Efl Teachers Implemented The Clt Approach Education Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

This study aims to further explore secondary school teachers attitudes and beliefs towards the CLT in EFL classrooms, using the country of Libya as example. We do not yet know what are the teachers doing in their classrooms? Are they implemented the CLT approach or just follow the traditional methods such as GTM and ALM. What are the challenges they encounter to implement the CLT approach? To fully answer these questions, we need to examine the teachers' actual practice in the classroom; what the teachers' attitudes and beliefs about the implementation of the CLT approach. Therefore, the next chapter of this study investigates the following questions:

1. Are secondary EFL teachers implemented the CLT approach?

2. What are the teachers' attitudes and beliefs towards the CLT approach?

3. What are the challenges that the teachers encounter to implement the CLT approach in their classrooms?

In view of the requirements that CLT places on learners as well as teachers, the previous chapter has shown that there is much to be asked about their ability to adopt this approach in EFL settings and still much to investigate here. This chapter is organized in the following order. It begins with a brief description of the rationale for using a quantitative and qualitative research design. This is followed by a description of the participants. It concludes with a description of the procedure for data collection and analysis which includes the instruments used to collect the data for this study: The first instrument was a questionnaire, the second instrument was a teachers' structured interview and the third was classroom observation. A discussion of the rationale for choosing structured interviews. The chapter ends with the data analysis procedures used for analyzing the collected data.

3.2 The triangulation methods

A questionnaire, structured interview and classroom observation were used in this study to collect data. The combination of these three methods is considered to be of a vital important. Attitude is multiple of dimensions that include components of cognition, emotion and behaviour as well as situational constraints. This means that study of attitudes towards CLT should address these four components and the relation between them. In order to give a full picture of teachers' beliefs and attitudes towards the CLT approach, this study applied methodological triangulations. Triangulation takes different forms such as seeking different sources and utilizing different methods, but the underlying concept is that several different data sources and/or analytic approaches should be used (Brown, 2001). The main purpose of this chapter is to explain the nature and rational of the methodological triangulation that has been used in this study.

3.3 Rationale for use of qualitative and quantitative research design

Before explaining how data were collected, I will present a rationale for using two different types of research design. Where the study relied on teachers' beliefs, feelings and experiences about the use of the CLT approach in Libyan classrooms, mixed methods were employed for this study; a qualitative and quantitative research methods. Creswell (2008: 62) identified 'mixed methods designs' as "procedures for collecting, analyzing, and mixing both quantitative and qualitative data in a single study…". Qualitative research is recommended for studies concerned with complex phenomenon. This mixed approach give to elaborate, clarify, explain and confirm data (Jang et al, 2008: 221). Seliger (1989: 118) says that "Qualitative methods originally developed from the methodologies of field anthropologists and sociologists concerned with studying human behaviour within the context in which that behaviour would occur naturally and in which the role of the researcher would not affect the normal behaviour of the subjects". A qualitative method is suitable for describing second language acquisition in its natural context. In addition, qualitative research is "to stress the unique strengths of the genre for research that is exploratory or descriptive, that assumes the value of context and setting, and that searches for a deeper understanding of the participants' lived experiences of the phenomenon" (Marshall & Rossman, 1999: 60). More importantly, qualitative research is characterized by its lengthy involvement with participants, through observations and semi structured interviews. Qualitative research has been well established in most academic fields of study. It has a unique approach to research that can draw on multiple sources and on people's views and opinions of specific experiences.

In this specific study, it was important to use a qualitative research approach due to the reliance on individual perceptions of a particular situation. "Qualitative researchers are intrigued with the complexity of social interactions as expressed in daily life and with the meanings the participants themselves attribute to these interactions" (Marshall & Rossman, 1999: 2). Not only was it important to document teachers' perceptions, but also to determine how their environment, in this case an EFL environment, affected and shaped such perceptions since, as mentioned previously, the context and setting in which teachers experience the phenomenon affects their overall perceptions.

Quantitative methodology introduced analysis of the data in numerical and statistical forms. Henning (1986: 702) points out that "without some resource to quantitative methods, some marriage of words and numbers, it is inconceivable that the investigation of language acquisition will ever be said to belong to the realm of scientific inquiry". For Henning, quantitative analysis helps the researcher go beyond the data and generalize to other areas of study. Other researchers using qualitative methods have realized this need to quantify their data (Spada, 1987, Flyman-Mattsson, 1999). According to the research questions the use of both qualitative and quantitative methods are applied in data analysis.

There is a need to really understand what is going on in EFL secondary schools in Libya and to describe the teachers' attitudes and beliefs to obtain a clear picture of what is going on in EFL classrooms. Quantitative analysis gives benefits to quantify, generalize, and compare the data with other similar studies. It is also useful to draw conclusions, and make more accurate judgments. By employing this mixed method, teachers are able to show their knowledge and experience towards the implementation of the CLT approach and explain the challenges encounter them in its implementation in their classrooms.

The use of triangulation to validate data is an important aspect of qualitative research and serves to guarantee credibility in reporting findings. According to (Amores, 1997: 521) triangulation is "the collection and comparison of data from two or more separate observations or illustrations of the behaviours being studied".

For this study, data were gathered in three ways: through questionnaires given to participants, observations of classroom activities and procedures, and interviews with participants that helped validate both the answers in the questionnaires and observations. Triangulation of methods is important in order to get a full picture of teachers' attitudes and beliefs towards the CLT approach, this study employed the methodological triangulation mentioned in the paragraph above. As Brown (2001) reported, triangulation takes different forms such as seeking different sources and utilizing different methods, but the underlying concept is that several data sources and/ or different analytic approaches should be used.

