International Journal of Bilingual Education and

Bilingualism

ISSN: 1367-0050 (Print) 1747-7522 (Online) Journal homepage: http://www.tandfonline.com/loi/rbeb20

Content and language integrated learning in the

Netherlands: teachers’ self-reported pedagogical
practices

Evelyn van Kampen, Wilfried Admiraal & Amanda Berry

To cite this article: Evelyn van Kampen, Wilfried Admiraal & Amanda Berry (2018) Content
and language integrated learning in the Netherlands: teachers’ self-reported pedagogical
practices, International Journal of Bilingual Education and Bilingualism, 21:2, 222-236, DOI:
10.1080/13670050.2016.1154004

To link to this article: https://doi.org/10.1080/13670050.2016.1154004

© 2016 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

Published online: 07 Mar 2016.

Submit your article to this journal

Article views: 2820

View Crossmark data

Citing articles: 3 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rbeb20
INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM, 2018
VOL. 21, NO. 2, 222–236
https://doi.org/10.1080/13670050.2016.1154004

Content and language integrated learning in the Netherlands:

teachers’ self-reported pedagogical practices
Evelyn van Kampen, Wilfried Admiraal and Amanda Berry*
Leiden University Graduate School of Teaching, Leiden University – ICLON, Leiden, Netherlands

ABSTRACT ARTICLE HISTORY

In recent years, a surging uptake of content and language integrated Received 15 July 2015
learning (CLIL) has permeated the European context. This article Accepted 8 February 2016
presents the outcomes of a study about the self-reported pedagogical
KEYWORDS
practices of CLIL teachers in the Netherlands. To investigate these CLIL; bilingual; classroom
teachers’ pedagogies, a questionnaire was designed, validated and, pedagogies; practicing
subsequently completed by 297 teachers (218 CLIL teachers and 79 teachers; the Netherlands
regular teachers) from secondary schools across the Netherlands. This
allowed a detailed investigation of the nature and range of pedagogies
of CLIL teachers as well as differences between CLIL teachers of different
subject disciplines. The results of the CLIL teachers were compared to a
sample of regular teachers. Three main findings emerged from the
study. First, CLIL teachers report using more pedagogical approaches
focusing on providing students with diverse input and scaffolding than
approaches focusing on developing students’ language and their
knowledge of disciplinary literacies. Second, the subject discipline of a
CLIL teacher seems to influence the pedagogical approaches they report
using. Third, there is a notable difference between the kinds of
pedagogies reported by CLIL teachers and their regular counterparts.

1. Introduction
Content and language integrated learning (CLIL) is commonly described as a ‘dual focused edu-
cational approach in which an additional language is used for the learning and teaching of both
content and language’ (Coyle, Hood, and Marsh 2010, 1). CLIL is often referred to as an ‘umbrella’
term, that is difficult to define at both theoretical and practical levels because it includes many var-
iants depending on the specific context in which it is implemented (Cenoz, Genesee, and Gorter
2014). The main aim of CLIL as an educational approach is to improve students’ foreign language
competence whilst simultaneously facilitating their learning of subject matter.
The use of CLIL has spread greatly in the past two decades, especially in Europe, and is most often
used in schools offering a form of bilingual education (Pérez-Cañado 2012). These schools offer stu-
dents the opportunity to study some subjects through a language other than their first language. This
means, in practice, that subject teachers take on an additional role, teaching their subject in the
context of a foreign language. Research into the specific pedagogical practices employed by CLIL tea-
chers has been limited to-date and, as such, has been identified as a key area in need of further inves-
tigation (Pérez-Cañado 2012; Dalton-Puffer and Smit 2013).

CONTACT Evelyn van Kampen [email protected]

*
Present address: School of Education, RMIT, Victoria, Australia.
© 2016 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://
creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the
original work is properly cited, and is not altered, transformed, or built upon in any way.
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 223

Considering that a key component of quality CLIL education lies in teachers’ classroom pedago-
gies, this study seeks to contribute to broadening the knowledge base about the nature and
range of practices employed by CLIL teachers. It does so by investigating the self-reported pedago-
gies of CLIL teachers in the Netherlands.

2. Context of study
Dutch bilingual schools provide a rich context for CLIL research as the Netherlands has offered this
form of education for over 25 years, with currently over 120 schools involved (see www.
europeesplatform.nl/tto). A key characteristic of the Dutch CLIL context is its highly institutionalized
approach. One organization, the EP-Nuffic, coordinates all activities related to CLIL education in the
Netherlands. All members of the Network of Dutch Bilingual Schools have endorsed a Standard for
Bilingual Education (De Graaff and van Wilgenburg 2015), which defines the basic requirements for
schools with a bilingual stream. The Standard includes a Competency profile for teachers, outlining
the competencies expected from teachers in bilingual education that includes knowledge about the-
ories underpinning CLIL, an ability to identify essential elements of a CLIL approach in one’s own
subject domain, and the use of CLIL-related pedagogical approaches that, for example, support
language production and the development of language learning strategies.
In the Netherlands, students are streamed into one of three main secondary school types: VMBO,
HAVO and VWO, following the completion of primary school. VMBO is the four-year pre-vocational
stream, HAVO is the five-year intermediate stream and VWO is the six-year pre-university stream.
Initially, CLIL education was only offered at VWO level. However, Dutch CLIL education is slowly
expanding to include different school types (HAVO and VMBO) and levels (primary education). In
Dutch bilingual secondary schools, there is a greater focus on CLIL teaching in the lower forms com-
pared with the upper forms because all students need to complete the Dutch-language national
exams (with the exception of foreign language subjects) in their final year of school. Bilingual
HAVO and VWO schools are expected to offer at least 50% of lessons through the foreign language.
Most CLIL teachers are not native speakers of the foreign language; neither do they have a back-
ground in language pedagogy (De Graaff et al. 2007).

