I want a summary of the article ( find the article attached) no reference
Bilingual children's long‐term outcomes in English as a second language: language environment factors shape individual differences in catching up with monolinguals SPECIAL ISSUE ARTICLE Bilingual children’s long-term outcomes in English as a second language: language environment factors shape individual differences in catching up with monolinguals Johanne Paradis and Ruiting Jia Department of Linguistics, University of Alberta, Canada Abstract Bilingual children experience more variation in their language environment than monolingual children and this impacts their rate of language development with respect to monolinguals. How long it takes for bilingual children learning English as a second language (L2) to display similar abilities to monolingual age-peers has been estimated to be 4–6 years, but conflicting findings suggest that even 6 years in school is not enough. Most studies on long-term L2 development have focused on just one linguistic sub-domain, vocabulary, and have not included multiple individual difference factors. For the present study, Chinese first language-English L2 children were given standardized measures of vocabulary, grammar and global comprehension every year from 4 ½ to 6 ½ years of English in school (ages 8½ to 10½); language environment factors were obtained through an extensive parent questionnaire. Children converged on monolingual norms differentially according to the test, with the majority of children reaching monolingual levels of performance on the majority of tests by 5 ½ years of English exposure. Individual differences in outcomes were predicted by length of English exposure, mother’s education, mother’s English fluency, child’s use of English in the home, richness/quality of the English input outside school and age of arrival in Canada. In sum, the timeframe for bilinguals to catch up to monolinguals depends on linguistic sub- domain, task difficulty and on individual children’s language environment, making 4–6 years an approximate estimate only. This study also shows that language environment factors shape not only early-stage but also late-stage bilingual development. Research highlights • One of very few longitudinal studies on the long-term oral language development of bilingual children learning English as a second language. • The only study to date examining multiple individual difference factors predicting variation in long-term outcomes of English second language children across multiple linguistic sub-domains. • One of very few studies examining linguistic sub- domains other than vocabulary in long-term out- comes of English second language children. • Unique findings include a complex answer to the question of ‘When do English second language children catch up with their monolingual peers?’ Introduction Research with monolingual children has shown that variations in their language environment, e.g. in the amount of input, quality of the input and the frequency and complexity of linguistic structures, influence their rate of language development (Ambridge, Kidd, Rowland & Theakston, 2015; Fernald & Weisleder, 2011; Hart & Risley, 1995; Hoff, 2006; Huttenlocher, Waterfall, Address for correspondence: Johanne Paradis, Department of Linguistics, 4-57 Assiniboia Hall, University of Alberta, Edmonton, AB, T6G 2E7, Canada; e-mail:
[email protected] © 2016 John Wiley & Sons Ltd Developmental Science 2017; 20: e12433 DOI: 10.1111/desc.12433 Vasilyeva, Vevea & Hedges, 2010; Lieven, 2010). A key difference between bilingual and monolingual children is that there are more potential sources of variation in their language environment because bilinguals would experi- ence all the sources of variation that monolinguals would in addition to the variation inherent to learning two languages instead of just one (Paradis & Gr€uter, 2014). Most evident is that bilingual children receive less input in each of their languages, on average, than monolingual age peers learning each language. Furthermore, the input in each language to bilingual children is seldom equal, relative amounts of input can change over time and according to social and educational circumstances, and the timing of input in each language can be staggered because one might be learned later than another. Much research has shown that bilingual children’s rate of development is indeed sensitive to quantity and quality of input as well as to the frequency and complexity of linguistic structures; moreover, interactions emerge between these factors within and across the two languages (V.M. Gathercole, 2007; V.M. Gathercole & Thomas, 2009; Hoff, Welsh, Place & Ribot, 2014; Paradis, 2010a; Paradis, Nicoladis, Crago & Genesee, 2011b; Paradis, Tremblay & Crago, 2014; Rispens & DeBree, 2015; Smithson, Paradis & Nicoladis, 2014; Thomas, Williams, Jones, Davies & Binks, 2014). Overall, research with bilingual children can contribute important insights into how variations in children’s language environment shape their course of development. The term bilingual can be applied to any child learning two languages, but the timing and sociolinguistic context of their language learning can be different and this, in turn, can impact their language development and out- comes in each language (V.M. Gathercole, 2007; V.M. Gathercole & Thomas, 2009; Paradis, Genesee & Crago, 2011a). Bilingual children from newcomer (immigrant and refugee) backgrounds typically speak a minority language as their first language (L1), and learn the societal language as their second language (L2), mainly through preschool and school (these children are often referred to as ‘ELLs’ – English language learners – in Canada and the United States). The majority of research on individual differences in bilingual development has focused on the preschool and early elementary school years and/or on simultaneous bilinguals who began to learn both languages from birth or in the toddler years. We know