Keywords

1 Introduction

1.1 Contextualisation and Definitions of Key Terms

“Affectivity” is an umbrella term that has been widely used to refer to a range of individual variables that have been perceived to interact with second language acquisition (SLA). Indeed, research has foregrounded affective factors so as to account for the fact that language gains, in whatever learning context, do not always follow clear patterns. Thus, there is evidence pointing to such factors and ultimate achievement in the target language being highly interrelated (Gardner 1985; Bernaus 1994; Dörnyei 2001; Masgoret and Gardner 2003; Bernaus et al. 2004; Polat and Schallert 2013). This has led to the growing visibility of combined “mixed-methodologies” (Allen and Herron 2003; Ellis 2008), especially in the wake of the publication of Firth and Wagner (1997, re-published 2007), somehow uniting the strengths of quantitative research with a qualitative dimension which enquires into those “affective factors” that are believed to influence SLA, including (language) attitudes, beliefs (and/or opinions), and motivation.

Richards et al. (1992) and Richards and Schmidt (2002) regard attitudes as expressions of positive or negative feelings towards a language. Language attitudes, therefore, are abstractions to be inferred from stated beliefs or observed patterns of behaviour, reflecting the perceived simplicity or difficulty of learning, degree of importance, elegance, or social prestige of a target language, which may play an important role in the language learning process.

For their part, Richards and Schmidt (2002: 297) consider beliefs (often also referred to as, or together with, “opinions”) to be relatively stable sets of ideas and attitudes about such aspects as language learning, teaching methodology, personal abilities or goals in language learning, which may sometimes impede the acceptance of new ideas and practices. As a result, such beliefs may influence learners’ attitudes and motivation during the language learning process.

Finally, Gardner defined motivation as a “combination of effort plus desire to achieve the goal of learning the language plus favourable attitudes toward learning the language” (1985: 10). Accordingly, motivation can only be artificially separated from attitudes or beliefs (Gardner 2006). Research has distinguished two main types of motivation, namely, instrumental and integrative motivation (Brown 2000). The former involves concepts of practical value for learners such as career promotion, business opportunities, prestige, power, access to scientific and technical information or just passing a test or exam (Gardner 1983: 203; Saville-Troike 2006: 86). On the other hand, integrative motivation has focused on the individual’s need to belong. This, referred to as “integrativeness” by Gardner (1983), assumes the existence of an increasing identification process with a valued community on the part of the learner (Dörnyei 2009: 22–23).

However, the internationalisation of English has problematised the very notion of that “valued community”, which has led Dörnyei and Csizér (2002) to reformulate “integrativeness” in terms of their “L2 motivational self system” (already applied in foreign language acquisition research—see Polat and Schallert 2013). Thus, integrativeness now becomes an identification process with “the ideal L2 Self” (Dörnyei 2009), an ideal image of oneself as a proficient L2 speaker which learners are assumed to have and which reinforces their integrative disposition. Interestingly, in the Dörnyeian model the ideal L2 Self is seen to feed on both attitudes towards members of a (possibly international) L2 community and purely instrumental aspects such as future professional success (Dörnyei 2009). Consequently, the re-defined concept of integrativeness is now seen as the gateway to the learners’ intended effort to study their language of choice. In this way, the L2 motivational self system blurs the clear-cut distinction between intrinsically- and extrinsically-oriented motivation.

1.2 CLIL and Affectivity

Interestingly, experts have been quick to point out that CLIL may play a role in the promotion of positive affective factors. Thus, Banegas (2012: 113) highlights the motivational benefits to be drawn from carefully-implemented CLIL programmes. Casal and Moore (2009), for their part, attribute the higher language gains found among CLIL learners to motivational issues. In particular, experts have pointed to CLIL as providing that “comprehensible input +1” which, according to Krashen’s (1987) influential learning/acquisition model, learners should find challenging yet motivating (Wilhelmer 2008). By placing language as a tool (and not only as the target) in the L2 learning process, CLIL programmes may give rise to holistic, cognitively-engaging classroom activities (Brewster 2004: 26), which may in turn have important side-effects such as improved learner confidence, motivation and autonomy (Coyle 2006a, b), especially when lack of interest seems to have characterised secondary education contexts like Spain’s for at least the last two decades (Alcalde et al. 1994).

