Students ’ Preferences on the Use of Mother Tongue in English as a Foreign Language Classrooms : Is it the Time to Re-examine English-only Policies ?

In literature on bilingual teaching, different perspecttives exist for and against the use of first language (L1) in English as a Foreign Language (EFL) classrooms. There is a continuing matter of debate on whether L1 contributes to or precludes the learning of a second language (L2). Numerous studies have been conducted on this topic, but no clear consensus exists on whether L1 should be banned or its inclusion in EFL classrooms should be allowed. A significant body of literature has explored this phenomenon from teachers’ perspectives, and an adequate number of studies have explored the phenomenon from the students’ perspectives. However, few studies have identified the reasons for which students need such an inclusion. More importantly, no studies seem to have considered demographic variables that may directly influence students’ perceptions of the use of L1 in their (L2) classrooms, that is; educational background and their language proficiency level. This study primarily investigated whether students’ educational background and their L2 proficiency influenced their perceptions of the use of Turkish in their L2 classrooms as well as their perceptions and needs for the use of L1 in their classrooms. The study was conducted on a sample of 303 Turkish learners of EFL at English Preparatory School of European University of Lefke in Northern Cyprus. Data were collected using a questionnaire. Analysis of the data indicated that EFL students had high positive perceptions toward the inclusion of L1 in their L2 classes and that their perceptions were affected by their demographic characteristics. Students with lower level of L2 proficiency were also found to have more positive perceptions toward the use of L1. Furthermore, the particular issues where students needed L1 were also identified. Implications for language teachers and policy makers are discussed.


Introduction
Controversies for incorporating mother tongue (L1) in English as a Second Language (ESL) and English as a Foreign Language (EFL) classrooms have been a longstanding discussion in the current literature.Many arguments for and against the use of L1 were made by a significant number of professionals; some of them have considered L1 to be beneficial in ESL and EFL classrooms (i.e., Burden, 2001;Chick & McKay, 1999;Schweers, 1999), whereas some have considered it to be a hindrance (i.e., Cianflone, 2009;Macdonald, 1993).In contrast, a third group of researchers suggest a criterion for when and what amount of L1 should be used in EFL classrooms (i.e., Elridge, 1996;Voicu, 2012).In other words, they agree with the inclusion of L1 in EFL classrooms, though seeking a limit for its use.However, there is a common perception that seems to dominate educational institutions that the use of L1 should be avoided in EFL classrooms, and many institutions have an English-only policy.This idea is often supported by the view that learning of L2 is strongly determined by maximum exposure to that language, so L1 should be banned from classrooms.However, several researchers (i.e., Jenkins, 2005) have criticized this perception, and a movement toward the inclusion of L1 in L2 classrooms has begun to attract attention.All these conceptions, however, seem to be rather naïve as there is neither a persuasive body of research nor satisfying empirical work indicating that abandoning or allowing the use of L1 in EFL classrooms would better contribute to students' learning of a L2 (Swain, Kirkpatrick, & Cummins, 2011).
Such lack of empirical work also seems to be one of the reasons for such disunity and inconclusiveness among the professionals on the phenomenon.A lack of studies conducted in different contexts and considering different variables (i.e., students' perceptions, students' educational background, and students' level of language proficiency) when studying the phenomenon seem to be another weakness of the current research, which needs special attention (Schweers, 1999).
Considering this, this study contributes to the literature by exploring EFL students' perceptions of the use of L1 (Turkish) in their EFL classes in English Preparatory School (EPS) of European University of Lefke (EUL), Northern Cyprus.In addition, it investigates whether the use of L1 differs according to the language proficiency level of the students and whether their educational background-that is, their past learning experiences in schools-impacts their perceptions of the use of L1 in the present context.Furthermore, this study identifies students' needs when demanding for L1.We believe that exploring the phenomenon from students' perspective would enrich the findings in the area as we believe that both the views of educational professionals as well as students' views and factors influencing their views must be essentially considered.Expectantly, the findings of this study will guide professionals when designing educational policies and help them reconceptualize their arguments according to the needs of students-the active agents of the classroom and the learning process.

