Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 1.
Published in final edited form as: Child Dev. 2018 Jun 19;90(6):e729–e744. doi: 10.1111/cdev.13103

Family Turbulence and Child Internalizing and Externalizing Behaviors: Moderation of Effects by Race

Sean R Womack a,*, Lindsay Taraban b, Daniel S Shaw b, Melvin N Wilson a, Thomas J Dishion c
PMCID: PMC6301136  NIHMSID: NIHMS972577  PMID: 29921025

Abstract

This study examined the impact of residential instability and family structure transitions on the development of internalizing and externalizing problems from age 2 through 10.5. Child’s race was examined as a moderator. Caregiver reports of internalizing and externalizing behaviors were obtained on 665 children at ages 5 and 10.5. Early-childhood residential and family structure transitions predicted elevated internalizing and externalizing problems at ages 5 and 10.5, but only for Caucasian children. These findings suggest that residential and family structure instability during early childhood independently contribute to children’s later emotional and behavioral development, but vary as a function of the child’s race. Community organizations (e.g., WIC) can connect turbulent families with resources to attenuate effects of residential and family structure instability.

In the context of child development, turbulence has been most commonly defined as any events that cause temporary or permanent disruption to the day-to-day structure or routines of a child’s life (Adam, 2004). Turbulence often takes the form of acute events that directly involve changes in the child’s environment, such as changes in family structure or moving residences. Transitions in caregivers or residence lead to periods where normal roles are disrupted and new routines are established, and are often experienced as stressful by the child (Jackson & Warren, 2000). While isolated exposure to such disruptive events may be relatively benign, prolonged exposure to chronic residential mobility or family structure instability can stress the family system and ultimately have negative implications for a child’s social, emotional, behavioral, and cognitive development (for a review, see Jellyman & Spencer, 2008). Among families with limited parenting resources, poor interfamily relationships, and limited economic resources, weathering turbulence may strain already vulnerable areas of the family system, exacerbating existing problems within the family.

Economically marginalized families are at a heightened risk for both residential and family structure instability. According to the U.S. Census Bureau, in 2009 26.5% of families living below the poverty threshold moved compared to just 11.7% of families living at or above 150% of the poverty line (Ihrke, Faber, & Koerber, 2011). Moreover, sociological research suggests that poor families tend to move sporadically and for more negative reasons (e.g., job loss, eviction) than more affluent families, potentially adding to the stress of a move (Fitchen, 1994). Further, poverty and parenting-figure transitions are themselves related, as economic stress has been associated with a greater likelihood of relationship dissolution (Cooper, McLanahan, Meadows, & Brooks-Gunn, 2009). Moreover, parents in lower-SES families are more likely to cohabitate with partners than marry, relative to parents in higher-SES houses (Manning & Brown, 2006). This is important given that compared to married couples, unmarried cohabiting partners are more likely to separate, putting their children at higher risk for experiencing the dissolution of their parents’ relationship (Cavanagh & Huston, 2006). For example, it is estimated that 50% of children born to unmarried cohabiting parents will experience the dissolution of their parent’s relationship, compared to just 20% of children born to married parents (Kennedy & Bumpass, 2008).

Some researchers have combined residential mobility and family structure instability into a single construct of “environmental unpredictability,” which has been linked to maladaptive development (e.g., Belsky, Schlomer, & Ellis, 2012). Although residential mobility and family structure instability share common elements of environmental turmoil, they have important differences that make it useful to study the constructs independently. Specifically, family structure instability involves a change in the composition of the family household and changes in support within the home, while residential mobility involves disruptions in the external world, including changes in proximity to extrafamilial support systems and other community resources, as well as disruptions within the home related to moving and reorganizing one’s belongings. Thus, it is possible that the unique stress of these dissimilar types of transitions may be differentially associated with child development outcomes.

An Ecological Perspective on Child Development

Consistent with Bronfenbrenner’s (1979) ecological systems theory, exposure to turbulence can disrupt multiple related subsystems at critical developmental periods, leading to a myriad of negative child outcomes. For example, frequent residential mobility can create disruptions that reverberate across the microsystem and mesosystem through changes in proximity to social ties (e.g., family, friends, teachers, neighbors) and community resources (e.g., schools, community centers), causing stress for both children and their parents (Wilson, Tolson, Hinton, & Kiernan, 1990). Similarly, changes in family structure may modify a family’s parenting hierarchy, day-to-day routines, and rules or disciplinary practices, as parenting figures enter or leave the family system.

In addition, the timing of turbulent events in relation to a child’s developmental stage is crucial. Exposure to chronic turbulence in early childhood may disrupt the parent-child relationship at a time when the child is more dependent on caregivers relative to later years (Shaw & Bell, 1993). Parents in highly turbulent families devote significant time and energy to the logistics of securing housing or dealing with changes in their own relationship status. A child vying for depleted parental attention may escalate disruptive behaviors, increasing the frequency of coercive parent-child interactions, leading to later behavior problems (Scaramella & Leve, 2004; Shaw et al., 1998). Alternatively, in the context of a highly turbulent household, some children may withdrawal from parenting figures, leading to symptoms of anxiety (Rodriguez, 2011).

In middle childhood, stress from turbulence may have different mechanisms of impact on child behavior. For example, stress from turbulence may compromise parents’ abilities to monitor their children, creating opportunities for children to become more involved with deviant peers and engage in delinquent behaviors. Research suggests that parents of residentially mobile families tend to be less aware of who their children’s peers are, relative to non-mobile families (South and Haynie, 2004). Additionally, school-aged youth with turbulent home lives are at an elevated risk to attempt suicide, which may be a product of increased incidence of being a victim of bullying following a move to a new school (Haynie, South, & Bose, 2006a). To date, however, there has been little research on the impact of turbulence at different developmental periods.

Family Structure Instability

Transitions in parenting figures during childhood have been repeatedly linked to child emotional (Frojd et al., 2012) and behavioral problems (Capaldi & Patterson, 1991; Cavanagh & Huston, 2006; Shaw, Winslow, & Flanagan, 1999). The mechanism by which family structure transitions impact child development is complex and likely dependent on a number of individual and contextual factors (Amato, 2000; Goodnight et al., 2013; Martinez & Forgatch, 2002). For example, a change in economic resources following a separation may financially restrict a family, forcing a newly single parent to enter the workforce or work overtime, creating more unsupervised time for children to engage in problematic behaviors (McMunn et al., 2001; Ram & Hou, 2003). Alternatively, parents may become depressed or engage in less attentive parenting practices as they cope with the dissolution of a previously close relationship (Martinez & Forgatch, 2002). Conflict between parents prior to or during a separation also has consistently been linked to child emotional and behavior problems (Davies & Cummings, 1994; Ingoldsby et al., 1999; Shaw, Emery, & Tuer, 1993). Finally, while the addition of a new cohabiting parenting figure to the home may bring an increase in financial resources to the family, this new entity also brings a restructuring of household routines, which may be stressful for children adjusting to a new parenting hierarchy (Tolson & Wilson, 1990; Wilson, Tolson, & Kiernan 1989).

Much of the extant research on family structure instability is based on the “frequency” hypothesis, which suggests a linear relationship between the number of transitions children experience and subsequent behavior problems (Capaldi & Patterson, 1991). However, such models do not take into account the relationship the transitioning partner has with members of the household. By omission, frequency models suggest that transitions involving a biological father have an equivalent psychological impact on children as transitions involving a short-term, live-in boyfriend. One alternative to a frequency approach is the use of a weighted approach, where transitions involving primary caregivers, biological parents, and step-parents would be weighted more heavily than transitions involving non-marital, cohabiting partners who may be less involved in parenting or have a weaker attachment to the child.

