ELF in the Iranian education system: Exploring teacher trainers’ and pre-service/in-service English teachers’ attitudes

The present study was aimed at exploring English teacher trainers’ and preservice and in-service English teachers’ attitudes toward English as a lingua franca (ELF) in the Iranian education system. To amass the data, 68 pre-service and 118 inservice English teachers and 21 teacher trainers filled out an adapted ELF questionnaire and sat semi-structured interviews. The results of the study revealed that the participants expressed diverse attitudes to different aspects of ELF. However, generally, they showed further inclination toward native speakerismalthough some traces of flexibility were observed in their attitudes toward the mainstream English language teaching (ELT). Further, the results of one-way ANOVA showed that overall there was no significant difference between the three groups’ attitudes toward ELF. Moreover, the results of Kruskal–Wallis Test also demonstrated no significant difference between the three groups’ attitudes to each aspect of ELF. The findings of the study imply that the practitioners’ fixed mindsets cannot change unless a significant change occurs in the education systemof the country in general and teacher training programs in particular. Subjects: Teachers & Teacher Education; Continuing Professional Development; Education Policy & Politics

ABOUT THE AUTHORS Ahmad Asakereh is a Ph.D. candidate in Applied Linguistics at Razi university, Kermanshah, Iran. His areas of interest include teaching English language skills, English as an international language, English as a lingua franca, and critical pedagogy. He is the author of some articles in his areas of interest and has presented both in and outside the country.
NouroddinYousofi assistant professor in Applied Linguistics/TEFL at Razi University, Kermanshah, Iran. His main research interests include Language teaching and Translation. He has published and presented papers locally and internationally.
Hiwa Weisi, assistant professor in Applied Linguistics at Razi University of Kermanshah. His main research interests include Sociolinguistics, Discourse Analysis and research engagement in higher education. He presented and published many papers in his areas of interest in national and international conferences and journals.

PUBLIC INTEREST STATEMENT
English is no longer considered a foreign language used by non-native speakers to communicate with only native speakers. In today's world, individuals with diverse lingua-cultural backgrounds across the world use it to communicate with each other. Therefore, exerting emphasis only on normative and fixed rules can be problematic in such multilingual and multicultural settings. However, many educational systems around the world disregard the current status of English in today's world and consider the norms common among native speakers of English as the point of reference in English language teaching. To highlight this issue in the Iranian education system, the researchers examined Iranian practitioners' attitudes toward English as a lingua franca (ELF). The results revealed that the participants were more or less in favor of the normative rules common among native speakers of English.

Introduction
Around 380 million people use English as their first language while over a billion people use it as an additional or a second language to communicate mainly with people coming from diverse linguistic and cultural backgrounds. Consequently, those who speak English as their first language are in the minority (Clyne & Sharifian, 2008). Demographically and geographically, English as an international lingua franca is a unique one, as to date no other lingua franca has been used as widely as English across the world (Jenkins, 2018). Baker (2012) holds that as English is commonly used by non-native English speakers as a main means of communication in diverse contexts, it is of paramount importance for language learners and teachers to pay close attention to diverse contexts in which English is used as a means of communication. In such multilingual and multicultural settings, it is therefore neither sufficient nor authentic to consider only particular linguistic and cultural elements common in English-speaking communities, while ignoring other contexts in which individuals with diverse linguacultural backgrounds are engaged in communication using English.
To incorporate English as a lingua franca (ELF) in English classes, English teachers need to reexamine and rethink the status quo by focusing on the sociolinguistic and sociocultural reality of the English language, getting further exposure to various English varieties used across the world, engaging in class discussion on globalization, focusing on intelligibility rather than accuracy and exerting emphasis on communicative strategies (Dewey, 2012). This hence requires practitioners to shift their attention from the fixed model of native speakerism to a further flexible one. Teacher training programs, however, are still in favor of Anglophone model of English language teaching (ELT). Dewey (2015) believes that teacher training programs persistently perpetuate the premise that English language rules are fixed and are not subject to any change. This implies that English language teachers are obliged to follow monolithic lingua-cultural norms practiced by those speaking English as their first language. However, the common default approach to ELT stands in contrast with the current status of the English language. The current study, thus, put the spotlight on the teacher training programs in Iran, a neglected context, to provide further insights into the status of ELF in the Iranian education system by exploring teacher trainers' and preservice/in-service teachers' attitudes toward ELF.
A bulk of studies (e.g. Coskun, 2011;He & Zhang, 2010;Khatib & Monfared, 2017;Moradkhani & Asakereh, 2018;Sadeghi & Richards, 2015;Tajeddin, Alemi, & Pashmforoosh, 2017) has been conducted on teachers' attitudes toward accent and culture in light of EIL paradigm in the Iranian and non-Iranian contexts, the results of which showed more or less a normative pedagogical model was considered desirable by the participants of the study. However, attitudinal studies on ELF in Outer-Circle contexts especially in the Iranian context seem to require further attention. Moreover, it seems the preceding attitudinal studies on ELF had a narrow approach to ELF as they focused merely on one aspect of ELF, for example, linguistic or cultural aspect and disregarded other aspects. In the present study thus attempts were made to have a more in-depth investigation into different aspects of ELF in the Iranian education system by considering not only in-service English teachers' attitudes toward ELF but also those of pre-service teachers and teacher trainers. Basically, teacher trainers play a significant role in raising pre/in-service teachers' awareness of ELF (Vettorel, 2015). It can thus be illuminating to ascertain to what extent pre/in-service teachers and teacher trainers see eye to eye about ELF, as divergent views can impinge upon the implementation of ELF in teacher training programs. Additionally, the effect of teacher training programs on pre-service teachers' mindsets can come into light by exploring their attitudes toward and awareness of ELF after years of studying at the Iranian teacher training universities. Moreover, exploring in-service teachers' attitudes toward ELF can also highlight the extent the contexts in which they are teaching can affect their attitudes toward ELF. Borg (2011) maintains that teachers' professional development is affected by the level of their familiarity with new theories and trends in language teaching profession. On the other hand, teachers are considered as active decision makers who are largely influenced by their attitudes, previous experience, and knowledge. It is therefore believed that if a new policy or a paradigm is implemented without taking teachers' attitudes into account, it is very likely to lead to failure. Therefore, teachers' attitudes toward ELF,