3. Theoretical framework
3.1. Effective CLIL pedagogy
While CLIL is defined very broadly both in research and practice, it can be distinguished from other
forms of bilingual education, immersion education and content-based instruction on a number of
points: CLIL is about using a foreign language or lingua franca that is generally not regularly used
outside of the classroom; students are usually exposed to CLIL teaching only once they have
already acquired literacy skills in their mother tongue; CLIL teachers are generally subject specialists
and non-native speakers of the target language; and CLIL lessons are usually timetabled as regular
lessons within the school curriculum (Dalton-Puffer 2011, 183–184).
The broad definition of CLIL and its use in varied contexts also means that it is not associated with
one specific form of pedagogy. Rather, CLIL functions as ‘a type of instruction that fuses the best of
subject matter and language teaching pedagogies’ (Morton 2010, 97). Coyle (1999) proposed a holis-
tic conceptual framework for CLIL, the so-called 4C’s model, which combines different aspects associ-
ated with CLIL teaching: Content (subject matter), Communication (language to learn and use),
Cognition (learning and thinking) and Culture (social awareness of self and others). Coyle’s framework
was developed to support beginning CLIL teachers in organizing lessons and planning units of work.
Another issue associated with the broad definition of CLIL is the difficulty to distinguish it from
‘good teaching’ generally. CLIL pedagogy is typically associated with innovative, student-centered
pedagogical approaches. Focusing on student-centeredness is considered necessary for effective
224 E. VAN KAMPEN ET AL.

subject and foreign language learning to occur. However, arguably, this focus is not unique to CLIL
teaching. Rather, what makes CLIL teaching distinguishable from ‘good teaching’ is its focus on inte-
grating subject teaching with language teaching. CLIL teachers are expected to carry out an extra role
in that they need to ensure that the content they teach is understood by students through the
medium of a foreign language, that is, they are language teachers at the same time they are teachers
of specific subject matter (Coyle, Hood, and Marsh 2010).
Theoretical models of CLIL pedagogy are generally based on approaches to effective foreign
language learning. These approaches range from simple to more complex. The most simple, the
so-called ‘language bath’ approaches (Dalton-Puffer 2007, 3), reflect the view that, simply by immer-
sing students in the target language, they will learn it. In contrast, more elaborate approaches draw
on a much wider theoretical base. For example, Swain’s (1995) output hypothesis, which claims that
learners will only learn a language once they are forced to produce their own written or spoken
output in that language, and Lyster’s (2007) focus on attention to form and corrective feedback
for the development of learners’ accuracy in the foreign language.
Westhoff’s (2004) penta-pie model illustrates an even more elaborate theoretical perspective. This
model emphasizes that effective foreign language pedagogy should facilitate five aspects: (1)
exposure to meaningful input; (2) meaning-focused processing; (3) form-focused processing; (4)
opportunities for output production and (5) awareness and use of language learning strategies. De
Graaff et al. (2007) transformed the penta-pie into a CLIL observation tool for effective second
language pedagogy, and argued that in CLIL lessons a mix of all five aspects should be present.
Recently, there has been an increased emphasis on the teaching of disciplinary literacies in CLIL
contexts, including the discourse and genres associated with particular subject disciplines (e.g.
Morton 2010; Meyer et al. 2015). For example, in learning about science, students need to understand
and produce different kinds of texts such as laboratory reports, or descriptions of phenomena. It is
important that either in their Science lessons or English lessons students are introduced explicitly
to the purposes and key linguistic features of these text types as, especially in a foreign language
learning context, knowledge about these forms of language cannot be assumed (Morton 2010).
Through a focus on disciplinary literacies, more integrated subject-specific content and language
teaching and learning is expected to occur (e.g. Meyer et al. 2015).

3.2. CLIL teachers’ classroom pedagogies

As emphasized in the introduction, teachers’ classroom pedagogies have been identified as a key
area for further CLIL research. Relatively, little is known about the kinds of pedagogical designs
and approaches that characterize CLIL lessons and to what extent these differ from regular subject
teaching and foreign language teaching (Dalton-Puffer and Smit 2013). What follows is a synthesis
of research that focuses on CLIL pedagogies in secondary education.
Several researchers have investigated the nature of CLIL pedagogies as innovative and student-
centered. Coonan (2007) conducted research over three years with a group of language and
subject teachers enrolled in a two-year post-graduate CLIL training program in Italy, using teacher
logs, questionnaires, focus groups and interviews. Her study revealed that the teachers considered
CLIL as an ideal environment for pedagogical innovation and that it fosters greater teacher awareness
of student learning needs. In comparative lesson observations, in which CLIL and regular content
lessons in three subjects of six Swiss classes were followed over a period of two years, Badertscher
and Bieri (2009) found no difference in overall lesson design. Nikula (2010) studied to what extent
teaching through a foreign language affects a teacher’s interactional and instructional style by observ-
ing one Finnish biology teacher’s CLIL and regular lessons. In CLIL lessons, the teacher used less varied
and subtle language as well as shorter monologues compared with regular lessons. Also, in these CLIL
lessons, students often adopted a more active role asking questions and making comments. Pérez
Cañado (2014) investigated the perceived training needs of in-service CLIL teachers from various
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 225

