3.1 Classroom discourse

Classroom interaction in CLIL contexts appears to be one of the more salient topics of research. Escobar Urmeneta (Reference Escobar Urmeneta2009), in her work on the importance of teacher scaffolding in CLIL environments, concludes that linguistic support is necessary for students to be able to understand teachers’ expectations. The author suggests that precise teacher discourse should be both contextually and linguistically meaningful for the learners, especially in the CLIL context. Teacher input is also one of the focal points for Vallbona González's (Reference Vallbona Gónzalez, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) study. Based on results of tests administered to fifth and sixth graders in Catalonia, with various sections and exercises including listening comprehension, dictation, cloze test, and writing, the author concludes that students’ increased exposure to the target language incurred in CLIL classrooms is as relevant to the test results as quality of input, language modelling and teaching strategies.

Both teacher and students are considered in the study by P. Moore (Reference Moore2007). This author reflects on meaning-making in the CLIL classroom, especially in the organization of discourse, the signalling of comprehension and the expression of power or solidarity through discursive interaction. Building from this analysis, the author suggests rich and varied language use should be proactively modelled and promoted. Similarly, Bentley provides an outline of how teachers can empower students in CLIL classes to use the target language ‘beyond the conversational level of basic daily transactions’ (Bentley Reference Bentley2007: 138).

Student discourse is the focus of other studies although it varies as to which aspect of language learning in the CLIL classroom is highlighted. E. Moore & Nussbaum (Reference Moore, Nussbaum, Escobar Urmeneta and Nussbaum2011) apply multimodal qualitative analysis as an integrated means of exploring the interaction that takes place in CLIL situations. They present rich descriptions of the participants’ orientation of action, interaction and use of cognitive and linguistic resources to provide insight into effective CLIL interaction. Evnitskaya & Aceros (Reference Evnitskaya and Aceros2008) use a conversation analysis approach to look at pair work interaction. They find that the adoption and alternation of discursive roles such as ‘language expert’ and ‘non-language expert’ has a positive impact on students’ performances, especially if these roles are not imposed by authority (the teacher), but negotiated within the pair work. Consequently, the authors conclude that this type of interaction in CLIL contexts should be encouraged, since it is advantageous to learning both language and content.

Fuentes & Hernández (Reference Fuentes and Hernández2011) also consider the role of collaboration and expert roles in group work within CLIL environments. The authors analyse the interaction carried out by students of differing levels of competence in the target language (English) in a secondary education chemistry class. They first track the students’ peer-scaffolding and their contributions to the collaborative group work, and then compare those episodes with the teachers’ previous expectations of the students’ capabilities and participation. Interestingly, the results indicate that contributions made by students with fewer competences were significantly more abundant and of higher linguistic quality (that is to say communicatively comprehensible to others) than had been expected by the teachers.

Gallart & Planas (Reference Gallart, Planas, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) use categories from Mercer's (Reference Mercer2004) framework for interpreting communicative acts to categorize the use of student code-switching during a maths class in English. The authors conclude that, in CLIL contexts such as the one in the study, the learners’ use of code-switching facilitates ‘exploratory talk’ and ‘dialogic and interactive communication’ (p. 38). Similarly, E. Moore (Reference Moore, Guasch and Milian2010) examines the use of English as the vehicular language for the teaching and learning processes in CLIL classes at a university in Catalonia. The author is particularly interested in exploring why and when multilingual practices emerge within this context and how these link to the learning process. She concludes that the interactional format influences the emergence of code-switching and multilingual practices, these being more prevalent in group work than in teacher-fronted activities.

Dalton-Puffer (Reference Dalton-Puffer, Escobar Urmeneta and Nussbaum2011) uses the model proposed by Zydatiß (Reference Zydatiß and Vockrodt-Scholz2007) as a means of delving into the complexity of classroom interaction in a CLIL context. Taking examples from a secondary history class on World War I, she examines data to look at meaning structures, and micro- and macro-structures of explanations, and comes to the conclusion that ‘explanations employ a wide range of semantic relations’ (p. 136) although there is also evidence that the students lack the procedural knowledge to formulate explicit explanations.

3.2 Focus on content and/or language

Given that the CLIL approach emphasises integration of content and language, it is perhaps not surprising that the dichotomy of focus between the two occurs in a significant part of the published research. In their study on students’ gains in the target language in secondary science CLIL classes, Escobar Urmeneta & Sánchez Sola (Reference Escobar Urmeneta and Sánchez Sola2009) use a ‘pre-test/treatment/post-test research design’ (p. 65) to determine the ‘efficacy of the [CLIL] sequence in promoting the simultaneous learning of language and science’ (p. 69), in particular fluency and lexical repertoire of the teaching unit. Their data were collected from six classes with a total of 217 students, stemming from multiple choice tests and written production. The software programmes Advanced Text Analyser and P_lex were used to obtain t-test results in the following categories: lexical recognition; total number of unique words; type-token ratio; number of sentences per text produced; total number of words produced by informant; and lexical sophistication. The authors posit that despite an apparent lack of progress in some of the categories, there is significant improvement in students’ overall fluency and linguistic repertoire.

Looking at language acquisition in CLIL, and as part of a study aimed at providing support for secondary school CLIL teachers in the area of social studies, Llinares & Whittaker (Reference Llinares and Whittaker2007) apply both a rhetorical and linguistic framework to data that were collected from two Madrid-based schools. Following the theoretical lines of Martin, Matthiessen & Painter (Reference Martin, Matthiessen and Painter1997), and Halliday (Reference Halliday and Matthiessen2004), the spoken data were analysed for processes, circumstances, clause complexes and modality. The authors come to the conclusion that the CLIL students did acquire some of the linguistic features of the discipline (social studies).

In a comparison of two classes (a ‘traditional’ language learning class and a CLIL class), Ruiz de Zarobe (Reference Ruiz de Zarobe, Monroy and Sánchez2008) finds that the CLIL students obtained slightly better results in content, language organization, vocabulary, language use and mechanics of the target language, although the advantage was most significant for the latter category. The author finds that the amount of exposure to and use of the target language is pivotal to success in language learning in CLIL environments, especially for written communication.