Arguments for and against the Use of L1 in EFL Classrooms
Carefully examining the literature suggests that three popular views dominate the area on the use of L1 in EFL classrooms.Among the optimistic group, Brookes-Lewis (2009), for example, stated that the inclusion of L1 in L2 classrooms is unavoidable, and it may also positively contribute to the learning of the target language.Similarly, Storch and Wigglesworth (2003) mentioned that "the use of L1 may provide learners with additional cognitive support that allows them to analyze language and work at a higher level than would be possible where they restricted to sole use of their L2."According to these researchers and many others (i.e., Burden, 2001;Cianflone, 2009), the use of L1 should be allowed in classrooms owing to its contribution on learners' cognitive development, and the learning of second language should model the learning of first language.Furthermore, they believed that a distinction between first and second language is necessary to indicate the difference to the students (Burden, 2001).In contrast, the pessimistic group (Harmer, 2001;Swain et al., 2011;Littlewood & Yu, 2011) argued that inclusion of L1 in EFL and ESL classrooms do not considerably impact students' learning of the target language.Moreover, some of them (i.e., Littlewood & Yu, 2011) were also of the view that allowing the use of L1 do not strengthen learners' cognition but may result in inappropriate transfer of the bad language habits from first language to the target language; thus, it should be abandoned.Those who shared the view of using monolingual approach (hindrance of first language) in classrooms, however, were strongly criticized as such an approach was considered to be impractical, and maximum exposure to the second language approach alone was considered to be inadequate for second language learning (Cook, 2001).
Numerous researchers, on the other side, have argued that if L1 is used appropriately and necessarily in EFL classrooms, it is beneficial rather than a hindrance (Afzal, 2013;Elridge, 1996;Spahiu, 2013;Voicu, 2012).Instead of taking strong stances toward the inclusion or exclusion of L1 in EFL classrooms, these researchers preferred to suggest criteria for when and what amount of first language should be used.Such criteria are often based on the aspects such as the purpose of using L1, students' language proficiency level, and duration of the foreign language classes of learners (Afzal, 2013).Based on the criteria available in the literature, approaches and techniques that focus on guiding teachers to appropriately use L1 in classrooms are also proposed (i.e., code-switching).However, researchers, in the last two decades, have focused on determining how and when to use L1 and agreeing on a policy regarding the extent to which students and/or teachers will use L1 such that a balance can be achieved (Schweers, 1999).

Research on the Use of L1 in EFL Classrooms
The literature addresses that the majority of the studies conducted on the use of L1 in EFL classrooms focused on teachers of EFL and considered teachers' opinions on whether the L1 should be used.Studies by Afzal (2013), Jenkins (2005) and Khati (2011), for example, examined the opinions of the teachers who observed positive development on students' learning of ESL when their first language is used in the classroom.Few studies, however, focused on the opinions of the students.Schweers (1999), for example, found that Spanish students mostly preferred their teachers to understand their first language.His study also revealed that 88.7% of the students in the study favored the use of Spanish in the class as it helped them to learn the second L2 better.The students in the same study also favored the use of their first language (up to 39%) during the lessons.Rodrigues and Oxbrow (2008) also found that students believed that using their L1 in L2 classroom helped them to improve their L2 learning.The majority of the students in the study also mentioned that they preferred their teachers to explain grammar rules and give instructions in their L1.Their preference toward their teacher explaining the similarities and differences between the first and second language was also an interesting finding of this study.A more comprehensive study conducted by Bouangeune (2009) revealed that students who received instructions in their L1 showed a better improvement in English than the second group who received instruction in the second language.Finally, a study conducted in Croatia by Kovacic and Kirinic (2011) examined students' and teachers' preferences toward using Croatian in English classrooms.Both participant groups in the study were found to have positive perceptions regarding the use of Croatian in the class; 56% of students and 45% of teachers considered the use of L1 in the classroom to be beneficial and important.
A considerable number of similar recent studies exploring students' perspectives have been conducted, which often suggest the benefits and assistance of first language use in EFL classrooms, although the application of monolingual approach in second language teaching classrooms is still popular.However, current studies on this phenomenon seem to neglect certain aspects.First, although studies on students' and teachers' preferences of L1 use in EFL classrooms exist, to our knowledge, none of these studies have considered variables such as students' English language proficiency levels and students' educational background, which would be indicators of the reasons of their perceptions.We believe that investigating such issues would provide deeper insights into the phenomenon and enrich the findings in the area.Therefore, we explore students' perceptions toward the use of L1 in their EFL classrooms by linking this to their educational background, which can be perceived as their educational habits, as well as their language proficiency (i.e., whether the need for L1 decreases along with the increase in students' proficiency level).We believe that the findings of such a study will effectively guide policy makers, curriculum designers, and other relevant professionals when designing their second language programmes as well as re-examine the validity of English-only policies.

