Print   |   Save

An Ethnographic Approach to Assessing CLIL

Prue Goredema
Humboldt University of Berlin


This article provides an overview of a comprehensive ethnographic study forming part of a doctoral project in English didactics undertaken at Humboldt University of Berlin. The longitudinal study focuses on thirty two Year 9 learners who are taking Biology and Geography with English as the language of instruction. Student progress is tracked over the course of one academic year in order to investigate in detail how Content and Language Integrated Learning (CLIL) is practiced. Focus areas include the social context, pedagogical considerations, the students’ perspectives and learning strategies, the nature of CLIL classroom discourse, and the quality of performance on written tasks when language and subject learning are combined. With an ethnographic approach, multiple data sources and methodologies are employed: qualitative approaches for surveys and interviews with participants, and quantitative analysis of the data gathered from the students’ manuscripts. The study is on-going.


Keywords: Sociolinguistics, SLA pedagogy, discourse analysis, academic language proficiency

1. Introduction

In the Berlin secondary school in which this study is set, 32 Year 9 pupils are studying Biology, Geography, and eventually History through the medium of English. In Germany, a strategic drive to improve student competence in languages accelerated the spread of bilingual teaching programmes that had been cautiously introduced under different socio-political conditions in the late 1960s (Zydatiß, 2007: 1). CLIL in Germany has largely enjoyed a positive reception. Some language practitioners argue that language acquisition is enhanced when students are exposed not so much to the grammar of the target language, but when they are afforded the opportunity to use the language in practice (Marsh, 2001).  Studies comparing the performance of CLIL learners to that of students in mainstream programmes where English as a Foreign Language (EFL) is not combined with a content subject have largely yielded encouraging conclusions about the effectiveness of CLIL in fostering target language competence (Zydatiß, 2007; Wannagat, 2009; Várkuti, 2010).  In the German Institute for International Educational Research (DIPF) study of 2003-2004, the language skills of around 11,000 students at 220 schools were assessed through various methods, including classroom testing and surveys (Deutsches Institut für Internationale Pädagogische Forschung:1). The findings suggested that there is much room for improvement in English language instruction if students are to attain the requisite standards.

Subsequent research into CLIL theory and methods has been fruitful. Lyster takes a counterbalanced approach, arguing for the need to “[systematically integrate] content-based and form-focused instructional options” (2007: 3-5), and corpus-based approaches to language pedagogy are gaining ground (Granger, Hung, and Petch-Tyson, 2002). Nevertheless, an all-encompassing study specific to the German context is lacking, and therefore this avenue of exploration is being pursued in the present study. As Kiely asserts, there remains a number of “challenges to producing a clear picture of the learning value of initiatives such as CLIL through cohort-based summative assessment profiles” (2009:,para. 29). Descriptions of classroom dynamics and language performance become more meaningful when placed in their social context. So whilst identifying the nature of the second language acquisition prevalent in the CLIL classroom is a major point of interest, it is equally important to examine the social and pedagogical realities that inform the entire process.  Effective CLIL research necessarily entails enquiry into a number of disciplines, “uniting a much wider field of research than is associated with language learning per se, including learning theories, language learning theories, intercultural and social processes and [providing] a lens through which integrated learning can be interpreted” (Coyle, 2007: 558).  This list is of course not exhaustive.

Page 57
Page 58

Several avenues of enquiry have purposefully been pursued in order to derive a comprehensive picture of how CLIL is carried out in practice. Firth and Wagner (1997) argue for the need for Second Language Acquisition (SLA) research to include the social interaction dimension so that contextual information forms part of the final analysis. It is an idea favoured by Bakhtin (1986) where he notes that the sociocultural, historical, and economic environment feed into the acquisition and use of language. Language cannot be studied in isolation, divorced from its context. The present study is longitudinal, tracking the progress of 32 Year 9 learners over the course of one academic year (2010-2011). Describing CLIL in context is a key aim, and thus a “thick description” as defined by Clifford Geertz will be presented in this otherwise mixed-methods investigation incorporating multiple research tools (1973: 5).

Focus areas include:

  • The social context in which the additional language is being learnt
  • Pedagogical considerations: the nature and scope of the challenges that the teachers face, and the novel approaches undertaken in conducting CLIL classes
  • Student perspectives and the role of learner autonomy in negotiating meaning
  • Teacher-student interaction in the construction of content and language knowledge
  • The development of lexico-grammatical performance as manifest in the students’ written assignments

Although the study is on-going, preliminary findings suggest that this broad, ethnographic approach will provide insight into the intricacies of integrating content and language learning.