Linguistic elements and pedagogical model in ELF
In ELF settings interaction between individuals is not based on fixed sociolinguistic rules; rather, individuals are engaged in code-switching, paraphrasing and repetition to make their message intelligible (Dewey, 2014). In ELF contexts, language learners should be exposed to diverse discourses in which non-native speakers use English as a means of communication, and mutual intelligibility should be given special attention (Sifakis, 2004). The variation of ELF contexts makes it extremely difficult to predict what cultural and linguistic elements one will encounter in ELF interactions. However, English teachers need to encourage language learners to focus on intelligibility in various contexts (Hynninen & Solin, 2018). Smith and Nelson (1985) define intelligibility as recognition of a/an word or utterance, and draw a distinction between intelligibility and comprehensibility which refers to the meaning of a/an word/utterance. Deterding and Gardiner (2018) also assert that not all the pronunciation features are necessary for the intelligibility of speech in ELF contexts. Some phonological features including the use of weak forms and connected speech (i.e. elision, assimilation) common in normative rules used by native speakers cause problems for intelligibility in ELF interactions. There are some further features playing no significant role in intelligibility in ELF communication. The replacement of constant sounds such as "T" and "D" by "S" and "Z", word stress, and intonation to name a few are among those features presenting no obstacles to intelligibility in ELF contexts (Jenkins, 2009). On the other hand, second language learning theories, which have been challenged vehemently since the advent of ELF, capitalize on the importance of achieving native-like proficiency.
2.3. Ownership of the English language and the role of English speakers' mother tongue in ELF Bayyurt (2017) points out that an international language like English used by individuals with diverse lingua-cultural backgrounds is no longer owned exclusively by those who speak English as their first language. Crystal (2003) also asserts that one of the consequences of a global language like English is that either no one can be considered as its owner or all its users can be regarded as their owners. In the same vein, Marlina and Xu (2018) also maintain that everyone speaking the language owns it. Generally, in ELF contexts, native speakers constitute only a small number of English language users, which has led to the rejection of the traditional conception that English speakers need to conform to the normative models of native speakers (Huang, 2018).
In the mainstream ELT, language learners' English is considered corrupt and deficit, and language learners need to strive to eliminate all traces of their first language to get closer to native speakers of English, something which is frowned upon in ELF research (Ranta, 2018). Baker (2018) points out that in ELF communication, pragmatic strategies such as code-switching, repetition, accommodation are not viewed as linguistic deficiencies; rather, it is considered as successful communication between multilingual speakers. In ELF interactions, items and linguistic elements from English speakers' first language sharing the same L1 can thus be observed. However, this does not lead to communication breakdown as they adopt different strategies such as negotiating and pre-emptive strategies to resolve any possible miscommunication (Cogo & House, 2018).

The role of culture in ELF
ELF goes beyond the ideology that the English language and its culture are fixed and immutable (Dewey, 2014). McKay (2000) points out that in the contexts of ELF where English is used by people with diverse lingua-cultural backgrounds, the focus of instruction should be on multiculturalism rather than biculturalism. As ELF is associated with multicultural and multilingual contexts, providing a connection between a specific language and culture in such settings can be considered naïve and unrealistic (Baker, 2009). ELF users are not supposed to be familiar with all diverse cultures in ELF communication rather they need to be aware of the dynamic nature of ELF communication. However, knowledge of a specific culture and differences between cultures can also be of paramount importance. ELF is more compatible with the contemporary and poststructuralist views of culture, and ELF is not considered culturally neutral. It is a fallacy to consider intercultural communication between people to be a neutral social practice as in intercultural communication, cultural and meaning negotiation are rampant. However, this does not mean that there is a fixed culture associated to ELF; rather, in ELF, English users may keep moving across and between local, national, and global cultures in a more dynamic fashion (Baker, 2018).

Communication goal in ELF
The main concern of communicative competence model in the mainstream ELT, proposed by Dell Hymes, is communication based on the normative rules common in a particular speech community, namely, native speakers of English rather than communication between English users from around the world (Seidlhofer & Widdowson, 2017). English in Expanding Circle countries (e.g. Iran, Korea, China, etc.) is no longer a foreign language as people in these countries use English to communicate with non-English speakers rather than native speakers for various purposes. Basically, ELF is defined as a means of communication used by people whose first languages are different. Even though the majority of ELF scholars approve this definition, some (e.g. House, 2003) exclude native speakers of English from ELF settings, while the majority of scholars (e.g. Jenkins, 2006) believe native speakers are part of this scenario and cannot be excluded, something with which the researchers of the current study are in agreement. However, they believe that nonnative speakers no longer need to conform to the norms common among native speakers of English when communicating with them. Hence, agreement or disagreement on the inclusion of native speakers of English in ELF settings was a main factor diverging ELF scholars' views on the notion of ELF.

The significance of teacher training programs and teachers' awareness of and attitudes toward ELF
No change can be observed in ELT classes as long as English language teachers are left to their devices. Teacher training programs need to nurture critical teachers and raise their awareness of sociolinguistic reality of the English language (Dewey, 2012). Vettorel (2015) believes that English language teachers need to prepare language learners to accept and get accustomed to the English varieties, something which entails modification in the mindsets of both teachers and students. This responsibility partly rests on the shoulder of teacher trainers to raise pre-service and in-service teachers' awareness of ELF and to encourage them to think critically about the Anglophone models adopted for ELT. ELF necessitates redefining the traditional definition of ideal language teachers and language learners who put a great deal of efforts into acquiring cultural and linguistic norms practiced by native speakers of English (Blair, 2015). Marlina (2014) believes that the purpose of ELF pedagogy is raising students' awareness of different English varieties, encouraging students to appreciate all varieties of English and showing respect to different cultures. However, in order to explore the implementation of ELF pedagogy in the classroom, teachers' critical attitudes need to be taken into consideration (Bayyurt & Sifakis, 2015), as disregarding teachers' attitudes can lead to failure to apply any educational policies. Even though teachers can not apply any significant changes in a top-down curriculum conforming to the monolingualism, they can raise students' awareness of English varieties with the aim of helping them make an informed decision about the existing English varieties (Wang, 2015). Dewey (2015) also points out that it is of utmost importance to explore teachers' awareness of ELF and their attitudes toward ELF. Basically, implementing ELF without considering teachers' awareness and attitudes toward it can result in little change in practice (Dewey, 2012;Wen, 2012). In order to contribute to the shift of paradigm from native speakerism to ELF in practice, critical reflection and voices on the implementation of ELF from practitioners should be examined thoroughly (Marlina, 2014).

Studies on English teachers' attitudes toward ELF
Regarding the significance of the implementation of ELF in English classes, a number of studies (e.g. Bayyurt & Sifakis, 2015;Deniz, Özkan, & Bayyurt, 2016;Vettorel, 2016) have been conducted to examine the impact of teacher training programs on teachers' attitudes toward ELF. Further, the existing literature also provides attitudinal studies, which are more relevant to the current study. Some studies (e.g. Ren, Chen, & Lin, 2016) addressed language learners' attitudes toward ELF. However, as the current study deals with English language teachers' attitudes toward ELF, the spotlight is thus put on studies dealing with English teachers' attitudes to ELF. İnceçay and Akyel (2014), for example, explored the attitudes of 100 Turkish English teachers and 10 teacher educators toward ELF. The findings revealed that more than half of them were in favor of normative model of ELT. They considered ELF as a deviation from the Standard English in that they believed that it would cause confusion in international communication. They further believed that not only cultural elements of English-speaking countries but also Turkish culture should be included in English classes as Turkish culture can motivate English language learners and make learning more interesting for Turkish language learners. Although interviews and a questionnaire were adopted to amass the data, the interviews addressed merely the participants' knowledge of ELF. This resulted in insufficient excerpts from the interviews and lack of in-depth presentation of the findings. Moreover, Curran and Chern (2017) conducted a study in Taiwan to examine four groups of pre-service English language teachers' attitudes toward ELF (i.e. graduates, students majoring and minoring in English and interns). The results of the study indicated that the interns and English majors showed a further tendency toward native-like proficiency, while the students minoring in the English language were more in favor of nonnative English varieties. As the study was quantitative, it did not provide an in-depth explanation for the participants' attitudes toward the principles of ELF. The authors thus relied on pure conjecture to explain the reasons behind the participants' attitudes. Further, the only attitudinal study on ELF conducted in the Iranian context was carried out by Barzegar Rahatlou, Fazilatfar, and Allami (2018). In this study, in-service English teachers' attitudes toward some issues discussed in ELF such as their views of their accent and the ownership of the English language, and their attitudes toward their own English were examined. Although the results of the study showed that the participants did not possess straightforward attitudes toward ELF, they tended to be in favor of normative English. Despite the fact that the study can be considered as the first attitudinal study on ELF in the Iranian context, the participants were recruited only from private language sectors in three cities, and public school teachers were left out of the study. Moreover, in the study, some items of the questionnaire explored the participants' attitudes toward their own accent, which may not represent their attitudes toward non-native accents in general and ELF in particular as some teachers may believe that they possess native-like accents; therefore, they do not disclose their real attitudes toward ELF in general and non-native accents in particular. Furthermore, in the legitimacy of non-native English section of the questionnaire, the focus was only on accent, which may misrepresent the broad focus of ELF.
Following a thorough study of the literature, it was revealed that although some studies (e.g. Barzegar Rahatlou et al., 2018;Bayyurt & Sifakis, 2015;İnceçay & Akyel, 2014) have addressed English teachers' attitudes toward ELF across the world, as far as the researchers are concerned, insufficient research has been conducted in the Iranian context. Moreover, to the best of the researchers' knowledge, in both the Iranian and non-Iranian contexts, few research studies, if any, have compared teacher trainers' and pre-/in-service teachers' attitudes toward ELF. The noted gaps in the literature thus motivated the conduct of the study with the aim of providing further impetus for the implementation of ELF in the Iranian education context.

