INTRODUCTION

Learning English Language and assimilating to global society are synonymous. In this globalized world, labour market becomes very competitive, irrespective to open (e.g., Bangladesh), close (e.g., Iran) and island (e.g., Cyprus) economies, where desire to secure a high paid job requires extra skills (e.g., proficiency in English) including academic qualification. From such perspective, those skills are getting priority which enhance individual’s communication capacity where the English language has become indispensable for communication in the international arena (Crystal, 2003) and no country can afford to ignore it. Knowing the English language uplifts individual’s socio-economic status by providing advantages in the labour market, higher education and honour in the society of non-English speaking country like Iran. Of being high demand, services of English language learning are being offered by many, including both private and public, institutions. To ensure high quality (i.e., international standard), both native and non-native teachers are being employed in those institutions. Because of indistinctive inherent attributes (e.g., capability, background knowledge), these two types of teachers have advantages and disadvantages at their services which has significant consequences to the English proficiency of the student. Therefore, the linkage between the types of teacher and performance of learner is pecuniarily important that has yet to be studied in Iran.

As part of rapid changes, as in the world, Iran is exposed to the mass media and has easy access to science and the internet which promote necessity for Iranians, especially the new generation, to learn the English language. It has a gate-keeping function for access to higher education and certain types of employment more than anything else. In the context of Iran, English is considered as a foreign language, so, the EFL learners mainly have the opportunity to learn the language in classes as they have little chance to acquire it in society or in a stress-free environment. Therefore, this study has chosen to focus on the teachers’ contributory role. In particular, it aims to investigate the perceived dichotomy between Native Speaking English Teachers (NSETs) and the non-native speaking English teachers (NNSETs) to examine its impact on the learning and learners of English in Iran.

THEORETICAL FRAMEWORK

Much research has been done to clarify the significant role of NSETs and NNSETs in terms of teaching English to EFL learners. Identifying the teacher who is more qualified than other to teach English language is inconclusive (Celik, 2006; Cook, 1999; Medgyes, 1992; Graddol, 2006). While the NSETs have the advantage of knowing the target language (Cook, 1999; Graddol, 2006), the NNSETs can communicate the student better (Seidlhofer, 1999). The NSETs can be better teachers in using spoken form of language and teaching oral skills (Arva and Medgyes, 2000), in contrast, the NNSETs are familiar with the structure of the target language and the learners’ mother tongue, they can solve the learner’s problems better than their native counterparts in terms of interference of learner’s mother tongue and negative transfer (Medgyes, 1994).

Considering both the written and oral proficiencies tests, Derivry-Plard (2005) study on a homogenous group of French post-secondary EFL learners indicates that students taught by the NNSETs performed better in both tests which contradicts with other tenets (Seidlhofer, 1999) that 'native' speakers be better teachers of oral skills" (Derivry-Plard, 2005). However, Carless (2006) study reveals that secondary school students taught by native teachers are more successful than those taught by their non-native counterparts that NSETs have upper hand or are superior to NNSETs in terms of teaching EFL, but contradicts the theory of English as a global language presented by (Crystal, 2003). Therefore, the issue of teacher’s ‘nativeness’ remains a matter of concern in non-native countries like Iran.

English native speakers have the rights and responsibilities not only of controlling the forms and the norms of English globally but also of dominating theory and practice of its teaching and research (Nayar, 1994) on one hand and on the other, native speakers do not always speak according to the rules of their standard national languages; they display regional, occupational, generational, class-related ways of talking that render the notion of a unitary native speaker artificial (Kramsch, 1997). The obvious contradiction and the implication of the studies reviewed above make the researcher to inquire the inconclusiveness of the debate whether types of teachers (native or otherwise) have any effect on the student’s performance in the context of Iran.

EFL COUNTRY CONTEXT: IRAN

The present educational system of the country comprises three levels: primary (5 years), guidance (3 years) and secondary (4 years). Upon completing the secondary level with a diploma, students have to pass the university entrance exam to study in the university.

Presently, English is taught as a foreign language in Iran, and is introduced at the level of guidance school, when the children aged 12. The teaching of the language is then continued into the secondary school and university. At the first two levels (guidance school and secondary school), English is a compulsory subject and students have to attend English classes three hours per week. The public schools have only local Iranian teachers appointed by the Ministry of Education.

As the teaching/learning issue has still not been properly addressed in the public schools, language institutes have been established in Iran as adjuncts to those schools with the aim of overcoming the weaknesses in the system of learning English. This has led to the private language institute administrators hiring language teachers who are preferably native speakers of English. The most common weaknesses of Iranian EFL teachers are their poor English proficiency, lack of ability to hold English conversations in class and their unfamiliarity with educational technologies (Maftoon et al., 2010).

The issue of NSET and/or NNSET can be seen in the perceptions held by administrators of private language institutes who hire English teachers. Their views are usually reflected in their advertisements for teachers. Many advertisements request that only NSETs apply for the position of English language teachers that disadvantages NNSETs in the job market (Liu, 1999). The discrimination against NNSETs and the priority given to the NSETs are disposed mostly by private language institutes and parents of students (Celik, 2006).

RESEARCH DESIGN AND SAMPLE SELECTION

A quantitative method was adopted in this study. An established language institute and two of its sub-branches located in Yazd province, Iran, were randomly selected as the research sites. Three NSETs and three NNSETs (3x3) and 90 newly enrolled female students who are homogenous in attributes (e.g., age, socio-economic status and language proficiency background of the learners and gender) were selected through convenience sampling method (Creswell, 2008) to conduct the study.

