanics. Indeed,
most of the positive associations between marriage and parenting
found for the overall sample appear to be driven by Hispanics,
because few significant associations were found for Whites and
Blacks.
As compared to Hispanic married families, the quality of
mother– child interactions was particularly negative in Hispanic
cohabiting and divorced families. Significant differences were
found between married and cohabiting families, and between married
and divorced families, for six of the seven outcomes considered.
Effects were consistent across Mexicans and non-Mexican
Hispanics. These results are consistent with Mechanic and Hansell’s
(1989) work documenting that Hispanic adolescents, relative
to Whites or Blacks, were more sensitive to the negative effects of
parental conflict in the context of divorce. However, the reasons
why detrimental effects of cohabitation and divorce might be more
acute for Hispanics are unclear.
One possibility is that Hispanic women in these relationships
face unique economic and cultural constraints. Hispanic families
may be characterized by more traditional gender role beliefs
(Oropesa & Gorman, 2000), such that Hispanic women may be
expected to shoulder more of the burden of keeping house and
raising children (Shelton & John, 1996). Particularly for cohabiting
couples, who are less likely to pool resources (Bauman, 1999),
it is possible that Hispanic women find themselves with fewer
economic resources but more household responsibilities. Traditional
gender role beliefs may also inform the burden of divorce;
though the stigma may be lessening, divorced Hispanic women
may face social censure not present in other cultures (Oropesa &
Landale, 2004). Social censure could mean that divorced Hispanic
women feel more isolated from their community, thereby exacerbating
parental stress and anxiety. Clearly, more work needs to be
done as to how the particular social and cultural context surrounding
nonstandard family arrangements influences mother– child interactions.
Limitations to our study should be noted. First, as noted earlier,
our models can at best demonstrate associations between family
structure and mother– child interactions, but they cannot demonstrate
causality. The associations between family structure and
mother– child interactions may represent the presence of unobserved
characteristics which are correlated both with marital status
and parenting practices. Most likely, these unobserved characteristics
would cause an upward bias in estimates; that is, findings
would overattribute to marriage effects that are actually caused by
personal factors. Though our overall results indicate positive associations
between marriage and parenting, these effects are concentrated
among Hispanics, with few significant associations
found for Whites or Blacks. If our findings for marriage are
upwardly biased by omitted variables, then this bias must differ by
racial and ethnic status. This does not mean that we have produced
unbiased coefficients but does indicate that our results for Whites
and Blacks are inconsistent with the most likely omitted variable
bias explanation.
Second, previous studies have indicated that single-mother and
cohabiting families experience multiple transitions (Osborne &
McLanahan, 2007), and it is possible that there are family transitions
that occurred between the two waves of data that were not
observed. Unfortunately, we cannot quantify the extent of this
problem, though we note that only 15% of cohabiting and singlemother
families experienced a transition between 24 and 48
months, so the fraction experiencing more than one transition is
likely to be smaller. Third, the ECLS-B does not contain information
on potential mediating mechanisms, such as maternal depression
or social support, which could explain the association between
family structure and child well-being. We were able to test the
effect of three mechanisms: family transitions, SES, and relationship
quality. As noted earlier, these measures had little effect on
our estimates. It is possible, of course, that including other mediators
would change our results.
Finally, the observational measures of mother– child interaction,
though used in other ethnically diverse samples, such as the Early
Head Start evaluation (Love et al., 2005), have only recently begun
to be validated among children of color. It is possible that the
outcomes examined have different meanings across cultures. Even
if the outcomes are culturally biased, we have minimized this
problem by focusing on within racial and ethnic group comparisons,
rather than between racial and ethnic comparisons.
Despite these limitations, this study highlights the importance of
considering race and ethnic differences in family structure. In
keeping with previous studies, our results do indicate a benefit of
biological married-parent families over other, unmarried family
structures. At the same time, however, these results were not
uniform across ethnic groups and were concentrated among Hispanics.
Results from this study suggest that we should not expect
marriage to be associated with higher child functioning in all
families.
References
Acs, G., & Nelson, S. (2002). The kids are alright? Children’s well-being
and the rise in cohabitation. Washington, DC: Urban Institute.
Amato, P. R. (2005). The impact of family formation change on the
cognitive, social, and emotional well-being of the next generation.
