Clil Learning Affective Factors

CLIL Learning: Achievement Levels and Affective Factors Jaana Seikkula-Leino Faculty of Education, Teacher Training School, Turku University, Finland The aim of the study was to investigate how successfully pupils had learned content in content and language integrated learning (CLIL) and to assess pupils’ affective learning factors, such as motivation and self-esteem, in CLIL. Learning was presented in terms of achievement level, which was described as the relationship between measured levels of intelligence and school success. The study indicated that there were no major differences in learning, whether the language used in instruction is the pupils’ mother tongue or a foreign language; pupils of different intelligence levels had similar chances to succeed in both cases. However, it was discovered that there may not be as many overachievers among the pupils in CLIL as among the pupils in the Finnish language instruction group. Achievements in the Finnish language, i.e. the mother tongue, were not negatively affected by CLIL. In terms of the affective factors, the study indicated that CLIL pupils had a low self-concept in foreign languages, although pupils had a strong motivation to learn.

doi: 10.2167/le635.0 Keywords: achievement levels, affective factors, content and language integrated learning (CLIL), content learning

Introduction CLIL teaching has been increasing rapidly in Finland and in some other countries as well, due to European integration and internationalism. This has created major challenges for language teaching and there is also a greater need for communicative competence. To meet these requirements some new solutions for language teaching have been developed, to enable a pupil to spend more time for language learning and to develop his communicative skills intentionally. In CLIL, a pupil usually studies some subject or a content theme in a foreign language. English is the most popular foreign language in Finland, but German, French and Russian are also used as languages of instruction. Finland’s two official languages are Finnish and Swedish, the latter being a minority language in Finland. This gives, in addition to CLIL, the opportunity to offer traditional immersion teaching where pupils whose mother tongue is Finnish study subjects and themes in Swedish. The differences between CLIL and immersion are the following factors: in CLIL pupils usually learn to read and write through their mother tongue whereas in immersion the foreign language has a role in this cognitive development. This is maybe the most noticeable difference between these two ¨ types of instruction (Malmstrom, 1993). Immersion teachers are bilingual and
C

0950-0782/07/04 328-14 $20.00/0 LANGUAGE AND EDUCATION

328

2007 J. Seikkula-Leino Vol. 21, No. 4, 2007

CLIL Learning: Achievement Levels and Affective Factors

329

their mother tongue is the language which is foreign for their pupils (Swain & Lapkin, 1982: 5). When pupils start their school in immersion, any former knowledge of this foreign language is not required. It is even recommended for pupils not to have any previous knowledge of the foreign language in question. Thus, all pupils would have the same starting level (Vesterbacka, 1991b: 64–65). In CLIL, there are no common requirements for pupils’ and teachers’ foreign language skills. In immersion, at least 50% of teaching should be through the foreign language whereas in CLIL, according to the European Union, at least 25% of teaching should be carried through a foreign language. Immersion includes the objective for pupils to become fluent in both languages. In CLIL, there is variation according to the goals concerning foreign language learning. Both immersion and CLIL focus primarily on foreign language learning, which is perhaps the greatest similarity between these two teaching types. Immersion teaching was developed in Finland during the 1980s. CLIL teaching was introduced most noticeably in the 1990s. This was made possible by the new Finnish school laws in the 1980s and 1990s. The proliferation of different forms of CLIL and immersion has raised some questions. Could the combination of a foreign language and the language which is the pupils’ mother tongue (that are both used as a media for learning) have some impact on the learning of content or the mother tongue, for example? The following article will present research results that are linked to the learning, self-esteem and motivation of the pupils in CLIL. As CLIL has not been widely studied, research on immersion teaching is also included to support the analysis and observations. Learning in immersion teaching Pupils who have attended immersion have demonstrated even better performance than those who have studied in their native language. Studies conducted in Spain and Catalan have focussed on the immersion pupils’ success in a foreign language, their mother tongue and mathematics. Immersion pupils were compared to a so-called normal class. All pupils had a low social background. The skills of the immersion pupils exceeded those of the normal-class pupils in both mother tongue and mathematics (Sampera, 1994: 13). Some other studies have also shown that immersion pupils may be better in verbal and non-verbal communication skills, cognitive skills and divergent thinking than so-called normal-class pupils (Vesterbacka, 1991a: 24). Pupils have also been reported demonstrating similar success in both immersion classes and classes that are studying in their native language. According to Canadian immersion studies, the children’s mother tongue, English, does not suffer from French immersion. It has also been observed that the immersion pupils were not weaker in the subject taught than the other pupils (Cummins, 1995; Genesee, 1988). It has also been shown that immersion has not disturbed the intellectual development of the pupils (Genesee, 1976). Results that are not in line with the above studies have, however, been obtained in studying the mathematical skills of pupils. For instance Ribes (1993), who studied ten-year-old immersion pupils, pointed out that immersion pupils did not manage worse than so-called normal-class pupils as regards

