Cultural Socialization and School Readiness of African American and Latino Preschoolers

Abstract

Cultural socialization practices are common among ethnic minority parents and important for ethnic minority child development. However, little research has examined these practices among parents of very young children. In this study, we report on cultural socialization practices among a sample of parents of low income, African American (n = 179) and Latino (n = 220) preschool-age children in relation to children’s school readiness. Cultural socialization was assessed when children were 2½ years old, and child outcomes assessed one year later included pre-academic skills, receptive language, and child behavior. Children who experienced more frequent cultural socialization displayed greater pre-academic skills, better receptive language, and fewer behavior problems. This association did not differ by child gender or ethnicity. The implications of these findings for the development of parent interventions to support school readiness are discussed.

There are persistent ethnic disparities in academic achievement with ethnic minority children more likely to experience early academic failure (Lee & Burkham, 2002). Data from the Early Childhood Longitudinal Survey indicate that African American and Latino kindergarteners are more likely to enter kindergarten in the lowest quartile of academic readiness in both reading and math than any other ethnic group except American Indians/Alaska Natives (West, Denton, & Germino-Hausken, 2000). These differences have many long-term implications, including differential rates of high school graduation. According to national statistics, in 2011, the high school drop-out rate was 7% for non-LatinoBlacks and 14% for Latinos, compared to 5% for non-LatinoWhites (U.S. Department of Education, 2013). Ethnic disparities in academic achievement are due in large part to a higher prevalence of risk factors affecting ethnic minority families such as poverty, single parent households, and low parental education.

Parents of color face unique challenges in raising healthy children not only due to the higher prevalence of risks but also due to unique contextual factors resulting from a history of oppression and current experiences of racism and discrimination (Harrison, Wilson, Pine, Chan, & Buriel, 1990; Pachter & Garcia Coll, 2009). It has been contended that socialization practices of ethnic minority parents are shaped by these contextual factors in ways that are distinct from majority parents (Garcia Coll et al., 1996; Harrison et al., 1990; Ogbu, 1981). One such set of parenting practices is ethnic-racial socialization, or socialization practices transmitting information regarding race and ethnicity to children (Hughes et al., 2006). Evidence indicates that ethnic-racial socialization is associated with a range of positive outcomes for ethnic minority children, primarily in the domains of ethnic identity development and other psychosocial outcomes (Bennett, 2006; Burt, Simons, & Gibbons, 2012; Caughy, O’Campo, Randolph, & Nickerson, 2002; Hughes, Hagelskamp, Way, & Foust, 2009; Neblett, Banks, Cooper, & Smalls-Glover, 2013; Rodriguez, Umaña-Taylor, Smith, & Johnson, 2009; Umaña-Taylor & Guimond, 2010). Hughes et al. (2006) delineated four types of ethnic-racial socialization: cultural socialization, preparation for bias, promotion of mistrust, and egalitarianism/silence about race. In this paper, we focus specifically on cultural socialization, which includes practices aimed at teaching children about and instilling pride in their cultural heritage. Cultural socialization is by far the most common ethnic-racial socialization practice (Brown, Tanner-Smith, Lesane-Brown, & Ezell, 2007; Thornton, 1997) and the one most consistently associated with positive outcomes for ethnic minority children (Hughes et al., 2006).

Specifically, we examine whether cultural socialization during early childhood is associated with better cognitive, language, and behavioral indices of school readiness among ethnic minority preschoolers. Although much of the cultural socialization research has focused on ethnic identity development of youth, there is an emerging literature suggesting cultural socialization is also associated with better academic engagement and achievement during adolescence (Huynh & Fuligni, 2008; Smalls, 2010; Wang & Huguley, 2012) as well as better cognitive, behavioral and academic outcomes during preschool and early elementary school (Brown, Tanner-Smith, & Lesane-Brown, 2009; Caughy, Nettles, O’Campo, & Lohrfink, 2006; Caughy et al., 2002). Theory regarding how cultural socialization may contribute to child competence is an area of research that is particularly underdeveloped. Hughes et al. (2006) theorized that cultural socialization was related to better behavioral and cognitive outcomes because of improved child self-esteem. Similarly, we previously theorized that culturally-anchored socialization contributes to better cognitive development for children of color through dual mechanisms of improved self-esteem and increased opportunities for parent-child engagement and cognitive stimulation (Caughy et al., 2002). Cultural socialization as a means of increasing opportunities for cognitive stimulation is consistent with the literature demonstrating that more cognitive stimulation both inside and outside the home is associated with better cognitive outcomes for children (Bradley et al., 1989; Votruba-Drzal, 2006). Cognitive stimulation associated with cultural socialization could translate into better language development and pre-academic skills, with greater opportunities for parent engagement and improved self-esteem associated with greater behavioral competence.

