A large body of literature demonstrated White-Hispanic disparity in math skills of children in their yearly years(Reardon and Galindo 2009). Research argued that White-Hispanic gaps in math skills were relatively greater at the beginning of school years and the gaps gradually narrow down as they progress through their elementary school years. However, the gaps vary depending on ethnic identity and parents’ immigration status and their use of English language at home. Lowest level of math skills was found at the entry level of kindergarten among Mexican and Central American origins - particularly first- and second-generation immigrants - and those from homes where English is not spoken(Reardon and Galindo 2009). Many other studies also demonstrated children’s language proficiency in English and home language work as a buffer in reducing the gap of math achievement among kindergarden children(Halle et al. 2012).
A different stream of literature focused on the participation in child care settings as reducing the gaps in math achievement. Several research suggested the impact of center-based care on Latino children’s language skills(Ansari and Winsler 2012; Magnuson et al. 2004; Morrissey and Vinopal 2018). Reserarch argued that both language skills and school readiness skills of Latino children were greater when they placed in centre-based childcare (CBC) compared to their placement in the family child care (FCC)(Ansari and Winsler 2012). The math skills of kindergarten and first grade children were higher among those who attended in child care center in their early lives than those who did not(Magnuson et al. 2004).
Therefore, previous studies demonstrated the individual impact of center-based care and use of home language and their association with math skills of children in early school years. Still there is a knowledge gap on how center based care and home language influence each other to understand White-Hispanic disparity in math skills. In other words, whether or not the effect of center-based care increases the likelihood of speaking English at home which in turn affect White-Hispanic disparity in math skills.
Do White and Hispanic kindergarten children differ in terms of home language , center-based care, and math skills?
Do White and Hispanic kindergarten children differ in terms of the association between home language and math skills?
Do White and Hispanic kindergarten children differ in terms of the association between center-based care and math skills?
Whether or not home language mediate the association between center-based care and math skills of White and Hispanic children.
Whether or not Center-based care predict White and Hispanic children’s use of language at home.
The public use dataset on the Early Childhood Longitudinal Study‐Kindergarten Class of 1998-99 (ECLS‐K) was used in the present study. This dataset was released by U.S. Department of Education, National Center for Education Statistics, 1999). In total 21,204 children were included in ECLS-K study who came from both private and public schools attending either full-day or part day in the kindergarten program in the fall and spring (1998-99) and they were followed through 8th grade (2007). The present study only focused on kindergarten year (fall and spring, 1998-99). Only White and Hispanic racial groups were included in this study. Multiracial children, children with disabilities and children who did not meet the criteria for math skills assessment were excluded in this study. This included 11,114 children initially. After selecting the variables and covariates of interest, children having complete sets of information were included. This excluded 21% (n≈2352) of the initial sample. Thus, 8786 children were included in this study.
The information in ELCS-K (1998-99)study was collected from kindergarteners, their parents, teachers and schools all across the United States. Children’s assessment was done by trained evaluators at school and information was collected from parents over telephone. Teachers and school administrator were also asked to complete questionnaires.
Race/ethnicity Child race/ethnicity (Hispanic and White) was the part of composite variables. Hispanic children included both categories whose ethnicity were known or unknown. Both White and Hispanic children with multi-racial identities were excluded.
Home language Children were asked whether they speak Non-English at home or not.
Center-based care The parents were asked whether or not their children had ever in center-based care.
Children’s math skills Children’s reading and math skills at kindergarten entry were measured by overall (Item‐response theory) IRT‐scaled scores on math assessments. Children were individually administered each test by an examiner who asked developmentally appropriate questions about count, number, shape, relative size, ordinality/sequences, add/subtract, multiply/divide, place value, rate& measurement, fraction, area, volume, and math standard error of theta.
Covariates Covariates were selected based on theoretical rationale and findings from previous studies. They were used for descriptive analysis as well as controlling subsequent analyses. Specifically, child covariates included child age (in month) during assessment (range: 57–79; median = 68; M = 68, SD = 4.19), gender (49% boys vs. 51% girls), kindergarten (96.6% first-time kindergartener, vs 3.4% second-time kindergartener ), Household covariates included number of siblings in the household (range: 0 to 10; median = 1.00; M = 1.24, SD = 1.08), total number of household members (range: 2–13; median = 4.00; M = 4.50, SD = 1.25), SES(13% First quintile, 17% Second quintile, 20% Third quintile 23% Fourth quintile and 27% Fifth quintile), family types (74.22% Both parents & sibling, 10.27% Both parents & no sibling, 9.63% Single parent & sibling,4.99% Single parent & no sibling and 0.89% Other), total number of books (range: 0–200; median = 75; M = 84, SD = 60.55),. In addition, other psycho-social measures such as :approach to learning(range: 1.2–4.0; median = 3.2; M = 3.1, SD = 0.46), self-control(range: 1.00–4.00; median = 2.80; M = 2.90, SD = 0.48), sad/alone(range: 1.00-3.80; median = 1.50; M = 1.50, SD = 0.37), impulsivity(range: 1.00–4.00; median = 2.00; M = 1.90, SD = 0.62) were also included.