Future of Children, 15, 75–96.
Andreassen, C., & Fletcher, P. (2007). Early Childhood Longitudinal
Study, Birth Cohort (ECLS-B) psychometric report for the 2-year data
collection (NCES 2007–084). Washington, DC: Institute of Education
Sciences, U.S. Department of Education.
Bauman, K. J. (1999). Shifting family definitions: The effect of cohabitation
and other nonfamily household relationships on measures of poverty.
Demography, 36, 315–325.
162 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Beck, A., Cooper, C., McLanahan, S., & Brooks-Gunn, J. (2009). Relationship
transitions and maternal parenting (Working Paper 2008–12-
FF). Princeton, NJ: Princeton University.
Belsky, J. (1984). The determinants of parenting: A process model. Child
Development, 55, 83–96.
Belsky, J. (1990). Parental and nonparental child care and children’s
socioemotional development: A decade of review. Journal of Marriage
and the Family, 52, 885–903.
Belsky, J. (1999). Interactional and contextual determinants of attachment
security. In J. Cassidy & P. Shaver (Eds.), Handbook of attachment:
Theory, research, and clinical applications (pp. 249–264). New York:
Guilford.
Berlin, L., & Cassidy, J. (1999). Relations among relationships: Contributions
from attachment theory and research. In J. Cassidy & P. R. Shaver
(Eds.), Handbook of attachment: Theory, research and clinical applications
(pp. 688–712). New York: Guilford Press.
Bornstein, M. H. (Ed.). (1995). Handbook of parenting. Mahwah, NJ:
Erlbaum.
Bradley, R. H., Corwyn, R. F., McAdoo, H. P., & Coll, C. G. (2001). The
home environment of children in the United States: Part 1. Variations by
age, ethnicity, and poverty status. Child Development, 72, 1844–1867.
Brown, S. L., & Booth, A. (1996). Cohabitation versus marriage: A
comparison of relationship quality. Journal of Marriage and the Family,
58, 668–678.
Bumpass, L., & Lu, H.-H. (2000). Trends in cohabitation and implications
for children’s family contexts in the United States. Population Studies,
54, 29–41.
Cabrera, N. J., Shannon, J. D., West, J., & Brooks-Gunn, J. (2005).
Parental interactions with Latino infants: Variation by country of origin
and English proficiency. Child Development, 77, 1190–1207.
Capaldi, D. M., & Patterson, G. R. (1991). Relation of parental transitions
to boys’ adjustment problems: I. A linear hypothesis. II. Mothers at risk
for transitions and unskilled parenting. Developmental Psychology, 27,
489–504.
Carlson, M. J., & McLanahan, S. (2006). Strengthening unmarried families:
Could enhancing couple relationships also improve parenting?
Social Service Review, 80, 297–321.
Conger, R. D., Conger, K. J., Elder, G. H., Jr., Lorenz, F. O., Simons, R. L.,
& Whitbeck, L. B. (1992). A family process model of economic hardship
and adjustment of early adolescent boys. Child Development, 63,
526–541.
Conger, R. D., Patterson, G. R., & Ge, X. (1995). It takes two to replicate:
A mediational model for the impact of parents’ stress on adolescent
adjustment. Child Development, 66, 80–97.
Cooper, C., McLanahan, S., Meadows, S., & Brooks-Gunn, J. (in press).
Family structure transitions and maternal parenting stress. Journal of
Marriage and Family.
Cox, M. J., Owen, M. T., Lewis, J. M., & Henderson, V. K. (1989).
Marriage, adult adjustment, and early parenting. Child Development, 60,
1015–1024.
Cummings, E. M., & Davies, P. T. (2002). Effects of marital conflict on
children: Recent advances and emerging themes in process-oriented
research. Journal of Child Psychology and Psychiatry and Allied Disciplines,
43, 31–63.
DeGarmo, D. S., & Forgatch, M. S. (1999). Contexts as predictors of
changing maternal parenting practices in diverse family structures: A
social interactional perspective to risk and resilience. In E. M. Hetherington
(Ed.), Coping with divorce, single parenting and remarriage: A
risk and resiliency perspective (pp. 227–252). Hillsdale, NJ: Erlbaum.