An important gap in the literature is data regarding ethnic-racial socialization among parents with very young children. Of the 45 papers reviewed by Hughes et al. (2006), only five focused on children under the age of 6 years (Branch & Newcombe, 1986; Caughy et al., 2002; Hughes & Chen, 1997; McKay, Atkins, Hawkins, Brown, & Lynn, 2003; Spencer, 1983). Of those, only one (Caughy et al., 2002) focused exclusively on children under the age of 6 years. All five studies included only African American families. A search for studies published since the Hughes et al. review identified only three papers examining cultural socialization of children under the age of 6, and all relied on the same data set (the ECLS-K), which measured ethnic-racial socialization with a single question (Brown & Lesane-Brown, 2006; Brown et al., 2009; Lesane-Brown, Brown, Tanner-Smith, & Bruce, 2010).

Findings reported from our previous work in Baltimore (Caughy et al., 2002, 2006) and by Brown et al. (2009) suggest the beneficial effects of cultural socialization for cognitive and behavioral development are evident as early as preschool. If so, cultural socialization may be an important target for supporting school readiness among ethnic minority children. However, evidence of the positive effects of cultural socialization during early childhood is limited to these three reports and is also limited to studies of young African American children. Whether cultural socialization is associated with better cognitive and behavioral outcomes in other ethnic groups is an open question. The purpose of this investigation was to examine cultural socialization in a sample of low income African American and Latino caregivers in relation to the school readiness of their young, preschool-aged children. First, because of the dearth of available data on cultural socialization practices with children in the early preschool years, we examine the frequency of cultural socialization and whether it differed by child and household characteristics. In terms of relations with school readiness, we hypothesized that higher levels of cultural socialization would be associated with better cognitive, language, and behavioral school readiness. In addition, we examined whether associations between cultural socialization practices and child outcomes were moderated by child gender and/or ethnicity. Although there are no explicit reasons to anticipate relations with cultural socialization would differ by ethnicity, given limited evidence of gender differences (Caughy et al., 2006), it is important to examine this question because of the paucity of research on ethnic-racial socialization and child outcomes among young Latino children.

METHODS

Participants

Participants were drawn from a larger study on self-regulation skills and school readiness among low income, ethnic minority preschoolers conducted in a large urban area in the southwestern United States. Families were recruited through a variety of community-based recruitment efforts. Children were between 29 and 31 months of age, the age at which self-regulation skills, the primary focus of the larger study, begin to demonstrate dramatic development. Children had at least one parent who was either African American or Latino, a family income at or below 200% of the federal poverty level, and had not been hospitalized at birth for more than 7 days.

A total of 407 families were enrolled and completed the initial home visit. Primary caregivers reported their ethnicity and the ethnicity of their child during this visit. Respondents were first asked if they were Hispanic and then asked which racial group(s) they identified with: Black/African American, White/European American, Asian, American Indian, or Other. Caregivers could check as many racial groups as they liked. Similar questions were posed regarding the child’s ethnicity. We excluded 8primary caregivers who were white, non-Hispanic. Of the remaining 399 caregivers, 198 (49.5%) were Latino, 181 (45.2%) were Black/non-Latino, and 20 (5%) were multiracial/multiethnic. Of the children in the sample, 185 (46.5%) were non-Black Hispanic, 170 (42.5%) were non-Hispanic Black, and 44 (11%) were multiracial/multiethnic. Multi-ethnic children who were Hispanic were classified with the Hispanic group.