Labajos Miguel & Martín Rojo (Reference Labajos Miquel, Martín Rojo, Muñoz, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) compare two language immersion classes: one designed for newly arrived students in the Madrid area who do not speak Spanish and one aimed at introducing bilingual education (Spanish-English) in primary education in the same geographical area. Taking a critical sociolinguistic ethnography focus, the authors look at three types of data: (1) teaching materials, (2) institutional documentation, and (3) participants’ representations of the activities. The authors conclude that there was a relative balance between content and language pedagogical focus but that these results must always be considered within the social and economic parameters in which the CLIL classes took place.

3.3 Teacher roles and teacher education

The studies mentioned previously have inevitably overlapped with areas in which teachers need specialized training in order to meet the challenges of CLIL contexts. Focusing on teacher discourse, Clegg (Reference Clegg2007) analyses the language used by educators in CLIL classrooms in order to determine moments when greater language support is needed and how this can be provided by the teacher (e.g. through specific teacher discourse). The analysis includes looking at specific genre types that might be required in texts used in CLIL. The author argues that this analysis is essential for CLIL teacher preparation. Clegg reminds the readers of the importance of reflection on text genres, especially which types can best facilitate learning in these environments.

Following Halliday's ideational function (Reference Halliday1970), Dafouz (Reference Dafouz2007) analyses teacher talk in a university-level engineering classroom. Her findings point to the vital role of the instructor's discursive features such as extensive use of pronouns and avoidance of modal forms (in order to avoid discourse complexity) in the construction of knowledge in a CLIL classroom. This issue is central to the work of Sainz Osinaga et al. (Reference Sainz Osinaga, Ozaeta, Garro, Perez, Egizabal, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) as well, although their study is based in primary education. The authors analyse the discourse function of teachers’ instructions in science and social science CLIL environments and find that the teachers’ discourse focuses mostly on curricular content rather than on language. This shortfall of focus on metalinguistic aspects is seen as a disadvantage to the language learning process. The need for more metalinguistic reflection in CLIL classrooms is also highlighted by Pérez-Vidal's (Reference Pérez-Vidal2007) analysis of teachers’ input in CLIL classes. She concludes that in the classes in the study described in her article, there was not enough focus on form (FoF) in the discourse employed by the teacher.

Complementary to this, Whittaker & Llinares (Reference Whittaker, Llinares, Escobar Urmeneta and Nussbaum2011) find that the sequencing of tasks in CLIL classrooms can have significant impact on student performance. These authors apply a systemic functional approach to analyse the discourse produced by students during class discussions, written texts and interviews. Their results imply that the teacher's planning and the way in which a CLIL teaching sequence is presented has a significant impact on the learning process.

Along similar lines, Tapias Nadales (Reference Tapias Nadales2011) uses an ethnographic approach to explore group-work dynamics in a science classroom, based on multi-modal data of student-teacher interaction, student-student interaction and written student output. The author examines differences between task (teacher designed materials) and activity (what students do to carry out the proposed tasks communicatively) in order to understand how the acquisition of content and language merge in CLIL learning events. Tapias Nadales argues that the student interpretation of the goals of the group work often differs from the teacher's instruction and that in this particular case there was deviation from the teacher's task objectives (in their reports on the history of atoms, the students’ focal point was more on artistic output – how to represent atoms graphically – than on language usage). This leads the author to propose that in CLIL classrooms the need for more precise instruction in order to bring the students’ attention to both content and language is particularly important.

A key teacher competence for successful CLIL implementation, according to published research results, is the capacity to collaborate across disciplines. For example, Horillo Godino (Reference Horillo Godino, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) examines fragments of discourse between two teachers (one content and one language) in order to discern key ‘points for discussion’ between experts ‘who have to engage in a process of cross-curricular collaboration to develop CLIL teaching sequences’ (p. 42). Similarly, Canet Pladevall & Evnitskaya (Reference Canet Pladevall, Evnitskaya, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) examine the collaboration and reflection of teacher planning when preparing CLIL materials and implementation. Collaboration is also a key theme in the article by Sagasta et al. (Reference Sagasta, Pedrosa, Madinabeitia, Nazabal, Barnes, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011) although its focus is on collaboration between student-teachers and teacher trainers in a CLIL teacher education course.

As CLIL classes become more mainstream, action research by teachers involved in this approach is becoming more common place. This is evidenced by several of the aforementioned articles and in work by Arnau (Reference Arnau, Escobar, Evnitskaya, Moore and Patiño2011) and Pallarés Monge (Reference Pallarés Monge2011). Arnau's study is applied to social studies teachers in Catalonia who are working within immersion classrooms for immigrants. The author reflects on the triangulation between a support teacher, a regular classroom teacher and a teacher in ‘reception classrooms’ for immigrants, and surmises that these circumstances (triangulated collaboration between different education agents) ensure more effective pedagogical interventions for CLIL environments.

Taking a longitudinal, qualitative approach, Pallarés Monge's study explores a CLIL maths programme implemented in a secondary school in Barcelona. After outlining the tasks and materials used for the CLIL sequence, the author analyses the participation and organization roles of students during cooperative group work. He emphasizes the ‘natural way’ in which the students organize themselves to complete the task and how they willingly engage in the target language to complete the group work.

Similar research by CLIL teachers in their classrooms examines the need for teachers to create learner-centred environments for effective CLIL implementation. In her research, Eixarch (Reference Eixarch2011) studies three CLIL teaching units for science in English in a secondary institution. She examines the use of a portfolio as a tool for teacher reflection, the feedback between teacher and student as an instrument for self-regulation, and self-regulation activities in group work. The author concludes that in order for the CLIL project to be more learner-centred, teachers must involve the students in the assessment process and this requires previous training on self-assessment. Cots & Clemente (Reference Cots, Clemente, Escobar Urmeneta and Nussbaum2011), looking at higher education CLIL courses, obtain similar results. They deduce that pedagogical changes are necessary in order to ensure efficient learner-centred approaches that will help students to ‘deal effectively with both content and language’ (p. 182).

