A Study on CLIL Secondary School Teachers in Spain: Views, Concerns and Needs

. Content and Language Integrated Learning (CLIL) is a methodological approach that is growing very fast in many European countries, particularly in Spain. The implementation of bilingual programmes in primary, secondary and tertiary education has produced significant changes that have had direct consequences on all educational stakeholders, including teachers, parents and students. In the case of CLIL teachers, research has often addressed their training needs, and actions towards preparing them for successful classes have been proposed. However, few studies have focused on their concerns and views of bilingual programs. Despite the fact that many researchers have acknowledged the importance of understanding CLIL teachers’ views and beliefs, thus hoping for more studies on those issues, this is not yet one of the major research targets. In this study CLIL secondary school teachers in Spain were approached in order to identify the problems they encounter when implementing CLIL. An online questionnaire with both open and close questions was designed administered to informants across Spain. The findings reveal that, overall, the difficulties teachers encounter when implementing a bilingual programme are multiple, and many informants believe that the bilingual programme in English needs a comprehensive reform in Spain.


Introduction
Content and Language Integrated Learning (CLIL) dates back to the 90s. This methodology has been widely implemented across Europe and it is now extending to other parts of the world like Asia (cf. Ikeda 2014;Yang 2015). The term CLIL was proposed by the University of Jyväskylä and the European Platform for Dutch Education to describe educational methods with a dual-focused educational approach in which an additional language is used for the learning and teaching of both content and language (Maljers, Marsh & Wolff 2006). Schools and other educational institutions such as universities provide the main opportunities for people to learn languages, and the European Union needed to find an approach that could achieve their goals. CLIL started in Europe as an educational method aimed at meeting European demands regarding multilingualism and the development of European citizens' linguistic and intercultural skills and abilities. Following the 2002 European Council Meeting held in Barcelona, Member States were urged to foster the mastery of basic skills by teaching at least two foreign languages from an early age. This recommendation has been commonly referred to as the Barcelona goal of the "mother tongue +2". In 2008 the Council of the European Union adopted the Resolution on a European Strategy for Multilingualism with a view to enhancing mobility across Europe, and intercultural education among others.
In the 2014 Council Conclusions on Multilingualism and the Development of Language Competences, the Council of the European Union invited EU member states to make a strong commitment to improving multilingualism and enhancing the quality and efficiency of language learning and teaching, by teaching at least two languages in addition to the main language(s) of instruction and by exploring innovative approaches to the development of language competences. According to the Council, the ability to communicate in a language other than one's mother tongue is one of the key competences which citizens should seek to acquire. Studies results showed that the level of language skills of many young people in the EU could be improved (Council of Europe, 2014, p. 2).
There has been a rapid spread of CLIL in Europe but especially in Spain, both in bilingual and monolingual settings. In the Autonomous Community of Madrid, for example, the bilingual programme has been running for 15 years. In Madrid the bilingual programme started in the 2004-2005 academic year in 26 public primary schools; it was extended to semi-public primary schools (concertados) in [2008][2009], and then to public secondary schools in 2010andto Bachillerato (Sixth Form) in 2014-2015. In vocational schools, it was implemented in 2016-2017. The first students sat the PAU exams (exams to enter university) in 2016. In the 2016-2017 academic year, 45% of public primary and secondary schools in the Autonomous Community of Madrid were part of the bilingual programme. In the 2017-2018 academic year, the number of bilingual public primary schools was 369 and that of bilingual public secondary schools was 152 ("D.G. de Educación concertada, Becas y Ayudas al Estudio", 2017).
Another autonomous community where the implementation of bilingual programmes has been extensive is Andalusia, as evidenced by the Andalusian government's Plurilingualism Promotion Plan (Plan de Fomento del Plurilingüismo) for primary and secondary schools which was initiated in 2015. In the 2020-2021 academic year a total of 1192 public schools in this region offered bilingual education (Consejería de Educación y Deporte. Junta de Andalucía, 2020).
Plenty of voices have celebrated the positive aspects of CLIL and bilingual programmes (Lorenzo, Casal & Moore, 2010), while negative opinions of CLIL are few. Some critical voices do appear, such as Bruton (2011aBruton ( , 2011bBruton ( , 2013 who backs his claims by mentioning students' motivation and lack of self-esteem when they are not confident in the language of instruction, teachers' inability to impart knowledge at the levels required, or the fact that CLIL is not always egalitarian. Bilingual programmes are also the subject of ongoing debates in the Spanish national press (Pérez-Barco 2020 ;Recio 2021;Rius 2018;Sánchez 2019;Torres Menárguez 2018. As Cenoz, Genesee & Gorter (2014) indicate, 'Even though the development of CLIL has stimulated research on [sic] content and language integrated learning, there are important empirical gaps in our understanding of its effectiveness' (p. 257). And they further explain: Fundamental issues about the effectiveness of CLIL remain unexamined. Specifically, much, if not most, research on CLIL has been conducted by ESL/EFL scholars who have compared CLIL and non-CLIL groups of learners and reported higher achievement in English for CLIL learners (Coyle 2007a;Järvinen 2007;Lorenzo, Casal & Moore, 2010). Although these results provide general support for CLIL [….], they do not establish a clear causal link between integrated language and content teaching and learner outcomes. CLIL instruction usually entails more contact hours with the target language during the school day, and it could be this extended exposure to the target language that is the crucial variable (Tedick and Cammarata 2012) (p. 257).
From a general perspective, it seems that though there have been many publications on teacher needs (cf. Doiz & Lasagabaster 2017;Pérez Cañado 2016), very few of them actually deal with secondary teachers in bilingual education programmes. Although CLIL programmes have attracted much scholarly interest and some researchers have stressed the fact that CLIL teachers' opinions should be taken into account (Hüttner, Dalton-Puffer & Smit 2013;Hüttner 2018), research has concentrated primarily on issues related to students' performance, whether content is acquired in the same way as when taught in the student's native language, and the effects of CLIL on mother tongue (especially writing), and content subject competence (cf. Hunt 2011;Dalton-Puffer 2011;Llinares & Whittaker 2007;Whittaker, Llinares & McCabe 2011).
In the same vein,  insisted that researchers should investigate lay beliefs about the perceived success (or lack of it) of CLIL at different educational institutions. In their article 'Content and Language Integrated Learning: A research agenda' (2013), they highlighted the fact that CLIL methodology for teaching a foreign language requires further investigation, and outline action research that must be carried out in areas highly pertinent to CLIL but so far underrepresented by scholars. And they proceed to name a few examples: These are, firstly, policy issues, comprising policy statements as well as stakeholders' perceptions of CLIL and its success; secondly, classroom discourse as the prime site for the investigation of CLIL practices and their implications for the learning process; and, thirdly, classroom pedagogy, with the focus on potential differences between CLIL and non-CLIL settings (p. 545).
It is mandatory to listen to the professionals working in bilingual programmes to get insights into and knowledge of how CLIL is functioning on a daily basis in an effort to find ways to improve it. For example, issues stakeholders at bilingual schools face in their daily practice have been widely avoided. Apart from language policy regulations, CLIL stakeholders such as teachers, members of the school board (headteachers, secretaries and the heads of studies), parents, students, bilingual programme coordinators, etc. are an important strength in the development of innovative bilingual programmes and their voices should be heard. What do they think about using another language to teach/learn contents? Do they think bilingual programmes in Spain have been successful? What are their needs? Investigations should be carried out on the beliefs about the perceived collaboration or lack of it among teachers and other stakeholders in bilingual secondary schools in Spain. CLIL teachers work hard in an effort to manage bilingual programmes on top of other chores such as teaching, taking care of student discipline, and paying attention to diversity in the classroom. Overall, a thorough diagnosis of CLIL teachers' needs and concerns must be implemented.

