Development of first language literacy skills in bilingual education
Paper presented at the European Conference on Educational Research, University of Hamburg, 17-20 September 2003
The study investigated the possible negative effects of bilingual teaching on the development of children's literacy skills. In the bilingual classes (Finnish / English) examined, approximately 20 % of the instruction was given in English. The control group studied exclusively in Finnish. At the beginning of first grade (at the age of seven), pupils' level of school readiness was significantly higher in the bilingual classes than in the monolingual classes. This was due to the entrance procedures for bilingual classes. After two study years, the literacy skills of the test group were significantly better than those of the control group. On the other hand, when observing pupils who started first grade with a poor level of school readiness, or those who started with an excellent level of school readiness, there was no significant difference in the development of literacy skills between bilingual and monolingual groups. Bilingual education has not had a negative effect on the pupils' development of literacy skills.
In Finland traditional language teaching has been criticised for providing the pupil with neither the necessary skills nor the courage to communicate with speakers of other languages. One method aimed at developing more versatile language skills is the use of a foreign language as the medium of instruction in other lessons than actual language lessons. In Finland there has been, in recent years, a vast increase in bilingual teaching, However, many are concerned that teaching through a foreign language has negative effects on the development of children's literacy skills in the mother tongue. The aim of my study was to find out whether the pupils' reading and writing skills develop equally well in bilingual as in monolingual classes.
Content based language learning
The terminology related to bilingual education is as yet unestablished. The term bilingual education indicates the presence of two languages in the school, but it is used to refer to various differently constructed models of instruction. (Baker 2000, 90-91) Other terms that are used to refer to bilingual models where language and content are integrated include content based language teaching, language-based content instruction, language sensitive content instruction and language- or content-enhanced teaching. In Finland the general term CLIL, short for content and language integrated learning, is used to refer to the Finnish models of bilingual education. (Järvinen 1999, 15-16)
The leading principle of Content based language learning (CLIL) is integrating language teaching with the teaching of other subjects. Pupils acquire information via a foreign language and simultaneously develop their language skills. Thus the focus is not on language learning but on acquiring the informational content of various subjects. (Brinton 1989, 2) Moreover, the language learning process is more efficient when a foreign language is used to refer to genuine subject matter. (Richards & Rodgers 2001, 207)
In CLIL the pupil is allowed to hear a sufficient amount of the foreign language and to use it in a meaningful context. In order to facilitate understanding and the language learning process, teaching methods are of an illustrative and concrete character. Teaching is more lucid and more concise than when it is carried out in the pupils' mother tongue. (Nikula & Marsh 1997, 52-54) Same concepts are brought up several times in different manners. Phrases and expressions related to the school day, repeated day after day, form an important part of the teaching, as do games, songs, stories and nursery rhymes, which are all efficient means of learning new words and phrases. (Merisuo-Storm 2002, 31)
In recent years there has been a vast increase in bilingual teaching in Finland. According to a study by Nikula and Marsh (1997), schools seek to make their language teaching more efficient in order to prepare their pupils for a society that lays a strengthening emphasis on internationality. CLIL enables the pupil, once he or she reaches university level, to participate in courses taught in a foreign language and to use the language in an accurate and versatile manner. The pupils' good command of foreign languages has its advantages also in the short run. Many schools have established international relationships. The number of immigrant and exchange students has rapidly increased, and many schools take part in international cooperation programmes. Moreover, many schools seek to increase their appreciation by offering bilingual teaching; in fact, many have succeeded because CLIL attracts interest and is increasingly popular. Parents encourage schools to initiate bilingual teaching, as they wish their children to obtain a good basis for language proficiency and courage to communicate in foreign languages.
The teaching through a foreign language is mostly implemented on a relatively small scale. In some schools the foreign language is used merely during one or two hours a week or in short sessions throughout the week. (Järvinen 1999, 24) At the moment the foreign language used as a medium of instruction is almost exclusively English. Nevertheless, the English language is practically the only common denominator for bilingual teaching in Finland. There is great variation in the amount of foreign language used, in the subjects of instruction selected to be taught in it, in teaching methods and materials, in the aims of the teaching, in the teachers' background, and in the admission procedures. (Mustaparta & Tella 1999, 13)
Varied opinions have been expressed about suitable selection criteria. Many hold the opinion that teaching in a foreign language does not suit all pupils. However, it appears to be difficult to define who such pupils are. The results of the research available are contradictory. The Finnish Ministry of Education recommends that it should be carefully considered whether bilingual teaching suits children who have linguistic learning difficulties or a poor command of their first language. (Marsh, Marsland & Nikula 1999, 37)
In Finland, the main cause for concern related to bilingual teaching is its possible negative effect on young pupils' literacy skills. A strong command of the mother tongue is considered the basis for all learning. Therefore in bilingual classes pupils are usually not taught to read or write in the second language until they have achieved a satisfactory level of literacy in Finnish. Furthermore, in first grade a large part of the subject matter is taught in both languages rather than exclusively in the foreign language.
