Learning Spanish Sociocultural Background

Learning Spanish as a Heritage Language: The Role of Sociocultural Background Variables Janet S. Oh Department of Psychology, California State University Northridge, 18111 Nordhoff Street, Northridge, CA 91330-8255, USA Terry Kit-fong Au Department of Psychology, The University of Hong Kong, Pokfulam Road, Hong Kong This study represents a first effort to identify sociocultural background variables that predict the successful mastery of a heritage language by adult learners. While there are well-documented sociocultural models for other second language learners, thus far there are no similar published models for heritage language learners. We focused on Latino college students learning Spanish (N 5 55) and the following background variables: cultural identification, cultural participation, and use of Spanish outside of class. Language assessments focused on learners’ mastery of Spanish accent and grammar. Results indicate that these background variables are related to mastery of Spanish accent, and that use of Spanish outside of class is related to mastery of Spanish grammar. The findings have important implications for language instruction and point to a number of important future research directions in the field of heritage language learning and instruction. Keywords: adult language learner, cultural, heritage language, language instruction, language learning, Latino

The immigrant-background population in the United States is growing rapidly, and much of this population comes from backgrounds in which a non-English language is spoken. Although the growing immigrant population continues to bring a variety of non-English, heritage languages to the US, many of the children and grandchildren of these immigrants have limited or no ability to speak their heritage language. In fact, heritage languages are usually lost by the second or third generation of immigrants in the US (Fishman, 1978; Krashen, 1996; Veltman, 1983), and this process seems to be accelerating, with heritage languages now more often being lost during the second generation (i.e. the first generation born in the US; Fillmore, 2000). In many cases, immigrantbackground adults who lost their heritage language as children – or never spoke the language – want to learn or relearn the language in adulthood. In the present study, we focused on one such group, Latino college students (i.e. those from Latin American cultural backgrounds) who were (re)learning Spanish, in order to better understand what factors might predict successful mastery of their heritage language. 0790-8318/05/03 229-13 $20.00/0 LANGUAGE, CULTURE AND CURRICULUM

229

# 2005 J. S. Oh and T. K. Au Vol. 18, No. 3, 2005

230

Language, Culture and Curriculum

Latinos in the US: Spanish as a Heritage Language Latinos are the fastest growing ethnic minority group in the United States (Marotta & Garcia, 2003). In 2002, 37.4 million people in the US were of Latin American descent (13.3% of the population), 15 million of whom were foreign born (Ramirez & de la Cruz, 2003). In fact, Latinos now make up about half of the foreign-born population (53.3%; Larsen, 2004). Spanish is therefore the most common heritage language in the US, with 28.1 million speakers in the year 2000 (Shin & Bruno, 2003). There is some evidence that Spanish as a heritage language is maintained at higher rates than other heritage languages in the US, at least among the second generation. Portes and Hao (1998) found that while only 29% of secondgeneration Asian American adolescents reported being able to speak their heritage language ‘very well’ or ‘well’, 82% of their Latino counterparts reported the same. However, even with Latino adolescents’ much higher proficiency in their heritage language, Latino and Asian American students exhibited similar levels of preference for English over their respective heritage languages: 71% of Latino and 74% of Asian American students reported that they preferred to speak English over their heritage language. Therefore, although there may be higher levels of heritage language proficiency among Latino students as compared to other linguistic minority groups, their preference for English indicates that they are still much more likely to use English rather than Spanish, and thus are not likely to pass on their heritage language to their own children. In fact, the pressures to learn English are so strong that it is common for Latino parents to struggle to raise their children as SpanishEnglish bilinguals rather than English monolinguals (Bayley et al., 1996). Adult language learners and the sociocultural context In many cases, adults who preferred speaking English over their heritage language as children and/or adolescents want to (re)learn their heritage language in adulthood. Unfortunately, as many adult language learners know from personal experience, it is often very difficult to master a language learned after childhood. As numerous studies in this area have shown, early language experience is crucial for achieving native-like mastery of a language, especially in the domains of accent and grammar (e.g. Johnson & Newport, 1989; Long, 1990; Oyama, 1976; Williams, 1980). However, much of this work focuses on first-time, adult learners of a language, and relatively little is known about adult heritage language learners (HLLs). Adult HLLs are likely to differ from other adult second language learners (L2Ls) in two key ways that may affect their success in learning the target language. First, they are more likely to have had some prior exposure to the language, especially during early childhood. Adult HLLs, because of their immigrant background, often come from homes in which they were exposed to their heritage language from an early age. However, with the rapid rate of loss of heritage languages in the US, their early exposure is often limited – to just passive, overhearing experience with the language and/or to experience that is restricted just to the early years (usually only until starting school).