Demographic characteristics of the families are displayed in Table 1. The majority of caregivers were mothers, although African Americans were more likely than Latinos to have a non-maternal primary caregiver. African American families were also less likely to include two parents and more likely to have a household income below 50% of the federal poverty level. Latino caregivers were less likely to have completed high school. In addition, Latino caregivers were primarily foreign-born (74.2%), and most foreign-born caregivers were from Mexico (95.7%). Maternal language proficiency was determined using two questions how well they spoke English and Spanish on a four point scale (1 = very poorly, 2 = poorly, 3 = well, 4 = very well) (Marin & Gamba, 1996). Caregivers who reported they spoke the language well or very well were considered proficient in that language. Most caregivers (62%) were Spanish-dominant, while a third were bilingual. The average age of children at the first home visit was 29.78 months (sd = .62, range 28 – 31 months), and the average age of primary caregivers was 29.25 years (sd = 7.65, range 17 – 64 years). A second home visit was completed one year later, when the child was 3½ years old. The follow-up rate was slightly higher for Latino (93%, n = ) compared to African American children (86%, n = ), χ² (2) = 7.17, p< .05. The average age of children at this visit was 41.57 months (sd = 1.24, range 38 – 47 months).

Table 1.

Characteristic of study sample (N = 399).

	Child Ethnicity				χ2
	African American (N = 179)		Latino (N = 220)
	N	%	N	%
Child gender
Boy	98	54.7	116	52.7	.16
Girl	81	45.3	104	47.3
Child race/ethnicity
African American, non-Latino	170	95.0	0	.0	370.06***
Latino	0	.0	185	84.1
Multiracial/multiethnic	9	5.0	35	15.9
Primary caregiver race/ethnicity
African American, non-Latino	173	96.6	8	3.6	351.11***
Latino	0	.0	198	90.0
Multiracial/multiethnic	6	3.4	14	6.4
Primary caregiver relationship to child
Mother	158	88.3	214	97.3	15.38***
Father	10	.6	3	1.4
Grandmother/Aunt	11	6.1	3	1.4
Family structure
Nuclear	37	21.0	132	60.0	149.17***
Single parent	45	25.6	9	4.1
Nuclear/extended	4	2.3	49	22.3
Single parent/extended	38	21.6	22	10.0
Single parent plus partner	30	17.0	4	1.8
Other	22	12.5	4	1.8
Primary caregiver’s education
Less than high school	26	14.6	95	43.4	41.41***
High school	81	45.5	79	36.1
More than high school	71	39.9	45	20.9
Average family income
Less than 50% federal poverty level	95	55.2	33	15.1	74.93***
50–99% federal poverty level	38	22.1	116	53.0
100–149% federal poverty level	22	12.8	47	21.5
150+% federal poverty level	17	9.9	23	10.5
Neighborhood ethnic composition
African American	64	39.6	12	5.5	84.83***
Hispanic	34	19.4	125	56.8
Non-Hispanic White	8	4.6	13	5.9
Ethnically mixed	69	39.4	70	31.8
Latino families only
Caregiver language proficiency
English			10	4.9
Spanish			128	62.1
Bilingual			68	33.0
Nativity (caregiver)
U.S. born			56	25.5
Foreign born			164	74.5
Country of origin (foreign-born caregivers)
Mexico			157	95.7
Central America			6	3.6
South America			1	.1

^**p < .01;

^***p < .001

Procedures

All interviews were conducted by an individual of the same ethnicity as the respondent, and Spanish-speaking caregivers were interviewed in Spanish. Survey items were translated into Spanish by a native speaker on the research team and then back-translated into English by a non-native speaker. When assessing Spanish-speaking and bilingual children, a fully bilingual child assessor maintained flexibility in administering assessment items in the alternate language in situations in which the child could not answer the question when administered in their primary language.

Measures

Cultural socialization

Cultural socialization was assessed at the first home visit using the five- item Cultural socialization subscale of the Parents’ Messages to Children About Race scale (Hughes, 2003; Hughes & Chen, 1997). Example items include “celebrated cultural holidays of your ethnic/racial group” and “talked to your child about important people or events in the history of your ethnic group”. For each item, the parent was first asked “Have you ever…?” and if yes, asked to report the frequency in the last 12 months (Never, Once, 2–3 times, 4–5 times, 6–7 times, 8+ times). Each item was recoded into a 6 point Likert-scale ranging from 0 (Never at all) to 6 (8+ times in the last year). The internal reliability of the scale was .72and did not differ significantly between African Americans (.73) and Latinos (.70).