Method
This study adopted a quantitative methodology to obtain numerical data for wider generalization.

Research Questions
This study seeks answers to the following research questions to enlarge the knowledge base in the area: 1-What are students' perceptions toward the use of Turkish in their EFL classrooms?2-Concerning the demographic characteristics of the students, what are the differences in students' perceptions toward the inclusion of Turkish in EFL classrooms?3-Regarding the proficiency level of the students and their past experiences in learning English, what are the students' perceptions toward the use of Turkish in their EFL classrooms?4-What are the students' needs of the use of Turkish in EFL classrooms?

Research Setting
This study was conducted at EPS of EUL in Northern Cyprus.Every academic year, students from various countries, mainly from Turkey, begin their education at EUL. Students who are offered a place at EUL and who will study at departments wherein medium of instruction is English are asked to prove their English language proficiency as required by the university (i.e., by providing a certificate of achievement or by attending to university's English proficiency exam).Those who are unable to prove their English language proficiency are placed at EPS of the EUL.EPS offers one-year program of training in English that is 25 hours per week (which lasts for approximately 34 weeks) to prepare these students for the required levels of English to continue studying in their departments in the following years.Students who are enrolled in the English training programme are often categorized into two different levels of proficiency: beginner and elementary.On some occasions, however, students with a higher proficiency level of English language are also placed in appropriate group levels, such as pre-intermediate or intermediate.Students whose levels of language proficiency are determined as beginner and elementary are often described by the institution as having none or slight knowledge or skills of the English language.In every eight weeks at the institution, exams are held for each level, and the students who become successful at any level move to an upper level (i.e., a student who began at the beginner level advances to the elementary level if he/she passes the exams).
English language training offered by EPS is general English.The curriculum employed at EPS is heavily formed according to the communicative approach, and the school applies a strict English-only policy.This means that neither the teachers nor the students of the institution can use their L1 during classes, but they have to use L2 in all aspects of academic work.The syllabi of the school and the course books used during the classes are also designed and selected according to this policy, and the directorate of the institution is stern regarding the use of L1 at the institution.

Participants
In the 2014-2015 academic year, 357 students were enrolled in EPS.These included students from Turkey and Turkish Republic of Northern Cyprus (TRNC), whose L1 was Turkish, and students from other various countries, whose L1 varied (i.e., Kazakhstan and Azerbaijan).Those who were not from Turkey and TRNC and whose L1 was not Turkish, however, were not invited to participate in the study as they were few; the intention was to reveal information regarding the Turkish students.Consequently, 303 Turkish participants were selected with simple random sampling method with 95% confidence intervals and sampling error of 2%.Demographic characteristics of the students who participated in the study are presented in Table 1.Distribution of sociodemographic characteristics of the students are indicated in Table 1.
Table 1 indicates that 46.2% of the students who participated in the study were aged 20 years or below, 36.6% of them were aged 21 years or below, and 17.1% of them were aged 25 years or older.Of the participants, 20.4% were female and 79.5% were male.Considering the type of high schools from which the students graduated, 29.3% of the students graduated from either College, Science, or Anatolian high schools, 41.2% of the students graduated from state high schools, and 29.3% of the students graduated from vocational high schools.In terms of the level of graduation, 6.2% of the participants graduated with a low level of GPA, 26.4% graduated an average level of GPA, and 67.3% graduated with a high level of GPA.88.1% of the students who participated in the study are the citizens of the Republic of Turkey, and 11.9% are the citizens of TRNC.In terms of the longest duration of stay in a residential area, 78.5% of the students lived in the city centre, whereas 21.4% of the students mentioned their previous residential areas as town or village.
Table 2 provides information regarding students' level of proficiency and their educational background.
The distribution of the English language proficiency level and educational background in English of students who were enrolled in EPS by the time of this study are provided in Table 2.In addition to their level of proficiency, students were also asked to state the level of school in which they began learning English.45.21% of them signified that they started learning English in the primary school, 28.38% of them in the secondary School, 8.25% in the high School, and 18.15% of them first started learning English in the university.10.89% of the participants stated that they were previously enrolled in an English foundation course.The students were also asked to evaluate their level of English before they attended the EPS program; 64.36% of the students evaluated their level of English as low, and 28.71% of them mentioned that it was at an average level.Only 6.93% of the students indicated that they had a high proficiency level of English.While 76.90 % of the students who participated in the study were pleased to receive English education at the university, 23.10% expressed their dissatisfaction with this.30.36% of the students who participated in the study also stated that they knew another language other than English.

