Elsevier

Early Childhood Research Quarterly

Volume 31, 2nd Quarter 2015, Pages 19-33
Early Childhood Research Quarterly

Building vocabulary in two languages: An examination of Spanish-speaking Dual Language Learners in Head Start

https://doi.org/10.1016/j.ecresq.2014.12.006Get rights and content

Highlights

  • We examined English and Spanish vocabulary among Head Start Dual Language Learners.

  • In fall and spring, children knew more about the language they heard most at home.

  • Classroom instruction predicted Spanish and English vocabulary learning in Head Start.

  • For English learning, the role of instruction was strongest for those with low initial skills.

Abstract

This study examines the English and Spanish vocabulary skills that young Dual Language Learners (DLLs) bring to Head Start, as well as their vocabulary learning over the year. Further, we isolate the unique contributions of various child, family, teacher, and classroom factors to these skills. Participants were drawn from a recent cohort of the Head Start Family and Child Experiences Survey. Results show that, for both Spanish and English vocabulary, child and family factors, especially the prevalence of each language in the household, play a role in initial skills and end-of-year skills. The quality of the language of classroom instruction also predicts Spanish and English vocabulary learning over the year for all children; in English, this relation is significantly greater for children with the lowest initial skills. Findings elucidate potential leverage points for intervention to improve Spanish and English vocabulary outcomes during Head Start for these vulnerable early learners.

Introduction

As the population of native Spanish-speaking children in American classrooms grows, our education system needs to find ways to support these children's language skills, both in English and in their home language. Targeting these outcomes in the earliest years of school will help to position children for later success in their academic careers and, thereafter, in the workforce (Barnett, 2011). The Head Start preschool program can play an essential role in supporting this early learning, as it serves the nation's most needy children, a disproportionate number of whom are Spanish-speaking Dual Language Learners (DLLs). This paper examines the experiences of a large sample of Spanish-speaking DLLs who are representative of Spanish-speaking children across the nation who began Head Start in 2006–2007 and remained enrolled through spring 2007. Analyses explore the extent of children's receptive vocabulary knowledge in English and Spanish at the beginning and end of the preschool year as well as the degree to which child, family, and school factors predict vocabulary competence in one or both languages.

Vocabulary development is essential for communicating with others, learning to read, and succeeding in school more broadly (Dickinson et al., 2010, LeFevre et al., 2010, Menting et al., 2011, Spere and Evans, 2009). Children learn the meaning of words through multiple exposures in meaningful contexts, such as rich conversations with adults, siblings, and peers (Hart and Risley, 1999, Hirsh-Pasek and Golinkoff, 2012). For many children in poverty, these meaningful exposures may be infrequent due to limited access to resources in the home (Bradley, Corwyn, McAdoo, & Coll, 2001) and community (Hindman, Miller, Froyen, & Skibbe, 2012), as well as limited conversation with adults (Dickinson and Tabors, 2001, Hart and Risley, 1995, Heath, 1983, Hoff, 2003). As a result, typically developing children in poverty often enter kindergarten with vocabulary and other language skills more than a full standard deviation below middle-income peers (Bulotsky-Shearer et al., 2012, Hindman et al., 2010, Lee and Burkam, 2002). Those who start out behind rarely catch up (Fernald, Marchman, & Weisleder, 2013).

Early vocabulary learning is essential, and especially challenging, for young children in poverty who are learning English as well as a different home language. In American schools, more than 10% of students are DLLs, and in some states (e.g., California) this number reaches nearly 30% (National Center for Education Statistics, 2013). Most DLLs are native speakers of Spanish (National Center for Education Statistics, 2012), and this group is expected to grow through the next decade (Lopez and Gonzalez-Barrera, 2013, Ortman and Shin, 2011, Shin and Kominski, 2010). DLL children are disproportionately over-represented among the ranks of America's poor, with the majority (66%) coming from households with incomes below 200% of the Federal poverty level (EPE Research Center, 2009).