School readiness

School readiness was assessed at age 3½ years using the Bracken School Readiness Subscale (SRS) of the Bracken Basic Concept Scale-Revised (BBCS-R), the Peabody Picture Vocabulary Test (PPVT), and the Child Behavior Checklist (CBCL). The Bracken SRS assesses knowledge of colors, letters, numbers, and shapes and includes a Spanish adaptation (Bracken, 1998). The PPVT (Dunn & Dunn, 1981) was used to assess receptive language skills. For Spanish-speaking children, the Spanish equivalent of the PPVT, the Test de Vocabularioen Imagenes Peabody (TVIP, (Dunn, Padilla, Lugo, & Dunn, 1986)), was used. The CBCL (Achenbach, 1991) is a measure of child behavioral competence based on parental report. Higher scores on the CBCL indicate more behavioral problems. The validity and reliability of the Bracken, TVIP, and CBCL with Latino children has been demonstrated (Bracken et al., 1990; Dunn et al., 1986; Rubio-Stipec, Bird, Canino, & Gould, 1990).

Caregiver sensitivity

Characteristics of caregiver behavior were assessed at Time 1 during an interaction modeled after the NICHD Study of Early Child Care and Youth Development (SECCYD)(NICHD ECCRN, 1999). Interactions were independently rated using five-point versions of the SECCYD global rating scales (1 = low; 5 = high) to measure parenting qualities of sensitivity, intrusiveness, detachment, cognitive stimulation, positive regard, and negative regard (Owen et al., 2010). The team coding videos in Spanish included three bilingual raters. To ensure reliability between teams, both groups met together weekly to code an English-speaking dyad. Interrater reliability was calculated using an intraclass correlation coefficient (ICC) (Shrout & Fleiss, 1979) based on double coding 28% of the cases and ranged from .79 to .85 across the scales. A confirmatory factor model indicated a single factor for five of the parent behavior measures (sensitivity, cognitive stimulation, positive regard, negative regard, and detachment) fit the data best, χ² (2) = 1.58, p = .45, CFI = 1.0, TLI = 1.0, RMSEA = .00, and this model fit both ethnic groups equally well. A sensitivity composite was created by summing these five indicators (with negative regard and detachment inverted).

Caregiver, household, and neighborhood characteristics

Household characteristics included caregiver education, number of children in the household older than the target child, family income-to needs ratio, and neighborhood ethnic composition. Family income-to-needs ratio at each time point was calculated by dividing family income by the federal poverty level for a family of that size and averaged across time points. Neighborhood ethnic composition data were obtained from the 2010 census for the families’ zip code and classified as predominantly African American (average percentage of African American residents 72%, range 52–89%), Hispanic (average percentage of Hispanic residents 64%, range 43–82%), non-Hispanic White (average percentage of non-Hispanic White residents 63%, range 51–74%) or ethnically mixed (30+% of two or more ethnic groups or 20+% of three or more ethnic groups). The distribution of families by neighborhood type is displayed in Table 1. A higher proportion of Hispanic participants (57%) lived in predominantly Hispanic neighborhoods compared to the proportion of African American participants living in predominantly African American neighborhoods (40%). African American participants were more likely to live in either Hispanic or non-Hispanic White neighborhoods compared to the proportion of Hispanic participants living in African American or non-Hispanic White neighborhoods (24% vs. 11%).

Results

The average cultural socialization score was 1.16 (sd = 1.24), indicating most parents engaged in cultural socialization practices about once in the last year. Approximately 30% of caregivers reported never engaging in cultural socialization. Differences in cultural socialization practices by child and household characteristics are displayed in Table 2 for the whole sample and stratified by ethnicity. Cultural socialization was more prevalent among caregivers of boys, African American caregivers, and families living at or above 100% poverty. Although there were no overall differences in cultural socialization by neighborhood ethnicity, there appeared to be differences among African Americans specifically. Combining African Americans living in Hispanic and in non-Hispanic White neighborhoods, caregivers living in these neighborhoods used cultural socialization practices more frequently than African American caregivers living in African American or ethnicity mixed neighborhoods, F (2, 170) = 3.38, p< .05.

Table 2.