Data Collection Instruments
In the study, questionnaires are used as the main instrument for data collection.The questionnaire comprised two sections.The first section of the questionnaire, which was prepared by the researcher, involved 12 questions to identify students' demographic characteristics.In the second section of the questionnaire, there were 7 questions that were adapted from the study by Sevik (2007), who has also adapted his questionnaire from the surveys prepared by Schweers (1999) and Tang (2002).The questions in the second section were related to the use of L1 in EFL, in this case; Turkish.Main focus and the aim of the questions in the second section was first to identify students' requirements of Turkish in the classroom as well as their level of satisfaction for such an inclusion, and second to understand the reasons for and contribution of the use of Turkish to their target language learning in the classroom.See Appendix A for the questionnaire used.

Data Collection Procedures
Questionnaires were conducted during January-February 2015.Before the students began filling the questionnaire, they were informed regarding the aim of the survey, and necessary instructions were provided to them regarding ways to answer the questions.To obtain higher response validity, the students were explained that they did not have to write their names on the questionnaire and that their responses would be kept confidential.
After obtaining their consent, the questionnaires were distributed to the students and collected back after they filled them out.

Data Analysis
The data obtained from the questionnaires were analyzed using Statistical Package for the Social Sciences version 19.0 for windows evaluation.
To determine students' demographic characteristics, their previous foreign language experiences, and their perceptions toward the use of L1 in EFL classroom, a table of frequency was used as a tool to indicate the differences.Square analysis was used to compare the results obtained from the comparison of the students' sociodemographic characteristics with their perceptions toward the use of Turkish in EFL classrooms regarding their previous foreign language experiences.The results are shown in crosstabs.

Results and Discussion
In this section, findings of the study regarding students' perceptions toward the use of L1 in their classrooms, the impact of their demographic characteristics, and the influence of their English language proficiency level on their perceptions are presented.The circumstances when the students needed to use L1 or to receive input in L1 are also explored.Findings are given under each research question posed.

Students' Perceptions of Using Mother Tongue (Turkish) in EFL Classrooms (Research Question 1)
Table 3 shows the distribution of the students' perceptions toward the use of Turkish in their EFL classroom.
According to the table, 69.97% of the students believed that Turkish should be used in their classrooms while 30.03% disagreed with this.Similar to Schweer's (1999) findings (88.7% said Yes), the majority of the students (69.97%) in this study would like Turkish to be used in their EFL classrooms.On the other hand, in a similar study by Kovacic and Kirinic (2010), the need for the Croatian in EFL classrooms was revealed to be 100%.Although the participants in the Croatian study were at the intermediate and upper-intermediate level, they fully requested Croatian to be used in EFL classrooms.The main reason for this could be students' proficiency level of English as it can be assumed that as the level increases, complexity in the language arises, and consequently the demand for L1 can increase.However, this study suggests no significant differences for the demand for L1 use in terms of students' level of language proficiency (elementary, pre-intermediate, intermediate).