Method
The main objective of this research is to approach content subject secondary teachers in Spain in order to first of all find out more about their personal characteristics and background, including training, language proficiency, experience. Then this work also aims at obtaining information related to those teachers' attitudes towards teaching content subjects in English, and also identifying the problems they encounter when implementing CLIL at their schools with regard to coordination and communication with the following stakeholders: • school board, • bilingual programme coordinator at the school, • different departments (Maths, English, Science, Technology, etc.) at the school, • rest of teaching staff (both CLIL and non-CLIL teachers).
For the present study, an anonymous online questionnaire was designed as a survey tool (see Appendix). The questionnaire was a Google form in Spanish and respondents were reached via social media or email. The time frame within which the survey was completed was approximately a year. The questionnaire was composed of 18 items and consisted of 2 sets of questions: the first ones were background questions where identification (subject) variables were contemplated. The variables related to the individuals' characteristics (age, gender) and also the type of institution they were working at (public-private-semi-public (schools sponsored by a public voucher system)), their place of work (within Spain) and the content subjects they were teaching in English.
The following set of questions was made up of opinion and value questions to check the informants' current level of English as well as their feelings about bilingual programme coordination. The questionnaire included short-and long-answer questions, and some close-and some open-response items.
Informants were 86 secondary school teachers spread across Spain and the subjects taught in English included Technology, Arts and Crafts, PE, Geology and Biology, Maths, Physics and Chemistry, Ethical Values, and Natural and Social Science.
The statistics resulting from the responses obtained were then subjected to tests aimed at finding out whether there were some significant correlations between the variables that were studied. For that purpose, Pearson's chi-square test was chosen as it had the potential of providing us with reliable data as to how the aforementioned variables interacted with each other.