In this regard, research suggests that younger students appear to show a stronger desire to learn, and more positive attitudes towards language learning in general, and English in particular, than their older counterparts (Tragant and Muñoz 2000). These findings are supported by Cenoz (2001), who concludes that younger students generally hold more positive attitudes towards the foreign language than older learners due to psychological and educational issues. In other words, there seems to be a general decline in positive attitudes towards foreign language learning in students from the highest grades, which in the Spanish context would predictably affect the last years of compulsory secondary education (CSE). Interestingly, higher education contexts do not seem to be affected by this problem (Karahan 2007; Yassin et al. 2009).

There seems to be, however, remarkably little research available on CLIL and its possible effects on affectivity, particularly in those specific contexts in which CLIL is used in bilingual settings (see Lasagabaster 2015 for a full report). This is relevant since evidence has emerged of other variables interacting with affectivity in language learning contexts. One such variable is the learner’s language profile since, apart from the fact that research points to the acquisition of additional languages possibly being fostered in multilingual speakers (Lasagabaster 2015), multilingualism has also been linked to multicultural attitudes (Bourhis et al. 1981; Cenoz and Gorter 2011). This partly accounts for the growing interest in CLIL programmes on the part of the European institutions in their effort to promote plurilingualism (North 2000; Council of Europe 2001). Among other findings, Lasagabaster and Sierra (2009) have found that positive attitudes to not only English but also Spanish and Basque are higher among CLIL than non-CLIL students. However, such results should also be taken with some caution. Thus, Seikkula-Leino (2007), reporting from Finland, finds that English CLIL primary-school students may combine low self-concepts in the foreign language (not significantly different from those found among their non-CLIL peers) with a strong motivation to learn. Lasagabaster (2015) also reviews examples from other (mainly Asian) contexts showing that the viability of English CLIL instruction in bi- or multilingual settings is highly dependent on attitudinal factors at macrosocial levels.

Finally, even though research is still scanty in the area (Anya 2011: 442), students’ attitudes also seem to be related to gender issues. Thus, although there is no general agreement on this as yet, female students are mostly reported to have more positive language learning attitudes and be more strongly motivated than male students (Merisuo-Storm 2007; Pavlenko and Piller 2008), although such differences have not always proved significant (Weseley 2012; Henry and Cliffordson 2013).

1.3 Aims

This chapter aims at exploring the role CLIL may play in the development of affective factors in learners of L3 English. The novelty of the study resides in that, despite the relevance of affective factors, research that compares such factors among both CLIL and formal instruction (FI) students remains scarce (Lasagabaster and Sierra 2009); secondly, most of the literature on attitudes, beliefs and motivation in the field of SLA has generally focused on university contexts (Weseley 2012), whilst the present study will centre on CSE; and finally, attention will also be paid to the learners’ gender and language profile, in order to measure their possible impact on attitudes.

More specifically, our study addresses the following research questions:

RQ1:

Does learning context (CLIL vs. non-CLIL) play a role in the development of affective factors?

RQ2:

What is the impact of CLIL programmes on affective factors related to the content subject taught through English?

RQ3:

Does the participants’ language profile have an effect on their interest in language learning?

RQ4:

Does learning context in combination with the learner’s gender influence the development of affective factors?

Our study’s methodological considerations are presented in Sect. 2 below. The main findings can be found in Sect. 3, duly subdivided into four subsections, each addressing a separate research question. Section 4 provides a discussion of the main findings. Finally, some general conclusions will be presented.

2 Method

2.1 Participants

The present research was designed as a longitudinal study. Its participants (all from the COLE project pool) were two groups (CLIL vs non-CLIL) of Catalan-Spanish bilinguals enrolled in CSE at five state-run schools in the Balearic Islands (see Chapter “Learning English and Learning Through English: Insights from Secondary Education”). They were asked to complete a questionnaire tapping into the affective factors signalled above, namely, attitudes, beliefs and motivation (ABM). Participants completed the questionnaire at the beginning of year 2 of CSE (T1, marking the onset of the CLIL programme) (CLIL n = 93; non-CLIL n = 77; age 13–14); additionally, a subsample of participants (CLIL n = 85; non-CLIL n = 66) completed their questionnaires at a second collection time (T2, end of year 3 of CSE; age 14–15). This second collection time coincides with T3 within the general COLE project research design (see Chapter “Learning English and Learning Through English: Insights from Secondary Education”).

CLIL students had 6 h of English classes per week (3 h of EFL plus 3 h of content subjects through English). For their part, non-CLIL students only had 3 h of EFL lessons per week. In three of the participating schools, CLIL programme admission criteria included the student’s general academic record, their previously obtained EFL grades, and / or their performance in an EFL placement test. Table 1 shows the gender distribution per group.