The data were collected by a conducting questionnaire, structured interview and observing teachers in their classrooms using Communicative Orientation of Language Teaching (COLT). For the questionnaire, as Genesee & Upshur (1996), noted that the questionnaires are most useful when employed periodically and when relatively systematic and uniform feedback is desired from students, parents, or teachers, for example, before instruction begins and at the end of major units of instruction or an entire course of instruction. At these times, the information they provide is valuable for planning and assessing whole courses or units. They also added that if they are structured rather than open-ended, the answers to questionnaires can be easily quantified. Gay (1992) mentioned that an advantage of questionnaires over interviews is that they can be administrated to a bigger sample at one time and are less time consuming. Genesee& Upshur (1996), further say that questionnaires provide concrete and fixed records of subjects' answers. As for interviews, they are superior to questionnaires in finding out about individuals (i.e. teachers') beliefs and attitudes. Oppenheim (1992) mentioned that interviews allow more to be said about the research issues than is usually mentioned in a questionnaire.

For the class observation, the reason behind its use is that it is seen as the best way to examine what teachers actually do in their classrooms, and how they use communicative activities in the classroom. The study investigates CLT in secondary schools in Libya and the main research question that this study is aiming to address is why is CLT not working and what are the solutions?

Having presented an overview of the methodology used in this study, I will first describe the setting and participants.

3.4 Sample

There are two types of sampling, probability (random sample) and non- probability (non-random/ purposive sample). Probability sampling means that every member of the population gets an equal opportunity of being selected in the sample as the selection is completely out of the researcher's control. Whereas for non-probability sampling, the researcher intentionally decides either to include or exclude a section of the wider population (Gay & Airasian, 2003: Cohen et al 2007, Bryman, 2008). In this study the probability sample (random sample) was implemented.

3.5 Setting

For the current research, the primary location of the research was Libyan secondary EFL teachers from 12 schools in the city of Misurata city (see Table 3.1 below). I chose Misurata as it was easy for me to gain access to schools to recruit all the participants needed for my research. A friend enabled me to get a permission to gain access to participants and collect the data, which would also have been difficult had I not worked at secondary schools. The locations for the research were state secondary schools randomly selected in the city of Misurata. The schools names are (see Table 3.1): Shuhada Yedder; Arrwesat; Abbad; Aljazeera; Arramala; Azzarouq; Gaser Ahmed; Karzaz; Abu Rawaiya; Addafniya; Al-Gheran and Azzaweya where I conducted the questionnaires, classroom observations and interviews. All locations are in Misurata city, 210 kilometres east of Tripoli, the capital.

Table 3.1 School's locations

School Name


Shuhada Yedder












Gaser Ahmed

Gaser Ahmed



Abu Rawaiya

Abu Rawaiya





3.5 Participants

Sample size has been defined by Larson-Hall (2010) as "the number of participants in a study" (p: 401). The participants in this study consist of twenty-four participants who comprised Libyan secondary teachers of English as a foreign language teaching in the state schools. All participants in this study are males as it is more convenient for a male in Islamic society to sit observe and interview males with ease. Islam religious prohibits females to sit with or talking individually to a non- relative male. As Rees and Althakhri (2008: 127) stated it, Islam has affects people's beliefs, behaviours and practices in Arab countries. Thus, females are often unable to participate in the interview even if they wish to. The married female teachers need to get the permission from their husbands and those unmarried ones need to get the permission from their family. However, getting this permission is very difficult (see Metcalfe 2008; Rees and Althakhri 2008). All teachers are answered the questionnaires, interviewed and observed.

As soon as I got the permission from the Ministry of Education, I decided to divide the teachers into three groups according to their experience (see Table 3.2 below): Group 1 contains 13 teachers with most experiences (54.17%); Group 2 contains 6 teachers with medium experiences (25%) and Group 3 contains 5 teachers with least experiences (20.83%). I further divided the teachers in to two groups according to their Education background: Group 1 consists of 16 teachers (66.67%) with Education background and 8 teachers (33.33%) with no Education background. Two teachers from every school were randomly selected. I was the sole investigator who conducted questionnaire, interview the participants and observed them. The participants were selected by purposively sampling Bogdan & Biklen (2007). This selection of: (a) BA holders in English or linguistics (b) at least one year of teaching experience. Even though all these teachers have experience in teaching English, they might have different attitudes as to the effectiveness of CLT in their teaching contexts. Even though the participant teachers reported that CLT was introduced recently, only few participants had received CLT training courses.

Table 3.2 Participants Experience background

Teachers with most experiences (more than 10 years)

Teachers with medium experiences (from 5 to 10 years)

Teachers with least experiences (less than 5 years)




To compare any differences between teachers with Education background and with no background, 16 participants in this study were teachers with Education background (66.67) and 8 teachers with no Education background (33.33%), (see Table 3.2). Since I was well aware that the teachers' level of English might limit the information they provided, I administered the interviews in Arabic and provided an Arabic translation for the English questionnaire so the participants could express their thoughts and ideas fully in their native Arabic. Of all twenty four participants, their ages ranged from 23 to 60 with an average of 41.5 years old.

Table 3.2 shows Participants' Educational background


With Education background

With no Education background




3.6 Research instruments

The main aim of this study is to uncover the EFL teachers' attitudes and beliefs towards CLT and to achieve this as I stated above, I adopted a triangulation methodology: 1) survey questionnaire, 2) structured interview and 3) classroom observation. According to Peng et al. (1991: 105), cited in Usunier (1996: 148), "... with the use of a combined quantitative- qualitative approach, we shall improve our capacity in revealing the holistic, naturalistic, and inductive aspects of the phenomena under investigation".

3.6.1 Three methods of study

Questionnaires are used mostly to collect data on phenomena such as attitudes, motivation and self conception which are not easily observed. They are also used to obtain background information about the research subject such as age, previous background in language learning, number of language spoken and years of studying the language (see Oppenhein 1992; Richard & Lockhart 1994: 10; Seliger & Shohamy 1989: 172). The purpose of the questionnaire survey in this study is to assess the teachers' reported attitudes toward CLT, whether the teachers are implemented the CLT or not and what are the constraints encounter them to implement it. The q can elicit information from the teachers who cannot be interviewed or observed because they are too numerous to be approached individually (Oppenheim 1992). In addition, the use of the random selection of the teachers makes the results more recognizable to the wider population.