Demographic differences in cultural socialization practices

	Full Sample		African American		Latino
	M (SD)	F	M (SD)	F	M SD)	F
Child Gender
Boy	1.28 (1.34)	4.10*	1.46 (1.49)	2.38	1.13 (1.19)	1.57
Girl	1.03 (1.11)		1.15 (1.19)		.94 (1.04)
Primary caregiver’s race/ethnicity
African American	1.36a (1.37)	4.27*	1.35 (1.38)	2.08	1.65 (1.20)	.17
Latino	.99a (1.10)		---		.99 (1.10)
Multiracial/multiethnic	1.08 (1.17)		.53 (.47)		1.31 (1.32)
Primary caregiver’s education
Less than high school	.94b (1.09)	9.83***	1.40 (1.35)	3.17*	.82e (.98)	6.88**
High school	1.04c (1.14)		1.05d (1.19)		1.02f (1.10)
More than high school	1.59b,c (1.43)		1.60d (1.51)		1.57e,f (1.33)
Family poverty status
<100% federal poverty level	1.08 (1.17)	4.56*	1.24 (1.26)	2.05	1.26 (1.23)	3.78+
100%+ federal poverty level	1.38 (1.41)		1.59 (1.67)		.94 (1.07)
Older children in the household
None	1.17 (1.32)	.22	1.29 (1.50)	.04	1.04 (1.08)	.83
1–2	1.13 (1.18)		1.32 (1.17)		1.00 (1.18)
3 or more	1.24 (1.31)		1.37 (1.56)		1.12 (1.05)
Neighborhood ethnic composition
African American	1.26 (1.32)	.62	1.20 (1.28)	2.25+	1.53 (1.52)	.84
Hispanic	1.18 (1.30)		1.77 (1.72)		1.02 (1.12)
Non-Hispanic White	1.32 (1.29)		1.88 (1.46)		.98 (1.10)
Ethnically mixed	1.05 (1.12)		1.13 (1.19)		.98 (1.06)
Caregiver language proficiency	---	---	---	---
English	---	---	---	---	1.20 (1.47)	1.27
Spanish	---	---	---	---	.89 (1.02)
Bilingual	---	---	---	---	1.13 (1.16)

Note. For variables with more than two groups, post-hoc comparisons using a Bonferonni correction were conducted to determine the pairwise differences. Means that share the same subscript are significantly different from one another.

⁺p < .10;

^*p < .05;

^**p< .01;

^***p < .001

Intercorrelations among the study variables are displayed in Table 3. Cultural socialization practices at age 2½ were associated with higher Bracken SRS and PPVT/TVIP scores and lower CBCL Total Problem behavior scores at age 3½. These associations appeared to be stronger for African American children relative to Latinos.

Table 3.

Intercorrelations of study variables

	Full Sample
	1	2	3	4	5	6	7
1 Cultural socialization	1.00
2 Bracken SRS	.171**	1.00
3 PPVT/TVIP	.136*	.448**	1.00
4 CBCL Total problem behaviors	−.134*	−.154**	−.094+	1.00
5 Caregiver sensitivity composite	.015	.188**	.342**	−.127*	1.00
6 Caregiver education (1 = < HS)	−.118*	−.264**	−.017	.126*	.010	1.00
7 Avg family income-to-needs ratio	.070	.296**	.235**	−.143**	.204**	−.155**	1.00

	African American Children
	1	2	3	4	5	6	7
1 Cultural socialization	1.00
2 Bracken SRS	.190*	1.00
3 PPVT/TVIP	.194*	.593**	1.00
4 CBCL Total problem behaviors	−.190*	−.060	−.012	1.00
5 Caregiver sensitivity composite	.085	.235**	.298**	−.150	1.00
6 Caregiver education (1 = < HS)	.035	−.311**	−.188*	.104	−.136	1.00
7 Avg family income-to-needs ratio	.098	.314**	.245**	−.211**	.096	−.228**	1.00

	Latino Children
	1	2	3	4	5	6	7
1 Cultural socialization	1.00
2 Bracken SRS	.134+	1.00
3 PPVT/TVIP	.147*	.364**	1.00
4 CBCL Total problem behaviors	−.056	−.252**	−.206**	1.00
5 Caregiver sensitivity composite	.021	.191**	.282**	−.178**	1.00
6 Caregiver education (1 = < HS)	−.164*	−.265**	−.034	.110	−.085	1.00
7 Avg family income-to-needs ratio	.089	.308**	.142*	−.109	.225**	−.260**	1.00

^*p < .05;

^**p < .01

A multivariate model for the whole sample as well as stratified by child ethnicity was fit to estimate the relation between cultural socialization and school readiness after adjusting for family demographic characteristics and caregiver sensitivity (Table 4). Continuous variables were centered at the mean. For the sample as a whole, cultural socialization at age 2½ was associated with better Bracken and PPVT/TVIP performance and fewer behavior problems at age 3½ even after adjusting for potential confounders. The effect sizes were modest, ranging from .11to .12 based on a one standard deviation increase in cultural socialization practices. When examined separately by child ethnicity, cultural socialization practices were associated with better Bracken performance and fewer behavior problems for African Americans but with greater receptive language among Latinos. However, comparisons across groups indicated that the coefficients associated with cultural socialization did not differ significantly by child ethnicity.