Spanish-speaking DLLs in poverty face the vocabulary-learning challenges outlined above in their native language and often enter kindergarten with Spanish vocabulary skills that, when assessed on standardized measures, fall more than one standard deviation (SD) below the average skills of more socioeconomically and culturally diverse Spanish speakers (Davison, Hammer, & Lawrence, 2011). Further, these children have varied opportunities to hear and practice English at home (Place & Hoff, 2011), so that they enter school with English skills that may approach 2SDs below that measure's native-speaking (monolingual) standardization sample mean (Davison et al., 2011). Catching up in vocabulary is challenging in one language, let alone two (Biemiller, 2004), and generally requires considerable time and support in both languages (Hoff et al., 2012). Consequently, by fourth grade, fully 69% of DLLs cannot read in English at even a basic level, far higher than the 28% of native English speakers who struggle to read (National Center for Education Statistics, 2013). Over time, DLLs face higher risks of chronic underperformance in all content areas and have higher rates of drop out (Aud et al., 2011, Hernandez, 2011, Thomas and Collier, 2002). Outcomes are even worse for DLLs who begin school with relatively lower English skills and/or family socioeconomic status (Kieffer, 2008).

A promising leverage point in the effort to improve the vocabulary skills of young Spanish-speaking DLLs in poverty is the federally funded Head Start program for low-income children and families. Currently, an estimated 37% of the children served by Head Start are of Hispanic/Latino backgrounds, and the majority of those are DLLs (US Department of Health and Human Services, 2013b). English language and vocabulary are a key learning goal of the Head Start program, especially for DLLs (US Department of Health and Human Services, 2013a), and there is some encouraging evidence that Head Start is effective in meeting this goal. For example, the Head Start Impact Study (Puma et al., 2012, US Department of Health and Human Services, 2010) showed that children, whether native speakers or DLLs, who spent one year in Head Start outperformed peers on English receptive vocabulary (ES = 0.09 for those attending at age 4, and 0.18 for those attending at age 3), with four-year-olds also showing continued benefits in first grade vocabulary (0.09) and third grade reading (0.11). However, findings from the Head Start Family and Child Experiences Survey (FACES) study are more mixed, showing that most children make meaningful but small gains (i.e., 2.5 standardized score points) in vocabulary during Head Start, although DLLs learned substantially less English vocabulary from fall to spring than native English speakers (Hindman & Wasik, 2013).

There is considerable support in the field for DLLs’ continued development in their native language as well, because (a) this language has essential cultural value in their families, (b) children may be able to use their native language to build complex knowledge (which limited vocabulary makes difficult in English), and (c) this new knowledge may translate to English language learning (August & Shanahan, 2006). Supporting Spanish-language learning for Spanish-speaking DLLs is not currently an explicit goal of Head Start (US Department of Health and Human Services, 2013a), likely because of resource constraints related to staffing and materials (Halle, Hair, Wandner, McNamara, & Chien, 2008). Therefore, DLLs may have varied access to Spanish-speaking teachers and peers in their classrooms. As a result, some data suggest that, similar to other early education programs (Pacini-Ketchabaw & Armstrong de Almeida, 2006), Head Start may not improve Spanish vocabulary for Spanish-speaking DLLs (Puma et al., 2012).

Thus, Head Start has at least modest benefits for the English vocabulary learning of Spanish-speaking DLL preschoolers in poverty, but there appears to be room to enhance vocabulary growth for these children in both languages. An essential first step in understanding how best to support Spanish-speaking DLLs’ vocabulary learning in Head Start is to conduct observational studies to understand these children's experiences with Spanish and English, both at home and school, and to construct a comprehensive model of the extent to which various factors uniquely and cumulatively contribute to vocabulary development in both languages.

Ecological systems theory (Bronfenbrenner, 2005) suggests that children's learning could be linked to a broad constellation of factors, especially those attendant to the child, family, and classroom, making both independent and cumulative contributions to development. Where vocabulary is concerned, this holistic framework is complemented by empirical evidence pointing to specific variables of import, outlined below. To our knowledge, there are currently no large-scale studies of Spanish-speaking DLLs in Head Start or other programs that, informed by Bronfenbrennarian theory and empirical vocabulary research, simultaneously examine a comprehensive collection of potential vocabulary predictors, including objective observations of classroom instruction (Hammer et al., 2011a, Hammer et al., 2014). Below, we explain the particular array of relevant factors through which we operationalize this ecological approach.