Observation is fundamental and highly method in all qualitative inquiry (Marsh & Rossman 1997: 107). Mcmillan & Schumacher (2201: 273) write:

As a technique for gathering information, the observational method relies on a researcher's seeing and hearing things and recording these observations, rather than relying on subject's self-report responses to questions or statements.

Many researchers (Weir & Roberst 1994: 164) have regarded the use of classroom observation as the most appropriate method for the checking of the implementation of a teaching method.

In this study, the purpose of the observation is to get direct information regarding the teachers' actual behaviour in the classroom settings. Analysis of the data serves to find out their states beliefs and attitudes are reflected in their behaviour and what it is that made their classroom teaching traditional or communicative. Another advantage of using classroom observation, according to Mcmillan & Schumacher (2201: 273) "is that the researcher does not need to worry about the limitation of self-report bias, social desirable or response set, and the information is not limited to what can be recalled accurately by the subject". As far as this study is concerned, the most important advantage in the classroom observation can guide to the social desirability factor. A q survey conducted to ask teachers about their attitudes toward CLT may draw favourable responses. Although teachers were asked honestly, there is a possibility not to admit that they have negative attitudes toward this popular method. Through classroom observation, however, teachers' behaviour in the classroom can be accurately observed.

Interview can be used to explore general topics to help uncover the participant's experiences, interpretation and meaning (Marchal & Rossman 1999: 108). But unlike questionnaire, the qualitative in depth interview is able to explore in depth information from interviewees. In this study, interviews are carried out to elicit reports of the teachers' classroom practices and the rationale behind their practices. The use of the interview enables the interviewer to note and record the information that goes beyond the q and classroom observation. In addition, the interviewer is available to clarify questions that the respondents cannot understand, because the interviewer is directly communicating with them. Interviews allow for the clarification and ambiguity in questions and responses.

Although each method has its limitations, the combination of them can overcome them. Although are useful in terms of their extensive focus, they are limited in that they take a snapshot of development in one moment in time, under very specific conditions. Questionnaires tell little about the actual process in the classroom instructions are carried out. Classroom observation can present rich descriptions which can be used to understand teaching practices. However, the limitations lie on the fact that they do not reveal the rationale underlying the practices. Interviews are useful for investigating beliefs and situational constraints that the teachers encounter in using CLT. However, limitations lie on the fact that interview do not inform about the actual classroom practices. Thus, using a single method cannot elicit complete information about teachers' attitudes toward CLT, while the use of the combination of the three methods can overcome the limitations of each individual method.

Most previous studies have investigating teachers' attitudes towards CLT using either classroom observations (Nunan 1987); interview Li 1998 or questionnaire Karavas Doukas 1993, 1996). Most of these studies adopted a single research method to study one or two aspects of attitudes and have their limitations. A combination of the three methods allows this study to go further to the previous ones and ensure that adequate attention is paid to the description of beliefs, attitudes and overt behaviours as well as situational constraints. Questionnaire

Genesee and Upshur (1996) state that "questionnaires are most useful when employed periodically and when relatively systematic and uniform feedback is desired from students, parents, or teachers-for example, before instruction begins and at the end of major units of instruction or an entire course of instruction. At these times, the information they provide is valuable for planning and assessing whole courses or units". (1996: 127).

They also added that if structured, questionnaires can be easily quantified. Gay (1992) pointed out that an advantage of questionnaires over interviews is that they can be administrated to a bigger sample at one time and are less time consuming. Genesee and Upshur also stated that questionnaires provide concrete and fixed records of subjects' answers.

According to Gass and Mackey (2007: 148) questionnaires are " written instruments that present all participants with the same series of questions or statements, which the participants then react to either through providing written answers , marking Likert-style judgements or selecting options from a series of statements".

This study's aim is investigate secondary EFL teachers' beliefs and attitudes towards the implementation of the CLT approach and to identify the constraints encounter them to implement it in their classrooms. According to (Cohen et al, 2007; Gass & Mackey, 2007) questionnaire is a suitable method to collect data for as it gives all the participants the same questions.

I administered thirty questionnaires to twelve secondary schools in Misurata city. Of the twenty nine returned questionnaires, five had to be discarded as answers were left blank.

From March to May 2010, I administered a 5- point Likert questionnaire to twenty four Libyan EFL secondary school teachers in six schools in Misurata. A Likert scale questionnaire was selected as a primary means of data collection because, as Fraenkel and Wallen (1996:129) put it, "... it is possible to discover attitudes by asking individual to respond to a series of statements of preference... The pattern of responses is then viewed as evidence of one or more underlying attitudes. The questionnaire, which consisted of twenty four items, was adapted from Karavas- Doukas's (1996) and anchored by "strongly agree" and "strongly disagree". Karavas (1996: 195) says that the Likert type scale (or method of summated ratings) is the most widely used method of scale construction for three reasons: first, its relative ease of construction; second, its use of fewer statistical assumptions; and third, it defers to other scaling techniques, as no judges are required in that. Karavas says distinguishes between Likert scale with the two other very common techniques of attitude scale construction (Thurstone & Guttman), a Likert scale does not use experts to judge which statements are most appropriate for the attitude scale. Besides, it does not use a laborious procedure called Scalogram analysis to select the most appropriate items for inclusion in the scale. In addition to the 24 items, the questionnaire also asked about participants' demographic information and asked open- ended questions in which the participants had to define the term communicative language teaching either in English or Arabic. Also they were asked to indicate whether they would be interested in being interviewed and/or in allowing me to observe a class of their choice where they would illustrate principles of CLT. A professor in linguistics reviewed and commented on the questions, and I modified them accordingly in order to neutralize loaded meanings in some of the original items. Of the thirty questionnaires distributed, twenty nine completed ones were returned, accounting for 96.67% of the population. However, five of the returned questionnaires were dropped out because most answers left blank and didn't seriously answer the whole questions. Finally, the data obtained from the survey questionnaires were quantitatively analyzed using descriptive statistics, such as means and standard deviation (SD). I used the Microsoft Excel Package in order to facilitate the analysis procedures. The questionnaire constructions