While previous research has consistently linked transitions in parenting figures to poor emotional and behavioral outcomes, there are notable differences in the consistency of these findings by race. For example, Shaw and colleagues (1999) found that African American children experience a higher frequency of separations among parenting figures than do Caucasian children from ages 1.5 to 5 years, but that Caucasian children in to-be-divorced families showed significantly higher rates of externalizing problems compared with African American children in to-be-divorced families. Likewise, Fomby and Cherlin (2007) found that African American children in the National Longitudinal Study of Youth experienced more parenting figure transitions compared to Caucasian children, but transitions only predicted higher levels of externalizing problems among Caucasian children. Thus, research has not only shown that the structure of African American families is more fluid compared to Caucasian families, but also indicates that African American and Caucasian children respond differently to these transitions (Fomby & Cherlin, 2007; Shaw et al., 1999).

Residential Instability

While less frequently studied in relation to child behavior outcomes compared to family structure instability, exposure to residential transitions has also been linked to childhood internalizing (Jellyman & Spencer, 2008; Simpson & Glenn-Fowler, 1994; Tiesler et al., 2013) and externalizing problems (Jellyman & Spencer, 2008; Parente & Mahoney, 2009; Simpson & Glenn-Fowler, 1994). Although moving is often related to opportunities (e.g., moving to a better school district, moving for a new job), families living in poverty tend to move in parallel or downward trajectories (e.g., to comparably disadvantaged or poorer neighborhoods) and are more likely to move because of negative reasons (e.g., eviction, job loss) relative to wealthier families (Fitchen, 1994). Additionally, frequently moving families are likely to be disconnected from community resources, making it difficult to utilize sources of economic and psychological support.

Moving may influence child behavior in different ways depending on the child’s developmental stage. For example, frequent mobility in the early years may detract from the time parents spend interacting with their children at a developmental period where positive parent-child interactions are critical to healthy development (Bowlby, 2008). Alternatively, mobility during the school-age years may involve changing schools and moving away from neighborhood peers (Dewitt, 1998). Children who move schools or neighborhoods frequently tend to be socially isolated and be less popular within their new peer networks (South & Haynie, 2004), and may be drawn to more antisocial peers who are open to accepting new peers into their social circles (Haynie, South, & Bose, 2006b). Additionally, among families who move, parents tend to have less knowledge of who their children’s peers are, increasing opportunities for mobile children to engage with deviant peers (South & Haynie, 2004). While most children eventually break into existing peer networks and parents learn who their children’s friends are (Haynie et al., 2006b; South & Haynie, 2004), chronically moving families may never become fully immersed in social networks within the communities that they live.

The Present Study

There is a sound foundation of research consistent with the notion that exposure to multiple residential and parenting figure transitions represents a significant disruption in children’s lives by stressing the family system. Previous studies have consistently linked repeated residential and parenting figure transitions to both internalizing (Frojd et al., 2012) and externalizing behaviors (Cavanagh & Huston, 2006; Tiesler et al., 2013). Still, many questions remain unanswered. To date, there has been little research examining the impact of turbulence at different developmental periods on child behavior and mental health outcomes. Similarly, the extant body of turbulence literature has primarily focused on Caucasian, middle class families, limiting the ability of researchers to determine how variations in culture and economic resources respond to turbulence. The limited research on low-SES and minority families is surprising based on national statistics that suggest low-SES families move more frequently and have more transitions in family structure. Previous studies have focused on the independent effects of transitions in parenting figures or residence, but have not explored the simultaneous effect of changes in parenting and residence. As residential mobility and family structure instability often co-occur, it is important to tease apart the effects of each type of transition while accounting for the effects of the other to determine if one type of transition is more impactful at different developmental periods or on specific types of child problem behavior (i.e., externalizing versus internalizing). Finally, previous research has typically been retrospective, limiting the ability for researchers to infer the nature of relations between turbulence in the home and prospectively ascertained later problem behaviors.

We seek to fill several gaps in the growing literature on family turbulence by prospectively examining the influence of both parenting figure and residential transitions on emotional and behavior problems in early childhood and pre-adolescence using a diverse sample of high-risk youth. The present study has four aims. Aim 1: We test the effects of residential and family structure transitions in early-childhood (birth–5 years) on pre-kindergarten (age 5) internalizing and externalizing behaviors.

Aim 2: We examine associations between early- and middle-childhood (5–10 years) turbulence on pre-adolescent (age 10.5) internalizing and externalizing behaviors. In the middle-childhood model, we also test for residual effects of early-childhood residential mobility and family structure instability by controlling for age 5 internalizing behaviors.

Aim 3: We compare models for African American and Caucasian children to determine if the effects of turbulence are moderated by ethnicity. As the structure of African American families tends to be more fluid and transitions more normative, we hypothesize that the adverse effects of family structure transitions will be stronger predictors of emotional and behavioral problems for Caucasian children relative to African American children. To date, no studies that we could find have examined the moderating effect of race on the relationship between residential mobility and child behavior outcomes. Therefore, we seek to add to the existing literature by testing the moderating effect of race on this relationship.

Aim 4: To explore potential differences between a frequency and weighted model of family structure transitions, we test associations between the number of family structure transitions and maladaptive behaviors as well as associations between a weighted family structure transitions score and maladaptive behaviors.

Method

Participants

Participants were 731 families with children between the ages of 2 years 0 months and 2 years 11 months, recruited between 2002 and 2003 from Women, Infant, and Children Nutritional Supplement (WIC) centers in and around the metropolitan areas of Pittsburgh, PA (urban), Eugene, OR (suburban), and Charlottesville, VA (rural). These data were part of the Early Steps Multisite Project, a trial of the Family Check-Up intervention (Dishion et al., 2008). Families were screened for eligibility based on child behavior (e.g., early conduct problems or high-conflict relationships with adults), family problems (maternal depression or daily parenting challenges), and sociodemographic (low education achievement and low family income) risk factors. To be eligible, families had to score one standard deviation above the mean on two out of these three risk domains. Of the 1,666 families that were screened, 879 met the eligibility requirements (52% in Pittsburgh, 57% in Eugene, and 49% in Charlottesville), and 731 (83.2%) agreed to participate (88% in Pittsburgh, 84% in Eugene, and 76% in Charlottesville).

Of the 731 families recruited, 272 (37%) were recruited in Pittsburgh, 271 (37%) in Eugene, and 188 (26%) in Charlottesville. Target children (TC, 49% female) were on average 29.9 months (SD = 3.2) at the age 2 assessment. Across sites, children belonged to the following racial groups: 27.9% African American, 50.1% European American, 13.0% biracial, and 8.9% other races (e.g. American Indian). Thirteen percent of participants identified as Hispanic. During the 2002 – 2003 screening period, more than two thirds of those families enrolled in the project had an annual income of less than $20,000, and the average number of family members per household was 4.5 (SD 1.63). Forty-one percent of the sample had a high school diploma or general education diploma (GED), and an additional 32% had 1 – 2 years of post-high school training at the time of the screening. Following the baseline assessment at age 2, half of the families were randomly assigned to receive the Family Check-Up (see Dishion et al., 2008). Therefore, intervention status is controlled for in all analyses.

The current study restricted analyses to the 665 families (91.0% of the initial 731 families) whose parents reported the TC’s race as Caucasian or African American. At the age 5 assessment, data were collected from 612 PCs and 340 ACs and from 568 PCs and 398 ACs at the age 5 and 10.5 assessments, respectively. Selective attrition analyses revealed that removed families had a similar number of moves and similar family transitions scores at all ages as retained families. Families not included in the present study did have lower age 10 externalizing scores (t = 2.22, p < .05), lower levels of PC education at TC age 2 (t = 4.67, p < .01) and at TC age 5 (t = 3.32, p < .01) education, lower age 5 PC depressive symptoms (t = 2.40, p < .05), lower household chaos at ages 2 (t = 4.56, p < .01) and 5 (t = 3.69, p < .01), and lived in safer neighborhoods between 2 and 5 (t = 5.94, p < .01) and 7 and 10 (t = 6.52, p < .01).