Background to the context of the study
Prior to the Islamic revolution in 1979, Iran had a strong relationship with England and America. The boom in the English language in the country was partly due to their strong bond with the Englishspeaking countries, which resulted in the establishment of two principal English centers, namely British Council and Iran-American society. The movement toward internationalization was also another driving force for the widespread focus on the English language. However, with the advent of the Islamic revolution, all traces of western cultures were removed from the ELT, while the traditional teaching methodology more or less remained the same. With the advent of communicative approach, some modification in the curriculum was required to meet the needs of the students in today's world. However, policymakers seem to be ambivalent about the role of English language in the education system of the country. On the one hand, they exert emphasis on the role of English and on the other hand they do not provide any strong policy for English teaching in the education system of the country (Aghagolzadeh & Davari, 2017). Notwithstanding Islamic republic of Iran's deteriorating relationship with English-speaking countries especially the United States following Islamic revolution in 1979, English is flourishing in the country. Its significant role in international communication has been appreciated even by the leader of the Islamic Republic of Iran, Ayatollah Ali Khaminei, whose quote is presented in the preface of the newly-designed ELT textbook series. In 2012, policymakers decided to make some significant changes in ELT textbooks to encourage language learners to use English communicatively. More specifically, the locally designed ELT textbooks focus on the four macro language skills and are heavily culture-bound, which eliminate almost all cultural elements practiced in English-speaking countries. Apparently, this can be considered a step toward ELF paradigm, which frowns upon the dominance of cultural elements practiced in English-speaking communities. However, due to the lack of trained teachers to implement such policies in the classroom, the new policy has been doomed to failure. Allocating limited time (2 to 3 hours weekly) in the curriculum to teaching English and traditional testing system which still follows the old system of testing have compounded the problem (Davari & Aghagolzadeh, 2015).

Participants
A total of 186 pre-service and in-service public school English language teachers (i.e. 68 pre-service (23 female and 45 male) and 118 in-service (76 female and 42 male)) was recruited from across the country based on a cluster random sampling. Pre-service teachers' and in-service teachers' age ranged from 20 to 25 (M = 22) and from 23 to 50 (M = 41), respectively. In-service teachers with teaching experiences ranging from 1 to 30 (M = 15) were teaching at junior and senior high schools across the country. Given their academic degrees, 2 had Associate of Arts degrees, 53 were holders of Bachelor degrees, 57 had master degrees, and 6 were holders of Ph.D. degrees. Regarding their majors, 96 majored in English teaching, 10 in English Translation, 5 in English Literature, and 7 in Linguistics. From among the in-service -teachers, only 58 (49%) had a degree from teacher training universities while 60 (51%) did not study at any teacher training universities.
In addition, pre-service English language teachers were prospective teachers engaged in the third/fourth-year teacher training programs in Farhangiyan universities (i.e. state teacher training universities) to achieve a BA degree in English language teaching. Moreover, 21 teacher trainers (i.e. 15 males and 6 females) whose age ranged from 38 to 57 (M = 45) and who were teaching at Farhangiyan universities with teaching experience ranging from 5 to 36 (M = 13) were recruited for the present study. Given their academic degrees, 18 were Ph.D. holders and 3 had Master degrees. With respect to their majors, 15 majored in English teaching, 3 in Linguistics, 1 in English Translation, and 2 in English Literature.

Instruments
The design of the study was a sequential complementarity mixed-method, in which qualitative and quantitative data collection methods were adopted to explore different layers of the areas under investigation (Riazi & Candlin, 2014). In order to have an insight into the participants' attitudes toward the area under study, the qualitative and quantitative data were mixed at the level of data analysis. In essence, quantitative studies on teachers' attitudes can be consolidated by qualitative methods to provide more exhaustive findings (Borg, 2012).
To investigate pre-service and in-service teachers' and teacher trainers' attitudes toward ELF principles, Curran and Chern's (2017) ELF questionnaire was adapted. However, as the original questionnaire did not cover different aspects of ELF and some items were not appropriate for the Iranian cortexes, some changes were applied to the questionnaire. The current 30-item questionnaire included items revolving around ELF-related concepts such as accent and pedagogical models (items 1 to7), communication goals (items 8 to 9), ownership of the English language (items 10 to 12), culture in ELT (items 13 to16), the role of students' mother tongue in ELT (items 17 to 20), native-like structures (items 21 to 24), English idiomatic expressions and collocations (items 25 to 28), and ELF-awareness (items 29 to 30) (e.g. Canagarajah, 2006;Jenkins, 2002;Vettorel & Lopriore, 2013;Seidlhofer, 2001;Syrbe & Rose, 2016). Moreover, the questionnaire was based on a five-Likert scale ranging from strongly disagree (1) to strongly agree (5), which agreement or disagreement with the items on the part of the participants revealed the extent they conformed to ELF principles. More specifically, their agreement with items 1, 2, 3, 5, 6,8,11,15,17,20,22,23,24,25,26,27,28 indicated their disinclination to the implementation of ELF in the ELT classroom while their agreement with items 4, 7,9,10,12,13,14,16,18,19, and 21 revealed their tendency toward the implementation of ELF in the ELT classes. In addition, items 29 and 30 examined the participants' awareness of the notion of ELF. Prior to the study, the items of the questionnaire were confirmed by two experts in the field who were well-versed in ELF-related concepts and the questionnaire was piloted with 50 samples from the same population. Based on the feedback received from the participants in the pilot study and the experts, some repetitive, irrelevant and vague items were removed from the questionnaire. More specifically, the questionnaire was modified in two stages. The first stage, in which irrelevant and repetitive items were omitted, was carried out prior to its polite-study with the participants. For example, items dealing with English varieties common only in Outer-Circle countries (e.g. I think different English varieties like Indian English, Singaporean English, and South African English are as acceptable as the English varieties used in English-speaking countries) and pragmatics (e.g. students should follow the way of greeting, requesting, inviting, and refusing used by native people in English-speaking countries.) which are mainly discussed in EIL paradigm rather than ELF paradigm were omitted from the questionnaire. Furthermore, in this stage repetitive items (e.g. English textbooks for Iranian students should present only cultural elements of English-speaking countries (England, America, Australia, etc.); Customs practiced by English-speaking countries such as Thanksgiving and Halloween should be part of language learning for Iranian students.) were merged into one item. The second stage, the focus of which was mainly on the clarity of the item, took place after the pilot study. In this stage, attempts were made to provide vivid examples for some items which some participants found unclear (e.g. students' grammatical mistakes should not be corrected as long as they are intelligible to others (e.g. dropping "s" from third person singular in simple present tense)).
Following a statistical analysis, its Cronbach's Alpha was estimated to be .99, indicating a high internal consistency. Moreover, a semi-structured interview guide, revealed their rationalization on ELF-related issues, was adopted to have an in-depth understanding of the participants' attitudes toward and awareness of ELF. Prior to the main study, the interview guide was confirmed by two experts in the field. Subsequently, it was piloted with one in-service, one pre-service and one teacher trainer to check its comprehensibility and clarity. Accordingly, some modifications were applied to the interview guide. For example, specialized terms such as Expanding, Outer and Inner Circles, which required detailed background knowledge on the part of the participants, were removed from the interview guide based on their comments in the pilot study. Furthermore, to help the participants provide as much information as possible, the interview was conducted in Persian, the participants' mother tongue. Prior to the pilot study, the interview guide was translated from English into Persian by the first researcher. In addition, in order to check the accuracy of the translated version and to ensure that the original meanings were maintained, backtranslation was carried out by an expert in translation from English into Persian and vice versa, which indicated that there was no significant mismatch between the two versions.