The questionnaire starts with a demographic part for personal information of the teachers such as place of graduation; sex and teaching years. The first part seeks information about Importance of grammar instructions (statements 1, 3, 12, 17, 18 and 23). The second part of the questionnaire is asking for the role of the teacher (statements 4, 5, 7, 8, 11, 16, 19, 20 and 24). The third part is for Group and pair work activities (statements 2, 9, 13, 21 and 22). The final part is seeking information for Scale of error correction (statement 6, 10, 14 and 15).

In order to encourage the teachers to complete answering the whole questionnaire, I itemised the statements in a Likert scale which consists of five rates from 'strongly disagree' (SD) 'disagree' (D), 'uncertain' (U), 'agree' (A) and 'strongly agree' (SA). Cohen et al (2007: 331) refers to the likert scale layout as "economical of space". Larson Hall (2010: 395) also points out to it as "typical". The questionnaire consists of a total of 24 statements as shown in Table 3.3

Table 3.3 Part 1 (Means (M), Standard Deviation (SD)). Number of participants is 24 teachers



Standard Deviation

1- Grammatical correctness is the most important criterion by which language performance should be judged.



3. Grammar should be taught only as a means to an end and not as an end in itself.



12. Knowledge of the rules of a language does not guarantee ability to use the language.



17. By mastering the rules of grammar. Students become fully capable of communicating with a native speaker.



18. For most students, language is acquired most effectively when it is used as a vehicle for doing something else rather than studied as a primary end in itself.



23. Direct instructions in the rules and terminology of grammar are essential if students are to learn to communicate effectively.



Table 3.4 Part 2 (Means (M), Standard Deviation (SD)). Number of participants is 24 teachers



Standard Deviation

4- When a learner comes to the language classroom with little or no knowledge of the language, he/she cannot be expected to suggest what the content of the lesson should be or what activities are useful for him/her.



5. Training learners to take responsibility for their own learning is not feasible when learners are not used to such an approach.



7. The teacher as "authority" and "instructor" no longer describes all the teacher's roles in the language classroom.



8. The learner-centered approach to language teaching encourages responsibility and self- discipline to development to full potential.



11. 1n a large class, it is not possible to organize teaching so as to meet all individual needs.



16. The teacher as transmitter of knowledge is only one of the many different roles he/she performs during the course of a lesson.



19. The role of the teacher in the language classroom is to impart knowledge through activities such as explanation, writing and example.



20. Tasks and activities should be negotiated and adapted to suit the students' needs rather than determined by the teacher in advance.



24. To meet all the needs and interests, the teacher must supplement the textbook with other materials and tasks.



Table 3.5 Part 3 (Means (M), Standard Deviation (SD)). Number of participants is 24 teachers



Standard Deviation

2. Group work activities are essential for cooperative relationships and productive interactions among students to emerge.



9. Group work allows students to explore problems for themselves and thus have some measure of control over their own learning.



13.Group work activities are not an efficient use of planning and teaching time.



21. Small group work may occasionally be useful to vary the routine, but should be secondary to sound formal instruction by a competent teacher.



22. Group work activities have limited value since it is very difficult for the teacher to monitor the students' performance and prevent them from using their mother tongue.



Table 3.6 Part 4 (Means (M), Standard Deviation (SD)). Number of participants is 24 teachers



Standard Deviation

6. For students to become effective communicators in the foreign language, the teachers' feedback must be focused on the appropriateness and not the linguistic form of the students' responses.



10. Unless all their grammatical errors are corrected, students will be at risk of imperfect learning.



14. Since errors are a normal part of learning, too much time should not be spent in correction.



15. The communicative approach to language teaching produces fluent but inaccurate learners.


1.02 Validity and Reliability of Teachers' Questionnaire

Various kinds of validity checks can be assigned by researchers to consider the validity of the research methods used. Although all types of validity can significantly contribute in the success of any research, content validity was more relevant and important for this questionnaire. Construct (content) validity implies that the instrument used should cover the topic under investigation fairly and comprehensively (Gass & Mackey, 2007: 4).

A good content (construct) validity of questionnaires can be done through other academics' reflection on their contents and structures (Cohen et al 2000; Gay & Airasian 2003; Bryman, 2008). Thus, the views of the teachers' questionnaire of this study were viewed by Dr Martha who supervised me on this study. Different reviews and discussions also made by my PhD colleagues who made many comments. In addition, a PhD friend revised the Arabic translation version for accuracy (see appendix. ). Interviews

As Gay (1992) pointed out that the interview is a good way to obtain in-depth data of people's beliefs, feelings and attitudes. The need for interviews of participants in the current research is for the comparison between participants' beliefs and their actual practice. Breakwell (1990) distinguishes between two kinds of interviews: structured and unstructured. Unstructured interviews deal with the topics not the questions. No questions; order, but depends on the flow of the conversation. This kind of interviews allows participants to say as much as they like which is good, but it difficult to compare the interviewees. They are also difficult to analyze. However, in structured interviews the interviewer questions the respondent exactly the same questions, and in the same order (Crowl 1993: 124). This is good to compare between subjects and cover all topics. For the above reasons, the researcher decided to conduct structured interviews, following a set of guidelines that need to be followed for setting questions (Breakwell, 1990; Oppenheim, 1992; Genesee and Upshur 1996). They should not include complex words or lead the interviewee. They should not be long, or challenge the interviewee's memory by asking him/her to recall old information, and they should not be able to be interpreted in more than one way. Oppenheim (1992) refers to the idea that interviews allow more to be said about the research issues than is usually mentioned in a questionnaire. Interviews are often used in the literature for investigating perceptions and knowledge (Karavas-Doukas, 1996; Zhang, 1997; Sato & Kleinsasser, 1999; Sakui, 2002). Also the researcher can clarify the misunderstandings, and observe the interviewees and take notes of his/her behaviour. Finally, interviews give in-depth information gathering and flexibility where information is impossible to obtain by any other method (Seliger & Shohamy, 1989; Gay, 1996). The structured interview consisted of fourteen questions about Communicative Language Teaching such as using group work, and other activities, using audiovisual aids, and teaching grammar. An Arabic version was distributed to the participants to ensure participants understood them. Teachers' responses were written down while they replied and they were also recorded. .