Results
The first question was if the informants had any theoretical knowledge of or previous training in bilingual education. The requirements set by central authorities for CLIL teachers vary from country to country. In Spain, in addition to a teaching qualification, it is necessary to hold a certificate proving thorough knowledge of the target language. The minimum level required is usually CEFR B2, but there are some variations across Autonomous Communities, which are responsible for setting additional requirements. Some Autonomous Communities require the so-called habilitación lingüística, which can be roughly translated as 'Linguistic Capability Certification' and which can be obtained as part of the subject qualification or through a certificate issued by the Escuelas Oficiales de Idiomas (public language institutes) or other accredited institutions (e.g. the Valencian Community, the Region of Murcia, the Principality of Asturias, the Canary Islands and Cantabria). Other Autonomous Communities require specific training in CLIL methodology (e.g. Cataluña, La Rioja, Community of Castile and León, the Valencian Community, the Region of Murcia, the Canary Islands and Extremadura). (Eurydice 2017, pp. 91-92). In some Autonomous Communities of Spain further requirements are needed, and teachers must participate in specific training courses on CLIL methodology (Eurydice 2017, p. 93;Manzano Vázquez 2015).
CLIL teachers (in Spanish profesores de áreas o disciplinas no lingüísticas (ADNL)) must be qualified teachers in both their subject and a foreign language.
According to the Guía Informativa para centros de enseñanza bilingüe ('Guide for bilingual educational institutions') (2013) by the Regional Government of Andalusia, one of CLIL teachers' most important tasks is to adapt the curriculum of their content subject by incorporating aspects related to the second language they use as a medium of instruction while always respecting the minimum contents of each subject. Ideally, in order to teach content subjects in a foreign language, teachers in bilingual schools should have a good knowledge of both the subject taught and the language in which it is taught. They are also expected to be familiar with the requirements of CLIL methodology. Surprisingly, more than half of the content subject teachers who participated in this study had no previous training in bilingual teaching. As for question number two, namely the level of English according to the Common European Framework of Reference for Languages (CEFR), most informants hold a B2 or C1 certificate. Figure. 2. Level of English.
As concerns question three, years of experience using CLIL, most informants had between 1 and 10 years of teaching experience using CLIL methodology. However, despite the fact that most of the informants hold a B1 or a C1 certificate and have been teaching in bilingual programmes for some years, nearly half of them (46,90%) do not feel comfortable teaching in a language that is not their mother tongue. Question number four tackled a very complex issue, namely, self-perception about teaching content in a foreign language. Some of the most important stakeholders for the success of bilingual programmes are non-native teachers who have to teach contents in a foreign language and, sometimes in monolingual settings like most regions in Spain. CLIL teachers' responsibilities and challenges are numerous and readjustments in their methodologies are needed (cf. Moate's (2011) investigation on the effects of FL-mediated teaching on teachers' professional integrity).
A good number of respondents point to their poor command of and fluency in the English language. Others complain about their lack of linguistic resources. That leads many to pay more attention to their linguistic skills than to the contents they teach. In this sense, they believed they waste a lot of time that could be employed on designing and improving the contents to be taught. The fact that English is not their mother tongue makes them feel insecure. Some even pointed out that teaching in a foreign language makes the teaching look 'theatrical and unnatural' because nuances are lost in the explanation of concepts. Some informants argued that contents have to be taught well in the mother tongue so that they can be assimilated correctly. Finally, some believed that teaching content subjects in a foreign language means extra work without any reward. Overall, many believed CLIL methodology hinders their teaching profession.
The rest of the questions in the questionnaire focused specifically on communication/coordination with other stakeholders at the bilingual secondary school where they were working.
The bilingual programmes project in Spain has yielded changes that have affected and still affect many aspects of the educational institutions (organization, personnel, and functions, changes in the teachers' schedules and the contents, and also methodological adaptations) which means greater coordination between all the stakeholders at the school.
Question number five was meant to ask informants if there existed collaboration among all the stakeholders in the bilingual programme (school board, bilingual programme coordinator, CLIL teachers, departments, etc.). 58,6% of the teachers that participated in the study believed there was coordination between all the stakeholders in the bilingual programme at their school. Those who maintained there was no coordination highlighted the fact that their timetables did not include slots for coordination meetings.
Question number six focused on coordination between CLIL teachers and the school board. Research on coordination between school boards and CLIL teachers in the bilingual programmes is non-existent, though some research works such as Laorden Gutiérrez & Peñafiel Pedrosa's (2010) have discussed school administrations' attitudes towards the bilingual project. The present study reveals that secondary schools differ from one another in terms of coordination between school boards and bilingual programme teachers or at least there might be no consensus among the CLIL teachers that took part in the study. While 40% of the informants stated that there was coordination between the school board and the CLIL teachers at their schools, the same amount of teachers had a different impression of their schools.
