Family and Cultural influences on Low-income Latino Children’s Adjustment

Abstract

This study examined family and cultural influences on adjustment among ninety low-income Latino middle school children (46% girls; Average age = 11.38, SD = .66) and their primary caregivers (93% Female; Average age = 36.12, SD = 6.13). All participants identified as Hispanic/Latino, with 75% of families identifying as Mexican-origin Latino, and 77% of parents and 32% of children identifying as immigrants. Hierarchical linear modeling analyses revealed that family reframing interacted with familism, with high levels of both associated with fewer psychological symptoms, while passive appraisal is linked to worse functioning. Results are discussed with regard to the implications of this research for preventive interventions with families in poverty.

Latino youths experience proportionately more depression, anxiety, and delinquency than do non-Hispanic white youths and are overrepresented among poor children (Surgeon General, 2000). Poverty contributes to these negative outcomes for children, including symptoms of depression, anxiety, aggression, and poor scholastic performance (Grant et al., 2003). Poverty affects children and families largely through poverty-related stress (PRS), a collection of stressors that frequently co-occur in the lives of low-income families, including exposure to violence, economic stress, deteriorating family relationships, and discrimination (e.g., Wadsworth et al., 2008). The accumulation of these situations and events creates a “context of stress,” whereby stressors build on and exacerbate each other (McLoyd, 1990), contributing to poor psychological functioning (Wadsworth et al., 2008).

Latino families who can successfully cope with stress by using family coping strategies work to maintain positive family relationships and structure, thereby protecting against upheaval in the family system. Family coping is comprised of a family’s strategies and behaviors aimed at strengthening the family as a whole, maintaining emotional stability and well-being of family members, using family and community resources to manage a situation or event, and making efforts to problem-solve family hardships created by stress (McCubbin, McCubbin, Thompson, & Thompson, 1998). Specifically family coping includes strategies that maintain a positive outlook (family reframing), encourage family participation in traditional spiritual or religious activities (family spiritual support), encourage positive family relationships and getting help from relatives (family social support), and manage community resources to assist the family (family mobilizing support). A family’s passive appraisal of a situation is considered a maladaptive strategy.

Family coping is likely to be particularly relevant within the Latino family context due to the traditional emphasis on loyalty and sharing with family (Cuellar, Arnold, & Gonzales, 1995). Research examining family coping has not yet been conducted in the context of PRS, but has found positive effects in adjusting to other types of stress, such as a child’s developmental disability and parental chronic illness (e.g., Korneluk & Lee, 1998; Lustig, 2002). Immigrant families typically hold strongly to traditional ideals, including valuing and staying loyal to the family, participating in traditional cultural activities, and teaching cultural beliefs to children. Family coping strategies may be particularly beneficial for these traditionally oriented families. Indeed, family pride predicts greater engagement in family coping strategies such as reframing and spiritual support (Hanline & Daley, 1992). As families acculturate, coping as a family may remain essential for those who maintain an emphasis on family and culture, while coping individually may be more useful for children in families with lower levels of familism or family ethnic socialization. Thus, these cultural factors may moderate the effectiveness of various family coping strategies, especially those that are consistent with positive attitudes toward the family or traditional values—family reframing, family spiritual support, and family social support.

This study examined main and interactive associations of PRS, family coping, immigrant/generational status, family ethnic socialization, and familism on child mental health. We predicted that family coping would have direct associations with child mental health. In addition, we hypothesized that family coping would moderate the negative effects of PRS on child mental health. Further, we predicted that among families with high degrees of familism and family ethnic socialization, family coping would be most helpful. In sum, the hypotheses of the current study were: (1) positive family coping strategies (e.g., family reframing, family spiritual support) would be related to fewer child internalizing and externalizing symptoms, (2) family coping would moderate the effects of PRS on internalizing and externalizing symptoms, and (3) cultural orientation (familism, family ethnic socialization, and immigrant status) would moderate the effects of family reframing, family spiritual support, and family social support, such that these strategies would have stronger effects for families with stronger cultural orientation.