The present study included structured interviews with twenty four participants. A structured interviews, as Merriam reported (1998: 74), "is guided by a list of questions or issues to be explored, but neither the exact wording nor the order of the questions is determined ahead of time". Merriam further suggests that a structured interview is flexible enough to allow the research to respond to the situation at hand, to the emerging world-view of the participants, and to new or unforeseen ideas on the topic.

I opted for a structured interview because through this method I was able to uncover teachers' attitudes and beliefs towards CLT. In other words, the interviews helped me develop insights on how the participants interpret and define their world. Through interviews I gathered descriptive data in the interviewees' own words (Bogdan& Biklen, 2007). For this study, a structured interview was chosen for individual interviews in order to gather a comparable data, across participants (Bogdan & Biklen, 2007). The rationale for selecting structured interview as a data collection method is that it provides a compromise between what Dornyei (2007) refers to as two extremes, unstructured and structured interviews. Even though the researcher prepares a set of guiding questions and prompts prior to the interviews, the format of the structured interview is open-ended, giving a chance to the respondents to explore on issues raised in the interview questions. In other words, as Dornyei (2007) states, the questions provide guidance and directions, and the interviewer is free to follow up on any interesting developments that may arise during the interview session. Since I did not want to limit the depth and breadth of the interviewee's narrative, the interview guide simply provided guidance and directions to the interviews (Dornyei, 2007).

Interviews as a data collection method have other advantages. First, interviews allow one to gather a large amount of data quickly, Bogdan& Biklen (2007); Marshall& Rossman, (1999). Given that I had over four months to work with the participants, structured interviews were the easiest and quickest way to obtain a large amount of data on the teachers' attitudes and beliefs. Second, the researcher could request immediate clarification or pose follow-up questions after the interviews, Marshall& Rossman (1999).

In order to recruit the participants, I first read through the returned questionnaires and sorted out which indicated an interest in interviews. The same participants who completed the questionnaire were interviewed. Six participants dropped out for financial reasons as they wanted me to interview them in their home towns and costs prohibited this.

The participants were asked where they prefer to be interviewed and the appropriate time for them it. After a week time a final arrangement were made and the participants were contacted regarding the place and time of interviews (Limerick et al. 1996), (see appendix).

On the day of the interview, a review of the interview questions was intensively made. At the beginning of each interview, the researcher introduced himself to the interviewees in a friendly atmosphere I followed the interview processes by letting them feel at ease and relax; I offered them a hot or cold drink of their choice. At the beginning of each interview, I emphasized once again that the purpose of the interview was to talk about and share their knowledge and experiences in teaching English as a foreign language in Libyan secondary schools, and their beliefs and attitudes, rather than to assess their knowledge of CLT. I asked for their permission to tape- record the interview (see appendix). While I made sure that each interview should last about an hour at the most, I did not stop those participants who were highly engaged in their conversations with me on the topic, but rather I tried to see which direction the interviews went. I assured them that their names would not be disclosed. All of interviewees were informed to withdraw or stop the interview at anytime (Berg, 2009). The interviews were conducted in Arabic language due to the interviewees' willing (Gass & Mackey, 2007: 135) and to enable the conversations to flow naturally (Briggs 1986). At the end of the interview, the participants were asked to review their responses carefully as they may use inaccurate or miss any data (Bell, 2005). After I was done with the twenty-four interviews, I transcribed and translated them into English for inclusion in the thesis appendix and for reference in the text of this thesis. Then each interviewee verified each English script. Finally, I am able to summarize and transcribe the data in a comprehensive way (Cohen et al, 2007). Interview design

I decided to prepare a set of question of 36 divided into 4 topics to elicit the teachers' beliefs and attitudes towards the CLT approach to help answering the research questions (see Chapter 2). The interview topics were chosen by consulting some PhD theses (Embark 2011; Al-Nouh 2008) and I made some modification to them. The questions were reviewed and checked many times and then were divided into 4 topics, as I stated before, the first topic is "Teaching language in the classroom". It covers 7 questions aiming at knowing the teachers' knowledge and practice of the CLT in their classrooms. The second topic is about "CLT implementation". It consists of 5 questions aiming at discovering whether the teachers implement the CLT and if not what are the challenges they encounter, it aims at knowing about the resources availability and it aims to check the training that teachers may undertake. The third topic is "Instructions". This topic covers 11 questions, aiming at knowing how teachers teach grammar instructions and how they teach fluency and accuracy. The fourth topic is "Error correction". It consists of 6 questions aiming at knowing how teachers correct students' errors. The fifth topic is "Assessments. It covers 4 questions aiming aat finding out how teachers testing communicative competence. The final topic is "New curriculum". It aims at knowing why students make errors in speech and the possibility to teach the four skills equally. The interview questions are 36 distributed to 24 teachers:

Teaching language in classroom

What do you think is the most important skill in teaching FL, listening, speaking, reading or listening?


Do you think EFL learners should be pushed to communicate in English from the beginning of language learning

Is it possible that students can interact in English with each other if put in pairs/groups?

Is it possible to speak English during the whole lesson without using your L1?