Spanish as a Heritage Language

231

Clearly, early experience with a language is important for mastering that language. In fact, there is growing evidence that childhood language experience, even in the case of these types of limited experience, can help in the acquisition of that language later in life. Adult learners of Korean who spoke the language only during early childhood have better accents than first-time, adult learners of the language (Oh et al., 2003). Additionally, adult learners of Spanish who had overheard the language regularly during childhood (but only rarely spoke it, if at all) had better accents than adult learners without such childhood exposure to Spanish (Au et al., 2002; Knightly et al., 2003). Hence, it is clear that the early language experience that many HLLs bring to the classroom sets them apart from their classmates. Another key way in which HLLs may differ from L2Ls is the sociocultural context in which they are learning the target language. There is a well-established line of research studying the impact of various sociocultural background variables for adult L2Ls (e.g. Cle´ment et al., 2003; Gardner et al., 1999; Masgoret & Gardner, 2003). These studies have focused on sociocultural background variables such as opportunities to use the target language, motivation for learning the target language, and attitudes toward the target language group. Generally, they have found that sociocultural variables such as these are at least moderately related to achievement in second language learning (e.g. Masgoret & Gardner, 2003), or to willingness to communicate in the second language (e.g. Cle´ment et al., 2003). However, this research focuses on L2Ls, and relatively little is known about how these types of background variables may play a part in the success of adult HLLs. The present study is an effort to identify background variables that may play a part in adult HLLs’ success. While there are well-documented sociocultural models of language learning for L2Ls, thus far there are no comparable published models for HLLs. In fact, it is as yet unclear what the relevant and important sociocultural background variables might be. Background variables such as motivation for learning the language and identification with the language group are likely to play a part in the language learning success of HLLs as well, but for different reasons. The present study represents a first effort in identifying these variables as they relate to the success of HLLs. We focused on the following background variables as possible predictors of successful language mastery in Latino college students learning Spanish: identification with Latino culture, participation in Latino cultural activities, and use of Spanish in out-of-class settings. We chose these variables to cover a range of social and cultural background variables that may affect HLLs’ mastery of the target language, and because there is evidence that roughly analogous background variables play a part in the success of L2Ls. These analogous variables are, respectively: attitudes toward the target language group (Masgoret & Gardner, 2003), frequency and quality of contact with the language group (Cle´ment et al., 2003), and opportunities to use the target language (Flege et al., 1999; Gardner et al., 1983). Additionally, identification with the heritage culture has been mentioned by Korean HLLs as one of the reasons why they want to learn their heritage language (Cho et al., 1997). For our assessment of Spanish language mastery, we focused on mastery of Spanish accent and grammar, since late language learners typically encounter the most difficulty acquiring native-like proficiency in these two aspects of language (e.g. Johnson & Newport, 1989; Long, 1990).

