And Still WE Rise: Parent–Child Relationships, Resilience, and School Readiness in Low-Income Urban Black Families

Abstract

The Family Stress Model acknowledges forms of resilience in the face of hardship; however, few studies have emerged on the potentially positive role of familial relationships in the academic, psychological, and prosocial success of impoverished Black children. The current study evaluates how parent–child relationship conflict and financial stress are associated with children’s school readiness (i.e., academic, psychosocial, and socioemotional indicators). Latent profile analyses, incorporating financial stress, general stress, and parent–child relationship variables were used to test whether varying family stress profiles differentially predicted children’s school readiness in Black families with children entering kindergarten (N = 292). Findings revealed 4 latent classifications with profiles of low, moderate, moderate/high, and high/moderate stress and conflict variables, respectively. Whereas the low-profile was associated with the most desirable school readiness indicators overall, children in the high/moderate-profile were rated as significantly more psychosocially and socioemotionally prepared for school than their moderate/high-profile counterparts. Families with less conflictual parent–child relationships had more optimal school readiness relative to families with higher conflict and less financial strain. The findings of the current study have the potential to contribute to theories of poverty and parent–child relationships, as well as guide therapeutic services focused on family relationships through school- and community-related programs for impoverished urban Black youth and their families.

Substantial evidence suggests that poverty, both acute and chronic, leads to less desirable outcomes for all who experience it; particularly children (see Raver, Roy, & Pressler, 2015). Black children are more likely to encounter these effects because of the higher rates of poverty Black families endure (Pew Research Center, 2015). Despite harrowing statistics, there are many Black children who grow up in poverty and “beat the odds,” demonstrating strengths and positive outcomes not predicted by the theoretical literature on poverty (see Brown, Barbarin, & Scott, 2013). Developmental and family researchers can learn a great deal from Black children and their families who show resilience (Boyd-Franklin & Karger, 2012), or positive life outcomes in the face of such adversity (Luthar & Cicchetti, 2000).

The Family Stress Model

Impact of Poverty on Families

Poverty is one of the most pervasive predictors of psychological maladjustment for parents and their children (Bierman et al., 2010; Leventhal & Brooks-Gunn, 2011). Moreover, almost half (42%) of all Black children under the age of 6 are impoverished relative to 37% of Hispanic and 14% of White children (Economic Policy Institute, 2014). As advanced by McLoyd (1990) in the Family Stress Model (FSM), the stress of poverty precipitates parental psychological distress, poor relationships with parents, and subsequent child problems. Although these relationships have been robustly associated in several studies (Anderson, 2015; Nievar & Luster, 2006; Pinderhughes, Nix, Foster, & Jones, 2007), it is important to understand how challenges from poverty impact Black family life and school-readiness characteristics, in particular (Duncan & Magnuson, 2002).

Psychological distress.

Adults who are impoverished often have a single income, low-wage jobs, and/or several family members depending on the income or aid (Reyes, 2008). The stress from poverty has been directly associated with parental depression in diverse populations (O’Neil, Wilson, Shaw, & Dishion, 2009) and within economically segregated neighborhoods (Ludwig et al., 2012). Moreover, the confluence of historical factors relevant to Black caregivers (e.g., poverty, inequality, and discrimination) poses particular challenges for Black mothers who often draw less upon their social support networks and rely on fewer resources than others (Belle & Doucet, 2003). Low-income mothers have some of the highest rates of depression, which not only impacts their personal well-being, but also contributes to being less responsive and more inconsistent and hostile in their parenting practices (Tomlinson, 2010). Furthermore, the psychological stress that is faced by low-income parents may present a challenge for engaging in school involvement and academic socialization due to fatigue (Conger, Ge, Elder, Lorenz, & Simons, 1994). Depressed Black parents model fewer social interactions for their children as a result of this financial hardship (Leiferman, Ollendick, Kunkel, & Christie, 2005).

Parent–child relationships.

Poverty limits families’ access to material goods and also complicates the relationships between family members (McLoyd, 1990; Pinderhughes et al., 2007). For low-income mothers, depression has been found to be significantly and robustly associated with negative parenting behaviors (Lovejoy, Graczyk, O’Hare, & Neuman, 2000). Black parents who have elevated depression symptoms report greater conflictual relationships with their child relative to their nondepressed peers (Lee, Lee, & August, 2011). Conflictual parent–child relationships also account for the association between parental distress and child outcomes, emphasizing the effects of parental stressors on subsequent child outcomes (Anderson et al., 2015). The quality of parent–child relationship has been associated with early cognitive functioning, including literacy development (Bergin, 2001). The relationship between parent and child also impacts the child’s prosociality as a function of the limited opportunities for relationship skill-building and self-management (Lindsey, Colwel, Frabutt, Chambers, & MacKinnon, 2008). Studies have found that poverty is related to mother–child relationship quality, parenting practices, and maternal involvement, and these proximal variables are indirectly related to children’s internalizing problems (e.g., Brody & Flor, 1998). Similarly, Myers and Taylor (1998) found that maternal psychological distress, high family stress burden, and coercive parenting practices contributed to urban Black children’s externalizing behaviors.