Spanish-speaking DLLs in US schools are learning two languages. Consequently, it is important to explore how development in these two languages is linked over time (Branum-Martin et al., 2009) and what variables explain change in either or both.

Theory (Cummins, 1979) suggests that children could leverage information learned in one language – either conceptual or phonological – to support understanding in another language (Barac and Bialystok, 2012, Bialystok et al., 2003, Kim et al., 2014); however, empirical data imply that these links are complex. Some work finds that young children who know and/or learn more Spanish vocabulary also know and/or learn more English vocabulary (Uccelli & Páez, 2007) and develop stronger English decoding and comprehension skills (Dressler and Kamil, 2006, Miller et al., 2006). Along the same lines, English vocabulary knowledge has been linked to Spanish vocabulary and/or reading knowledge (Carlisle et al., 1999, Hammer et al., 2007, Miller et al., 2006).

However, another body of evidence indicates that early learning in one language does not predict learning in the other (Hammer et al., 2014). For example, some observational work shows that DLLs’ Spanish vocabulary skills are linked to skills such as processing speed only in Spanish, not English, although English vocabulary is predictive of English processing speed (Marchman, Fernald, & Hurtado, 2010). Examining a related but distinct outcome, Gottardo and Mueller (2009) found that Spanish vocabulary skills do not directly support later English reading comprehension, although English vocabulary skills do. Further, intervention work shows that Spanish instruction supports Spanish vocabulary learning but not English vocabulary (Barnett et al., 2007, Durán et al., 2010, Farver et al., 2009) or English speaking or reading proficiency (Cena et al., 2013).

Methodological disparities including differences in the predictor (i.e., expressive or receptive vocabulary, reading) and outcomes targeted (e.g., vocabulary, processing speed, reading fluency), the study design (i.e., cross-sectional or longitudinal), and the age of children in the sample likely underlie at least some of these varied findings (Branum-Martin et al., 2009). In the current study, consistent with Cummins’ theory, we test how vocabulary in either language at the beginning of Head Start is linked to vocabulary in either language at the year's end.

A wealth of research in Head Start and similar contexts suggests that several child background factors likely predict children's Spanish and English vocabulary knowledge at the beginning of the year, as well as (but perhaps to a lesser degree) their learning over time. Much work finds that girls develop vocabulary knowledge more rapidly than boys (Huttenlocher et al., 1991, Huttenlocher et al., 2010), entering and potentially leaving preschool with stronger skills (Hindman et al., 2010, Ready et al., 2005, Winsler et al., 2014); gender may play a complex role among DLLs in households where parents use more of their native language with girls than boys (Hammer, Davison, Lawrence, & Miccio, 2009). Learning-related disabilities are often linked to lower skills at school entry and to slower learning over time (Elklund et al., 2013, Skibbe et al., 2008). Finally, older children are commonly found to enter preschool with stronger vocabulary knowledge, having had more time to learn words, although evidence is mixed regarding whether learning over time is greater among older (Hindman & Wasik, 2013) or younger children (Puma et al., 2012). The current study therefore accounts for child gender, disability status, and age.

Among family factors, more years of parent (especially maternal) education has often emerged as a robust predictor of child vocabulary knowledge (Kreisman, 2012, Letts et al., 2013), especially for English (Bohman, Bedore, Peña, Mendez-Perez, & Gillam, 2010). Further, as described above, higher levels of family poverty might undermine vocabulary in fall (Bradley et al., 2001) and/or throughout the year (Kieffer, 2008) in both languages (Bohman et al., 2010, Kim et al., 2014) Finally, for DLLs, the relative prevalence of Spanish and English in the household, and parents’ competence and comfort with these languages, is linked to initial knowledge and learning over time, with children showing higher knowledge in the more prevalent language(s) (Hammer et al., 2003, Hammer et al., 2014, Place and Hoff, 2011). Fundamentally, it is not clear how prevalent Spanish and English are in Head Start children's homes, although evidence suggests that there could be significant variation (Hammer et al., 2003, Hammer et al., 2011b, Hammer et al., 2012). This study simultaneously examines maternal education, family poverty, and home exposure to either language as predictors of vocabulary.