Descriptive analysis was conducted to see the differences of White and Hispanic Children in math skills, home language, and center-based care. Linear regressions were conducted to examine White and Hispanic children’s differences in math skills in terms of home language and center-based attendance. Two different Logistic regressions were conducted. First one was conducted to examine whether center-based care predict home language of White and Hispanic Children. Second one was conducted to predict White and Hispanic Children’s likelihood to obtain center-based care. All of the above regression models were further simulated by using Zelig statistical software of R.
White and Hispanic differences in social interaction skills, home language, center-based care and math skills obtained from descriptive analysis. Both linear and logistic regressions were used to estimate the differences after controlling for covariates. White children with mean age who speak English at home and who attended in center-based care was the reference group in all regression models. The results suggested that White and Hispanic kindergarten children differ significantly in their Math skills, use of home language and attending in center-based care. After controlling for covariates, Hispanic children had significantly lower math scores than White children.The mean difference in math scores of White and Hispanic children was (d = -1.6, sd=0.3).`
White and Hispanic children also differ in terms of use of home language. Hispanic children were less likely to speak English at home than White children. The probabilities of speaking English language at home were 71% and 99% in case of Hispanic and White children respectively. So the difference was 27%.
White and Hispanic disparity was also found in the probabilities of attending in center-based care. The results suggested that Hispanic children (84%) were significantly less likely to attend in center-based care than White Children (88%). The difference was 4.7%.
The results suggested that White and Hispanic disparity in math scores vary depending on their use of Language at home. The mean difference in math score was (d = -1.6, sd=0.31) when both groups of children speak English at home. However, this difference was increased when compared with Hispanic children who speak non-English at home. The mean difference in math score of White children who speak English at home and Hispanic children who speak non-English at home was (d=3.2, sd=0.37).
Similarly, White and Hispanic disparity in math scores vary depending on their attend in center-based care. The mean difference in math score was (d = -1.6, sd=0.32) when both groups of children attended in center-based care.` However, this difference was increased when compared with Hispanic children who did not attend center-based care. The mean difference in math score of White children who attended in center-based care and Hispanic children who did not attend center-based care was (d=2.2, sd=0.4).
In addition, the effect of center-based care on math skills vary depending on the use of language at home especially in case of Hispanic children. Among Hispanic children who attended center-based care, those who used English at home had greater math skills than those who used Non-English at home.
The results from Logistic regression analysis suggested that center-based care increases Hispanic children’s probability of speaking English at home. Hispanic children had 10% greater likelihood of speaking English at home when they attended in center-based care compared to when they did not. White and Hispanic disparity in use of home language was found to be greater if White children were compared with Hispanic children who did not attend in center-based care than if White children were compared with Hispanic children who attend in center-based care. White and Hispanic disparity in use of English as home language was 38% if White children were compared with Hispanic children who did not attend in center-based care. White and Hispanic disparity in use of English as home language was 27% if White children were compared with Hispanic children who attended in center-based care. So, the center-based care accounted for 11%of the variation in use of English Language at home.
The above results suggested that both home language and center-based care accounted for the White-Hispanic disparity in math skills. However, home language was found as a mediator in the association between center-based care and math skill in case of Hispanic children. So, center based care might not be effective in reducing White-Hispanic disparity in math skills if Hispanic children were not proficient or not frequently use English language in their daily lives at home. So, center-based care might be effective in reducing White-Hispanic disparity in math skills only if Hispanic children are proficient and frequent in using and practicing English at home. Therefore, it can be concluded that by using English language at home, Hispanic children become more proficient in speaking, using, and understanding English as White children which helps them to understand the math problem and thereby increase their math scores.
The finding of this study can be used in understanding and reducing the White-Hispanic disparity in math skills. The present study can be replicated by using experimental research can be designed to get more controlled effects of home language and center-based care on White-Hispanic disparity in math skills.
The present study can be replicated by using more complied and complete dataset to avoid list-wise deletion.
List-wise deletion was used in the present study which excluded 21% of the total sample.
Ansari, Arya, and Adam Winsler. 2012. “School Readiness Among Low-Income, Latino Children Attending Family Childcare Versus Centre-Based Care.” Early Child Development and Care 182 (11): 1465–85. https://doi.org/10.1080/03004430.2011.622755.
Halle, Tamara, Elizabeth Hair, Laura Wandner, Michelle McNamara, and Nina Chien. 2012. “Predictors and Outcomes of Early Versus Later English Language Proficiency Among English Language Learners.” Early Childhood Research Quarterly 27 (1): 1–20. https://doi.org/10.1016/j.ecresq.2011.07.004.
Magnuson, Katherine A., Marcia K. Meyers, Christopher J. Ruhm, and Jane Waldfogel. 2004. “Inequality in Preschool Education and School Readiness.” American Educational Research Journal 41 (1): 115–57.
Morrissey, Taryn W., and Katie Vinopal. 2018. “Center-Based Early Care and Education and Children’s School Readiness: Do Impacts Vary by Neighborhood Poverty?” Developmental Psychology 54 (4): 757–71. https://doi.org/10.1037/dev0000470.
Reardon, Sean F., and Claudia Galindo. 2009. “The Hispanic-White Achievement Gap in Math and Reading in the Elementary Grades.” American Educational Research Journal 46 (3): 853–91.