The students (90) were divided into two equal groups, each of which was further sub-divided into groups of three, with 15 students in each group. Sampling procedure and the sample size (15) for each class is based on Creswell (2008). The reason for including 3x3 groups of NSET vs. NNSET teacher is that it improves statistical results. All six teachers were selected by the institutes which filtered all their teachers using various tests in order to employ only those who were appropriate and well-qualified.

JUSTIFICATION OF SELECTING PRIVATE LANGUAGE INSTITUTE (PLI)

The rationale behind the selection of PLI as the site of the study was that in public schools usually teach about language meaning just reading comprehensions and structure and words in isolation (out of the context) where as in private Language institute the emphasis is on teaching the language as a system, paying attention to all four skills namely listening, speaking, reading and writing. The second point is that the learners have access to native speaking English teachers in some private language institutes however, in public school the students are deprived of such opportunity. Third point is that a comprehensive comparison between native and non-native teachers would be possible since both types of teacher are involved in teaching procedures only in private language institutes.

TEACHING METHODOLOGY, MATERIALS AND MEASURING INSTRUMENTS

To conduct the educational experiment, the two theories of language and learning were considered. Natural Approach, a theory of language, demonstrates communication as the primary function of language and it is best suited for the beginners and is designed to help them become intermediates (Krashen and Terrell, 1983; Richards and Rodgers, 1990).

As regards to the theory of learning, due to the age of the learners, a combination of behaviouristic approach which finally led to comprehension hypothesis proposed by Krashen (2004) was adopted. Accordingly, the Total Physical Response (TPR) was selected as an approach to teaching for the first three weeks. Asher (1977) believes that “abstraction should be delayed until students have internalized a detailed cognitive map of the target language”. This trend continued up to the time the learners gained some basic familiarity with the sound system, at which point there was a shift towards the communicative approach to teaching and learning proposed by Hymes, Swain and Canale in the 70‘s. The aim of this approach was to help foreign language learners communicate in the target language.

With regard to teaching materials, during the experiment, the students of all six groups were exposed to a set of textbooks entitled New Parade I and II by Herrera and Zanatta (2001) along with another textbook, On Your Mark, by Davy (2000), which have been recently used in some private language institutes in Iran. The experiment consists of 50 sessions of 100 min three times per week.

Data were collected via a semi-direct oral test and administered to the learners in a language laboratory and their responses recorded for scoring procedures. The reliability of the test was computed through a pilot study using intra- and inter-rater procedures (76 and 71.5%, respectively) and by applying the Pearson Product formula.

The test, which had been pre-recorded, comprised three equal parts, each comprising eight items (Appendix A). It was broadcast in the laboratory and the respondents had to respond to the items in English and their responses were recorded. The instructions in the test were in Persian (the learner’s L1) in order to enable the respondents to better understand what they were supposed to do, but the rest of the test was in English. Two NSETs scored subjectively via listening to the tape individually and the head researcher himself, an NNSET, transcribed the test verbatim. The average of the three scores thus given was considered as the raw score of each individual.

FINDINGS OF THE STUDY

Initially, to see whether there were differences among the mean scores of the three classes supervised by the three NSETs, the one-way ANOVA was applied. Table 1 shows the result of ANOVA which indicates that there was no significant difference between or among them. Statistical analysis indicates that the p value of observed scores (amount of statistical significance), 0.552, is larger than the predetermined alpha level (probability = 0.05), meaning that the three groups taught by NSETs had obtained similar scores on the oral-test and the difference is not statistically significant (Table 1).

Table 2 depicts the statistical analysis that determined whether there were significant differences among the three classes taught by the NNSETs.

Table 1:	Differences among the groups taught by NSETs

Table 2:	Differences among the groups taught by NNSETs

Table 3:	Data analysis related to the semi-direct oral test
NG: Learners taught by NSETs, NNG: Learners taught by NNSETs

Table 4:	Semantic errors of NGs and NNGs
Smaller mean indicates fewer errors

Table 5:	Grammatical errors of NGs and NNGs
Smaller mean indicates fewer errors

Table 2 indicates that there was no significant difference among the three groups taught by the NNSETs. In other words the amount of significance (0.771) (Table 2) is greater than the predetermined alpha level 0.05; this means that the three groups had scored similarly on the oral-test and the difference is not statistically significant.

As evident from the collected data, both groups of learners exhibited good oral performance; however, as the results of the t-test and the means of the two groups of the learners indicate, learners taught by NSETs performed better than those taught by NNSETs (Table 3).

The analysis of the transcribed version of the semi-direct oral test showed that the grammatical errors of both groups were higher in number than their semantic errors (Table 4, 5). This may indicate that all learners, including the participants of the study, were more powerful in terms of receptive (understanding meaning) rather than productive skills (producing grammatical sentences). Therefore, for the better achievement of the learners, teachers should pay specific attention to the receptive skills when teaching. This idea was raised by Widdowson (1977) more than three decades ago; he stressed that both receptive and productive skills should walk hand-in-hand. This means that a good speaker needs to be a good listener; similarly, a good writer should be a good reader.

Based on the means of the two groups located in Table 4 and 5, one can conclude that in terms of both meaning and grammar, learners taught by NSETs made fewer errors than those taught by NNSETs.