Classroom instruction is perhaps the most essential teacher-related factor to consider. In the United States, most early childhood classroom instruction includes at least some (and usually primarily) English (August & Shanahan, 2006), and research has generally explored effects of instruction on young learners’ English skills (Hammer et al., 2011a, Hammer et al., 2011b). Evidence from Head Start and other high-need preschool programs suggests that more frequent explicit instruction in English vocabulary supports English learning (Marulis and Neuman, 2010, Marulis and Neuman, 2013). However, work on the FACES data (Hindman & Wasik, 2013) indicates that, among children in Head Start, more frequent instruction may actually be inversely linked to vocabulary, likely because teachers emphasize instruction when children are particularly struggling but may not significantly improve their learning through these efforts.

It is uncertain how many classrooms (in Head Start and other settings) provide any access to explicit or implicit instruction in Spanish by teachers and/or peers. Given the concentration of DLLs in various parts of the country (Pew Research Center, 2012) and the efforts of Head Start to recruit local residents as staff (Zigler & Muenchow, 1994), Spanish could be quite common, at least in some areas. Teachers’ and peers’ use of Spanish in the classroom can support children's Spanish vocabulary learning in preschool (Burchinal et al., 2012, Fry, 2008, Tang et al., 2012). Indeed, data show that children who participate in Spanish–English or Spanish-only programs make gains in Spanish vocabulary and other skills, at least without detriment to English learning (Barnett et al., 2007, Cena et al., 2013, Farver et al., 2009) and potentially with benefits for English (as above). Similarly, even using mostly English but infusing Spanish strategically, such as to explain new words (Lugo-Neris, Jackson, & Goldstein, 2010) or asking families to read classroom books with children at home in their own language or in English (Roberts, 2008), can benefit Spanish vocabulary.

Whatever the language of instruction, the global quality of instructional language matters as well, given that greater word learning appears dependent upon rich, clear, and sophisticated explanations of words and concepts and complex feedback to children (Burchinal et al., 2009, Castro et al., 2011, Hindman and Wasik, 2013, Mashburn et al., 2008). Evidence from FACES suggests that, among all Head Start learners (i.e., native English speakers as well as DLLs), the quality of instructional language is far more predictive of vocabulary learning than the frequency of vocabulary instruction (Hindman & Wasik, 2013). This study accounts for vocabulary instruction frequency and overall instructional language quality, as well as use of Spanish, specifically among DLLs.

Children may learn more about vocabulary from more educated and experienced teachers (Connor et al., 2005, National Institute of Child Health, 2002), if modestly (Early et al., 2006, Early et al., 2007). This matter demands particular attention in Head Start, given recent increases in teachers’ levels of education – indeed, in 2009, approximately 50% of teachers held bachelor's or advanced degrees, and more recent data suggest that this figure has risen above 60% – and teachers’ varied but generally substantial experience, averaging nine years (Moiduddin, Aikens, Tarullo, West, & Xue, 2012). In addition, full-day programs may provide greater benefits than half-day programs (Hall-Kenyon et al., 2009, Votruba-Drzal et al., 2008), if extra hours in care provide additional, high-quality learning opportunities. The current study examines how each of these factors is linked to early English and Spanish vocabulary among Spanish-speaking DLLs.

Ecological systems theory (Bronfenbrenner, 2005) and recent research (Connor, Morrison, Fishman, Schatschneider, & Underwood, 2007) also posit possible interactions between factors in a child's environment. Empirical evidence shows one relevant effect in Head Start: high-quality instructional language may be most linked to vocabulary learning for those who enter preschool with the lowest vocabulary skills (Hindman and Wasik, 2013, Justice et al., 2005). To date, research has not examined this issue among Spanish-speaking DLLs in Head Start, for whom it would be important to carefully explore such interactions in both languages (i.e., whether the role of instruction in English vocabulary learning is greater for children who enter with low English skills, and/or whether the role of instruction in Spanish vocabulary learning is greater for children who enter with low Spanish skills). The current study explores both of these possibilities.