232

Language, Culture and Curriculum

Method Participants The research reported here is part of a larger study of adult Spanish language learners. All participants were recruited from second-year, college-level Spanish language classes at a large university in Southern California. For the purposes of this study, only those participants who had at least one Latino parent were included in the analyses, N ¼ 55 (mean age ¼ 20.5 years, six participants did not report age; 36 female, 18 male, one participant did not report gender). This included 28 participants with two Latino parents, and 27 participants with one Latino parent (for most of the latter group, the other, non-Latino parent was either white or of mixed racial background). Participants ranged from having no prior experience with Spanish before taking a Spanish language course in high school or college to having spoken Spanish until at least adolescence. Because we were interested in the unique influence of sociocultural variables rather than prior linguistic experience, early experience with Spanish was controlled for in the analyses (see below). Materials and procedure All participants first completed a detailed background questionnaire asking about their language background, identification with Latino culture, participation in Latino cultural activities, and use of Spanish outside the classroom. They then took part in a language abilities assessment. They finished with a brief follow-up interview asking about their language background. Background characteristics Early language experience All participants answered a detailed language background questionnaire asking about their experiences with Spanish up to the time of the study. This questionnaire included general questions about language background (e.g. participant’s first language, parents’ language abilities), along with specific questions about the quantity (how much they heard, were spoken to, and spoke Spanish) and quality (words/short phrases/sentences; extent of mixing with English) of their experience with Spanish since birth. A follow-up interview offered a chance to clarify questionnaire responses. Based on this information, participants were classified into one of the following learner groups: (1) novice learner-began in college; (2) novice learner-began in high school; (3) minimal childhood exposure; (4) childhood overhearer; (5) childhood addressee; (6) childhood speaker; and (7) regular speaker (see Table 1 for sample sizes by learner group). Novice learners did not have any regular exposure to Spanish prior to beginning Spanish language courses in college or high school. Those with minimal childhood exposure heard Spanish occasionally during childhood, but never on a regular basis (only a couple of hours a month; this usually consisted of visits with a Spanishspeaking grandparent). Childhood overhearers regularly overheard others speaking Spanish during childhood, but they were only addressed in Spanish minimally and/or spoke Spanish minimally, if at all. Childhood addressees were regularly addressed in Spanish during childhood, but they

Spanish as a Heritage Language

233

Table 1 Sample sizes by learner group Learner group

N

Novice learner: college

2

Novice learner: high school

2

Minimal childhood exposure

4

Childhood overhearer

18

Childhood addressee

6

Childhood speaker

18

Regular speaker

5

Total sample size

55

spoke Spanish minimally, if at all. Childhood speakers regularly spoke Spanish up to about age 10, but only spoke the language minimally, if at all, after that. Regular speakers regularly spoke Spanish throughout most of their lives, until at least age 14. These groupings are based on learner groups from the larger sample. Two coders initially grouped participants into these groups. Intercoder reliability was excellent (Cohen’s kappa ¼ 0.93). Where there were disagreements, a third coder was brought in to resolve the dispute. This ordinal variable was used as our measure of early language experience in the analyses. Cultural identification To measure identification with Latino culture, we used the affirmation and belonging subscale of the Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992). The MEIM asks participants to rate their agreement on statements about their ethnic identification, using a four-point Likert scale, ranging from one (strongly disagree) to four (strongly agree). Of the original 20-item MEIM, we used the five affirmation and belonging subscale items as a measure of cultural identification (‘I am happy that I am a member of the group I belong to’). Responses were averaged (with reverse coding where necessary) to create one cultural identification score (Cronbach’s alpha ¼ 0.73). Cultural participation Participation in Latino culture was measured with the two-item MEIM ethnic behaviours subscale (‘I am active in organisations or social groups that include mostly members of my own ethnic group’), and three modified items from the MEIM other group orientation subscale. The other group orientation subscale asks about participation in cultural groups other than one’s own. For the purposes of this study, we modified three of these items to ask about participation in Latino cultural activities. For example, ‘I like meeting and getting to know people from ethnic groups other than my own’ was modified to ‘I like meeting and getting to know people from Latino background.’ Additionally, to ensure that participants were referring to participation in Latino cultural activities (rather than another culture with which they might identify), all cultural participation items were modified to refer specifically to Latino culture. For example, ‘I participate in cultural practices of my own group, such as special food, music, or customs’ was modified to read