Pena Díaz & Porto Requejo (Reference Pena Díaz and Porto Requejo2008) administered two sets of questionnaires to CLIL teachers to determine some of their conceptions and anxieties about this approach to language teaching. The results indicate that, on the whole, the teachers were ‘extremely positive’ about the experience (p. 157), in particular because of the increase in number of FL class hours and the quick adaptation to the process by their students. However, the teachers expressed some concerns about the level needed in the target language to carry out these classes, the extra work entailed and a lack of methodological knowledge. When asked about the role of the first language (L1) in second language (L2) learning, questionnaire results indicated that the teachers’ beliefs were contradictory. Seventy per cent of the cohort felt that this type of language programme can interfere with the development of L1, even though they were participating in a CLIL experience. The authors underscore the fact that despite these misgivings the teachers are highly motivated to continue participating in the project.

Along similar lines, Fernández Fernández et al. (Reference Fernández Carballo-Calero2005) carried out a study into soon-to-be-CLIL teachers’ attitudes about the upcoming perspective of teaching through this approach. The researchers administered questionnaires and follow-up interviews of teachers from 25 schools in the Madrid area. The study principally looked at the teachers’ motivation, their opinions about preparatory courses for teaching through CLIL and their expectations of what CLIL teaching entails (this study focuses on a first phase of preparing teachers, not on actual implementation). Results of the research indicate that the teachers were generally enthusiastic about the opportunity to teach through CLIL, in large part due to the self-perception of being innovative, or ‘pioneers’. Their enthusiasm was only slightly tempered by a perceived lack of information and planning on a more administrative level.

Toledo, Rubio & Hermosín (Reference Toledo, Rubio and Hermosín2012) administered a 45-question Likert-scale survey to explore the perceptions of students enrolled in a CLIL higher education course. They address the learners’ level of satisfaction, academic performance, improvement in the target language, difficulty and effort of participating in a CLIL course, and general assessment of the experience. While the majority of the students indicated that they were aware of the learning opportunity and felt positive about the process, those who had enrolled for reasons other than personal motivation to take part in a CLIL course had more negative attitudes.

Using questionnaires and focus group discussions, Feixas et al. (Reference Feixas, Codó, Couso, Espinet, Masats, Vila, Blasco, Cano, Gilar, Lledó and Mañas2009), carried out a study on student and teachers’ perceptions of CLIL teaching practices in a teacher education programme. One of the principal points that emerged concerning the student perspective is that the participants did not consider the use of English as an impediment to the learning of the content, despite the inherent difficulties of using an L2 as the vehicular language for communication in the classroom. The students highlighted the added challenge of expressing themselves correctly in writing and speaking in the target language but at the same time indicated that this helped them probe related vocabulary in more depth. On the other hand, in the questionnaires and focus group discussions, the teachers emphasized the lack of teaching materials with a specific focus on the lexicon needed for the subject, bringing out an apparent contradiction between students’ perceptions of communicative gains versus the teachers’ focus on difficulties related specifically to structural language features.

3.4 Reflections

In the overview of research concerning CLIL in Spain, there does not appear to be any clear-cut core area of investigation. This may be due, in part, to the fact that while the application of the approach appears to be extensively accepted worldwide, definitions of what constitute CLIL differ widely. Barwell (Reference Barwell2005: 143) defines the approach as ‘the teaching and learning of both language and subject areas (. . .) in the same classroom, at the same time’, placing emphasis on a balanced integration of instruction of both content and language. Dalton-Puffer (Reference Dalton-Puffer2007: 1) emphasizes the language as medium, rather than goal of the instruction when she states that CLIL ‘refers to educational settings where a language other than the students’ mother tongue is used as a medium of instruction’. Mehisto, Marsh & Frigols (Reference Mehisto, Marsh and Frigols2008: 12) see CLIL as ‘an umbrella term covering a dozen or more educational approaches’.

Some authors have even argued that CLIL is a variable of content-based language teaching (CBLT), or immersion teaching (see Netten & Germain Reference Netten and Germain2004, Reference Netten, Germain, Dooly and Eastment2008, Reference Netten and Germain2009) and therefore is not really a new concept (Wesche Reference Wesche and Kaplan2010). It can be argued – rather tongue-in-cheek – that CLIL was even practised by ancient Roman upper-middle classes, who preferred to have their children educated in Greek. Under a wide definition of CLIL, philology courses at most Spanish universities have used this approach for decades.

Under the broad use of the term, some research publications apply the label of CLIL to English-mediated language classes or English for specific purposes (ESP) courses (e.g. Oliva Girbau Reference Oliva Girbau, Escobar Urmeneta, Evnitskaya, Moore and Patiño2011, González Rodríguez & Borham Puyal Reference González Rodríguez and Borham Puyal2012). Arguably, investigation into learning environments that have historically used a foreign language as the means of communication, but which are now labelled CLIL, distort the picture of the gains accrued through these contexts.

CLIL research in the Spanish context, at times, seems to encroach on topics which traditionally have fallen within long-standing disciplines in the fields of SLA or AL: code-switching, focus on form, teacher talk, teacher education, for example. This makes it difficult to differentiate between the results that have emerged from enquiry within these recognized fields of study and investigation into CLIL outcomes. There are ample examples in SLA and AL that discuss the relevance of communicative approaches in the classroom, that argue for the need of significant and calibrated discursive input, thereby supporting findings that discursive teacher scaffolding is essential for student learning and so forth. This lack of referencing to similar studies in SLA and AL implies a dichotomy between research in CLIL and research in language teaching in general. This delineation should be interrogated more critically as many of the outcomes of the studies seem to beg the question of how and why these results should be considered as different from good teaching practices in any field (language learning or content learning). It is recommended that researchers try to address what makes CLIL innovative and how it differs, if at all, from other FL teaching and learning.