Demo, D. H., & Cox, M. J. (2000). Families with young children: A review
of research in the 1990s. Journal of Marriage and the Family, 62,
876–895.
De Wolff, M. S., & van IJzendoorn, M. H. (1997). Sensitivity and attachment:
A meta-analysis on parental antecedents of infant attachment.
Child Development, 68, 571–591.

Erel, O., & Burman, B. (1995). Interrelatedness of marital relations and
parent– child relations: A meta-analytic review. Psychological Bulletin,
118, 108–132.
Federal Interagency Forum on Child and Family Statistics. (2007). America’s
children: Key national indicators of well-being, 2007. Retrieved
from http://www.childstats.gov/americaschildren/eco1.asp
Fields, J. (2004). America’s families and living arrangements: 2003.
Washington, DC: U.S. Census Bureau.
Garcia-Coll, C., & Pachter, L. M. (2002). Ethnic and minority parenting. In
M. Bornstein (Ed.), Handbook of parenting (2nd ed., Vol. 4, pp. 1–20).
Mahwah, NJ: Erlbaum.
Gibson-Davis, C. M. (2008). Family structure effects on maternal and
paternal parenting in low income families. Journal of Marriage and
Family, 70, 452–465.
Graefe, D. R., & Lichter, D. T. (1999). Life course transitions of American
children: Parental cohabitation, marriage, and single motherhood. Demography,
36, 205–217.
Graefe, D. R., & Lichter, D. T. (2002). Marriage among unwed mothers:
Whites, Blacks and Hispanics compared. Perspectives on Sexual and
Reproductive Health, 34, 286–294.
Hao, L. (1996). Family structure, private transfers, and the economic
well-being of families with children. Social Forces, 75, 269–292.
Hetherington, E. M. (1999). Should we stay together for the sake of the
children? In E. M. Hetherington (Ed.), Coping with divorce, single
parenting, and remarriage (pp. 93–116). Mahwah, NJ: Erlbaum.
Hetherington, E. M., & Clingempeel, G. (1992). Coping with marital
transitions: A family systems perspective. Monographs of the Society for
Research in Child Development, 57, Serial No. 227.
Hetherington, E. M., & Jodle, K. M. (1994). Stepfamilies as settings for
child development. In A. Booth & J. Dunn (Eds.), Stepfamilies as
settings for child development (pp. 55–79). Hillsdale, NJ: Erlbaum.
Hetherington, E. M., & Stanley-Hagan, M. (1999). The adjustment of
children with divorced parents: A risk and resiliency perspective. Journal
of Child Psychology and Psychiatry, 40, 129–140.
Hogan, D. P., Eggebeen, D. J., & Clogg, C. C. (1993). The structure of
intergenerational exchanges in American families. American Journal of
Sociology, 98, 1428–1458.
Hogan, D. P., Hao, L.-X., & Parish, W. L. (1990). Race, kin networks, and
assistance to mother only families. Social Forces, 68, 797–812.
Ispa, J. M., Fine, M. A., Halgunseth, L. C., Harper, S., Robinson, J., Boyce,
L., et al. (2004). Maternal intrusiveness, maternal warmth, and mother–
toddler relationship outcomes: Variations across low-income ethnic and
acculturation groups. Child Development, 75, 1613–1631.
Kreider, R. M., & Fields, J. (2005). Living arrangements of children: 2001.
Washington, DC: U.S. Census Bureau.
Lerman, R. (2002). Impacts of marital status and parental presence on the
material hardship of families with children. Washington, DC: Urban
Institute.
Love, J. M., Kisker, E. E., Ross, C., Raikes, H., Constantine, J., Boller, K.,
et al. (2005). The effectiveness of early Head Start for 3-year-old
children and their parents: Lessons for policy and programs. Developmental
Psychology, 41, 885–901.
Manning, W. D., & Lichter, D. T. (1996). Parental cohabitation and
children’s economic well-being. Journal of Marriage and the Family,
58, 998–1010.
Manning, W. D., Smock, P. J., & Majumdar, D. (2004). The relative
stability of cohabiting and marital unions for children. Population Research
and Policy Review, 23, 135–159.
Martin, J. A., Hamilton, B. E., Sutton, P. D., Ventura, S. J., Menacker, F.,
Kirmeyer, S., & Mahews, M. S. (2009). Births: Final data for 2006.