330

Language and Education

mathematical reasoning skills, but the immersion pupils showed relatively weaker mechanical skills and understanding of mathematical concepts. Correspondingly, a wide scope mathematics study conducted on second grade pupils showed that immersion pupils fared worse than other classes in mathematics (Gaya, 1994: 30–35). I have presented studies that show that the children’s mother tongue skills are very similar both in immersion classes and in so-called normal classes. For example, in Catalonia, the pupils in immersion did not differ from the group that studied in its mother tongue as regards their skills in their mother tongue, i.e. Spanish. However, what was notable was that during the first three years, the skills of the immersion pupils were slightly lower in their mother tongue than those of their peers in the so-called normal class. By the end of the third year, the differences had been levelled out. According to a study conducted in Barcelona, the immersion pupils in the fourth grade (9–10-year-olds) demonstrated the same skills in their native language as other Spanish pupils of their age (Gaya, 1994: 38–42). Dodson and Thomas (1988) have also observed in their studies on Welsh immersion pupils that they generally developed their skills similarly to their peers. However, the capacity to grasp concepts was not as strong among immersion pupils as among the pupils who studied only in their mother tongue. Vesterbacka (1991a: 166–168) has conducted immersion research in Finland and noted that the Finnish skills of immersion pupils have not been weaker than those of the pupils who are taught in Finnish. As a typical feature of linguistic development it can be said that immersion pupils who were taught in Swedish were able to write even more creatively than those who were taught in Finnish. Meril¨ainen (2002) has also reached similar results: fourth-graders who were studying in English had a good grasp of basic concepts of biology in Finnish. According to Elomaa (1996), immersion pupils in the fourth grade had lower Finnish skills than those who studied in Finnish. Let us state that immersion pupils were slightly more skilled in some fields than pupils studying in their mother tongue. Immersion studies often point out that the pupils who may be lagging behind tend to reach the same level as their peers at some point (cf. Laur´en, 2000). Learning in CLIL Research on CLIL is relatively young. However, some studies in this field have been conducted. The pupils’ mother tongue skills have been studied in the first and second grades. The language class succeeded better than the so-called normal class in the reading speed test. However, in the sections testing mechanical writing skills, reading comprehension and vocabulary, the language class did not do as well as the class studying in Finnish (Koivum¨aki & Stara, 1994). H¨am¨al¨ainen (1998) has reached similar results when studying the development of vocabulary in the pupils’ mother tongue from the second to fifth grades. Especially the pupils in upper classes demonstrated remarkably weaker vocabulary skills than their peers in Finnish language classes. It has to be noted that these results are important as the research group was fairly large. The study included 139 pupils in CLIL and 170 pupils in Finnish language teaching. The vocabulary test included mainly vocabulary connected to nature and the environment.

CLIL Learning: Achievement Levels and Affective Factors

331

The study conducted by Qiang (2000) observed the development of the cognitive skills of Tibetan children in bilingual teaching. It showed that the pupils who were taught in their mother tongue succeeded better in it than those who had participated in bilingual teaching. Moreover, especially older pupils studying in a foreign language lagged behind their peers as regards content, i.e. mathematics. It has to be noted that it is somewhat difficult to compare the results concerning teaching in a foreign language in this case because the cultural starting point varies from the Western and Finnish background. However, the study was wide as it included altogether 508 pupils who were in the second and fourth grade. The pupils’ potentially different backgrounds were taken into consideration. Pupils have been observed to learn very well in teaching content in a foreign language. For example a continuing study made by Merisuo-Storm (2000) indicated that the development of the pupils’ reading and writing skills during their first year at school did not show any remarkable difference between those who were taught in Finnish and those who were taught in a foreign language. The study also aimed at surveying how weaker and more skilled pupils developed in these different teaching methods, and no remarkable differences were noticed. The study included 80 pupils who were studying in a language class. The comparison group included 59 pupils who were studying everything in Finnish. Merisuo-Storm (2002) reached similar results in a more comprehensive study which monitored pupils during their first two years at school. Pupils studying in a foreign language developed better reading skills than those who were studying in Finnish. This study included 134 pupils in CLIL. It is to be noted that teaching in a foreign language accounted for approximately 20% of all teaching. Moreover, the pupils had studied a few years in CLIL. The research made by Rahman (2001) also supports the above conclusion that the pupils’ skills in their native language develop well in CLIL. The study looked into the spelling of compound nouns of sixth graders in CLIL. Pupils in CLIL succeeded better than pupils learning in their mother tongue when they had to write essays in which the focus was on the correct spelling of compound nouns. Affective factors The self-concept of pupils in language-oriented classes has been studied with 7–10-year-old pupils and it has been observed to be very positive. The study also looked into the linguistic self-concept of the pupils, which was very positive as ¨ well. Gender differences were minimal (Hyvonen & Lahdenranta, 1994: 112– 116). School motivation has been studied in immersion classes, pupils in CLIL and in so-called normal classes. The pupils’ foreign language was Swedish and they were in the second or third grades. The immersion pupils showed the most motivation. Pupils in CLIL were more motivated than the pupils being taught in Finnish. All classes explained success mainly with internal factors. Pupils in CLIL attributed their success to competence whereas those in immersion and in Finnish classes emphasised the will to try. Pupils in Finnish-speaking classes often linked success to the task being easy, which indicates feelings of