Table 4.

Multivariable linear regression of school readiness outcomes at age 3½ on racial socialization practices at age 2½

	Full Sample
	Bracken SRS		PPVT/TVIP		CBCL
	b (se)	t	b (se)	t	b (se)	t
Constant	85.86 (.90)	95.11***	86.45 (.91)	95.29***	47.54 (.72)	65.88***
Caregiver education (1 = < HS)	−6.03 (1.53)	−3.94***	1.38 (1.54)	.90	2.34 (1.22)	1.91+
Family income-to-needs ratio	5.76 (1.29)	4.45***	4.24 (1.30)	3.26**	−1.88 (1.03)	−1.81+
Neighborhood ethnicitya	3.95 (1.87)	2.11*	1.61 (1.88)	.86	−.44 (1.50)	−.29
Caregiver sensitivity composite	.60 (.20)	3.00**	1.18 (.20)	5.87***	−.33 (.16)	−2.04*
Cultural socialization practices	1.24(.56)	2.20*	1.29 (.57)	2.27*	−.94 (.45)	−2.08*

	African Americans
	Bracken SRS		PPVT/TVIP		CBCL
	b (se)	t	b (se)	t	b (se)	t
Constant	87.80 (1.60)	55.02***	83.99 (1.41)	59.44***	46.25 (1.13)	40.90***
Caregiver education (1 = < HS)	−10.25 (3.63)	−2.83**	−4.10 (3.21)	−1.28	.91 (2.57)	.36
Family income-to-needs ratio	5.94 (2.01)	2.95**	4.02 (1.78)	2.26*	−3.06 (1.42)	−2.15*
Neighborhood ethnicitya	3.16 (2.96)	1.07	5.30 (2.62)	2.03*	−.27 (2.09)	−.13
Caregiver sensitivity composite	.74 (.34)	2.22*	.94 (.30)	3.18**	−.34 (.24)	−1.43
Cultural socialization practices	1.68(.95)	1.80+	1.28 (.83)	1.54	−1.30 (.66)	−1.96+

	Latinos
	Bracken SRS		PPVT/TVIP		CBCL
	b (se)	t	b (se)	t	b (se)	t
Constant	84.04 (1.17)	71.99***	88.13 (1.30)	67.95***	48.87 (1.05)	46.42***
Caregiver education (1 = < HS)	−3.50 (1.67)	−2.09*	1.56 (1.86)	.84	2.03 (1.51)	1.35
Family income-to-needs ratio	6.04 (1.84)	3.28**	2.37 (2.05)	1.16	−1.00 (1.66)	−.60
Neighborhood ethnicitya	3.96 (2.51)	1.58	−1.58 (2.79)	−.57	.13 (2.26)	.06
Caregiver sensitivity composite	.55 (.28)	1.97+	1.12 (.31)	3.62***	−.56 (.25)	−2.22*
Cultural socialization practices	.81 (.72)	1.13	1.65 (.80)	2.07*	−.29 (.64)	−.45

^aCoded 1 for African American children living in predominantly Hispanic or non-Hispanic white neighborhoods and for Hispanic children living in predominantly African American or non-Hispanic white neighborhoods.

⁺p < .10;

^*p < .05;

^**p < .01;

^***p < .001

Possible effect moderation by child gender or neighborhood ethnicity was examined by entering product variables for these factors by cultural socialization into the models for each of the school readiness outcomes. None of the interactions by gender or by neighborhood ethnicity were significant.

As a sensitivity test, we re-fit the multivariate models after excluding any children (n = 44) and caregivers (n = 20) who were multiracial/multiethnic. Results for the Bracken and PPVT/TVIP were unchanged. For the CBCL, after excluding multiracial/multiethnic children and caregivers, the positive association between the cultural socialization and CBCL scores was somewhat attenuated, b = −.80, se(b) = .48, t = −1.68, p< .10.

