Comparing local and international English teacher training courses: Lessons learned

Abstract This article presents an overall exploratory comparison of the English teacher training courses (TTCs) held in Iranian private language institutes and Certificate in Teaching English to Adults (CELTA). Data were gathered through Iranian institutes’ websites; interviews with Iranian English teachers, English teachers holding CELTA certificate, and teacher trainers; as well as questionnaires filled out by institutes’ supervisors and CELTA holders. Content analysis and document analysis were conducted to reveal the similarities and differences between CELTA and Iranian TTCs. Results showed that Iranian TTCs were quite different from CELTA, especially at implementation and evaluation stages. The findings revealed that Iranian TTCs neglected the trainees’ needs, limited the teachers to a series of practical steps in teaching, did not provide enough opportunities for teaching practice, and evaluated the trainees’ performances subjectively. However, they enjoyed systematic planning, contained practical techniques for teaching the language components and skills, and were based on institutes’ needs. Iranian teacher trainers are advised to involve the trainees more in the implementation of the course through doing assignments, to provide the trainees with video recordings of the experienced teachers’ classes for observation, and develop an objective criterion for assessing the trainees’ teaching performances.


PUBLIC INTEREST STATEMENT
There are hundreds of English private language institutes across Iran which hold teacher training courses (TTCs) to train the teacher applicants. However, it is not clear if these institutes follow international standards in teacher training and train successful teachers. Because the Certificate in Teaching English to Adults (CELTA) is internationally recognized and is very popular in Iran, this paper compared Iranian TTCs with CELTA. Results showed that although Iranian TTCs were similar to CELTA in planning, they could not provide the trainees with enough opportunities for practicing teaching and observing successful teachers' classes, and the examiners evaluated the trainees' performances without any objective criteria. However, Iranian participants were satisfied with the convenient schedule, practical teaching techniques, and friendly atmosphere. Like previous studies, these results emphasize that Iranian TTCs need to be adapted for the Iranian context and be based on the trainees' needs rather than following the international courses blindly.