Procedure

In-home assessments were conducted at child ages 2, 3, 4, 5, 7.5, 8.5, 9.5, and 10.5 with PCs, TC, and alternate caregivers (ACs), when available. Assessments were conducted between 2002 and 2014 and involved a variety of tasks including questionnaires, interviews, videotaped observations, and examiner impressions. All participants were financially compensated following each assessment. Custodial parent’s written consent was obtained prior to administration of any measures at each assessment. A Certificate of Confidentiality was obtained from the National Institute of Health to offer protection of participants’ confidentiality and encourage honest reporting. Institutional review board approval was obtained for all screening and assessment procedures.

Measures

Independent Variables

Residential Transitions

Changes in residence were assessed at each home assessment. Participants were asked to list all of the addresses the TC had lived at since the family’s last assessment (usually within the past year). Because we asked about addresses going back to the last assessment, we were able to fill in gaps if a family had missed a wave of data collection. In line with the goals of the present study to examine the relationships between changes in physical residence, we counted any change in residence as a move. Finally, moves were summed to create a total residential mobility score from early-childhood (birth – 5 years) and middle-childhood (5 – 10.5 years).

Parenting Figure Transitions

Parenting figure transitions were coded using responses from the demographic interview administered at each wave. First, we established whether or not the PC had a full-time live-in partner at each assessment wave using a question on the demographics interview. Next, we utilized two questions that asked about inter-wave partner transitions to assess relationship instability between assessments. We then integrated the changes in live-in partner status from wave to wave with the inter-wave partner instability to create a total partner instability variable. Finally, 62 children experienced a change in PC at some point between the age 2 and 10.5 assessments. A raw family structure transitions score was calculated for each family by summing the total number of partner and PC changes from early-childhood (birth – 5 years) and middle-childhood (5–10.5 years). We then created weighted family structure transitions scores for early- and middle-childhood by weighting each transition involving a non-biological parent as 1, biological (non-PC) parent or step-parent transitions as 2, and PC transitions as 3.

Dependent Variables

TC Internalizing and Externalizing Behaviors

TC internalizing and externalizing behaviors were assessed using the Child Behavior Checklist (CBCL), which is a parent report of emotional and behavior problems in children over the past 6 months (Achenbach & Rescorla, 2001). Responses are on a 3-point Likert scale ranging from Not True (0), Somewhat or Sometimes True (1), or Very True or Often True (2). PCs and, when possible, ACs completed the 115-item CBCL at TC age 5 and the 76-item CBCL at TC age 10.5. The CBCL yields an internalizing factor, which is comprised of the Anxious Depressed, Withdrawn Depressed, and Somatic Complaint subscales, and an externalizing factor, which is comprised of the Rule-Breaking, Aggressive Behavior, and Attention Problem subscales. For the internalizing factor, Cronbach internal consistencies ranged from .81 to .90 for the age 5 and 10.5 assessments for PC and AC reports. For the externalizing factor, Cronbach alphas ranged from .86 to .95 for the age 5 and 10.5 assessments for PC and AC reports, indicating good internal consistency for both factors across time and informant. PC and AC CBCL reports were modestly correlated on the internalizing (r’s = .295, and .430 at ages 5 and 10, respectively, p < .001) and externalizing (r’s = .418, and .579 at ages 5 and 10, respectively, p < .001) factors. To minimize reporter bias, PC and AC scores were averaged, with PC reports used alone when AC reports were not available.

Covariates

Demographic Variables

Demographic data including gross monthly income at ages 2 and 5, PC’s educational attainment at ages 2 and 5, intervention status, and site location were included as covariates in analyses. PC’s education was coded on a 9-point scale from no formal schooling through a graduate degree. Intervention status was dummy coded as 0 = control and 1 = intervention. Site location was accounted for by creating two dummy coded variables to represent site with Oregon families treated as the comparison group. TC’s race was collected at the age 2 assessment during the demographics interview and dummy coded as Caucasian = 0 African American or biracial = 1. TC’s gender (49% female) was also collected at the age 2 assessment.

PC depressive symptoms

PCs completed the Center for Epidemiologic Studies Depression Scale (CES-D), a widely used measure of depressive symptomology (Radloff, 1977), at the age 2 and 5 assessments. Participants indicated how frequently they had experienced a list of 20 depressive symptoms over the past week (e.g., I had trouble keeping my mind on what I was doing and I had crying spells). Response options ranged from Rarely or none of the time (0–1 days) to Most or all of the time (5–7 days). Cronbach alphas suggest good internal consistency (α’s = .88 and .92 at ages 2 and 5, respectively).

Chaos in the home

Household chaos was measured using 15 items from the Confusion, Hubbub, and Order Scale (CHAOS; Matheny, Wachs, Ludwig, & Phillips, 1995). The CHAOS assesses the level of noise, confusion, and routine disorganization in a household. Example items include “It’s a real zoo in your home” and “You can’t hear yourself think”. Participants responded True or False to each item. The measure was scored by summing the number of True items. PCs completed the CHAOS scale at ages 2 and 5. Cronbach alphas were .81 and .84 at ages 2 and 5, respectively, indicating adequate internal consistency.

Neighborhood Dangerousness

PCs completed the 25-item Me and My Neighborhood Questionnaire (MMNQ; Pitt Mother and Child Project, 2001) at each assessment. The MMNQ assesses a parent’s perceptions of their neighborhood affiliation, cohesion, violence, and disorder. The 15-item dangerousness subscale was used to assess perceptions of neighborhood dangerousness. Sample items on the dangerousness subscale include “A family member was robbed or mugged” and “I saw or heard about a shooting gallery near my home.” Response options were on a 4-point Likert Scale ranging from Never to Often. Cronbach Alphas range from .87 to .92 on the neighborhood dangerousness scale between ages 2 and 10.5, indicating good internal consistency. Early-childhood neighborhood dangerousness was assessed by taking the mean of PC reports from ages 2 to 5, while middle-childhood dangerousness was assessed by taking the mean of PC reports from ages 7.5 to 10.5.

Data Analysis

Descriptive statistics and intercorrelations were calculated between all variables using SPSS version 24 (IBM Corp., 2016). Mean differences on each variable for African American and Caucasian families were calculated using t-tests and differences by site location were calculated using a Tukey HSD test. To test Aim 1 that turbulence in early childhood predicted early childhood behavior problems, child externalizing and internalizing behaviors at age 5 were regressed onto family structure and residential transitions from ages 0–5. To test Aim 2, that there would be independent effects of early- and middle-childhood turbulence on pre-adolescent problem behaviors, internalizing and externalizing behaviors at age 10.5 were regressed onto early- and middle-childhood family structure and residential transitions. To test Aim 3, that the associations between turbulence and problem behaviors would be moderated by race, chi-square difference tests were conducted. Specifically, we freely estimated separate regression models for African American and Caucasian families while individually constraining the paths between number of moves and family structure transitions and the outcome variables. Finally, to test Aim 4, that there would be differences between a frequency and weighted model of family structure transitions, we ran separate regression models using the raw number of family structure transitions and a weighted family structure transitions score. Family income, PC education, TC gender, intervention status, site location, PC depressive symptoms, household chaos, and neighborhood danger were included in the models as covariates. All regression analyses and chi-square difference tests were conducted in Amos version 23 (Arbuckle, 2014). Missing data were treated using full information maximum likelihood estimation in Amos.

Results

Descriptive Statistics and Intercorrelations

Between birth and age 10.5, 89.9% of the children in the study experienced at least one move and 79.4% of children experienced at least one family structure transition. Turbulence was so common among families in the study that only 5.0% of children did not experience a single move or family structure transition, while 74.3% of children experienced both residential mobility and family structure instability. However, there was considerable variability in the amount of turbulence children experienced, providing additional support for our approach of looking at the frequency of transitions rather than the presence or absence of turbulence. In relation to moves, children moved an average of 2.25 times between birth and age 5 and an average of 1.60 times between ages 5 and 10.5. There was no difference by race with regards to the number of early-childhood moves, but African American families moved more frequently than Caucasian families during middle childhood (t596.67 = 4.01, p < .01). Regarding family structure transitions, the average child in the study experienced 1.84 transitions in early childhood and 1.92 transitions during middle childhood, while the average transitions score was 3.26 for early-childhood and 2.18 during middle childhood. In line with precious research, African American children experienced more family structure transitions than Caucasian children during both early childhood (t589.95 = 3.72, p < .01) and middle childhood (t572.85 = 3.50, p < .01). See Table 1 for all descriptive statistics and group differences.

