Book-sharing and Heritage Language Vocabulary Development Among Dual Language Preschoolers

• In Event: Poster Session 13
  In Poster Session: PS 13 Topic - Language, Communication

Abstract

Introduction
Heritage language (HL) is closely linked to one’s ethnic background. Research has supported that HL maintenance has personal benefits, such as identity formation and socialization of heritage culture (Fishman, 2001; Tse, 1996). Loss of HL can impair dual language learners’ (DLLs’) cultural identity development and even familial bonds (Ennser-Kananen, 2012; Fillmore, 2000). It also hinders bilingual development which affects DLLs’ cognitive and socio-emotional development (Bialystok, 2015; Kuhl, 2009). Much of the past literature has focused on DLL’s English development, but given the benefits and lack of studies in HL maintenance, it is important to examine DLL’s HL development. DLLs’ language and literacy experience at home can influence their language development (Bronfenbrenner & Morris, 2006) and is related to their language proficiency (Hammer et al., 2012). Previous research has found that parents’ book-sharing language use and practices predicted children’s vocabulary outcomes (Malin et al., 2014; Quiroz et al., 2010). However, previous studies have focused on monolingual English-speaking families. Furthermore, few studies have examined different immigrant groups from the same areas to understand possible similarities and differences between immigrant groups. How shared-book reading practices help maintain and develop DLLs’ HL for Spanish- and Cantonese-speaking families will be the focus of this study.

Methods
A total of 37 Mexican American (MA) and 37 Chinese American (CA) DLLs and their families were recruited from Head Start programs in Northern California. All parents primarily used Spanish or Cantonese during shared book-reading. Transcripts were coded using a coding scheme adapted from Luo & Tamis-LeMonda (2017) in which each parental statement was coded as yes/no, referential, behavioral or inferential (Cohen’s kappa = 0.75). Utterances describing name, feature or location of objects were coded as referential; utterances describing characters’ actions were coded as behavioral; utterances that require inferences, interpretations, internal states, manufactured speech or deductive thinking were coded as inferential. Children’s HL expressive vocabulary was collected with the Picture Vocabulary subtest from Woodcock-Johnson IV Tests of Oral Language (Schrank et al., 2014) and receptive vocabulary with the Peabody Picture Vocabulary Test (Spanish: Dunn et al., 1986; Chinese: Lu & Liu, 1998). Raw scores were used as there were no standardized versions for bilingual Chinese populations in the U.S.

Results
There were no significant differences in DLLs’ vocabulary outcomes and their parents’ book-sharing styles, except for yes/no statements, between MA and CA groups. Multiple linear regression analysis results revealed that 1) yes/no and behavioral statements were negatively associated with expressive vocabulary (F(9, 61) = 2.82, p < .01) and 2) yes/no, behavioral and inferential statements were positively associated with receptive vocabulary (F(9, 54) = 3.26, p < .01), after controlling for families’ per capita income, parent level of education, child’s age, cultural group and reading frequency. The findings suggest that parents’ book-sharing styles, particularly yes/no, behavioral and inferential, can influence DLL preschoolers’ HL vocabulary outcomes from low-income families. Implications for home literacy practice will be discussed.

Authors

• Emily Mak, University of California - Davis
  Presenting Author

• Ezra Mauer, University of California - Berkeley
  Non-Presenting Author

• Yuuko Uchikoshi, University of California - Davis
  Non-Presenting Author

• Rufan Luo, Arizona State University
  Non-Presenting Author

• Qing Zhou, University of California - Berkeley
  Non-Presenting Author