Do you speak English all the time with your learners, during classroom time?

Do you use group and pair work activities in your lessons and when?

CLT implementation:

Is it possible to implement CLT in Libyan secondary schools

Is it possible to implement other methods rather than CLT approach?

Are there any constraints encountering the implementation of CLT?

Are resources available in Libyan secondary schools?

Do you undertake training courses?


Can learners acquire the rules of grammar subconsciously through just hearing input?

Is it possible to teach grammar through communicative activities without explanation or drilling?

Do you think using language for a real communicative purpose will result in fluency and accuracy?

How do you use English for a real purpose in the classroom?

Do you explain grammar rules or do you teach them implicitly through pattern drilling?

Do you think the teacher in the foreign language classroom is to provide target language input?

What role do you assume in your classroom, a provider of knowledge or a guide of students activities?

If you aim for communicative competence. How would you focus on accuracy?

Is it possible to teach fluency before accuracy or accuracy before fluency or both at the same time? Which is more difficult to do and why?

Which is more difficult to do accuracy or fluency? And why?

Do you teach fluency before accuracy or both at the same time?

Error correction

What do errors represent?

Which errors do you think you need to correct, those affecting meaning or those affecting language?

Is it possible to ignore students' errors during a communicative activity? And which errors would you ignore?

Which errors would you ignore during a communicative activity?

How often do you correct your pupils' errors?

Do you immediately correct them or do you leave them after the activity is finished?


How can one assess the development of linguistic competence?

Is it possible to test students' communicative competence?

How do you test communicative competence?

Which part of communicative competence do you test, grammatical, discourse, sociolinguistic, or strategic?

New Curriculum

Why do learners make more errors in spontaneous speech than they do on written exams?

Do you think your students after learning English as a foreign language will end up being communicatively competent?

Is it possible to teach listening, speaking, reading and writing equally well? Classroom observations

For classroom observation, I adopted the COLT (Frohlich, Spad and Allen. 1985) scheme to observe the same teachers of the total sample being interviewed. Classroom observation was chosen to collect data as Gay (1992) describes it as the only way to get more objective and accurate information about teachers' practices in the classroom than an interview or a questionnaire. Also, as Crowl (1993) says, "There are numerous forms of behaviour that can best be measured by direct observation rather than by paper and pencil tests or by questionnaires. In education, one of the most common forms of behaviour that is best measured by direct observation is behaviour in a classroom setting", (Crowl, 1993: 125). Besides some studies like Orafi & Borg (2008); Al-kwaiter (2001) report that there is inconsistency between what are actually teachers saying and what they are doing in classrooms.

As far as the methodology is concerned, a qualitative method is used in this study to find out whether teachers are implementing the CLT approach in their classrooms. The qualitative method was chosen in this study it focuses on discovering truth and underlying meanings (Babbie, 2004). According to Gay (1996) behaviour occurs in a context and to understand the behaviour we need to understand the context in which it occurs. However, quantitative research collects numerical data to explain, predict, or control the phenomenon being investigated (Gay, 1996). The combination is useful to compensate for respective strengths and weaknesses. Seliger and Shohamy (1989) point out that the combination of the two methods in field work is an ideal cyclic process.

As for the observation, it is necessary to assess human skills and behaviour. It allows the collection of detailed and complex information that might not be possible to obtain using other methods such as questionnaires (Genesse and Upshur, 1996). The need to investigate teacher practices and beliefs, according to Woods (1996), stems from thinking that teachers are not transparent entities who fulfil curriculum plans and goals as prescribed by their authors, but who filter, digest, and implement the curriculum depending upon their beliefs and environmental contexts. Also and more importantly, as Ruiz-Funes (2002: 3) argues, teachers' "own experiences, beliefs, and practices have not been adequately recorded". Classroom observation becomes vital as it is through the classroom "the prime elements of learning and teaching, ideas and ideologies, policies and plans, methods and materials, learners and teachers- all mix together" (Kumaravadivelu, 2003). According to Seliger & Shohamy (1989), researchers could participants in the observations where they participate in the act that they are describing; or non-participant where they observe and record notes of the observed behaviour (Seliger and Shohamy, 1989). For me, I am non- participant observer in the present study.

The instrument used in this study is an adapted version of the COLT scheme (Communicative Orientation of Language teaching), Allen et al. (1984) & Frohlich (1995). The major aim of COLT is to capture differences in the communicative orientation of L2 classrooms. The present study has similar concerns. It sets out to measure the extent to which instructional treatment in Libyan secondary EFL classes may be characterized as communicative. However, since the COLT scheme had been developed in a different context, some modifications were introduced to adapt it to the Libyan teaching situation. I shall go into these changes and the rationale behind them later on in this chapter.

The COLT scheme is a real-time instrument which enables the observer to describe pedagogical events as they occur at the level of task type, class organization, content, skill and materials use. It includes specific categories which distinguish communicative from non-communicative approaches. They are, thus, variables which serve to "measure the extent to which an instructional treatment may be characterized as communicatively oriented". Frohlich et al. (1985: 29).

As (Al- Nouh, 2009) classifies it, COLT Observation Scheme is divided into two parts: Part 'A' is about classroom activities, (related to the teachers and pupils' activities); part 'B' is about teacher-pupil interaction, (related to teachers and pupils' activities).