Procedure
After validating the questionnaire, it was administered to the participants online and in person. On the front page of the questionnaires, the instructions were presented clearly in order to diminish any possible misunderstanding. In addition, in the online version, the first researcher's contact information was provided to make any further inquiry on the part of the participants feasible and in the face-to-face administration, the first researcher was present on the day of the questionnaire administration to answer the participants' possible questions about the items. Moreover, the researcher ensured the participants that their personal data and their responses would remain strictly confidential and would be used only for the research purposes. Following the administration of the questionnaire, 20 in-service and pre-service English teachers (i.e. 10 pre-service and 10 in-service) and five teacher trainers, who were willing to participate in the semi-structured interviews sat follow-up interviews. As the participants came from different parts of the country, telephone interviews, in Persian (i.e. official language of the country), were conducted. The first researcher requested informed consent from the participants prior to recording their voices. On average, each interview took around 35 minutes.

Quantitative analysis
Both descriptive and inferential statistics were utilized to analyze the data. To answer the first research question, mean and percentage of the responses to the items of the questionnaire were calculated to ascertain the extent the participants conform to ELF principles. Furthermore, to answer the third research question, following the test of normality, one-way ANOVA was run to see if there was any statistically significant difference between the three groups' attitudes. Furthermore, to compare their attitudes to each section of the questionnaire, Kruskel Wallis was run.

Qualitative analysis
Given the second research question, having amassed the qualitative data through a mobile callrecorder application, the first researcher transcribed them verbatim; subsequently, they were translated from Persian into English and the translated and the original versions were checked by an expert to ensure its accuracy. To analyze the data, a three-stage process of open coding, axial coding, and labeling was followed (Merriam, 2009). More specifically, in the first stage, each segment which was relevant for answering the posed research question was highlighted and was given a specific code. In the second stage, the related codes were combined together and given specific codes. In the final stage, the recurring theme was extracted and labeled accordingly. In order to ensure that the recurring themes reflected the participants' views, they were requested to check them and provide comments on them. Their comments showed that there was no significant discrepancy between the extracted themes and what they maintained in the interviews.

Results and discussion
Research question 1: What are the Iranian pre-service and in-service English language teachers' and teacher trainers' attitudinal responses to ELF-related issues?
The participants' attitudinal responses to ELF-related issues are presented in eight sections below. Generally, their responses indicated that more or less they were in favor of native speakerism. Detailed findings are presented in Tables 1 to 8.
Research question 2: How do Iranian English teacher trainers and pre-service and in-service English language teachers rationalize their attitudes toward ELF-related issues?
To answer the second research question, a set of semi-structured interviews was conducted. The quantitative findings were followed by the qualitative ones immediately to provide readers with the participants' rationalizations for their attitudinal responses.