Child problems.

Through both direct (Evans, 2004) and indirect (Anderson, 2015) associations, financial stress is associated with school-related characteristics and trajectories in youth (see Bierman et al., 2010; Evans & English, 2002; Joe & Davis, 2009). Poverty hampers school readiness, or “the multidimensional concept that considers behavioral and cognitive aspects of the child’s development as well as the child’s adaptation to the classroom” (Parker, Boak, Griffin, Ripple, & Peay, 1999, p. 413). In addition to completing fewer years of schooling, impoverished youth have fewer academic skills upon school entry (Ryan, Fauth, & Brooks-Gunn, 2006) critical for later achievement (Duncan et al., 2007). Moreover, even within low-income samples, racial disparities have been found in academic exposure, including math-preparedness (Hill, 2011), preschool vocabulary (Duncan, Ludwig, & Magnuson, 2007), and literacy and oral language (Britto, Brooks-Gunn, & Griffin, 2006).

With regard to psychosocial functioning in Black families (e.g., internalizing and externalizing problems; Oravecz, Koblinsky, & Randolph, 2008), Dallaire and colleagues (2008) found that poverty predicted young children’s depressive symptoms, even after accounting for parental education and negative parenting behaviors. Similarly, Li, Nussbaum, and Richards (2007) found that poverty and other neighborhood stressors predicted higher rates of internalizing and externalizing symptoms in urban Black youth. Furthermore, social-emotional learning (SEL) is one aspect of socioemotional development that is being increasingly identified as an important developmental strength for children with regard to competency in self-awareness, self-management, social awareness, relationship skills, and responsible decision-making (Zins & Elias, 2006). Although few studies have explored poverty’s link to SEL outcomes explicitly, Raver and Knitzer (2002) surmised that poverty is negatively associated with child self-control and that up to 27% of low-income kindergarteners pose classroom problems such as arguments or fights as a function of the severity of their economic stress.

Coping With Poverty: Black Family Resilience

Although the intersection of race and poverty in the United States puts Black children at greater developmental risk due to additive forms of discrimination (e.g., racial, financial, environmental, etc.; Sameroff, 2006), it also presents an opportunity for those Black families to counter the challenges of poverty. Although these families may all be exposed to poverty, not all may experience the same levels of stress. Given the myriad poverty-related risk factors that contribute to compromised child academic, psychosocial, and socioemotional outcomes, it is understandable why the evidence regarding impoverished Black children supports a narrative that fails to adequately investigate those who go on to lead successful lives. Their story, however, deserves to be told.

Resilience.

Theorists have identified familial resilience as a way to better conceptualize the reciprocal processes between parent and child by focusing on familial processes (e.g., parent–child interactions) that occur in the face of adversity (Patterson, 2002). Whereas the FSM indicates that poverty is a risk that can produce less desirable outcomes in children, the theoretical model also incorporates moderators that the family resilience literature has found to be protective of stressors (Garmezy, 1991; Walsh, 2003). Variability within any sample is to be expected; thus, some economically impoverished families have indicated relatively and objectively low financial stress from poverty (Werner, 1989). Understanding parent–child relationships are especially useful for younger and less autonomous children entering school. However, limited research explains the successful components and outcomes for Black children and their families at school entrance (see Boyd-Franklin & Karger, 2012; Nicolas et al., 2008).

Low stress.

Although the majority of variable-centered analyses indicate that poverty is predictive of stress indices, several studies using person-centered approaches (e.g., latent profile analysis [LPA]) provide evidence for relatively low-stress subsamples within economically impoverished populations. Goudie, Narcisse, Hall, and Kuo (2014) explored financial and psychological stressors in a study of parents caring for children with a disability and found three typologies (i.e., high financial and psychological, financial alone, and low stress). Within the three groups, approximately two out of three low-income caregivers would be classified as having low financial and psychological stress (e.g., low stress profile). Although those families who made less than the federal poverty threshold were less likely to be classified as low stress, the two profiles were not mutually exclusive. In addition, Brody and colleagues (2013) examined poverty risk, psychological adjustment, and psychophysiological stress outcomes (i.e., allostatic load) in Black adolescent youth. Within five profiles, findings suggested that approximately one in four youth were characterized by teachers as psychosocially and academically competent relative to their peers, yet no differences were reported between the youth based on level of poverty exposure. Thus, children and adults may experience high poverty, yet report or be observed to exhibit little reactions to stress.

Parent–child relationship.