*More than one option can be chosen
Considering the students' views of why Turkish should be used in EFL classroom, 62.38% of them signified that it would help the understanding of the difficult issues.48.18% of the respondents also mentioned that it made it easier to understand new vocabulary items, and 26.07% of them said they felt at ease, comfortable, and were less stressed.Finally, 42.90% of the participants said that they felt less lost if Turkish were used in their classroom.A similar finding was also found in Dujmovic's ( 2007) study where students considered that L1 helped them to understand difficult concepts and vocabulary items better by 30% and 90%, respectively.
When this study's findings are compared with those of the study by the Kovanic and Kirinic (2010), it can be seen that the percentages in this study are lower.It seems that the belief that as the students' proficiency level decreases, the demand for L1 increases is supressed with the view that as English gets more complicated, the demand for L1 can increase at a parallel rate.Furthermore, Dujmovic (2007) also indicated the percentages of the students who feel at ease, comfortable, and were less stressed and less lost as 36% and 43%, respectively, which are similar to those of this study.Schweers (1999) also indicated the percentage of the students who feel less lost as 68.3%, whereas it is 42.90% in this study.This is perhaps because Schweers (1999) did not consider the language proficiency level of the students' who participated in his study, and there is a possibility that his participants' level of proficiency was lower or higher than this study's participants' proficiency level.
While 48.84% of the students in the study seem to be satisfied with the use of L1 by their teachers, 45.54% can be seen to be sometimes satisfied with this situation, and 5.61% stated that they were not pleased with this situation at all.Since the results related to the item three indicated a much higher positive view with regard to the students' preference of L1 use by their teachers, this might indicate the drawback of the item one, where 30% of the students said Turkish ought not to be used in class.On the other hand, only 5.61% of the participants gave a negative response to item three, which was a very similar statement.The reason for this could be that the item one was too structured (limited to two answers), whereas item three included the option of "sometimes" where the ones who gave negative responses to item one preferred to choose "sometimes" when answering item three.
It can now be viewed as a more reliable and valid item, and thus its findings should be considered as the main finding for the preferences for the use of Turkish in EFL classrooms instead of item one's findings.

Findings of the Correlation between the Students' Preferences of Turkish and Their Demographic Characteristics (Research Question 2)
Table 4 shows the comparison of the students' level of preferences of Turkish to be used in EFL classrooms according to their demographic characteristics.
According to the participants' responses toward the use of L1 in their EFL classrooms, 63.57% of the students who are between the ages of 20 or younger, 72.07% of the students who are aged 21-24, and 82.69% of the students who are between the ages 25 or above, expected the use of Turkish in their EFL classrooms.It has been found out that there is a meaningful statistical difference between the students' view of including Turkish and their group of age (p < 0.05).Students in higher age group seem to expect Turkish to be used more than the other age groups do.This might be due to the fact that adult learners desire to comprehend every utterance made by the teacher (Schweers, 1999).Table 4 suggests that while 50% of the female students expect Turkish to be used in EFL classrooms, this ratio was 75.10% for male students.It has been proved a significant statistical difference exists between the expectations of female and male students toward the use of Turkish in EFL classrooms (p < 0.05).Therefore, it can be said that male students expect Turkish to be used in EFL classrooms more than female students do.A similar finding was also existent in the study of Adamu (2002), where females were found to have more positive attitudes toward speakers of the target language than males.In this sense, female students might expect their teachers to use the target language more than male students as they have more positive attitudes toward the speaker of the target language, in this case; their teacher.
When the students' expectations of Turkish to be used in EFL classrooms are considered according to their type of high schools they graduated from, no meaningful statistical difference has been found (p > 0.05).61.80% of the students from Science/Anatolian/College high schools, 76.0% of the students from state high schools, and 69.66% of vocational high school students expect Turkish to be used in their classes.Therefore, regardless of the type of high school participants graduated from, it can be said that they share similar opinions toward the inclusion of Turkish in their EFL classrooms.
With regard to the students' level of graduation, it has been explored that there is no statistical difference in students' preferences of Turkish to be included in their EFL classrooms (p > 0.05).68.42% of the students who graduated with a low GPA, 73.75% of the students who graduated with an average level of GPA, and 68.63% of the students with a high level of GPA shared similar opinions about the presence of Turkish in EFL classrooms, and the difference between the views of students is not statistically significant.