Hyattsville, MD: National Center for Health Statistics.
McLanahan, S., & Sandefur, G. D. (1994). Growing up with a single
FAMILY STRUCTURE 163
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
parent: What hurts, what helps. Cambridge, MA: Harvard University
Press.
McLoyd, V. C. (1990). The impact of economic hardship on Black families
and children: Psychological distress, parenting, and socioemotional development.
Child Development, 61, 311–346.
McLoyd, V. C. (1998). Socioeconomic disadvantage and child development.
American Psychologist, 53, 185–204.
Mechanic, D., & Hansell, S. (1989). Divorce, family conflict, and adolescents’
well-being. Journal of Health and Social Behavior, 30, 105–116.
Miller, B. C., Fan, X., Christensen, M., Grotevant, H. D., & van Dulmen,
M. (2000). Comparisons of adopted and non-adopted adolescents in a
large, nationally representative sample. Child Development, 71, 1458–
1473.
National Institute of Child Health and Human Development Early Child
Care Research Network. (1999). Child care and mother– child interaction
in the first 3 years of life. Developmental Psychology, 35, 1399–
1413.
Nock, S. L. (1995). A comparison of marriages and cohabiting relationships.
Journal of Family Issues, 16, 53–76.
Oropesa, R. S., & Gorman, B. R. (2000). Ethnicity, immigration, and
beliefs about marriage as a “tie that binds”. In L. Waite, C. Bachrach, M.
Hindin, E. Thomson, & A. Thornton (Eds.), The ties that bind: Perspectives
on marriage and cohabitation (pp. 188–211). New York: de
Gruyter.
Oropesa, R. S., & Landale, N. S. (2004). The future of marriage and
Hispanics. Journal of Marriage and Family, 66, 901–920.
Osborne, C., & McLanahan, S. (2007). Partnership instability and child
well-being. Journal of Marriage and Family, 69, 1065–1083.
Popenoe, D., & Whitehead, B. D. (2006). The state of our unions 2006.
Piscataway, NJ: Rutgers University.
Shaw, D. S., Winslow, E. B., Owens, E. B., & Hood, N. (1998). Young
children’s adjustment to chronic family adversity: A longitudinal study
of low-income families. Journal of the American Academy of Child and
Adolescent Psychiatry, 37, 545–554.
Shelton, B. A., & John, D. (1996). The division of household labor. Annual
Review of Sociology, 22, 299–322.
Snow, K., Thalji, L., Derecho, A., Wheeless, S., Lennon, J., Kinsey, S., et
al. (2007). User’s manual for the ECLS-B longitudinal 9-month–
preschool restricted-use data file and electronic codebook. Washington,
DC: National Center for Education Statistics.
Stewart, S. D. (2007). Brave new stepfamilies. Thousand Oaks, CA: Sage.
Thomas, A., & Sawhill, I. (2005). For love and money? The impact of
family structure on income. Future of Children, 15, 57–74.
Thomson, E., Mosley, J., Hanson, T. L., & McLanahan, S. S. (2001).
Remarriage, cohabitation, and changes in mothering behavior. Journal
of Marriage and Family, 63, 370–380.
Voydanoff, P., & Donnelly, B. W. (1998). Parents’ risk and protective
factors as predictors of parental well-being and behavior. Journal of
Marriage and the Family, 60, 344–355.
Whitbeck, L. B., Simons, R. L., Conger, R. D., Wickrama, K. A. S.,
Ackley, K. A., & Elder, G. H., Jr. (1997). The effects of parents’
working conditions and family economic hardship on parenting behaviors
and children’s self-efficacy Social Psychology Quarterly, 60, 291–
303.
Yeung, W. J., Linver, M. R., & Brooks-Gunn, J. (2002). How money
matters for young children’s development: Parental investment and
family processes. Child Development, 73, 1861–1879.
Received July 21, 2008
Revision received July 20, 2009
Accepted July 27, 2009

E-Mail Notification of Your Latest Issue Online!
Would you like to know when the next issue of your favorite APA journal will be available
online? This service is now available to you. Sign up at http://notify.apa.org/ and you will be
notified by e-mail when issues of interest to you become available!