Method

Participants

Ninety low-income Latino middle school children (46% girls; Average age = 11.38, SD = .66) and a primary caregiver (93% Female; Average age = 36.12, SD = 6.13) participated in the study. All participants identified as Hispanic/Latino (75% Mexican-origin Latino, 1% Guatemalan, 2% biracial (Hispanic and Caucasian), 22% Latino/Hispanic without additional descriptors). Among participants, 77% of parents and 32% of children identified as immigrants, 9% of parents and 45% of children identified as first generation Americans, 5% of parents and 13% of children identified as second or third generation, and 9% of parents and 10% of children indicated that their families had resided in the U.S. for many generations. The average monthly income reported was $1806.18 (SD = 906.69) with the average family size being 5.02 (SD = 1.45). All children participating in the study were receiving free or reduced lunch, a poverty indicator. Parental occupation was coded using Hollingshead’s (1975) 100-point scale with higher scores indicating higher prestige jobs. The average coded maternal occupation score was 18.59 (SD = 14.41; range 10–70), with 64% of occupations coded at 10 (e.g., cleaning services, homemaker). The average coded paternal occupation score was 26.79 (SD = 11.79; range 10–70), with 63% of occupations coded at 20 (e.g., construction, food services).

Procedure

A focus group was conducted to ensure cultural relevance and appropriateness of the study questions and measures. Study participants were recruited through Denver-area middle schools with above 90% enrollment in the free or reduced lunch program and with greater than 95% of children identifying as Latino/Hispanic and through summer programs geared toward low-income youths. Researchers attended school meetings, parent nights, registration, and school/program banquets, coming into contact with approximately 150–175 eligible families (yielding an estimated participation rate of 50–60%). At these events, parents were invited to participate and sign up for the study after a brief presentation describing participant involvement and general study aims. Of those that made initial contact, 90% completed the study. During appointments, bilingual research assistants obtained informed consent from parents and assent from children and then administered questionnaires and a short interview to parents. Among participants, 77% of parents and 23% of children chose to complete questionnaires in Spanish. Participants received financial compensation. Research was conducted in compliance with the University of Denver’s Institutional Review Board.

Measures

Demographic Information

Demographic information was collected on surveys and through brief parent interviews.

Poverty-related stress

Children and parents completed the Multicultural Events Schedule for Adolescents (MESA; Gonzales, Tein, Sandler, & Friedman, 2001), capturing stressful events that have happened within the family. Five subscales (57 items) used in previous research to create a reliable composite were used: economic strain, family moves and transitions, exposure to violence, family conflict, and daily hassles/discrimination (Wadsworth et al., 2008). The MESA was developed and validated using an ethnically diverse, low-income population and has demonstrated adequate test-retest reliability (r = .71). The subscales were summed for an overall PRS score. Cronbach’s alphas in this sample were .88 (child report) and .75 (parent report).

Parents were also interviewed using the Economic Hardship Questionnaire (EHQ; Lempers, Clark-Lempers, & Simons, 1989), an 11-item measure that assesses the number of constraints the family felt as a result of economic hardship in the last six months (e.g., “We have had to sell possessions to make ends meet”). Cronbach’s alpha was .88 in this sample. For parent report of PRS, EHQ and MESA scores were standardized and averaged. Scores from the EHQ and MESA were significantly correlated (r = .45).

Family Coping

Children and parents completed the Family Crisis Oriented Personal Evaluation Scales (FCOPES; McCubbin, Olson, & Larsen, 1987), designed to assess family coping strategies and adjustment to life stressors. Participants were instructed to complete the measure thinking about the stressful events on the PRS measure and what their family does to manage them. The FCOPES is a 30-item measure with the following subscales (example items in parentheses): Acquiring Social Support from relatives (Seeking advice from relatives), Family Reframing (Defining the problem in a more positive way so that we do not become too discouraged), Spiritual Support (Attending church services), Mobilizing Support (Seeking assistance from community agencies), and Passive Appraisal (Believing if we wait long enough, the problem will go away). The FCOPES has been used and validated with Spanish-speaking populations. Past reliability ranged from .71 to .86. Cronbach’s alphas in this sample ranged from .60 (Passive Appraisal) to .86 (Family Reframing). Sandler, Tein, and West (1994) suggest that coping strategies are best understood in the context of other coping strategies, and Connor-Smith, Compas, Wadsworth, Thomsen, and Saltzman (2000) recommend proportion scoring to take into account other responses to stress. Thus, raw scores were utilized in primary analyses where the effects of other strategies were controlled for in the analysis. Among correlations, proportion scores were utilized.