Introduction
Doubtless, teachers play a substantial role in education, so it quite stands to reason that their wellpreparedness would directly influence their students' academic achievements. Teacher preparation and the desperate need for teacher's development have caught the researchers' attention for the last two decades (Hammadou, 2004;Lee, 2007;Richards & Farrell, 2005). Emphasizing the situation-specific nature of English language teaching, Townsend and Bates (2007) argue that it is essential we employ teachers who have the skills and abilities for teaching in a specific situation because having qualified teachers is what all the students around the world deserve. Furthermore, Sandres and Horn (1998, p. 19) rightly state that 'the single most important factor in determining student academic success or failure is the classroom teacher'. In order to keep up-to-date and improve professionally, Ballantyne, Sanderman, and Levy (2008, p. 10) argue that 'there is a pressing need for education for teachers at all stages in their careers, which aims to prepare or upgrade teachers' knowledge and skills'.
There are various ways in which English teachers can improve their teaching skills and develop professionally. Attending TTCs; doing action research; consulting methodology books, journals, and magazines; cooperating and collaborating with colleagues; team teaching and observation; and joining teachers' groups and associations are among the most common ways (Harmer, 2002). Whatever the method of professional development, it is necessary to evaluate EFL teacher training programs in order to improve their impact on teachers' performance (Rea-Dickens & Germaine, 1993). Payne (1994) considers educational evaluation necessary because it greatly helps to improve the program during the development phase, makes rational comparison of competing programs possible, and leads to effective decision making in its own turn.
Due to the drawbacks of Iranian system of English Education at high schools (Maftoon, Yazdani Moghaddam, Gholebostan, & Beh-Afarin, 2010), too many private language institutes have been established across Iran, and these are the English language institutes which play the main role in teaching English in the Iranian context (Mesri, 2009;Sadeghi & Richards, 2016). Private language institutes are mushrooming across the country, and thousands of Iranians take English classes in these institutes for various reasons such as general conversation courses, writing and grammar classes, IELTS and TOEFL preparation courses, and courses in English for Specific Purposes. However, several researchers argued that these institutes were not successful in reaching their goals and meeting the language learners' needs (Mirhosseini & Khodakarami, 2015;Sadeghi & Richards, 2015). Sadeghi and Richards (2015) added that most of Iranian private institutes fail in teaching spoken English due to weaknesses in their curriculum, limitations of the classroom basedlearning, and lack of efficient and competent teachers. To be more precise, they pointed out English teachers in these institutes were not provided with any training regarding the nature of spoken English and the complexity and technicality of teaching a speaking course effectively.
These language institutes usually hold 40-60 hours of TTCs one to four times a year, but very few of them hold these TTCs in order to employ the English teachers they need. The participants of these courses are given a certificate at the end of the course, which in some cases allows them to work as English teachers in these language institutes. However, it seems that there is no unified procedure for holding TTCs in these institutes, and every language institute and teacher trainer run their own TTC. In spite of the fact that all these institutes aim to train English teachers for the Iranian context, their curricula and teaching methodology differ considerably (Rezaee & Ghanbarpour, 2016). Since most of these language institutes do not approve of the TTC certificates issued by other institutes, teacher applicants have to attend several TTCs in different institutes in order to find a job as an English teacher. That such discrepancy of opinion should exist amongst the institutes on a subject of so much importance is greatly to be lamented.
Although TTCs in Iranian private language institutes are held differently (Rezaee & Ghanbarpour, 2016), there is no organization responsible for monitoring and/or evaluating these TTCs. Furthermore, there exists little research in this area describing the procedures in holding these TTCs (Akbari & Yazdanmehr, 2011;Razmjoo & Riazi, 2006), evaluating their impacts on teachers' beliefs (Abasifar & Fotovatnia, 2015), investigating their curriculum (Rezaee & Ghanbarpour, 2016), or comparing them with internationally recognized TTC courses. Considering the aforementioned problems and to fill the said research gap in literature, this study compares the common procedure in holding TTCs held in Iranian private language institutes with CELTA.
CELTA is a pre-service TTC run by Cambridge University. The Cambridge CELTA is the most widely recognized initial qualification in teaching English as a foreign language (TEFL), recognized by language schools in dozens of countries around the world. Over 10,000 candidates gain the qualification each year. The CELTA is, in short, the qualification teacher applicants need to have if they plan to seek employment as an English language teacher in a quality language school and is the most widely recognized TEFL course. It introduces the applicants to the principles of effective teaching, helps them understand more about the English language, presents a range of practical skills for teaching English to adult learners, gives them hands-on teaching practice with real students, and builds their confidence in the classroom. The trainees are assessed on their ability through teaching for 6 hours at two levels of proficiency and their four written assignments. It lasts for 120 hours, and successful candidates receive one of three grades: Fail; Pass; Pass B; Pass A. CELTA covers the following topics: language awareness; the learner, the teacher and the teaching/learning context; planning for effective teaching of adult learners; classroom management and teaching skills; and resources and materials.
In particular, the study aims to compare the most common procedures in holding an English TTC in Iranian institutes with CELTA in terms of planning, implementation, and evaluation stages. The researchers chose CELTA for this purpose since both Iranian and CELTA TTCs are designed to prepare applicants with no teaching experience for teaching English in an EFL context, all Iranian language institutes approve of CELTA certificate, and many Iranian language institutes follow CELTA in holding their TTCs. In order to find out about the similarities and differences between the Iranian TTCs and the said international course, this study addresses the following research question.
What are the similarities and differences between CELTA and TTCs held in Iranian Private Language Institutes in terms of planning, implementation, and evaluation stages?

Participants
The participants of the current study comprised of 6 English teachers, 6 teacher trainers, 37 institute supervisors, and 18 Iranian English teachers holding CELTA certificate. First of all, from among Iranian cities where private language institutes held TTCs lasting for at least 40 hours, three cities were purposefully selected (Isfahan, Kermanshah, and Boushehr). Then, in each city, two language institutes were chosen through the convenience sampling. Finally, from each institute, the researchers invited one teacher trainer and one English teacher for the interview. Thus, six English teachers and six teacher trainers from six different institutes were interviewed in order to provide a clear picture of holding a TTC in the Iranian context. Besides these interviewed participants, 37 institutes' supervisors participated in the research by responding to the openended questionnaire designed by the researchers. These supervisors either taught the TTC in their own institutes or were quite aware of the TTC held in their institute. They worked in different institutes situated in the three cities mentioned before. Table 1 displays the descriptive information about the 6 teacher trainers, 6 English teachers, and 37 supervisors.
On the other hand, in order to collect data regarding the implementation of CELTA, six Iranian English teachers with CELTA certificates were interviewed. These interviewed participants were coded as CELTA holder A (CHA) to CELTA holder F (CHF). Furthermore, a questionnaire with six open-ended questions was designed in order to obtain data from those Iranian English teachers with CELTA certificate who were not ready for the interview. The open-ended questionnaire which included the same questions as the semistructured interview was e-mailed to 12 Iranian English teachers with a CELTA certificate (8 male and 4 female teachers). These CELTA holders were coded as CELTA holders 1-12 (CH1-CH12). They had taken CELTA abroad, more specifically in Turkey, Greece, and Armenia. Table 2 summarizes the background information of the participants who provided data with regard to CELTA.
To meet the ethical requirements of research with human participants, all the participants were asked to sign the consent form, were assured that the interview content would be kept safe and confidential, and were told that the data would be solely used for research purposes. They participated in the research project voluntarily and could withdraw from the project whenever they wanted. Furthermore, the participants and the institutes were coded in order not to reveal their real identities. The supervisors were coded as S1, S2, S3, to S37; English teacher trainers were coded as TTA, TTB, TTC, TTD, TTE, and TTF; English teachers were coded as ET1, ET2, ET3, ET4, ET5, and ET6; interviewed CELTA holders were coded as CHA, CHB, CHC, CHD, CHE, and CHF; and CELTA holders who filled out the questionnaire were coded as CH1, CH2, CH3, to CH12.

