EFL Teachers' and Learners' Perceptions of Code-switching: The Role of Learners' L2 Proficiency Levels

,

Code-switching (CS) has been a significant linguistic phenomenon in bilingualism research.For instance, CS is considered a predictable linguistic behaviour in contexts such as English as a second language (ESL) or foreign language (EFL).Moreover, we must review some definitions of CS to gain a complete understanding of this linguistic phenomenon.For instance, Gumperz (1982) found that CS is "the juxtaposition within the same speech exchange of passages of speech belonging to two different grammatical systems or subsystems" (p.59).Furthermore, Milroy (1995) concluded that CS is the alternative use by bilinguals of two or more languages in the same conversation.Thus, CS means alternating from one language to another in the same discourse (Nunan & Carter, 2001).Nguyen (2014) suggested that CS was viewed as showing a lack of linguistic competence, since bilinguals could not keep the two languages apart properly.
In recent years, CS has been viewed as a meaningful linguistic phenomenon in contexts such as ESL or EFL, and has attracted considerable attention (Alkhudair, 2019).Learners often consider CS to be a direct approach to acquiring a foreign language because of the sense of relaxation it produces by reducing their affective filter (Krashen, 1982).Currently, CS is no longer viewed as deficient language behaviour.Instead, it is considered a systematic and grammatically structured linguistic phenomenon (Nguyen, 2014).

Purpose and Significance
CS is a common linguistic phenomenon in EFL contexts in Saudi Arabia, where learners alternate between the English and Arabic languages for various reasons.This research investigated teachers' and learners' perceptions of CS in EFL contexts.Many studies have been conducted on teachers' perceptions of using one's first language (L1) in EFL classrooms.Also, many previous studies discussed learners' perceptions of CS while neglecting the role of learners' second language (L2) proficiency levels.Therefore, this study also investigated the relationship between learners' perceptions of CS and their L2 proficiency levels.
This study added to the literature on teaching English as a foreign language by investigating teachers' and learners' perceptions of CS, particularly in the EFL contexts in Saudi Arabia.Moreover, it offered some critical insights into the role of learners' L2 proficiency levels by investigating the relationship between learners' perceptions of the use of CS and their L2 proficiency levels.

Research Questions
For the purpose of the study, EFL is English as a foreign language, while L1 means learners' first language or mother tongue (Saville-Troike & Barto, 2017).The research aimed to address the following questions: (1) What are teachers' perceptions of CS in EFL classrooms?
(2) What are learners' perceptions of CS in EFL classrooms?
(3) Is there a correlation between learners' L2 proficiency levels and their perceptions of the use of L1 in EFL classrooms?