234

Language, Culture and Curriculum

‘I participate in Latino cultural practices. . .’. Responses were averaged (with reverse coding where necessary) to create one cultural participation score (Cronbach’s alpha ¼ 0.80). Spanish use outside the classroom As an additional measure of the sociocultural context in which HLLs are learning Spanish, we asked participants about the frequency of their use of Spanish outside the classroom in high school (HS) and currently, in college. Participants rated how often they took part in a variety of activities on a fivepoint scale, ranging from zero (not at all) to four (everyday). Items included speaking Spanish with a variety of people, including family, friends, and co-workers (four items), reading various types of Spanish-language media (three items), writing in Spanish (three items), listening to or watching Spanish-language radio or TV (two items), and singing in Spanish (one item). For each of the activities, participants indicated how often they took part in the activity, responding separately for HS and current use. These were summed across all of the various activities to create one HS and one current Spanish use score (Cronbach’s alphas ¼ 0.85 and 0.88, respectively). Language abilities assessment The language abilities assessment measures we report here were part of a larger language assessment battery. The assessment was run on a computer, using the PsyScope program (Cohen et al., 1993). Accent assessment (1) Narrative accent rating: participants were asked to tell a story from an abridged version of a picture book from Mercer Mayer’s ‘Frog, Where Are You?’ series. These wordless picture books have been widely used for eliciting narratives from children and adults in various languages (e.g. Berman et al., 1994). Participants were first given a chance to familiarise themselves with the book before telling the story. They audio-tape recorded their own story telling, and these narratives were later rated by two native Spanish speakers for accent. Raters used a five-point scale to rate each narrative, ranging from one (very strong foreign accent, definitely nonnative) to five (no foreign accent, definitely native; scale adopted from Bongaerts et al., 1997). Inter-rater reliability was very high (intra-class correlation ¼ 0.97, p , 0.001). The two raters’ accent ratings were therefore averaged for one narrative accent rating score. (2) Phoneme accent rating: participants’ production of specific Spanish phonemes was also rated. Each participant was asked to produce a target word embedded in the following sentence frame: Diga [target word] por favor (‘Say [target word] please’). Target phonemes were all Spanish stop consonants and appeared in either word-initial or word-medial position (always in a stressed position). Forty-eight native Spanish speakers were recruited to rate participants’ pronunciation of the target phonemes. Using the same five-point rating scale as the narrative accent raters, raters in this task were asked to rate each speaker’s accent, focusing on the target phoneme in the sentence (these ratings are only available for 41 of the participants in the study). Again, inter-rater reliability was

Spanish as a Heritage Language

235

very high (intra-class correlation ¼ 0.98, p , 0.001), and phoneme accent ratings are therefore an average of ratings across all raters, and across all of the different phoneme types. Grammar assessment (1) Narrative grammar rating: two other native Spanish speakers rated participants’ narratives of the ‘Frog, Where Are You?’ story for grammar, using a five-point rating scale similar to that used for accent ratings. Inter-rater reliability was again very high (intra-class correlation ¼ 0.95, p , 0.001), and narrative grammar ratings are averages across the two raters’ scores. (2) Grammaticality judgments: a grammaticality judgment task was included to assess participants’ ability to detect grammar errors. Participants heard a native Spanish speaker say a sentence, and then had to judge whether it was grammatical or ungrammatical. There were 33 pairs of grammatical-ungrammatical sentences which contained errors involving a variety of grammatical categories: number, gender, and person markers in noun phrases, and verb tense and aspect marking. Each sentence was presented separately in random order. Grammaticality judgment scores reflect the proportion of items participants answered correctly.