Data collection procedure
The researchers collected data from several sources. First of all, the information provided on the websites of the private language institutes holding English TTCs in Isfahan, Kermanshah, and Boushehr was gathered. It was necessary to add the institutes in Tehran and Shiraz because there was not enough information available on the websites of institutes situated in these three cities. In fact, the researchers needed to collect more data to reach data saturation point. Having coded the information available on the websites of nine institutes, one of the researchers found the main categories of the data. After that, he analyzed the information on the other websites in order to find out if there were any new themes or subcategories in this regard. Data collection continued up to the data saturation point, that is to say, the information on the websites of 34 institutes was collected. Data saturation point is the stage where no new information is found related to the themes under study, and the relation between the themes is established and proved (Strauss & Corbin, 1998).
Next, the researchers conducted semi-structured face-to-face open-ended interviews with the teacher trainers (see Appendix A) and English teachers (see Appendix B) working in the six chosen institutes. This semi-structured interview consisted of open-ended questions and focused on the planning, implementation, and evaluation stages of these TTCs. The interviews were held in the institutes' offices, lasted for 10-15 minutes, and were recorded. The interviews' questions focused on issues such as the trainees' selection, the course content, the way the courses was taught, trainers' qualifications, trainees' assignments, course duration, strengths and weaknesses of the course, and the evaluation criteria of the trainees' teaching practice.
Finally, the researcher utilized an open-ended questionnaire to collect information from the institutes' supervisors. This questionnaire consisted of the interview's questions held with the teacher trainer, but with some slight differences (see Appendix C). It was distributed among 45 institutes' supervisors who were not available for or willing to participate in the face-to-face interview. However, 37 of the supervisors responded to the questionnaire, yet 8 supervisors did not respond to the questionnaires due to their hectic schedule.
On the other hand, in order to collect the required data about CELTA, the researcher utilized the same data sources and data collection methods. In other words, information on the websites of CELTA centers, semi-structured interviews with CELTA holders, and open-ended questionnaires were the sources of information. However, the researchers took advantage of CELTA documents too since they had no access to CELTA trainers in Iran. First of all, the information available in the CELTA syllabus, CELTA trainee's book, and the websites of CELTA centers was collected. Then, the researchers conducted face-to-face semi-structured openended interviews with six Iranian English teachers who had taken CELTA abroad. Finally, a researcher-made open-ended questionnaire which contained the same questions as the interviews was administered among 12 Iranian CELTA holders (see Appendix D).

Reliability and validity issues
The researcher who was in charge of data collection was consistent in data collection across different institutes and with different persons. This is a necessary condition for qualitative reliability which is also known as dependability (Creswell & Miller, 2000;Gibbs, 2007). The researcher utilized the same sources of data for the Iranian TTCs and CELTA course, adopted the same strategies for analyzing the documents, utilized the same interview type and questions for the national and international TTC holders, and used the same techniques for coding the data gathered in both contexts. In order to make sure about the qualitative validity which is also known as trustworthiness, authenticity, and credibility (Lincoln & Guba, 2000), the researcher followed the validity strategies suggested by Creswell's (2009) study. To begin with, the researcher tried to improve the research credibility through triangulation or 'discovery of commonalities' (Kimchi, Polivka, & Stevenson, 1991, p. 364).
The researcher used open-ended semi-structured interviews, open-ended questionnaires, and document analysis. Furthermore, he integrated the views of English teachers, teacher trainers, and institutes' supervisors. Thus, different data collection techniques and various data sources were taken into consideration in order to reach sound conclusions and valid justifications. Finally, the researcher employed member checking, that is, the results of the data analysis were given to some of the teachers and course trainers so that they could attest to the authenticity of the findings.