Table 1.

Descriptive statistics by site and ethnicity

Construct Total
(n = 568)
Mean (SD) Range		 Virginia
(n = 135)
Mean (SD)		 Oregon
(n = 219)
Mean (SD)		 Pittsburgh
(n = 214)
Mean (SD)		 Caucasian
(n = 262)
Mean (SD)		 African American
(n = 176)
Mean (SD)
Family Income (2) 3.78 (1.92) 1–11 3.91a (1.88) 3.84a (1.81) 3.63a (2.04) 4.12y (1.99) 3.35x (1.83)
Family Income (5) 5.08 (2.56) 1–13 5.10a (2.45) 5.19a (2.53) 4.95 (2.65) 5.49y (2.54) 4.69x (2.56)
PC Education (2) 5.19 (1.14) 2–8 4.93a (1.11) 5.22b (1.23) 5.35b (1.02) 5.35x (1.11) 5.21x (1.01)
PC Education (5) 5.44 (1.19) 2–9 5.07a (1.16) 5.57b (1.22) 5.58b (1.11) 5.55x (1.11) 5.51x (1.10)
PC Depressive Symptoms (2) 16.75 (10.66) 0–55 14.69a (9.49) 16.65ab (10.78) 18.27b (11.09) 16.69x (10.89) 17.44x (10.43)
PC Depressive Symptoms (5) 14.81 (11.32) 0–53 14.60a (11.56) 14.02a (10.86) 15.72a (11.59) 15.40x (12.17) 14.77x (10.62)
Household Chaos (2) 5.29 (3.60) 0–15 4.93a (3.48) 5.94a (3.46) 5.34a (3.81) 5.99y (3.59) 4.87x (3.56)
Household Chaos (5) 4.99 (3.75) 0–15 4.92a (3.78) 4.70a (3.50) 5.33a (3.95) 5.41x (3.71) 4.82x (3.83)
Neighborhood Danger (2–5) 7.51 (6.24) 2.75–35.50 5.34a (5.00) 6.07a (4.59) 10.46b (7.24) 6.54x (5.05) 9.40y (7.07)
Neighborhood Danger (7.5–10.5) 6.36 (6.25) 2–40.75 4.03a (4.44) 5.11a (4.62) 9.12b (7.55) 5.59x (5.17) 8.00y (7.30)
Moves 0–5 2.28 (1.89) 0–13 2.17a (1.69) 2.73b (2.23) 1.89a (1.51) 2.38x (2.09) 2.16x (1.59)
Moves 5–10 1.80 (1.82) 0–13 1.99a (2.00) 1.73a (1.77) 1.74a (1.73) 1.59x (1.71) 2.01y (1.88)
Parent Transitions Raw 0–5 1.75 (2.32) 0–16 1.65a (2.27) 1.57a (2.09) 2.01a (2.54) 1.46x (2.09) 2.21y (2.54)
Parent Transitions Raw 5–10 1.87 (2.67) 0–17 1.75ab (2.49) 1.52a (2.37) 2.30b (3.01) 1.49x (2.28) 2.27y (2.93)
Parent Transitions Score 0–5 3.26 (3.72) 0–20 3.21a (3.41) 3.17a (3.89) 3.38a (3.75) 3.09x (3.97) 3.47x (3.37)
Parent Transitions Score 5–10 2.18 (3.19) 0–16 1.97a (2.64) 1.80b (2.86) 2.65a (3.70) 1.78x (2.20) 2.67y (3.44)
Externalizing 2 18.32 (6.12) 4.50–37.50 17.57 a (6.05) 18.50a (5.62) 18.44a (6.66) 18.65x (5.81) 17.83x (6.55)
Internalizing 2 11.29 (5.78) 0.50–35.00 10.97a (5.39) 10.71a (5.06) 12.03a (6.56) 10.75x (5.40) 12.11y (6.25)
Externalizing 5 10.64 (6.63) 0.00–33.50 9.96ab (6.52) 9.55a (6.02) 11.73b (6.97) 10.12x (6.40) 11.27y (6.87)
Internalizing 5 5.45 (3.98) 0.00–20.50 5.64a (4.99) 5.02a (3.77) 5.69a (3.69) 5.37x (4.04) 5.54x (3.92)
Externalizing 10.5 9.56 (8.12) 0.00–47.50 7.96a (7.97) 9.52a (7.71) 10.45a (8.52) 9.31x (7.70) 9.88x (8.63)
Internalizing 10.5 7.57 (6.08) 0.00–30.50 6.68a (5.80) 8.08a (6.28) 7.48a (6.02) 8.04x (6.35) 6.99x (5.69)
Open in a new tab

Note. Different superscripts indicate significant differences (p < .05) between groups. For Primary Caregiver (PC) Education: 2 = 7th grade or less, 3 = junior high, 4 = partial high school, 5 = high school/GED, 6 = partial college, 7 = junior college, 8 = standard college graduate, 9 = graduate degree; Income (annual): 1 = < $4,999, 2 = $5,000 – $9,999, 3 = $10,000 – $14,999, 4 = $15,000 – $19,999; 5 = $20,000 – $24,999, 6 = $25,000 – $29,999, 7 = $30,000 – $39,000, 8 = $40,000 – $49,999, 9 = $50,000 – $59,999, 10 = $60,000 – $69,999, 11 = $70,000 – $79,999, 12 = $80,000 – $89,999; 13 = > $90,000.

In terms of inter-correlations, the number of moves and transitions in family structure during early childhood were both positively correlated with the number of moves and family structure transitions during middle childhood, indicating that experiences of instability are related to continued instability in later childhood. Additionally, the number of moves and transitions in family structure were concurrently related, suggesting that moves and family structure changes occur together at a higher rate than chance. See Table 2 for all inter-correlations between the independent and dependent variables.

Table 2.

Intercorrelations between Independent and Outcome Variables

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
1. Moves 0–5 -
2. Moves 5–10.5 .332*** -
3. Family Transitions Raw 0–5 .198*** .215*** -
4. Family Transitions Raw 5–10 .078† .216*** .241*** -
5. Family Transitions Score 0–5 .178*** .197*** .749*** .203*** -
6. Family Transitions Score 5–10.5 .082* .282*** .221*** .844*** .188*** -
7. Internalizing (2) .051 .069† .046 .116** .118** .093* -
8. Externalizing (2) .090* .064† .014 .018 .113** .022 .555*** -
9. Internalizing (5) .125** .089* .102* .074† .124*** .050 .354*** .254*** -
10. Externalizing (5) .061 .062 .131** .079† .113*** .046 .270*** .471*** .600*** -
11. Internalizing (10.5) .106* .066 .044 −.010 .067 .013 .278*** .300*** .455*** .396*** -
12. Externalizing (10.5) .118** .045 .152** .104* .138** .077† .235*** .369*** .338*** .575*** .641*** -
Open in a new tab

Note.

†

denotes p < 0.1,

*

denotes p < 0.05,

**

denotes p < 0.01,

***

denotes p < .001.

Turbulence Predicting Age 5 Behaviors

Main effects of turbulence

There was a significant main effect of raw (β = .098, p < .05) and a trend level effect of weighted (β = .075, p < .10) family structure transitions on age 5 externalizing behaviors. Similarly, there was a significant main effect of weighted family structure transitions on age 5 internalizing behaviors (β = .084, p < .05). Finally, there was a significant main effect of early childhood residential mobility on internalizing behaviors in the model including raw family structure transitions (β = .120, p < .01) and the model including weighted family structure transitions (β = .116, p < .01).

