Code Switching. The Relationship between personality traits and attitudes toward switching behaviour

Academic Paper 2018 37 Pages

English Language and Literature Studies - Linguistics


Table of Contents


Purpose of the Study

Statement of the Problem

The Description of Participants in Term of PT
Data Collection Instruments

Procedure of the Study

Research Design

Data Analysis

The Relationship between the Students’ Extroversion Sub-scale of PT and their Attitudes toward CS
The Relationship between the Students’ Neuroticism, Conscientiousness, Agreeableness, and Openness Sub-scales of PT and their Attitudes toward CS
The Overall Attitudes of Students toward CS Regarding their Gender

The Relationship between the Five Main Sub-scales of Students’ PT and their Attitudes toward CS
The Relationship between the Students’ PT and their Attitudes toward CS Regarding Gender
Overall Attitudes of Students toward CS Regarding their Gender



Appendix A: Big Five Inventory Questionnaire (John & Srivastava, 1999)

Appendix B: Questionnaire of Students’ Attitudes toward Teachers’ CS (Mingfa, 2011)


The purpose of this study was to investigate the relationship between MA English students’ personality traits (PT) and their attitudes toward university teachers’ code switching (CS) in Urmia, Iran. In addition to that purpose, the correlation between each sub-scale of PT including (extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience) and teachers’ CS was analyzed. Finally, the overall attitudes of MA TEFL English students toward CS behavior were discussed as well. To this end, 150 MA English students (70 males and 80 females) from State and Azad universities of Urmia City participated in this study. Two instruments were used for data collection for this aim. In order to measure students’ PT, the Big Five Inventory designed by (John & Srivastava, 1999) was administered. Secondly, to measure Students’ attitudes toward teachers’ CS, the questionnaire developed by Mingfa Yoa (2011) was used. According to the results, no significant relationship was found between the PT of students and their attitudes toward teachers’ CS. Furthermore, there was not a significant relationship between students' PT and their attitudes toward CS regarding the five sub-scales of PT. The findings of the study indicated that the majority of students had similar attitudes toward CS phenomenon. Their overall attitudes were positive toward teachers’ CS, and the majority of students agreed with CS in EFL settings. As a result, it was revealed that CS is an acceptable behavior in EFL context from MA TEFL students’ perspectives.

Keywords: personality traits, MA students, attitudes, code switching, university teachers, Iran


Nowadays, learning English is considered as a necessity almost all around the world (Hamers & Blanc, 2000), and the ability to communicate in English as an international Language (Sharifian, 2009) or global Language (Crystal, 2003) has turned into one of the significant objectives of people for different reasons. As a result, English is the distinguished language of the contemporary century in the variety of subjects throughout the world (Crystal, 2003). Since English is an international language, there is a big trend among people to learn English as a second language (SL/L2) or foreign language (FL) for a variety of reasons, respectively (Crystal, 2003; McKay, 2002). With regard to English as a second language (ESL) and English as a foreign language (EFL) contexts, Richards and Schmidt (2013) clarified that EFL context refers to a setting in which people learn English in a formal setting without any contact with the target language (TL) outside, while in ESL context, English is routinely used in daily life. Iran is one of the countries where English is learned as a foreign language. In fact, it is not only the foreign language in schools (Rahimi & Eftekhari, 2011) but also the medium of instruction in university (Khajavi & Abbasian, 2011).

In EFL or ESL contexts, there are some situations that speakers may combine some or more segments of first language (L1) and L2 together for several possible reasons. As Hamers and Blanc (2000) stated, the combination of linguistic items between L1 and L2 are the concern of many researchers who deal with the issue of CS phenomenon in bilingualism contexts. According to Trudgill (2000), switching among speakers is a means to handle the situation as they prefer, and to convey a subtle difference in or shade of meaning in their interaction. Ellis (1985) believed that certain difficulties are possible just because of the difference between learners’ L1 and L2 during the process of CS. “Behaviorist learning theory believes that the old habits of the L1 inevitably interfere with the process of learning the new habits of the L2” (Ellis, 1985, as cited in Du, 2016, p. 16). Accordingly, it is predicted that “the similarities between the L1 and L2 facilitate L2 learning, while the differences between the two languages lead to negative transfer and errors” (Ellis, 1985, p. 16) in CS process.