Data analysis procedure
In order to analyze the collected data and to find the main themes and categories, the researchers employed grounded theory approach. After careful examination and constant comparison of the texts and based on inductive reasoning, the researchers identified the main themes and categories of the data. In qualitative data analysis, the researchers need to take three steps of open coding, axial coding, and selective coding (Strauss & Corbin, 1998). But, before proceeding to coding the data, the researchers had to transcribe the interviews verbatim.
The first step in data analysis was open coding, which is a systematic analytic process through which concepts were identified and their properties and dimensions were discovered in the data (Strauss & Corbin, 1998). In order to code the data, the researcher transcribed the data, read the texts two times carefully, and then assigned those pieces of text which were similar in nature or meaning to one category. In his doing so, the researcher analyzed a whole sentence or paragraph and specified its main idea because he had read the previous literature and had identified several categories in this regard. After choosing names for the categories according to the expressions of the respondents and previous literature, the researcher continued the data analysis giving the same name to each object that enjoyed similar properties with one of the categories. Once the first and second interviews were coded, the researcher had found the main themes and categories of the data. Having found the major themes and categories of the data, it was time to ensure the coding reliability. To this end, the researcher used an external code check to ensure reliability in coding. In other words, the researcher asked one of his colleagues to code the second interview using the list of codes already identified. The researcher did the rest of coding himself since according to Lombard, Snyder-Duch, and Bracken (2002), there was enough inter-rater agreement (87%).
In the next stage which is called axial coding, the researcher answered questions such as where, when, why, who, how, and with what consequences in relation to the established categories. In fact, he discovered how the categories were related to their subcategories along the lines of their properties and dimensions. Finally, the researcher analyzed the data in order to determine the main theme of the research and organize all the other themes and categories around this central theme. The core or central category which illustrated the key idea of the research was planning, implementation, and evaluation of the teacher training course. The results of the data analysis are shown in Table 1, summarizing all the categories and subcategories in holding the TTC in both contexts.

Results and discussion
Analysis of the data about the TTCs in Iranian private language institutes and CELTA showed that there were three stages involved in holding a TTC in both contexts: planning, implementation, and evaluation stages. There were nine main categories in both contexts as follows: aims, trainees' selection, teacher trainer, course content, syllabus, teaching policy, trainees' involvement, feedback, and evaluation. These main categories had quite different subcategories and qualities in these two contexts, except for the last three categories, that is to say, trainees' involvement, feedback, and evaluation. However, even these categories were seemingly similar in these contexts, but there were different mechanisms and policies involved therein. Table 3 shows the main findings of the study.

Aims
The findings revealed that most of the Iranian institutes announced four to six aims which were mostly similar to the content rather than aims. Generally, there were three categories in relation to aims. Some of the institutes (N = 20) aimed to prepare and train English teachers who could teach adults in the Iranian context. These institutes held TTCs lasting for about 30-40 sessions and covered language acquisition theories, teaching methods, and practical teaching techniques. Another group of institutes (N = 7) planned their TTCs according to the institute's needs, held TTCs when they needed to employ English teachers, and focused on the techniques needed for teaching a special book series, such as Top Notch, Four Corners, or Touch Stone. Yet another aim mentioned by four of the English teachers and three of the trainers was that 'a large number of Iranian private language institutes held TTCs just for financial purposes'. These institutes held TTCs when there were enough trainees registered, and they were not strict about the candidates' proficiency and qualifications.
Quite contrary to the Iranian TTCs, CELTA pursued three clear and specific aims which were manifested in the syllabus. By comparison, the goals in CELTA were 'acquire essential subject knowledge and familiarity with the principles of effective teaching, acquire a range of practical skills for teaching English to adult learners, demonstrate their ability to apply their learning in a real teaching context' (CELTA Syllabus and Assessment Guidelines, 2016, p. 2). As can be seen, CELTA aimed to familiarize the trainees with the knowledge of teaching principles and techniques (theory), to teach them the practical techniques (knowledge of practice), and finally to give them the chance to apply these techniques in a real situation (putting the theories into practice).