Table 1 Participants’ gender distribution
Full size table

As can be seen, female participants clearly outnumber males within groups at both collection times [T1 = 92 (54.12 %) vs. 78 (45.88 %) and T2 = 86 (56.95 %) vs. 65 (43.05 %)].

2.2 Research Instruments

As stated in Chapter “Learning English and Learning Through English: Insights from Secondary Education”, a questionnaire (ABM) was developed to capture the opinion of both groups of students at both T1 and T2. The questionnaire was adapted from another questionnaire used by the SALA-COLE research group in the SALA (Study Abroad and Language Acquisition) Project, with a Cronbach’s alpha measuring internal consistency between .74 and .93 (see Pérez-Vidal 2014).

The ABM questionnaire was divided into three main sections, each focusing on the following: attitudes (19 items), students’ beliefs and opinions on the learning of English (20 items), and motivation (14 items). Each section included a set of items which were only answered by CLIL students, as they concerned their experience in this programme: items 14–19 (first section), items 11–20 (second section) and items 9–14 (third section). The questionnaire was administered in Catalan since this is the preferred language of communication at the participating schools.

Participants were also asked to complete a language profile questionnaire (see Chapter “Learning English and Learning Through English: Insights from Secondary Education”), which was only administered at T1. It elicited factual data concerning the participants’ personal details, L1 and L2 language use, EFL instruction, learning experiences, (foreign language-related) habits and contact with EFL speakers.

2.3 Analysis

Students were asked to indicate the extent to which they agreed or disagreed with the statements of the two first sections of the ABM questionnaire on a 1–5 Likert-type scale, ranging from 1 (“totally agree”) to 5 (“totally disagree”). The third section asked students to choose 1 out of the 5 different options included in each of the items. Results were analysed using the Statistical Package for the Social Sciences (SPSS) programme.

3 Results

3.1 Learning Context and Development of Affective Factors

3.1.1 Language Attitudes

T-test results from questions 1–13 (section 1 ABM questionnaire) indicate that, although there were differences between both groups of students, such differences were not statistically significant. The comparison of the means between the two groups reveals, however, the tendency of the non-CLIL group to have more negative learning attitudes at T1 (see Table 2). It should be noted that the highest mean scores (5) show negative attitudes towards English, whereas the lowest scores show positive attitudes towards the language. For the sake of readability, items 1, 7 and 8 (in which, contrary to all other items, high scores signal positive attitudes) have been transformed and recoded so that the values obtained in all the items can be compared in a consistent manner. Interestingly, research findings show that the non-CLIL group score slightly higher than their CLIL peers on these aspects at T2.

Table 2 Learning attitudes: descriptive statistics
Full size table

The only significant difference between both groups was found in item 1 (“I am studying English because it is a compulsory subject”) at T2 (t = −2.212; p < 0.05), with CLIL students expressing clearly lower agreement with this item.

Time, however, is an intervening variable as far as the development of language attitudes is concerned. Thus, statistically significant differences were found among students as a whole (CLIL and non-CLIL students altogether) in items 4, 5, 8, 10 and 11 (p < 0.05) as time progressed (T1 vs. T2). Results indicate that students’ overall opinions improved considerably at T2 in the aforementioned items: “In general I like English music and I want to understand it” (item 4); “I like watching English films and to be able to understand them” (item 5); “I want to travel abroad and speaking English will be useful” (item 10); and “I want to speak English because I want to communicate with people from different countries” (item 11). The exception was item 8, “I like the English language but I do not like the English lessons”, which obtained lower ratings at T2.

3.1.2 Beliefs on the Learning of English

Section 2 ABM aimed at gathering information on both CLIL and non-CLIL students’ opinions on various aspects of the English language learning process. Results show that both groups of students share similar views. In fact, the t-test performed only revealed statistically significant differences in item 5 (t = 2,116; p < 0.05): “I get nervous when I have to speak English”. When speaking English is required, non-CLIL students report higher degrees of anxiety than their CLIL counterparts.

Across time (T1 vs. T2), statistically significant differences were only found for item 7, “I would like to get to know more English language speakers”. At T1 (t = −2.729; p < 0.05), CLIL students show greater interest in this aspect, registering values closer to the positive end of the scale (1 = totally agree; 2 = agree) (Table 3). At T2, nevertheless, results indicate that both groups show greater interest in meeting native English speakers. This growing interest is more apparent among non-CLIL students who, at T2, show even greater willingness to get to know native English speakers than CLIL students, although this difference is not significant. In fact, no statistically significant differences were observed between groups at T2.