164 GIBSON-DAVIS AND GASSMAN-PINES
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

1
Step by Step Directions for the
Journal Article Review Paper
1. Know the purpose of this assignment.
One main objective of this assignment is to learn how to find first hand information on a topic
using a research study. (Research studies are originally published as articles in Academic Journals.
Newspapers, magazines etc. get the information from the articles. So when you hear of a study on the
news, it’s possible it was misinterpreted or misleading). With this assignment, you will get to see the
first-hand, direct source of study information. You will then gain experience summarizing the study
and writing the information in your own words, in a way that your friends/family could understand.
Finally, you will have the opportunity to critique how the study was done, and suggest changes for
future research.
2. Getting started – Finding a Journal Article on a Research Study.
The most important part of starting and writing the Journal Article Review Paper is that you
find a journal article that explains one specific study on any topic that has to do with child
development. To find a suitable journal article, you need to use the “Directions for Obtaining Journal
Articles for Research” (on page 3). It is a requirement that you find your article from the Palomar
Library. You cannot find information like you need just off of the internet, so please use those
directions provided!
3. Double Check to make sure this article is on one specific study.
The quickest way to see if your article is explaining one specific study is to skim through the
article and look for paragraph headings including “methods” “results” and “discussion” that usually
means they are focused on one study. The discussion section is the easiest to understand usually (they
tend to sum things up there and use more everyday language) So, I recommend reading that section
first, then going back and reading the entire article.
Also, read through the “Paper Requirements” document (on page 4). Print it out and have it in
front of you while searching for an article. If you find the answers to the questions in an article, the
article should work for you. If not, the article is probably not describing just one specific study.
4. You found an Article! Now be sure to save it to your computer.
The website will let you email it to yourself, or save it directly to your computer. Make sure
you save it electronically. (Do not just save the link as the link may become invalid)
5. Email a copy of your Journal Article to the Professor
(This is required! Email by the Due Date listed in the syllabus!)
Be sure to save it to your computer first, then email me a copy. Do not use the feature off the
website to send it to me, I will not be able to tell that article is from you and easily reply back to you.
Do not email me a link to the paper. Email me the article as an attachment.
Within 2 days you should receive an email back, approving your journal article. (I am just checking to make sure you are
on the right track, and have a study that will work)
6. Start Your Paper
Once your article is approved, print the article. Use the “Paper Requirements” (on page 4) to
know what you need to write about. Highlight that information in the article. (I expect that when I see the
printed copy of the article at the notebook check that it will be highlighted and written on). When writing your
2
paper, be sure to include all of the information asked. Make sure you are putting all that information
in your own words. You cannot take sentences, or even parts of sentences that are in the article and
write them in your paper… this is Plagiarizing! DO NOT use quotes, I want to read a paper all
written by you! This paper should be in APA format, and a minimum of 6 pages long (that is
including the title page, abstract page, and reference page)
Remember to put your paper in APA format and follow all of the Writing Tips (available on
Blackboard). Papers that are turned in and are not in APA format will be returned without a grade.
7. Proofread Your Paper
Make sure you have included all necessary information listed on the “Guidelines for Journal
Article Review Paper.” Use the items in the APA Information & Writing Tips section for APA
guidelines and to improve your grammar and writing style. Also, have someone else read through
your paper to help find typos and make sure all of your sentences are clear and make sense. Make
sure all of your information is in your own words!
If you would like to have the professor give feedback, email a rough draft of your paper along with
the journal article at least 3 days (preferably 5 days) prior to the day the assignment is due. Be sure to
note in your email that that is your rough draft of your paper and you are requesting feedback.
(Otherwise it might be assumed that you are just turning in a final draft early.)
8. Turn in Your Journal Article Review Paper and a Copy of the Journal Article
(Submit by Due Date listed in the Syllabus)
Turn in your paper and article via email to chdvonline@cox.net . In the subject line put “your
name, Review Paper.” Include your course number in the email as well. Send both your paper, as a
Microsoft Word document (.doc) and a copy of your journal article each as a file attachment in one
email. Put a hard copy of your paper, with your journal article stapled to it, into your notebook.