Results Mean responses and range of responses for each variable are presented in Table 2. There were a broad range of responses to the sociocultural background questions. On the cultural identification scale, participants ranged from relative lack of identification with Latino culture to strongly identifying with the culture. On average, participants seemed to identify moderately with Latino culture. Results were similar for cultural participation. In terms of Spanish use, there was a wide range of responses. Recall that for each scale, participants rated 13 items on a scale ranging from zero to four, resulting in a maximum possible total score of 52. No one reported using Spanish in all of the contexts Table 2 Descriptive statistics and reliability measures for study variables N

Range

Mean

Standard error

Reliability

Cultural identification

55

1.6– 4.0

3.29

0.08

0.73a

Cultural participation

55

1.4– 4.0

3.10

0.09

0.80a

Spanish use: HS

55

0– 32

9.20

0.98

0.85a

Spanish use: Current

55

0– 41

12.16

1.18

0.88a

Accent-narrative

55

1.0– 5.0

3.52

0.14

0.97b

Accent-phoneme

41

2.5– 4.7

3.77

0.07

0.98b

Grammar-narrative

55

1.0– 5.0

3.21

0.14

0.95b

Grammar judgments

55 0.48– 0.92

0.70

0.07

N/A

Sociocultural variables

Language assessments

a

Cronbach’s alpha. Intra-class correlation (all p , 0.001).

b

236

Language, Culture and Curriculum

everyday in HS or currently, but rather most of our participants reported doing at least some of the activities some of the time. In terms of the language ability assessment, there was also a wide range, from relatively poor Spanish language ability to native-like abilities. On average, our participants appeared to fall in the middle of each of the rating scales, and correctly judged a moderate number (70%) of grammaticality items. Relationships between background variables and language ability measures As a first step, bivariate Pearson correlation coefficients were calculated between the background variables and the language ability measures. These reveal that all of the variables are related to the accent ratings, and that HS and current Spanish use are related to grammar abilities (see Table 3). The present study focused on the unique influence of sociocultural background variables. As many researchers have shown, early language experiences play a critical role in the mastery of a language (e.g. Johnson & Newport, 1989; Oyama, 1976; Williams, 1980). It is also clear that various sociocultural variables play a part in the success of adult L2Ls (e.g. Cle´ment et al., 2003; Masgoret & Gardner, 2003). However, it is as yet unclear what the influence of such sociocultural variables are for adult HLLs. Therefore, to investigate the unique influence of these sociocultural variables for adult HLLs’ mastery of a language (above and beyond the influence of early language experiences), we calculated correlation coefficients that controlled for variation due to early experience with Spanish. That is, partial correlations between each of the sociocultural background variables and each of the language ability measures were estimated, controlling for variation due to early experience with Spanish (Table 4). These partial correlations reveal that mastery of Spanish accent is positively related to cultural identification (for narrative accent ratings only), cultural Table 3 Bivariate correlations between background variables and language ability measures Early language Cultural Cultural Spanish use: Spanish use: experience identification participation High school Current –

0.22

0.22

0.25

0.45

Accentnarrative

0.47

0.35

0.34

0.40

0.48

Accentphoneme

0.57

0.35

0.42

0.50

0.42

Grammarnarrative

0.53

0.27

0.22

0.29

0.44

Grammar judgments

0.52

0.21

0.06

0.20

0.41

Early language experience



p , 0.01;  p , 0.05.

Spanish as a Heritage Language

237

Table 4 Partial correlations between background variables and language ability measures, controlling for early Spanish experience Cultural identification

Cultural participation

Spanish use: High school

Spanish use: Current

Accent-narrative

0.33

0.35

0.47

0.40

Accent-phoneme

0.24

0.36

0.51

0.24

Grammar-narrative

0.17

0.21

0.36

Grammar judgments

0.06

20.23

0.10





0.32 0.17

p , 0.01;  p , 0.05.

participation (for both types of accent ratings), HS Spanish use (both), and current Spanish use (narrative accent ratings only). Grammar ratings on the narrative task are positively related to Spanish use, both in HS and currently, but not to any of the other background variables. Grammaticality judgments are not reliably related to any of the background variables. Relationships among background variables In addition to the relationships between the sociocultural background variables and Spanish language mastery, it is important to understand the relationships among the background variables. Bivariate Pearson correlation coefficients among the background variables are presented in Table 5. These correlations indicate a strong positive relationship between cultural identification and cultural participation (r ¼ 0.49, p , 0.01), and between cultural participation and both HS and current Spanish use (r ¼ 0.49 and 0.47, respectively, both p , 0.01). Additionally, HS and current Spanish use were strongly positively related to one another (r ¼ 0.61, p , 0.01). Table 5 Bivariate correlations among background variables

(1) Cultural identification (2) Cultural participation (3) Spanish use: high school (4) Spanish use: current 

1

2

–

0.49 –

3

4

0.16

0.17

0.49

0.47

–

0.61 –

p , 0.01.