Scholars have distinguished resilience from risk by examining how proximal family factors, such as parental well-being, positive relationships with a caregiver, and supportive parenting practices, can positively impact youth development (Gutman & Midgley, 2000; Myers & Taylor, 1998). Emotionally responsive caregiving has been found to mediate risk factors and promote positive outcomes for economically impoverished children with high financial and environmental stressors (Egeland, Carlson, & Sroufe, 1993). Further, Myers and Taylor (1998) found that low levels of an element of parent–child conflict, that is, hostile/rejecting parenting, was a key distinguishing factor between profiles of stress-resilient families (e.g., high family stress burden but no clinically significant behavioral problems) and low stress-healthy families (e.g., low familial stress burden and no clinically significant behavioral problems). Even though the stress-resilient families reported higher distress and risk variables, they showed less hostility with their child relative to the low stress-healthy cluster.

A number of family level beliefs and practices impact the relationship between parent and child (e.g., parental expectations; Urdan, Solek, & Schoenfelder, 2007, meaning making; Patterson, 2002, positive outlook; Buikstra et al., 2010, and spirituality; Mattis & Mattis, 2011). In particular, parenting practices within Black families have been identified as culturally appropriate styles which consider the child in his or her larger context (Hill, Murry, & Anderson, 2005; Nobles, 2007). Although a host of familial factors contribute to protective environments, low parent–child conflict has been identified as one of the key components in modeling psychosocially beneficial behaviors for young children (Low & Stocker, 2005). Lower levels of parent–child conflict have been evident for Black families relative to other racial groups in several comparative studies (e.g., Barber, 1994); however, fewer within-group analyses have been conducted to explore variations in Black parent–child relationships (see Nobles, 2007).

Child well-being.

Dyadic and interactive processes from parent to child, especially when the child is young, can help to model the generational transmission of positivity regarding less-thanideal financial circumstances (Mattis et al., 2016). Interventions focused on behavioral changes in young children often integrate families to reinforce skills and systems important for both parent and child (e.g., Early Steps Project; Dishion, Kavanagh, Schneiger, Nelson, & Kaufman, 2002), including the improvement of parent–child relationships. Positive parent–child relationships, moreover, have been found to promote academic performance (Gutman, Sameroff, & Eccles, 2002), psychological adjustment (Zimmerman, Ramirez-Valles, & Maton, 1999), socioemotional learning (Brown et al., 2013), and reduce problem behaviors in youth (Weist, Freedman, Paskewitz, Proescher, & Flaherty, 1995).

Analytical techniques that explore variations in groups indicate a range of adjustment outcomes for at-risk youth (Buckner, Mezzacappa, & Beardslee, 2003; Connell, Spencer, & Aber, 1994). In an LPA of impoverished youth, approximately one in four fit into a “resilient” profile with high levels of stress yet low levels of adjustment problems (Brody et al., 2013). In addition, “stress-resilient” profiles of youth included high family stress but few behavioral problems in youth compared to two other profiles (Myers & Taylor, 1998).

The Current Study

Protective factors within the family resilience literature (e.g., parent–child relationship) may be particularly important to our understanding of preschool to early elementary age children because of the strong influence of family on child outcomes for this developmentally receptive age group (Healy, Sanders, & Iyer, 2015). This is especially relevant because early parenting behaviors and beliefs have been found to be most critical in determining academic readiness for Black and low-income families (Hill, 2001). Whereas scholars agree that the parent–child relationship can be important for successful development and entry into school, much is left to be desired in the empirical literature regarding the impact of parent–child relationships in low-income Black families preparing for school entry (Anderson, 2015; Iruka, Burchinal, & Cai, 2010). In particular, parents who have less conflictual relationships with their children may be modeling behaviors that will better prepare them for the psychosocial expectations of and relationships with their teachers (Fantuzzo, McWayne, Perry, & Childs, 2004).

Thus, two research questions exist for the current study. To continue the investigation of varying analytical approaches to the FSM (Anderson, 2015), the first research question is: do distinct profiles of families with differential types of risk exist within the sample? In particular, are there families who—despite experiencing high levels of financial and personal stress—experience less than expected conflictual parent–child relationships? In accordance with the literature identifying three to five differential profiles of familial risk variables, the emergence of a low, medium, and high stress and conflict profile as well as a high stress and low conflict (e.g., “resilient”) profile is hypothesized. The second question explores associations between the profiles and school readiness, that is: do these profiles differentially predict school readiness (e.g., academic, psychosocial, and socioemotional constructs)? It is hypothesized that school readiness will be most optimal in the low stress/low conflict profile, yet differences will emerge between the high stress/high conflict profile and the high stress/low conflict profile.

Method

Participants

Participants included 292 self-identified Black families in a medium-sized urban southeastern city. The sample participated in three waves of a longitudinal randomized control trial of an SEL afterschool program for kindergarteners attending low-performing schools in high-risk neighborhoods from 2013–2015. Both randomly assigned treatment and control families were included in the sample for this study. No distinctions were made between the groups since data reported in this study were collected prior to the intervention. Target schools were included within the study based on their poverty risk status (e.g., Title 1 designation, or high percentage of low-income children within the school), and any family within the school who submitted an application was accepted into the program.