332

Language and Education

helplessness and low motivation (Filppula, 1996). Heinil¨a and Paakinen (1997) have reached similar results as well. Pupils in CLIL enjoyed studying in a foreign language and they were motivated to study a foreign language. It is worth noting that the studies on immersion and CLIL are not wholly comparable, as these teaching methods are somewhat different. CLIL practices also vary by school and research results concerning immersion are not totally applicable to CLIL, even though there are similarities. The results cannot be aligned to situations in which a child is learning in a different language but where no attention is paid to the development of his mother tongue skills. Many researchers in the USA, the Scandinavian countries and Germany have observed that when children participate in bilingual teaching as immigrants or as members of a linguistic minority group, their performance is not as good as that of their peers (cf. Hoffmann, 1991). However, these pupils can reach the same level as pupils learning in their native language. This requires that pupils have obtained bilingual teaching for 4–7 years (Thomas & Collier, 2002).

The Purpose of the Study This study (Seikkula-Leino, 2002) had two major purposes. The first aim was concerned with answering the following question: ’How well have CLIL pupils learned content compared to non-CLIL pupils and how well specific subjects such as mathematics and Finnish language as a mother tongue, were learned viewing content from this standpoint?’ Learning was presented in terms of achievement level, which was described as the relationship between measured levels of intelligence and school success. The purpose for adjusting school success to intelligence was to look at learning more individually. The second objective of this study was to examine the affective factors of learning. The aim was to assess CLIL pupils’ affective learning factors like motivation and self-esteem. Motivation was presented in terms of causal attributions as well as instrumental, integrative and cognitive orientation towards foreign language learning. In addition, the general self-esteem of the pupils and their academic and foreign language self-concept were taken into consideration.

Theoretical Framework The theoretical framework of the study first focused on general learning and then on different aspects of foreign language and CLIL. Learning in CLIL was seen in this study mainly as a constructive learning process (Julkunen, 2002; Meril¨ainen, 2002; Rauste von Wright, 1997) where the acquisition of a foreign language and the communicative use of it formed the basis for language learning (e.g. Ellis, 1985; Krashen & Terrell, 1984). Learning was presented in terms of achievement level, which is described as the relationship between measured levels of intelligence and school success. If a pupil does not reach the level of school success predicted by his intelligence, he could be defined as an underachiever and in the opposite situation he could be defined as an overachiever. A pupil whose achievements are in line with his intelligence level can be called an achiever according to his abilities. In this study, intelligence was seen as theoretically grounded to the traditional intelligence theories such

CLIL Learning: Achievement Levels and Affective Factors

333

as Spearman (1904, 1927) and Thurstone (1938) but also to some new theories like Sternberg (1986) who stresses more on the active process of intelligence. Moreover, the theoretical framework was built on the affective factors of learning, like Borba’s (e.g. 1989, 1994) self-esteem theory, Burns’ (1979, 1982) and Shavelson et al.’s (1976) self-concept theory, Laine and Pihko’s (1991) self-concept theory of language learning and Weiner’s (1974, 1984, 1992) causal attribution theory. According to Borba (e.g. 1989, 1994), self-esteem consists of five basic building blocks: security, selfhood, affiliation, mission and competence. The study sought for the self-concept of the pupils in the said sectors of self-esteem. Thus, it was possible to demonstrate what kind of self-esteem pupils in CLIL have. Self-esteem is closely linked to self-concept. The main difference between self-concept and self-esteem is that the latter is also connected to the individual emotional factors. As opposed to self-esteem, self-concept is a more objective description of oneself. Self-concept includes, for instance, social, physical and emotional self-concept as well as a learning self-concept (Burns, 1979, 1982; Shavelson et al., 1976). The section of the learning self-concept included in this study is the foreign language self-concept. The foreign language self-concept includes all the knowledge, ideas, ideals and evaluation of oneself as a language learner – the instrumental, integrative and cognitive orientation towards foreign language learning (Laine & Pihko, 1991: 15–16). In addition to self-esteem, the aim was, hence, to highlight the foreign language self-concept of the pupils in CLIL in more detail. The causal attribution theory of Weiner (1974, 1984, 1992) is based on the interpretation of an individual’s way to explain his successes and failures. These successes and failures can be explained through various causal attributes. Success at school can be evaluated in the following ways: I succeeded because I was good; I failed because I did not try hard enough. Especially, if the outcome is unexpected, people tend to look for reasons and explanations. An individual attributes his success/failure mainly to four different factors: competence, attempt, the level of difficulty or facility of the task and coincidence. The various models of explanation of failure and success have an impact on the goal-oriented performance and hence, on the lowering, maintenance or rise of motivation. This study aimed at assessing the motivation of pupils in CLIL from the standpoint of causal attribution.