Regarding Matrix Language Frame Model (MLFM) established by Myers-Scotton (1997), it is confirmed that one language is considered to be dominant over another in CS process. As a result, the dominant language is called matrix language (ML), while the second language is called embedded language (EL) mainly due to the insertion of its words in the former language. Accordingly, ML is expected to be English, whereas EL is expected to be Persian language in this study.

CS phenomenon has been defined by different researchers regarding their aims of the study. For instance, CS is defined as the "alternation of two languages within a single discourse, sentence, or constituent" (Poplack, 1980, p. 583). Accordingly, Gardner-Chloros (2009, p. 4) defines CS as the “combination of more than one language or dialect in the same conversation or sentence by bilingual people." Nunan and Carter (2001, p. 275) describes it as a “phenomenon of switching from one language to another in the same discourse”. According to Hamers and Blanc (2000), the concept of CS is mostly related to bilingualism setting in which knowing more than one language and being able to use them in a fluent way is considered, and is generally observed in EFL classroom process (Sert, 2005) or ESL one (Brice, 2000). By considering the classroom context, mainly the EFL one, the insertion of L1 is possible only if all the participants have the same L1. Otherwise, students and their instructors may encounter difficulties in communication, and they may not achieve effective results due to their use of CS in classrooms (Sert, 2005).

Among the several aspects of the study in CS phenomenon, one of the prominent parts is studying about various states of perspectives or attitudes toward CS in language learning contexts. For instance, a variety of viewpoints can be reported in monolingual versus bilingual settings (Ellis, 1985). CS has extensively been examined from many researchers’ perspectives. There are both supporting and opposing views on CS in the classroom context. Accordingly, some EFL instructors completely confirm it as a useful strategy, whereas some have contrastive views toward CS (Sert, 2005). It is verified that teachers can use CS to clarify the unknown materials to learners in EFL settings, and avoid misunderstanding among the students (Sert, 2005). It is also believed that CS is a bilingual tool that paves the way to start conversation and communication (Romaine, 1995). It is confirmed that CS as an unconscious behavior happens purposefully among the communicators to convey their meanings (Myers-Scotton, 2002). Cook (2000) and Stern (1992) also have positive attitudes toward CS phenomenon in L2 context because the ‘English Only’ context may cause lack of motivation among learners, only due to lack of comprehensible input in TL context (Widdowson, 2003). In contrast, some scholars such as Labov (1971) and Hughes, Shauness, and Brice (2006) reported negative attitudes toward CS and considered it as a weakness behavior in pedagogy. Consequently, it is believed that teachers should focus on ‘English only’ or monolingual teaching style (Chaudron, 1998; Ellis, 1984, 1985; Phillipson, 1992). Correspondingly, Krashen (1987) and Littlewood (1981) also support monolingual schooling. Accordingly, Wardhaugh (2010) stated that the monolingualism approach is an “accepted norm in so many parts of the Western world that it is often assumed to be a world-wide phenomenon, to the extent that bilingual and multilingual individuals may appear to be unusual” (p. 96).

Purpose of the Study

As Ellis (1985) mentioned, interest in language study in the classroom setting has grown regularly, and has been motivated by the recognition that successful results may depend on the language used by the teacher and the type of interactions happen in the classroom, whether as a subject lesson or a language lesson. Among the variety of interests in different language study contexts, studying about language attitude as one of the most important facts in applied linguistics can help scholars to judge peoples’ social class, group membership, and some other items by the means of language used in different social and educational settings (Garrett, 2010). As Perloff (2003) mentioned, the evaluation of a language can result in a variety of attitudes toward it. As Baker (1992) confirmed, the term ‘language attitude’ is an umbrella term that consists of several language-based items such as attitudes toward different language dialects, preference, and usage. CS as a typical phenomenon in language attitudes is considered as a practice of switching between two or more versions of languages in a single conversation (Gardner-Chloros, 2009). Accordingly, it is important to have a close look at CS from different perspectives (Gardner-Chloros, 2009). There are a large number of different attitudes about CS among the scholars. Recently, increasing interest in CS has paved the way to conduct a variety of investigations to know different angles of this phenomenon (Gumperz, 1982). Although CS has extensively been examined from educators and researchers’ perspectives, only few analysts have surveyed the students' viewpoints toward CS in Iran, specifically in Urmia City.