The sample included 292 school-age children (M_age = 5.42, SD = .33, range = 4.75–6.38 years), including 164 girls (56%) and 128 boys (44%), and their primary caregivers (N = 292; henceforth referred to as parents). Most parents were female (95%, N = 277) and ranged in age from 20 to 76 years (M_age = 31.49, SD = 8.35). The majority of parents were biological mothers (87%, n = 255), followed by grandmothers (6%, n = 17), fathers (5%, n = 14), and aunts (2%, n = 5). Within the sample, 170 parents (58%) indicated being the only caregiver within the home. Of the parents, approximately two thirds completed (n = 116; 41%) or had less than (n = 75; 26%) a high school education. With respect to income, 98% of the sample (n = 273) had children who received free or reduced lunch at school and 240 participants (82%) reported receiving additional public assistance (e.g., temporary assistance for needy families).

Although precise income data were unavailable for the participants, information about families at the school and in the city broadly emphasize their limited economic wherewithal. Students at the schools in the city where this study was conducted were exposed to high levels of academic, economic, and social risk. The vast majority (>90%) of students at these schools were eligible for free or reduced-price lunch. Families within this city had a median family income (~$39,653) that was just over half the national average (~$63,211), and are considered relatively impoverished by national comparison (see Anderson, 2015, for demographic characteristics).

Procedure

Data were collected through several methods and approved by the Institutional Review Board. Most children were directly assessed on academic measures at a summer assessment camp held prior to the first academic year in which they were enrolled in the study. Both consent and assent were obtained during the enrollment of the child in the camp. Child assessment times were typically no longer than one hour. In addition, parents met with research team interviewers who collected data on various family level factors (e.g., family socioeconomic indicators, household structure), parent–child relationships, and child psychosocial and socioemotional functioning (referred to as the parent quantitative interviews). Parents received $40 gift cards for completed interviews. If families were unavailable for testing in the summer, the research team collected measures in the early fall at school or another convenient location.

Measures

Financial strain.

Financial stress was measured through financial strain (Kessler, Turner, & House, 1988) and based on the answers to three questions with a 5-point rating scale (e.g., “How difficult is it for you to live on your total household income right now?”) ranging from 1 (not at all difficult) to 5 (very difficult). Financial strain yielded acceptable reliability for this sample (α = .75).

Perceived Stress Scale.

Parent’s perceived stress was assessed by the 14-item Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983), which evaluated parents’ feelings throughout the past month. Items (e.g., “In the last month, how often have you been upset because of something that happened unexpectedly?”) had a response ranging from 1 (never) to 5 (very often). Relatively high internal reliability was established for the PSS within this sample (α = .82).

Parent–Child Relationship/Parenting Stress.

The 13-item Parent–Child Relationship/Parenting Stress (unauthored and compiled for the Early Childhood Longitudinal Study-Kindergarten) measuring parent and child conflict (e.g., “I am usually too busy to joke and play around with [CHILD]” or “[CHILD] does things that really bother me”) ranged from 1 (not at all true) to 4 (completely true), and caregivers could also indicate that they did not know or would prefer not to answer. Four of the responses were reverse-coded. The Parent–Child Relationship/Parenting Stress yielded average reliability (α = .70) for this sample.

Woodcock-Johnson-III Tests of Achievement.

For academic readiness, reading was assessed through the Letter-Word Identification subtests and mathematics through the Applied Problems subtest. In the Letter-Word Identification, the word red is presented in print, and the administrator would ask the child “What does this word say?” The Woodcock-Johnson-III Tests of Achievement (Woodcock, McGrew, & Mather, 2001) is a widely used, individually administered assessment battery that measures general cognitive abilities and achievement in individuals with standardized performance relative to his or her same-age population. Reliability of the scales could not be established within this sample given limited responses due to increasing item difficulty.

Differential Abilities Scales-II.

Children’s verbal skills were measured using the Differential Abilities Scales-II (Elliott, 2007) Verbal Comprehension and Naming Vocabulary subtests. As an example, a child may be asked to sort toys and present the requested item (e.g., a watch). The Differential Abilities Scales-II is a comprehensive clinical instrument that assesses cognitive abilities important to learning. This sample yielded acceptable (naming vocabulary; α = .80) and unestablished (verbal comprehension) validity within this sample.

Social Skills Improvement System.

Parents were asked about children’s problem behaviors through the Social Skills Improvement System (Gresham & Elliott, 2008). Specifically, the Internalizing and Externalizing scales within Competing Problem Behaviors were used. Items included, “How often does s/he act sad or depressed?” (internalizing) or “How often does s/he fight with others?” (externalizing), on a scale of 1 (never) to 4 (almost always). Internal consistency reliability varied within this sample for internalizing (α = .65) and externalizing (α = .85) problems.

The Devereux Student Strengths Assessment.