countries, using a survey. A main finding was that teachers considered their current knowledge to be
the lowest, and training needs the highest, regarding the theoretical underpinnings of CLIL.
In the Dutch CLIL context, Koopman, Skeet, and de Graaff (2014) reported that the few studies
investigating pedagogy have tended to do so from a ‘process-product’ perspective, with teacher
behavior (process) used to predict learner outcomes (product) (e.g. Admiraal, Westhoff, and de Bot
2006). De Graaff et al. (2007) used the observation tool they constructed for effective second
language pedagogy in CLIL (see Section 3.1) to analyze a set of nine CLIL lessons from a variety of
school subjects in three Dutch schools. They found that, over all the lessons they observed, teachers
used almost the whole range of indicators of effective CLIL pedagogy. Little evidence was found for
facilitating noticing of and explaining problematic and relevant language forms. Koopman, Skeet,
and de Graaff (2014) conducted a small-scale study with six CLIL subject teachers from three different
Dutch bilingual schools to investigate these teachers’ knowledge of language pedagogy. They con-
cluded that the teachers generally had a complex view of what language teaching involved, and that
specific areas for further professional development should include: teaching language ‘chunks’ (i.e.
specific sentence constructions and lexical collocations), focus on language form and understanding
students’ developmental processes in relation to second language acquisition.
There are few empirical studies investigating teaching of disciplinary literacies in CLIL classes. This
is not surprising given it is a relatively new focus of research and models of how to do so have only
recently begun to emerge (e.g. Meyer et al. 2015). Morton’s (2010) research is one of the few empirical
studies explicitly focusing on the application of genre-specific teaching by CLIL teachers. Using audio,
video and interview data from three history lessons in two bilingual schools in Spain, Morton
observed that in none of these classrooms was there any explicit focus on the linguistic features
of history genres, either as texts to be understood or deconstructed, or as texts to be produced by
the students. Interestingly, the textbooks used by the classes in Morton’s study required students
to use a variety of genres in written exercises without offering any scaffolding. Some assignments
required only very limited written production by students, an aspect that other studies about CLIL
practices in Europe (e.g. Dalton-Puffer 2007) have also highlighted.

4. Research questions
The motivation for this study emerges from the above described situation that we do not yet know
much about secondary CLIL teachers’ pedagogies, and that what little research exists is rather dispa-
rate. If we want to support more effective teaching and learning in CLIL, then we need to know more
about the characteristics of current teaching practice. To this end, it is relevant to produce a compre-
hensive account of what teachers say they do in their teaching. Hence, the following research ques-
tions guided the current study investigating the self-reported pedagogical practices of CLIL teachers
in the Netherlands:

(1) What are the main characteristics of self-reported pedagogical practices of CLIL teachers and how
do they differ from those of regular teachers?
(2) To what extent are differences in self-reported pedagogies of CLIL teachers related to the subject
disciplines they teach and how does this differ from regular teachers?

5. Methods
5.1. Participants
The study was conducted from November until December 2014. Teachers from all Dutch secondary
schools offering CLIL education at VWO level (N = 111) were invited to participate. In addition, partici-
pants were also recruited from a smaller, regional sample of 40 regular VWO secondary schools that
do not offer CLIL education. Some of the VWO schools also offered education at HAVO and VMBO
226 E. VAN KAMPEN ET AL.

level. Approximately 1200 teachers (based on an average of 8 teachers per school) were indirectly
invited to participate, mainly via school coordinators.
Table 1 provides background information about all of the participating teachers. Over half of the
teachers have more than 10 years teaching experience. The range of subjects taught by the partici-
pating CLIL teachers is representative of the Dutch CLIL context in that, for example, fewer creative
(arts based) and physical subjects are offered through English. Note that we have included English as
a separate subject because English teachers are considered to have a special role in Dutch CLIL
schools in that their main emphasis is language acquisition and use, whereas CLIL teachers in
other subject areas have a main emphasis on subject-specific content knowledge. Also, English tea-
chers in CLIL schools are expected to possess specific competences that allow them to support and
complement the teachers of other school subjects in developing students’ knowledge of subject-
specific language. Hence, including English-language teachers in the analyses, but keeping them sep-
arate from the teachers of the other school subjects allows us to investigate the differences. We also
invited teachers of Other Foreign Languages to take part in the survey to allow us to investigate poss-
ible differences between this group and other groups. However, too few CLIL teachers of Other
Foreign Languages participated to allow a meaningful comparison.
Appendix 1 provides additional background information about the CLIL teachers. The number of
years of CLIL teaching experience of the participating CLIL teachers varies greatly (from 0 to 11 or
more) as does the number of hours per week that the teachers report teaching CLIL classes (from
1 to 11 or more). In terms of their entry into CLIL teaching, the majority indicated that they were
asked to teach in the CLIL section by school management. Few of the CLIL teachers followed a
course about CLIL teaching as part of their higher education (either pre- or in-service). Most of the
teachers are not native speakers of English and do not have a bi- or multilingual background. At
the same time, they rated their level of English as high, although speaking, reading and listening
ability were rated consistently much higher than writing ability.

5.2. Instrument
5.2.1. The CLIL questionnaire
To investigate the self-reported pedagogical practices of CLIL teachers, an English-language online
questionnaire – the Content and Language Integrated Learning Questionnaire (CLIL-Q) – was designed

Table 1. Background variables considered for CLIL and regular teachers.

Variable All teachers (%) (N = 296) Regular teachers (%) (N = 78) CLIL teachers (%) (N = 218)
Subject discipline
English 27.1 20 29.8
Other foreign languages 8.4 26.7 1.5
Mathematics and sciences 28.6 21.3 31.3
Social sciences 26.7 28 26.3
Creative and physical 9.2 4 11.1
Gender
Female 46.9 54.7 44.1
Male 53.1 45.3 55.9
Age
<25 1.4 1.3 1.4
25–29 14.1 10.5 15.4
30–39 27.6 18.4 30.8
40–49 25.9 26.3 25.7
50–59 24.8 35.5 21
60 or over 6.2 7.9 5.6
Years teaching experience
0–2 8.2 3.9 9.7
3–5 12.9 13 12.9
6–10 19 14.3 20.7
>10 59.9 68.8 56.7
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 227

as, to our knowledge, there were no existing suitable questionnaires. The content of the CLIL-Q was
developed on the basis of existing research about effective CLIL pedagogies (summarized in Section
3). We strived to translate the insights from previous research into items that were as specific as poss-
ible, providing examples to make more abstract approaches easier to interpret.
The CLIL-Q contained two parts and a combination of open and closed questions. Part A asked
participants for background information (see Table 1 and Appendix 1). Part B asked participants to
indicate, on a 5-point Likert scale, for 42 items how often they use particular pedagogical
approaches in their lessons. CLIL teachers were asked to focus on their lessons in bilingual edu-
cation only. For each item, a space was provided for teachers to comment on their ratings,
helping us to identify whether they, for example, do not apply a pedagogical approach often
because they do not see the need for it or because they do not have the resources and/or knowl-
edge to do so. Additionally, teachers in both CLIL and regular education were asked to answer the
following open question: ‘Are there differences in the teaching approaches you use to teach these
classes? Please elaborate.’ Coordinators from the invited schools received a reminder email two
weeks after their initial invitation to forward to teachers who had not yet completed the
questionnaire.
The CLIL-Q was piloted twice by groups containing both CLIL and regular teachers and was sub-
sequently adapted based on the feedback received. The main change made to the questionnaire
after piloting was that initially the CLIL-Q asked teachers to provide two ratings for each item –
the importance for their teaching and how satisfied they were with how they applied a particular
practice in their teaching. We changed this to just one rating per item because the pilot group indi-
cated that the questionnaire was too lengthy and they found it difficult to distinguish between the
aspects of importance and satisfaction. The second pilot group provided us with ideas about how to
improve the wording of some items.