2. Societal context

Before delving into the classroom study, it is necessary to take stock of the social environment in which this CLIL programme is situated as this is a key aspect of ethnography. In 1995, the European Commission issued a white paper entitled Teaching & Learning: Towards a Learning Society stating that apart from their mother tongue, citizens should learn a further two European languages because “proficiency in several Community languages has become a precondition if citizens of the European Union are to benefit from the occupational and personal opportunities open to them in the border-free Single Market” (EUR-Lex, 1998).  It is in this socio-political setting that CLIL has taken root (Marsh 2002).  Whilst CLIL programmes throughout the continent are undertaken in different languages, in Germany, much like in the rest of Europe, English is by far the preferred language for bilingual instruction, accounting for 60%. The other significant languages used are French (11 %) and Spanish (3%) (Vásquez, 2007: 97). Clearly, the role of English as the de facto language of international commerce, trade, science, and diplomacy is reflected in these figures. Marsh, Maljers and Hartiala (2001: 19) attribute this to “the pressures of global mobility and European integration.” It would appear that market forces are currently determining the interpretation of the EU language policy. Unabated, the spread of English may pose a threat to other languages, thus bringing forth questions of justice, democracy, and social inclusion. Those not competent in the language of prime importance would surely soon find themselves at the fringes of society and unable to participate in the political, social, and economic discourse that determines the continent's common future. Robert Phillipson states that in much of Europe, competence in English is becoming a prerequisite for access to higher education and employment, in tandem with preferred forms of communication in a national language (in Craith, 2007: 70). Phillipson (2003: 7) also questions whether “the language policies in place at the national and international levels will ensure that cultural and linguistic diversity are maintained and will flourish” (2003: 7). Whilst the relative merits of these language education policies are enthusiastically and vociferously debated by researchers and policy makers, those who are ultimately to be served by these strides are largely unaware of the language policy goals and how they are to be actualised. A survey of prevailing attitudes towards language learning within the school community was considered but ultimately abandoned in favour of a literature-based assessment of the societal context.

Page 58
Page 59

3. Institutional setting

With the peculiarities of the social context elucidated through the survey, the next portion of the ethnographic study entails examining how CLIL has been applied at this particular school. The schooling system in Germany is distinct in its four streams, each leading to different school leaving qualifications: Hauptschule, Realschule, Gesamtschule, and Gymnasium. Thelatter is thetop tier where on successful completion of the thirteenth year of schooling, students are granted the Abitur, a qualification that is a prerequisite for university study (German International School, Lohmer and Eckhardt, 2010). The relative merits or injustices of this system are beyond the purview of this discussion.  What is important to note is that the study is being conducted at a Gymnasium.According to Wolfgang Hallet, classical CLIL programmes are more commonly carried out in Gymnasium and Realschule schools, where the express aim is “to increase [students’] possibilities in the labour market and the acquisition of bicultural competence” (Vásquez, 2007: 99). The suggestion here is perhaps that it is in the Gymnasium stream that the innovative CLIL programme is likely to find fertile ground.

Nevertheless, CLIL is not a monolith, one-size fits all endeavour: There are different models, ranging from single modules offered over a few weeks, to intensive programmes where several subjects are taught in the target language over a number of years. The school where the study is situated is a centrally located state-run secondary school. With enrolment close to 900 students, it is a multicultural school where the national language—German—isn’t necessarily the home language of all the students. Exchange programmes with China and Palestine are offered annually, and there is a wide assortment of extracurricular and sporting activities, thus offering students stimulation outside of the formal environment of the classroom. Since the school had long had a culture of providing comprehensive learning routines, introducing CLIL was a natural progression, and it has been offered since the 2000/2001 academic year. The CLIL subjects are introduced cumulatively, starting with Geography in Year 8, Biology in Year 9, and History in Year 10. The students are thus given ample time to adapt to the approach. Prior to taking CLIL classes, students have received several years’ of English as a second language instruction. For some students in the sample, English was first introduced in Year 3; however, it was only in Year 5 that the focus on grammar became more pronounced as is typical in Berlin schools (Schwabe, 2010). The suggestion here is that despite the novelty of entering the CLIL classroom, the students are nevertheless familiar with the foundational aspects of the language of instruction.  In order to be considered for the CLIL class at this school (hereafter S), the students generally need to have maintained at least a B average in the preceding school year. The class therefore consists, in large part, of academically gifted students, or at least those who routinely work hard enough to gain top marks. Nevertheless, the CLIL approach calls for a rethinking or adaptation of some teaching methods that ordinarily serve well in language and content lessons alike.