Moderation by race

According to parent reports, after accounting for all family and sociodemographic covariates, the number of moves from birth to 5 significantly predicted higher levels of externalizing behaviors at 5, but only for Caucasian youth (X2 = 4.588, p = .032). The moderation effect of race on the association between early childhood moves and age 5 externalizing behaviors was significant in the model including the raw family structure transition score and the model including the weighted family structure transitions score (see Table 3).

Table 3.

Moves and Raw Family Structure Transitions Predicting Early-Childhood Behaviors

Standardized Beta Coefficients Constrained Path Analysis
Path Full Sample Caucasian African American X2 P-Value
Moves (0–5) → Externalizing (5) .034 .175** −.039 4.588 .032
Family Transitions Raw (0–5) → Externalizing (5) .098* .142* .054 .997 .318
Moves (0–5) → Internalizing (5) .120** .158** .062 .725 .394
Family Transitions Raw (0–5) → Internalizing (5) .058 .016 .091 .784 .376
Moves (0–5) → Externalizing (5) .035 .113* −.088 5.482 .019
Family Transitions Weighted (0–5) → Externalizing (5) .075† .069 .080 .147 .701
Moves (0–5) → Internalizing (5) .116** .147* .077 .304 .581
Family Transitions Weighted (0–5) → Internalizing (5) .084* .122* .028 .989 .320
Open in a new tab

Note. The Full Sample column corresponds to the main effect of the model, while the Caucasian and African American columns correspond to the moderation effects. For parsimony, covariates are not included in this table. However, analyses accounted for family income (2), primary caregiver (PC) education (2), intervention status, target child gender, site location, household chaos (2), PC depressive symptoms (2), and neighborhood danger (2–5).

†

denotes p < 0.1,

*

denotes p < 0.05,

**

denotes p < 0.01.

Turbulence Predicting Age 10.5 Problem Behaviors

Main effects of turbulence

There was a trend-level main effect of early childhood moves on age 10.5 externalizing behaviors, controlling for age 5 externalizing behaviors (β = .077, p < .10). This trend suggests that moves in early childhood predict late-childhood externalizing behaviors above and beyond the effects of early childhood moves on age 5 externalizing behaviors. There was also a trend-level main effect of weighted family structure transitions on age 10.5 externalizing behaviors (β = .083, p < .10).

Moderation by race

In models including raw family structure transitions, race was found to moderate the relationship between early-childhood moves and age 10.5 internalizing (X2 = 5.340, p = .021) and externalizing (X2 = 5.583, p = .018) behaviors. Additionally, there was a trend-level moderation effect of race on the relationship between the raw number of family structure transitions and levels of age 10.5 externalizing behaviors (X2 = 2.727, p = .099). Finally, race moderated the association between the number of early-childhood family structure transitions and age 10.5 externalizing behaviors, controlling for age 5 externalizing behaviors (X2 = 4.975, p = .026). In the aforementioned instances, turbulence predicted maladaptive behaviors for Caucasian, but not African American youth. See Table 4 for all standardized betas and results of the chi-square difference tests for models using raw family structure transitions.

Table 4.

Moves and Raw Family Structure Transitions Predicting Pre-Adolescent Behaviors

Standardized Beta Coefficients Constrained Path Analysis
Path Full Sample Caucasian African American X2 P-Value
Moves (0–5) → Externalizing (10) .083† .196** −.050 5.583 .018
Family Transitions Raw (0–5) → Externalizing (10) .086† .159** .010 2.727 .099
Moves (0–5) → Δ Externalizing (5–10) .077† .127* .022 1.008 .315
Family Transitions Raw (0–5) → Δ Externalizing (5–10) .026 .109* −.068 4.975 .026
Moves (5–10) → Externalizing (10) −.041 −.074 .010 .851 .356
Family Transitions Raw (5–10) → Externalizing (10) .041 .043 .013 .175 .676
Moves (5–10) → Δ Externalizing (5–10) −.044 −.085 .001 1.184 .277
Family Transitions Raw (5–10) → Δ Externalizing (5–10) .047 .085 .027 .860 .354
Moves (0–5) → Internalizing (10) .062 .114† −.023 5.340 .021
Family Transitions Raw (0–5) → Internalizing (10) .020 .088 −.123† 1.972 .160
Moves (0–5) → Δ Internalizing (5–10) .013 .059 −.044 1.410 .235
Family Transitions Raw (0–5) → Δ Internalizing (5–10) −.007 .073 −.153** 7.006 .008
Moves (5–10) → Internalizing (10) .050 −.086 .072 .003 .953
Family Transitions Raw (5–10) → Internalizing (10) −.044 .049 −.022 .884 .347
Moves (5–10) → Δ Internalizing (5–10) .039 .019 .083 .326 .568
Family Transitions Raw (5–10) → Δ Internalizing (5–10) −.047 −.061 −.034 .330 .565
Open in a new tab

Note. The Full Sample column corresponds to the main effect of the model, while the Caucasian and African American columns correspond to the moderation effects. Early-childhood mobility and family structure transitions were included in the same model as middle-childhood transitions to allow for the detection of age effects of turbulence on age 10 behaviors. For parsimony, covariates are not included in this table. However, analyses accounted for family income (5), primary caregiver (PC) education (5), intervention status, target child gender, site location, household chaos (5), PC depressive symptoms (5), and neighborhood danger (7–10). Additionally, we controlled for internalizing and externalizing behaviors at age 5 to detect changes in internalizing and externalizing behaviors between ages 5 and 10, respectively.

†

denotes p < 0.1,

*

denotes p < 0.05,

**

denotes p < 0.01.

Differences between raw and weighted family structure transitions

Notably, there was no longer a moderating effect of race on the path between weighted family structure transitions in early childhood and pre-adolescent internalizing and externalizing behaviors. However, there was still a trend-level main effect of early-childhood weighted family structure transitions on age 10.5 externalizing behaviors (β = .083, p < .10) and the path remained significant for Caucasian youth (β = .125, p < .05), but not African American youth (β = .009, p > .05). See Table 5 for all standardized betas and results of the chi-square difference tests for models using weighted family structure transitions.

Table 5.

Moves and Weighted Family Structure Transitions Predicting Pre-Adolescent Behaviors

Standardized Beta Coefficients Constrained Path Analysis
Path Full Sample Caucasian African American X2 P-Value
Moves (0–5) → Externalizing (10) .083† .194** −.048 5.444 .020
Family Transitions Weighted (0–5) → Externalizing (10) .083† .124* .009 1.296 .255
Moves (0–5) → Δ Externalizing (5–10) .078† .129* .017 1.205 .272
Family Transitions Weighted (0–5) → Δ Externalizing (5–10) .024 .072 −.033 1.715 .190
Moves (0–5) → Internalizing (10) .057 .114† −.032 2.325 .127
Family Transitions Weighted (0–5) → Internalizing (10) .036 .053 −.080 2.235 .135
Moves (0–5) → Δ Internalizing (5–10) .010 .062 −.057 1.825 .177
Family Transitions Weighted (0–5) → Δ Internalizing (5–10) −.004 .014 −.080 1.393 .238
Moves (5–10) → Externalizing (10) −.039 −.059 .013 .640 .424
Family Transitions Weighted (5–10) → Externalizing (10) .028 .048 −.009 .477 .490
Moves (5–10) → Δ Externalizing (5–10) −.044 −.061 −.001 .600 .439
Family Transitions Weighted (5–10) → Δ Externalizing (5–10) .046 .052 .034 .091 .762
Moves (5–10) → Internalizing (10) .041 .048 .068 .002 .969
Family Transitions Weighted (5–10) → Internalizing (10) −.012 −.041 .019 .511 .475
Moves (5–10) → Δ Internalizing (5–10) .033 .026 .077 .176 .675
Family Transitions Score (5–10) → Δ Internalizing (5–10) −.016 −.031 .006 .253 .615
Open in a new tab

Note. The Full Sample column corresponds to the main effect of the model, while the Caucasian and African American columns correspond to the moderation effects. Early-childhood mobility and family structure transitions were included in the same model as middle-childhood transitions to allow for the detection of age effects of turbulence on age 10 behaviors. For parsimony, covariates are not included in this table. However, analyses accounted for family income (5), primary caregiver (PC) education (5), intervention status, target child gender, site location, household chaos (5), PC depressive symptoms (5), and neighborhood danger (7–10). Additionally, we controlled for internalizing and externalizing behaviors at age 5 to detect changes in internalizing and externalizing behaviors between ages 5 and 10, respectively.