Table 3 Beliefs (item Q2.7): “I would like to know more English people”
Full size table

3.1.3 Motivation Towards Learning English

Section 3 ABM enquired into motivational issues. CLIL students admit to being more highly motivated during the English lessons (item 8), but differences between CLIL and non-CLIL students were not significant over time (T1 and T2), all students’ responses gathering at the positive end of the scale (1 = extremely motivated, and 2 = highly motivated) (Table 4).

Table 4 Motivation (item Q3.8): “My motivation in the English lesson is very high… very low”
Full size table

Likewise, both groups of students acknowledge either studying “a great deal” (value = 1) or at least doing a “fair enough” amount of study work (value = 2) (item 3), with no significant differences between groups over time.

As far as the identification of the key motivating factors in the EFL class is concerned (item 5), the following aspects were ranked in order of importance at both times: “marks”, “group work”, “activities”, “teaching method”, and “amount of work”. The time variable does not seem to affect such factors. This also applies to the identification of the less motivating factors related to the EFL class, which seemed to remain stable across time (item 6). The following were ranked in order of importance: “amount of work”, “activities” and “teaching method”.

Students were also asked to identify the aspect they liked the most about the English language (item 7). In this respect, both groups of students report that “the number of things that English enables them to do” is the most motivating factor for them (Fig. 1). This interest increases over time (51.6 % vs. 57.3 %). Lagging quite far behind this, students also acknowledge being motivated by such aspects as “the way English sounds”, “English people” and “English culture”. The first of those three aspects (“the way English sounds”) becomes an even more appealing option at T2 (20.8 % vs. 26.6 %). On the contrary, the students’ interest seems to diminish across time regarding the other two aspects, namely “English-speaking people” (12.6 % vs. 8.4 %), and especially their “culture” (10.7 % vs. 2.8 %).

Fig. 1
figure 1

Motivation: “What I most like about English”

Full size image

Results also indicate that students identify two major reasons for learning English (item 2): “getting a good job” (T1 39.13 % vs. T2 29.17 %) and “being able to communicate with people around the world” (T1 33.54 % vs. T2 43.75 %). This last reason scores higher at T2, that is, as students’ command of the English language increases. Participants also admit to “liking English” as a relevant factor, although this obtains similar values over time.

3.2 CLIL and Affective Factors: The Content Subject Taught Through English

3.2.1 Attitudes

Section 1 ABM contained a sub-section (items 14–19) exclusively addressing CLIL students’ language attitudes. T-test results indicate that there are no statistically significant differences among CLIL students regarding this issue. Time does not seem to affect their opinions to a significant extent either. On the whole, CLIL students seem to share more homogeneous attitudes towards learning than non-CLIL students.

Table 5 shows the descriptive statistics for the aforementioned items. As can be seen, CLIL students show a positive attitude towards all the aspects concerning English, since the mean values of all the items are situated at the positive end of the scale (1 = totally agree, and 2 = agree).

Table 5 CLIL participants’ attitudes at T1 and T2
Full size table

3.2.2 Beliefs

The results discussed here have been drawn from questions 11–20 in section 2 ABM. As was the case with attitudes, no significant differences were observed among CLIL students over time except for item 20: “It is difficult for me to understand the CLIL subject”, which obtained more positive results at T2 (Table 6).

Table 6 Self-perceived difficulty to understand the CLIL subject: T1 vs. T2
Full size table

3.2.3 Motivation

CLIL students’ motivation regarding their CLIL subject was addressed in Section 3 ABM (items 9–14). Concerning the CLIL subject (item 13), students report being either “very highly” (T1 15.4 %, vs. T2 24.8 %) or “highly” motivated (T1 61.8 % vs. T2 46.8 %), with little overall difference over time. Students also report studying “a lot” (T1 17.9 % vs. T2 26.6 %) or “hard enough” (T1 64.3 % vs. T2 56 %) for their CLIL subject (item 10), again with little difference across time.

As regards the major motivating aspects related to the CLIL subject (item 11), on the one hand, students identify “group work” and “teaching method” at T1 and T2 respectively. In general, T2 scores are lower for all the aspects except for “teaching method”. “Marks” are also relevant, especially at T2, being placed second in order of importance after “teaching method” (Fig. 2).