4. Pedagogical concerns

The ethnographic approach taken allows for a detailed description of how the CLIL teachers use English to teach the content subject, in this case Geography and Biology. At S, these subjects are taught by two teachers, hereafter TG (Geography teacher) and TB (Biology teacher), both of whom are voluntarily cooperating with the researcher by sharing their time, insight, and experiences with this approach in the didactics of English. There is no especially designed CLIL syllabus: the content to be taught is that which is taught to all students in the Berlin administrative area, and as such the syllabus is devised with L1 (German) instruction in mind. This poses a number of practical and methodological challenges for the CLIL teacher since the material must be adapted for the English-medium CLIL classroom without the loss of substance, depth, and nuance. First published by the Berliner Landesintitut für Schule und Medien in 2006, the current syllabus stipulates that the goal of education has long been for learners to acquire knowledge for the successful completion of everyday tasks and tasks they will face later on in their careers.    The Biology syllabus is designed to adhere to this goal, and it encompasses three competence areas: awareness of the theoretical and scientific principles taught, competence in communication, and the ability to use scientific learning to evaluate phenomena in practical, technical, and social contexts (Berliner Landesintitut für Schule und Medien, 2006). Regardless of the language of instruction, these goals must ultimately be met; therefore, in assessing CLIL, there is a need to move beyond a mere description of the practices in place towards evaluating their efficacy.

Page 59
Page 60

The participating CLIL teachers, TG and TB, are quick to point out the fact that the subject—for example, Biology—is the prime focus of the lesson. The use of English is almost incidental. This leads to what might be dubbed the CLIL paradox: striving to develop the learner’s knowledge and cognitive skills whilst facing the ‘handicap’ of the learner’s underdeveloped linguistic skills. The themes that the students cover (for example the structure and function of the human eye and the mammalian nervous system) are tackled with more depth and sophistication than can easily be communicated in simple language. Cummins (2000) presents an insightful treatise into the distinction between Basic Interpersonal Communication Skills (BICS) and Cognitive Academic Language Proficiency (CALP). A student’s ability in BICS need not signal equal competence in content-based tasks. The CLIL teacher thus has to dedicate much time and effort to reiterating and reinforcing content areas that are not easily understood. Coyle explains that “where the language level of learners is lower than their cognitive level, the learning environment must take into account this mismatch through ensuring that cognitive progression is maintained by accessing content through a lower linguistic level gradually working towards higher linguistic demands” (2005: 552). This approach has already been observed in practice in the present study: The teacher’s role as mediator is central to the CLIL classroom.

How the teachers meet these demands is a delicate operation, one that is continuously being explored in the teacher interviews that are carried out throughout this study. The dearth of textbooks for the CLIL classroom merely exacerbates the problems of tailoring material for the CLIL classroom. As it stands, “few publishers dare to publish specific CLIL materials as the market is still very small and not economically viable” (Vásquez, 2007: 103). The paucity of teaching materials places extra pressure on the teacher in terms of preparing each lesson. This is particularly the case with Biology. Whilst there are some textbooks available, these are not always age appropriate, or their focus or approach may be unsatisfactory or not in line with the course specifications. Using books imported from the UK is a useful stop-gap measure; however, using textbooks intended for L1 students in a different educational environment invariably poses further constraints in terms of meeting the CLIL students’ needs. The informants, TG and TB, are not native speakers of English (NNS), so despite their qualifications and experience, occasionally they feel inadequate in doing some tasks, such as readily offering synonyms and explaining some concepts without recourse to German. A further finding thus far is that further insight on task types that simultaneously foster cognitive and linguistic development is needed.