5.2.2. Four factors identified

Principle component factor analysis with oblique rotation was conducted on the Likert-scale items of
Part B of the CLIL-Q to determine the underlying factors. Previously, this led to a factor solution with
six factors with an eigen value higher than 1. A decision was made to settle for a four-factor solution
because the reliability of a five- or six-factor solution was not high enough. The four-factor solution
was reached by removing items that did not load onto one of the four factors and by removing items
that did not fit with the main ideas that each of the four factors was identified to stand for. The sol-
ution reached contains 28 items and explains 50.1% of the variance.
Appendix 2 shows the rotated factor loadings for each of the four factors. Table 2 reports descrip-
tive statistics for each of the identified factors, including the reliability scores (Cronbach’s α). The four
factors identified relate to various clusters of pedagogies identified in the literature review (Section 3).
Factor 1, ‘Literacies’, consists of items related to the idea that in CLIL teaching, it is important that
students become knowledgeable about the specific language used in particular subject disciplines.
Factor 2, ‘Language’, consists of items related to the idea that it is important that CLIL teaching
focuses on developing students’ written and spoken ability in the foreign language through which
the subject is taught. Factor 3, ‘Scaffolding’, consists of items related to the idea that it is important
that CLIL teachers provide scaffolds to enable learners to better understand both the content and the
language to which they are exposed. Factor 4, ‘Input’, consists of items related to the idea that for CLIL

Table 2. Descriptive statistics and reliability scores per factor.

Factor No. of items M SD Cronbach’s α
Literacies 5 3.04 0.97 .83
Language 9 3.07 0.87 .81
Scaffolding 7 3.58 0.67 .71
Input 7 3.74 0.74 .81
228 E. VAN KAMPEN ET AL.

teaching to be effective it is important that teachers provide learners with both creative and diverse
input that is tailored to their learners’ content and language ability.

5.3. Data analysis

Descriptive statistics, qualitative content analysis and multivariate analysis of covariance (MANCOVA)
were used to answer the first research question. MANCOVA was used to seek an answer to the second
research question. The MANCOVA to answer the first research question included whether the teacher
was a CLIL teacher or a regular teacher as the independent variable, the four types of pedagogical
approaches as dependent variables and the background variables listed in Table 1 and Appendix
1 as covariates. The MANCOVA to answer the second research question included subject discipline
as the independent variable, the four types of pedagogical approaches as dependent variables
and the background variables listed in Table 1 (for all teachers) and Appendix 1 (for CLIL teachers)
as covariates. We conducted two separate MANCOVAs to investigate the relationship between the
self-reported pedagogies of CLIL teachers and their subject discipline, and of regular teachers and
their subject discipline as both groups differed in the number of teachers of particular school
subjects.

6. Results
6.1. Characteristics of the self-reported pedagogical practices of CLIL teachers compared
to regular teachers
The left column of Table 3 reports descriptive statistics for each of the four pedagogical approaches
for CLIL teachers. The mean scores for each approach reveal a clear division in that the Scaffolding and
Input approaches have higher scores than do Literacies and Language. Paired sample t-tests were con-
ducted between each combination of approaches to compare within the group of CLIL teachers for
differences between the four approaches. These tests revealed that all mean scores differ significantly
(with α = 0.05).
The right column of Table 3 reports descriptive statistics for each of the four pedagogical
approaches for regular teachers. When comparing for differences in the self-reported pedagogical
practices of CLIL teachers and regular teachers, our analysis showed a significant multivariate main
effect of type of teacher (CLIL or regular; Wilks’ λ = .82, F(4, 236) = 13.24, p < .001, η 2 = .18). Significant
univariate effects were obtained for: Language (F(1, 239) = 32.33; p < .001; η 2 = .12), Scaffolding (F(1,
239) = 3.92; p = .049; η 2 = .02); and, Input (F(1, 239) = 5.67; p = .018; η 2 = .02). For these three pedago-
gical approaches, the CLIL teachers scored higher on average than did the regular teachers. However,
for Literacies, the difference was not significant.
The results of the open question, which asked teachers who teach both CLIL and regular classes to
elaborate on the extent to which the approaches they use in these two contexts differ, provide
further evidence of how CLIL teachers characterize their pedagogical practices in CLIL lessons and
how unique they consider these practices to be. On the basis of the results, the following picture
emerges. Out of 107 teachers who responded to this question, 77 reported a great deal of difference
between the approaches they use in CLIL compared to their regular lessons. However, 11 did not give

Table 3. Descriptive statistics per pedagogical approach for CLIL and regular teachers.
CLIL teachers (N = 218) Regular teachers (N = 78)
M (SD) M (SD)
Literacies 2.97 (0.99) 3.23 (0.89)
Language 3.18 (0.77) 2.76 (1.05)
Scaffolding 3.65 (0.63) 3.40 (0.72)
Input 3.80 (0.72) 3.60 (0.77)
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 229

a reason, or gave a reason unclear to us. Of the 66 who gave a reason, these were spread across the
four categories of pedagogical approaches, with two main categories of differences being Language
and Scaffolding. Within these two categories, there was considerable variation in teachers’ responses.
Of the 30 teachers who identified no or hardly any difference in the pedagogies they use in CLIL and
regular classes, 15 provided a clear reason. In the following section, we synthesize and discuss all of
the teachers’ responses. It should be noted that the percentages of teachers who made specific com-
ments are generally small because their responses were spread across the different categories and
also within the categories there was a lot of variation in how teachers reported their CLIL teaching
differed.
The most frequently reported difference by the 77 teachers who mentioned there was a difference
between teaching CLIL classes and regular classes relates to Language (26%). They identified more
focus on communication in their CLIL classes, that is, encouraging students to produce spoken
output, such as through interactions between students in the foreign language. Mostly, this was
reported in terms of more opportunities for oral communication. For example, one teacher explained:
I teach biology and there are no differences in the teaching approaches when we do practical work, while in
theory lessons there are. In regular lessons I give a short lecture. In the CLIL lessons I try to find forms where
the students need to speak to each other.