Accent and pedagogical models
As can be seen from Table 1, item 1 is the only item in this section in which the means for the three groups exceeded 3, representing 52% of the pre-service, 54% of the in-service teachers, and 48% of the teacher trainers either strongly agreed or agreed that educational materials need to follow native-speaker models. Further, the main difference between the means of the pre-/in-service teachers' responses (Ms = 3.61, 3.47, respectively) and those of the teacher trainers (M = 2.42) was observed in item 6. While 68% of the pre-service and 65% of the inservice teachers believed that classroom materials should provide a single pedagogical model, either British or American, only 24% of teacher trainers believed so. Item 2 also demonstrated a difference between the means of the pre-service (M = 3.27), the in-service (M = 3.14) teachers' responses and those of the teacher trainers' (M = 2.71). Fifty-two percent and 48% of the preservice and the in-service teachers, respectively, either strongly agreed or agreed that one should sound like a native speaker to be considered a proficient English speaker while only 24% of teacher trainers believed so. This shows further leniency toward English varieties on the part of the teacher trainers. The responses to item 3 indicate that the majority of the three groups were tolerant toward accented English, which was manifested through the disagreement of 56% of the pre-service, 65% of the in-service and 62% of the teacher trainers with item 3. However, item 7 showed that 58% of the pre-service, 61% of the in-service teachers and 52% of the teacher trainers either strongly disagreed or disagreed with the idea that phonological features which do not affect intelligibility can be ignored. Item 5 also represented a difference between the means of the pre-service (M = 3.50) and the in-service (M = 3.57) teachers' responses and those of the teacher trainers' responses (M = 2.61), which was manifested through 59% of the pre-service and 69% of the in-service teachers' agreement with the idea that introducing different English accents can cause confusion in the classroom while only 28% of the teacher trainers agreed with this item. However, item 4 showed that 56% of the in-service teachers agreed with exposing students to various non-native accents while only 37% of the preservice teachers and 24% of the teacher trainers showed their tendency toward this item. This part of the findings indicating more or less tendency toward native-like pedagogical models especially on the part of pre-service and in-service teachers, which is partly in line with the findings of İnceçay and Akyel's (2014) Curran and Chern (2017)   Basically, this tendency toward native-like proficiency can have significant effects on teacher training programs, testing centers, and international publishers. This can even impact upon the employment market of English teachers regardless of their teaching qualifications, which results in English native speakers' advantage over non-native speakers with diverse linguistic and cultural backgrounds (Phillipson, 2016). However, the significant issue is that whether practitioners are fully aware of the possible impact of their tendency toward native-like proficiency on the English language teaching profession in today's world. Increasing their awareness of the consequences of their attitudes toward native-like proficiency can encourage them to reflect on their strict and rigid ideologies about the English language as this misconception can put their future profession as non-native English teachers in jeopardy.
To have an in-depth understanding of their reasons behind their responses to the items of the questionnaire, the three groups' vantage points on native and non-native English were explored thoroughly in a set of semi-structured interviews.
Basically, pre-service teachers believed English spoken by native speakers is superior to English spoken by non-native speakers, and they themselves attempted to observe the norms of native speakers although some believed this may not be achievable. They noted that English spoken by native speakers is marked by fluency, appropriate vocabulary, fewer mistakes, and correct pronunciations, accents, and structures.
Pre-service 1: Accent and pace of delivery are among factors distinguishing natives from nonnatives. When we see those around us like our professors teaching English, it is clear that they are not natives, the vocabulary and pet phrases they use are not English, they cannot be found in movies and lectures, it is clear most of them think in Persian when speaking English.
In-service teachers were also in favor of English spoken by native speakers, and they strived to follow the norms and rules common among natives as they found them Standard. They believed mainly phonological and linguistic features distinguish native speakers from their non-native counterparts.
In-service 1: Intonation is one of the issues distinguishing native speakers from non-native speakers. Pronunciation and word stress usually are not observed by non-native speakers. Structures and vocabulary used by native speakers are different from those used by non-native speakers. I like English spoken by native speakers more because it does not contain mistakes. We try to become native-like but we can't. The joy of language is its naturalness which is attractive for teaching.
The teacher trainers believed that accent, fluency, and word choice are among the main factors distinguishing native speakers from their non-native counterparts. They were strongly in favor of English spoken by native speakers. They believed that if students get exposed to the mainstream pedagogical models, they will be more successful in their future communication, and they pointed out that teachers following such models receive more respect from their students.
Teacher trainer 1: If I want to employ an English teacher, I will definitely try to employ someone whose language proficiency is at least close to native speakers because if students get exposed to the original language, they will definitely be more comfortable in their future communication.
Having said that, although some pre-service teachers had negative attitudes toward non-native accents as they considered them imperfect and unintelligible, the majority of the pre-service teachers revealed that they had no problem with non-native accents; however, they attempted to follow American accent rather than British accent, which they found more difficult. They maintained that if students invest their efforts in other aspects of the English language such as vocabulary, grammar, collocations, and so on, it can be more fruitful.
Pre-service 2: Previously, I highly preferred American accent. Because of difficulty in pronouncing some words in British accent, I did not prefer British accent. But after studying at academic level I came to realize that accent is not important. If your speaking enjoys fluency and accuracy, it is enough.
The in-service teachers had positive attitudes toward intelligible non-native accents. They indicated that one should not get obsessed by accents and should focus his/her attention on the other aspects of the English language. However, they noted that they experience difficulty with some non-native accents such as Indian English. Although they were aware of variations in English accents across the world, they deluded to their inclination toward native-like accents.
In-service 2: Whatever accents except for Indian accent, it is not intelligible for me whatsoever. But native-like or non-native accents are not much important to me. I just cannot comprehend Indian accent. As a teacher, you should not be obsessed by the idea whether one is using a native-like accent.
Interestingly, even though the majority of the participants clearly noted that they approved of accented English, they preferred to sound like a native speaker. This shows that it is not enough to focus only on whether teachers approve of non-native accents. Researchers need to go beyond this and explore teachers' deep ideology to have a more comprehensive picture of their real personal inclination toward native speakerism.

Communication goals in ELF
Given Table 2, dealing with communication goals in ELF, the highest mean for item 8 was related to in-service teachers' responses, representing 70% of them agreed or strongly agreed that an English teaching program for the Iranian students should prepare them to communicate mainly with those speaking English as their first language, while 40% of the pre-service teachers and 38% of the teacher trainers agreed or strongly agreed with the item. However, further disagreement from 53% of the pre-service teachers and 52% of the teacher trainers with item 9, addressing communication mainly with non-native speakers, was reported.
In the interviews, pre-service teachers provided further information by explaining that the main goal of language teaching is communication with both native speakers and non-native speakers, no differentiation should be considered in this case. However, they asserted that if native speakers are considered as the main target group, students will encounter no problem in their future communication with non-native speakers.
Pre-service 3: We should teach English in a way that enables students to communicate with native speakers and non native speakers. Those who can communicate with native speakers, they are 100% able to communicate with non-native speakers. It is better we consider native speakers as a target group, so we can have both.
Although they were aware of communication goals in today's world, it seems pre-service teachers believed there would be a magic transition from learning how to communicate with native speakers to learning how to communicate with non-native speakers, which shows they mistakenly considered the two contexts homogeneous while giving the priority to communication with native speakers. Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers The majority of the in-service teachers believed that as English is an international language, native speakers should not be considered as the only target.
In-service 3: as it is selected as an international language, we use it to communicate, no difference between natives and non-natives. We want to have communication with both.
Teacher trainers maintained that the aim of learning English can be as varied as students' objectives. However, they believed ultimately the aim of language learning can be communication with both native speakers and non-native speakers. They maintained that one who is learning English need to have intercultural competence and get familiar with global village.
Teacher trainer 2: The ultimate objective of English language teaching is communication with both groups of natives and non-natives. Consequently, we can have local characteristics but we get familiar with linguistic and cultural characteristics of native speakers. Table 3, addressing the ownership of the English language, demonstrates that the lowest means were allocated to the pre-service and in-service teachers' and teacher trainers' responses to item 11 (i.e. Ms = 2.11, 2.42 and 2.14, respectively) showing that only 13% of the pre-service, 23% of the in-service teachers and 20% of the teacher trainers agreed or strongly agreed that nativespeakers are the owners of the English language. Responses to item 10 indicate that the majority of the three groups (i.e. 60% of the pre-service, 64% of the in-service teachers, and 71% of the teacher trainers) believed that the English language belongs to everyone speaking the language. Further, response to item 12 revealed that 50% of the pre-service, 56% of the in-service, and 57% of the teacher trainers also agreed or strongly agreed that the English language does not belong to anyone.

Ownership of the English language
Even though a few pre-service teachers believed native speakers own the English language, as they believed natives are the only ones having the rights to bring changes to their language, the majority referred to English as a language spoken by many non-native English speakers, who can contribute to its change. They, therefore, believed it has no owner.
However, the results of the interviews with in-service teachers showed that half of them believed that the English language belongs to native speakers in general and to England in particular. They referred to their strong policy and historical hegemony over the world and their English-speaking ancestors. They asserted that the English rules are also derived from native English speakers rather than non-native speakers. On the contrary, almost the other half of the in-service teachers either believed that the English language belongs to all English users or believed no one owns the English language as all countries use the language.
In-service 4: All of those use the language, write books, and make movies manipulate the language, which means one country, person or limited groups are not the owners.
The teacher trainers did not regard native speakers as the absolute owners of the English language, and they exerted emphasis mainly on English language users rather than the owners of the language as they believed that all language users, be it natives or non-natives, can bring changes to the language.
Teacher trainer 3: We should put the traditional view aside. The owners of the language are the users. If we look at it from this angle, there are many English varieties. Nonnatives outnumber native speakers. If we look at it from this angle, non-natives

3.52
Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers are the owners. On the other hand, generally and intrinsically, native speakers are the owners. I cannot say for sure who the owners of the language are.