Discussion This study was a first systematic effort to identify sociocultural background factors that play a part in adult HLLs’ successful language mastery. In particular, we focused on Latino college students who were (re)learning Spanish. Latinos are the largest linguistic minority group in the US. This statistic, coupled with second-generation Latinos’ high rate of preference for English over Spanish (Portes & Hao, 1998), indicates that growing numbers of Latino adults will not be able to speak their heritage language. If as adults they

238

Language, Culture and Curriculum

want to (re)learn Spanish, what factors may play a part in their mastery of the language? We already know that childhood experience with Spanish helps the adult language learner in acquiring a native-like accent in the language (Au et al., 2002; Knightly et al., 2003). In this study, we focused on the unique influence of sociocultural background factors which may also play a part in adult HLLs’ success. Mastery of Spanish accent was positively related to all of the sociocultural background variables. Those who more strongly identified with Latino culture, who participated more often in Latino cultural activities, and who used Spanish more often outside the classroom in high school and currently (in college) had better accents. HLLs who feel more positively about Latino culture may be more motivated to speak Spanish ‘like a native speaker,’ and therefore may try harder to attain a native-like accent. Likewise, those who take part in more Latino cultural activities and feel positively about spending time with other Latinos may want to be able to speak like a native speaker, in order to ‘fit in’ more easily with their cultural group. Moreover, interacting with other Latinos could improve one’s accent through hearing others’ and practising one’s own accent. As a result, the more often HLLs actually take advantage of these and other opportunities to speak Spanish, the more native-like their accent. Mastery of Spanish grammar was assessed through a rating of productive abilities in a narrative task, and through a grammaticality judgment task measuring receptive abilities. HLLs who used Spanish outside the classroom more often, in high school and currently, had better productive grammar abilities. While cultural identification and cultural participation were related to mastery of Spanish accent, they were not related to Spanish grammar production. It may be that while positive feelings toward one’s cultural group and interacting with members of that group are enough to enhance accent, HLLs need to actually take advantage of those and other opportunities and speak the language in order to enhance their grammar abilities. It is unclear why none of the background variables were related to receptive grammar abilities as assessed by the grammaticality judgment task. It would seem logical that since grammar production was associated with these background variables, receptive grammar abilities would be to some extent as well. One possibility is that the Latino students who take advantage of more opportunities to speak Spanish may have a better feel for what ‘sounds right,’ thereby resulting in native-like grammar production. However, such benefits may favor commonly-used expressions, and may not necessarily be comprehensive. Therefore, compared to those who use Spanish less, they can sound better grammatically when they can choose what they say, without necessarily being better at detecting a wide-ranging and comprehensive collection of morphosyntactic errors. The relationships among the sociocultural background variables themselves reveal that their relationships with HLLs’ mastery of the language may be more complicated than simple bivariate relationships. The interrelatedness of the background variables indicates that it may be more appropriate to look at the influence of these background variables in the context of a larger model of HLLs. A more advanced statistical technique (perhaps path analysis) would