Finnish and Canadian research has shown that immersion education also benefits the pupils' first language literacy skills. However, no research has been available about the effects of CLIL on literacy skills.
The Aim of the Study
In CLIL the content of all teaching has to be concised in order to leave sufficiently time for teaching through a foreign language. The risk that this involves is that there may not be sufficiently time left for the children to practise their literacy skills. The main aim of my study was to find out whether studying in a bilingual class affects the development of pupils' literacy negatively or whether these pupils achieve the same level as their peers who study exclusively in Finnish. Special attention was paid on the development of pupils who started school with either a poor or an excellent level of school readiness, in order to find some guiding principles as to whom CLIL suits, and to find out whether it can hinder some pupils' learning. Secondly, the study aimed to find out how successful those pupils who achieved excellent literacy skills, and those who achieved only a poor level, were in learning English.
The experimental group (N=78) consisted of three bilingual classes in three schools in South-Western Finland. In these classes approximately 20 per cent of all teaching was in English. The control group (N=58) likewise consisted of three classes, of the same schools; these classes studied exclusively in Finnish. The development of all these pupils was observed from the beginning of first grade to the end of second grade. In the spring of the second study year, two new bilingual classes (N=54) who started first grade in the same schools at the beginning of that study year were included in the study.
The study was quantitative and longitudinal, and it included three measurements at intervals of one year. The purpose of the initial measurement was to define each participant's individual starting level. The second measurement aimed to evaluate the pupils' literacy skills in the spring semester of the first study year and the third measurement at the end of second grade. The results of the second and third measurement were compared with those of the initial measurement. The other areas covered in the third measurement were the English skills of the experimental group and the literacy skills of the new bilingual first grade classes. The test results of the new first grade pupils were compared with those of the measurement performed one year earlier.
The aim of the study was to answer the following questions: (1) What is pupils' starting level at the beginning of first grade in bilingual and in monolingual classes? (2) How does the development of first language literacy skills of pupils studying in bilingual classes differ from that of pupils who study exclusively in Finnish? And (3) how do the English skills of those pupils who perform either very well or poorly in reading and writing tests develop during first and second grade?
Six tests were used as indicators. The initial test (Poussu-Olli & Merisuo-Storm 1999) was divided into five sections; each of these sections consisted of several items. The items in the general section were aimed at measuring the pupil's understanding of number, of sentences and of phoneme-grapheme correspondence, his or her ability to continue phrases and lines of patterns, and his or her ability to find synonyms and rhyming words. The other four sections measured auditive and visual perception, mathematical skills, and memory. The two reading tests (Merisuo-Storm & Poussu-Olli 2000 and 2001) were used to measure the accuracy and speed of reading both aloud and soundlessly, and reading comprehension. The test used in the second grade included tasks that dealt with synonyms and opposites, and rhyming words. The writing tests (Poussu-Olli & Saarni 1998, Merisuo-Storm 2001) included, apart from writing from dictation tasks, items that measured the pupil's ability to perceive phoneme-grapheme correspondence, auditive and visual perception, listening comprehension skills, and memory. The English test (Merisuo-Storm 2001), measured the pupil's ability to recognise written English words, and heard words and phrases.
Nearly all of the measurements were performed in class, only the reading aloud test for each pupil separately. Normally no more than one lesson was used during the same day for the measurements, although the reading aloud test occasionally extended to the following lesson. Thus the group tests required in first grade at least three separate occasions, and in the spring semester of second grade at least four. The test results were analysed using an SPSS program.
The initial test measured the pupils' school readiness, including their auditive and visual perception, mathematical skills, and memory. In this test both the total scores and the scores in each five sections of the test were considerably higher in bilingual classes. This was due to the entrance procedures for bilingual classes and to the family background of the children who applied to study in them. In the entrance procedures emphasis was normally laid on the applicants' linguistic skills; therefore it is not surprising that in the initial test the pupils who were studying in bilingual classes had considerably more success in the tasks that measured phonological and phonemic awareness. It should be pointed out that parents who wished their children to study in a bilingual class were often interested in languages and in reading and literature; they had given their children plenty of linguistic stimulation, and from them the children had adopted a positive attitude towards literature. However, it should also be pointed out that although the pupils studying in the bilingual classes showed, as a group, a higher starting level in regard to many of the skills and abilities that the tests measured, there were pupils also in the control group who showed excellent performance in the initial test.