Findings of the Correlation between the Students' Preferences of the Use of Turkish Regarding Their
Proficiencya Level and Past Experience with English.(Research Question 3) According to Table 5, 13.70% of the elementary level students mentioned that the use of Turkish has no or very little influence on learning English language.35.62% of the students seem to think it has fairly much influence, and a plurality (50.68%) think that it has quite a lot or a lot influence on learning English.30.96% of the pre-intermediate level students think that Turkish has no or very little influence on learning English language, whereas the ratios of the students who think Turkish has fairly much and quite a lot/a lot influence on learning English language are 26.90% and 42.13, respectively.33.33% of the students studying at intermediate level think that Turkish has no or very little influence on learning English language, whereas 33.33% of them think it has fairly much influence, and 33.33% think it has quite a lot or/a lot influence on English language learning.It has been explored that there is a meaningful statistical difference in students' perspectives towards the contribution of Turkish to learn English (p < 0.05).Students studying at elementary level think that Turkish contributes to English language learning more than pre-intermediate and intermediate students think it has.Unlike the finding in Table 4, which found that there is no difference between the preferences of the students about L1 use and their level of proficiency, here, when the contribution of L1 is the subject matter, students with low level of English think that it has more contribution to learn the target language.This can be interpreted as students with low level of English would like to hear more explanation at the beginning of their new language learning journey to grasp everything their teachers mention in the classroom.They might seek for suggestions on how to learn more effectively, and psychologically, and they may need to feel more secure, as well as requiring to be double checked by their teachers on whether they understand the subject matter.
According to the students' previous schools in which they began to learn English, there is a meaningful statistical difference in perspectives toward the contribution of Turkish to learn English language (p < 0.05).
Students who began to learn English at primary and secondary schools think that Turkish has less contribution to learn English when compared to students who began to learn English at high school or in university who share the view that Turkish has much influence on learning English language.The reason for this could be that the students who started to learn English earlier could not achieve the benefits of such inclusion, so that they are less optimistic about this issue.
22.56% of the students who evaluated their level of English as low before attending university think that Turkish has no or very little influence on learning English, 28.21% of them think it has fairly much influence, and 49.23% of them shared the opinion that Turkish has quite a lot or a lot influence on English language learning.34.48 % of the students who evaluated their level of English as average before attending university believe that Turkish has no or very little influence on learning English language, 33.33% of them believe it has fairly much influence, and 32.18% of them think it has quite a lot or/a lot influence on learning English language.In addition, 38.10% of the students who evaluated their level of English as high revealed that Turkish has no or very little influence on learning English language, 28.57% think it has fairly much influence, and 33.33% of them view Turkish as the key source of learning English language.A meaningful statistical difference is highlighted when students' previous experiences in English are associated with their views on the influence of Turkish on English language learning (p < 0.05).Of the students who evaluated their level of English as low before attending university, 22.56% tend to see Turkish more helpful than the students who had an average or high English level before attending university education.It can be due to the students' low background in English which pushes them to feel insufficient and less secure in the classroom when compared to the other level of students; thus, they perhaps would like to hear more of L1 in their EFL classrooms.
According to the students' knowledge of other languages, there is no significant difference in their views about the influence of the use of Turkish on English language learning.

Students' Needs of the Use of Turkish in EFL Classrooms (Research Question 4)
Table 6 illustrates students' views of the possible reasons about whether Turkish should be used or not in their classrooms.Turkish should be used, while 18.2% disagreed with this.72.9% of the respondents agreed with the item 2 whereas the ratio of the ones who disagree is 27,1%.79.9% of the students agreed that "to explain complex grammar points," Turkish should be used, whereas 20.1% disagreed with the statement.While 64% of the students agreed that Turkish should be used to "explain different concepts or ideas", 36% of them seemed to disagre.69% of the students disagreed with the statement that in order "to give instructions" Turkish should be used, while 31% of the participants agreed.67.3% of the students also seem to agree that teachers should use Turkish while "giving suggestions on how to learn more effectively" while 32.7% of them disagreed with it.The students' ratios who agreed with the statements that Turkish should be used to "summarize the materials covered" and "to joke around the classroom" are also 49.9% and 52.1%, respectively.
With regard to the similar studies conducted by Schweers (1999) and Jan, Li and Lin (2014), this study found parallel results to the occasions in which the students think mother tongue should be used.Jan, Li and Lin (2014), for example, found that to define new vocabulary items, 76.9% of participants requested for Chinese, and in the present study, 81.80% of respondents mentioned that they would like Turkish to be used for this purpose.Jan, Li and Lin (2014) also found out that 72.5% and 75.3% of the participants would like Chinese to be used in EFL classrooms to explain complex grammar points and to explain different concepts or ideas respectively, similar to this study's findings, in which the ratios are revealed as 79.90% and 64% for the items 3 and 4, respectively.The least important factor for the students for the Turkish to be used in the classroom is that "receiving instructions" (31%) and the other less important occasions are "summarizing the materials already covered" and "joking around the classroom" with the percentages of 46.90 and 52.10, respectively.It can be assumed from these findings that students only prefer Turkish to be used in EFL classrooms for acquiring the subject matter.When these findings are compared to Schweers' (1999) findings of the students' perspectives about these occasions for the mother tongue to be used, only for the purpose of explaining difficult concepts, 86.2% of the respondents agreed to use L1, and the other percentages for the other purposes are 20.2% maximum.