Culture in ELT
With respect to Table 4, dealing with the role of culture in ELT, the three groups of the participants' responses to the related items showed that they gave special attention to local, target, and non-English cultural elements in ELT. However, the means of their responses showed slight differences in their preferences for the aforementioned cultural elements. The highest mean (M = 4.32) of the pre-service teachers' responses, for example, was given to item 15, representing the importance of getting familiar with cultural elements of native speakers. In addition, the highest mean (4.32) of the in-service teachers' responses (94%) was presented in item 13, emphasizing the importance of local culture. Interestingly, none of the teacher trainers disagreed with items 13, 14, 15, and 16, showing their inclination toward all cultural elements common in the Kachru's three Circles.
The pre-service teachers expressed diverse views with respect to the inclusion of culture in ELT. Generally, they believed culture and language are closely interwoven and exerted emphasis on cultures common in English-speaking communities. Further, some believed local culture can also be included in English teaching as this can make language learning more interesting and relevant; however, they emphasized that culture of non-English-speaking courtiers should be considered to a lesser extent as students are learning the English language and they themselves are Iranians.
Pre-service 4: If language learning is combined with culture, students can learn language easily.
Local culture and target culture should be considered. Getting familiar with cultures of non-English-speaking countries can be helpful. But we are learning English and we are Persian. We are in contact with these two cultures.
The in-service teachers also believed that culture should not be divorced from language as they are closely interwoven. However, they emphasized that cultural elements which are against the morals and norms of the society should be left out. The majority also highlighted cultures common in local and non-English-speaking countries as they believed learning about different cultures can make language learning more interesting. In addition, they asserted that students may need to talk about their own cultures when engaged in international communications.
In-service 5: Culture should be included in language teaching but I believe cultures of different countries should be included when you teach the English language as an international language; cultures should be international as well.
The teacher trainers also strongly believed that target culture should not be excluded from the language, and they gave the priority to the cultural elements of native speakers. They believed if cultural elements of native speakers are ignored, language learners may encounter significant problems with their communication.
Teacher trainer 4: The culture of the second language should be taught. Culture is sometimes considered the fifth skill. I consider Phillipson's views on linguistic imperialism extreme because today we believe in intercultural competence. If we ignore the target culture, this competence will have no meaning.
They also added that English teachers should not confine students only to cultural elements of native English speakers, as students may need to get engaged in international communication, where it is necessary to get familiar with diverse cultural elements. Sharifian (2016) holds that the notion of intercultural communication in ELF is different from the one common in the mainstream ELT in which cultural miscommunication between native speakers It is important that students be able to use English to talk about their own culture and customs. 14 It is important that English teachers help students to better understand cultures and customs of people from other countries with whom students are likely to communicate. It is important to raise students' awareness of cultural differences and encourage them to talk about such differences. Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers and non-native speakers is emphasized. Exposing language learners to various cultural elements, which are not against the moral codes of the community, can provide further opportunities for students to learn from the cultures. Kumaravadivelu (2012) also emphasized that in 21 st century language learners need to have cultural liberty, which means that they should not only learn about diverse cultures but also learn from them. However, Baker (2009) believes in ELF contexts, it is not sufficient to acquire knowledge about a particular culture. Basically, English language users need to develop their ability in negotiation, something which cannot happen unless teachers abandon their fixed ideology about the role of native speakers' cultures in ELT. Table 5 demonstrates that although the three groups of the participants gave a special role to students' L1 in English classes, the in-service teachers and teacher trainers in comparison to pre-service English teachers were more lenient when it came to the use of students' L1 in English classes. More specifically, 54% of the in-service and 57% of the teacher trainers disagreed with item 17, addressing avoidance of students' mother tongue in the classroom, while 38% of the pre-service disagreed with the item. Regarding code-switching (item 18), 41% of the pre-service teachers, 66% of the in-service teachers, and 48% of the teacher trainers showed a tendency toward code-switching in English classes. The use of literal translation by students addressed in item 19 was also supported by 38% of the pre-service teachers, 57% of the in-service teachers, and 52% of the teacher trainers. However, the majority of participants (i.e. 54% of the pre-service teachers, 53% of the in-service, and 52% of the teacher trainers) believed ultimately students should be discouraged from using their L1 in English classes.

Students' mother tongue in ELT
In the interviews, the pre-service teachers also were not against using students' mother tongue in the classroom and some believed it can have a positive impact on students' learning. However, they maintained that it should not be used excessively in English classes and should be adopted as little as possible by the time students reach high levels of language proficiency. They further asserted that if students switch to their mother tongue, they attempt to provide them with English equivalents to help them with learning English equivalents.
Pre-service 5: I think limited use of students' mother tongue; students should not get accustomed to hearing the Persian equivalents of every English sentence. I think it is necessary to some extent which means very limited. If a student switch to his mother tongue when performing a task I let him to finish his/her task meanwhile I write down what I have seen and at the end I will bring it to their attention.
The in-service-teachers were also against English-only policy in English classes specifically at elementary levels as they believed this can put students under pressure and make English classes uninteresting. However, generally, they believed the use of students' mother tongue should be decreased by the time students reach advanced levels, and judicious use of students' mother tongue should be in any language teachers' agenda.
In-service 6: In my opinion, students' mother tongue should be used judiciously. It should not be banned altogether. Sometimes students' mother tongue can help them to compare some issues with their mother tongue and learn easily specifically at lower levels.
The teacher trainers also were not totally against the use of students' mother tongue in the classroom; however, they believed cautious should be exercised not to use it excessively, as they believed it is one of the techniques of ELT rather than the only one. They also considered it a facilitator technique in ELT.
Teacher trainer 5: Generally, I introduce projects in which students' mother tongue is used less.
I believe translation is one of the techniques from among thousands of techniques and if translation is used properly as a teaching technique, it can be fruitful. Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers Although the participants appreciated students' mother tongue, they believed its use should be decreased when students reach higher levels of language proficiency. This shows that the participants mainly associated the use of students' mother tongue to their linguistic deficiency. However, non-native speakers may refer to code-switching to represent their identity or to show solidarity with an interlocutor or they may accommodate to the interlocutors with different first languages (Jenkins, Cogo, & Dewey, 2011). Marawu (2018) asserts that code-switching helps language learners to perceive their L1 as an important tool for communication in multilingual settings. In ELF contexts, code-switching is not viewed negatively and it is considered as a way of presenting speakers' identity and expressing solidarity with the interlocutor (Seidhofer, 2005). Table 6 shows that although the majority of the participants (i.e. 41% of the pre-service, 56% of the in-service teachers and 67% of the teacher trainers) believed that students' grammatical mistakes should not be corrected as long as they are intelligible to others, 77% of the preservice, 76% of the in-service teachers and 81% of the teacher trainers agreed or strongly agreed that students should be encouraged to use native-like English structures. When it comes to the level of their tolerance of non-native structures, the lowest mean (M = 2.97) was associated to the in-service teachers' responses to item 23, showing further tolerance of non-native structures in comparison with that of the pre-service teachers' (32%) and that of the teacher trainers' (33%). The same scenario can be seen in item 24 where the means of preservice teachers' and teacher trainers' responses to the item were 3.04 and 3, respectively, while that of in-service teachers' was 2.94, indicating that the majority of the first two groups did not accept non-native structures. This can be attributed to the fact that as in-service teachers are engaged with teaching school students with various language proficiencies, they are more tolerant of non-native structures and have a more realistic understanding of the issue.