Methodology The study subjects included 217 pupils from grades 5 and 6 in a Finnish comprehensive school; 116 of them were enrolled in CLIL classes. There were ten classes in this study. In the CLIL classes, 40–70% of instruction was carried out in English. The starting point for defining under- and overachievers was to highlight how well the pupils fared in different teaching methods in view of their competence (i.e. their potential intelligence). Thus, it was possible to control the way pupils may have been selected to the CLIL classes (the majority of pupils had been selected to the CLIL classes through entrance examinations and thus, it is probable that their school performance is higher than average). Regression procedures have been carried out in this study, in order to make

334

Language and Education

accurate inferences about achievements on the basis of intelligence. Hence, the pupils were aligned and their actual learning could be studied more accurately. The intelligence tests used were Raven’s non-verbal intelligence test and Wechsler’s vocabulary test, which measures mainly verbal intelligence. Tests concerning school success were tests on mathematics and Finnish as a mother tongue. The mathematics test was based on the national mathematics assessment aimed at sixth graders and it was modified to suit the purposes of this study to be appropriate also for fifth graders. The school success test of Finnish as a mother tongue was drafted on the basis of the international reading skill assessment of the IEA. Since this study aimed at looking into the writing skills of the pupils as well, the fairly wide-scope open question section of the reading comprehension test was used for that purpose. In order to assess overall school performance, the pupils’ grades both in Finnish as a mother tongue and mathematics were also taken into consideration. The indicators for self-esteem, lerning self-concept and foreign language selfconcept were drafted in the Likert form (scales 1–5). The self-esteem indicator was based on the theory of Borba. The learning self-concept indicator was based on the self-concept indicator created by Burns and Shavelson. The foreign language self-concept indicator was created by adjusting the indicator drafted by Laine and Pihko (1991: 124).

Results Grouping individual achievement levels and content learning in CLIL Pupil’s individual achievement levels were classified in this study and pupils were divided into three different groups: underachievers, achievers and overachievers. Differences in achievement levels between CLIL and non-CLIL classes can be seen in Figure 1. In conclusion, it can be said that there were no significant differences in general learning between the two groups, CLIL and non-CLIL classes, when comparing them on the basis of their achievement levels. A CLIL pupil could just as well

Figure 1 Pupils’ different achievement levels in CLIL and in non-CLIL classes

CLIL Learning: Achievement Levels and Affective Factors

335

Figure 2 Pupils’ different mathematical achievement levels in CLIL and in non-CLIL classes (n = 217)

be an underachiever as a pupil who had attended teaching where the mother tongue had been used as the language of instruction. However, some differences could be detected, for example, CLIL classes were more heterogeneous than nonCLIL classes. Moreover, the results also indicate that intellectually weak, quite weak, good and excellent pupils could perform as well in both forms of teaching. On the other hand, significant differences existed when comparing the pupils’ mathematical achievements, which is illustrated in Figure 2. Figure 2 illustrates the significant difference ( p = 0,000***) between the groups. There were more mathematical achievers in CLIL classes than in non-CLIL classes. Furthermore, CLIL pupils tended to overachieve less than pupils in non-CLIL classes. According to this study, pupils in CLIL classes could learn mathematics according to their intelligence but in non-CLIL classes pupils are more likely to perform above their potential intelligence. There was no significant difference between CLIL and non-CLIL pupils in their learning of their mother tongue. However, data on tests demonstrated that in non-CLIL classes, pupils were strongly overachievers, meaning that pupils overachieved in both subjects – Finnish language and mathematics, whereas in CLIL classes pupils who overachieved were either overachievers in the mother tongue or in mathematics, but not in both subjects. Affective factors Self-esteem No significant differences were found in measured self-esteem between the two groups. However, CLIL pupils felt that they had worse knowledge of foreign languages than pupils in non-CLIL classes. CLIL pupils also evaluated themselves as weaker foreign language learners than pupils in non-CLIL classes ( p = 0,000***).

Motivation CLIL pupils demonstrated strong motivation to learn in general, including the learning of foreign languages despite their low self-esteem in relation to that

336

Language and Education

of pupils in non-CLIL classes. Pupils in CLIL still wanted to achieve more external goals than internal ones ( p = 0,040*) even though motivation for reaching internal objectives seemed to develop by age.