Child Mental Health

Children completed the Youth Self Report (YSR) and parents completed the Child Behavior Checklist (CBCL), both of which are part of the Achenbach System for empirically based assessment (Achenbach & Rescorla, 2001). The YSR and CBCL yield scores for Internalizing symptoms and Externalizing symptoms. There is evidenced to support the reliability and validity of the YSR and CBCL as a measure of children’s adjustment (Achenbach & Rescorla, 2001).

Culture

Parents and children completed a revised version of the Familial Ethnic Socialization Measure (FESM; Umaña-Taylor, Yazedijan, & Bámaca-Gómez, 2004). This 12-item measure assesses the degree to which cultural or ethnic practices occur within the family (e.g., “My family teaches me about our family’s ethnic/cultural background”). Cronbach’s alphas ranged from .92 to .94 with ethnically diverse samples (Umaña-Taylor et al., 2004). Cronbach’s alphas in this sample were .91 for child report and .90 for parent report.

Parents and children also completed the Familism Scale (Gil, Wagner, & Vega, 2000). This 7-item scale has been used with multiple Latino samples and includes items “We are proud of our family,” and “We share similar values and beliefs as a family.” It has attained previous reliability of .87. Cronbach’s alphas in this sample were .89 for child report and .90 for parent report.

Results

Data were checked for skewness, kurtosis, and extreme outliers before primary analyses were conducted. Correlations and descriptive statistics are reported in Tables 1. T-tests were conducted to examine differences across child gender and across immigrant status. Child report of Family Social Support (t (df= ) = −2.47, p = .02) was significantly different, with girls reporting significantly more family social support (M = 30.44, SD = 7.35), compared to male children (M = 26.61, SD = 7.32). Passive appraisal and internalizing symptoms were significantly different across immigrant status (t (df= )= −2.75, p = .01; t (df= ) = −2.27, p = .03), with immigrant parents reporting higher levels of passive appraisal and child internalizing symptoms (M = 12.36, SD = 3.23; M = 11.20, SD = 7.46) compared to non-immigrant parents (M = 10.10, SD = 3.53; M = 6.90, SD = 8.04). Associations with SES were also examined. There were no significant correlations between income or parental occupation with parent or child report of the primary variables. Income-to-needs (a ratio of income to the federal poverty line based on family size) was significantly correlated to parent-report of family passive appraisal (r = −.23, p = .03), and child-report of internalizing and externalizing symptoms (r = −.23, p = .03; r = −.22, p = .04). Primary analyses were repeated controlling for gender and income-to-needs, and findings were highly consistent. Immigrant/generational status was tested in primary analyses for main and interactive effects, but was not significant.

Table 1.

Correlations and Descriptive Statistics

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.
1. PRS	.14	.19+	−.24*	.05	.01	−.02	−.03	−.23*	.26*	.31**
2. Family Acquiring Social Support	.20+	.02	−.54**	−.41**	.11	−.21*	.07	−.03	.23*	.18+
3. Family Reframing	−.05	−.49**	.28**	−.10	−.37**	−.05	−.02	.12	−.27*	−.26*
4. Family Spiritual Support	−.13	−.45**	−.12	.33**	−.34**	−.30**	.07	.02	−.14	−.12
5. Family Mobilizing Support	.02	.32**	−.46**	−.25*	−.04	−.28**	−.11	−.06	.20	.12
6. Family Passive Appraisal	−.09	−.42**	−.14	−.04	−.39**	.15	−.05	−.07	.02	.13
7. Family Ethnic Socialization	.13	−.01	.27**	−.01	−.20+	−.15	.13	.53**	.05	−.12
8. Familism	−.10	−.08	.36**	.21*	−.21*	−.36**	.53**	.23*	−.07	−.26*
9. Internalizing Symptoms	.34**	.17	−.18+	−.09	.08	.03	−.02	−.23*	.12	.66**
10. Externalizing Symptoms	.28**	.30**	−.21*	−.20+	.15	−.05	−.15	−.34**	.62*	.38**
Parent Report M	.01	28.75	30.56	14.46	13.13	11.83	43.74	30.04	10.20	7.50
Parent Report SD	.88	7.35	6.17	3.82	3.91	3.42	10.18	5.34	7.77	6.69
Child Report M	6.11	28.36	29.59	13.93	11.33	12.81	43.49	30.22	12.70	8.52
Child Report SD	5.77	7.54	6.27	3.58	4.43	3.51	10.79	5.35	8.14	7.31