In light of the need for a more comprehensive model of vocabulary learning among Spanish-speaking DLLs in Head Start, we utilize the recent (2006) cohort of the Family and Child Experiences Survey (FACES) dataset to examine several research questions.

  • (1)

    What is the nature and variability of Head Start DLLs’ exposure to Spanish and English in their first preschool year? We anticipated that children's home language experiences would largely feature Spanish, but that a portion of the sample would speak some or even mostly English. We also anticipated that some classrooms might include teachers and/or peers speaking some Spanish.

  • (2)

    In the fall of Head Start, what child and family factors predict Spanish and English receptive vocabulary? We hypothesized that each of the child and family factors outlined above might support fall skills.

  • (3)

    What child, family, and classroom factors predict receptive vocabulary learning in either language over the year (i.e., spring vocabulary knowledge, accounting for fall vocabulary knowledge)? We hypothesized that all factors outlined above might support spring skills, but that classroom factors – especially quality of instructional language – might be of particular import.

  • (4)

    Does the contribution of high-quality instructional language to vocabulary learning during preschool depend in part on children's initial skills in English, Spanish, or both? We hypothesized that the quality of instructional language might play a greater role in learning for children with lower initial skills in either language.

Section snippets

Method

The current study drew on the Family and Child Experiences Survey (FACES), 2006 cohort. Given this study's emphasis on unpacking the predictors of vocabulary learning among Spanish-speaking DLLs, we focus only on children whose parents responded “yes” to a question on the family background survey (specifically, “Is any language other than English spoken in your home?”) and then specified that this language was Spanish.

Home language(s)

Fifty-seven percent of mothers reported that their family spoke mostly Spanish at home, while 29% spoke a combination of English and Spanish and 14% spoke mostly English. Nearly all mothers reported reading well (43%) or very well (49%) in Spanish. However, regarding English, 14% of mothers reported no understanding and 55% reported very little understanding, while few understood it well (21%) or very well (10%). There was a small, significant correlation between Spanish and English competence (

Discussion

This study revealed new information about the home and classroom experiences of Spanish-speaking DLLs in Head Start, including factors that predict their vocabulary development in both languages as they begin the program and throughout the preschool year. Most children spoke either Spanish or a mix of Spanish and English at home, and most also had at least some access to Spanish in the classroom from teachers and, potentially, from peers. Children made modest gains in English vocabulary skills,

Conclusions

This secondary data analysis of the FACES data (2006 cohort) revealed that Spanish-speaking Dual Language Learners (DLLs) enter Head Start with low receptive vocabulary in Spanish and very low receptive vocabulary in English, and they make modest gains throughout the year in English but fewer gains in Spanish. Children have stronger initial and year-end skills in the language that is most prevalent in their home. Higher quality classroom instruction also contributes to Spanish learning, as well

References (123)

  • J. Huttenlocher et al.

    Sources of variability in children's language growth

    Cognitive Psychology

    (2010)
  • D. Kreisman

    The source of Black–White inequality in early language acquisition: Evidence from early Head Start

    Social Science Research

    (2012)
  • S. Aud et al.

    The condition of education 2011 (NCES 2011-033)

    (2011)
  • D. August et al.

    Developing literacy in second-language learners: Report of the National Literacy Panel on language-minority children and youth

    (2006)
  • R. Barac et al.

    Bilingual effects on cognitive and linguistic development: Role of language, cultural background, and education

    Child Development

    (2012)
  • W.S. Barnett

    Effectiveness of early educational intervention

    Science

    (2011)
  • D.C. Berliner

    Meta-comments: A discussion of critiques of L.M. Dunn's monograph bilingual Hispanic children on the US mainland

    Hispanic Journal of Behavioral Sciences

    (1988)
  • E. Bialystok et al.

    Developing phonological awareness: Is there a bilingual advantage?

    Applied Psycholinguistics

    (2003)
  • A. Biemiller

    Teaching vocabulary in the primary grades: Vocabulary instruction needed

  • T.M. Bohman et al.

    What you hear and what you say: Language performance in Spanish–English bilinguals

    International Journal of Bilingual Education and Bilingualism

    (2010)
  • R.H. Bradley et al.