Discussion The learning of CLIL pupils did not differ significantly from the way pupils learn when they study in their mother tongue. Yet some differences can be observed: the results seem to indicate that pupils who are taught in their native language tend to overachieve more strongly than those in CLIL. Moreover, it can be observed that the pupils who are strongly overachievers, i.e. both in Finnish as a mother tongue and in mathematics, are hardly present among CLIL pupils. Hence, teaching in a pupil’s mother tongue provides the pupil with more opportunities to reach maximum results. However, one must bear in mind that CLIL pupils learn a foreign language to a very high standard, which is unlikely to happen to such extent in teaching conducted only in Finnish. When the learning results are separately observed in Finnish as a mother tongue and mathematics, this study shows that as regards Finnish as a mother tongue, the pupils’ learning results were similar in both Finnish and CLIL classes. CLIL pupils overachieved even more strongly than those in Finnish teaching, even though the difference was not remarkable. Another outcome worth mentioning is that in one class where more than 75% of all teaching was conducted in a foreign language, 29% of the pupils were overachievers. The class in question was very large and the pupils had not been selected on the basis of general competence; the requirement for enrolment had been a good knowledge of the English language. The classroom teacher of this class was a native speaker of English who never addressed the pupils in Finnish. The formal teaching of Finnish as a mother tongue was given by a specialist teacher of Finnish as a mother tongue. Hence, the results of this study strongly support the idea that the mother tongue skills of pupils learning in a foreign language were not weaker than the skills of those learning in their mother tongue. Moreover, it must be noted that both weak and highly intelligent pupils had learned Finnish as a mother tongue to a fairly similar level: there were no great differences between underachievers, those learning according to their competence and overachievers in the groups of intellectually weak, quite weak, quite good and very good pupils. In view of previous research on the development of a pupil’s mother tongue skills either in CLIL or in Finnish teaching, it can be said that this study supported the results obtained by, for instance, Cummins (1995), Genesee (1988), Gaya (1994: 38–42), Meril¨ainen (2002) and Vesterbacka (1991b: 166–168) in their research on immersion as well as the research conducted by Merisuo-Storm (2000, 2002) and Rahman (2001) on CLIL, which showed that the learning of a pupil’s mother tongue is not disturbed by the participation in CLIL. The previous paragraph brought into consideration the research results concerning one class, showing that some pupils may demonstrate very strong developments in their mother tongue even though their teaching is conducted mainly in a foreign language. Such results remind us of those obtained by Sampera (1994) and Vesterbacka (1991a: 24) according to which the skills of

CLIL Learning: Achievement Levels and Affective Factors

337

immersion pupils in their mother tongue may even generally speaking exceed those of other pupils. It is possible that the pupils’ thinking and metalinguistic skills have developed as they have had to operate between two languages (cf. Cummins, 1995). However, we must note that the class in question also included slightly more underachievers than other classes in average. This provides a reason to suspect that some pupils may not have had linguistics skills strong enough to cope with such a wide scope teaching in a foreign language. On the other hand, several other reasons can be traced behind such underachievement (cf. Butler-Por, 1987), wherefore, in this case, it is difficult to draw any conclusions as regards the underachieving pupils in question. The results indicate that pupils in CLIL had fairly good mathematical skills. Both weak and talented pupils had learned relatively well according to their level. However, it is possible that the teaching of mathematics in a foreign language has an impact on how well the pupils manage to proceed in the subject and hence, it may not be possible to reach top results in relation to the potential competence of the pupils. Such results are supported by earlier studies on immersion which have shown that the mathematical skills of the pupils were almost similar in the immersion classes and in the so-called normal classes (cf. Cummins, 1995; Genesee, 1988). Nevertheless, this study does not support the assumption that pupils studying in a foreign language would succeed better in mathematics than pupils who study maths in their mother tongue (cf. Sampera, 1994). To some extent, it can be assumed that CLIL pupils do not reach maximum results in mathematics (cf. Gaya, 1994: 30–35; Qiang 2000). Pupils themselves report to have experienced problems in understanding the teaching of mathematics (Julkunen, 1999, 2000). Moreover, the teachers’ section of this study indicated that the teachers sometimes feel they have problems reaching the basic goals. Since the medium of teaching is a foreign language, the way in which a pupil learns the foreign language may have an impact on their ability to learn mathematics also. The affective filter has an impact on learning a foreign language (Krashen, 1982, 1985; Ellis, 1990): when teaching is conducted in a foreign language, it is probable that the affective filter has also an impact in the learning of content. In addition, pupil’s skills in the foreign language itself steer the learning of content. Even though the pupils in CLIL did not differ from their peers studying in their mother tongue, Finnish, as regards self-esteem, it was notable that their self-evaluation was more clearly marked by self-criticism. Learning in a foreign language may have steered the development of the pupils’ self-esteem to some extent, especially as the foreign language self-concept of the pupils was significantly weaker in relation to that of the pupils studying in Finnish. The pupils in CLIL felt much weaker in their understanding, reading, writing and speaking skills of the foreign language than the pupils who were studying in Finnish. Moreover, they felt that they were generally weaker as language learners than the pupils who were taught in Finnish. These results are remarkable especially when we take into consideration that in CLIL classes the pupils indeed have a good knowledge of the foreign language. Pupils in CLIL are forced to face the difficulties involved in learning content in a foreign language. As the teachers involved in this study have described, CLIL is very demanding for the pupils. Hence, learning in CLIL is in principle different