less about sources of individual differences in the long-term outcomes of bilingual children learning English as a second language (L2), and such research would have both theoretical and applied relevance. Input-driven accounts of language development, such as Usage-Based theory, predict that language input and environment factors would continue to influence language development and use even at later stages. This is because a central assumption of Usage-Based theory is that linguistic systems – at the individual and the societal level – are not static or uniform but instead are always being shaped to some extent, and by similar factors that shape early development (Behrens, 2009; Bybee, 2010; Ellis, 2008; Tomasello, 2003). Therefore, research exam- ining the influence of language input and environment factors on late-stage L2 acquisition would test this Usage-Based theory prediction. Regarding applied rele- vance, in contrast to popular beliefs about speedy rates of English L2 development in newcomer children, even after years of exposure to English in school these children can still lag behind monolinguals (e.g. Cobo- Lewis, Pearson, Eilers & Umbel, 2002; Paradis et al., 2011a). Understanding how long it takes for bilingual children to ‘catch up’ with their monolingual peers in the L2 – if all of them do so and under what circumstances – has broad societal interest because newcomer children’s proficiency in the L2 lays the foundation for academic success and integration (Cummins, Mirza & Stille, 2012; OECD, 2006; Saunders & O’Brien, 2006). Furthermore, research examining long-term L2 outcomes across different linguistic sub-domains and with sources of individual differences could challenge a simplistic inter- pretation of what ‘catching up’ means. Accordingly, the goal of the present study is to examine the English L2 development of bilingual chil- dren in the upper elementary school years to understand (1) whether they converge with monolingual norms for L2 abilities across different linguistic sub-domains/tasks in this timeframe and (2) how language environment factors shape their L2 development at this stage. Individual differences in early English L2 development Much recent research has focused on child-external, language environment factors, as well as child-internal factors, as sources of individual differences in bilingual children’s early English L2 development. Not surpris- ingly, virtually all studies have found that amount of English input as measured by time in English preschool/ school positively predicts higher English L2 abilities (e.g. Blom, Paradis & Sorenson Duncan, 2012; Bohman, Bedore, Pe~na, Mendez-Perez & Gillam, 2010; Chondro- gianni & Marinis, 2011; Cobo-Lewis et al., 2002; Collins, O’Connor, Su�arez-Orozco, Nieto-Casta~non & Topple- berg, 2014; V.M. Gathercole, 2007; Paradis, 2011). Higher family socioeconomic (SES) background and greater richness of the L2 environment outside school (e.g. frequency and diversity of reading, media use, organized activities and playing with friends in the L2) are input quality factors associated with stronger L2 © 2016 John Wiley & Sons Ltd 2 of 15 Johanne Paradis and Ruiting Jia abilities in the early stages of development (Bohman et al., 2010; Collins et al., 2014; V.M. Gathercole, 2007; Golberg, Paradis & Crago, 2008; Hammer, Komaroff, Rodriguez, Lopez, Scarpino et al., 2012; Jia & Fuse, 2007; Cobo-Lewis et al., 2002; Paradis, 2011). When the impact of English spoken at home is considered sepa- rately, findings become more complex. Bohman et al. (2010) and Hammer et al. (2012) found differences in the impact of English input vs. output at home such that child output – use of English – was more predictive than the child hearing English at home. The impact of languages spoken at home on bilingual children’s L2 development is also influenced by parental fluency in the L2, which in turn is modulated by parental education (Chondrogianni & Marinis, 2011; Hammer et al., 2012; Hoff et al., 2014; Saunders & O’Brien, 2006; Winsler, Burchinal, Tien, Peisner-Feinberg, Espinosa et al., 2014). Low English fluency among parents with lower education levels could provide partial explanation for why English input at home is not always facilitative, and why SES background has an influence on children’s L2 development. In conjunction with these environmental factors, child-internal cognitive mechanisms like verbal short- term memory and non-verbal analytic reasoning also predict individual differences in bilingual children’s English L2 abilities (Collins et al., 2014; Farnia & Geva, 2011; Paradis, 2011). Bilingual children’s two languages are also a resource for each other in development. Studies have shown positive cross-language transfer for grammatical features, as well as cross-domain and cross- language influence at the language–literacy interface (Blom et al., 2012; Paradis, 2011; Pasquarella, Chen, Lam, Luo & Ram�ırez, 2011; Ram�ırez, Chen & Pasquar- ella, 2013; Zdorenko & Paradis, 2008). Paradis (2011) is particularly relevant because a sub- set of the participants from this study was followed longitudinally in the present study. Paradis (2011) examined language environment and child-internal pre- dictors of 169 English L2 children’s vocabulary and grammatical abilities; children had a mean of 20 months of exposure to English in preschool/school and ranged in age from 4;10 to 7;0. Regression modeling revealed that child-internal factors, such as analytic reasoning (non- verbal IQ), verbal short-term memory (non-word repe- tition), L1 typology and age, predicted more variance than the environmental factors such as length of exposure to English, richness of the English environment and maternal education (a measure of SES background); however, all of these factors were significant predictors of children’s English abilities. One objective of following a cohort of these children longitudinally was to ascertain whether the proportion of variance in children’s lan-