Literature Review
A considerable amount of literature has focused on teachers' and learners' perceptions of CS in EFL classrooms.For instance, previous studies have confirmed that teachers and learners showed positive attitudes and acceptance of CS in EFL classrooms (e.g., AbdelMagid & Mugaddam, 2013; Al Adnani & Elyas, 2016; Al-Amir, 2017; Al-Balawi, 2016; Alkhudair, 2019; Al-Nofaie, 2010; Yao, 2011).Al-Nofaie (2010) argued that CS is an unavoidable phenomenon.She examined the perceptions of Saudi teachers and students regarding the use of CS in a Saudi intermediate school for females.Several studies used only one or two data collection instruments to investigate teachers' and students' perceptions of CS in EFL classrooms.However, Al-Nofaie (2010) investigated this phenomenon using questionnaires, interviews, and four observations of one classroom.As a result, most teachers who participated in her study showed a positive attitude towards CS.Also, teachers were forced to use their L1 to explain complex tasks to weak students.Therefore, low-level students were a strong stimulus for the teachers to code-switch.CS is considered a linguistic phenomenon that indicates the use of two languages within the same conversation.Yao (2011) investigated and analysed teachers' and students' attitudes towards CS in EFL classrooms in a local secondary school in China.The target audience consisted of 52 English teachers and 100 students.A quantitative approach was applied in this research using the Statistical Package for the Social Sciences (SPSS) to analyse the results of a Likert-type questionnaire.The questionnaire aimed to investigate attitudes about teachers' CS in EFL classrooms and determine when, how, and why teachers code-switched and for what functions.
Moreover, each teacher was interviewed to elicit information about their backgrounds, training, and teaching philosophy.In addition, the teachers were asked about their teaching views, especially regarding the role of CS in EFL classrooms.The findings indicated that both students and teachers had similar opinions on most question items.Also, such consistency showed that teachers and students have a similar positive attitude towards teachers' CS in EFL classrooms.
AbdelMagid and Mugaddam (2013) found that CS by EFL teachers was unavoidable and that L1 was crucial as a necessary part of language teaching and learning.However, this is inconsistent with the study by Al-Adnani and Elyas (2016).Al-Adnani and Elyas (2016) aimed to investigate the purposes and effects of teachers' code-switching on learners' achievement of speaking skills in the EFL context and students' attitudes towards CS.In Saudi Arabia, there was little research (primarily qualitative) on the impact of CS on students' achievement in speaking L2, and there was not much focus on this topic applied to Saudi students to investigate CS in EFL contexts.A quantitative design was used to collect and analyse students' questionnaires and final speaking grades.The primary tools were observation, questionnaire, and students' final speaking grades, which served as a supporting tool.In addition, the researchers suggested that CS should not be applied in EFL classrooms.Moreover, the results showed a positive attitude towards teachers' CS as a learning facilitator in the classroom.Also, there was a clear difference between the learners' grades whose teachers used CS and those who did not.Furthermore, 80% of EFL teachers in Saudi Arabia used L1 (Arabic).Teachers showed positive attitudes towards using L1, such as checking comprehension, explaining new vocabulary, and introducing grammar (Al-Balawi, 2016).Additionally, previous studies have primarily focused on the correlation between teachers' perceptions of CS and L2 proficiency levels.For instance, Al-Amir (2017) investigated teachers' perceptions while using L1 in EFL contexts and discussed the correlation between teachers' perceptions about using L1 in EFL contexts and teachers' English language proficiency levels.Further, the researcher used a quantitative approach and elicited data through a questionnaire adapted from Manara's (2007) study and analysed using SPSS.However, this research investigated EFL teachers' perceptions and excluded students' perceptions.She also indicated that most teachers agree to use L1 in their EFL classrooms.However, there is no correlation between teachers' English language proficiency level and their perceptions of L1 use.
In contrast, Al-Amir (2017) did not consider the relationship between students' perceptions of CS in EFL classrooms and their L2 proficiency levels.In another study, Alkhudair (2019) used a quantitative approach that included two structured questionnaires for female professors (n = 20) and students (n = 60) to collect and analyse data via SPSS.The researcher designed the questionnaire based on the literature reviewed in previous studies.Additionally, the results revealed a range of positive attitudes towards CS by professors and students.The students showed considerable acceptance and considered CS to be a learning facilitator.However, Alkhudair (2019) applied only one instrument (i.e., a questionnaire), and the questionnaires included only two scales, which restricted participants' responses.In the researcher's opinion, such a questionnaire would be more effective if five scales or options were added, such as (1) strongly disagree, (2) disagree, (3) neither agree nor disagree, (4) agree, ( 5) and strongly agree, to elicit teachers' and students' perceptions of CS.
Studies on CS focused on teachers rather than students.Therefore, one of the research gaps that this study filled was addressing CS functions by both teachers and students.Hussein et al. (2020) aimed to identify CS's primary functions and roles among Jordan's EFL high school teachers and students.They applied a mixed method that included observing two EFL teachers in 12 classes and distributing an Arabic Likert-type questionnaire to 330 high school students.The survey consisted of two parts.The first included four questions on age, gender, class, and school.The second part included 15 items to elicit data on students' perceptions of their CS functions and their effect on classroom processes and procedures.Like the findings obtained by other researchers, the results of the classroom observations indicated seven functions of teachers' CS.The teachers encouraged and guided the students by inserting Arabic words and phrases as an affective function.Some teachers repeated sentences or instructions in Arabic to attract students' attention, and they translated words and tasks into Arabic.
Such research was useful for this study because it showed one peculiar finding: teachers code-switched because of inadequate mastery of the target language.In addition, two data collection instruments were used in this study, making it an authentic and valuable resource.This research addressed a reasonable number of students.However, it only focused on two EFL teachers during the observations.Moreover, the number of practical classes was insufficient.

Research Methodology
This section outlines the research design, participants, data collection methods, and ethical practices during the research.

