Abstract
This study investigates change in the home language environment of Language Minority (LM) learners between kindergarten and eighth grade, drawing on data from the National Center for Education Statistics’ Early Childhood Longitudinal Study–Kindergarten (ECLS-K) Cohort. The analytic sample included 1,538 children whose parents reported a language other than English spoken at home to any extent and for whom relevant data were available for both time points. Mantel-Haenszel chi-square tests show that patterns of language use in the home are dynamic, suggesting that such patterns among the growing school-age LM learner population cannot be treated as time-invariant.
The generally low academic achievement of students who come from homes in which a language other than English is spoken, commonly referred to as Language Minority (LM) learners, is well documented (e.g., August & Shanahan, 2006). Yet our knowledge of whether, and to what extent, patterns of language use in the home change over time remains limited. The process of language shift in itself is not a new phenomenon, as patterns of language use have been found to change across time and across generations, with shifts toward English monolingualism in the U.S. context (e.g., Rumbaut, Massey, & Bean, 2006; Veltman, 1983; Zentella, 1997). The challenge is that the bulk of the language use data has been collected from cross-sectional surveys or from one-time case studies. Because these studies typically compare multiple generations of immigrants, they rarely capture changes in language use over the course of individuals’ schooling.
Longitudinally documenting the dynamics of language exposure and use among LM learners is a critical next step to supporting improved models of second language acquisition and informing educational decisions about these learners. However, only a few relevant longitudinal studies have been conducted to date (e.g., Portes & Rumbaut, 2006; Reese, Garnier, Gallimore, & Goldenberg, 2000; Suárez-Orozco, Suárez-Orozco, & Todorova, 2008). Portes and Rumbaut found substantial shifts toward English among children of immigrants from adolescence to young adulthood, suggesting that shifts can occur within one generation, even among the foreign born. Because the large majority of today’s school-age LM learners are U.S.-born (August & Shanahan, 2006) and are thus instructed in U.S. classrooms from kindergarten on, the children’s shifts to English can be expected to occur all the more quickly. However, the extent to which this occurs remains an open empirical question.
The Present Study
This study investigates change in the home language environment of LM learners between kindergarten and eighth grade, drawing on data from the National Center for Education Statistics’ Early Childhood Longitudinal Study–Kindergarten (ECLS-K) Cohort (Tourangeau, Lê, Nord, & Sorongon, 2009). ECLS-K selected a nationally representative cohort of 22,104 children who entered kindergarten in 1998 and followed them through eighth grade. In the fall of kindergarten and in the eighth-grade year, parent interviews were conducted to collect information on various aspects of the children’s home environment. Specifically, parents reported on whether a language other than English was spoken at home, and if so, the frequency with which this language was spoken by the mother to the participating child and by the child to the mother. The analytic sample included 1,538 children whose parents reported a language other than English spoken at home to any extent and for whom relevant data were available for both time points. In this sample, 92.5% of the children were born in the United States and 62.3% of the mothers were born outside the United States.
Results
Overall, there was considerable instability in the frequency with which both mothers and children used a language other than English at home between kindergarten and eighth grade. Table 1 displays the changes in mother-to-child and child-to-mother language use between kindergarten and eighth grade, specifically showing the following:
Half of the sample demonstrated stability in language use.
More than one third of the sample shifted toward less frequent use of the native language.
About 13% of the sample shifted toward more use of the native language.
Changes in Reported Home Use of Language Other Than English Between Kindergarten and Eighth Grade (n = 1,538)
Note. Stable = no change in reported home use of language other than English; less = less reported home use of language other than English; more = more reported home use of language other than English.
Mantel-Haenszel chi-square tests indicated that these shifts were statistically significant (for further details, see supplemental document available on the journal website). The only exception was the “very often” category for child-to-mother use, as there was somewhat greater stability in this category across the two time points (see supplemental document). Additional subgroup analyses indicated that these patterns were largely the same for children from different language backgrounds (see supplemental document).
Discussion
A defining feature of the LM learner population is that LM learners are exposed to and/or use a language other than English at home. Educators and researchers need to gain a better understanding of the extent to which LM learners’ patterns of home language use are stable or dynamic (i.e., continuously changing) over time to inform educational decisions that account for the complex influences of home environments and to enable the creation of theoretical models of language learning that are more nuanced in their description of these developmental changes.
Our results show that patterns of language use in the home are dynamic. Only half of the sample’s patterns of home language use remained stable from kindergarten to eighth grade. For the other half, the predominant shifts—as might be expected, and in line with the longitudinal work of Portes and Rumbaut (2006)—were toward more English use. However, for some of the families, we found, albeit to a much lesser extent, shifts toward more use of the native language. Taken together, these results suggest that the home language background of today’s largely U.S.-born LM learner population cannot be treated as time-invariant. For educators, this means that obtaining data on home language use upon school entry is necessary but insufficient for providing instructional contexts that match students’ needs. For researchers, this means that designs, analyses, and theoretical models must account for the time-varying nature of LM learners’ patterns of home language use.
It is clear that English proficiency is necessary for academic success in U.S. schools and in the world labor market (e.g., Tienda & Mitchell, 2006), but it is unlikely that gaining English proficiency needs to be synonymous with native language loss. Aside from the importance of maintaining family ties, the ability to speak more than one language will likely result in greater economic opportunities in an increasingly globalized world (for a discussion, see Portes & Rumbaut, 2001). Thus the need for longitudinal work that focuses on understanding the complex and dynamic nature of home language use across the school years, as it relates to other sociodemographic factors (e.g., income, parental education), becomes all the more pressing. Specifically, a critical next step will be to investigate how changes in students’ reported home language use relate to other variables such as academic achievement and social mobility over time.