In assessing written work, developing an effectively streamlined assessment strategy is essential since the teacher must consider the language and content of any given student manuscript. A grey area lies in deciding when low attainment in a test is a result of poor mastery of the content rather than poor understanding of the language through which the content is delivered. As Leung and Mohan (in Kiely, 2009: para 2) state, “A teacher’s planning should provide opportunities for both learner and teacher to obtain and use information about progress towards learning goals.”  Each task is an opportunity for “learners to demonstrate their knowledge, understanding, and skills” (op.cit). TG has expounded on the importance of variety in task types in order to elicit meaningful information on how the student is faring. Preliminary findings from interviews with TG and TB show that CLIL poses unique problems for the teacher, and these are summarised in Figure 1.

Figure 1: The challenges of CLIL instruction—the teacher’s perspective

Access to training CLIL certification is only a starting point.
On-going teacher support often lacking.
A NNS instructor inevitably faces language hurdles which might prove disadvantageous for the students.
Access to resources Limited range of materials available for teaching
Lack of time Lesson planning is time-consuming because the source material must be modified.
A bilingual glossary of key terms must be prepared for each lesson.
The limited time allocated for lessons precludes in-depth exploration of topics.
Page 60
Page 61

Data collection in this section of the project is chiefly through teacher interviews and lesson observation. These methods, coupled with audio recording of lessons and analysis of the proceedings, ascertain whether and/or how the teachers actually circumvent the aforementioned challenges in their day-to-day work.

5. Student views

In addition to input from the teachers, the perspectives of the CLIL users themselves form an integral part of the assessment of CLIL. Since the study involves minors, parental informed consent has been granted, and anonymity has been assured. Naturalistic, unobtrusive observation has been deemed the most viable way of gathering detailed information without conspicuously interfering with the learners or causing them discomfort. In order to present a “thick description” replete with insightful detail in the tradition of Geertz, it is necessary to engage directly with the students from time to time.  This is primarily done in two ways: ascertaining attitudes, expectations, and experiences through structured and unstructured interviews and questionnaires; and examining the dynamics of the classroom.  This combination of etic and emic views reveals much insight into how these learners, when left to their own devices, adopt creative problem solving strategies as they appropriate English and use it in a personal and constructive way. The constructivist approach to education, propounded by Lev Vygotsky, takes into account the role of social interaction in cognitive development and in the construction of meaning.  To make sense of new information, the learner builds on existing knowledge (Vygotsky, 1978). In the CLIL classroom, the process is compounded since the student simultaneously learns new subject-related information alongside new linguistic features: lexis, syntax, and the semantic thread that binds these. Coyle, Verdú and Valcárcel stress the importance of content “which should be both logically and psychologically meaningful,” and they aver that the teacher’s role as mediator is critical in bolstering students’ construction of meaning (2002: 16). The social environment also contributes to the overall experience. Since this is a longitudinal study, it will be of interest to note how/if the already observed behaviours might change over the course of the academic year. The diagram below (Figure 2) presents a summary of the focus of the on-going student interviews that accompany the study.

Figure 2: Construction of meaning: avenues of investigation

Figure 2
Page 61
Page 62

6. Classroom conversation analysis

Having gathered information on the societal and school setting, pedagogical considerations, and the students’ attitudes and experiences, the next consideration is how all these factors interact in the classroom. This is achieved through periodical audio taping, transcription, and analysis of lessons in both Biology and Geography. Rogers explains that in Critical Discourse Analysis, the “analyst’s goal is to study the relationships between language form and function and explain why and how certain patterns are privileged over others” (2004: 4). Education researchers within the emerging CLIL tradition have employed CDA and Conversation analysis (CA) methodologies to parse the content of classroom discourse (for example, Dalton-Puffer, 2007; Dalton-Puffer and Smit, 2007). This not only elicits information on the type and complexity of discourse patterns, but on whether the specified curriculum outcomes are being met.

The routines of classroom interaction include:

  • The incidence of Initiation—Feedback—Response (IFR) cycles
  • Rhetorical modes
  • Non-verbal cues
  • Communicative competence
  • Emergence of CALP.

These aspects have been selected so that the routines of classroom interaction might be analysed in depth. As grounded theory practitioners assert, in charting new terrain, a judicious approach is favourable, thus allowing theory to emerge from the gathered data (Rhine, 2009). Researchers working in this tradition typically avoid approaching the study with a foregone conclusion in mind.