Research Design
A quantitative approach was utilised in this research to collect data from 40 Saudi EFL teachers and 50 Saudi undergraduate EFL learners to investigate their perceptions while using CS in EFL classrooms.

Subjects
Because this study investigated Saudi teachers' and learners' perceptions of CS in EFL classrooms, 40 Saudi EFL teachers and 50 Saudi undergraduate EFL students from Saudi universities participated in the research.The teachers and students were selected randomly from Saudi universities.

Data Collection Methods
This study used modified two Likert-type questionnaires adopted from Alkhudair (2019) to record teachers' and students' perceptions of CS in EFL classrooms.The first questionnaire elicited teachers' perceptions of CS in EFL classrooms.In contrast, the second questionnaire elicited students' perceptions of CS in EFL classrooms.The second questionnaire was translated into learners' L1 (Arabic).In addition, 32 out of 50 learners took the Oxford Online Placement Test to record the effects of L1 in EFL classrooms.

Ethics
Participation in this study was voluntary.After approval from the research supervisor and the ethical committee at Taif University, the researcher wrote the questionnaires via Google Forms.In addition, the researcher developed an informed consent form for the participants to fill out before answering the questionnaires.The participants were informed that they could withdraw at any time, and their survey responses were anonymous.

Procedure for Data Collection
Alkhudair (2019) developed a questionnaire based on previously reviewed studies with only two responses or options.This study included five responses or options in the questionnaire to elicit more reliable and accurate data from teachers (see Appendix A) and learners (see Appendix B).Due to the restrictions caused by Covid-19, the questionnaires were sent to the participants via email, Twitter, and WhatsApp.SPSS was used to determine the frequencies, percentages, and mean scores.Learners' L2 language proficiency was measured using the Oxford Online Placement Test.Such online placement tests enhanced the reliability and accuracy of the results.The tests were automatically graded and marked with scores from 0 to 120 and were eventually sent to the researcher.Then, the researcher divided the learners according to their test scores.
The test results included the Common European Framework of References (CEFR) level (Pre-A1 to C2) and the time the participant took to complete the test.SPSS was used to calculate the Pearson Correlation Coefficient and determine if there was a correlation between learners' L2 proficiency levels and their perceptions of using L1 in EFL classrooms.Two variables were applied: 1) learners' perceptions of CS in EFL classrooms and 2) the results of learners' Oxford Online Placement Tests.The purpose was to determine the relationship between the two variables.For example, SPSS was used to calculate the Pearson coefficient.The formulas returned a value between -1 and 1, where result (1) indicated a strong positive relationship, and minus one (-1) indicated a strong negative relationship.A zero (0) indicated no relationship.

Findings of the Study
To record EFL teachers' perceptions of CS in EFL classrooms, 40 teachers answered a Likert-type questionnaire adopted from Alkudair (2019).The teachers were asked, "Do you agree with the CS strategy professors use in the classroom?"Table 2 shows the responses, indicating that 85.6% of teachers agreed and 14.4% disagreed.Furthermore, 85.6% believed that CS enhances the learning of the English language, while 14.4% disagreed.Moreover, 64.4% of the teachers preferred to communicate with students in English and Arabic to make them feel confident in exploring their ideas.Also, 69.6% of the teachers agreed that using a mixture of languages leads to weakness in students' English.In contrast, however, 74.4% of the teachers agreed to prevent students from switching to their L1 because it would ruin students' English vocabulary.Furthermore, only 63.6% of the teachers agreed to allow their students to switch to L1 (Arabic) when they did not know English's equivalent words.The following Table 1 summarises the results of teachers' responses.To investigate EFL learners' perceptions of CS in EFL classrooms, 50 learners from Saudi universities answered a Likert-type questionnaire adopted from Alkudair (2019).The learners were asked a series of questions regarding CS.As shown in Table 4, 77.6% of the learners agreed with professors using their mother tongue and English in the classroom.Additionally, 85.2% of the learners agreed that CS is a valuable tool that helps them understand complicated matters easily.However, 88.4% of the learners agreed that when the professor switches from English to Arabic, the lesson becomes more complicated.Moreover, 76% of the learners agreed that when the professor switches from English to Arabic, it affects their English language proficiency.
The learners agreed that CS enhances their English language learning (80.8%).Furthermore, 84.4% of the learners agreed that when the professor switches from English to Arabic, it helps them participate and become more active in class.They said that CS by professors saves time and effort in looking for a specific meaning.In addition, 82.8% of the learners agreed that using CS in the classroom ensures that they do not lose any ideas or information.However, 84.4% of the learners agreed that when the professor switches from English to Arabic, they consider him a non-professional professor since he violates English classroom rules.Moreover, 83.6% of the learners agreed that when the professor switches from English to Arabic, they feel that they are not proficient in English.Furthermore, SPSS was used to calculate the Pearson correlation coefficient to investigate if there was a correlation between learners' L2 proficiency levels and their perceptions of the use of L1 in EFL classrooms.As a result, there were significant positive and negative correlations between learners' perceptions of CS and their test scores depending on the statements.For instance, a positive correlation was found where learners with high test scores felt that the professors' CS made them insufficient to learn English.However, learners with high test scores did not favour CS to enhance their understanding and learning of English.Therefore, this indicated a significant negative correlation.
Similarly, students who preferred that professors use CS scored low on the test.The students most likely to participate in class due to CS also scored low on the test.Finally, most students who believed that CS improved their understanding of concepts in class also scored low on the test.When the professor switches from English to Arabic, it helps me to participate and be more active in the class.Use of code-switching by professors saves my time and effort in looking for specific meaning.* Correlation is significant at the 0.05 level (2-tailed).