Lastly, in the CLIL research published in Spain, there appears to be a predominance of holistic studies based on single events or classes (case studies, ethnographic studies, phenomenological studies) – many of which are based on explanatory or descriptive analyses. Indubitably, investigation into specific phenomena within a real-life context is of great value to education research, nonetheless CLIL research in Spain has reached a point where aggregate knowledge would be of great benefit to the current state-of-the-art. This would require large-scale, longitudinal, contrastive studies of results stemming from research thus far.

4.1 Focus on learner (early language learning and the age factor)

Unlike many other European countries, Spanish children begin learning a foreign language in formal settings at a very early age, usually around six (but in some regions, as young as three years old). Not inconsequently, there are established research groups and investigators of international renown whose area of expertise lies in language learning and the age factor. It is important to note that international studies in SLA focusing on the age factor cover a wider scope than research regarding young learners; however, YLL is a recognized feature of these studies. In the research covered here, the age factor is integrated into the field of early language learning.

Research into the impact of age on language learning has been the focus of several studies in Spain during the last decade (see Roca de Larios & Manchón Ruiz Reference Roca de Larios and Manchón Ruiz2006 for an overview of national and international studies on the age factor and language learning). Muñoz et al. (Reference Muñoz, Pérez-Vidal, Celaya, Navés, Torras, Tragant and Victori2003) argue that the most effective means of introducing a foreign language to young learners is through content immersion (CLIL). Cenoz (Reference Cenoz2003) looks at the age factor within multilingual contexts and finds that there are no negative effects of exposure to more than one new language for young language learners, even though it requires long-term planning and an appropriate teaching approach.

In a cross-country comparative study, Muñoz, Tragant & Torras (Reference Muñoz, Tragant and Torras2010) consider the incipient oral productions of 243 second-grade pupils from seven different European contexts, including Spain. The study is part of a larger European project entitled ‘Early Language Learning in Europe (ELLiE)’. Six schools in each of the seven countries were chosen for a longitudinal study of six students in each school (high-, medium- and low-level learners, as selected by the teachers). Following an interview and a vocabulary test, the students took part in a role-play task in the target language. Transcriptions of the language production were then analysed for total number of nouns and verbs and ratios of noun and verb together in order to measure pupils’ lexical and grammatical development. The results of the study show that young learners use nouns more often than verbs and that they make some use of memorized formulaic target language, although this may be incomplete (e.g. ‘where toilet?’). Their data also demonstrate that seven- and eight-year-old students are able to distinguish between L1 and FL, which helps them produce more complex output and form hypotheses of the meaning of unknown target language.

In a longitudinal study of nine beginner learners from different age groups, Alonso-Vázquez (Reference Alonso-Vázquez2004) compares the speed of each individual's learning process of negation in EFL. The participants were interviewed and recorded monthly for nine months and transcripts were analysed for incidences of negation use. To minimize the effects of learning ‘fluctuations’ the author analysed the data for ‘consolidated learning’ (p. 14). The indicators used were incidences of negation using ‘no + V’ ‘don't V’ and ‘aux. neg. form’. Equations stemming from graphic representation of the negations allowed comparative analysis of the ‘evolution of the learning process of the different participants’ (p.14). The author found ‘no detectable relationship between age and efficiency of the learning process’ (p. 25). Based on the homogeneous learning process of each individual, the author argues that other factors are more important than age.

In a comparative study of native (English) and non-native young language learners, Moya Guijarro & Jiménez Puado (Reference Moya Guijarro and Jiménez Puado2004a, Reference Moya Guijarro and Jiménez Puado2004b) look at the development of interlanguage (Selinker Reference Selinker1972) of 50 girls and 35 boys with Spanish as their L1. The students received approximately ten hours of English classes per week. They were video-recorded periodically from the age of three till the age of eight. The transcriptions were analysed for morphological aspects such as use of –ing, plural form with –s, Saxon genitive, third person singular in present tense, irregular plural forms, regular and irregular past forms, forms of negation (no, not, none) and auxiliary forms. Percentages of correct and incorrect use of these forms were then calculated. These results were compared with studies of language development in learners of English as an L1. The authors conclude that there is an apparent similarity in the process of language acquisition of L1 and L2 in morphological aspects, even though the development of the FL in the Spanish context is lengthier than the native speaker's development of the same morphological features. Both groups quickly develop the progressive form and –s inflection for plural nouns, while the –s ending for third person takes longer to learn. According to the authors, this becomes an endemic problem, above all for the Spanish students learning English as a foreign language. There are noticeable differences between the two groups in the acquisition of negative forms. The authors propose that this may be due to the more complex system of negation in English, resulting in more difficulties for the Spanish learner of English. They emphasize the fact that the incorrect use of the target language should not be seen as a ‘failure’ since it is an important part of the learner's development of their interlanguage.

Looking at YLL in multilingual settings, Perales (Reference Perales2006) examines language acquisition of English as a third language by three groups of young bilingual language learners (four-, eight-, and eleven-year-olds who speak Spanish and Basque). In the study, the author analyses differences in development of the use of negation in English and finds that the two older groups outperformed the youngest learners. He then suggests the need for further research to determine whether these results are due more to the type of instruction rather than age differences, given that the younger group had not received the same type of language instruction input as the other groups.

Coyle (Reference Coyle2005) interrogates the widespread belief that children have a natural capacity for acquiring a foreign language, coming to the conclusion that young learners’ attention needs to be selectively directed to formal elements of the linguistic system of the target language. Based on the model proposed by Ellis (Reference Ellis1997), the author suggests that focused attention is important for learners of all ages – as well as being a key element for better teaching practice.