Only 8% (of the 26%) of teachers mentioned explicitly that more ‘communication’ includes more
focus also on writing in their CLIL classes. Of the 26% of teachers, 9% added that they focus more
on providing specific language instruction in the lower forms than the upper forms. In terms of feed-
back on communication, interestingly, only 6% of teachers mentioned increased emphasis on cor-
recting students when they make language mistakes in CLIL classes, compared with their regular
classes. Exclusive use of English in their CLIL classes was mentioned by 9% of teachers. Of those tea-
chers whose approaches to teaching their regular and CLIL classes do not or hardly differ, 33%
emphasized that language is equally important in their regular lessons. As one teacher put it: ‘This
is important for ALL students … because a major part of learning is about acquiring language.’
With regard to Input, of the 77 teachers who mentioned there was a difference, 12% indicated that
their teaching is much more ‘method driven’ (i.e. textbook-based) in regular classes, compared with
their CLIL classes in which they select their own materials and design more assignments themselves.
One teacher explained: ‘Regular lessons come from the method and are much more structured
according to this method. Teaching CLIL classes is much more intuitive than regular lessons.’
Especially English teachers emphasized that in their regular lessons they use more Dutch language
because the textbooks for regular classes make use of Dutch. One teacher explained: ‘I feel I am
limited in my teaching strategies. I use less CLIL assignments, because I need to prepare regular stu-
dents for Dutch-English translation elements (method and department choices).’ Focusing on devel-
oping the so-called European and International Orientation of students, which is a major goal of
Dutch bilingual education, was mentioned by 11% of teachers as an important element of their
CLIL lessons. As one teacher stated: ‘The link between the content of the lesson and current affairs
is much stronger in CLIL classes, where we select our own materials.’
Regarding Scaffolding, 29% mentioned that they provide more scaffolding to students in their CLIL
classes, compared with their regular classes. Of these, 14% of the teacher responses focused on
making glossaries or personal idiom files with students to facilitate their learning of relevant vocabu-
lary. The remaining responses described task-related scaffolds, such as providing more tools to
analyze a text or produce a summary, and teaching-related scaffolds, such as using gestures and
visual aids and explaining problems a number of different times using different approaches.
Comparatively few teachers (8%) mentioned approaches to Literacies. Responses were varied and
included getting students to notice the specificities of subject-related text types and how to structure
written texts for their subject discipline. Interestingly, 20% of those teachers who indicated there is no
difference between their CLIL and regular classes stated that this is because of the nature of the
subject they teach, for example: ‘Drama is a very CLIL subject’ and ‘Mathematics is for both the same’.
230 E. VAN KAMPEN ET AL.

Teachers also gave responses that were more related to the Dutch CLIL context, than to pedago-
gical approaches. Of those teachers that mentioned that there was a difference, 12% of teachers indi-
cated that they design their CLIL lessons to focus more on higher level cognitive skills. One teacher
explained this as: ‘For CLIL students I try to be on the top of Bloom’s taxonomy. For regular students I
tend to stay in the middle.’ Another teacher explained:
In my CLIL classes I encourage the students more to think critically and see things from different perspectives. This
has to do with the books that are used, but also with the fact that CLIL students need more challenges to keep
them motivated and working.

A further interesting point to note is that, of those teachers who stated there was no or hardly any
difference, 40% emphasized that over time they have noticed that they increasingly use the pedago-
gical approaches from CLIL classes, such as more focus on student interaction, in regular classes as
well and find that they work well.

6.2. Relationship between the self-reported pedagogies of CLIL teachers and their subject
disciplines compared to regular teachers
Table 4 reports descriptive statistics for CLIL teachers and regular teachers per subject discipline. We
did not include the three CLIL teachers of ‘Other Languages’ nor the three regular teachers of Creative
and Physical subjects because these numbers are too low for a meaningful comparison with teachers
of other subject disciplines.
The MANCOVA to test the relationship between CLIL teachers and their subject discipline revealed
a significant multivariate main effect for subject discipline, (Wilks’ λ = .38, F(12, 365.41) = 13.66,
p < .001, η 2 = .28). Given the significance of the main test, the univariate effects were examined. Sig-
nificant univariate effects were obtained for: Literacies (F(3, 141) = 16.70, p < .001, η 2 = .26); Language
(F(3, 141) = 20.44, p < .001, η 2 = .30); Scaffolding (F(3, 141) = 6.04, p = .001, η 2 = .11) and Input (F(3, 141)
= 7.57, p < .001, η 2 = .14).
As the variable subject discipline contains four subject disciplines for CLIL teachers (English, Math-
ematics and Sciences, Social Sciences and Creative and Physical subjects), a post hoc multiple com-
parison was run to interpret how they differed. The post hoc test by Scheffe revealed that for the
Literacies approach, English teachers scored significantly higher than teachers from all other disci-
plines (Scheffe, p < .05). Teachers from the Creative and Physical subjects scored significantly lower
than teachers from all other disciplines. For the Language approach, the post hoc test revealed
that English teachers scored significantly higher than teachers from all other disciplines (Scheffe,
p < .05). For the Scaffolding approach, the post hoc test showed that teachers from the Creative
and Physical subjects scored significantly lower than both Social Science teachers and Mathematics
and Science teachers (Scheffe, p < .05). Interestingly, for this approach English teachers did not score
highest. Finally, the post hoc test revealed that for the Input approach, English teachers and Social