†

denotes p < 0.1,

*

denotes p < 0.05,

**

denotes p < 0.01.

Discussion

The present study adds to existing literature on childhood turbulence by examining the effects of residential mobility and transitions in family structure on child behaviors from early through middle childhood. Specifically, we examined differences between Caucasian and African American children on the impact of turbulence in early and middle childhood on parent-reported child internalizing and externalizing behaviors at ages 5 and 10.5. Exploring differences in behavior outcomes of turbulence by race is a salient issue given differences in both prevalence and outcomes of turbulence observed between Caucasian and African American youth in the present study and in previous research (Fomby et al., 2007; Ihrke et al., 2011; Shaw et al., 1999). Additionally, we examined the effects of residential mobility and family structure instability in early and middle childhood to determine if there are different effects of turbulence at different developmental periods.

Our findings indicate that after accounting for several factors that may influence our variables of interest, higher levels of residential and family structure transitions predicted higher levels of caregiver-reported internalizing and externalizing behaviors. Family structure transitions were associated with higher levels of externalizing behaviors, which is consistent with previous research finding family structure instability to be associated with disruptive behaviors in children (D’Onofrio et al., 2005; Goodnight et al., 2013; Martinez and Forgatch, 2002). Likewise, residential mobility in early childhood was associated with higher levels of externalizing behaviors, consistent with extant literature (Fowler et al., 2014; Tiesler et al., 2013).

However, the associations between residential and family structure instability and subsequent maladaptive behaviors were only significant for Caucasian youth, indicating that race may moderate the association between early-childhood turbulence and later behaviors. With regards to family structure transitions, the results of the present study are consistent with others documenting the effects of the number of transitions on behavior problems for Caucasian but not African American children (Fomby & Cherlin, 2007; Shaw et al., 1999). Specifically, although African American youth in the present study experienced more family structure transitions throughout childhood compared to Caucasian youth, family structure transitions were a significant predictor of subsequent internalizing and externalizing behaviors for Caucasian youth only. That family instability predicted elevated emotional and behavioral problems in Caucasian but not African American children extends findings by Shaw and colleagues (1999), who found parenting figure transitions in early childhood to predict early-childhood externalizing problems, but only in Caucasian children.

In the present study, the moderating effect of race on the relationship between early-childhood family structure transitions and pre-adolescent behavior problems was no longer present when a weighted family structure transition score was used. These findings suggest that, to some extent, accounting for the relationship of the transitioning partner to other members of the household may explain the differences in behavior outcomes previous research has documented in Caucasian versus African American youth (e.g., Fomby & Cherlin, 2007). Research suggests that African American youth are more likely than Caucasian youth to experience multiple family structure transitions (Fomby & Cherlin, 2007; Shaw et al., 1999), but those transitions are also more likely to involve non-biological, non-married cohabiting parenting figures (Kelly Raley & Wildsmith, 2004) who may be less involved in day-to-day parenting. Therefore, African American children may be experiencing family structure transitions that do not have lasting effects on a child’s psychological functioning, while the traditional structure of Caucasian families makes it more likely that Caucasian children are experiencing the loss of a biological or step-parent.

A frequency approach may inflate associations between instability and behavior problems for Caucasian youth treating transitions in biological or step-parents in Caucasian families equivalently to non-biological, non-marital cohabiting partners in African American families. The weighted approach accounts for some of those differences, at least partially leveling the playing field. It should be noted that, in the present study, the early-childhood weighted transitions score still only predicted elevated internalizing and externalizing behaviors for Caucasian and not African American children.

Additionally, the present study is the first that we are aware of to test the moderating effect of race on residential mobility outcomes. The number of moves in early childhood predicted higher levels of externalizing behaviors at age 5 and 10.5 and higher levels of internalizing behaviors at age 10.5 for Caucasian but not African American children, suggesting that Caucasian youth may also be more vulnerable to the negative effects of early residential mobility.

Taken together, these results indicate that turbulence in early childhood has a greater impact on subsequent behavior problems for Caucasian children than for African American children. One possibility for the less substantial impact of turbulence on African American children is that extra-familial members (e.g., aunts, uncles, grandparents) in African American families play a more prominent role in parenting children relative to Caucasian families, and thus may buffer the effects of turbulence in early childhood (Sarkisian & Gerstel, 2004; Wilson, 1989). Therefore, because transitions in parenting figures may cause a lesser degree of distress and disruption in African American compared to Caucasian families, African American children may show lower levels of problem behavior in response to such transitions. In African American families, the availability of extra-familial parenting figures may minimize the loss of a caregiver following a relationship dissolution. Additionally, extra-familial parenting figures living outside the home may offer a residence of stability for children who are growing up in chronically moving families.

Extending beyond the home, the African American community as a whole tends to be tightly knit (Harrison et al., 1990; Wilson & Tolson, 1990). In highly turbulent neighborhoods, African American families may be more likely to assist one another with connecting to the community and local resources, minimizing some of the stressors associated with moving.

The Developmental Timing of Turbulence

With regards to the developmental timing of turbulence, in the present study only residential mobility and family structure instability in early childhood predicted elevated internalizing and externalizing behaviors at ages 5 and 10.5. Additionally, there appear to be residual effects of residential mobility and family structure instability in the first 5 years of life. Specifically, in the present study both residential mobility and family structure instability in early childhood predicted higher levels of pre-adolescent externalizing behavior, accounting for age 5 externalizing behaviors and number of transitions between ages 5 and 10.5. These findings, in conjunction with previous research (Goodnight et al., 2013; Rumbold et al., 2012) indicate that early childhood is a period of sensitivity to environmental change and that exposure to environmental instability may have long-term implications for a child’s emotional and behavioral development.

From a cognitive developmental perspective, the stressors of early residential mobility may disrupt cognitive development and, in particular, executive functioning (Schmitt, Finders, & McClelland, 2015). Poor executive functioning may limit a young child’s ability to successfully adapt to their changing environment (Miyake et al., 2000), leading to maladaptive coping strategies, including internalizing or externalizing behaviors. Research suggests that exposure to high levels of stress in early childhood is related to lower levels of inhibitory control at school entry (Brown, Ackerman, & Moore, 2013), and that poorer inhibitory control is associated with maladaptive socio-emotional development (Rhoades, Greenberg, & Domitrovich, 2009). Therefore, children who experience high levels of turbulence in early childhood enter the transition to school with poorer emotional or behavior regulation, which may precipitate a cascade of social and academic difficulties conducive to maintaining or escalating disruptive behaviors (Moilanen, Shaw, & Maxwell, 2010).

Limitations and Directions for Future Research

While the present study offers many contributions to the existing body of literature, it is not without limitations. First, when recruited the Early Steps Multisite sample was predominantly low SES, which limits the generalizability of these findings to more affluent samples. Families in this study experienced much higher levels of residential instability and family structure changes than typical American families (Ihrke et al., 2011; Kennedy & Bumpass, 2008). Additionally, the results from the present study may not necessarily generalize to other types of families that experience high levels of instability, such as military families, because of the unique economic and social stressors associated with living in poverty.

A second limitation of the present study is that our measures of child behavior were generated from caregivers’ reports. Because caregivers are often directly involved in the transitions a family goes through, they themselves may be affected by housing or relationship instability. This could influence how they perceive their child’s behavior or may influence other factors such as their parenting strategies. Having the AC report in addition to the PC report is helpful in that it provides a perspective of another parenting figure, who may not be living in the home. However, AC reports were only available for approximately half of the sample, limiting our ability to have two caregiver reports for the majority of children.