Trainees' selection
The requirements which were specified for the candidates applying for the Iranian TTCs stood in total contrast to those of CELTA. Contrary to CELTA, Iranian private language institutes were not very strict about the qualifications of the candidates. After filling out the registration forms online or submitting a resume, the candidates had to take a written exam which was based on IELTS or TOEFL exams and was intended to determine the candidates' General English Proficiency. In 24 of the institutes, the written exam was reduced to vocabulary, grammar, listening, and reading sections; the writing section was removed. Three institutes (JIDA, SAMA, and IROX) did not require the candidates to take the written proficiency test at all. In many of the institutes' websites (27 out of 34), it was posted that 'the candidate should have an acceptable level of oral and written proficiency in English language' or 'be in intermediate level of English'. That is to say, the institutes did not specify any exact cutoff score on a standard proficiency test such as TOEFL and IELTS. To top it all, the exact results of the written tests were rarely announced (just in 3 institutes); as a result, the institutes could decide whom to invite for the TTC regardless of their score in the test. The next requirement for attending the TTCs was to attend an interview which lasted between 5 and 10 minutes. 'The candidates' fluency, accuracy, pronunciation, grammar, and motivations for becoming a teacher were assessed in the interview' (TTF). There was usually one interviewer involved in the interview, and the interview was not recorded for scoring purposes. The only exception was GOSA institute, where the interview was conducted in the form of group discussion and was video-recorded (Teacher Trainer C).
In CELTA, no formal academic entry requirements were needed, but the candidates were required to '[b]e at least 20 years old by the end of the course, have the potential to become effective teachers, and have a minimum of a high C1 level in English, both written and spoken that enables them to follow the course and complete all the assessed elements successfully' (CELTA Courses in Seville, 2016). Then, the candidates had to register in one of the CELTA-holding centers using online registration system. Next, they needed to perform a pre-interview task, which was designed to help the candidates prepare for the interview and to help the examiners get a clear picture of the CELTA candidate. This task consisted of different sections such as open-ended questions; writing, or rewriting a paragraph; correcting the spelling, punctuation, and grammatical mistakes; checking word class; predicting learner problems; correcting grammar and vocabulary errors; detecting pronunciation problems; finding differences in meaning; and recognizing word stress. After that, they were required to participate in a face-to-face, online, or on-the-phone interview (CELTA Courses in Seville, 2016). The interview lasted for around 40 minutes, and the questions focused on the candidates' reasons for taking CELTA and basic grammar. The interviewer aimed to ensure that the candidates did the pre-interview task themselves and to test their ability in re-framing and communicating the same concepts in different ways.

Teacher trainer's qualifications
There was no written document to reveal what qualifications a teacher trainer had to hold; thus, it was better to analyze the trainers' resumes and interviews to determine the needed qualification. The researchers came up with three qualifications in this regard. On the one hand, teacher trainers in Iranian institutes (GOOO, ANEL, and AFFF) were usually the experienced teachers of the institutes, having and they had between 8 and 10 years of teaching experience. The second criterion for becoming a teacher trainer was having an international teacher training certificate such as CELTA; the teacher trainers in GOSA and JIDA had CELTA certificate. Most Iranian teacher trainers had two to three TTC certificates held in Iranian language institutes, but none of them had attended any trainer training course. Finally, some of the institutes such as SAMA and KIII cared about the trainers' university degree in English majors and chose trainers who were PhD holders in TEFL. In these institutes, teacher trainers were PhD holders in TEFL, but some of them had no experience of teaching in these institutes.
On the other hand, having analyzed the information provided on the websites of CELTA centers about the CELTA trainers and tutors, it was found that the CELTA trainers were much more experienced than their Iranian counterparts. For example, the CELTA trainers in Seville had started their career as English teachers since 1987, 1988, 1992, 1996, and 1997. Most of them (four out of five trainers in Seville) had done their CELLTA and DELTA courses before starting their careers as teacher trainers. Furthermore, most of them (N = 4) had taught English and trained teachers in more than 6 countries such as China, Saudi Arabia, Russia, South Africa, Oman, Italy, Venezuela, Portugal, and the UK. (Meet the team, 2016).

Course content
The next element compared was the course content, which is one of the most important elements of a training course. This category consisted of three subcategories: topics, source books, and criteria for determining the content. It was revealed that the TTCs in 27 out of 34 institutes covered issues such as the English teaching methodology, classroom management skills, practical techniques for teaching different language skills and components, and error correction techniques. Teacher trainers suggested the following source books: Techniques and principles in language teaching by Larsen Freeman (five of the six institutes), Learning teaching by Jim Scrivener (2 institute), and The practice of English language teaching and How to teach English by Jeremy Harmer (3 institutes). The trainers had different criteria for choosing the course content. Two trainers (TTA and TTC) believed that teacher trainees needed to be familiarized with the English teaching methods. The second group of trainers (N = 2) focused on the institutes' needs; therefore, they focused on the practical techniques needed for teaching a special book series taught in that institute, using the Teacher Guides written by the book's authors. The last group of trainers (N = 2) who had CELTA or DELTA certificates based their content on CELTA source books. These teacher trainers strongly believed that 'CELTA is a comprehensive TTC and every one taking CELTA or attending a CELTA-based TTC can surely teach in the Iranian context'.
By comparison, quite different topics were taught in CELTA, being specified in detail. The main topics covered in CELTA were as follows: 'Learners and teachers, and the teaching and learning context, Language analysis and awareness, Language skills: reading, listening, speaking and writing, Planning and resources for different teaching contexts, Developing teaching skills and professionalism' (CELTA Syllabus and Assessment Guidelines, 2016, p. 3). Thus, topics included in CELTA were related to different aspects and persons involved in the teaching and learning of a language, while Iranian courses selected a limited number of topics mostly related to the teaching aspect, in the wake of which the learning and learner aspects fell into sheer oblivion. There were two categories of books among CELTA sources: grammar books and English teaching methodology books. Advanced English Grammar in Use by Raymond Murphy and Practical English Usage by Michael Swan were the main grammar books. Learning To Teach English by Peter Watkins; How to Teach English by Jeremy Harmer; Teaching Practice by Roger Gower, Diane Phillips, and Steve Walters; Learning Teaching by Jim Scrivener; and Practical Techniques for Language Teaching by Lewis and Hill were the most widely suggested methodology books.