Spanish as a Heritage Language

239

be ideal for this, but before researchers can move to such techniques, we need to better understand the particular background variables involved. That is, we have chosen a number of possible variables for such a model based on sociocultural models of language learning in L2Ls, but it is as yet unclear whether all of these are relevant, and whether there are other variables that may be relevant for HLLs though not for L2Ls. Researchers need to address these issues before they can move to a sociocultural model of language learning in HLLs. Although Latinos are the largest linguistic minority group in the US, there are growing numbers of other linguistic minority groups. Exploring the role of these sociocultural background variables among HLLs of other languages would help us to understand the broader applicability of such a model. Additionally, it would be informative to study HLLs as they begin their language courses. Participants in this study were enrolled in second-year college Spanish language courses, but many of them had been taking Spanish language courses since high school. We have largely talked about the background variables as predictors of Spanish language mastery, but it is also possible that learning Spanish has in turn affected the sociocultural context of their language learning. Following HLLs’ progress beginning in their first Spanish language course would not only help us to better understand these background variables as predictors of language mastery, but also how they may in turn be affected by the act of language learning. This study represents a first effort to better understand the sociocultural context of the HLL. While sophisticated models of the L2L exist, it is as yet unclear what variables should be included in a model of heritage language learning. In this study, we have identified some variables that appear to be related to language mastery in the HLL. Further research needs to be conducted in order to begin to develop a model of how these variables (and perhaps others) play a role in the mastery of a heritage language by adult learners. This kind of model would help language instructors to better understand how language learning by HLLs differs from L2Ls. Some language programs have already begun to design language courses specifically for HLLs (e.g. ‘Spanish for Home Background Speakers’ at Stanford University, ‘Spanish for Heritage Speakers’ at the University of California, Los Angeles), which reflect the differences in the learning needs of HLLs as compared with L2Ls. The results of this study provide some preliminary evidence that encouraging HLLs to explore their cultural identification, to participate in cultural activities, and to use the language in a variety of contexts outside of the classroom may help these language learners to successfully master the language. Further understanding of the sociocultural context in which HLLs learn the heritage language will provide important information for instructors designing these courses. Acknowledgements This work was supported by a National Science Foundation Predoctoral Research Fellowship and NIMH traineeship in Developmental Cognitive Science 5T32MH10926 awarded to Janet Oh, and NIMH grant MH56118, seed grants from University of California Linguistic Minority Research Institute, UCLA Institute of American Cultures, and UCLA Academic Senate

240

Language, Culture and Curriculum

awarded to Terry Au. We thank Leah Knightly and Sun-Ah Jun, who collaborated with us on related studies and offered invaluable help in this one. We also thank Celia Adame, Marilyn Alvarado, Brenda Alvarez, Argelia Andrade, Cecilia Arriaza, J. Patrick Barjam, Jessica Brauner, Mauricio Carvallo, Ivonne Cisneros, Susan Curtiss, Kristina Cutura, Olga Escamilla, Laura Espan˜a, Catherine Fountain, Ingrid Gooding, John Grinstead, Vilma Hernandez, Gerardo Ibarra, Patricia Keating, Sahyang Kim, Hyuck-Joon Lee, Reuben Lim, Ann Marroquin, Olivia Martı´nez, Karla Mayen, Nancy Miranda, Claudia Parodi, Jaime Paz, Katie Polsky, Irasema Ramos, Salome Rinco´n, Krista Rodriguez-Bruno, Laura Romo, Norma Salazar, Susan Schaffer, Sylvia Sherno, Minjung Son, Henry Tehrani, Motoko Ueyama, and John Whalen for their help with various aspects of this research. We thank the research participants for their time. Finally, we thank an anonymous reviewer for their very helpful comments. Correspondence Any correspondence should be directed to Janet S. Oh, Department of Psychology, California State University Northridge, 18111 Nordhoff Street, Northridge, CA 91330-8255, USA ([email protected]). References Au, T.K., Knightly, L.M., Jun, S.-A. and Oh, J.S. (2002) Overhearing a language during childhood. Psychological Science 13, 238–243. Bayley, R., Shechter, S.R. and Torres-Ayala, B. (1996) Strategies for bilingual maintenance: Case studies of Mexican-origin families in Texas. Linguistics and Education 8, 389– 408. Berman, R.A., Slobin, D.I., Aksu-Koc, A.A., Bamberg, M., Dasinger, L., Marchman V., Neeman, Y., Rodkin, P.C., Sebastia´n, E., Toupin, C., Trabasso, T. and von Stutterheim, C. (1994) Relating Events in Narrative: A Cross Linguistic Developmental Study. Hillsdale, N.J.: Erlbaum. Bongaerts, T., Van Summeren, C., Planken, B. and Schils, E. (1997) Age and ultimate attainment in the pronunciation of a foreign language. Studies in Second Language Acquisition 19, 447– 465. Cho, G., Cho, K.-S. and Tse, L. (1997) Why ethnic minorities want to develop their heritage language: The case of Korean Americans. Language, Culture and Curriculum 10 (2), 106– 112. Cle´ment, R., Baker, S.C. and MacIntyre, P.D. (2003) Willingness to communicate in a second language: The effects of context, norms, and vitality. Journal of Language and Social Psychology 22 (2), 190– 209. Cohen, J.D., MacWhinney, B., Flatt, M. and Provost, J. (1993) PsyScope: A new graphic interactive environment for designing psychology experiments. Behavioral Research Methods, Instruments, and Computers 25 (2), 257– 271. Fillmore, L.W. (2000) The loss of family languages by immigrant children: Should educators be concerned? Theory into Practice 39 (4), 203– 210. Fishman, J.A. (1978) Language Loyalty in the United States. The Hague: Mouton. Flege, J.E., Yeni-Komshian, G.H. and Liu, S. (1999) Age constraints on second-language acquisition. Journal of Memory and Language 41 (1), 78 – 104. Gardner, R.C., Lalonde, R.N. and Pierson, R. (1983) The socio-educational model of second language acquisition: An investigation using LISREL causal modeling. Journal of Language and Social Psychology 2, 51 – 65. Gardner, R.C., Masgoret, A.-M. and Tremblay, P.F. (1999) Home background characteristics and second language learning. Journal of Language and Social Psychology 18 (4), 419– 437.