Besides the importance of the investigations on language attitudes in educational settings, mainly in EFL contexts (Gardner, Lalonde, & Moorcroft, 1985), evaluation of some other critical items such as personality, affective, motivational, and demographic factors also plays significant role in successful teaching and learning. Among all of them, personality is considered to be the important one in evaluating the attitudes of people (Carrell, Prince, & Astika, 1996). Personality as a significant psychological feature can lead peoples’ behavior in having their own patterns of attitude, feeling, thought, and treatment toward somebody or something (Phares, 1991). According to Corr and Matthews (2009, p. 11), “Personality theory has been persistently concerned with the description of individual differences”. Correspondingly, learning about all individuals’ differences such as evaluating their personality, intelligence, attitude, etc. is considered to be a significant factor in effective learning of SL or FL (Gardner et al., 1985).

Because of the paucity of studies on investigating the features of personality and language attitude, it seems essential to conduct a research on these two constructs in Iran. It is believed that PT (McCrae & Costa, 1987) is an appropriate model for understanding the relationship between personality and various academic behaviors (Poropat, 2009). Appropriately, by considering the importance of personality and attitude in language teaching and learning context, it seems timely to investigate these two important features in CS behavior as well. Accordingly, the aim of the present study is to evaluate the relationship between PT of MA English students and their attitudes toward university teachers’ CS behavior in EFL context. Secondly, the correlation between five main sub-scales of PT and CS behavior is going to be discussed in this study. Then, the overall attitudes of students toward CS will be explored as well. In addition to the overall attitudes of students, the role of another important item such as students’ gender concerning CS behavior is going to be discussed in this study.

Statement of the Problem

CS is a very popular phenomenon that occurs almost in any language class. This situation is very evident in the EFL classes held in Iran. Although the amount of CS may vary from class to class, it is an integral part of every English classroom. The problem is that teachers do not very well know if they should use CS or not and if yes to what extent. By considering CS phenomenon, one of the most controversial issues in English Language Teaching (ELT) has been the question of using L1 in English language classes. By looking at the theoretical background of using L1 in L2 classroom, its usage has been debated for many years. Accordingly, there have been many arguments about its use in L2 learning context (Auerbach, 1993).

Some researchers advocate the use of L1 in L2 context (e.g., Larsen-Freeman, 2000; Schweers, 1999; Tang, 2002), while some others consider it as a wrong methodology (e.g., Harbord, 1992; Mori, 2004; Polio & Duff, 1994). Because of different attitudes and views toward CS (Auerbach, 1993), knowing and evaluating the attitudes of participants, especially learners toward CS is considered as a critical factor in language teaching. As Ellis (1984) stated, “language learners are not only communicators and problem-solvers, but whole persons with hearts, bodies, and minds, with memories, fantasies, loyalties, identities” (p. 39). It means that learners’ attitudes are leading factors in pedagogy (Fakeye, 2010). So, the importance of students’ attitudes toward CS is the first gap that was evaluated in this research.