Discussion of Results
CS is a predictable linguistic phenomenon in EFL contexts in Saudi Arabia.For instance, English language teachers can predict that some learners will switch codes from English (L2) to Arabic (L1), either due to their low-level L2 proficiency or learners' capacity to switch between English and Arabic subconsciously.Previous studies have confirmed that most teachers and learners show positive attitudes and tremendous acceptance of the use of CS in EFL classrooms (e.g., AbdelMagid & Mugaddam, 2013; Al Adnani & Elyas, 2016; Al-Amir, 2017; Al-Balawi, 2016; Alkhudair, 2019; Al-Nofaie, 2010; Yao, 2011).This study's findings are consistent with those of previous studies mentioned in the literature review.For instance, the teachers' and learners' perceptions of CS were positive since most agreed with the assessment questions.In addition, there was significantly strong agreement among the teachers and learners on CS strategies that enhanced English language learning.This finding suggests that teachers consider CS a useful pedagogical tool to improve students' English language learning.Also, this finding is consistent with Al Adnani and Elyas (2016), who found a positive attitude towards teachers' CS as learning facilitators in the classroom.In addition, the previous finding is consistent with Al-Balawi (2016).She reported how 80% of the EFL teachers in Saudi Arabia showed positive attitudes towards L1 and utilised CS to check comprehension, explain new vocabulary, and introduce grammar.However, the teachers indicated that CS prevents students from thinking and exploring independently.Moreover, they agreed that excessive use of CS leads to weakness in learners' L2 proficiency.As a result, the application of CS in EFL contexts must be controlled and guided by teachers.CS might be used in certain situations when necessary.
As Nguyen (2014) reported, learners' results showed that CS is a systematic and grammatically structured linguistic process.For example, 85.2% of the learners agreed that using an Arabic equivalent to explain a difficult English word was a valuable tool for comprehension.However, most students agreed that CS complicated the lesson when the professor switched from English to Arabic.Furthermore, most students agreed that when the professor switched from English to Arabic, it helped them participate and become more active in the class.Also, using CS by professors saved learners' time and effort in looking for specific meanings.This finding is consistent with Alkhudair (2019), who showed a range of positive attitudes towards using CS by professors and students.
Furthermore, the students showed considerable acceptance and considered CS to be a learning facilitator.As the learners indicated in the questionnaire, CS might be used in certain situations, such as explaining complicated matters.
These findings have implications for EFL teachers' use of controlled CS in the classrooms.For instance, they must use CS to explain complicated matters only when it is hard for learners to understand them in English.The findings of this study suggest that the learners do not prefer teachers' random use of CS.In contrast, they prefer the controlled use of CS to explain complicated matters and save learners' efforts in seeking specific meaning.Most students agreed that CS affected their English language proficiency in a negative way when the professor switched from English to Arabic.This finding suggests a relationship between learners' CS and their L2 proficiency levels.
In contrast, most students agreed that they considered the professor a non-professional when he switched from English to Arabic since he violated English classroom rules.Moreover, most students agreed that when the professor switched from English to Arabic, they felt that they were not proficient in English.
Regarding the correlation between EFL learners' perceptions of CS and their L2 proficiency levels, this study's results indicated significant positive and negative correlations between learners' perceptions and their Oxford Online Placement Test scores.For instance, a positive correlation was found when learners with high test scores believed the professors' CS made them less proficient in English.This finding suggests that EFL teachers need to minimise the use of CS strategies in the classrooms to improve learners' L2 proficiency.However, a negative correlation was found when learners who scored high on the test did not favour the use of CS to enhance their understanding and learning of English.This finding indicates that EFL learners find it useless to use L1 as a facilitator for learning the English language.In addition, this indicates that learners with higher L2 proficiency levels do not prefer CS in EFL classrooms.
Students who preferred professors' use of CS scored low on the test.Moreover, the students most likely to participate in class due to CS also scored low on the test.These findings are consistent with those of Al-Nofaie (2010).She found that low-level learners were a strong stimulus for teachers to code-switch.In contrast to earlier findings regarding learners with higher L2 proficiency levels, low-level learners preferred CS in EFL classrooms.Moreover, this study's results showed that a low L2 proficiency level stimulates learners to code-switch to Arabic, consistent with the previous findings of Al-Nofaie (2010).However, CS of high L2 proficiency level learners is considered a natural linguistic production of the mother tongue (Arabic).Therefore, such use of L1 while speaking in L2 is predictable and common since both languages exist in high L2 proficiency learners' linguistic competence.These findings suggest that the role of L1 is crucial for low-level learners and that their use of L1 is unavoidable.
Consequently, EFL teachers' use of CS in classrooms is inevitable.Most students who believed that CS improved their understanding of concepts in class also scored low on the test.Such positive and negative correlations do not suggest that one variable causes the other but tell us that two variables change together.