Syllabus
The next category was called syllabus, consisting of the course duration, number of sessions and syllabus specification. The TTCs in Iran lasted from 10 to 60 hours, were held 3 days a week, or just on the weekend. Thus, the courses usually spanned over 2-10 weeks. The classes were usually held in the evening on weekdays (7-10 pm) or on the weekend from morning to noon. In four of the institutes visited, there was no written document specifying the exact syllabus session by session. Just in two institutes, the trainees were provided with the course syllabus, specifying in detail what was going to be taught in each and every session.
In contrast, CELTA was significantly different from Iranian TTCs regarding the syllabus category. The subcategories identified in CELTA syllabus were different modes, number of sessions, and course elements. Candidates could take CELTA full-time lasting for 4-5 weeks, part-time taking from a few months to a year, or online lasting for a minimum of 10 weeks to a maximum of an academic year (CELTA Courses in Seville, 2016). Unlike TTCs in Iranian institutes, CELTA lasted for a minimum of 120 contact hours including the following seven components: input sessions, supervised lesson planning, teaching practice (six assessed hours), feedback on teaching, peer observation, observation of experienced teachers (minimum six hours), and consultation time (CELTA Syllabus and Assessment Guidelines, 2016, p. 3).

Teaching policy
Another important element of a TTC is the way in which the course aims are put into practice, this category was named teaching policy. Most of the TTCs in Iran followed a lecture-based model of presentation. Four of the trainers and five English teachers interviewed believed that 'the course they attended was lecture-based'. ET3 explained the course methodology as follows: 'the trainer talked about the way of teaching language skills and components for most of the class time. Then, one or two of us had a teaching practice called demo'. However, in two institutes (GOSA, GOOO), the trainees said 'we had at least 4-5 TPs during the course'. These two institutes held their courses like a workshop: the trainers demonstrated the way a skill or component had to be taught, and then they asked one or two of the trainees to have demos teaching the same skill or language component. Quite contrary to the Iranian TTCs, CELTA was quite practical, and all the aforementioned seven components of CELTA required the trainees' active involvement in the course. The candidates were taught the necessary knowledge and skills through the input sessions. Then, they prepared a lesson plan with the help of their tutor. Besides observing their peers teaching, the trainees observed experienced teachers classes. After that, they had a feedback session talking about the observations they did, focusing on the strong and weak points of the classes observed. Finally, they started teaching a real class with non-native students at two levels (CELTA Syllabus and Assessment Guidelines, 2016, p. 3). The following quotations from the CELTA holders attest to these points. 'We were provided with lots of information about language learning and teaching techniques. Everybody tried to be active in the input sessions' (CELTA holder B). CELTA holder B added that '[t]hey taught us how to prepare a lesson plan for our classes. I did not know much about this before taking CELTA. Besides lesson planning, we observed different classes and prepared reports on them. This, I think, was the most useful part of the course'. However, CELTA holder C believed that 'participating in the feedback sessions after the observation sessions was really useful'. Finally, he asserted that '[t]he main difference between our TTCs in Iran and CELTA was that we taught a real class with non-native students in CELTA, but in Iran the trainees teach their classmates'.

Trainees' involvement
The next category was the trainees' involvement in the course, which consisted of assignments, observation, and teaching practices. Only in one of the institutes following CELTA syllabus (JIDA) were the trainees required to do four writing tasks. In other Iranian institutes, the trainees just attended the class and took notes, not doing any reading or writing during the course.
Furthermore, just one of the institutes (JIDA) provided the trainees with the opportunity to observe three sessions of the experienced teachers classes and video-recorded classes during the TTC. Finally, the trainees were given the opportunity to practice-teach during the term 3-5 times on average, each one lasting for 10-15 minutes.
In stark contrast to the Iranian TTCs, the trainees were actively involved in the course through pre-course and during-the-course written assignments, assigned reading and research, observation, active participation in input sessions, preparing lesson plans, and teaching practices. The trainees in CELTA were required to dedicate a minimum of 80 hours for reading, research, doing the tasks, assignments, and lesson preparation. They were expected to do one pre-course task and write four assignments of about 750-1000 words related to the topics covered during the course. Besides, they had to do peer observation and to observe experienced teachers teaching classes of language learners for a total of 6 hours, three of which could be on video. Finally, they needed to practice-teach classes of the relevant age group and size for a total of 6 hours (CELTA Syllabus and Assessment Guidelines, 2016). Four of the CELTA holders (1, 4, 5, and 7) told: 'We were required to study a lot and prepare for the course. There were a lot of exercises and assignments'.'