Although numerous empirical and theoretical studies have been done in the area of CS, there have not been adequate researches in the Iranian bilingual settings (Fakharzadeh & Eslami-Rasekh, 2009). Therefore, the lack of research in CS field in Iran was the second significant gap of this study. Another important gap, especially on attitudes toward CS phenomenon, is the investigation of students’ PT, and includes questions such as how different individuals view the same linguistic phenomenon from their own perspectives. In other words, the present study aims to address and evaluate individual’s attitudes toward CS regarding their PT or individual differences. A variety of items in individual differences including personality, attitude, aptitude, and motivation are considered as the critical characteristics in successful learning of SL or FL (Gardner, 1985). Personality as one of the items in individual differences plays a critical role in making differences in how and what people learn (McCaulley, Natter, & Myers 1980), and it is also considered as an important factor in language learning (Carrell, Prince, & Astika, 1996). The relationship between personality and language learning is mutual, and they can affect and be affected by each other, respectively (Ellis, 1985). Thus, it is firmly believed that Language learning behavior of people can be varied according to their personality (Skehan, 1991). Therefore, the role of PT in students’ behavior and attitudes is so impressive (Dewaele, 2013; Dörnyei, 2005; MacIntyre, Clément, Dörnyei, & Noels, 1998; Öz, 2014; Pourfeiz, 2015). Furthermore, it is confirmed that some psychological features such as attitudes and personality have impressive influence on L2 learning (MacIntyre & Charos, 1996; MacIntyre, Clément, Dörnyei, & Noels, 1998; Öz, 2014). Dornyei (2014) also confirmed that personality is a determinative factor in people’s behavior and learning.

Participants typically are selected and grouped together according to gender, age, and language proficiency but rarely according to their attitudes toward the very linguistic phenomenon that is being studied regarding their PT. This is the third gap of this study. Furthermore, there have been a variety of studies on individual differences in different educational contexts (Contessa, Ciardiello, & Perlman, 2005; Fowler, 2002) but there have been few studies based on the ‘Big Five’ model in L2 context (Ellis, 2008). Therefore, the present study aims to use ‘Big Five Inventory’ model designed by John & Srivastava (1999) to answer this gap as well. Due to a variety of attitudes, this study attempted to investigate the attitudes of MA TEFL students toward CS with regard to their PT, and how these attitudes are influenced by their gender. It is assumed that exploring students’ views on CS concerning their PT will shed more light on the field of language learning and teaching. Ajzen (2005, as cited in Dehbozorgi, 2012, p. 41) believes that “attitude, like personality trait, is a hypothetical construct that is inaccessible to direct observation and must be inferred from measurable responses. These responses must reflect positive or negative evaluations of the attitude object”. So, knowing about students’ attitudes, either negative or positive will reveal the range of their preferences in CS behavior in EFL context.

Regarding the discussion proposed, following research questions are going to be discussed in details.

1. Is there any significant relationship between the students’ extroversion sub-scale of PT and their attitudes toward CS?

2. Is there any significant relationship between the students’ neuroticism sub-scale of PT and their attitudes toward CS?

3. Is there any significant relationship between the students’ conscientiousness sub-scale of PT and their attitudes toward CS?

4. Is there any significant relationship between the students’ agreeableness sub-scale of PT and their attitudes toward CS?

5. Is there any significant relationship between the students’ openness sub-scale of PT and their attitudes toward CS?

6. Are there any positive attitudes toward CS regarding students’ gender?

Based on a comprehensive survey of the literature, no similar study could be found in Iran. Therefore, this study could attract the Iranian scholars’ and educators’ attention to investigate about the relationship between learning and personality. Additionally, a study such as this will show the importance of individuality in educational settings, and help teachers and instructors to meet the educational needs of learners.



In order to conduct the study, the sample or participants were selected from among MA TEFL students at Urmia state University and Urmia Azad University. The sample group included 150 TEFL students ranging from 22 to 40 years in the above-mentioned universities in Urmia City. The participants of this study were 80 female and 70 male MA students in different semesters of the university. From among 80 female participants, 40 students were randomly selected from each university (40 students from Urmia state University and 40 persons from Urmia Azad University).

Again, from 70 male participants, 35 were equally selected through the random sampling from each university, which is considered as a “selection of participants relying totally on random basis and completely on chance in order to include subjects with similar characteristics to the population" (Dörneyi & Taguchi, 2010, p. 61). From two types of random sampling method in quantitative researches including simple and stratified, the simple random sampling type was selected to be used in this study because it is the most effective and acceptable way to select a representative enough sample from the whole population (Mackey & Gass, 2015). Accordingly, in this sampling method, each person has an equal chance to be independently selected in the study. Furthermore, it can enhance the degree of external validity and generalizability to the whole population (Cohen, Manion, & Morrison, 2013).