9. Wait for Your Grade
Within two days after the due date you will receive an email confirming that your paper was
received. If you turned in your paper on time, make sure to look for this email. Within one-two weeks
of the due date you will receive an email with your paper grade. It will be sent as an attachment on a
grading rubric. Any items highlighted on the rubric note feedback regarding your paper and should
help explain why you received the number of points you did on the paper. At this time, your grade
will also be posted in the “my grades” section on Blackboard.
3
Directions for Obtaining Journal Articles for Research – Palomar College
1. Go to http://www.palomar.edu/library/
2. On the column on the left under click on “Databases/Articles”
3. The next screen will ask you to choose from General Databases. Click on Open Database for
Academic Search Premiere (EBSCO) or, if you are picking a medical related topic also select
Alt HealthWatch, MEDLINE and CINAHL Plus. Then click on continue.
4. If you are on a computer off campus, you will then be asked to enter your Student ID # and
password. (This is the same information you give to log onto Eservices and Blackboard).
From the school library, it should open automatically.
5. Enter a topic you are interested in, (start broad if you are unsure of what you want to study)
Then put AND Methods. Example: Child AND Methods or Pregnancy AND Methods or
Autism AND Methods etc.
(attaching the words AND Methods to whatever topic you are searching is just a trick that will
greatly improve your search results and eliminate many articles that will not work for this
paper.)
6. To the left of the screen under Limit your results be sure to mark the “Full Text” box and
“Scholarly (Peer Reviewed) Journals” box. Don’t worry about any of the other boxed
7. Click on “Search”
8. Choose titles that fit what you are looking for. Make sure they are from academic journals
(sometimes there will be book reviews). It will also show you how long the articles are. Most
articles appropriate for this paper are 3-30 pages long. (If you see an article that is only one
page it will not work for this assignment)
9. Click on the title or HTML Full Text or PDF Full Text to open the article
10. If you did not find any articles that fit what you needed, be patient; try searching with other
key words. Start broad and then get more specific.
11. When you find an article(s) that you like, be sure to email them to yourself or save them to
your computer.
4
Paper Requirements – Journal Article Review Paper
Include ALL of the following:
Title Page in APA Format (page 1)
Follow the directions in the APA Essentials Document and make yours look just like the Sample APA Paper
on Blackboard
Abstract Page (page 2)
Follow APA format. Write one paragraph that summarizes your paper. An informational abstract states the
purpose, methods, scope, results, conclusions, and recommendations included in your report.
Body of paper (pages 3-4, or maybe 3-5 depending on how you write)
1st Paragraph = Introduction
Provide general background information on the topic before even mentioning your specific study. Define
and explain any special terminology used in the paper. (you might need another reference source to
complete this portion.
In the body of your paper,
Introduce the Study by its title. Explain the study. In YOUR OWN WORDS include the following:
Who conducted the research study?
When and where did the study take place? (or what year was it published?)
What was researched?
What populations were researched? (Who were the participants?) How did they pick participants? What
was the sample size?
How did they conduct the study?
What research method did they use? (Observation, case study, survey, interviews, experiment etc. – see chapter 1
of your text for descriptions)
What were the findings of the study?
How do the findings of this study apply to the field of Child Development? How can they be used? (this you
will have to think about on your own !)
*Stay objective throughout your paper. Do not give your opinion or use “I” until the conclusion.
Last paragraph = Conclusion – now give your opinions and use “I” Answer ALL the following:
Did the results surprise you? Why/why not? (Relate them to personal/professional experience related to the topic).
What do you think were the strong points of how this study was conducted? (not how the article was writtenfocus
on how they did the study)
What do you think were the weak points of how this study was conducted?
What factors were not taken into account?
If you were conducting a similar study, what would you do differently?
Reference Page in APA Format (last page) – Cite your article
Your reference page may only have the one article you responded to. That is fine; just make sure to put it in
APA format. If you used another source to help write your introduction page, include it too.
Type your paper in Times New Roman Font size 12, double-spaced. Use APA format for your entire
paper (see APA Essentials, and the sample Paper on Blackboard). In text citations are not needed for
this paper since all of your information is coming from just one source. Paper should be at least 6 pages
long, including title, abstract & reference page. Don’t forget to attach your journal article!
PLACE THIS ORDER OR A SIMILAR ORDER WITH US TODAY AND GET AN AMAZING DISCOUNT 🙂