Table 4. Descriptive statistics per subject discipline for CLIL teachers and regular teachers.
English Mathematics and sciences Social sciences Creative and physical Other foreign languages
M (SD) M (SD) M (SD) M (SD) M (SD)
CLIL teachers N = 59 N = 62 N = 52 N = 22
Literacies 3.69 (0.8) 2.72 (0.81) 2.89 (0.8) 1.75 (0.82) –
Language 3.94 (0.54) 2.77 (0.56) 2.83 (0.57) 2.94 (0.7) –
Scaffolding 3.59 (0.64) 3.82 (0.59) 3.74 (0.5) 3.17 (0.7) –
Input 4.15 (0.6) 3.42 (0.71) 4.02 (0.51) 3.29 (0.77) –
Regular teachers N = 15 N = 15 N = 21 – N = 20
Literacies 3.40 (0.77) 2.73 (0.68) 3.24 (0.88) – 3.70 (0.81)
Language 3.50 (0.87) 2.19 (0.58) 2.13 (0.92) – 3.47 (0.85)
Scaffolding 3.29 (0.76) 3.14 (0.67) 3.53 (0.65) – 3.54 (0.73)
Input 3.51 (0.75) 3.28 (0.83) 3.89 (0.76) 3.65 (0.65)
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 231

Science teachers scored significantly higher than teachers of Mathematics and Sciences and of Crea-
tive and Physical subjects (Scheffe, p < .05).
The MANCOVA to test the relationship between regular teachers and their subject disciplines also
revealed a significant multivariate main effect for subject discipline (Wilks’ λ = .44, F(16, 150.335) = 2.9,
p < .001, η 2 = .19). Given the significance of the main test, the univariate effects were examined. In
contrast to the CLIL teachers, significant univariate effects were only obtained for: Language (F(4,
52) = 7.8, p < .001, η 2 = .38). As the variable subject discipline contains four subject disciplines
(English, Other Foreign Languages, Mathematics and Sciences and Social Sciences), a post hoc mul-
tiple comparison was also run to interpret how they differed. The post hoc test by Scheffe revealed
that for the Language approach English teachers and Other Foreign Language teachers scored sig-
nificantly higher than teachers from the other two disciplines (Scheffe, p < .05).

7. Discussion
This study focused on investigating the self-reported pedagogical practices of CLIL teachers in the
Netherlands. Our first research question focused on determining the main characteristics of CLIL tea-
chers’ practices and how these differ from regular teachers. Our findings indicate significant gaps
between notions of effective CLIL pedagogy in theory and the pedagogical approaches reported
by these CLIL teachers. On the whole, the participating CLIL teachers reported lower scores on the
Literacies and Language approaches than they did on Scaffolding and Input. The CLIL teachers
reported least use of the Literacies approach, that underscores concerns raised by recent studies
(e.g. Morton 2010; Meyer et al. 2015) about the disciplinary literacies teaching competence of CLIL
teachers. The Literacies approach was also hardly mentioned in the teacher responses to the open
question, which confirms the quantitative results and indicates that, in general, promoting disciplin-
ary literacy of students is not something that is at the forefront of teachers’ minds when they consider
their pedagogical practices.
An important finding emerging from this study is that CLIL teachers report generally using fewer
approaches linked to the Language approach compared with Scaffolding and Input. This sits in con-
trast with the focus on students’ language development that is considered part of what makes CLIL
unique (Coyle, Hood, and Marsh 2010), and most theoretical models about effective CLIL teaching are
heavily focused on what is necessary for effective language teaching to occur. Interestingly, in the
open question, teachers from both CLIL and regular classes most frequently distinguished their
CLIL teaching from their regular teaching by mentioning the Language approach. Whilst most tea-
chers mentioned stimulating spoken output, few teachers mentioned a focus on language forms,
for example by correcting students’ language mistakes. This is consistent with the findings of De
Graaff et al. (2007) and Koopman, Skeet, and de Graaff (2014) that teachers’ focus on language
form is limited.
This raises an avenue for further research into this aspect of CLIL teachers’ pedagogy. For example,
it would be useful to describe in more detail what kind of feedback from subject teachers supports
students’ language development and how learning about feedback can be effectively supported
through pre- and in-service education.
When we compared the self-reported pedagogical practices of CLIL teachers to those of regular
teachers, our findings revealed that for all approaches except Literacies CLIL teachers scored signifi-
cantly higher than did the regular teachers. The fact that the difference between CLIL and regular
teachers is not significant for Literacies may be due to the fact that further developing students’ dis-
ciplinary literacies is currently considered to be of similar limited importance for all subjects, regard-
less through which language they are taught (e.g. Meyer et al. 2015). Especially for the Languages
approach, a high degree of variance was explained by the different teacher groups. This allows us
to explain the inconsistency between the quantitative and qualitative results for our first research
question: although CLIL teachers had relatively low scores on Language, the difference with
regular teachers is large. So, compared to regular teachers, CLIL teachers score high on the Language
232 E. VAN KAMPEN ET AL.