Fig. 2
figure 2

“What motivates me most in the CLIL lesson is…”

Full size image

On the other hand, the least motivating factors (item 12) seem to be “amount of work” and “tasks”, ranked in this order of importance at both times (T1 and T2). Lagging behind this, students also report “teaching methodology” and “marks” among the least appealing factors. However, “teaching methodology” appears to be seen in a more positive light at T2, whereas “marks” are seen as equally relevant at both times (T1 12.1 % vs. T2 12.4 %) (Fig. 3).

Fig. 3
figure 3

“What motivates me the least in the CLIL lesson is…”

Full size image

At T1, CLIL students’ willingness to study their CLIL subject (item 9) is mainly motivated by their wish “to have a good job”, although the second major reason reported is simply “because [they] like it”. In fact, this latter factor obtains the highest score at T2 (Fig. 4).

Fig. 4
figure 4

CLIL: Most motivating factors

Full size image

3.3 Language Profile: Interest in Language Learning

In order to ascertain the possible effect of the CLIL programmes under study on the learners’ interest in language learning, the participants’ language profile was examined. Firstly, attention was paid to the languages that the participants claim they usually speak to communicate with both their parents. Results show a very different language distribution among CLIL and non-CLIL participants. As can be seen in Fig. 5, Catalan-dominant speakers (i.e. those that use only this language in their daily communication with their parents) are far more representative in the CLIL group than in the non-CLIL group. Conversely, Spanish prevails in this latter group, which is also characterised by a much higher percentage of participants who use languages other than Catalan, Spanish or English in such habitual communication.

Fig. 5
figure 5

Participants’ language profile (Note: Mother L1 CLIL Language used by CLIL students to communicate with their mothers, Mother L1 non CLIL Language used by non-CLIL students to communicate with their mothers, Father L1 CLIL Language used by CLIL students to communicate with their fathers, Father L1 non CLIL Language used by non-CLIL students to communicate with their fathers)

Full size image

Additionally, the grades CLIL and non-CLIL participants obtain in the Catalan, Spanish and EFL subjects were looked into. The comparison of the means on a 4-point scale between both groups of students reveals that CLIL participants obtain higher marks in EFL. As for the two official languages in the Balearic Islands, CLIL students obtain higher grades in Catalan and slightly lower marks in Spanish (Table 7), although such differences are not significant.

Table 7 Comparison of CLIL vs. non-CLIL participants’ grades
Full size table

Table 8 shows the grades that CLIL students obtain in their CLIL subject on a 10-point scale. As can be seen, these are slightly lower than those obtained in their EFL course, although differences are not significant.

Table 8 CLIL participants’ grades in the CLIL subject
Full size table

Finally, results reveal statistically significant differences between CLIL and non-CLIL students concerning the learning of a second language other than English (t = −2.044; p < 0.05). Unlike their non-CLIL peers, the majority of CLIL students report speaking a second foreign language in addition to English.

3.4 Learning Context and Gender: Impact on Affective Factors

3.4.1 Female vs. Male Students’ Attitudes (CLIL and Non-CLIL)

Although no significant differences were found when comparing the possible interaction of gender and attitudinal factors in CLIL and non-CLIL students (as measured in Section 1 ABM, items 1–13), t-test results reveal statistically significant differences between male and female students as a whole in items 2(t = −2.498), 5(t = −2.388), 6(t = 2.048), 11(t = −2.236) and 12(t = −2.427) at p < 0.05 at T1. Significant differences were also found in items 2(t = −2.263), 4(t = −3.308), 6(t = 3.158), 10(t = −2.478) and 12(t = −2.254) at p < 0.05 at T2. As can be seen, the difference between both groups remains constant across time except for items 5 (“I like watching English films and being able to understand them”) and 11 (“I want to learn English to be able to communicate with foreign people”), which are only significant at T1. Additionally, results indicate statistically significant differences at T2 between male and female students in two other items: item 4 (“I like English music and I want to understand it”) and item 10 (“I want to travel abroad and learning English will help me”).

As Table 9 below shows, mean value results indicate that female participants have more positive attitudes towards the English language (item 2) and language learning in general (item 12) at T1 and T2. Female students also show greater interest in English films and seem to derive more pleasure from their ability to understand them (item 5). They also tend to show more visible appreciation of English as a tool enabling communication with foreign people (item 11). Male students, on the other hand, consider that the English language helps them understand videogames (item 6) and value this aspect higher than their female counterparts at both times.