Feedback
The next category found in the implementation stage was the feedback the trainees received on their teaching practices. In the Iranian context, the course trainer and other trainees gave oral feedback on the trainee's teaching practice in five out of six institutes. In one institute, the written feedback was provided on two sides of a paper. Good points of the TP were written on one side, while weak points of the TP were provided on the other side.
In accordance to CELTA regulations, every trainee had to observe his peers' and experienced teachers' classes and fill in special feedback sheets. Then, the trainer asked the trainee to prepare a report of the observed class and share notes of what s/he noticed in his own lessons and observed classes. In order to exchange the views and provide feedback, a feedback session was held and the strengths and weaknesses of previous day's lessons were discussed. At the end of the session, the trainees received the trainers' written comments on and evaluation of their own lessons. Besides the tutor, the trainees received feedback from peers who observed their classes (CELTA Syllabus and Assessment Guidelines, 2016).

Evaluation
The last category in holding the TTCs was evaluation in which the trainees' teaching performance and/ or assignments were evaluated. This category was divided into task, rater, evaluation criteria, and the result subcategories. In the Iranian institutes, the evaluation usually happened at the last session of the course. All the trainees were expected to teach part of a lesson, such as a dialogue, a reading passage, a listening comprehension, or a grammar exercise. This teaching practice which was the only task considered in the final evaluation lasted for 10-15 minutes. In addition to the TP's, the trainees in one of the institute (JIDA) had to prepare a report about their observation and to submit four writings for the final assessment. In five institutes, only the teacher trainer was involved in evaluating the teaching practice task. However, in one of the institutes, there were two raters doing the evaluation (GOSA). Like the previous category, only one institute (JIDA) used the CELTA checklist to evaluate the teaching performance of the trainees, the other raters evaluated the teaching performance holistically and subjectively. Finally, the trainee's performance was indee observed, not graded or scored. In other words, all the trainees attending the course were given the same certificate.
Unlike evaluation in the Iranian context, evaluation in CELTA followed a very clear procedure and was more objective, continuous, and integrated. To be exact, during-the-course activities were taken into account, and evaluation consisted of both teaching practices and written assignments. The trainees had to teach for a total of 6 hours, working with adult classes at a minimum of two levels of ability. Secondly, they had to complete four written assignments during the course, each one about 750-1000 words (CELTA Syllabus and Assessment Guidelines, 2016). The assignments and practical activities were internally assessed and externally moderated by a Cambridge English-approved assessor who sampled portfolios and teaching practice and discussed and agreed the grades for all the candidates. CELTA certificate was awarded to candidates who met the course requirements and whose performance met or exceeded the criteria in both assessment components. Their performance had to match ALL of the descriptors at a particular passing grade in order to achieve that grade. The grades were Pass A, Pass B, Pass, and Fail (CELTA Syllabus and Assessment Guidelines, 2016).