The Description of Participants in Term of PT

According to self-reported information gathered from 80 female students regarding their PT in five main categories including extroversion, agreeableness, conscientiousness, neuroticism, and openness, the mean scores were reported respectively as 29.38, 24.91, 29.47, 24.33, and 28.22. Accordingly, the standard deviations (SD) of these five categories were 5.24, 3.82, 4.83, 5.15, and 3.04, respectively. According to the mean scores found among the female participants concerning five main sub-scales of PT, conscientiousness sub-scale with mean score equal to 29.47 was the highest one, while neuroticism with mean score equal to 24.33 was the lowest sub-scale. It was revealed that female participants were more conscientious, and at the same time were less neurotic (See Table 1).

Table 1: Dimensions of the Five-Factor Model of PT with Reliability Scores among Female

illustration not visible in this excerpt

Among the 70 male students regarding their PT in five main categories including extroversion, agreeableness, conscientiousness, neuroticism, and openness, the mean scores were reported respectively as 30.44, 24.61, 28.16, 24.62, and 29.46. Accordingly, SD of these five sub-scales concerning the PT was 4.22, 3.61, 5.08, 4.52, and 3.50, respectively. Among the males, extroversion sub-scale with mean score equal to 30.44 is the highest one, while agreeableness with mean score equal to 24.61 is the lowest sub- category. As a result, it was revealed that male participants were more extroverts, and the same times were less agreeable (See Table 2).

Table 2: Dimensions of the Five-Factor Model of PT with Reliability Scores among Males

illustration not visible in this excerpt

Data Collection Instruments

In the present study, two following structured and standard questionnaires with close-ended questions (Cohen et al., 2013) were selected to collect data, mainly because of their facilities in data collection process (Dörneyi, 2003).

1. The Big Five Inventory model developed by John and Srivastava (1999).
2. The questionnaire of evaluating the attitudes of students toward teachers’ CS developed by Mingfa Yoa (2011).

In order to measure the relationship between students’ PT and CS issue, the Big Five Inventory model developed by John and Srivastava (1999) including five main sub-scales with 44 items was used to collect data regarding the students’ PT. The five domains of the ‘Big Five Inventory’ model are: 1) extraversion vs. introversion, 2) agreeableness vs. antagonism, 3) conscientiousness vs. lack of direction, 4) neuroticism vs. emotional stability, and 5) openness vs. closeness to experience. The questionnaire has been designed in Likert-type scale with five choices from 1 to 5. Participants were asked to rate each item by their level of agreement as follow: 5 = strongly agree, 4 = agree a little, 3 = neither agree nor disagree, 2 = disagree a little, 1 = strongly disagree (See Appendix A).

Secondly, to measure Students’ attitudes toward teachers’ CS, the questionnaire developed by Yoa (2011) was conducted among the participants. Attitudes of MA TEFL students to CS in EFL classrooms were analyzed under four categories including 1) attitudes to teachers’ persona in using CS (questions 1 to 5), 2) attitudes to CS in subject access (questions 6 to 10), 3) attitudes to CS in classroom management (questions 11 to 15), and 4) attitudes to CS for interpersonal relations (questions 16 to 20). Again, in this Likert-type questionnaire, the participants were asked to answer their level of agreement from 1 to 5 as follow: 5=strongly agree; 4=agree; 3=not sure; 2=disagree; 1=strongly disagree (See Appendix B). Furthermore, some demographic information of the participants as the moderator variables which refers to independent variables were also gathered in case they may be effective in their attitudes toward CS phenomenon, or maybe change a correlation of the variables (Mackey & Gass, 2015). As a result, students were asked to complete their demographic data including their gender and the type of their universities (State or Azad).

The important factor in applying a questionnaire for collecting significantly acceptable and statistical data is its reliability. Accordingly, the reliability of a questionnaire refers to “the consistency of the scores obtained—how consistent they are for each individual from one administration of an instrument to another and from one set of items to another” (Fraenkel & Wallen, 2008, p. 154). Appropriately, the two selected questionnaires concerning the ‘PT and CS phenomenon’ were completely standard and qualified ones. Their reliabilities has been confirmed by scholars, and are equal to 0.83 and 0.71 Cronbach Alpha coefficient, respectively. Regarding their reliabilities, it is revealed that both of the questionnaires are reliable enough for collecting quantitative research data in the present study. Finally, we asked from students to state their overall attitudes toward CS behavior in Iranian EFL context.