7. Assessing performance

The preceding studies consider the macro-features of CLIL its social context, its reception in the school, and how the teachers and learners surmount both expected and unforeseen challenges. The next essential consideration is the quality of the students’ performance in written work. Competence in written production in a targeted subject area and genre is an integral aspect of the learner’s overall competence. Hyland (2003: 18) explains that genre-based pedagogies offer students “explicit and systematic explanations of the ways language functions in social contexts. While knowledge acquisition is a central goal in the classroom, the development of the FL must nevertheless be monitored. Scholars distinguish between summative assessment—the measurement of a learner’s achievement—and formative assessment which serves to enlighten both the teacher and student on areas that need improving (Centre for Educational Research and Innovation, 2005: 13). In practice, the boundaries are often blurred.

At S, typical task types include:

  • Plenary and group discussions
  • Comprehension
  • Multiple choice questions
  • Cloze tests
  • Limited response
  • Guided writing
  • Describing, explaining, and interpreting data

The analyses are based on these questions:

  • Which task types generate information suitable for formative assessment?
  • Which task types generate information suitable for summative assessment?
  • Is there a relationship between a learner’s understanding of content and language performance?
  • What are the key features of the students’ texts?
  • Over the course of the academic year, are there any observable changes in the type and frequency of these features?
Page 62
Page 63

The selection of language features for evaluation is based on the ideals of academic writing as explained by linguist Hyland who lists high lexical density, high nominal style, and the use of impersonal constructions in a text where there is “a high proportion of content words in relation to grammar words such as prepositions, articles, and pronouns” (2006: 13-14). Snow (2010) concurs and includes conciseness and the use of “grammatical processes to compress complex ideas into a few words.” Noting the morphological and syntactic complexity of the texts complements this process.

The data are collected through intermittent assays, and then through triangulation, measures of lexical density (LD) and syntactic maturity are pitted against attainment.  Individual and composite (class) scores are noted. Today the error analysis methodologies pioneered by Corder in the 1960s have in large part been superseded by streamlined computer-based corpora, particularly in the investigation of lengthy texts. However, the task types in question here largely preclude this.  Figure 3 summarises the key areas of evaluation.

Figure 3: Assessing written production—elements selected for study

Figure 2

The longitudinal nature of the study facilitates comparison of performance throughout the testing period.

8. Conclusion

In conclusion, the project takes a comprehensive approach in order to present an evaluation which covers the social and pedagogical issues surrounding CLIL; the students’ perspectives and learning strategies, and an inventory of the features of their target language performance throughout the year. While an examination of attainment on class assignments is not overlooked, the focus of the study is primarily the students’ learning journey and acquisition of the English language. The vast scope inevitably calls for a mixed-methods approach where the various data sets complement each other. Few detailed ethnographic studies exist in the CLIL literature, thus making the present study timely. The work and contribution of these 32 students will hopefully be useful in showing whether CLIL indeed has “a major contribution to make to the Union’s language learning goals” (European Commission, 2003: 8).

Page 63
Page 64


Bakhtin, M.M.: 1986, Speech Genres and Other Late Essays (V. McGee, trans.).  University of Texas Press, Austin, TX.

Berliner Landesinstitut für Schule und Medien: 2006, Rahmenlehrplan für die Sekundarstufe I, Jahrgangsstufe 7-10. Hauptschule, Realschule, Gesamtschule, Gymnasium – Biologie. Senatsverwaltung für Bildung, Jugend und Sport.
, accessed: 3 December 2010)

Breidbach, S.: 2007, Plurilingualism, Democratic Citizenship in Europe and the Role of English, Language Policy Division DG IV – Directorate of School, Out-of-School and Higher Education. Council of Europe, Strasbourg.
(, accessed: 22 January 2010)

Centre for Educational Research and Innovation: 2005, Formative Assessment: Improving Learning in Secondary Classrooms. OECD Publishing, Paris.

Coyle, D.: 2007, Content and Language Integrated Learning: Towards a Connected Research Agenda for CLIL Pedagogies, The International Journal of Bilingual Education and Bilingualism 10(5), 543–562.

Coyle, Y, Verdú, M. and Valcárcel, M.: 2002, Teaching English to Children – Interactivity and Teaching Strategies in the Primary FL Classroom. Peter Lang, Frankfurt am Main.

Craith, M.N. (ed.): 2007, Language, Power and Identity Politics. Palgrave MacMillan, New York.

Cummins, J.: 2000, Language, Power and Pedagogy: Bilingual Children in the Crossfire. Multilingual Matters, Clevedon.