Conclusion
This study investigated EFL students' perceptions toward using L1 in their English classrooms and revealed that their educational background and English language proficiency level impacted their perceptions.The instances when the students demanded L1 were also of interest to this study.This study found that the majority of the students preferred L1 to be used in their English classrooms, as also found by others (Dujmovic, 2007;Schweers, 1999;Tang, 2002).In terms of the students' demographic characteristics and their perceptions, it is also found that their genders and level of graduation from previous schools impacted their perceptions.Female students, for example, were found to have more positive perceptions toward the inclusion of L1 in their English classrooms, and the students who graduated from their previous school with a low level tended to require more use of L1 in their classrooms.The reason for this could be that lower level students experience more complexity in comprehending in L2, and that they require more assistance through L1.The study also revealed that the students' past experiences with English and their current English language proficiency level also impacted their perceptions toward the use of L1.The findings address that students with a low background in English and who were at the elementary level thought Turkish had a contribution to EFL learning more than the other level of the students.With regard to the needs of the students that demanded for L1, defining new vocabulary items, explaining complex grammar points, and practicing the use of some phases and expressions were found to be the most striking aspects, as also found in the studies of Burden (2000), Dujmovic, (2007) and Schweers (1999).
When the overall findings of the study are evaluated in relation to the relevant literature, it can be concluded that new findings with regard to student language proficiency level and the demand for L1 use are emerging, which may show us important signs for considering existing English-only policies, perhaps by giving more consideration to what students require.One precauition that should be considered here, however, is that the multicultural classrooms involving non-Turkish students who have different L1's.Although this is a separate debate in the literature (Littlewood & Yu, 2011), it seems to stand as a main block in modifying the current English-only policies of the existing institutions.In terms of the needs where mother tongue is required by the students, issues emerging from the study are also worth considering by the teachers and the policy makers.Such consideration is believed to enhance, and perhaps, better articulate what students need and why, which would possibly lead to a better conceptualized learning between the students and the teachers.

Implications
The findings of this study may serve as a needs analysis for this specific program and programs of this kind in terms of better understanding student perceptions.Literature often suggests that students' perceptions and expectations should stongly be considered by the relevant people who are directly or indirectly in touch with them during their education (Harmer, 2001).In this sense, specific findings of this study should be considered, especially by the policy makers and the teachers.A clear finding of this study is that L1 is needed by lower level of students, and the majority of the students believe that using L1 in their English classrooms can positively contribute to their L2 learning.Thus, the English-only policies employed, in our opinion, should be re-examined in terms of the extent to which they consider student needs as well as to re-question their validity.The case of classrooms which include non-Turkish students, on the other hand, require more research on what to offer to them, as using L1 of a specific group of students in multicultural classrooms is another matter of problem.Further research with the teachers and other research considering other contexts would possibly provide more insights about the situation from a wider perspective, which would probably yield to a better examination and re-thinking of the existing policies.

Limitations
Although the present study gathered data from a representative sample within its own context, richer data would have been collected, which would possibly yield to different results.A potential contribution to the study would have been collectiong data also from the teachers, which would have also shown their views on the use of L1 when teaching L2.

Table 1 .
Demographic characteristics of the students

Table 2 .
Students' current level of proficiency in English and their previous experiences with English

Table 3 .
Students' perceptions toward the inclusion of Turkish in EFL classroom

Table 4 .
Comparison of the students' level of preferences of Turkish to be used in EFL classrooms according to their demographic characteristics

Table 5 .
Comparison of students' preferences of the influence of the use of Turkish on English language learning according to their current proficiency level of English and their previous experiences with English language

Table 6 .
Students' views of the possible occasions which Turkish can be used in EFL classrooms 8% of the students who participated in the study agreed with the statement "To define new vocabulary items,"