approach. This outcome supports the findings of Koopman, Skeet, and de Graaff (2014) that the
approaches to CLIL of teachers in the Netherlands are not limited to the ‘language bath’ conception
of language learning as teachers, for example, place emphasis on stimulating spoken output through
student interaction. However, the findings also show that teachers focus less on other relevant
aspects of language than they do on providing elaborated and scaffolded input.
The qualitative data from the open question also provided some important insights into the spe-
cificities of the Dutch context, particularly that regular lessons are much more textbook-driven com-
pared to CLIL lessons, and that CLIL lessons tend to focus more on higher level cognitive skills. A
further important insight relates to a kind of ‘transfer effect’ of CLIL teaching, that over time
several teachers report having incorporated approaches from their CLIL classes into their regular
classes because they have found them to be effective. This last point supports the claim made by
some researchers that CLIL teaching promotes the adoption of more learner-centered pedagogies
(cf. Dalton-Puffer 2011, 189).
Our second research question was concerned with finding out the extent to which differences in
the self-reported pedagogical practices of CLIL teachers are related to the subject discipline they
teach and how this compares to regular teachers. Our results show that for CLIL teachers, subject dis-
ciplines explained a great deal of variance for all four pedagogical approaches. English teachers
scored significantly higher than teachers of all other subject-groups for the Literacies, Language
and Input approaches. This result can be explained by the fact that the preparation of English tea-
chers includes learning about theories of second language acquisition and related pedagogical
approaches that many of the CLIL teaching models are based on. Given that in the CLIL bilingual
schools context English teachers have such a unique and different position in comparison with tea-
chers of other school subjects, the question may be raised as to whether future studies should include
English teachers as CLIL teachers. The same issue is also relevant for teachers of Other Foreign
Languages.
This study also found that CLIL teachers of the Creative and Physical subjects scored lowest on all
approaches except for Language. A possible explanation for this is that the theoretical models about
what constitutes effective CLIL pedagogy are often less relevant for teachers of such subjects as phys-
ical education, music and drama, because these subjects are less text based.
Our analysis of differences in the self-reported pedagogical practices of regular teachers related to
subject discipline shows that subject discipline significantly influences pedagogical practices.
However, contrary to findings from CLIL teachers, subject discipline only explained variance for
the pedagogical approach Language. For this approach, English teachers and Other Foreign
Language teachers scored significantly higher than teachers of Mathematics and Sciences and of
Social Sciences.
We consider that our findings about the significance of subject disciplines in explaining differ-
ences in the self-reported pedagogical practices of this group of CLIL teachers has an implication
for Dutch CLIL teacher education. In the Netherlands, teacher education programs are generally
based around teaching student–teachers about general models of effective CLIL pedagogy based
largely on theories of second language acquisition, such as Westhoff’s (2004) penta-pie model.
However, our results indicate that it may be relevant to also include more tailored elements, specifi-
cally about how to teach specific subject disciplines through CLIL.
A limitation of this study is that it focuses on the self-reported pedagogical practices of teachers
rather than their actual practices. However, as, in line with Trigwell and Prosser (1996), we consider
teaching approaches to consist of a combination of a teachers’ teaching intentions (why they adopt a
particular strategy) and a teachers’ teaching strategies (what they do), we think self-report is the most
effective way to provide us with an overview of these approaches. A further limitation is that partici-
pation in the CLIL-Q was voluntary so it could be that our sample of participating teachers is biased in
that, for example, teachers who were either very positive or negative about CLIL teaching chose to
complete the questionnaire.
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 233

8. Conclusion
This aim of this research is to contribute to better understanding the pedagogies of CLIL teachers in
the Netherlands by focusing on their reported use of various approaches associated with effective
CLIL teaching. The main conclusions from our study can be summarized as follows:

. CLIL teachers in the Netherlands perceive CLIL to be a unique pedagogy and its uniqueness lies to
a great degree in the fact that it seeks to integrate content teaching with language teaching.
. There appear to be significant gaps between what is considered effective CLIL pedagogy in theory
and the pedagogical approaches reported by these CLIL teachers. Further research should be con-
ducted into the Language approach, and also the claims of recent studies (Morton 2010; Meyer
et al. 2015) that teaching subject-specific literacies is an aspect that needs more explicit attention
in teacher education and teachers’ continuing professional development.
. More investigation is needed about how to interpret CLIL through different subject lenses and
how to help teachers recognize these elements in their subjects.

These conclusions, we believe can contribute to a wider, more informed discussion about CLIL tea-
chers’ classroom pedagogies and consequently serve to further bridge the current gap between
theory and practice in CLIL.

Disclosure statement
No potential conflict of interest was reported by the authors.

References
Admiraal, W., G. Westhoff, and K. de Bot. 2006. “Evaluation of Bilingual Secondary Education in the Netherlands: Students’
Language Proficiency in English.” Educational Research and Evaluation 12 (1): 75–93. doi:10.1080/13803610500392160.
Badertscher, H., and T. Bieri. 2009. Wissenwerb Im Content-and-Language-Integrated Learning. Bern-Stuttgart-Wien: Haupt.
Cenoz, J., F. Genesee, and D. Gorter. 2014. “Critical Analysis of CLIL: Taking Stock and Looking Forward.” Applied Linguistics
35 (3): 243–262. doi:10.1093/applin/amt011.
Coonan, C. M. 2007. “Insider Views of the CLIL Class through Teacher Self-Observation Introspection.” International
Journal of Bilingual Education and Bilingualism 10 (5): 625–646. doi:10.2167/beb463.0.
Coyle, D. 1999. “Theory and Planning for Effective Classrooms: Supporting Students in Content and Language Integrated
Learning Contexts.” In Learning Through a Foreign Language, edited by J. Masih, 46–62. London: CILT.
Coyle, D., P. Hood, and D. Marsh. 2010. Content and Language Integrated Learning. Cambridge, UK: Cambridge University Press.
Dalton-Puffer, C. 2007. Discourse in Content and Language Integrated Learning (CLIL) Classrooms. Amsterdam: John
Benjamins.
Dalton-Puffer, C. 2011. “Content-and-Language Integrated Learning: From Practice to Principles?” Annual Review of
Applied Linguistics 31: 182–204. doi:10.1017/S0267190511000092.
Dalton-Puffer, C., and U. Smit. 2013. “Content and Language Integrated Learning: A Research Agenda.” Language
Teaching 46 (4): 545–559. doi:10.1017/S0261444813000256.
De Graaff, R., and O. van Wilgenburg. 2015. “The Netherlands. Quality Control as a Driving Force in Bilingual Education.” In
Building Bilingual Education Systems: Forces, Mechanisms and Counterweights, edited by P. Mehisto and F. Genesee,
167–180. Cambridge: Cambridge University Press.
De Graaff, R., G. J. Koopman, Y. Anikina, and G. Westhoff. 2007. “An Observation Tool for Effective L2 Pedagogy in Content
and Language Integrated Learning (CLIL).” International Journal of Bilingual Education and Bilingualism 10 (5): 603–624.
doi:10.2167/beb462.0.
Koopman, G., J. Skeet, and R. de Graaff. 2014. “Exploring Content Teachers’ Knowledge of Language Pedagogy: A Report
on a Small-Scale Research Project in a Dutch CLIL Context.” The Language Learning Journal 42 (2): 123–136. doi:10.
1080/09571736.2014.889974.
Lyster, R. 2007. Learning and Teaching Languages through Content: A Counter-Balanced Approach. Amsterdam: John
Benjamins.
Meyer, O., D. Coyle, A. Halbach, K. Schuck, and T. Ting. 2015. “A Pluriliteracies Approach to Content and Language
Integrated Learning – Mapping Learner Progressions in Knowledge Construction and Meaning-Making.” Language,
Culture and Curriculum 28 (1): 41–57. doi:10.1080/07908318.2014.1000924.
234 E. VAN KAMPEN ET AL.