Table 9 Gender-based differences in attitudes: group statistics
Full size table

As can be seen in Table 9 below, female students show a more favourable attitude towards items 4 (“I like English music and I want to understand it”) and 10 (“I want to travel abroad and learning English will help me”) at T2.

3.4.2 Beliefs on English Language Learning: The Interaction of Learning Context and Gender

With regard to both CLIL and non-CLIL participants’ views on the learning of English (section 2, items 1–10), the only significant difference between male and female students was found in item 10 at T1: “In our community it is necessary to know how to speak English” (t = −3.055; p < 0.05). As can be seen (Table 10), mean value results show that female participants exhibit more positive views on this issue.

Table 10 Gender-based differences in beliefs: “In our community it is necessary to know how to speak English”
Full size table

However, this gender gap decreases over time since no significant differences were found at T2.

3.4.3 Gender and Motivation (CLIL and Non-CLIL)

Results reveal overall statistically significant differences between male and female students in item 3 (section 3 ABM): “In the English class I study a lot…nothing” (t = −2.273; p < 0.05) at T2. As can be seen (Table 11 above), mean value results reveal that, as time progresses, self-perceived study work is higher among females (1 = a lot; 5 = not at all). Male students, on the contrary, seem to lose interest over time and admit studying significantly less at T2.

Table 11 Group statistics: gender and motivation in CLIL and non-CLIL participants
Full size table

Female students also express higher motivation than their male counterparts (item 8), especially at T2, although the gender gap is not significant here. Overall results, however, suggest a slight decline in the participants’ interest and motivation over time.

3.4.4 Female vs. Male CLIL Students’ Attitudes, Beliefs and Motivation

As mentioned above, the three sections of the ABM questionnaire included a set of questions which addressed CLIL students exclusively. No statistically significant differences between males and females were reported across time among CLIL students regarding the issues concerned, indicating that both male and female CLIL participants held more homogeneous attitudes, beliefs and motivation than their non-CLIL counterparts.

4 Discussion

4.1 Learning Context (CLIL/Non-CLIL) and the Development of Affective Factors

With regard to the first research question, which enquired into the role that learning context may play in the development of affectivity, results show no statistically significant differences between CLIL and non-CLIL students concerning the affective factors measured. Thus, even though attitudes and beliefs do grow and motivation improves among CLIL participants between T1 and T2 (in line with claims put forward by Coyle 2006a or Wilhelmer 2008), this also holds true for non-CLIL students. An interesting finding was that all participants highlight the importance of English as a major tool enabling communication all around the world. This may be seen as a sign that both groups are aware of the status of this language as a lingua franca and that communication in English may well take place between non-native speakers of the language (McKay 2003; Llurda 2004; Ives 2010).

Among non-CLIL students, a tendency was detected to present more negative attitudes, already noticeable at T1. The only item for which significant differences have been found at T2 between CLIL and non-CLIL students (“I am studying English because it is a compulsory subject”) points to the latter being more instrumentally-motivated. Non-CLIL students, therefore, appear to be less intrinsically motivated to study English.

Time, however, seems to be a relevant variable for all students, and significant differences were found between T1 and T2 concerning different aspects, namely their interest in English music and films, and awareness of the usefulness of English as a tool that will enable them to travel abroad and communicate with a wide range of people.

Taken as a whole (i.e. both CLIL and non-CLIL), participants seem to show less appreciation for their EFL classes over time, in line with Tragant and Muñoz (2000), Cenoz (2001) and Lasagabaster and Sierra (2009). In light of this and other results discussed below, students’ negative attitudes seem to be related to language teaching methodologies. On the other hand, students’ increased contact with the foreign language seems to be linked to the formation of more positive attitudes towards it.

Results also seem to suggest that learning context does not play a major role in the development of participants’ affective variables. In fact, significant differences were only found across time regarding two aspects. One was the lower anxiety level claimed by CLIL students when having to speak English in class. This is probably due to the fact that CLIL students have been more amply exposed to, and encouraged to speak in English in class and, therefore, have greater confidence in their oral foreign language skills (see Lorenzo et al. 2011).

The other aspect for which significant differences were found concerns the learners’ willingness to meet more native English speakers. Although this wish is significantly stronger among CLIL students at T1, this interest grows in both groups at T2, and even more so among non-CLIL students (although this difference is not significant). This may be due to different reasons. CLIL students were already relatively highly motivated at T1 and therefore it could be argued there was little room for improvement in this respect.