Discussion

In this study, we examined the prevalence of cultural socialization among low income parents of African American and Latino preschool-age children. Although parents reported cultural socialization practices, they were in frequent, about once in the last year on average, with approximately 30% of caregivers reporting never engaging in cultural socialization. These results contrast sharply with those of our research in Baltimore in which we reported cultural socialization among 90% of African American parents of 3–4 year old children (Caughy et al., 2002). Hughes and Chen (1997) used the same measure as our study and reported an average cultural socialization score for parents of 4–5 year olds of 2.5, reflecting engagement in cultural socialization between 1–3 times in the last year. This compares to an average of 1.2 in our sample, corresponding to once in the last year or less.

These differences in prevalence may be a function of methodological differences and/or true differences in study populations. Over half of oursample was Latino, whereas Caughy et al. (2002) and Hughes and Chen (1997) included only African Americans. However, cultural socialization was similarly infrequent among the African Americans in our sample. In addition, our study in Baltimore (Caughy et al., 2002) included an economically diverse sample whereas the present study focused on low income families. Hughes and Chen limited their sample to two-parent families whereas less than 60% of the children in the present sample were living in two parent households. In addition, the measures of cultural socialization differed. Our Baltimore study used a measure developed by Stevenson (1999) for use with African Americans in which parents report the frequency with which they convey a set of attitudes to their children. In contrast, the measure used in the present investigation (Hughes, 2003; Hughes & Chen, 1997) measures the frequency of specific cultural socialization practices. Although examining both parental attitudes and behaviors have long traditions in socialization research, such differentiation has not been seen in the ethnic-racial socialization literature. However, the lower rate of cultural socialization in our study sample compared to that reported by Hughes et al. (1997) is more difficult to explain given the same measure was used. It may be that regional differences partly explain the different rates of cultural socialization across these two studies. For example, the location of the current study (Dallas) is significantly less segregated for both African Americans and Latinos relative to the location of the Hughes and Chen (1997) study (Mather, Pollard, & Jacobsen, 2011).

We found parents of boys more likely to report cultural socialization practices. Although many researchers have failed to find gender differences in cultural socialization (Caughy et al., 2002; Frabutt, Walker, & MacKinnon-Lewis, 2004; Hughes & Chen, 1997; Phinney & Chavira, 1995; Scott, 2003; Stevenson, Reed, & Bodison, 1996), some report cultural pride socialization is more common with girls (Bowman & Howard, 1985; Thomas & Speight, 1999). However, these studies were all with exclusively African American samples and, with one exception (Caughy et al., 2002), focused on older children and adolescents. Gender differences in cultural socialization may depend on developmental stage. Cultural socialization may be more frequent with boys when they are younger. For example, anecdotally, raising African American boys in a society prone to interpret their behaviors as threatening poses addition demands on parents (Curtis, 2012) which may explain why cultural socialization of boys is initiated earlier. However, gender differences in practices may change as children grow older.

We also found that, for African American caregivers, the frequency of cultural socialization was higher among families living in predominantly white, non-Hispanic or predominantly Hispanic neighborhoods. This difference was not confounded by family demographics, as African American caregivers living in white or Hispanic neighborhoods did not differ from other African American caregivers in terms of education or household income. Perhaps African American parents raising children in neighborhoods with few African American residents perceive a greater need to actively engage in cultural socialization because their children will not be exposed to it in the neighborhood. Only one other study, another of our Baltimore studies, examined neighborhood factors in relation to ethnic-racial socialization practices, but its examination did not include neighborhood ethnicity (Caughy et al., 2006). However, its location (Baltimore) was more segregated and less ethnically diverse than the location of the current study, evidenced by the fact that only 2% of participants in the Baltimore study lived in neighborhoods predominated by another ethnic group compared to almost a quarter of the African American participants in the current study. Although individual studies may not be of sufficient size, it would be useful to combine data across studies, link data to neighborhood context data from the census, and then systematically examine how ethnic-socialization practices vary across a range of geographic contexts.

Another aim of this study was to examine the association between cultural socialization and children’s school readiness. Here we found more frequent cultural socialization at this young age (2½ years old) was associated one year later with higher scores in pre-academic and receptive language skills and fewer behavior problems. These associations held after adjusting for the potentially confounding influence of sensitive, cognitively stimulating caregiving. Furthermore, we did not find these associations differed by child gender or ethnicity. Our findings echo Brown et al.’s (2009) findings from the ECLS-K and provide further evidence that such practices are important contributors to the school readiness of children of color. In addition, our findings extend those of Brown et al. (2009) by finding similar associations for Latino children.