The Devereux Student Strengths Assessment (DESSA; LeBuffe, Shapiro, & Naglieri, 2009) was administered to parents to assess the social-emotional competencies of children. The 45-item (e.g., “How often did the child cope well with insults and mean comments?”) measure ranged from 1 (never) to 5 (very frequently). The components of socioemotional learning were evaluated through the Self-Awareness, Social Awareness, Decision-Making, Self-Management, and Relationship Skills scales. Internal reliability varied from acceptable to excellent within this sample (αs = .69, .83, .88, .78, and .84, respectively).

Data Analyses

The current investigation used a person-centered analysis via LPA in MPlus 7.0 (Muthén & Muthén, 2007). LPA was used to examine potential profiles of individual families who share common characteristics of financial stress, general stress, and parent–child relationships. Given that various profiles were discovered, differences in membership were evaluated with respect to the predictor variables as well as the school readiness variables by multivariate analyses of variance (MANOVAs) in SPSS Version 23.0. Composite scores were created for academic (i.e., letter-word identification, naming vocabulary, applied problems, and verbal comprehension), psychosocial (i.e., internalizing and externalizing problems) and socioemotional measures (i.e., social awareness, decision making, relationship skills, self-awareness, and self-management). Since academic content had varying scale anchors, scores for each scale were standardized and the mean score was used to create the composite and the mean scores of the psychosocial and socioemotional measures were utilized for the composite.

Results

As a whole, participants reported little financial strain (M = 2.03, SD = .92) and almost never/sometimes experienced stress (M = 2.56, SD = .58). Parents also indicated that it was not at all/somewhat true that they experienced problems with their children (M = 1.49, SD = .37). Descriptively, parents reported that their children had few psychological problems (M = 1.59, SD = .39) and great socioemotional skills (M = 4.17, SD = .53). Children’s directly assessed academic scores reflected slightly below average performance for their age in applied problems (M_{StandardScore} = 93.82, SD = 9.92), letter writing identification (M_{StandardScore} = 98.81, SD = 12.04), verbal comprehension (M_Tscore = 43.54, SD = 6.40), and naming vocabulary (M_Tscore = 45.47, SD = 8.11). Table 1 provides the correlations, means, and standard deviations for all study variables.

Table 1.

Correlations, Ms, SDs, and Internal Validity Among Study Variables (N = 292)

Variable	1	2	3	4	5	6
1. Financial stress	—	.55**	.31**	−.11	.20**	−.16**
2. General stress		—	.46**	−.14*	.38**	−.32**
3. Parent–child relationship problems			—	−.13*	.45**	−.35**
4. Academic readiness				—	−.25**	.16**
5. Psychological readiness					—	−.52**
6. Socioemotional readiness						—
M	2.03	2.56	1.49	.00^	1.59	4.17
SD	.92	.58	.37	.72	.39	.53
Range	1.00–5.00	1.07–4.21	1.00–3.15	−2.30–1.96	1.00–3.17	2.29–5.00
α	.75	.82	.70	—	.88	.90

Note.

^{^}Denotes standardized score due to differential scores within measures.

^*p < .05.

^**p < .01.

LPA

LPA was conducted to determine whether classifications existed within the various stressors and relationships in low-income Black families. LPA was performed on a two-, three-, four-, and five-class model. Several fit indices were considered in the selection of profiles, including lower Akaike Information Criterion (AIC) and Bayesian Information Criteria (BIC) and higher entropy values. Given the preference of BIC in continuous variables such as those used within the LPA (Nylund, Asparouhov, & Muthén, 2007) as well as the standard of the minimum power criteria of 20 participants per cell (Cohen, 1992), the four-class model indicated the best fit (BIC = 1037.51; entropy = .76; please see Table 2) and was retained in further analyses. The latent classes within the four-class model were characterized relatively within the sample by: low levels of financial and general stressors and parent–child relationship problems (n = 121; low stress, low conflict [LSLC]), average levels of financial and general stressors and parent–child relationship problems (n = 111; moderate stress/moderate conflict [MSMC]), average financial and high general stressors with very high parent–child relationship problems (n = 20; moderate stress/high conflict [MSHC]), and very high financial stress and high general stress and moderate parent–child relationship problems (n = 40; termed high stress/moderate conflict [HSMC]). There were no significant differences between profiles in demographics including parent age, educational attainment, employment status, and child gender.

Table 2.

Latent Class Analyses Solutions

Variables	N Class	AIC	BIC	BLRT (p value)	Entropy	N Cell 1	N Cell 2	N Cell 3	N Cell 4	N Cell 5
FS/GS/PCR	5	2252.43	2333.32	27.21 (%.001)	.80	24	121	103	7	37
FS/GS/PCR	4	2271.64	2337.82	27.94 (<.001)	.76	20	111	121	40
FS/GS/PCR	3	2291.58	2343.06	72.56 (%.001)	.76	120	127	45
FS/GS/PCR	2	2356.13	2392.90	146.84 (%.001)	.67	165	127

Note. FS = financial stress; GS = general stress; PCR = parent-child relationship problems; AIC = Akaike information criterion; BIC = Bayesian information criterion; LMR = Lo-Mendell-Rubin; BLRT = bootstrapped likelihood ratio test. Bolded text represents final solution.

