Document Type : Original Article
Authors
Shiraz University
Abstract
Keywords
1. Introduction
1.1 Identity
Holland, Lachicotte, Skinner, and Cain (1998) conceptualize identity as lived and negotiated in; it is formed through participation in socially formed and culturally constructed worlds. However, one does not construct whatever identity she wishes, because she still must depend on cultural resources and recognition of others to be identified as a certain kind of person (Gee, 2001). Therefore, there is a range of possible identities available in a particular context, but these possibilities are also culturally and socially bounded. Further, identity is not a fixed and static construct; rather, it is ever changing and evolving. Multiple, often contradictory, identities are informed by self-understanding, which develops across time and space (Holland et al., 1998). In addition, the importance of language in the process of identity formation cannot be overlooked. As Weedon (1997) said, “language is the place where actual and possible forms of social organization and their likely social and political consequences are defined and contested. Yet it is also the place where our sense of ourselves, our subjectivity, is constructed” (p. 21). This point is particularly relevant in a language classroom while students learn to construct and negotiate their multiple identities in a new language.
1.1.1 Identity-based research in ESL/EFL
After the field of Second Language Acquisition (SLA) and Applied Linguistics took a social turn (Block, 2007; Johnson, 2006) and made the sociocultural perspective its primary point of investigation (Swain & Deters, 2007), SLA researchers became more interested in the link between identity (class, culture, race, gender, sexuality, etc.) and language learning (Bown, 2009; Flowerdew & Miller, 2008; Hansen & Liu, 1997; Morita, 2009; Ushioda, 2006). In the following section, a review on the research literature that addresses the interplay between identity and language learning from various sociocultural perspectives is given.
Identity orientations refer to the relative importance that individuals place on various identity attributes or characteristics when constructing their self-definitions (Cheek, 1989).
Regarding the existing paradigms on identity, there are two different models. Based on the first paradigm named as the Western, monocultural, cognitive view, the individual is considered as an independent, free and self-contained person. With respect to the second paradigm, the society plays an important role in the construction of the identity of the individuals. This paradigm is known as the constructivism and considers the human identity to be partly cognitive or individualistic and partly social (Brown, 2007). Two theories which support the second paradigm are social identity theory and cross-cultural theory. According to the social identity theory, identity is bipolar: social and personal (Tajfel, 1998). Regarding the cross-cultural theory, self or identity represents two ideas, namely, independent-self and interdependent-self (Markus & Kitayama, 1991). Cheek, Smith and Tropp (2002), benefiting from the mentioned paradigms and models, developed a questionnaire consisting of four aspects of identity, namely, personal, social, collective and relational.
1.1.1.1 Personal identity
People may prefer different identity orientations at different stages, with the specific preferences rooted in the way the focal defines him/herself. Moreover, a personal identity orientation is triggered and reflected when self is defined as a unique being. The individual self, also known as the personal, private or idiocentric, involves the formation of oneself as independent and unique, possessing a definite and clear boundary that distinguishes one from others.
1.1.1.2 Relational identity
The relational self encapsulates definitions of the self based on bonds with significant members, the quality of relationships, interpersonal roles, and traits that an individual shares with significant others. This view of the self places emphasis on interpersonal relatedness, intimacy and interdependence, and would commonly be connected with a psychological orientation to belong.
1.1.1.3 Social identity
The concept of social identity has significant potential to clarify and predict a diverse range of social behaviors. This potential originates from the idea that individuals’ social identity associates them with social groups and roles that influence their behaviors and actions. Social identity is described as a concept that is triggered and developed in a continuous exchange process among the individual and the group, which contains enduring central and tangential components.
1.1.1.4 Collective identity
Collective self, also known as social or socio-centric self, refers to self definitions developed from being a member of groups or social categories. Such views of the self goes hand-in-hand with the tendency to place emphasis on group association, in-group norms, roles and status as defined by collectives.
The definition of identity the researchers use is as follows. Identities are lived and negotiated in and through participation in situated sociocultural contexts. Identities are socially and culturally bounded and rely on language for individuals to construct a sense of who they are (Holland et al., 1998; Weedon, 1997). Building on such a view of identity from both sociocultural and poststructuralist perspectives to study college students who are enrolled in an EFL class in Iran, it might be said that these two different approaches, although coming from different epistemological traditions, complement each other in theorizing the relationship between identity and language learning; the former looks into the social nature of identity and learning while the latter stresses the multiplicity and complexity of identity. In the present study, this definition helps us better understand how Iranian college EFL students’ identities intersect with English language learning and achievement, and how their past experiences come into play in an English language classroom. Lastly, due to the importance of the two key concepts, namely, identity and language achievement on the one hand and the fact that in the literature no study in the academic context in Iran deals with the relationship between these two themes, the present study aims at discovering the extent of the relationship between them as well as the impact of the demographic factors on the two variables. As such, first the studies conducted on aspects of identity are discussed and then research done on the language achievement and proficiency are introduced.