Morton, T. 2010. “Using a Genre-Based Approach to Integrating Content and Language in CLIL.” In Language Use and
Language Learning in CLIL Classrooms, edited by C. Dalton-Puffer, T. Nikula, and U. Smit, 81–104. Amsterdam: John
Benjamins.
Nikula, T. 2010. “On effects of CLIL on a Teacher’s Language Use.” In Language Use and Language Learning in CLIL
Classrooms, edited by C. Dalton-Puffer, T. Nikula, and U. Smit, 105–124. Amsterdam: John Benjamins.
Pérez-Cañado, M. L. 2012. “CLIL Research in Europe: Past, Present and Future.” International Journal of Bilingual Education
and Bilingualism 15 (3): 315–341. doi:10.1080/13670050.2011.630064.
Pérez Cañado, M. L. 2014. “Teacher Training Needs for Bilingual Education: In-Service Teacher Perceptions.” International
Journal of Bilingual Education and Bilingualism. doi:10.1080/13670050.2014.980778.
Swain, M. 1995. “Three Functions of Output in Second Language Learning.” In Principle and Practice in Applied Linguistics,
edited by G. Cook and B. Seidlhofer, 125–144. Oxford: Oxford University Press.
Trigwell, K., and M. Prosser. 1996. “Congruence between Intention and Strategy in Science Teachers’ Approach to
Teaching.” Higher Education 32: 77–87.
Westhoff, G. 2004. “The Art of Playing a Pinball Machine. Characteristics of effective SLA-tasks.” Babylonia 3: 58–62.
 INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 235

Appendix 1. Background variables considered for CLIL teachers only.

Variable CLIL teachers (%) (N = 218)

Years CLIL teaching experience
0–2 22.4
3–5 26.8
6–10 28.8
11 or more 22
Hours per week teaching CLIL classes
1–2 10.8
3–4 16
5–6 19.3
7–8 16
9–10 9
11 or more 28.3
Currently teach regular and CLIL classes?
Yes 71.2
No 28.8
How became a CLIL teacher?
Applied for a job 22.9
Asked to teach in the CLIL section 15.4
Invited by school management 52.3
Other 9.3
Completed a university Master CLIL program?
Yes 15.1
No 84.9
Completed a University of Applied Sciences CLIL program?
Yes 3.2
No 96.8
Followed CLIL training at school?
Yes 55
No 45
Followed CLIL training by external institute?
Yes 46.3
No 53.7
Native English speaker?
Yes 18.5
No 81.5
Multi-lingual background?
Yes 17
No 83
Highest formal certified level of English?
C2 CEFR 44.2
C1 CEFR 15.9
Qualified English teacher 24.5
No formal qualification 15.4
Speaking ability
Poor 0
Below average 1
Average 14.8
Good 33
Excellent 51.2
Listening ability
Poor 0
Below average 0
Average 9.1
Good 34
Excellent 56.9
Reading ability
Poor 0
Below average 1.4
Average 9.7
Good 25.1

(Continued)
236 E. VAN KAMPEN ET AL.

Continued.
Variable CLIL teachers (%) (N = 218)
Excellent 63.8
Writing ability
Poor 0.5
Below average 4.8
Average 23.6
Good 38.5
Excellent 32.7

Appendix 2. Factor loadings of the pattern matrix based on principal component

analysis with oblique rotation for 28 items of the CLIL-Q (N = 296).

Item Literacies Language Scaffolding Input

1. Helping students to recognize common text structures relevant to the subject. .68
2. Providing students with tasks in which they learn to apply reading strategies. .68
3. Helping students to recognize text features specific to a subject. .61
4. Providing students with tools that help direct their attention to what is important .60
in information sources.
5. Helping students to notice the purpose for which subject-specific text types are .60
written.
6. Providing students with feedback about the fluency of their language. −.76
7. Providing students with feedback to produce more accurate forms of spoken −.71
language.
8. Encouraging students to provide feedback about language on their own and −.70
others’ work.
9. Formulating language goals when planning lessons. −.69
10. Activating students’ prior language knowledge about a topic. −.60
11. Grading students’ language use in oral assignments. −.58
12. Providing students with speaking scaffolds when setting speaking tasks. −.58
13. Urging students to only use the language of instruction in class. −.54
14. Working together with English teachers when preparing lessons. −.49
15. Asking students to explain their reasoning. .74
16. Using different kinds of questioning to help students understand content. .60
17. Providing students with tools that help them to organize, understand and .52
record what they observe.
18. Activating students’ prior content knowledge about a topic. .50
19. Making use of body language while teaching to enhance student .48
understanding.
20. Providing students with tasks that help them to use subject-specific terms that .48
are key for a lesson.
21. Asking students to change content from one form into another. .40
22. Selecting authentic materials for my lessons. .73
23. Creating my own lesson materials. .72
24. Providing students with different kinds of content. .71
25. Using different types of assessments. .63
26. Drawing links between study content and local and global issues. .53
27. Using visual aids while teaching. .38 .53
28. Providing students with input from a range of perspectives. .53
Note: Factor loading <.3 are not shown.