Specific characteristics of young learners come into play in the studies carried out by Agustín Llach & Barreras Gómez (Reference Agustín Llach and Barreras Gómez2007, Reference Agustín Llach and Barreras Gómez2009). These authors examine children's features in relation to the process they follow to acquire vocabulary in the foreign language. In both studies the authors find that the principal lexical errors are misspellings, omissions, borrowings and substitutions. They claim that information on recurrent error-making of YLLs provides teachers with important information about areas that may need reinforcement. Additionally, they also examine the main semantic fields of vocabulary produced by the young learners. These are leisure and games, followed by the lexicon related to school, home and family. The authors propose that having this information will help teachers make decisions about which content to include in their language teaching.

Learner characteristics, specifically gender, underscore a study by Jiménez Catalán & Terrazas Gallego (2005–Reference Jiménez Catalán and Terrazas Gallego2008), who investigate the receptive vocabulary acquisition in EFL of 270 fourth graders (L1 Spanish). Using a vocabulary levels test (VLT), the authors found that, generally, the overall receptive vocabulary of the pupils in the study was lower than 1,000 words, with a very low percentage of fourth graders within the 2,000 frequency band. They found a possible correlation between VLT scores and cloze test results, suggesting a relationship between vocabulary knowledge and L2 proficiency. At the same time, the study indicates that there was no significant difference in results between genders. The authors contend that this type of research can foment informed teacher decisions about when to introduce and how to use vocabulary according to learner levels although they also suggest that there is a need for more longitudinal research, carried out with this same group, in order to find out more about their receptive vocabulary development.

Jiménez Catalán co-authored another study (Ojeda Alba & Jiménez Catalán Reference Ojeda Alba and Jiménez Catalán2007) on gender differences in the acquisition and development of lexical competence of ten-year-old language learners. The study is based on the vocabulary used in English compositions by 119 girls and 152 boys. The analysis indicated that there were ‘80.16 word tokens and 37.18 word types for males and 88.49 word tokens and 41.22 word types per composition for females’ (p. 162). This confirms the authors’ hypothesis that ‘girls use a wider vocabulary than boys’ (p. 169).

4.2 Focus on pedagogical issues (teacher education, approaches and resources)

Research into YLL teacher education or practice is sparse. There are exceptions: for instance, Rodríguez Martín (Reference Rodríguez Martín2011) provides a case study of her teaching experience when applying a communicative approach with young learners. After analysing her own practice, the author postulates that it is possible to introduce language learning, through play, to children as young as three years of age. Also applying action research to her own context, Piquer Vives (Reference Piquer Vives2006) analyses questionnaires of parents, a teacher diary and video-recordings of interactions of different age groups (three-, four- and five-year-olds) receiving input through English. The author finds that the parents’ approval rate of this type of experience for their children is very high for the age-three group. She correlates these results with this group's higher rate of language improvement and enthusiasm for experimenting with the target language. Piquer Vives proposes that it is possible to introduce English to students as young as three as long as teachers use attractive and motivating strategies that include the reproduction of short, spontaneous and meaningful dialogue and the elimination of all activities related to formal language teaching.

Even though explicit research into YLL-focused teacher education is thin, there are studies in the Spanish educational context that can contribute to teacher training. Barreras Gómez (Reference Barreras Gómez2010) explains how the use of stories in EFL classrooms with young learners can scaffold the process of acquiring vocabulary in a contextualized, meaningful way, and then goes on to provide the theoretical background for the propositions made. The author argues that this approach not only promotes better language learning among young learners but ensures their enthusiasm for learning and stimulates their autonomy. In line with Piquer Vives's results, Vilà (Reference Vilà2006) argues that the basis for beginning FL instruction with students as young as three has been shown to be effective, as long as the teaching focuses on fun, motivating activities aimed principally at input and exposure. At the same time, there should be little pressure for language production at this stage.

Teaching approaches is another strand of research into YLL in Spain. Durán Fernández (Reference Durán Fernández2005) proposes that the implementation of a TBLT approach in primary education is more effective than traditional (Presentation, Practice and Production (PPP)) interventions in EFL classes. With a cohort of 62 students (33 students in the control group using PPP; 29 in the experimental group using TBLT), the author outlines a pre- and post-test study design. Results of the pre-test showed insignificant differences between the two groups in most of the areas of study, although the post-test showed an increase in reading comprehension for the experimental group. The author relates this to the fact that the experimental group was involved in intensive individual reading tasks.

Some studies into YLL examine specific resources available for the distinctive profile of the young learner. Jiménez Catalán & Mancebo Francisco (Reference Jiménez Catalán and Mancebo Francisco2008) carry out a quantitative analysis of the vocabulary input found in a corpus of four EFL textbooks from the last year of Spanish primary and secondary education. They survey the 50 most frequent words in each textbook to determine word type, word frequency, and frequency rank in order to assess the increase in the number of types and tokens at the end of each educational stage. They also explore the implications of word repetition in EFL textbooks for the acquisition of vocabulary. The study indicates a lack of commonality between the vocabulary focus of the textbooks. Subsequently, the authors argue that this inconsistency of exposure can affect the language input young learners are exposed to in formalized language learning settings.

Along similar lines, exploring text frequency, tokens, types and families, Alcaraz Mármol (Reference Alcaraz Mármol2011) applies a quantitative analysis of two textbooks which are widely used for young learners in EFL classrooms in Spanish schools. The author finds that one of the books is ‘lexically richer both in quantitative (number of tokens) as well as in qualitative (number of types and lemmas) terms’ (p. 14). The author does not claim that this means one course book is better than the other, or that the learners will necessarily receive more input or acquire more vocabulary. However, she suggests that teachers can carry out similar analyses for making an appropriate selection of their course book for young learners.

4.3 Reflections

During the process of compiling research for this review, several early studies into YLL (before 2003) were found, yet it appears that interest in advancing beyond the results of these studies is lagging. Moreover, the scarcity of research in Spain into teacher education for YLLs is worrying, given that these learners constitute a group of learners with specific needs and characteristics, one of the most important being that they are in key cognitive development stages. Teaching FL to a group of pupils who are in the process of acquiring their L1 (or in the case of bilingual regions, two L1s) requires specific methodology training.