Native-like structures
In the interviews, the pre-service teachers also exerted emphasis on native-like structures as they believed that there are some rules in the English language which need to be observed when one wants to speak proper English. Further, they believed progress in language learning will have no meaning if native structures are overlooked. Although some noted that the aim of language learning can determine whether it is necessary to follow native-like structures and indicated that non-native structures can also be comprehensible, they insisted on the native-like structures due to the aforementioned reasons.
Pre-service 6: I think learning these structures are necessary. Do we really learn a language only for the sake of communication? With two or three years of language learning, we can communicate what we mean. In that case making progress will be meaningless. We want something beyond communication.
In the interviews, contrary to what they reported in the questionnaire, in-service teachers believed that native-like structures should be followed specifically at higher levels of language proficiency although some expressed that they were cognizant of the fact that the main aim of language learning is communication. They asserted that using non-native structures can lead to imperfection, incomprehensibility and non-standard language. This discrepancy between the interviews and what they reported in the questionnaire can be attributed to their expectation from language learners at higher levels of language proficiency.
In-service 7: The main aim is communication. Learning native-like structures is better and it is necessary to learn native-like structures but in teaching English language as long as they do not affect communication it is ok but if they learn native-like structures they become perfect otherwise we may not be able to communicate what we mean, and misunderstanding takes place. Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers The teacher trainers believed that native-like structures should be learned and used although they indicated that some non-native structures are intelligible to others. They thought deviation from native-like structures could lead to misunderstanding, and they further indicated that as English teachers, it is necessary to learn native-like structures.
Teacher trainer 6: If we use Persian-like structures when speaking English, it is incorrect although communication may not be negatively affected. But as an English teacher and someone who is going to teach them definitely we should teach the correct structures. These native-like structures can form their linguistic characters.
The participants seem to be under the influence of the mainstream ELT, as they consider deviation from normative rules non-Standard. In many cases, some grammatical features are not crucial for interaction between English users; however, grammar is always associated with Standard English. Its marginal significance in interaction did not make it subordinate in mainstream ELT as using particular grammatical features, especially redundant ones, represents a social identity connected to a particular community and any disregard of these features excludes one from the community. The connotation associated with the notion of Standard can be a fixed and monolithic point of reference. It is believed that any variation therefore can corrupt this notion and put it in jeopardy (Canagarajah, 2007;Widdowson, 1994).

Native-like idiomatic expression and collocation
As can be seen from Table 7, the majority of the participants expressed their agreement (i.e. 75% of the pre-service, 80% of the in-service teachers and 86% of the teacher trainers) to item 25, exerting emphasis on the importance of native-like idiomatic expressions in ELT. Their responses to item 26 show that 35% of the pre-service and 46% of the in-service teachers disagreed or strongly disagreed with the item indicating unacceptability of replacement of one lexical element in an idiom with another lexical element, while the majority of the teacher trainers (43%) agreed with the item. However, 60% of the pre-service, 57% of the in-service teachers, and 53% of the teacher trainers agreed or strongly agreed that non-English idioms which are common in students' mother tongue should be avoided when students speak English.
Their responses to item 28, dealing with collocations, indicating that 66% of the pre-service, 62% of the in-service, and 52% of the teacher trainers agreed that students should avoid using collocations which are not common among those speaking English as their first language even if they are intelligible to others.
Some of the pre-service teachers believed that native-like idiomatic expressions are necessary for daily conversation and without idiomatic expression, our daily speaking sounds bookish and unnatural. In addition, they maintained that including idiomatic expressions in ELT makes English classes more interesting and attractive and makes our speaking even perfect.
Pre-service 7: Of course I think in this way we are supposed to have colloquial conversations not all of us will speak with ministers, ambassadors, and presidents of other countries. We are supposed to have colloquial conversation, things which are spoken on the street.
However, interestingly in the interviews, although the majority of pre-service teachers considered idiomatic expression as a secondary importance for students, they themselves attempted to learn them to reach native-like proficiency. Moreover, they were against using idiomatic expressions common in students' mother tongue.
Pre-service 8: I do not think it is necessary for school students to learn them. It can be included as an additional activity in their textbooks. Now we have some professors teaching us who  Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers seldom use them. This can be associated with the idea whether you want to sound like a native speaker. I myself try to learn them.
Moreover, even though some pre-service teachers considered collocations unnecessary for successful communication, the majority strongly believed that from the aesthetic point of view, learning collocations are necessary. Moreover, they believed that disregarding this can make one's speaking sounds weird, and they can also contribute to learning new vocabulary.
Pre-service 9: I think this is really necessary because disregarding this can ruin the English language.
If this was not necessary, the English language would be negatively affected. Not using collocation may not affect communication, but it will not have the expected beauty.
Basically, the in-service teachers did not consider learning idiomatic expressions as an indispensable part of language learning either as they believed communication can take place without using idiomatic expressions. Having said that, they believed using idioms can sometimes be helpful in communication. They thought initially students need to learn the basics of language before proceeding with learning idiomatic expressions.
In-service 7: I do not find teaching idiomatic expressions necessary for school students and students need to learn the basics of language and if reached a level which they can communicate with others then we can focus on idioms and phrasal verbs because there are many idiomatic expressions and someone who is not able to communicate how can he use them?
However, in the interviews, the in-service teachers unanimously found collocations necessary and considered it more important than idioms in ELT. They believed collocations are closely interwoven with language learning and cannot be overlooked. Further, they emphasized the negative impact of L1 on using correct collocations.
In-service 8: Learning collocations is necessary because if we do not learn collocations, our first language will intervene with the second language and this will affect our communication.
The teacher trainers found idiomatic expression as a means of communication and an indispensable element of language. Further, they believed idiomatic expressions can make language teaching and learning more interesting.
Teacher trainer 7: Idiomatic expressions are the indispensable part of the language and they are attractive to students. Students sometimes expect us to use them and this is the normal aspect of language otherwise it will make our speaking sound bookish.
The teacher trainers also considered collocations an important aspect of language as they provide language learners with correct and contextualized English.
Teacher trainer 8: Collocations provide more opportunity for language learning and as an indispensable part of language; collocation should be learned. I think collocations allow teachers to contextualize the language.
Although there was a discrepancy between what teachers reported in the questionnaires and interviews. In the interviews, they did not reject idiomatic expressions altogether and they thought they should be learned as an extra activity. The justification for their attitudes toward native-like idiomatic expressions can be mainly associated to their priority for learning other linguistic elements prior to learning native-like idiomatic expressions rather than excluding native-like idiomatic expressions in real-life communication in today's world.
Even though teaching idiomatic expressions can be interesting, they are of little use in international communication (Llurda, 2018). Some communication features common among native speakers of English can be unintelligible in ELF and impede communication. Among these features is the use of idiomatic expressions used either by native speakers or non-native speakers which can cause problems in the flow of communication as idioms are culturally loaded expressions. Making use of familiar idioms by both parties can, however, facilitate communication as speakers do not have to process the elements of an idiom bit by bit rather they process it as a whole, which reduces the processing burden. Nevertheless, a slight deviation from the established rules of idioms can represent a speaker as an outsider (Seidlhofer, 2009). Table 8 shows that the majority of the three groups were familiar with the notion of ELF. However, the pre-service teachers' responses to items 29 and 30 indicated the lowest means (Ms = 2.94 and 2.95, respectively) compared to those of the in-service teachers' (Ms = 3.47 and 3.56, respectively) and the teacher trainers' (Ms = 4.09 and 4, respectively), indicating that 41% and 43% of the pre-service were both familiar with ELF and know the difference between ELF and English as a foreign language (EFL), respectively. This can be attributed to the premise that many of the in-service teachers and teacher trainers might become familiar with the notion of ELF in their higher education studies.