338

Language and Education

in the way that the so-called meaningful learning, which is one starting point for ¨ 1981; Novak & Govin, 1995: 182), steers the way constructivism (cf. Engestrom, learning is constructed when taught in a foreign language. This means that at the same time as the pupil is learning content, he is also resolving communicative problems which are linked to operating in a foreign language. For the process to function, the pupil needs to be extremely active in the learning situation. It can also be seen that CLIL inevitably involves plenty of language that is above the current competence of the pupil (‘input-hypothesis’, cf. Ellis, 1990; Krashen, 1982; 1985: 2–3). Hence, the pupils may not typically always know all the words used in teaching. This makes CLIL a complicated and demanding situation for the pupil, requiring a good ability to concentrate (cf. R¨as¨anen, 1994). It is possible that the pupils may sometimes feel incompetent and inadequate, which surely affects their ideas of themselves as learners of a foreign language (cf. Laine & Pihko, 1991: 15–16). The experiences of failure in a certain specific field correlate significantly to a certain section in one’s self-concept (cf. Gage & Berliner, 1979). Thus, the pupils create a self-concept that is somehow biased in relation to the real self and illustrates the individual’s cognitive beliefs of himself (cf. Mc David & Harari, 1968). This leads to successful individuals considering themselves failures (Epstein, 1980: 96; Kagan & Havemann, 1980). ¨ The self-esteem of pupils in CLIL has not been studied much. Hyvonen and Lahdenranta (1994) have conducted a study related to this area and it has yielded results different from the current study. Their work observed that pupils in CLIL had a very strong self-concept and especially their foreign language self-concept was remarkably positive. However, the pupils involved in their research were 7–10-year-olds, which means that their results are not totally comparable to the ¨ results of the current study. The pupils in the study of Hyvonen and Lahdenranta were still clearly in their childhood, whereas the current study included pupils who were already in their early adolescence, which has an impact on their ¨ feeling of self and the development of their self-esteem. In addition, Hyvonen and Lahdenranta used a fairly small study group. All in all, this study indicates that pupils in CLIL were somewhat more motivated to study and to use a foreign language than pupils learning in Finnish. However, let us note that their enthusiasm was relatively moderate. The research results on the motivation of pupils in CLIL are more or less aligned to those obtained in some previous studies (cf. Filppula, 1996; Heinil¨a & Paakkinen, 1997).

Conclusion There are good opportunities to learn in CLIL, as pupils with an under or above average level of intelligence have learned content, even though CLIL teaching was implemented extensively in a foreign language. The results in this study did not support the conclusion that the development of one’s mother tongue could be negatively affected through the use of foreign languages. On the other hand, learning in CLIL can be so challenging that the maximal outcome of content learning is not always reached. It is a matter of choice for a student to study in CLIL because in content and language integrated learning there are rather good possibilities to develop the knowledge of a foreign language.

CLIL Learning: Achievement Levels and Affective Factors

339

The integration of a foreign language with content learning creates special challenges for learning. Furthermore, affective factors influence learning situations differently in CLIL than in non-CLIL classes. Pupils in CLIL had relatively low self concept in foreign languages. Therefore, CLIL teachers should be aware of the possibility of a weak self-concept in foreign languages among CLIL pupils and pay attention to giving some positive feedback about the pupils’ knowledge of a foreign language etc., thus, possibly strengthening pupils’ motivation for foreign language learning, even though in this study the results of motivation were rather positive. Besides this work, other studies have provided positive results concerning CLIL education in Finland in general. They have given an impetus to develop CLIL education both administratively and practically in Finland. CLIL and immersion teaching will be an official part of the Finnish comprehensive school system, following the new curriculum reform, which was in process during 2004 – 2006 and provides the formal norms for arranging content and language integrated teaching and immersion teaching. The curricula that are drafted locally have to be based on the national norms. In Finland, this curriculum reform will give a view of the future where content and language integrated teaching will be continued in the comprehensive school system and where its position is also administratively supported.

Suggestions for Further Study Especially from the point of view of CLIL it would be interesting to study the way pupils’ general thinking and metaskills develop in content and language integrated learning. This would yield more information about how pupils learn in CLIL. Moreover, possible research on pupils’ overall learning including a deeper perspective on the involvement of affective factors would complement on the results of this work. Correspondence Any correspondence should be directed to Dr Jaana Seikkula-Leino, Faculty of Education, Teacher Training School, Turku University, PL 13, Annikanpolku 9, 20611, Turku, Finland (jaana.seikkula-leino@utu.fi). References Borba, M. (1988) A K-8 Self Esteem Curriculum for Improving Student Achievement, Behavior and School Climate. Torrance: Jalmar Press. Borba, M. (1994) Home Esteem Builders: Activities Designed to Strenghten the Partnership between Home and School. Torrance: Jalmar Press. Burns, R.B. (1979) The Self-Concept Theory Measurement, Development and Behavior. New York: Longman. Burns, R.B. (1982) The Self-Concept Development and Education. London: Holt, Rinehart and Winston. Butler-Por, N. (1987) Underachievers in School. Issues and Intervention (pp. 18–24). Bath: Bath Press. Cummins, J. (1995) Canadian French immersion programs: A comparison with Swedish immersion programs in Finland. In M. Buss and C. Laur`en (eds) Language Immersion: Teaching and Second Language Acquisition from Canada to Europe (Research No. 192, Linguistics 30) (pp. 7–20). Vaasa: Vaasa University.