Third, because information on residential and relationship instability was collected retrospectively from the preceding year, it is possible that caregivers misremembered moves or relationship changes. However, the prospective nature of the Early Steps study offers a significant improvement over previous studies that have collected turbulence data based on recall extending as long as 17 years (e.g., Simpson & Glenn-Fowler, 1994).

Finally, our measure of residential mobility could be improved upon by ascertaining the reasons behind a move. It seems likely that moves occurring under stressful circumstances (e.g., eviction, job loss, divorce) would be more detrimental than moves occurring under positive circumstances (e.g., job promotion, moving to a bigger house or better neighborhood). Thus, future prospective studies should consider adding survey questions to determine why low-SES families are moving. Open-ended questions would provide excellent qualitative data that could inform research on the nature of housing mobility for poor Americans. It also would be important to replicate our findings that a weighted family structure instability score may account for the discrepancy in behavior outcomes between African American and Caucasian children suggested by a frequency model. Additionally, future research should consider potential mediating or moderating variables (e.g., child executive functioning, involvement of extra-familial parenting figures) that could explain the relationship between turbulence and behavior problems. For example, do larger parenting or community networks buffer against the negative effects of housing mobility or instability in family structure? Answering such questions could potentially explain the observed differences between Caucasian and African American children.

Conclusions

Results from the present study indicate that housing mobility and instability in family structure during early and middle childhood can negatively impact a child’s emotional and behavioral development. These findings held after accounting for several variables that may influence turbulence and both child and parental behavior (e.g., maternal depression, neighborhood dangerousness, family income). Exposure to turbulence in early childhood not only impacts behaviors in early childhood but can have lasting implications for mental health and behavior into late childhood. Finally, Caucasian children exposed to high levels of turbulence in early childhood appear to be at greater risk for later behavior problems than African American children. Based on the findings from the present study and previous research (e.g. Capaldi & Patterson, 1991; Jellyman & Spencer, 2008) that residential mobility and family structure instability predict maladaptive behaviors in children, community organizations catering to low-SES families with young children (e.g. Head Start, WIC, shelters for homeless families and domestic violence) should be aware that the families they work with are at a high risk for turbulence. Such organizations could take more aggressive steps to connect families who have moved to mental health resources to mitigate the negative effects of residential mobility on child adjustment. Additionally, while the present study did not find associations between turbulence in middle childhood and emotional or behavioral problems, other research has (Goodnight et al., 2013; Simpson & Fowler, 1994). Schools may want to take extra steps to provide new students moving into or within their district counseling resources to make sure the child acclimates to their new environment.

Acknowledgments

Support for this research was provided by the National Institute on Drug Abuse to the third, fourth and fifth authors (R01 DA023245, R01 DA022773). We also wish to extend our appreciation to the staff and research participants of the Early Steps Multisite Study.