Procedure of the Study

All participants were selected to participate in the study through simple random sampling model. At the beginning, the necessary permissions were taken from all course instructors to give the questionnaires to the students in their classes. All the students were also asked for permission orally to fill out the questionnaires during their courses. A brief demographic part was added to the questionnaires to gather certain required data about the participants. They included their gender, age, and type of their university. The names of participants were not mentioned in order to keep their personal information. Data collection period for this study lasted approximately one month from September to October in 2016. Prior to initiating the main study, and in order to evaluate the feasibility of conducting above-mentioned instruments, the researcher performed a pilot study on both questionnaires (PT and CS) to know whether they serve the research goals (Cohen et al., 2013). To do so, a pilot study conducted among the 20 students (10 males and 10 females). Subsequently, the efficiency of both questionnaires were confirmed as equal to 0.85 and 0.78 Cronbach alpha scores, respectively which represented a high range of validity. Finally, the questionnaires were distributed to students during the last 20 minutes of classes in two universities. Consequently, the formats of questions in both of questionnaires were closed-ended ones in order to facilitate the quantification and analysis of numerical date (Wei & Moyer, 2009).

Research Design

Along with evaluating the attitudes of participants (MA TEFL Students) about teachers’ CS behavior in EFL context, the present study is a form of quantitative and ex post facto correlational research design evaluating the correlation between the relationship between each sub-scale of PT (extraversion, agreeableness, conscientiousness, neuroticism, openness to experience) and CS issue. This is an ex post facto design because a pre-existing characteristic (personality) was used to form the groups. The study is ex post facto since no treatment is given to the participants and no change is made to them. The study has a quantitative design since the inferential statistics are utilized in order to find out the relationship between PT and CS. The main two variables of present study are namely, PT and CS which are both non-causal. In other words, the PT can be considered as the predictor, and the CS as the predicted or the response variable (also called target variable). Finally, the overall attitudes of students toward CS regarding their gender and the type of their universities were analyzed, qualitatively.

Data Analysis

As a quantitative and correlational research design, all the data were gathered through two standard questionnaires regarding the participants’ PT and their attitudes about CS issue. To analyze the obtained data, all data gathered from the questionnaires were entered into SPSS software, version 21 to evaluate the relationship between the each sub-scales of PT and students’ attitudes toward CS issue regarding their gender. In order to examine the assumption of linearity in the data, Loess line is used (Everitt & Dunn, 2001). This assumption says that the relationship between the two variables should approximately be a straight line. To examine this assumption, if the regression and Loess line match, this provides confidence in the assumption of a linear trend to the data. (Larson-Hall, 2010) As a result, no linear correlation was reported among the variables. Using Kolmogorov-Smirnov (K-S) to determine the normal or non-normal distribution of variables, it revealed the lack of normal distribution among the variables. Therefore, the non-parametric test of Spearman correlation instead of the parametric test of Pearson correlation was used to evaluate the gathered data.


The Relationship between the Students’ Extroversion Sub-scale of PT and their Attitudes toward CS

illustration not visible in this excerpt

Figure 1. The relationship between the students’ extroversion sub-scale of PT and their attitudes toward CS

illustration not visible in this excerpt

Figure 2. The relationship between the students’ extroversion sub-scale of PT and their attitudes toward CS with regard to gender

Table 3: Test of Normality for Extroversion Sub-scale of PT

illustration not visible in this excerpt

As Table 3 reveals, the scores on neither extroversion, nor CS were normally distributed (sig < 0.05); therefore, the researcher had to apply non-parametric statistics. Accordingly, the non-parametric test of Spearman was applied instead of Pearson test, which is a parametric one (See Table 4).



ISBN (eBook)
ISBN (Book)
File size
680 KB
Catalog Number
Institution / College
Urmia University – International Students Admission Department
code switching relationship




Title: Code Switching. The Relationship between personality traits and attitudes toward switching behaviour