Awareness of ELF
Following further inspection of the issue through the interviews, it was revealed that some of the pre-service teachers never heard the notion while others possessed little knowledge about ELF. They indicated that at university, no specific course was allocated to this notion and they maintained that their professors did not exert emphasis on this notion. However, a few noted that they got familiar with it superficially in one of their courses. They thought ELF is a defective version of the English language.
Pre-service 9: When one language is used for communication, is called lingua franca. There was in a course and my professor alluded to it. ESL is a language you acquire as a second language and lingua franca is just for communication, in which you do not need to know the language. You are only familiar with it to the extent that can meet your needs, but ESL means you know the language and you are familiar with its culture.
Half of the in-service teachers were not familiar with the notion whatsoever, and those who indicated their familiarity with the notion had very limited knowledge in this regard. They defined it as a language for international communication, and they emphasized that this notion was not highlighted during their academic studies.
In-service 9: I forgot it. It means the international language right? I do not have any specific information in the regard. I read only its definition and now even I forgot it. It is a language which two parties know and use it to communicate.
Apparently, the teacher trainers were not totally familiar with the notion of ELF, as they thought that it is a specific language common among a group of people and some believed it is a simplified version of language.
Teacher trainer 9: English as a lingua franca is a simplified language if I am not mistaken and it is used for business purposes. Previously it was used between countries. There was a shared business language which was simple in terms of grammar. The aim was that to exchange a commodity. Jenkins et al. (2011) point out that ELF proponents perceive English spoken by non-native speakers as different English rather than deficient and incompetent, while some view ELF as an impoverished code of communication which is common among those who do not have any other  Note: SD = strongly disagree, D = disagree, N = Neutral, SA = strongly agree, A = agree, M = mean, P = pre-service teachers, I = in-service teachers, T = teacher trainers alternative to communicate. However, research on ELF shows that ELF communication is replete with innovations, creativity, and communication strategies.
Sifakis (2017) asserts that English language teachers' awareness of ELF is a continuum ranging from no awareness to complete awareness. Generally, it seems the participants of the current study were mainly at the beginning of this continuum. However, following close examination of the continuum, it was revealed that it does not consider teachers who have false information about ELF, which can have significant effects on their attitudes toward ELF, as it was demonstrated in the present study. Therefore, to provide a more precise continuum of teachers' awareness of ELF, it may need to be revised.
Research question 3: Is there any significant difference between the Iranian pre-service and inservice English language teachers' and teacher trainers' attitudes toward ELF?
The results above showed some differences between the means of the participants' responses. To examine whether the differences are statistically significant, one way ANOVA was run. Table 9 indicates the descrptive statistics for pre-service/in-service teachers' and teacher tariners' attitudes toward ELF, and Table 10 demonstrates that there was no statistically significant difference between the three groups in terms of their attitudes toward ELF, F (2, 204) = .008, p = .99. This indicates that overall, the three groups expressed similar attitudes toward the English language in general and ELF in particular. Teacher trainers engaged in training would-be teachers can have a direct impact on the pr-service teachers' attitudes toward ELF and their awareness of ELF. However, as the findings revealed even teacher trainers were not totally  Teachers know about ELF and they attempt to implement it in the classroom as much as they are allowed Teachers are not aware of ELF at all but they implement it in the classroom unknowingly familiar with the notion, and it seems this might have affected the pre-service teachers' attitudes toward the notion. In-service teachers with various teaching experience were expected to be more in touch with the students' needs; however, they were also affected by the current education system of the country, in which the sociolinguistic reality of the English language is neglected in the teacher training programs.
The researchers of the present study did not stop there and proceeded with the data analysis to compare the three groups of the participants' attitudes toward each section of the questionnaire by running Kruskal-Wallis Test. As Table 11 indicates, no statistically significant difference was observed in the three groups' attitudes toward the individual sections of the questionnaire. That is, the pre-service teachers', in-service teachers' and teacher trainers' attitudes toward each section of the questionnaire did not diverge significantly.

Conclusion
The findings of the study revealed that the participants provided diverse attitudes toward ELF. However, more or less they tended to be in favor of native speakers' model. Jenkins (2000) maintains that when it comes to pronunciation and accent, it is an arduous task for practitioners and language learners to change their mindsets about so-called Standard accents. She believes one way out of this restricted attitude toward teaching pronunciation can be through raising teachers' awareness of the reality of the English language. Surprisingly, the findings of the interviews indicated that the participants were aware of the international role of English in today's world to some extent; however, they still inclined toward native speakerism. As Cavalheiro (2016) explained, English language teachers blindly follow the pedagogical model conforming to native speakers of English without questioning them despite the fact that teachers are aware of the international role of English in today's world. Cavalheiro believes that this is not due to unawareness, basically, they do not know how to implement some principles in the classroom to demonstrate its main sociolinguistic status in today's world. However, the findings of the study showed that the participants mainly associated normative rules with perfection, correctness, and Standard English; moreover, they considered English spoken by natives superior. This implies that accepting the sociolinguistic reality of the English language cannot lead to any significant changes in practitioners' ideologies. When the education system of the country and the teacher training programs accentuate the normative rules, no significant changes can be observed in such a context. ELF-aware teacher education can have a set of benefits for teachers such as raising their awareness of the reality of the English language, making them believe they are the users of language rather than the learners, encouraging them consider the needs of their own classes, and get engaged in critical reflection. However, such a program can contain some challenges such as teachers' resistance to change, stakeholders' expectations from teachers, and textbooks and school curricula (Sifakis & Bayyurt, 2018).
It should be noted that adopting ELF-aware approach in teacher training programs does not mean rejecting second language acquisition (SLA) theories; rather, this addition in teacher training programs can broaden teachers' horizons with respect to the status of the English language in today's world and can encourage them to put aside their conservative attitudes toward the English language. More specifically, ELF is not intended to restrict their choices; rather, ELF is about providing further choices and raising their awareness of the use of English in today's world to assist them in taking informed decision on the common normative rules (Cogo, 2012). The integration of ELF in English classes provides a more realistic picture of the English language in today's world. Students will no longer need to be the imitators of the normative rules and hegemonic cultural rules practiced in English-speaking communities (Siqueira, 2018).
Like any other studies, this study also contained some limitations. Although mixed-methods data collection (i.e. questionnaire and interviews) was used, future researchers can adopt classroom observation to examine to what extent the in-service teachers' and teacher trainers' beliefs match up with their practical teaching. Further, as this study only focused on the public sector, future researchers can compare practitioners ' attitudes teaching in both public and private sectors to ascertain to what extent contextual variations can impact upon their attitudes toward ELF. In addition, future researchers can conduct further research in diverse Expanding Circle contexts to illuminate the main impediments preventing practitioners to implement the principles of ELF in such contexts.