Conclusion and Recommendations
Based on the findings of this research, EFL teachers and learners show positive attitudes towards using CS in EFL classrooms.Moreover, the role of L1 is crucial for EFL learners with low L2 proficiency levels.As a result, EFL teachers' use of CS in EFL classrooms is inevitable.Based on the Pearson Coefficient Correlation results, we conclude that there are positive and negative correlations between EFL learners' perceptions of CS and their L2 proficiency levels.For instance, the correlation results indicate that learners with higher L2 proficiency levels do not prefer CS in EFL classrooms.
In contrast, learners with low L2 proficiency levels prefer CS in EFL classrooms.This study confirms that EFL teachers and learners embrace CS as a valuable teaching tool.However, they indicated that CS should be used sparingly to explain complex content and should not override the importance of English fluency in the classroom.
This study contributes to the existing body of teaching EFL by investigating teachers' and learners' perceptions of CS, particularly in EFL contexts in Saudi Arabia.Further, it offers some crucial insights into the role of learners' L2 proficiency levels by investigating the relationship between learners' perceptions of the use of CS and their L2 proficiency levels.
The limitations of this study were the small sample sizes of both groups.Therefore, the results may not be applicable to a larger target population of teachers and learners.Furthermore, only 32 of the 50 participant learners took the Oxford Online Placement Test due to the short time to conduct this research.Therefore, future research may involve larger sample sizes for both groups to develop more valid and reliable results.Finally, the research was limited to questionnaires as the only survey instrument.Future studies on this topic should elicit data by conducting structured interviews with participants.
Using code-switching in the classroom ensures me that I am not losing any ideas or information.is significant at the 0.01 level (2-tailed).

Table 1 .
Descriptive Statistics of Overall Teachers' Perceptions Ratings The overall mean and standard deviation of the teachers' responses were 3.68 and 1.3, respectively, indicating that EFL teachers agreed (73.6%) that CS was practical, acceptable, and led to improved expression without compromising the importance of the English language.

Table 2 .
Descriptive Statistics for Teachers' Perceptions Ratings by Item

Table 3 .
Descriptive Statistics of Overall Leaners' Perception Ratings

Table 4 .
Descriptive Statistics for Learners' Perception Ratings by ItemThe Oxford test scores were analysed for 32 students.The highest score on the test was 92, while the lowest score was 0.10.The students scored 29.46 on average, with a standard deviation of 26.12 marks.

Table 6 .
Correlation Between Learners' Perceptions and Oxford Online Placement Test Scores