The different starting levels of the two groups were taken into account when analysing the results of the other measurements. The study also looked at how pupils who showed a poor or an excellent starting level developed in bilingual and in monolingual classes. For this purpose the pupils were rank-ordered according to their total scores in the initial test and divided into quartiles of approximately the same number of pupils. As can be seen in Figure 1, most of the pupils who studied in bilingual classes fell into the highest quartile and the two middle quartiles, whereas pupils in the control group fell for the most part into the lowest quartile and the middle quartiles.
FIGURE 1. The percentual distribution of the pupils' total scores in the initial test in bilingual and monolingual classes
As the results of the first grade reading test show, the difference between the general levels of performance of the subject group and the control group did not diminish during the first study year; on the contrary, in the second semester the pupils studying in bilingual classes could read with greater speed and accuracy than the control group. These results were not exceptional: pupils studying in bilingual first grade in the following year achieved approximately the same level of reading speed and nearly the same level of accuracy as the previous group. However, within both the test group and the control group there was a huge difference between the highest and the poorest level of reading speed. Accuracy and speed are closely linked: pupils who are prone to errors are compelled to repeat words and to correct themselves often, and consequently their reading speed suffers. At the end of second grade the difference between the experimental group and the control group had further increased. The performance of the test group was significantly better especially in the reading comprehension test. Figure 2 displays the performance of both groups in the second grade reading comprehension test (E=excellent, VG=very good, G=good, F=fair, P=poor).
FIGURE 2. The percentual distribution of the pupils' marks in the reading comprehension section of the reading test in bilingual and monolingual classes
The reading tests included tasks that measured e.g. the pupils' auditive and visual perception, and memory. The initial test contained tasks that corresponded to these and in which the pupils of the bilingual classes succeeded considerably better than the control group. In the course of first grade the difference had somewhat diminished; this was predictable, in the light of earlier research. However, in the course of second grade the difference between the two groups' performance in regard to auditive perception and memory, surprisingly, increased. This is probably due to the fact that studying through a foreign language asks for more attentiveness; consequently the pupils learn to listen with attention. Furthermore, being compelled to try to learn to distinguish sounds, intonation and stress in a foreign language, these pupils had developed their auditory skills. Figure 3 shows the results of the auditive perception tasks.
FIGURE 3. The percentual distribution of the pupils' marks in the auditive perception section of the writing test in bilingual and monolingual classes
Memorising foreign words had developed the pupils' memory. Figure 4 displays the results of the memory tasks.
FIGURE 4. The percentual distribution of the pupils' marks in memory section of the writing test in bilingual and monolingual classes
The development of spelling skills shows the same tendency. In first grade there was no great difference between the test group and the control group in regard to the number of errors in the writing from dictation test, but at the end of second grade the pupils in the test group made considerably fewer spelling errors. The results of my study are consistent with the results of other research (e.g. Vacca et al. 1995, 148) that has shown a close parallel between reading literacy and writing literacy. Usually the same pupils were competent readers and writers. The percentual distribution of the scores of the writing from dictation test is presented in Figure 5.
FIGURE 5. The percentual distribution of the pupils' marks of the writing from dictation in bilingual and monolingual classes
Another question that I wished to answer was how the literacy skills developed in the case of pupils who showed, at the beginning of first grade, a poor or an excellent starting level. As was explained earlier, all participants were divided into four quartiles of approximately the same size according to their total scores in the initial test. Out of the pupils who fell into the lowest quartile, those who studied in bilingual classes achieved, at the end of second grade, better scores in the reading accuracy, speed, and reading comprehension tests than those who studied in monolingual classes. In the writing from dictation test those who studied in bilingual classes showed the same or a higher level of performance than those of the control group. The performance of pupils whose starting level was excellent did not differ greatly at the end of second grade regardless of whether they studied in a bilingual or a monolingual class.
Pupils' starting level seemed to have little effect on their performance in the English test. English skills appeared to be linked more closely with the pupil's level of first language literacy at the time of the English test; pupils who read and wrote well had been the most succesful in acquiring English. However, it can be stated that nearly all the pupils had acquired a considerable amount of English vocabulary over the two study years. They were also familiar with many phrases and capable of recognising written words. My impression in the test situations was that the pupils enjoyed using the English language and that they were proud of their knowledge of a foreign language. This supports the view that teaching through a foreign language can awaken childrens' interest in a foreign language and, consequently, make them eager to further develop their command of it.
My study, likewise earlier research, shows that there is a great variation in childrens' level of school readiness at the stage of entering primary school. This makes great demands on the teaching. The early stage of school life is important for the pupils' later development. Their experiences at this stage can determine their attitude towards school and their view of themselves as students. Defining the pupils' starting level is important in bilingual classes since studying in them is demanding, but it would be important to define all pupils' starting level at the beginning of first grade in order to give their school work a good start. An initial measurement could help the teachers to lay special emphasis on each pupil's individual needs; it would enable them to help each pupil develop his or her less advanced skills, and to allow all pupils to experience feelings of success. The measurement would also help teachers to meet the needs of pupils whose starting level is high; it would enable them to assign these pupils sufficiently demanding tasks and thus to allow them room for development and to increase their learning motivation.