References

  1. Achenbach TM, Rescorla L. ASEBA school-age forms & profiles. 2001. [Google Scholar]
  2. Adam EK. Beyond quality: Parental and residential stability and children’s adjustment. Current Directions in Psychological Science. 2004;13(5):210–213. [Google Scholar]
  3. Amato PR. The consequences of divorce for adults and children. Journal of Marriage and the Family. 2000;62:1269–1287. [Google Scholar]
  4. Arbuckle JL. Amos (Version 23.0) [Computer Program] Chicago: IBM SPSS; 2014. [Google Scholar]
  5. Belsky J. The determinants of parenting: A process model. Child Development. 1984;55:83–96. doi: 10.1111/j.1467-8624.1984.tb00275.x. [DOI] [PubMed] [Google Scholar]
  6. Belsky J, Schlomer GL, Ellis BJ. Beyond cumulative risk: distinguishing harshness and unpredictability as determinants of parenting and early life history strategy. Developmental psychology. 2012;48(3):662. doi: 10.1037/a0024454. [DOI] [PubMed] [Google Scholar]
  7. Bowlby John. A secure base: Parent-child attachment and healthy human development. Basic books; 2008. [Google Scholar]
  8. Bronfenbrenner U. The Ecology of Human Development: Experiments by Nature and Design. Cambridge, MA: Harvard University Press; 1979. [Google Scholar]
  9. Brown ED, Ackerman BP, Moore CA. Family adversity and inhibitory control for economically disadvantaged children: Preschool relations and associations with school readiness. Journal of Family Psychology. 2013;27:443–452. doi: 10.1037/a0032886. doi:10.1037=a0032886. [DOI] [PubMed] [Google Scholar]
  10. Capaldi DM, Patterson GR. Relation of parental transitions to boys’ adjustment problems: I. A linear hypothesis. II. Mothers at risk for transitions and unskilled parenting. Developmental Psychology. 1991;27(3):489–504. [Google Scholar]
  11. Cavanagh SE, Huston AC. Family instability and children’s early problem behavior. Social Forces. 2006;85:551–582. [Google Scholar]
  12. Cooper CE, McLanahan SS, Meadows SO, Brooks-Gunn J. Family structure transitions and maternal parenting stress. Journal of Marriage and Family. 2009;71(3):558–574. doi: 10.1111/j.1741-3737.2009.00619.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Davies PT, Cummings ME. Marital conflict and child adjustment: An emotional security hypothesis. Psychological Bulletin. 1994;116:387–411. doi: 10.1037/0033-2909.116.3.387. [DOI] [PubMed] [Google Scholar]
  14. Dewitt DJ. Frequent childhood geographic relocation: Its impact on drug use initiation and the development of alcohol and other drug-related problems among adolescents and young adults. Addictive Behaviors. 1998;23:623–634. doi: 10.1016/s0306-4603(98)00023-9. [DOI] [PubMed] [Google Scholar]
  15. Dishion TJ, Shaw DS, Connell A, Wilson MN, Gardner F, Weaver C. The Family Check Up with high-risk families with toddlers: Outcomes on positive parenting and early problem behavior. Child Development. 2008;79:1395–1414. doi: 10.1111/j.1467-8624.2008.01195.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. D’Onofrio BM, Turkheimer E, Emery RE, Slutske WS, Heath AC, Madden PA, Martin NG. A genetically informed study of marital instability and its association with offspring psychopathology. Journal of Abnormal Psychology. 2005;114:570–586. doi: 10.1037/0021-843X.114.4.570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Fitchen JM. Residential mobility among the rural poor. Rural Sociology. 1994;59(3):416–436. [Google Scholar]
  18. Fowler PJ, Henry DB, Schoeny M, Taylor J, Chavira D. Developmental timing of housing mobility: Longitudinal effects on externalizing behaviors among at-risk youth. Journal of the American Academy of Child & Adolescent Psychiatry. 2014;53(2):199–208. doi: 10.1016/j.jaac.2013.12.003. [DOI] [PubMed] [Google Scholar]
  19. Frojd S, Marttunen M, Kaltiala-Heino R. The effect of adolescent- and parent-induced family transitions in middle adolescence. 2012 doi: 10.3109/08039488.2011.628694. [DOI] [PubMed] [Google Scholar]
  20. Fomby P, Cherlin AJ. Family instability and child well-being. American Sociological Review. 2007;72:181–204. doi: 10.1177/000312240707200203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Goodnight JA, D’Onofrio BM, Cherlin AJ, Emery RE, Van Hulle CA, Lahey BB. Effects of multiple maternal relationship transitions on offspring antisocial behavior in childhood and adolescence: A cousin-comparison analysis. Journal of Abnormal Child Psychology. 2012;41:185–198. doi: 10.1007/s10802-012-9667-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Harrison AO, Wilson MN, Pine CJ, Chan SQ, Buriel R. Family ecologies of ethnic minority children. Child development. 1990;61(2):347–362. [Google Scholar]
  23. Haynie DL, South SJ, Bose S. Residential mobility and attempted suicide among adolescents: An individual-level analysis. The Sociological Quarterly. 2006a;47(4):693–721. [Google Scholar]
  24. Haynie DL, South SJ, Bose S. The company you keep: Adolescent mobility and peer behavior. Sociological Inquiry. 2006b;76:397–426. [Google Scholar]
  25. IBM Corp. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp; 2016. [Google Scholar]
  26. Ihrke DK, Faber CS, Koerber WK. Current Population Reports, P20–565. U.S. Census Bureau; Washington, DC: 2011. Geographical Mobility: 2008 to 2009. [Google Scholar]
  27. Ingoldsby EM, Shaw DS, Owens EB, Winslow EB. A longitudinal study of interparental conflict, emotional and behavioral reactivity, and preschoolers’ adjustment problems among low-income families. Journal of Abnormal Child Psychology. 1999;27:343–356. doi: 10.1023/a:1021971700656. [DOI] [PubMed] [Google Scholar]
  28. Jackson Y, Warren JS. Appraisal, social support, and life events: Predicting outcome behavior in school-aged children. Child Development. 2000;71:1441–1457. doi: 10.1111/1467-8624.00238. [DOI] [PubMed] [Google Scholar]
  29. Jellyman T, Spencer N. Residential mobility in childhood and health outcomes: A systematic review. Journal of Epidemiology & Community Health. 2008;62:584–592. doi: 10.1136/jech.2007.060103. [DOI] [PubMed] [Google Scholar]
  30. Kelly Raley R, Wildsmith E. Cohabitation and children’s family instability. Journal of Marriage and Family. 2004;66(1):210–219. [Google Scholar]
  31. Kennedy S, Bumpass L. Cohabitation and children’s living arrangements: New estimates from the United States. Demographic Research. 2008;19:1663–1692. doi: 10.4054/demres.2008.19.47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Manning WD, Brown S. Children’s economic well-being in married and cohabiting parent families. Journal of Marriage and Family. 2006;68(2):345–362. [Google Scholar]
  33. Martinez CR, Forgatch MS. Adjusting to change: Linking family structure transitions with parenting and boys’ adjustment. Journal of Family Psychology. 2002;16:107–117. doi: 10.1037//0893-3200.16.2.107. [DOI] [PubMed] [Google Scholar]
  34. Matheny AP, Wachs TD, Ludwig JL, Phillips K. Bringing order out of chaos: Psychometric characteristics of the confusion, hubbub, and order scale. Journal of Applied Developmental Psychology. 1995;16(3):429–444. [Google Scholar]
  35. McMunn AM, Nazroo JY, Marmot MG, Borehan R, Goodman R. Children’s emotional and behavioural well-being and the family environment: Findings from the Health Survey for England. Social Science & Medicine. 2001;53:423–440. doi: 10.1016/s0277-9536(00)00346-4. [DOI] [PubMed] [Google Scholar]
  36. Miyake A, Friedman NP, Emerson MJ, Witzki AH, Howerter A, Wager TD. The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: A latent variable analysis. Cognitive Psychology. 2000;41(1):49–100. doi: 10.1006/cogp.1999.0734. [DOI] [PubMed] [Google Scholar]
  37. Moilanen KL, Shaw DS, Maxwell KL. Developmental cascades: Externalizing, internalizing, and academic competence from middle childhood to early adolescence. Development and Psychopathology. 2010;22:635–653. doi: 10.1017/S0954579410000337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Parente ME, Mahoney JL. Residential mobility and exposure to neighborhood crime: Risks for young children’s aggression. Journal of Community Psychology. 2009;37:559–578. [Google Scholar]
  39. Pitt Mother & Child Project. Me and My Neighborhood. 2001. Unpublished measure. [Google Scholar]
  40. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1(3):385–401. [Google Scholar]
  41. Ram B, Hou F. Changes in family structure and child outcomes: Roles of economic and familial resources. The Policy Studies Journal. 2003;31:309–330. [Google Scholar]
  42. Rhoades BL, Greenberg MT, Domitrovich CE. The contribution of inhibitory control to preschoolers’ social-emotional competence. Journal of Applied Developmental Psychology. 2009;30:310–320. [Google Scholar]
  43. Rodriguez CM. Association between independent reports of maternal parenting stress and children’s internalizing symptomatology. Journal of Child and Family Studies. 2011;20(5):631–639. [Google Scholar]
  44. Rumbold AR, Giles LC, Whitrow MJ, Steele EJ, Davies CE, Davies MJ, Moore VM. The effects of house moves during early-childhood on child mental health at age 9 years. BMC Public Health. 2012;12(1):583. doi: 10.1186/1471-2458-12-583. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sarkisian N, Gerstel N. Kin support among Blacks and Whites: Race and family organization. American Sociological Review. 2004;69(6):812–837. [Google Scholar]
  46. Scaramella LV, Leve LD. Clarifying parent–child reciprocities during early childhood: The early-childhood coercion model. Clinical Child and Family Psychology Review. 2004;7(2):89–107. doi: 10.1023/b:ccfp.0000030287.13160.a3. [DOI] [PubMed] [Google Scholar]
  47. Schmitt SA, Finders JK, McClelland MM. Residential mobility, inhibitory control, and academic achievement in preschool. Early Education and Development. 2015;26:189–208. [Google Scholar]
  48. Shaw DS, Bell RQ. Developmental theories of parental contributors to antisocial behavior. Journal of Abnormal Child Psychology. 1993;21:493–518. doi: 10.1007/BF00916316. [DOI] [PubMed] [Google Scholar]
  49. Shaw DS, Emery RE, Tuer MD. Parental functioning and children’s adjustment in families of divorce: A prospective study. Journal of Abnormal Child Psychology. 1993;21:119–134. doi: 10.1007/BF00910493. [DOI] [PubMed] [Google Scholar]
  50. Shaw DS, Winslow EB, Flanagan C. A prospective study of the effects of marital status and family relations on young children’s adjustment among African American and European American families. Child Development. 1999;70:742–755. doi: 10.1111/1467-8624.00053. [DOI] [PubMed] [Google Scholar]
  51. Shaw DS, Winslow EB, Owens EB, Vondra JI, Cohn JF, Bell RQ. The development of early externalizing problems among children from low-income families: A transformational perspective. Journal of Abnormal Child Psychology. 1998;26:95–107. doi: 10.1023/a:1022665704584. [DOI] [PubMed] [Google Scholar]
  52. Simpson GA, Glenn-Fowler M. Geographic mobility and children’s emotional/behavioral adjustment and school functioning. Pediatrics. 1994;93:303–309. [PubMed] [Google Scholar]
  53. South SJ, Haynie DL. Friendship networks of mobile adolescents. Social Forces. 2004;83(1):315–350. [Google Scholar]
  54. Tiesler CM, Birk M, Kohlbock G, Koletzko S, Bauer C, Wichmann HE, et al. Residential mobility and behavioural problems in children: Results from GINIplus and LISAplus studies. Journal of Public Health. 2013;21:39–48. [Google Scholar]
  55. Tolson TF, Wilson MN. The Impact of Two- and Three-Generational Black Family Structure on Perceived Family Climate. Child Development. 1990;61(2):416–428. [Google Scholar]
  56. Wilson MN. Child development in the context of the Black extended family. American Psychologist. 1989;44(2):380. [Google Scholar]
  57. Wilson MN, Tolson TFJ. Familial support in the Black community. Journal of Clinical Child Psychology. 1990;19(4):347–355. [Google Scholar]
  58. Wilson MN, Tolson TFJ, Hinton I, Kiernan M. Flexibility and sharing of childcare duties in Black families. Sex Roles. 1990;22:409–425. [Google Scholar]
  59. Wilson MN, Tolson TFJ, Kiernan M. In: Harrison A, editor. Comparative analysis of nontraditional Black families on self-reports of childcare, socialization, and childrearing activities; 11th Empirical Conference on Black Psychology; Washington, D.C: NIMH; 1989. pp. 108–119. [Google Scholar]

RESOURCES