Note. Parent report is shown below diagonal. Child report is shown above diagonal. Cross-informant correlations are shown in bold.

⁺p < .10

^*p < .05

^**p < .01

Hierarchical Linear Modeling

Hierarchical linear modeling (HLM 6.03 software; Raudenbush, Bryk, & Congdon, 2004) was used to examine main and interactive associations of PRS, family coping, and cultural factors on symptoms. Hierarchical linear modeling was chosen as an ideal analytic strategy because it allows for measures to be nested within persons. Thus, both child reports and parent reports on each variable could be included within the same analysis. A similar approach has been used to model couple level data with male and female partner reports (e.g., Atkins, 2005). However, our model is more limited than the Atkins (2005) approach because we lack longitudinal data and therefore cannot include a level for time. Due to the more limited nature of our model, we also ran our analyses separate by reporter and found similar results. For the HLM model, level 2 consisted of the 90 children, while level 1 consisted of measures (two measures per variable). The final model with only main effects is as follows (where i refers to person and m refers to measure/reporter):

$$\text{Level}1:{(\text{Psychological}\text{Outcome})}_{\text{mi}}={\pi }_{0\text{i}}+{\pi }_{1\text{i}}{(\text{PRS})}_{\text{mi}}+{\pi }_{2\text{i}}{(\text{Family}\text{Social}\text{Support})}_{\text{mi}}+{\pi }_{3\text{i}}{(\text{Family}\text{Reframing})}_{\text{mi}}+{\pi }_{4\text{i}}{(\text{Family}\text{Spiritual}\text{Support})}_{\text{mi}}+{{\pi }_{5\text{i}}(\text{Family}\text{Mobilizing}\text{Support})}_{\text{mi}}+{\pi }_{6\text{i}}{(\text{Family}\text{Passive}\text{Appraisal})}_{\text{mi}}+{\text{e}}_{\text{mi}}$$

$$\begin{array}{l}\text{Level}2:{\pi }_{0\text{i}}={\beta }_{00}+r_{0\text{i}}\\ {\pi }_{1\text{i}}={\beta }_{10}\\ {\pi }_{2\text{i}}={\beta }_{20}\\ {\pi }_{3\text{i}}={\beta }_{30}\\ {\pi }_{4\text{i}}={\beta }_{40}\\ {\pi }_{5\text{i}}={\beta }_{50}\\ {\pi }_{6\text{i}}={\beta }_{60}\end{array}$$

The models were conducted for internalizing symptoms and externalizing symptoms. A main effects model was conducted first. Next, interaction terms between PRS and each family coping scale were entered. Interaction-terms, created from the product of PRS (centered) and each centered family coping variable were entered and tested sequentially. No significant interaction terms between PRS and family coping emerged. Finally, interaction terms were tested between family coping and cultural variables. Immigrant/generational status, familism, or family ethnic socialization were entered as independent variables tested separately to conserve power. In addition, two-way interaction terms including each family coping scale with immigrant/generational status, familism, and family ethnic socialization were tested sequentially. Non-significant terms were dropped from the models.