Dalton-Puffer, C.: 2007, Discourse in Content and Language Integrated Learning (CLIL) Classrooms. John Benjamins, Amsterdam.

Dalton-Puffer, C., and Smit, U. (eds.): 2007, Empirical Perspectives on CLIL Classroom Discourse. Peter Lang, Frankfurt am Main.

Deutsches Institut für Internationale Pädagogische Forschung: 2006, Unterricht und Kompetenzerwerb in Deutsch und Englisch. DESI, Frankfurt am Main.
(, accessed: 3 December 2010)

EUR Lex.: 1998, Council Resolution of 16 December 1997 on the early teaching of European Union languages. Official Journal of the European Communities. 98, C, 102.
( HTML, accessed: 20 January 2010)

European Commission: 2003, Language Teaching – Content and language integrated learning. European Commission Multilingualism.
accessed: 3 November 2009)

Firth, A. and Wagner, J.: 1997, On Discourse, Communication, and (Some) Fundamental Concepts in Second Language Acquisition Research Modern Language Journal 81(3), 285-300.

Geetrz, C.  1973, Thick Description: Toward an Interpretive Theory of Culture. In The Interpretation of Cultures: Selected Essays. Basic Books, New York.

German International School.
accessed: 3 December 2010)

Granger, S., Hung, J. and Petch-Tyson, S (eds.): 2002, Computer Learner Corpora, Second Language Acquisition and Foreign Language teaching. John Benjamins, Amsterdam.

Hyland, K.: 2006, English for Academic Purposes. Routledge, London.

Hyland, K: 2003, Genre based pedagogies: A social response to process, Journal of Second Language Writing 12(1), 17-29.

Kiely, R.: 2009, CLIL – The question of assessment.
accessed: 3 December 2010)

Leung, C. and Mohan, B.: 2004, Teacher formative assessment and talk in classroom contexts: assessment as discourse and assessment of discourse, Language Testing 21(3), 360-389.

Lohmer, B. and Eckhardt, T.: 2010, The Education System in the Federal Republic of Germany 2008. KMK, Bonn.
(, accessed: 3 December 2010)

Page 64
Page 65

Lyster, R.: 2007, Learning and teaching languages through content. – A Counterbalanced Approach. John Benjamins, Amsterdam.

Marsh, D.: 2001, Using Languages to Learn and Learning to Use Language. CLIL Compendium.  (, accessed: 12 November 2009)

Marsh, D., Maljers, A. and Hartiala, A.K. (eds.): 2001, Profiling European CLIL Classrooms: Languages Open Doors. University of Jyväskylä, Jyväskylä.

Phillipson, R.: 2003, English-only Europe? Challenging Language Policy. Routledge, London & New York.

Rhine, J.: 2009, What is Grounded Theory? Grounded Theory Institute.
( , accessed: 2 December 2010)

Rogers, R. (ed.): 2004, An Introduction to Critical Discourse Analysis in Education. Lawrence Erlbaum Associates, Mahwah, NJ.

Schwabe, T.: 2010, Personal Interview.

Snow, C.E.: 2010, Academic Language and the Challenge of Reading for Learning About Science, Science 328/5977, 450-452.

Várkuti, A.: 2010, Linguistic Benefits of the CLIL Approach:  Measuring Linguistic Competences, International CLIL Research Journal 1(3), 67-79.

Vázquez, G.: 2007, Models of CLIL: An Evaluation of its Status Drawing on the German Experience.  A Critical Report on the Limits of reality and Perspectives, Volumen Monográfico, 95-111.

Vygotsky, L.S.: 1978, Mind in society: The development of higher psychological processes. Harvard University Press, Cambridge Mass.

Wannagat, U.: 2009, Bilingualer Geschichtsunterricht im internationalen Fokus. Peter Lang, Frankfurt am Main.

Zydatiß, W.: 2005, Chronological sequencing.  Eine Methode zum intergrierten Sach-Sprach-Lernen. Content and Language Integrated Learning,  Der Fremdsprachliche Unterricht  Englisch 78, 50-53.

Zydatiß, W.: 2007, Deutsch-Englische Züge in Berlin (DEZIBEL). Eine Evaluation des bilingualen Sachfachunterrichts an Gymnasien: Kontext, Kompetenzen,Konsequenzen. Peter Lang, Frankfurt am Main.
accessed: 21 December 2009)