MANOVAs

MANOVA tests were conducted to detect differential means between groups. The MANOVAs detected significant differences between latent profiles in the stress and problem variables, F(9, 864) = 125.57, p < .001; Wilk’s Λ = .10. Financial stress, F(3, 288) = 332.95, p < .001, η² = .78, general stress, F(3, 288) = 120.61, p < .001, η² = .56, and parent–child relationship problems, F(3, 288) = 81.80, p < .001, η² = .46 were significantly different between profiles. Fisher’s least significant difference post hoc tests revealed significant differences between each of the profiles in financial stress, general stress, and parent–child relationship with the exception of MSHC and HSMC in general stress (all significant ps < .01; see Table 3). Differences were evident between groups in their naming conventions, such that any variable (e.g., financial stress) was statistically less (i.e., LSLC) or greater (i.e., HSMC) than the others relative to their name. Stress scores ranged from one to five, while parent–child conflict scores ranged from one to four. Thus, scores closer to five in stress (e.g., HSMC) and closer to four in parent–child conflict (e.g., MSHC) represented greater risk. Given MSHC’s lower financial stress and higher parent–child conflict relative to HSMC parents, differences between these groups are especially useful for understanding the impact of parent–child conflict on child school readiness.

Table 3.

Means and Standard Deviations for Profile Variables

Variable	LSLC (n = 121) M (SD)	MSMC (n = 111) M (SD)	MSHC (n = 20) M (SD)	HSMC (n = 40) M (SD)
Financial stress	1.29 (.34)a,b,c	2.23 (.45)d,e	1.96 (.62)f	3.76 (.54)
General stress	2.07 (.38)a,b,c	2.78 (.35)d,e	3.11 (.37)	3.16 (.50)
Parent–child relationship conflict	1.29 (.25)a,b,c	1.48 (.26)d,e	2.25 (.34)f	1.73 (.33)

Note. LSLC = low stress/low conflict; MSMC = moderate stress/moderate conflict; MSHC = moderate stress/high conflict; HSMC = high stress/moderate conflict. All differences are p ≤ .01.

^aDifferences between groups LSLC and MSMC.

^bDifferences between groups LSLC and MSHC.

^cDifferences between groups LSLC and HSMC.

^dDifferences between groups MSMC and MSHC.

^eDifferences between groups MSMC and HSMC.

^fDifferences between groups MSHC and HSMC.

Differences were also detected in school readiness variables that were assessed in children and reported by parents F(9, 843) = 7.05, p < .001; Wilk’s Λ = .81. Directly assessed child academic readiness was equivalent in all latent profiles, F(3, 281) = 2.23, p > .05, η² = .02, with the exception of a between-groups difference in HSMC and LSLC (p < .05; see Table 4). Group means significantly differed between parental measures of psychosocial problems, F(3, 288) = 16.20, p < .001, η² = .15. Parents in HSMC, MSMC, MSHC, and LSLC rated their children as being significantly different from each other in externalizing and internalizing problems (ps < .01). Parents in LSLC rated their children as having the least psychological problems (M = 1.45, SD = .32), whereas MSHC children were rated as having the most (M = 1.98, SD = .45). Between-profile differences were also found in socioemotional behaviors, F(3, 288) = 12.36, p ≤ .001, η² = .12. For overall socioemotional performance, each profile rated their children as significantly different from their peers (ps < .05) with the exception of MSMC and HSMC profiles. Differences suggest that parents in LSLC reported their children as having higher socioemotional skills than their peers (M = 4.34, SD = .44) while children in the MSHC profile were rated lowest in socioemotional skills (M = 3.70, SD = .62).

Table 4.

Means and Standard Deviations for School Readiness Variables by Profile

Variable	LSLC (n = 117) M (SD)	MSMC (n = 108) M (SD)	MSHC (n = 20) M (SD)	HSMC (n = 40) M (SD)
Academic^	.11 (.69)c	(.06 (.73)	.04 (.71)	(.19 (.73)
Psychosocial	1.45 (.32)a,b,c	1.62 (.37)d,e	1.98 (.45)f	1.75 (.44)
Socioemotional	4.34 (.44)a,b,c	4.12 (.48)d	3.70 (.62)f	4.02 (.65)

Note. LSLC = low stress/low conflict; MSMC = moderate stress/moderate conflict; MSHC = moderate stress/high conflict; HSMC = high stress/moderate conflict. All differences are p ≤ .05.

^{^}Denotes standardized score due to differential scales within measures.

^aDifferences between groups LSLC and MSMC.

^bDifferences between groups LSLC and MSHC.

^cDifferences between groups LSLC and HSMC.

^dDifferences between groups MSMC and MSHC.

^eDifferences between groups MSMC and HSMC.