Conclusion and implications
This study aimed to compare the most common procedure among Iranian private language institutes in holding an English TTC with the policies and practices involved in holding CELTA which is an internationally recognized TTC designed for almost the same purpose and candidates. First of all, it was found that the institutes holding TTCs acted quite differently from each other in planning, implementing, and evaluating the course. This lends support to Rezaee and Ghanbarpour's (2016) study who stated that although all the Iranian private language institutes pursue the same goal, they follow different procedures and cover different contents. Besides, in contrast to CELTA which trains teachers for any EFL context, most Iranian institutes aimed to train the applicants for teaching in their own institutes, so they focused on the practical techniques needed for teaching a special book series based on the institutes' needs. This is in line with the suggestions put forward by Bax (1995Bax ( , 1997Bax ( , 1998. They argue that predetermined content results in courses that are not sensitive to the trainees' particular teaching situation, because they are developed for a particular social and educational context. The second group of TTCs was held more often, lasted longer, and covered mostly theoretical aspects of TEFL. These results cast support to the earlier findings by Uysal (2012). In fact, these courses are based on the rationalist model of teacher training, focusing on the teaching of scientific knowledge to the trainees, and expecting them to use the acquired knowledge in their real teaching later. Day (1993) argues that this approach provides the trainees with received knowledge, which is of no practical use in terms of classroom experience.
Next, the findings indicated that in stark contrast to CELTA which followed certain international goals, the Iranian TTCs did not enjoy a systematic planning and did not follow clear and realistic objectives. This finding confirmed those of Uysal (2012) and Özer (2004) who found that participating teachers were not informed of the course aims beforehand. However, Wallace's (1991) and Kirkpatrick's (1998) assert that an effective training course must have a clear philosophy and objective, and the course content should reflect that. Furthermore, it was observed that most of the topics and course contents were related to the received knowledge of language teaching, while all the seven components of CELTA provided the trainees with experiential sources of knowledge and the trainees were very actively involved in the course. These results are partially in line with Bayrakçı (2009) and Odabașı-Çimer, Çakır, and Çimer (2010) who found that Turkish teacher training programs followed a pure transmission model, mostly focusing on theoretical knowledge, not allowing the trainees to take active participation in their learning, or implementing what they had learnt during the course.
Another big difference between CELTA and the Iranian TTCs was that Iranian teacher trainees did not have enough opportunities to practice-teach and do observation during the course. Coupled with the tight scheduling of the courses, this lack of experiential learning prevented the trainees from developing teaching skills. In fact, Iranian TTCs reduced the practical components of CELTA such as TPs, feedback sessions, observations, and experiencing real classes, which were the main sources of experiential knowledge for the trainees. These findings are in line with most of the studies conducted in the area of TTCs in the Middle East (Bayrakçı, 2009;Coskun & Daloğlu, 2010;Erozan, 2005;Odabașı-Çimer et al., 2010;Özer, 2004). Abaszadeh (2012), Hockly (2000), and Sandholtz (2002) also concluded that acquiring the teaching knowledge and skills does not finish by the end of short TTCs, and teaching skills are leaned gradually through teaching and practicing.
Still another difference was that CELTA aimed at providing the trainees at the first level with the knowledge of teaching principles and techniques (theory), to show them the practical techniques (knowledge of practice), and then help them apply these techniques in a real class (putting the theories into practice). To be more exact, CELTA trainees were taught how to move from knowledge and theory to practice, but the Iranian teacher trainees seemed to be expected to immediately apply what they learned in the theoretical part of the course in a class without much support and opportunities for practice. In fact, CELTA is based on a combination of the craft model through providing observation opportunities and feedback sessions and reflective practitioner model by discussing the main problems encountered by the trainees during their teaching, reflecting upon them, and finding solutions to them (Randall & Thornton, 2001). The knowledge gained through craft model is context-bound, and there is no room for generalization of the trainees' beliefs, because the danger is that trainees may use the same techniques they observe in other classes, hence obtaining unexpected results. On the other hand, the Iranian TTCs were based on the applied science model because the first sessions were input-based and teacher-centered. These sessions provided the trainees with received knowledge of teaching English, which is far from enough.
Thus, the researchers strongly recommended that Iranian language institutes set their aims in accordance with the course duration and the trainees' needs and proficiency level. In order to choose more proficient candidates, one standardized English proficiency exam like TOEFL or IELTS should be used to screen the TTC candidates. The researchers suggest that teacher trainers have teaching experience at different proficiency and age levels of private language institutes and have participated in at least one international trainer training course. Besides employing several trainers for teaching different topics of the TTC, the trainers should be up-to-date regarding the new trends and developments in teaching. Finally, course designers are strongly recommended to place fewer trainees in each class to give them more time for teaching practice, and to design longer courses to include the other essential topics included in international courses.
Another suggestion is to design different courses for English-major and non-English majors, since candidates who have studied other majors are not as proficient as English majors and have not passed any courses related to English teaching methodology. The other useful option is to assign the trainees to observe the experiences teachers' classes. However, since there are so many candidates attending Iranian TTCs and observing the classes might interrupt the normal routine of the class, it is wise to provide the trainees with the video recording of the experienced teachers' classes. It is also suggested that the TTCs focus on the teaching techniques and classroom management skills in general and not limit their contents to a specific course book which is the usual trend in most of the Iranian private language institutes. Iranian TTCs need to involve the trainees more in the course through writing three to four reports about the content of the course and observing video-files of experienced teachers' classes. The trainers are strongly advised to revise the content of their courses, to use up-to-date materials, avoid detailed analysis of teaching history, and finally to focus on the practical aspects of teaching.
Finally, Iranian TTC designers are suggested to take into consideration the following points. First of all, the evaluation stage should not be limited solely to the final 10-15-minute teaching performance, which does not happen in a real class. Besides teaching practice, the evaluation should be based on the trainees' assignments, the trainees' activities during the course, and the observations they do. The trainees need to make much more observations, either in real class or of video recordings, since they are not usually given much time for real teaching. The trainers should provide the trainees who are going to make the observations with a certain checklist consisting of different criteria and covering different aspects of teaching. Furthermore, there must be a specific and detailed rating scale for assessing the teaching performance of the trainees. Having two to three raters is another option that can help making a better and fairer judgment. Finally, the TTC certificates which are given to trainees must contain clear information about their teaching ability. In other words, the teaching ability and performance of the candidates should be measured exactly, i.e. presented as a scale or score.