In this sense, research interests and political interests do not always seem to converge. Research published in Spain which deals with teaching YLL tends towards small case studies of strategies or individual approaches or analysis of teaching materials. Moreover, as mentioned above, the volume of studies in this area is considerably lower than in the other areas. This sharply contrasts with the growing political impetus for YLL approaches in public education and highlights the need for more research into teaching and learning in this area and into specific teacher education for this profile.

As previously mentioned, currently, in several regions in Spain, FLs are introduced into formal education at younger and younger ages, sometimes as early as three. In many cases, these young students are taught by primary education teachers who are experts in teaching EFL but not in early childhood education (these teachers ‘cross borders’ between early childhood and primary). This is because national education laws stipulate that FLs must be taught by specialists.Footnote ⁴ However, the same law indicates that teachers specialized in early childhood education can teach FLs to young learners, if they can certify a B2 level of the Common European Framework for Languages (CEFR) (European Council 2001) in the target language – without any need for completing TEFL courses. This situation – that language specialists do not necessarily have the knowledge to deal with young learner needs and that the early childhood specialists are not trained in teaching FLs and do not need to certify a high command of English – attests to the need for more research into YLL teacher education.

A considerable amount of the research concerning YLL in Spain has focused on the ‘age factor’ although, as Perales points out (Reference Perales2006), it is still not clear whether the age factor or type of instructional input has more of an impact on the results of language learning. Likewise, it can be argued that older students may be more familiar, not only with the type of instructional input but also with the means of assessment used to gather the data in the studies, and this should be taken into consideration.

Finally, future research into YLL should bear in mind that there are many variables that come into play when a child learns FL, including but not limited to attitudes, strategies and motivations, amount of exposure to the target language, and family and school environments. Research focusing on only one variable (e.g. age) in language development should endeavour to take into account other contextualized factors in the language learning process. It would be beneficial to current and future language teachers to have access to more studies dealing with type, quality and amount of young learner interaction in the target language and the nature of tasks used in situations dealing with YLL.

5.1 Focus on tools

A large amount of the published research seems to be focused on specific tools or modes of communication and how they are used to teach FL/L2. Interestingly, the pedagogical use of wikis in the language classroom appears to have been particularly popular among Spanish researchers in the past few years. Díez-Bedmar & Pérez-Paredes (Reference Díez-Bedmar and Pérez-Paredes2009) piloted an online collaborative activity for the course entitled ‘English Grammar’ for pupils studying hospitality management. The wiki texts produced by Spanish students at the universities of Jaén and Murcia were then corrected by students at Leeds University (UK). The aim of the pilot project was to allow Spanish students of English to apply their writing and reading skills, complement their target language vocabulary while writing descriptions related to their course subject (for example, monuments, historical or artistic highlights,) and learn to become critical consumers of the target language. The study showed that, despite monitoring, some of the writing phases in the wiki did not function as anticipated due to lack of commitment from members of the working group. The authors outline the minimal conditions necessary for this type of interaction to be successful.

Also looking at the use of wikis, Gallardo Ballestero's (Reference Gallardo Ballestero2011) study, carried out in Andalusia with immigrant learners of Spanish, aimed to analyse the educational possibilities of the tool. The author identifies specific features of wikis which make them an appropriate tool for teaching L2 Spanish to immigrants. Similar practice was carried out by Franco González in Reference Franco González2012 in which immigrant students of L2 Spanish were engaged in collaborative writing through a task-based teaching approach.

The use of a virtual learning environment (VLE), more specifically the forum available in the VLE, is the focus of Rodríguez-Juárez & Oxbrow's (Reference Rodríguez-Juárez and Oxbrow2010) study on the correlation between learners’ English proficiency level, their type of motivation, and their participation in online communication. The authors first administered an initial online proficiency test to 39 university students enrolled in an English philology language and literature course at the Universidad de Las Palmas de Gran Canaria. They followed this with two Likert-scale questionnaires: one on general motivation for target language learning and one on interest in use of the VLE (Moodle). When calculating the results they also took into consideration students’ amount of participation in different types of virtual forums (open and private). As a measure of participation, the authors looked at frequency ratings which are available to teachers as a Moodle feature. The authors conclude that low levels of participation in the forums are not related to variables such as the students’ proficiency or motivation but to lack of students’ autonomy and teacher guidance. They suggest that participation could have been enhanced if teachers had taken the role of discussants in the forums and if the students’ comments had been evaluated. Furthermore, tracking of participation in the forum showed that a higher percentage of students read the posts but posted responses occasionally and that there was no discernible difference in participation of open versus private forums. Results of the overall study indicate that lower level students were more motivated to take part in the forum discussions. The authors make the proposal that advanced students need to be made aware of the need for extra practice in the target language and that more metacognitive knowledge, teacher feedback and the development of learner autonomy are recommended, since these were found to be the underlying factors for limited participation rather than proficiency level or motivation.

Using corpus-driven data of electronic lexical resources, along with corpora from in-class discussions during ESP courses (computers and business), Curado Fuentes (Reference Curado Fuentes2005) examines learners’ use of online language resources (dictionaries, fill-in-the-gap exercises, matching, concordances). The author looks into lexico-grammatical aspects of the language made available to learners through these resources and which are then reproduced by the learners through online case study discussions in order to evaluate the efficacy of online learning resources. Curado Fuentes concludes that the complementary use of ESP corpora (e.g. created through lexicon specifically used in ESP classes) can be important in the design of online language learning resources, and that these can then be used in tandem with communicatively oriented tasks.

Also looking at available technological resources for Spanish for specific purposes (medical lexicon and language use), Mendoza Puertas (Reference Mendoza Puertas2006) points out that this particular field has lagged behind in the gains made in TELL. In his article, the author underlines the importance technology holds for medical professionals and the subsequent need for integration of technology into the teaching of Spanish as a foreign language or ELE.