Final models with significant interaction terms are presented in Table 2. PRS was significantly related to more internalizing and externalizing symptoms. An interaction between familism and family reframing emerged significant for both outcomes, showing that familism enhanced family reframing and high levels of both were especially helpful. Family Passive Appraisal was related to more internalizing symptoms. Acquiring Social Support was related to more externalizing symptoms.

Table 2.

Hierarchical Linear Models

	Coefficient	SE	t-ratio		Coefficient	SE	t-ratio
Internalizing Symptoms Model				Externalizing Symptoms Model
Intercept	13.5664	4.2365	3.20**	Intercept	15.5200	3.5766	4.34**
Poverty-related Stress (PRS)	1.9108	.6249	3.06**	Poverty-related Stress (PRS)	1.5614	.4327	3.61**
Family Acquiring Social Support	.1722	.0993	1.73	Family Acquiring Social Support	.2008	.0893	2.25*
Family Reframing	−.3006	.1263	−2.38*	Family Reframing	−.1517	.1037	−1.46
Family Spiritual Support	−.0153	.2049	−.08	Family Spiritual Support	−.1070	.1395	−.77
Family Mobilizing Support	.1227	.1713	.72	Family Mobilizing Support	.0168	.1472	.11
Family Passive Appraisal	.4182	.1793	2.33*	Family Passive Appraisal	.2560	.1397	1.83
Familism	−.1384	.1125	−1.23	Familism	−.3400	.1009	−3.37**
Familism X Family Reframing	−.0247	.0076	−3.27**	Familism X Family Reframing	−.0259	.0100	−2.59**

^*p < .05

^**p < .01

Discussion

In addition to identifying strengths among poor Latino families, this study confirmed that PRS is associated with more symptoms of psychopathology (e.g., Wadsworth et al., 2008). Family reframing appears to be one helpful way that Latino families cope with poverty-related stress and is especially helpful for families with high degrees of familism. However, an interaction between PRS and family coping was not found, suggesting that although family reframing is linked to fewer symptoms across levels of stress, it does not fully buffer the effects of high stress. Family reframing strategies attempt to restructure stressors with active acceptance, define problems in a positive way, and maintain the belief that the family has the strength and power to solve problems. Families that can capitalize on these strategies have children with fewer symptoms of psychopathology. Ample research demonstrates that positive family relationships and functioning are helpful in the face of PRS (e.g., Gorman-Smith, Henry, & Tolan, 2004). However, PRS undermines effective parenting and disrupts family relationships (e.g., Conger et al., 2002). Family reframing may be an effective way of managing PRS that helps maintain positive family functioning. In this study, family reframing was most helpful among families with a sense value, pride, and closeness in the family (familism), with high levels of both familism and family reframing associated with the fewest symptoms. Familism increases parental monitoring and involvement with adolescents (Romero & Ruiz, 2007) and appears to encourage effective family coping. Among families that place high value on family, family reframing strategies are especially effective.

Although familism and family reframing were linked to positive child functioning, family passive appraisal and family social support were linked to more symptoms. The belief that the family has little control in managing PRS likely contributes to feelings of demoralization, hopelessness, and anxiety. Contrary to hypotheses, family social support was related to more externalizing symptoms. Family social support includes sharing difficulties with and seeking encouragement, advice, or help from relatives. Social support is often helpful, but people who are more distressed may also seek more support. For example, seeking support when coping with family conflict is often associated with distress or depression among children (e.g. Sandler, Tein, & West, 1994).

This study has limitations that are important to note. Parent and child reports were not associated across all the constructs included in this study, calling for additional research to replicate findings. Though sufficiently powered to detect medium or large effects, the relatively small sample size limits our ability to detect small effects. Also, reliabilities for the Passive Appraisal scale were relatively low, limiting interpretations of findings involving this scale. Longitudinal data are needed to understand the effects of family coping over time and limit our ability to infer directionality of influences.

Implications for Research, Policy, and Practice

Despite limitations, this study has important implications for intervention and prevention with low-income Latino children and families. This study highlights the importance of understanding cultural and family factors in risk and resilience processes for families dealing with PRS. Family-based interventions may simultaneously increase positive family interactions and support family coping, both of which protect against child psychopathology.