Another important question was if there was enough coordination or communication between the bilingual programme coordinator and CLIL teachers at the informant's school. Bilingual programme coordinators play an important role in the success of the implementation at any school. For example, according to Article 28 of the Order of 28 June 2011, which regulates bilingual education in schools in the Autonomous Community of Andalusia, bilingual public schools will count on a teacher responsible for the coordination of the bilingual programme/teaching, appointed from among the language teachers with a permanent job position at that school. The coordinator will work together with the school board. On some occasions, and by delegation, the coordinators must assume important functions and actions, among others: • To ensure the correct implementation of the methodological, curricular and organizational model. • To convene teaching staff meetings to coordinate the elaboration of the integrated curriculum of the languages in the framework of the educational project of the centre. • To propose and coordinate the activities of the teaching staff and, where appropriate, the departments in relation to the development of bilingual education. • To coordinate the different actions that are developed in the teaching centre in relation to bilingual education. • To inform and advise students' families. (cf. Guía informativa para centros de enseñanza bilingüe, 2013, p. 37, my translation). The study reveals that there is a tendency towards communication between the programme coordinator and CLIL teachers.
The following question was related to coordination or communication between the English Department and CLIL teachers at the school. The pedagogical CLIL models vary across countries, but in general CLIL promotes team teaching, that is a content teacher working together with a language teacher (Banegas 2011). In the Community of Murcia, for example, Article 4 of Order of 18 April from 2011, from the Consejería de Educación, Formación y Empleo ('Regional Ministry of Education, Training and Employment'), states that one of the tasks CLIL teachers must take on is creating teaching materials in coordination with the rest of the teaching staff, but especially with the English Department.
Nearly 60% of the informants in the study considered that there is no coordination with the English Department, and complained there are no hours allocated for coordination with English teachers during the official timetable. Some maintained that, even if there were, the members of the English Department would not get involved and change their methodology to coordinate with content subject teachers, except on specific occasions. Some respondents complained about the absence of assistance from the English department and believed that it should be a tool at the service of the rest of content subject departments. Some pointed out that many times it is not a matter of lack of good will on the part of staff in the English department, but that overloading the teaching staff with multiple bureaucratic tasks impedes the development of cooperation strategies. In general, teachers are of the opinion that bilingual programmes 1) require a change of mentality and work dynamics that not everyone knows how to bring about; 2) more time to meet and organize things correctly is needed; 3) some reduction of class hours to prepare materials is needed. Another important issue is if there is enough coordination or communication among CLIL teachers. In this respect, more than 50% believed that there is no coordination with other CLIL teachers, and the main reason is the lack of timetable slots for meetings. Some pointed out that, in order to do a good job in the bilingual programme, they all should have more time to get together and organize things correctly. CLIL teachers also need some reduction of class hours to prepare materials. Figure 9. Coordination among CLIL teachers.
As for collaboration between the different departments at secondary schools, interestingly enough, 58,7% of the teachers that participated in the study stated that there is no coordination between the different departments at their schools. Most complained there is no set time for meetings, and asked for a common strategy to avoid curriculum duplication in various subjects. Many believed that this lack of coordination has nothing to do with the schools being bilingual or monolingual, but instead is a common problem in the Spanish educational system. In relation to the responsibility for the success of the bilingual programme at their school, 57,8% of the informants believed that the responsibility falls mainly on both the CLIL teachers and the programme coordinators, whereas 42% of the respondents believed it falls on the students' performance. Some teachers also believed that, in order for the bilingual programme to be successful, families' collaboration is needed as long as it is not intrusive, as well as collaboration from students and highlighted the formative level of families. Besides, some maintained that the entire educational community must be convinced of the relevance and usefulness of the bilingual programmes, whose results can only be appreciable in the long term. As for the good functioning of the bilingual programme at their school, 57,1% of the informants agreed that more involvement on the part of the various departments is necessary whereas 42,9% believed more involvement is needed on the part of the programme coordinator, followed by the school board. Other stakeholders mentioned by the informants were: 1. Primary schools: Greater participation and collaboration with the primary schools the students are coming from is essential. 2. Students' families: More coordination with families is desired. 3. Educational authorities: achieving a better management of the ratio of bilingual groups is fundamental for the good implementation of CLIL programmes in class. Most teachers firmly agreed that, in order to do a good job, they should have time to get together and organize everything correctly, and to prepare material. The most important complaint throughout the survey was the lack of meeting hours in their timetables. Most teachers feel the need for tools and hours on their timetables consequently that all the members of the bilingual programme can coordinate. They also yearn for more CLIL materials (other than the textbooks) so that they do not have to waste a lot of time preparing them. There is no lack of involvement in general, and there are too many students per class, so there is a lot of work and little time to coordinate. Some informants also advocated a revision of the language assistant role in content subject classes. According to one informant: "Language assistants are useful in the English classroom but you can hardly cover the syllabus when you must use a third of your teaching hours with a teaching staff who is not trained in your subject or in any teaching methodology".