^fDifferences between groups MSHC and HSMC.

Discussion

The purpose of this study was to determine how parent–child relationships function in relation to financial and general stressors in the prediction of child school readiness (e.g., academic performance, psychosocial development, and socioemotional skills). With the FSM as a framework for this inquiry, we assessed how stressors often related to poverty (e.g., financial and general stress and parent–child relationship conflict) may have differing compositions when assessed through a person-centered approach. In families for whom parent–child conflict was relatively low, children were rated as more academically, psychologically, and socioemotionally prepared for school, even if the parents reported relatively greater financial stress.

With regard to the first research question, prior studies utilizing a profile approach to disaggregate the linear relationships between parental risk variables contributed to the hypothesis that at least four profiles of variable stress and parent–child conflict combinations would be present in this sample (Brody et al., 2013; Goudie et al., 2014; Myers & Taylor, 1998). Although mean scores were relatively low for stress and parent–child conflict measures generally, four unique profiles emerged, as expected, that elucidated differences between families with financial stress, general stress, and parent–child relationship problems. Although two of the four groups supported the hypothesis and appeared to be best explained by linear relationships with respect to their means (i.e., LSLC and MSMC), two other groups (i.e., MSHC and HSMC) emerged, representing approximately 21% of the study population with statistically higher than average reports of conflictual relationships or financial stress, respectively, based on the study means.

The association between the profiles and school readiness variables were the focus of the second research question. Significant differences were found between families’ stress/conflict profiles and their school readiness that were consistent with the hypothesis that lower parent–child conflict would predict more desirable school readiness. Findings indicated that children with parents in LSLC had the most desirable directly assessed child academic performance and parent-reported psychosocial and socioemotional readiness scores relative to other profiles. Yet, children with parents in MSHC had less desirable scores than families in both the MSMC and HSMC as predicted, implicating high levels of parent–child relationship conflict as a distinguishing factor from the similarly or higher stressed profiles. Higher conflict relationships have been demonstrated to be harmful to students for school entry in psychological and prosocial abilities (Connell & Prinz, 2002; Healy, Sanders, & Iyer, 2015; Ostrov & Bishop, 2008).

Close relationships with children are often heralded as a strength within the Black community (McAdoo & Younge, 2009). While Black families have consistently demonstrated less conflict between parents and children relative to families of other races (Barber, 1994; Davis-Kean & Sexton, 2009), even relatively higher parent–child conflict within group may contribute to less desirable school readiness. Thus, for MSHC families who were moderately stressed in financial and general concerns, relatively high conflict with their child may be a more challenging stressor than resource stressors (Gerard, Krishnakumar, & Buehler, 2006).

Although a profile did not emerge that was explicitly high stress and objectively low parent–child conflict, the “resilient” HSMC profile does provide insight into families with lower levels of parent–child conflict relative to their high levels of stress. Two times the number of families were categorized as HSMC relative to MSHC. It is possible that the HSMC parents may have experienced personal stressors but attempted to provide a context for their children that was protective of these stressors (Young, Polansky, Chalmers, Buttenweiser, & Williams, 1982). A variety of bonadaptive and maladaptive strategies exist to cope with financial stressors (Patterson, 2002), so clinical interventions can cultivate skills to contend with familial conflict (Walsh, 2003) and personal stress (McEwen & Gianaros, 2010). Whether the differences in stress is a function of measurement or impression management is a greater question of the field.

Given that stress is often assessed in three ways (e.g., identification, appraisal, and management; Lazarus & Folkman, 1984), a parent in this study may have identified, appraised, and/or managed her stressors differently from her counterparts; however, a self-report measure makes it difficult to assess which of the three components of stress is being perceived and reported by the parent. As an example, two parents facing the same objective stressor may rate their experience with the stress as relatively different. Conversely, two parents may face different challenging experiences, yet may interpret the stress from the experience as the same. Detailed measurement of the objective and subjective stressor experience can help to promote more accurate responses in future work addressing the management of poverty, general, and parent–child relationship stressors. In addition, if the participant perceives the researcher to differ in sociodemographic status, questions focused on such characteristics (e.g., stress due to poverty) may be challenging to collect verbally. Moreover, although participants may report subjective and perceived forms of stress, various types of measurements (e.g., allostatic load; Brody et al., 2013) may help to explain how stress manifests within the body. Finally, the exploration of qualitative accounts may unearth why parents indicated low levels of stress, including relative comparisons to others perceived to be more stressed by their financial situation.