    The home environments of children in the United States part I: Variations by age, ethnicity, and poverty status

    Child Development

    (2001)
  • L. Branum-Martin et al.

    Pictures and words: Spanish and English vocabulary in classrooms

    Journal of Educational Psychology

    (2009)
  • U. Bronfenbrenner

    Making human beings human: Bioecological perspectives on human development

    (2005)
  • M. Burchinal et al.

    Threshold analysis of association between child care quality and child outcomes for low-income children in pre-kindergarten programs

    Early Childhood Research Quarterly

    (2009)
  • V. Buysse et al.

    Effects of early education programs and practices on the development and learning of dual language learners: A review of the literature

    Early Childhood Research Quarterly

    (2013)
  • J. Carlisle et al.

    Relationship of metalinguistic capabilities and reading achievement for children who are becoming bilingual

    Applied Psycholinguistics

    (1999)
  • D.C. Castro et al.

    Promoting language and literacy in young dual language learners: Research, practice, and policy

    Child Development Perspectives

    (2011)
  • J. Cena et al.

    The impact of a systematic and explicit vocabulary intervention in Spanish with Spanish-speaking English learners in first grade

    Reading and Writing

    (2013)
  • C.M. Connor et al.

    Algorithm-guided individualized reading instruction

    Science

    (2007)
  • C.M. Connor et al.

    Teacher qualifications, classroom practices, family characteristics, and preschool experience: Complex effects on first graders’ vocabulary and early reading outcomes

    Journal of School Psychology

    (2005)
  • J. Cummins

    Linguistic interdependence and the educational development of bilingual children

    Review of Educational Research

    (1979)
  • J. Cummins

    “Teachers are not miracle workers”: Lloyd Dunn's call for Hispanic activism

    Hispanic Journal of Behavioral Sciences

    (1988)
  • Department of Health and Human Services

    “Friendly FACES” FACES findings: New research on Head Start outcomes and program quality

    (2006)
  • D.K. Dickinson et al.

    Language, reading, and classroom supports: Where we are and where we need to be going

  • D.K. Dickinson et al.

    Speaking out for language: Why language is central to reading development

    Educational Research

    (2010)
  • C. Dressler et al.

    First- and second-language literacy

  • L.L. Dunn et al.

    Peabody picture vocabulary test

    (2007)
  • L.M. Dunn et al.

    Test de vocabulario en imagines Peabody

    (1986)
  • D.M. Early et al.

    Teachers’ education, classroom quality, and young children's academic skills: Results from seven studies of preschool programs

    Child Development

    (2007)
  • K.M. Elklund et al.

    Predicting reading disability: Early cognitive risk and protective factors

    Dyslexia: An International Journal of Research and Practice

    (2013)
  • EPE Research Center

    Analysis of the U.S. Census Bureau's American Community Survey (2005–2007)

    (2009)
  • J. Farver et al.

    Effective early literacy skill development for young Spanish-speaking English language learners: An experimental study of two methods

    Child Development

    (2009)
  • A. Fernald et al.

    SES differences in language processing skill and vocabulary are evident at 18 months

    Developmental Science

    (2013)
  • R. Fry

    The role of schools in the English Language Learner achievement gap

    (2008)
  • T.A. Gibson et al.

    The receptive–expressive gap in the vocabulary of young second-language learners: Robustness and possible mechanisms

    Bilingualism: Language and Cognition

    (2012)
  • A. Gottardo et al.

    Are first- and second-language factors related in predicting second-language reading comprehension? A study of Spanish-speaking children acquiring English as a second language from first to second grade

    Journal of Educational Psychology

    (2009)
  • K.M. Hall-Kenyon et al.

    How do linguistically diverse students fare in full- and half-day kindergarten? Examining academic achievement, instructional quality, and attendance

    Early Education and Development

    (2009)
  • T. Halle et al.

    Predictors and outcomes of early versus later English language proficiency among English language learners

    Early Childhood Research Quarterly

    (2008)
  • C.S. Hammer et al.

    The effect of maternal language on bilingual children's vocabulary and emergent literacy development during Head Start and kindergarten

    Scientific Studies of Reading

    (2009)
  • Cited by (0)

    View full text