Discussion
After having obtained the findings presented earlier, we decided to establish correlations between some variables involved in this research. For that purpose, we resorted to Pearson's chi-square test, whose results seemed to indicate that there were statistically significant correlations between variables such as gender, age, language proficiency, previous experience, confidence while teaching in English and opinion regarding collaboration and coordination.

Gender and opinion on whether or not there is collaboration between stakeholders in the bilingual programme
Since our respondents included both men and women, we decided to start our correlation tests by trying to find out whether gender could determine responses to questions regarding previous knowledge or training in CLIL, confidence while teaching in a foreign language and collaboration between bilingual programme stakeholders.
The chi-square tests that were carried out led to the conclusion that there is a significant difference between men and women only when it comes to whether they believe there is collaboration amongst all the stakeholders in the bilingual programme. As a matter of fact, the tests revealed that male respondents mostly gave a negative answer, while female respondents mostly agreed. It therefore seems that women who teach in bilingual programmes are more likely to be satisfied with the amount of collaboration between teachers and educational stakeholders than their male counterparts.

Age
When it comes to age, we can say with some degree of certainty that answers to a couple of questions were correlated, to some extent, with the age groups to which respondents belonged. Overall, the respondents were divided into three age groups, namely participants aged under 35, those between 35 and 45 years old and finally informants who were above 45.

