The Acquisition of English by Immigrant School Learners in CataloniaAffective Variables and Cross-linguistic Influence

Résumé

Research on the acquisition of English by immigrants is abundant when it takes place in second language (SL) contexts. However, few studies deal with immigrant learners’ acquisition of English as a foreign language (EFL). In the Spanish autonomous community of Catalonia, with a very high rate of immigration, research on language acquisition by immigrant learners has focused on the two official languages, Catalan and Spanish. In contrast to this, the study hereby reported tackles EFL among immigrant secondary-school learners, with a focus on the participants’ affective variables (attitudes and perceived chances of success) and lexical cross-linguistic influence (CLI) in an oral and written narrative in EFL. The participants in our study are immigrant learners of three schools in the province of Barcelona (an area with the highest concentration of immigrants in Catalonia) who have at least two languages apart from English: L2 or L1 Spanish, L2 Catalan and their L1 when it is different from Spanish. The design of the study includes two different methods: a cross-sectional study with 92 learners divided into three groups (First cycle, ages 12-13; Second cycle, ages 14-15; and Post-compulsory education, ages 16-17), and a longitudinal study conducted with some of the cross-sectional participants, divided into two groups, with two collection times over two academic years: from grade 7 to grade 8 (n=10), and from grade 8 to grade 9 (n=14). The cross-sectional study includes a local group of learners in Second cycle for the comparison of their results in affective variables with those in the immigrant sample. The longitudinal study aims at investigating the role of hours of instruction (approximately 200 hours) in affective variables and CLI in EFL. The first Research Question in the study inquires into the participants’ indices of affective variables, which include attitudes and perceived chances of success, their factors and their relationship with achievement in EFL. The results in both the cross-sectional and longitudinal studies reveal that the participants have in general positive attitudes, slightly more favourable than those found in the local participants of the present study as well as in other studies conducted in Catalonia and other regions of Spain (Comajoan & Gomàriz, 2008b; Ibarraran, Lasagabaster & Sierra, 2008; Saravia i Terricabras, 2004; Uribe, Gutiérrez & Madrid, 2008). As for the immigrant participants' perceived chances of success, they reported having an expected school mark in English from Pass to Good, slightly inferior to those among their local peers. Similar findings have been found by previous research in Catalonia (Comajoan & Gomàriz, 2008a). The results on our immigrant participants’ perceived ability according to L1 showed that they did not believe in general that having an L1 different from Spanish and L2 Catalan had an influence on success in EFL. In the light of previous research, we assumed that affective variables led to higher indices of attainment in the target language through motivation. However, only two of the affective variables included in the study correlated with achievement in the three cross-sectional groups: attitudes towards learning English and expected school mark in English. Neither integrative attitudes (i.e. attitudes towards the target language community) nor instrumental attitudes correlated with indices of achievement. The lower scores obtained by integrative attitudes were coherent with the findings in Bernaus, Masgoret, Gardner & Reyes (2004), while the lack of correlations between integrative attitudes and achievement contradicts previous studies conducted by Gardner and associate researchers in Catalonia (Bernaus & Gardner, 2008; Gardner, 2007) and other national contexts (Clément, 1980; Gardner, 1960, 2000, 2007; Gardner, Day & MacIntyre, 1992; Gardner & Lambert, 1959, 1972; Gardner, Tremblay & Masgoret, 1997; Tremblay & Gardner, 1995). The comparison of the indices of affective variables between the three groups and the results cast by the longitudinal study indicate that the influence of school level and hours of instruction, which comprise increasing age and proficiency in EFL, on affective variables in our sample was weak. This is explained by the fact that the attitudinal constructs in the present study are more stable than other factors that depend on classroom variables, as was found in previous research (Gardner, Masgoret, Tennant & Mihic, 2004; Kuhlemeier, Van den Bergh & Melse, 1996; Lamb, 2007). It is argued here that social attitudes seem nonetheless more sensible to the influence of school level and hours of instruction than instrumentality or expected school mark in English and that the former can be influenced through increasing amount of exposure to the target language culture and the views exposed by the adults in the learners’ environment. The second Research Question of our study addresses amount of CLI and source selection in the participants’ oral and written production in English in addition to their factors. The results of the cross-sectional study show that school level is associated with less CLI in our sample, especially in the number of borrowings, which is in line with previous studies in the Catalan context (Celaya, 2006; Celaya & Ruiz de Zarobe, 2010; Celaya & Torras, 2001; Navés, Miralpeix & Celaya, 2005). Despite the fact that school level involves higher achievement, the statistical analysis showed that attainment in EFL was related to lower amount of CLI only in Second cycle, while type of CLI did not hold any consistent correlations with achievement. What is more, the role of hours of instruction in the longitudinal study, which is also related to higher achievement in EFL, was weak, as it only had a statistical significant effect on the decrease of amount of CLI in the oral production of the group followed from grade 8 to grade 9. Finally, the results in the study suggest that high indices of affective variables can lead to lower amounts of CLI but the comparison with the results in a previous piece of research (Agustín Llach & Fernández Fontecha, 2009) suggests that their effect seems to interact with school level and task effect. The findings on the participants’ use of other languages in EFL point to Spanish as, in general, the preferred source language in the written and oral data. Catalan was selected with lower frequency and no transfer from a mother tongue different from Spanish was found. In the choice of Spanish for instrumental roles (i. e. to address the interlocutor or to make comments on the task), source selection was determined mainly by the association of the interlocutor with the source language. In the use of transferred items to fill gaps in the narrative in EFL, the factors of source selection were different according to the participants’ L1. Among the native speakers of Spanish, the selection of this language was largely due to the role of the L1. Whereas among the rest of participants, transfer from Spanish and Catalan in detriment of the L1 was due to a combination of recency, proficiency and exposure to the L2, language distance and L2-status, which confirms a previous piece of research in Catalonia (Ciruela Castillo, 2007). Finally, the superiority of Spanish over Catalan in both L1 and L2 Spanish participants was largely due to the overwhelming presence of the former in the participants’ sociolinguistic environment, as in Ciruela Castillo (2007).