5.2 Focus on communicative function

Spanish researchers have used discursive analysis of the communicative strategies employed in academic email writing and Facebook postings – research which can influence indirectly the way genres of language are taught, especially writing (see Pérez-Sabater (Reference Pérez-Sabater2011) for a linguistic analysis of online communication genres). Morán (Reference Morán2008) applies conversation analysis to e-learning interactions. Looking at politeness strategies in potentially ‘threatening’ social online interactions, the author uses data from two different chats in order to observe online collaboration in e-learning through a framework of ‘Communities of Practice’ (Lave & Wenger Reference Lave and Wenger1991). Morán's analysis indicates that ‘new features’ of communication are evident in the online learning domain, providing important insight into specific online communicative competences which can be transferred to language teaching and learning situations.

With the aim of debunking previous assertions that CMC is somewhat ‘impoverished (. . .) in comparison with face-to-face communication’ (p. 73), Pérez-Sabater, Turney & Montero-Fleta (Reference Pérez-Sabater, Turney and Montero-Fleta2008) compare linguistic features of 100 emails written by both native and non-native English speakers. They focus on features such as formality of greetings and farewells, the use of contractions, politeness indicators and non-standard linguistic features. Based on their findings, the authors argue that non-native writers are more formal across all categories, possibly due to insecurity in the use of the target language, but both native and non-native communicators demonstrate awareness of the importance of greetings and farewells. The study also found a difference in the use of contractions between native and non-native writers (the former using contractions more often), indicating some contradiction in the tone of formality between greetings and the use of contractions in non-native email writing. As to politeness indicators, native email writers made more use of ‘semantic politeness indicators’ (p. 81). Pérez-Sabater et al. (Reference Pérez-Sabater, Turney and Montero-Fleta2008) conclude that there are significant parameters defining native and non-native varieties of email discourse. Although this study does not consider language teaching or learning in particular, it can provide input for designing language teaching materials and courses that integrate CMC into the teaching and learning of languages.

The focus on the potential of interaction (communicative events for language learning) is a significant strand of research carried out in the Spanish context (see Jordano de la Torre Reference Jordano de la Torre2011 for a synopsis of this evolution in teaching approaches in Spain). Many of the articles provide overviews of ways in which technology – especially the use of the internet and social media – can play a major role in supporting authentic communication in the language classroom, especially in courses following a communicative teaching approach. Calling on the premise of a socio-constructivist learning paradigm (in particular, basing arguments on the Vygotskian theory of Zone of Proximal Development, see Sotomayor García (Reference Sotomayor García2010)), research into ICT argues that CMC helps reduce learner anxiety and increase learner autonomy (Contreras Izquierdo Reference Contreras Izquierdo2008). Along these lines, Sánchez Tornel & Pérez Paredes (Reference Sánchez Tornel, Pérez Paredes, Hernández, Pérez-Paredes, Aguado Jiménez and Criado Sánchez2008) sustain the hypothesis that online interaction supports socio-constructivist learning episodes.

Research into the use of online collaboration (also known as telecollaboration, online international exchange and online intercultural exchange) indicates that this type of interaction can be important for intercultural learning (Vinagre Reference Vinagre2010). Basing her study on an online exchange carried out between the Department of Applied Languages at Nebrija University in Madrid and Dublin City University, Vinagre first describes the main aim of the project, namely, to improve students’ competences in the target language (English for students from Spain, Spanish for students from Ireland). Using qualitative data taken from the different modalities of online communication, Vinagre applies Byram's (Reference Byram1997) framework of intercultural communicative competences (ICC) to look at evidence of knowledge, attitude, and skills related to ICC. The results indicate that while some gain in ICC could be found in certain aspects of the exchange, the students did not change their own cultural perspectives nor were they able to integrate new perspectives from their peers. The author concludes that, while this type of exchange can be beneficial for intercultural learning, it should be monitored through in-class discussion and reflection.

In their article, Jauregi, Canto & Ros (Reference Jauregi, Canto and Ros2006) outline five written and oral communicative pair- and group-work video-conference tasks completed by a group of Dutch university students who were learning Spanish together with a group of Andalusian M.A. students who were studying to become teachers of ELE. The authors analyse the discourse that emerges in the video-conference sessions to determine how the expert group (the M.A. student-teachers) scaffolds discourse to co-construct the tasks. They then look at how both groups negotiate meaning to discuss intercultural issues or to write a text together. The authors conclude that the video-conference project made it possible to widen the horizons of the class and favour authentic communication in the classroom with beginner language learners (the Dutch students). This view is supported by the students who, when interviewed, were critical about the regular tasks they carried out in their face-to-face classes, suggesting that there should be more communicative tasks such as the video-conference so that they could use the language meaningfully and in an authentic context. Furthermore, according to the authors, the Spanish students also valued this experience positively as it offered them the possibility of experiencing, through active participation, the teaching methodology that had been presented to them in the M.A. teaching course.

5.3 Focus on participants (learners and teachers)

Research publications in Spain that delve into the participant role (versus highlighting the tool or the role of communication) tend towards studies of perceptions and attitudes of those involved in the TELL process. For instance, research into attitudes towards the use of computers as part of the language learning process (Fernández Carballo-Calero Reference Fernández Carballo-Calero2005) indicate that, generally, student attitudes are positive towards the use of TELL, although these may be nuanced. The study shows slightly more negative postures about technology as a support for teacher-student interaction, indicating that in specific situations, learners prefer face-to-face interaction over computer-supported interaction.

In a study carried out with beginner learners of ELE at Texas University, USA, Peart (Reference Peart2011) gathered class observation, student questionnaires, student online output, (from WebCT) and structured and semi-structured interviews with a teacher of ELE to determine if there was a correlation between vocabulary acquisition and amount of time spent working on practice exercises in the VLE (WebCT) and whether this was evident in examination marks. Above all, the author was interested in finding out if there is a ‘gender effect’ linked to learners’ use of WebCT and their results in vocabulary tests. The author found that the amount of time spent practising vocabulary in the online platform does affect the final results of male learners but there is no significant impact on the female students’ performance. She suggests the need for more research into different vocabulary learning strategies between male and female language learners and that this should be taken into account when designing online content and resources.