Table.3.3 COLT Observation Scheme

COLT Observation Scheme

Part "A"

Part "B"

Classroom activities

Teacher- pupil interaction

Categories include: participant organization, content, content control, student modality, and materials. For example, the categories start with 'Participant Organization' which refers to the way the students are organized Spada & Frohlich (1995). Part "A", as Spada & Frohlich divided them, refers to: participant organization, content, content control, student modality, and materials. 'Participant Organization' refers to the way the students are organized Spada & Frohlich (1995). It is divided into three patterns: 'Class', 'Group', and 'Individual'. The category 'Class' refers to the activities led by the teacher where the teacher interacts with the whole class or with individual students. It also applies to activities led by a student to another student or a student to class. Finally, it applies to whole class or individual groups in choral work. 'Group' refers to activities done in groups. 'Individual' refers to activities where students work alone. The second category is 'Content' which refers to the activities done by the teacher and students, (what they say, read, write or listen to in the classroom). Spada and Frohlich (1995:16) divided two other areas into: Language and Other topics, Language is further divided into: 'Form', 'Function', 'Discourse' and 'Sociolinguistic'. 'Form' covers grammar, vocabulary, pronunciation, spelling, and punctuation; 'Function' constitutes communicative functions such as requesting, apologizing, and explaining; 'Discourse' is how sentences are combined in a cohesive and coherent way, such as describing a process, e. g. what students did when they visited the moon Spada & Frohlich (1995: 17), and 'Sociolinguistic' is style appropriate to particular contexts Spada& Frohlich (1995). According to Spada & Frohlich (1995), 'Content control' refers to who chooses the topic or task which is the focus of instruction. Spada & Frohlich (1995) divided them into three subcategories: 'Teacher/text', 'Teacher/Text/Student' and 'Student'. 'Student Modality' refers to the skills involved in a classroom activity. 'Materials' were excluded from this study because they are not concerned whether teachers implement CLT or not. Savignon (2002) argued that even without authentic materials a teacher can still teach communicatively, as it depends on the teacher's understanding of CLT and how CLT happens. COLT, part '3', is about teachers' and students' verbal interaction. It is divided into two sections: teacher verbal interaction and student verbal interaction. The teacher verbal interaction categories include: target language off task-, target language on task, information gap; sustained speech; and reaction to form/message.

A sub-category, Pair-work, was added in the instrument used in the present investigation because, during the unstructured observations, this type of class organization was used in three of the classes observed. Furthermore, the textbook series English For Libya includes pair-work activities. It was thought that this category would help to see how far the teachers follow the directives in the textbooks and teacher's books. The sub-category Group/individual work of the COLT scheme was replaced by the sub-categories Group-work and Individual seat work. The tasks included in the textbooks always require learners to be organized in the same way, all learners working either individually, in pairs or in groups.

It was assumed that the sub-categories Pair-work and Group-work would help show whether the Libyan teachers provided their learners with the opportunity of developing their communication skills through actual use of the TL. In the literature on CLT, pair-and group-work are considered to be an important factor in developing learners' communicative competence (e.g. Long, 1990). The reason for this claim is that teacher centred approaches are thought to impose restrictions on learners' productive ability. In classes dominated by the teacher, students spend most of their time responding to questions and rarely initiate discourse (Allen et al. 1984).

The observation took place on two occasions for each of the 24 participants. Each observation lasted for 40 minutes, the same duration as the class itself. Therefore, the total is 48 classes, totalling 1,920 minutes. Classroom observations were carried out in six schools which were located in 12 areas in Misurata, as mentioned above: Shuhada Yedder; Arrwesat; Abbad; Aljazeera; Arramala; Azzarouq; Gaser Ahmed; Karzaz; Abu Rawaiya; Addafniya; Al-Gheran and Azzaweya. It might be argued that these areas do not represent the different possible sites where CLT is possibly being used. It should be pointed out that for the sake of representativeness of the sample, the researcher intended to visit more schools than those mentioned above, but, unfortunately, the Ministry of Education only allowed me to conduct the research just in the schools mentioned above. Nevertheless, according to King et al., (1987) a sample is representative when it includes people from the different categories of the population under investigation.

The major factors which were believed to influence classroom practices, and distinguish classes, were class level, teacher's experience and location of the school. Thus, the lessons observed for this study included (a) classes taught by teachers with Education background; b) teachers with no Education background; c) teachers with teaching experience and teachers with no teaching experience Table 3.4 below shows the classification of observations in terms of teacher experience (novice versus experienced), school location (urban versus rural areas) and class levels.

Table.3.4 Education and Experience background of Survey Participants:

Teachers with Education background

Teachers with no Education background

Teachers with most teaching experience

Teachers with medium teaching experience

Teachers with least teaching experience






The researcher used a sensitive Sony cassette recorder to pick up most sounds taking place in the classroom. A video recording was avoided as some were not willing to be video recorded. Spada and Frohlich (1995); Zotou, (1993) used audio recordings, and they reported that it was more convenient to use as it had better sound quality and easier to code the COLT categories. As I mentioned before, I played no role as participant observer. I interacted with the neither students nor the teachers during the observations. I entered the classroom with the teacher so not to distract the pupils' attention. The recorder was turned on and adjusted at the beginning of the class and ended after the teacher finished the lesson. I wrote down important classroom notes which would not have been picked up by the audio recording, for instance, gestures, the use of visual aids or the board. I took notes of the steps of the lesson using a digital clock as to what the teacher and students were doing and when to measure how much time was spent on a single activity. The recorder was used to pick up the teachers' and students' oral production during various activities. Then, I filled in the COLT Observation Scheme based on my notes and the recorded material.

I contacted schools and met teachers. I met with every teacher in his own school before conducting classroom observations with the head teacher. I explained the research procedures to them and their written consent was taken after the observation in order not to make them anxious. (see Appendix 7 for the consent form). Teachers were notified a week prior to their classrooms observation.

The observations took place from April to June 2010. A total of twelve secondary schools were involved in the study. The average is about 40 students in every classroom. They attended classes six days a week from September to June each year. Every teacher was observed twice, as mentioned above, and a week was left between the two observations as Spada and Frohlich (1995); Spada (1987) who used COLT in their studies to observe classrooms left one week interval between the two classroom observations to observe how teachers and students handle different themes in the syllabus. It was impossible to leave more than a week between the two classroom observations as student needed to prepare for the final examinations and start their summer vacations.