The results of the study show that, even when the different starting levels of the two groups were taken into account, pupils' literacy skills developed at least equally well in bilingual education as when the pupils study exclusively in Finnish. Furthermore, studying through one language or two languages did not appear to have any significant effect on the development of a pupil's literacy skills when his or her starting level at the beginning of first grade had been poor or excellent. Pupils who study in bilingual classes were especially advanced in reading comprehension skills; thus they were well prepared for successful studies later on. The results of my study clearly indicate that there is a close connection between the development of literacy and second language acquisition; pupils who achieved a poor level of literacy were less successful in their English studies than fluent readers and writers. Nevertheless, even pupils whose success was only modest had succeeded in aqcuiring a considerable amount of English. All the pupils appeared to consider their English studies important and to have a positive attitude towards them.
The results of my study support the view that when most of the teaching is carried out in the pupils' first language, CLIL does not affect the development of first language literacy skills negatively. Children are capable of maintaining the two languages separate. Moreover, they achieve an awareness of their mother tongue as a language and a communication system comparable to other languages.
Home has a remarkable influence on children's level of school readiness. The study has shown that the first few years in a child's life have the greatest effect on his or her learning results both in the mother tongue and in foreign languages. A child's future success in school can often be predicted from his or her level of familiarity with written language and with language usage at the stage of entering school. If the parents have read aloud to their child and engaged him or her in conversation, his or her vocabulary and other linguistic abilities are likely to be well developed. The child has a positive view of himself or herself as a learner and a student, and this encourages him or her to learn more. A foreign language used in class will not cause him or her anxiety; instead, he or she will actively seek to understand it.
The influence of home diminishes little when the child enters school. Parents whose children study in bilingual classes are usually interested in their children's studies and seek to help them to succeed in them. This is made apparent by the fact that these pupils' parents attend parent-teacher meetings at school more often than other parents. Moreover, they encourage their children to read and to write also outside school, which has a positive effect on the children's development.
It can be concluded, from the results of my study, that it is worthwhile to continue and to further develop CLIL in Finland. If the teaching is carefully planned, it will not affect the development of first language literacy negatively. However, at the moment there is a great variation in the realization of CLIL in Finland. A standardised curriculum would facilitate, for example, the mobility of pupils between schools and the availability of suitable teaching materials. My study focused on the first two school years, but is important to research likewise the development of older pupils who study in bilingual classes; in such research it would be possible to compare pupils' learning results in bilingual classes and other classes in regard to not only their first language but, moreover, the foreign languages which, or through which, they study.
Baker, C. 2000. The Care and Education of Young Bilinguals. An Introduction for Professionals. Clevedon: Multilingual Matters.
Brinton, D.M., Snow, M.A. & Wesche, M.B. 1989, Content-Based Second Language Instruction. New York: Newbury House.
Järvinen, H-M. 1999. Acquisition of English in Content and Language Integrated Learning at Elementary Level in the Finnish Comprehensive School. Turku: Turun yliopisto.
Marsh, D., Marsland, B. & Nikula, T. 1999. CLIL: A review of current thinking. In D. Marsh & B. Marsland (eds.) CLIL Initiatives for the Millenium. Report on the CEILINK Think-Tank. Jyväskylä: University of Jyväskylä, 34-45.
Merisuo-Storm, T. 2002. Oppilaan äidinkielen lukemisen ja kirjoittamisen taitojen kehittyminen kaksikielisessä alkuopetuksessa. Turku: Turun yliopisto.
Merisuo-Storm, T. & Poussu-Olli, H-S. 2000. Peruskoulun ensimmäisen luokan lukemistesti Siiliperhe. Naantali: Scribeo.
Mustaparta, A-K. & Tella, A. 1999. Vieraskielisen opetuksen järjestäminen peruskoulussa ja lukiossa. Helsinki: Opetushallitus.
Nikula, T. & Marsh, D. 1997. Vieraskielisen opetuksen tavoitteet ja toteuttaminen. Helsinki: Opetushallitus.
Poussu-Olli, H-S. & Merisuo-Storm, T. 1999. Koulutulokkaan alkutesti. Naantali: Scribeo.
Poussu-Olli, H-S. & Saarni, M. 1998. Peruskoulun ensimmäisen luokan kirjoitustesti Siili. Naantali: Scribeo.
Richards, J.C. & Rodgers, T.S. 2001. Approaches and Methods in Language Teaching. Second Edition. Cambridge: Cambridge University
Vacca, J.A.L., Vacca, R.T. & Gove, M.K. 1995. Reading and Learning to Read. Third Edition. New York: Harper Collins College Publishers.
This document was added to the Education-line database on 27 October 2003