However, it should be noted that effect sizes for these relations were modest (~.11). Although encouraging cultural socialization as part of parent education programs to improve school readiness may be beneficial, this emphasis is unlikely to have significant impact on its own. However, the benefits of cultural socialization may accrue and be more evident during later childhood/adolescence. For example, it has been reported the effect sizes of cultural socialization for youth ethnic identity are significantly greater (.50–.60), and the effect sizes for the relation between ethnic identity and academic achievement in adolescence have been described as medium effect sizes (Hughes, Witherspoon, Rivas-Drake, & West-Bey, 2009; Rivas-Drake, Hughes, & Way, 2009; Smith, Levine, Smith, Dumas, & Prinz, 2009). It is also possible the very low frequency of involvement in cultural socialization practices with these toddlers may account for the small effect sizes.

This study is the first to examine cultural socialization practices among Latino parents of preschoolers. Latino parents reported lower rates of cultural socialization compared to African Americans. This may signify different values placed upon cultural socialization between the two ethnicities but may also be due to ethnic differences in attributions of the age-appropriateness of such practices with young preschoolers. In addition, although stratified analyses suggested associations between cultural socialization practices and school readiness outcomes were weaker for Latino children, statistical comparisons across groups did not identify significant ethnic differences. Hughes and colleagues (2003) also reported higher rates of cultural socialization among African American compared to Latino parents, although the Latino groups in question were Puerto Rican and Dominican. In contrast, Phinney and Chavira (1995) did not find differences in frequency of cultural socialization between African American and Mexican American parents of adolescents. The Latino parents in the present study were largely Mexican-origin. The study sample for the Phinney and Chavira study was significantly smaller than ours, so lack of statistical power could explain these differences. It is also possible that African American parents initiate cultural socialization at an earlier age such that these ethnic differences may disappear as children age.

There are limitations of this investigation to keep in mind when interpreting the findings. The low income status of the study sample limits generalizability of findings to families with similarly few economic resources. It would be preferable to have longitudinal data from an economically diverse sample of ethnic minority parents to more fully explore how these practices relate to differences in family socioeconomic status and vary by child age. In addition, these findings cannot be generalized to Latino families who are not of Mexican-origin; as such, more research is needed regarding the impact of ethnic-racial socialization practices for young Latino children of other countries of origin.

A challenge of conducting research on ethnic-racial socialization is capturing the ethnic diversity of the families from whom data are collected. Although we excluded primary caregivers who were neither African American nor Hispanic, we chose to retain those who self-identified as multiethnic. One could argue (rightly so) that the ethnic-racial socialization practices of multiethnic caregivers may be substantially different from caregivers who are not multiethnic. Likewise, socialization practices that best support the development of a multiethnic child may differ in meaningful ways from those for children who are not multiethnic. However, excluding these caregivers from the analysis did not substantially change the results. The number of multiethnic caregivers and children in our sample was too small and varied to examine subgroup differences. However, given the increasing diversity of our population as well as the increasing rate of interethnic marriages (Lofquist, Lugaila, O’Connell, & Feliz, 2012), the lines between ethnic groups will become increasingly blurred in the future. Research on the contribution of ethnic-racial socialization to children’s development will have to evolve accordingly.

A number of researchers are calling for better integration of ethnic-socialization in theoretical models of positive development among ethnic-minority youth (Evans et al., 2012; Neblett, Rivas-Drake, & Umaña-Taylor, 2012). Evans et al. (2012) proposes that cultural socialization fosters elements of positive youth development including competence, confidence, character, connection, and caring. Longitudinal research across the span of development is needed to describe how cultural socialization practices evolve as children grow and, in turn, are associated with outcomes in multiple domains of positive youth development. Ours is the only study using a longitudinal design with a sample of children this young and demonstrating that higher rates of cultural association are associated with better child outcomes one year later. In addition, this is the only study we are aware of that has examined these processes for young Latino children. We are continuing to follow our sample into elementary school which will provide us with a unique ability to examine how cultural socialization practices during early childhood contribute to competencies in multiple domains for children of color as they develop.