Finally, students also report on those aspects they find most motivating about their target language. The results obtained point to both CLIL and non-CLIL students seemingly granting more importance to the amount of things that speaking English enables them to do, which may be taken as a sign of students being more instrumentally-motivated. However, this also needs to be interpreted in combination with other results such as the growing interest expressed by participants concerning English music and films, which points to the more integrative end of the motivation continuum. This could be read in the light of the L2 self theory lately put forward by Dörnyei (2009; see also Polat and Schallert 2013), which no longer neatly separates integrative and instrumental motivation, and points to the learner as gradually learning to see him/herself as a member of a target community which is transnational in nature (see also Yashima and Zenuk-Nishide 2008). This may at least help explain the fact that native English-speaking people and their culture(s) are of no particular interest to our participants.

4.2 CLIL: Effects on Affectivity Related to the Content Subject Taught Through English

Concerning the second research question, which looked at the effects of a CLIL programme on affectivity related to the content subject taught through English, no significant differences were found among CLIL students regarding their positive attitudes to CLIL subjects, which indicates that such attitudes are fairly homogeneous.

As for their beliefs towards the content subject taught through English, no significant differences were found between T1 and T2 except for one particular aspect, which indicates that CLIL students find it easier to understand their CLIL subject at T2 (Section 2 ABM, item 20). Thus, the greater and more meaningful exposure to English in CLIL programmes seems to be effective when it comes to increasing students’ language understanding, particularly as far as their receptive skills are concerned. This is something that had already been reported about language immersion programmes (Genesee 1994; Grabbe and Stoller 1997) and, more recently, about CLIL itself (Ruiz de Zarobe 2015).

Results also point to CLIL students being remarkably motivated in their EFL lessons, reporting individual hard work, with no significant differences across time. This may be taken as further evidence that they were highly motivated in the first place. Therefore, CLIL might not play a major role in increasing an already high motivational level (see Trenchs-Parera and Juan-Garau 2014, who find a very similar pattern in a higher education context).

As regards motivation in relation to their CLIL subject, there is little variation across time, although scores tend to be slightly lower at T2. It is worth noting, however, that the scores obtained by the “teaching method” are higher at T2, which might be an indicator of CLIL students’ growing appreciation of CLIL. Surprisingly, “teaching method” is also cited among those least motivating factors about the CLIL subject, but it should be added that, even when considered a drawback, methodology is on the whole seen in a more positive light at T2.

Additionally, evidence has also been found that CLIL students’ willingness to study their content subject through English is driven by a mix of instrumental and integrative motivation factors at T1, although the latter clearly prevail at T2. Thus, it can be posited that the CLIL programmes under study have a positive effect on the students’ motivation, particularly of the integrative kind, again in line with Coyle (2006a) or Wilhelmer (2008). Further support of this thesis can be found in the fact that CLIL students increasingly enjoy having to use and communicate in English in their CLIL subject.

4.3 Language Profile and Interest in Language Learning

In order to enquire into our third research question, which considered the possible relationship between language profile and interest in language learning, the participants’ language profile was examined. Results show that the CLIL and non-CLIL groups are different as regards their L1s, with Catalan prevailing over Spanish among the former whilst the reverse applies to the latter. The percentage of users of languages other than Catalan, Spanish and English is also much higher among non-CLIL participants. This points to the non-CLIL participants in our sample being more ethnically diverse than their CLIL counterparts.

Additionally, CLIL students obtain higher marks in Catalan (the schools’ vehicular language) and English, and slightly lower marks in Spanish, even though these differences are not significant. This may be related to the different language background found among participants.

The different language profile found between both groups of participants may be an indicator of extrinsic factors (e.g. socio-economic context and cultural capital) which may have played a role in the participants’ academic performance, including their language-related subjects (for a fuller discussion, see Boada et al. 2011). Such performance, therefore, might be related to aspects and factors other than learning context (CLIL or FI, in our case). Indeed, the greater ethnic diversity detected among the non-CLIL students may in turn be illustrative of the high immigration rates that have of late characterised Majorcan society (see Chapter “Learning English and Learning Through English: Insights from Secondary Education” for further details). The greater diversity of cultural traditions, economic and linguistic backgrounds found among non-CLIL participants may therefore have had an impact on their school performance and other related issues. In this respect, it is important to highlight that results reveal statistically significant differences between CLIL vs. non-CLIL students as far as the learning of an additional foreign language other than English is concerned (t = −2.044; p < 0.05). Unlike non-CLIL students, the majority of CLIL participants report speaking a second foreign language in addition to English. Therefore, CLIL students present a richer, more complex language profile which, linked to the socio-economic and cultural capital issues referred to above, may be related to higher motivation and linguistic performance.