In their teacher-focused research, Durán Fernández & Barrio Barrio (Reference Durán Fernández and Barrio Barrio2007) collected questionnaire responses from early childhood and primary education teachers of English in 100 schools. The authors find that all of the teachers in the study agree that technology must be integrated into current teaching and learning processes. The questionnaire answers showed that all of the participants in the study had had at least minimal experience with the use of technology in their teaching, but that only 27% of them used computers ‘creatively’ in their everyday teaching practice. Given these apparent contradictions, Durán Fernández & Barrio Barrio conclude that there is a need for more training for teachers in how to use technology in their classrooms.

Ramírez Verdugo & Alonso Belmonte (Reference Ramírez Verdugo and Alonso Belmonte2007) carried out research on teachers’ expectations and preconceptions about the use of the internet. Questionnaires were administered to six Spanish EFL primary teachers before and after taking part in an innovative pedagogical project aimed at complementing textbook-based instruction with the use of internet-based materials (digital stories, songs and games). The authors compare the results of the questionnaires and conclude that the teachers felt the technological materials offered their learners the opportunity for improving their listening skills (ability in understanding the gist of real oral messages) and also had a positive effect on their pronunciation. Nonetheless, the teachers felt that there was a lack of access to appropriate materials (some were too simple, others too difficult). The teachers in the study indicated that there was a need to prepare their learners to work more autonomously from the beginning. The findings lead the authors to call for more and better pre- and in-service teacher training in order to help teachers to (a) fully integrate the use of technology into their language teaching practice, (b) develop the critical skills necessary for using technology effectively, and (c) know how to aptly assess the results of ICT use.

Many of the conclusions drawn from research into ICT and teacher education support Area Moreira's (Reference Area Moreira2008) call for a ‘pedagogical innovation’ so that technology is embedded into the teaching process, rather than serving as an ‘added-on’ component. Research on the role of technology in teacher training has also focused on specific teaching skills or resources (for example, Raigón Rodríguez & Gómez Parra (Reference Raigón Rodríguez and Gómez Parra2005) look at how technology can serve as an important resource for teaching specifically targeted lexicon). Antoniadou (Reference Antoniadou2011) provides an in-depth longitudinal analysis of language teacher development mediated through CMC. Basing her study on a wide, multimodal database (face-to-face interactions, online synchronous and asynchronous interactions, and social-media output), Antoniadou examines the computer-supported interactions between two transatlantic groups studying to become foreign language teachers. The participants took part in a year-long task-based telecollaborative project designed to expand the learning opportunities of both groups through international, intercultural engagement with online peers. Through the use of analysis of ‘learning episodes’ the author traces moments of professional development of teacher competences, facilitated through virtual collaboration. Results from the qualitative data lead Antoniadou to conclude that the effective integration of telecollaborative activities in pre-service teacher education programmes can enhance teacher learning in significant ways, including content knowledge, language knowledge, as well as the ability to better apply critical reflection to one's own teaching practices.

Finally, research in Spain has delved into the challenges faced during the implementation of TELL – a list which frequently includes technical difficulties (Erguig Reference Erguig2009; Ortega Martín, Macrory & Chretien Reference Ortega Martín, Macrory and Chretien2010), time limitations, and lack of professional training in the efficient use of technology in the classroom (Ruiz Madrid Reference Ruiz Madrid2006; Durán Fernández & Barrio Barrio Reference Durán Fernández and Barrio Barrio2007).

5.4 Reflections

It is quite frequent, when considering technology and education, to envision a bright future in which language teaching and learning are optimized through the integrated use of ever-expanding technological resources, and this optimism (as well as the challenges to be faced) is frequently voiced in the literature published by Spanish experts in the field of TELL. Nonetheless, as Hubbard (Reference Hubbard2008) reminds us, it is also easy to imagine a more dismal future in which ‘technology remains weakly integrated (. . .) and research remains fragmented’ (p. 176). An essential step to ensure the former, cheerier vision of future language teaching and learning is, inevitably, carrying out investigation in the field. In this sense, Spanish researchers appear resolute; however, there is a need to move away from indiscriminate embracing of technologies as the panacea of educational ills towards a more critical research that aims to improve understanding of the underlying process of TELL. As was pointed out over a decade ago in a Eurydice (2001) report, ‘the relations between ICT and education are complex and cannot be reduced to the simple availability of material resources within the educational process’ (p. 15). Studies that further a deep understanding of the intricate relationship between technological tools and language learning are essential.

There is ample descriptive documentation of TELL practices published in Spain, along with numerous articles, chapters, and books on pedagogical proposals for its use. On the other hand, research in this area, as well as studies on MALL, appears to be more widely published in international arenas, less so in Spain, and may result in less diffusion of study outcomes for a local audience, including local, regional and national policy makers.

The issue pointed out previously about CLIL research published in Spain can similarly be detected in TELL studies: the prevalence of holistic studies based on single events (e.g. case studies, classroom practice). These descriptive analyses are important for advancing knowledge in the field but would benefit from complementary longitudinal, large-scale research that departs from results established up until now.

Research thus far has demonstrated that student and teacher perceptions of the use of technology are generally positive, although admittedly a seamless fusion of technological resources into current teaching practices has its challenges. Further studies are needed in order to delineate more exact parameters for effective, integrated language learning with technology so as to guide teachers and policy makers in this transition. Researchers should try to identify the necessary enabling environments that must exist if technology is to be used ‘effectively’ in language teaching. Also, as the use of technology in the language classroom becomes more mainstream, there is a need for studies into the impact of these resources on learning and achievement as well as on how to monitor and evaluate the development of multiple literacies (not only language but digital competences, for instance). Finally, questions of equity should be explored, such as whether there are differential impacts of technology use in identifiable sub-sets (e.g. gender, socioeconomic, linguistic or cultural groups) and how to ameliorate disadvantages, if any.