340

Language and Education

Dodson, C.J. and Thomas, S.E. (1988) The effect of total L2 immersion education on concept development. Journal of Multilingual and Multicultural Development 9 (6), 467– 485. Ellis, R. (1985) Understanding Second Language Acquisition (2nd edn). Oxford: Oxford University Press. Ellis, R. (1990) Instructed Second Language Acquisition (pp. 57–59). Oxford: Blackwells. Elomaa, M. (1996) Kielikylpylasten ensikieliset kirjoitelmat nelj¨annell¨a luokalla. In M. Buss and C. Laur´en (eds) Kielikylpy: Kielitaitoon K¨ayt¨on Kautta (Research No. 13) (pp. 71–80). Vaasa: Vaasa University. ¨ Engestrom, Y. (1981) Mielek¨as oppiminen ja opetus (pp. 8–9). (Publication Series B: 17). Helsinki: Government Stationary Office. Epstein, S. (1980) The self-concept: A review and the proposal of an integrated theory of personality. In E. Staub (ed.) Personality: Basic Aspects and Current Research (pp. 81–132). Englewood Cliffs, NJ: Prentice-Hall. Filppula, H. (1996) Koulumotivaatio kieliluokassa ja kielikylpyluokassa. Unpublished EdMa thesis, University of Turku, Teacher Training School. Gage, N.L. and Berliner, D.C. (1979) Educational Psychology (pp. 187–189). Chicago: Rand Mc Nally. Gaya, A.B. (1994) Evaluating immersion programmes: The Catalan case. In C. Laur´en (ed.) Evaluating European Immersion Programs: From Catalonia to Finland (Research No. 185, Linguistics 27) (pp. 24–27). Vaasa: Vaasa University. Genesee, F. (1976) The role of intelligence in second language learning. Language Learning 26, 267–280. Genesee, F. (1987) Learning through Two Languages: Studies of Immersion and Bilingual Education (p. 180). Cambridge, MA: Newbury House. Heinil¨a, H. and Paakkinen, J. (1997) Vieraskielisen opetuksen arviointi peruskoulun ala¨ asteella: Kielenoppimisen n¨akokulma. Unpublished EdMa thesis, University of Turku, Teacher Training School. Hoffman, C. (1991) An Introduction to Bilingualism (pp. 118–119). London: Longman. ¨ Hyvonen, T. and Lahdenranta, P. (1994) Puolalan ala-asteen kieliluokan oppilaiden min¨akuva ja asenteet. Unpublished EdMa thesis, University of Turku, Teacher Training School. ¨ H¨am¨al¨ainen, M. (1998) Aidinkielen kehittyminen vieraskielisess¨a opetuksessa. Sanavarastotutkimus. Unpublished EdMa thesis, University of Turku, Teacher Training School. Julkunen, K. (2000) Pupils’ opinions on content and language integrated learning (CLIL). In K. Julkunen and A. Haapala (eds) Learning and Instruction in Multiple Context and Settings II. Proceedings of the Third Joensuu Symposium on Learning and Instruction. Faculty of Education (Research No. 79, Faculty of Education) (pp. 134–145). Joensuu: Joensuu University. ¨ oppiminen ja opettaminen vieraalla kielell¨a. In M-L Julkunen, K. (2002) Sis¨altojen Julkunen (ed.) Opetus, Oppiminen ja Vuorovaikutus (pp. 97–108). Helsinki: WSOY. Kagan, I. and Havemann, E. (1980) Psychology, an Introduction (p. 452). New York: Harcourt Brace Jovanovich. Koivum¨aki, K. and Stara, P. (1994) Puolalan koulun kieliluokkien opetuksessa k¨aytett¨av¨an englannin kielen m¨aa¨ r¨a ja tilanteet sek¨a oppilaiden a¨ idinkielen taidot. Unpublished EdMa thesis, University of Turku, Teacher Training School. Krashen, S. (1982) Principles and Practice in Second Language Acquisition (pp. 30–32). Oxford: Pergamon. Krashen, S. (1985) The Input Hypothesis: Issues and Implications (p. 3). New York: Longman. Krashen, S. and Terrell, T. (1984) The Natural Approach. Language Acquisition in the Classroom (2nd edn). Oxford: Pergamon Press. Laine, E. and Pihko, M-K. (1991) Kielimin¨a ja sen mittaaminen. Publication Series of Institute of Educational Research A No. 47. Jyv¨askyl¨a: The University of Jyv¨askyl¨a. Laur`en, C. (2000) Kielen taitajaksi. Kielikylpy k¨ayt¨ann¨oss¨a◦ (pp. 128–134). Jyv¨askyl¨a: Atena. ¨ P. (1993) Spr˚akbadet/Svenska Spr˚akklassen i Abo. En kartl¨aggning. Avhandling Malmstrom, 10 sv. Flerspr˚akighet och didaktik (pp. 20–22). Vasa: Vasa Universitet.