Age and previous knowledge or training in CLIL
Though there was no statistically significant difference between respondents who were under 35 and those aged between 35 and 45 and neither was there any difference between informants aged above 45 and the group right under them (35-45), the tests revealed a statistically significant difference between the youngest (under 35) and oldest (above 45) participants. More specifically, while younger participants tended to have a negative answer, most older participants were positive when asked about whether they had some theoretical knowledge or previous experience in CLIL. This is no surprise, as teachers are often encouraged to take courses in order to improve on their skills. Nevertheless, such courses are often aimed at more experienced teachers who hold a stable position at the educational institution and need refresher courses so as to be aware of latest methodologies and improve their skills. Younger teachers are more likely to be in the early years of their career (as discussed later) and might not have had the opportunity to take such courses for many reasons, including the nature of their contracts that do not entitle them to such benefits or the (wrong) assumption that they do not need such courses because they have just graduated. This might explain why they generally lag behind their older colleagues in terms of previous knowledge or training in CLIL.

Age and opinion on whether or not there is coordination and communication amongst CLIL teachers and bilingual programme coordinators
Age also determined to some extent the answer to the question as to whether participants believed there is enough collaboration between CLIL teachers and coordinators. Here again, though the differences between some age groups were not that significant, we can say, based on the data obtained after subjecting our data to Pearson's chi-square test, that although both age groups would likely have a positive answer, participants aged above 45 would most likely provide a positive answer than their counterparts aged between 35 and 45 (see Table 3 below). This same situation was experienced between respondents under 35 and those between 35 and 45 when it comes to their opinion on whether or not there is enough coordination and communication amongst CLIL teachers themselves (see Table 4 below). The percentage of positive answers amongst respondents under 35 was much higher than that of respondents aged between 35 and 45.

Language proficiency
Before going any further in our analysis of statistics it is worth stating that four language proficiency levels were identified in our survey, namely B1, B2, C1 and C2. This explains why there were some statistically significant differences in responses when comparing some groups and not others.

Language proficiency and previous knowledge of or training in CLIL
When it comes to the question about whether they have theoretical knowledge or some previous training in CLIL, there were no statistically significant differences between the responses provided by respondents at B1 and B2 proficiency levels, who would most likely provide a negative answer to the question. In addition, though C2 respondents would more likely provide a more positive answer than their B1 respondents, the difference was not statistically significant, and neither was the difference between C1 and C2 respondents who all tended to have a positive answer. Nevertheless, there were statistically significant differences between participants at the B1 level on the one hand and their counterparts at C1. While the former indicated that they had no theoretical knowledge or previous training in CLIL, most C1 respondents indicated that they did have some knowledge of and training in CLIL. Pearson's chi-square test results revealed a similar statistically significant difference between respondents belonging to the B2 proficiency group who mostly gave a negative answer to the question and the ones in the C1 group whose majority did the opposite. The same finding was obtained when describing B2 and C2 respondents.
Therefore, we can conclude with very little room for doubt that there was a positive correlation between language proficiency and previous knowledge or training in CLIL. Teachers who were highly proficient in English would most likely state that they had some previous knowledge and experience in CLIL, and the contrary was true of low-proficiency teachers. Furthermore, since there was some degree of dependence between age and previous knowledge of or training in CLIL, with the youngest teachers being likely to have little knowledge and training with regard to bilingual education, we can somehow conclude that younger teachers might also lack proficiency (at least when it comes to using the foreign language to teach) than their older colleagues

Language proficiency and confidence while teaching in non-native language (English)
Language proficiency was also correlated with confidence while teaching a language that is not one's mother tongue, and based on the relevance tests carried out, we can say with a high degree of certainty that proficiency level affects confidence when teaching a foreign language. More specifically, there were no statistically significant differences between B1 and B2 respondents who all tended to feel less confident and neither were there statistically significant differences between informants belonging to the C1 and C2 categories who overall felt confident teaching in English. However, there were clear statistically significant differences between B1 respondents who mostly provided a negative answer to the question regarding their confidence, and C1 and C2 respondents (respectively) who assured of their confidence while teaching in English. Similar findings were obtained when comparing B2 respondents to their C1 and C2 counterparts, respectively.
The aforementioned findings are not that surprising, as more proficient speakers tend to feel more comfortable communicating in a language than less proficient ones.