In order to meet the validity criteria, classroom observation should satisft the following conditions:

Observations provide an accurate record of what took place. This means that observation measures should indicate significant features in behaviour.

Observation data are relevant to the characteristic features of the program.

Observation data are complete in that the whole program is fairly reflected (Weir & Robert, 1994: 172)

In this study, some measures were taken to satisfy the first condition. The first measure was the use of a structured instrument which includes low-inference categories. According to Malamah-Thomas (1987) such categories are not subject to divergent interpretations, which increases the validity and accuracy of observational data. The second measure was to ensure the categories were mutually exclusive. Mutual exclusivity of categories was tested when the instrument was piloted. A third measure was to avoid informing teachers that they were going to be observed. The intention here was to allow the researcher to attend a lesson that had been prepared under normal circumstances. The fourth measure was to control for the observer effect. The researcher was a non-participant observer. He always enjoyed the classroom at the same time as the teacher and the students. He sat at the back and avoided any interruption or behaviour likely to attract the attention of the teacher or the students while the lesson was proceeding. Weir & Robert (1994: 173) stressed that the presence of the observer may not present an important threat to the validity of the data because classes tend to revert to normal when observer becomes a familiar presence when teachers and students become engaged in their tasks. That is because much classroom discourse is based upon strongly habituated routines followed by both teachers and students.

Bearing in mind the second condition suggested by Weir & Robert, which states that observational data must be relevant to the characteristics of the programme, a conscious effort was made to include categories that represent different features of CLT. These characteristics were determined from a review of the ESL/EFL literature on this methodology. A conscious effort was also made to adapt the instrument to the Libyan context and to the purposes of this study.

With respect to the third condition stipulating that the researcher should make sure that the whole programme is fairly reflected, an effort was made to observe different types of classrooms. According to King et al. (1987), three major variables are likely to differentiate classes, namely, class level, teacher experience and location of the school. In the present study, the sample included classes from first, second and third levels of the secondary schools. The sample also included classes which were taught by teachers with/out Education background and others with most, medium and least experience (see Table).

3.7 Data Analysis Procedures:

3.7.1 The survey questionnaire:

The data were organized and summarized to be meaningful. In this quantification stage, I assigned numerical codes to non-numerical categories of demographic information that could be counted for further tabulation, and/or interpretation. After the collected data were coded and categorized, I analyzed them by using SPSS 7.0 Version for Windows (Statistical Package for the Social Sciences). Descriptive statistics were used for the data analysis. These included frequencies, percentage, mean and standard deviation.

3.7.2. The interview:

Prior to the analysis of the structured interview data, I asked the interviewees to read the translated version of the interview scripts for verification. Then, I began the analysis by reading them carefully several times (Athanases & Heath, 1995), I returned frequently to the research question in order to refocus my work. In so doing, I pulled out salient themes that emerged, cognizant of the important fact that "... the critical task in qualitative research is not to accumulate all the data you can, but to 'can' (i.e., get rid of) most of the data you accumulate" (Wolcott 2001:44). Specifically, I wrote down units of information on separate sheets. The sheets were subsequently coded and sorted centring on situational factors (who, what, when, where, and why). The data thus obtained helped confirm or refute the analysis of the survey questionnaire. More importantly, the interview data helped narrow down the focus of the analysis.

The researcher herself/himself is an instrument in qualitative inquiry (Creswell, 1998), and his/her subjectivity plays a role in the study (Marshall & Rossman, 1999). Bearing this in mind, I made a conscious effort to unpack my subjectivity to understand the biases my own specific backgrounds that I brought to the study (Bogdan & Biklen, 2007). Recognizing my ambivalent feelings towards CLT, I tried to reduce the effects my biases can have on data. For instance, I employed triangulation of methods (questionnaire, classroom observation and structured interview) in order to minimize biases and achieve a more accurate interpretation of the data. At the same time, I used our commonality to forget a trusting relationship. I also used our common ethnic, professional, and linguistic backgrounds to establish rapport. For example, rather than starting each interview in English (which would be the accepted language of communication among English language practitioners in Arabic), I began each interview in Arabic and switched to English only after the participants chose to conduct the interviews in English. Thus I attempted to minimize the social gap between the participants and myself and to form solidarity with participants.

As Marshall& Rossman (1999) stated, the success of interviewing hinges to a large extent on the interviewer's interpersonal skills. Thome's (1983) suggests that a researcher cannot simply be a "sponge like observer", and Marshall & Rossman's recommend that "when the researcher [is] minimally intrusive and present for a short period of time, building trusting relations must proceed in conjunction with gathering good data" (p.80), I decided to begin building a trusting relationship with the participants by attending two of the classes that the participants taught. In these occasions, I was able to introduce myself to the participants and students as well.

3.7.3 COLT data analysis

Structured observation is more reliable than unstructured observation when generalizability of findings is the aim of a given study (King et al., 1987). Structured observation uses predetermined categories of description which can yield statistically analyzable data and allows for statistical inference to identify objective patterns and trends. It provides an objective standard of judgment to bring to evaluation findings and allows for generalizing results from a sample to a larger population. According to Weir and Roberts (1994), three measures can be used to ensure the reliability of observational data and the generalizability of results, namely, stability coefficient, observer agreement coefficient and repeated measure reliability coefficient. Stability coefficient compares measures made by the same observer of the same event on different occasions. Weir & Roberts (1994) stress that this method is not appropriate for describing a class in a school on different occasions since these events are not comparable. Besides, "although in training this measure may indicate consistency in judgment (i.e., reliability), it would not necessarily be valid (i.e., the observer may be consistently inaccurate in the completion of a checklist or in coding classroom talk)" Weir& Roberts (1994: 172). An observer agreement coefficient consists in comparing observations made by different observers of the same event occurring on a single occasion. This measure was not applicable to the present study which was carried out by a single researcher. The third measure, repeated measure reliability coefficient (more often used in other fields of enquiry, such as language testing), consists in administering the same in