CLIL Learning: Achievement Levels and Affective Factors

341

Mc David, J. and Harari, H. (1968) Social Psychology Individuals, Groups, Societies (p. 222). New York: Harper and Row. Meril¨ainen, M. (2002) Biologian perusk¨asitteiden omaksuminen suomen kielell¨a englanninkielisess¨a kielikylpyopetuksessa. Tapaustutkimus Hollihaan koulun 4. e-luokassa. Unpublished EdLic thesis, The Institute of Applied Linquistics, University of Jyv¨askyl¨a. Merisuo-Storm, T. (2000) Ensiluokkalaisen a¨ idinkielen lukemisen ja kirjoittamisen taitojen kehittyminen kielirikasteisella luokalla. Unpublished EdMa thesis, University of Turku, Teacher Training School. Merisuo-Storm, T. (2002) Oppilaan a¨ idinkielen lukemisen ja kirjoittamisen taitojen kehittyminen kaksikielisess¨a alkuopetuksessa (Publication Series C: 185). Turku: Painosalama. Novak, D. and Gowin, D.B. (1995) Opi oppimaan. (P. Lehto-Kaven, trans.) Tampere: Tammer-Paino. Original work published 1984. Qiang, Y. (2000) Bilingual Education, Cognitive Development and School Achievement. A Study of the Bilingual Programs for Tibetan Children. Institute of International Education. Stockholm: Stockholm University, Gotab. Rahman, H. (2001) Kaksikielisen (suomi-englanti) ja suomenkielisen kuudennen luokan a¨ idinkielen kirjoitelmien vertailua yhdyssanojen oikeinkirjoituksen osalta. Unpublished EdMa thesis, University of Helsinki, Teacher Training School. Rauste von Wright, M. (1997) Opettaja Tienhaarassa. Konstruktivismia K¨ayt¨ann¨oss¨a (pp. 17–19). Jyv¨askyl¨a: Atena. R¨as¨anen, A. (1994) Nyt opitaan vierailla kielill¨a. Opettaja 21, 16–17. Sampera, J.A. (1994) The Catalan immersion programme: Assessment and recent results. In C. Laur´en (ed.) Evaluating European Immersion Programs. From Catalonia to Finland. (Research No. 185, Linguistics 27) (pp. 13–26). Vaasa: Vaasa University. Seikkula-Leino, J. (2002) Miten oppilaat oppivat vieraskielisess¨a opetuksessa? Oppilaiden suoriutumistasot, itsetunto ja motivaatio vieraskielisess¨a opetuksessa. (Publication Series C:190). Turku: Painosalama. Shavelson, R.J., Hubner, J.J. and Stanton, G.C. (1976) Self-concept: Validation of construct interpretations. Review of Educational Research 46 (3), 407–441. Spearman, C. (1904) General intelligence, objectively determined and measured. American Journal of Psychology 15, 201–293. Spearman, C. (1927) The Abilities of Man. London: Macmillan. Sternberg, R.J. (1986) A triarchic theory of intellectual giftedness. In R.J. Sternberg and J.E. Davidson (eds) Conceptions of Giftedness (pp. 233–243). Cambridge: Cambridge University Press. Swain, M. and Lapkin, S. (1982) Evaluating Bilingual Education: A Canadian Case Study. Clevedon: Multilingual Matters. Thomas, W.P. and Collier, V.P. (2002) A national study of school effectiveness for language minority students’ long-term academic achievement. http://www.crede. org/research/llaa/1.1es.html-46k. Accessed 10.04.06. Thurstone, L. (1938) Primary Mental Abilities. Psychological Monographs, 1. Chicago: The University of Chicago Press. Vesterbacka, S. (1991a) Elever i Spr˚akbadskola, Social Bakgrund och Tidig Spr˚akutveckling [Summary, children in immersion programme, home-background and early language development: Issues and strategies]. Newbury Park, CA: Sage. Vesterbacka, S. (1991b) Kielen omaksuminen kielikylpyluokassa. In C. Laur´en (ed.) Kielikylpymenetelm¨a: Kielen K¨aytt¨o Mielekk¨aa¨ ksi (Tutkimuksia No. 1, T¨aydennyskoulutuskeskus) (pp. 64–77). Vaasa: Vaasa University. Weiner, B. (1974) Achievement Motivation and Attribution Theory. Morristown, NJ: General Learning Press. Weiner, B. (1984) Principles for a theory of student motivation and their application within an attributional framework. In E.E. Ames and C. Ames (eds) Research on Motivation in Education (pp. 15–68). London: Academic Press. Weiner, B. (1992) Human Motivation. Metaphors, Theories and Research. Newbury Park, CA: Sage.