Previous experience
When it comes to previous experience, the results obtained following the tests we performed revealed that there could be some statistically significant differences between respondents, depending on how long they have been teaching in bilingual programmes. First of all, it is worth mentioning that respondents were divided into four groups, namely those who had less than a year of experience, then respondents whose experience ranged between 1 and 5 years, 5 and 10 and finally those respondents with more than 10 years of experience. Overall, we realised that more experienced informants would more likely indicate that they have some theoretical knowledge and previous training in CLIL than their less experienced colleagues. This goes hand in glove with what was obtained earlier, when we correlated age with previous knowledge or training in CLIL

Previous experience and previous knowledge or training in CLIL
There was a significant difference between respondents holding between 5 and 10 years of experience, on the one hand, who mostly gave a positive answer to the question as to whether or not they had some previous experience and training in CLIL, and on the other hand respondents with an experience of less than a year, whose answers were mostly negative. The same statistically significant difference was obtained when comparing respondents with more than 10 years of experience to those with less than a year, or their counterparts whose experience ranged between 1 and 5 years. There were no statistically significant differences between respondents falling within contiguous categories describing years of experience, which points to the fact that a broad difference in terms of years of experience would likely determine the type of answer to expect.

Previous experience and confidence while teaching in non-native language (English)
Previous experience also determined to a greater extent whether or not respondents felt confident while teaching in a foreign language. In general, we realised that there was a positive correlation between years of experience and confidence level, though not all cases were statistically significant. Nevertheless, when comparing respondents with less than a year of experience and their counterparts having between 1 and 5 years of experience, the chi-square test clearly indicated that there was a statistically significant tendency for more experienced respondents (1 to 5 years) to provide a positive answer to the question and for less experienced ones (less than 1 year) to provide a negative answer. This very statistically significant finding was obtained after comparing respectively respondents with 5 to 1 years and those having more than 10 years of experience, who would mostly be more confident, to respondents with less than a year, who, as mentioned earlier, lacked confidence when teaching in English. It is worth mentioning here that, after the one-year mark, there were no statistically significant differences between respondents. This is not surprising either, as based on the findings presented earlier, respondents with more experience would have most likely received some training in CLIL coupled with the fact that practice definitely makes perfect.