Spanish as a Heritage Language

241

Johnson, J.S. and Newport, E.L. (1989) Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology 21, 60– 99. Knightly, L.M., Jun, S.-A., Oh, J.S. and Au, T.K. (2003) Production benefits of childhood overhearing. Journal of the Acoustical Society of America 114 (1), 465– 474. Krashen, S. (1996) Under Attack: The Case Against Bilingual Education. Culver City, CA: Language Education Associates. Larsen, L.J. (2004) The Foreign-Born Population in the United States: 2003. Current Population Reports, P20 –551. Washington, D.C.: US Census Bureau Long, M.H. (1990) Maturational constraints on language development. Studies in Second Language Acquisition 12 (3), 251–285. Marotta, S.A. and Garcia, J.G. (2003) Latinos in the United States in 2000. Hispanic Journal of Behavioral Sciences 25 (1), 13 – 34. Masgoret, A.-M. and Gardner, R.C. (2003) Attitudes, motivation, and second language learning: A meta-analysis of studies conducted by Gardner and associates. Language Learning 53 (1), 123– 163. Oh, J.S., Jun, S.-A., Knightly, L.M. and Au, T.K. (2003) Holding on to childhood language memory. Cognition 86 (3), B53– B64. Oyama, S. (1976) A sensitive period for the acquisition of a nonnative phonological system. Journal of Psycholinguistic Research 5 (3), 261– 283. Phinney, J.S. (1992) The Multigroup Ethnic Identity Measure: A new scale for use with diverse groups. Journal of Adolescent Research 7 (2), 156– 176. Portes, A. and Hao, L. (1998) E Pluribus Unum: Bilingualism and the loss of language in the second generation. Sociology of Education 71, 269– 294. Ramirez, R.R. and de la Cruz, G.P. (2003) The Hispanic Population in the United States: March 2002. Current Population Reports, P20– 545. Washington, D.C.: US Census Bureau. Shin, H.B. and Bruno, R. (2003) Language Use and English-Speaking Ability: 2000. Census 2000 Brief, C2KBR– 29. Washington, D.C.: US Census Bureau. Veltman, C. (1983) Language Shift in the United States. Berlin: Mouton. Williams, L. (1980) Phonetic variation as a function of second-language learning. In G.H. Yeni-Komshian, J.F. Kavanagh, and C.A. Ferguson (eds) Child Phonology (pp. 185– 215). New York: Academic Press.