Implications

Whereas the negative impact of poverty on children and families is well documented and should not be minimized, it is important to also understand the strengths that families living in poverty have and the strategies that they have developed to counteract environmentally adverse circumstances. Although outside of the scope of the current investigation, exploration of the qualitative ways in which “resilient” families in the HSMC profile were able to buffer their parent–child relationship from otherwise deleterious financial and general stressors may contribute to our understanding of familial coping strategies. To be sure, Wadsworth and colleagues (2013) have recently begun their exploration of the secondary coping strategies employed by low-income families through quantitative work in the Adaptation to Poverty Related Stress (APRS) model by furthering our understanding of potentially useful thought processes that can contribute to such a family resilience framework. Continued exploration of the APRS model, particularly as it pertains to family therapy with an explicit focus on financial stressors and problem-solving strategies, would be of great potential benefit to impoverished families struggling with maladaptive coping and personal stress (Wadsworth et al., 2013).

Furthermore, the recognition that Black families are not a monolith and have varying stressors and school entry characteristics is of import to this study. Literature that conceptualizes the impact of poverty on families indicates that teachers and schools may be primed to believe that the vast majority of impoverished students may face challenges in school (Bomer, Dworin, May, & Semingson, 2008). Although the literature is abundant regarding the impact of poverty on academic achievement and behaviors, this study also highlights the importance of perceptions of family stressors on varying school entry characteristics that may differentially prepare children for school and are malleable via health and school-based initiatives (e.g., CA “First 5”).

The characterization that the majority of families who live in poverty are stressed and maintaining problematic relationships was not supported within the current study (see Baldwin, Baldwin, & Cole, 1990). The vast majority of families were represented in profiles with either moderate or low levels of stress and parent–child conflict relative to the sample mean and midpoint of the scale. Because families living in impoverished areas have been found to have greater levels of stress than their counterparts (Conger et al., 2002), these unexpected relatively low and moderate stress scores reported by parents over the waves in the study provided the impetus for exploring relationships between parent–child interactions and child outcomes.

Although this sample did not report particularly high stress, both the LPA classifications (i.e., HSMC) and a range including the highest degree of difficulty experienced by the participants (e.g., M = 5.0) indicate approximately 25% of participants perceiving high levels of financial and general stress within the sample. It is important to note the potential protective benefits of lower perceived stressors by participants facing similar life circumstances. Within qualitative interviews, participants often determined their level of stress based on comparisons to those within their proximal environment. Thus, what may be perceived as “objectively” stressful to researchers is “subjectively” manageable to participants responding to survey inquiries. However, even within this relativity, children whose parents were in the MSMC group were rated as significantly more prepared for school than their MSHC counterparts. This suggests that stress reduction may be important in the lives of those in poverty given the associations between stress and conflictual parent–child relationships. Although a reduction in poverty overall is ideal, efforts focused on the specific reduction of stressors may help to free resources to bolster parent–child relationships. Although the access, utilization, and quality of mental health care within low-income areas is fraught with challenges, there may be ways to encourage parental components alongside after school programs or interventions, such as group therapy, exercise, or mindfulness practices (Coatsworth, Duncan, Greenberg, & Nix, 2010). Further, programs targeted at children or the family may address individual parent factors (e.g., depression) and familial dynamics (e.g., how to support a parent when she is depressed) to improve parent functioning and subsequent child school readiness.

In tandem, better understanding cultural factors that interact with socioeconomic risk (e.g., discrimination) would provide greater context for impoverished Black families (Anderson et al., 2015). Beyond universal protection, scholars argue that there are specific family based cultural factors (e.g., “no-nonsense” parenting, racial socialization, etc.) that can help to serve as protective in the lives of ethnic minority youth (Smith-Bynum, Anderson, Davis, Franco, & English, 2016; Williams et al., 2014). Such factors, however, are often not captured by universal parent–child relationship measures nor assessed in families with younger children (Anderson, 2015). Studies with a specific focus on Black families should investigate how these cultural components can impact relational dynamics between parents and children.

Limitations

Some shortcomings exist within the current study. This is the first wave from three cohorts of a longitudinal study, and the cross-sectional results only explore the kindergarten school-year. As the project progresses, academic, behavioral, and prosocial outcomes will be available for these youth throughout second grade allowing for longitudinal and growth-curve modeling. Second, the quantitative interviews were conducted and recorded verbally, raising possible concerns of social desirability between families and researchers. Although the members of the study team wanted to ensure that literacy would not be a hindrance, recent studies have demonstrated that technology (e.g., iPad or laptop computer) can assist in the collection of data to encourage candid responses and a sense of privacy. In addition to reducing social desirability, this method may ameliorate other potential reporting biases (e.g., those due to cross-cultural concerns that may arise when study populations make meaning of the cultural and/or sociodemographic differences between themselves and researchers).

Conclusion

Poverty can prove challenging to any person individually, but family systems within impoverished environments can be particularly strained by limited resources, energy, and greater conflict in parent–child dyads. Although evidence suggests that families with the relatively lowest financial strain, general stress, and parent–child conflict have children with more skills in parent-reported and directly observed school readiness criteria, families differentially reported some of the stressors of poverty that were associated with varying degrees of school readiness. It is our hope that their varying profiles encourage us to think more critically about the diversity within Black families to propose comprehensive theories and services that can support the relationships between parents and school-age children in urban low-income Black families.