Conclusions
The conclusions of this research try to set up the agenda for further research on CLIL teachers' demands. One main conclusion that can be drawn from the above statistics is the fact that younger teachers need more attention in order to fully fit within the bilingual programme. As a matter of fact, most of them are often less experienced than their older counterparts and might lack adequate training when it comes to teaching content subjects in a foreign language. In addition, a good number of younger teachers have the feeling that there is not enough collaboration and coordination amongst all educational stakeholders. Therefore, it would be necessary for school authorities to make sure that they have younger teachers in mind when offering professional development courses. In addition, as we mentioned earlier, younger teachers' lack of experience might also affect their fluency and confidence while teaching content subjects in a foreign language. Therefore, they should receive support from their peers and training that might help them improve their oral performance in the classroom.
All these issues must be tackled by educational authorities and headteachers in an effort to help teachers and bilingual programme coordinators at schools to carry out successful bilingual programmes. In this line, the UNED is presently coordinating the Erasmus+ project BiMo (Bilingualism in Monolingual Contexts) whose main aim is, with the help of secondary school headteachers and teachers, to deepen the views, impressions and needs of parents in monolingual societies with implemented bilingual education programmes, and to establish synergies from societies with greater experience in everyday bilingualism and bilingual education programmes in a foreign language (https://bimo.pixel-online.org/). As Cenoz, Genesee & Gorter (2014) point out: 'We believe that it is time for CLIL scholars to move from celebration to a critical empirical examination of CLIL in its diverse forms to better identify its strengths and weaknesses in different learning contexts. It is important that there not be just more research, but rather more critical research on CLIL' (p. 262).
According to the informants that participated in the study, overall, the difficulties teachers encounter when implementing a bilingual programme at their school are multiple, and many believe that the bilingual programme in English needs a comprehensive reform in Spain. Some issues for further analysis are the following: 1. The role of the language assistant: According to the Community of Madrid's 'Language assistant guide' (academic course 2018-2019), the role of the language assistant consists in reinforcing the students' oral skills collaborating in the learning-teaching process in the foreign language subject and any other subject taught in English. According to the Regional Government of Andalusia's Guía informativa para centros de enseñanza bilingüe ('Guide for bilingual education centers') (2013), the work of CLIL teachers is reinforced by the presence of the language assistant who, among other tasks, must help CLIL teachers in designing teaching materials and bringing students closer to the culture of the country where the FL is spoken. CLIL teachers are supposed to maintain fluid contact with the language assistants at all times. In this sense, some informants considered that the role of the language assistant should be reviewed, and suggested that it would be more convenient to use the economic resources to pay for language assistants rather than to reduce the number of students per classroom. They also proposed limiting the conversation assistants to the English classroom. Informants also complained that most of the language assistants lack the teaching experience they need, do not have an adequate 'teaching spirit' and are not really integrated into the life of the school. Fewer and fewer teachers want to have language assistants in class because many are not prepared, they do not have any knowledge about the content subject taught, and it means more work for the content subject teacher. One informant maintained: 'the selection should be more focused on people with a teaching vocation that do not come to Spain only to party and travel around Europe'. 2. Lack of homogeneous groups: Some informants believed that, in order to progress, groups should include students with a homogeneous level of English. Some students come from non-bilingual schools. 3. Lack of coordination with primary schools (up to sixth grade, 12yrs). In order to attain better results, there should be communication with the primary schools the students are coming from. 4. Lack of time for designing CLIL materials, applying for European programmes (Erasmus+), etc. Teachers see Eramus+ programmes as a way of promoting mobility and international encounters, both for teachers and students. But participation in Erasmus+ programmes requires a lot of work, effort and dedication and that is why most of the time this is carried out by teachers without family responsibilities since there is no reduction in the official schedule to allow for preparing this type of projects. Informants also complained because time reductions are often used for bureaucratic tasks related to tutoring, administrative duties, etc. Some highlighted the fact that without this reduction the coordination between different teachers and departments is unfeasible. If this is not taken into consideration, teachers can hardly work in teams, and detect strengths or weaknesses in their work. 5. Parents' expectations: Parents tend to choose a secondary school for their children only because it is bilingual, without actually knowing what to expect from bilingual programmes. According to many informants, parents and families should be informed prior to choosing the school where they are going to enrol their children. 6. Teachers' linguistic skills update: There is no regulated training according to the levels of the ADNL teaching staff, so there is little linguistic update. Informants complained that the linguistic update is often done by them in their free time and with their own money. Teachers in the study also explained that, although most teachers at their school have a C1 level, they have no possibility of maintaining it or improving it with the current training programmes. (At the Andalusian Official School of Languages, for example, free C1 exams were organized for the first time in June 2017). 7. A form of discrimination and selection: Many informants felt that, in general, the negative aspect of bilingualism is that it is a covert form of student selection and discrimination, since it marginalises those who cannot function in bilingualism often because they come from disadvantaged social contexts. Many informants explained that they had noticed that the students with the best results (in terms of knowledge of the subject and competence in English), usually have some English language academic support outside the high school, such as private English lessons in private language schools. The main concern is with those families that cannot afford those lessons. In addition, students, increasingly, come to high school with a higher level of English certified by official and authorized bodies; so that, when going up to higher forms, the level difference between those who do not study or progressively pass official language tests during the following courses of the ESO (Educación Secundaria Obligatoria, compulsory secondary education in Spain) becomes evident. Some also maintained that the positive segregation of bilingual students is a problem for the life of the school. 8. Competence in the mother tongue: There are many students who have serious communication and understanding difficulties in Spanish, which is why the situation is worse in English. Students from other countries who have to learn two languages at the same time in a context of little immersion in the vehicular language are also incorporated into the high school. 9. Lack of resources for students with special needs: There is no teaching staff specialized in bilingual therapeutic pedagogy in most bilingual high schools so the students who need this educational support do not take part in bilingual subjects. The solution sometimes is for these students to study only contents in Spanish. Some informants explained that in a classroom where both the content subject teacher and the language assistant speak English, the teacher's difficult task is for the special need students not to feel excluded. 10. Lack of methodological training and competence in English: Many informants complained that many teachers consider themselves capable of teaching 'through CLIL' when what they really do is to teach in another language. Therefore, more CLIL training is needed. Besides, since in many regions in Spain the minimum requirement is a B2 certificate, many teachers do not bother to improve their level to a C1 or C2 and there is no linguistic update.