4.4 Do Learning Context and Gender Influence the Development of Affective Factors?

Regarding our fourth research question, and taking both CLIL and non-CLIL students in combination, significant differences were found between male and female participants concerning attitudes towards English as a foreign language. This may be seen to be in line with Block’s post-structuralist view that gender is a “multilayered, graded” phenomenon, “grounded in social interaction” (2007: 866). The L2 self would therefore also be gendered, and this variable may interact with others such as “race, ethnicity, age, […] and social class” (Block 2007: 869) in the development of attitudes, beliefs and motivation. This is in agreement with previous research carried out by Brecht et al. (1995), Kinginger (2004), Isabelli-García (2006), Merisuo-Storm (2007) or Pavlenko and Piller (2008).

In our study, females seem to be more fully aware of the importance of being able to communicate in English, considering the importance of the tourism sector and the size of the resident tourist population in the Balearic Islands. This, which may point to a higher level of instrumental motivation among females (it is difficult to get a job in Mallorca if applicants do not speak English), may also be attributed to greater social sensitivity and eagerness to embrace a wider society than their own. In this respect, it should be noted that research has concluded that “many women around the world see learning English as a way of liberating themselves from the confines of gender patriarchy” (Pavlenko and Norton 2007: 677). This would again signal a rather blurred line separating instrumental and integrative motivation. In fact, overall female participants express remarkably more positive attitudes towards language-related issues which point to female learners being more integratively-motivated than their male counterparts. This is in line with most of the research conducted in the area (see, e.g., Sunderland 2000; Kobayashi 2002; Schwieter 2008).

However, it is worth noting that no significant gender-related differences were found among CLIL students concerning attitudes, beliefs and motivation. This finding is also consistent with the available literature showing that gender-based attitudinal differences seem to wane in the so-called “immersion” (Baker and MacIntyre 2000) and content-based language learning programmes (Lasagabaster and Sierra 2009).

Conclusions

The main aim of this study was to provide insights into affective factors concerning English and foreign language learning among secondary school students in two different learning contexts (CLIL vs. non-CLIL) in Mallorca. Overall, the following findings can be highlighted. First, although CLIL students tend to have more positive attitudes and beliefs than their non-CLIL peers, such differences are not significant; secondly, motivation grows among both CLIL and non-CLIL participants, reaching higher levels among the former, but CLIL students are already highly motivated at T1, leaving less room for improvement at T2; thirdly, there is evidence that CLIL and non-CLIL groups are different as regards their L1s, Catalan being more widely represented among students in the CLIL group, and Spanish and languages other than English being clearly more representative in the non-CLIL group; finally, significantly more positive overall attitudes are found among female participants, although gender-based differences are not statistically significant within the CLIL group.

To conclude, a possible reading of such findings could be as follows. To start with, CLIL seems to prove beneficial regarding the development of attitudes, beliefs and motivation. However, this cannot be taken as the sole factor playing a role here. Thus, the CLIL and non-CLIL students in our sample behave differently in that the former show far more homogeneous patterns regarding affectivity and are already more highly motivated prior to treatment.

Secondly, CLIL students report lower anxiety levels when speaking English in class. CLIL does seem to have played a role in increasing their confidence since this methodology substantially increases the participants’ exposure to the English language. However, this does not seem to result in significantly higher grades in EFL, as reported by participants.

Thirdly, the different language background of the CLIL and non-CLIL groups (and the implications this may have in terms of social background and cultural capital) should also be considered when accounting for the more positive attitudes that CLIL students show towards foreign language learning in general. In this regard, the CLIL participants’ higher motivational standpoint at T1 above may be seen as an indicator of the possible streaming of the best students into the CLIL group (common practice within CLIL programmes throughout Europe, including Spain—see Eurydice 2006; Bruton 2011). This interpretation is largely congruent with the information provided by three of the schools involved in this research, which use the student’s general academic performance as a screening factor for admission into CLIL pilot programmes.

Finally, female students in general present more positive attitudes than their male counterparts, but gender-based attitudinal differences do not prove to be statistically significant within the CLIL group. Although this affords different readings, it may also suggest that CLIL programmes may contribute to the neutralisation of widely-reported gender-based differences concerning attitudes, beliefs and motivation in foreign language learning, in line with the successful results reported in this regard by the UK Department of Education (2011) in relation to a pilot CLIL programme developed jointly at primary schools in the UK, France and Spain.