1.2.1 Studies on aspects of identity
From the 1980s to the present, many studies have been conducted on the impact of aspects of identity on and their relationship with other variables in different social educational contexts, sometimes with controversial findings and results. For the first time, Cheek and Busch (1982) found social identity to be positively correlated with measures of public self-consciousness, sociability, and institutional and altruistic selves. Personal identity was positively correlated with private self-consciousness, need for uniqueness and achievement-oriented self.
Cheek and Hogan (1983), also, found that personal identity correlated significantly and meaningfully with guilt feelings than with shame, whereas social identity correlated significantly and meaningfully with shame than with guilt. Frantz (1985) found that although social identity was negatively correlated with independence of judgment, personal identity was positively correlated with it.The combination of the two identity scales significantly predicted independence of judgment scores, demonstrating a pattern similar to that of Hogan and Cheek (1983). While personal and social identities were both positively correlated with private and public self-consciousness, Penner and Wymer (1983) found that personal identity showed a stronger relationship with private self-consciousness and social identity was more strongly correlated with private self-consciousness. Social identity was also positively correlated with self-monitoring. Replicating Cheek's (1982) use of a composite of personal identity and private self-consciousness, Wymer and Penner (1985) found high scores on both inner-directedness and social skills to predict higher levels of congruence between self- and peer-ratings. Respondents low on social skills and high on inner-directedness had higher levels of attitude-behavior congruence. Furthermore, respondents high on personal identity also tended to have higher attitude-behavior congruence than those low on personal identity. Besides, Leary et al. (1986) found respondents high in personal identity to consider personally-relevant job characteristics as significantly more important than respondents low in personal identity, while respondents high in social identity rated socially-relevant job characteristics significantly higher than those low in social identity. Moreover, Barnes et al. (1988) found that respondents high in personal identity were more apprehensive about self-evaluation and less apprehensive about social evaluation than were respondents low in personal identity. Furthermore, respondents high in social identity grew more apprehensive about social evaluation than did low social identity respondents.
Lamphere and Leary (1990) reported that personal identity had significantly positive correlations with private self-consciousness and a new endogenic orientation scale whereas social identity had significant positive correlations with public self-consciousness, the self-monitoring scale, and a new exogenic orientation scale. In addition, Luhtanen and Crocker (1992) found that social and collective identities were both positively correlated with the total score and identity subscale of their new collective self-esteem scale. Cheek et al. (1994) reported in their study that while there were no significant differences between Asian-Americans and European-Americans with respect to personal and social identities, Asian-Americans were significantly higher in collective identity than European-Americans were.
Leibowitz et al. (2005) conducted a study on the relationship among identity, language and teaching and learning at a higher education institution in the Western Cape. It showed how language, both as proficiency in the dominant medium of communication and as discourse, is a key component of identity in a higher education institution. The interviews demonstrated how, according to lecturers and students, language and discourse function as primary influences on individuals’ acculturation and integration into the academic community. According to the interviewees, language as a marker of identity is interwoven with other aspects of identity. It is both a resource and a source of identification and affiliation. The research demonstrated that dialogue and self-reflection can be facilitated via research into identity, teaching and learning, and that this can be beneficial for both the interviewees and the research team.
Razmjoo (2010) conducting a study that demonstrated no significant relationship between language achievement and the aspects of identity; that is, none of the identity aspects is a predicting variable for language achievement in the Iranian context.
1.2.2 Studies on language achievement/proficiency
According to Richards and Schmidt (2002), a language achievement test is designed to measure how much of a language learners successfully learned with specific reference to a particular course whereas a language proficiency test is defined as the degree of skill with which a person can use a language, such as how well a person can read, write, speak, or understand language and the test is not linked to any particular course of instruction. A large number of domestic and overseas studies are done on language proficiency/achievement performance and the factors affecting learners’ language proficiency/achievement.
Yashima and Zenuk-Nishide (2008) carried out a study on the area of proficiency in order to determine the effects of learning contexts on language proficiency. The results showed that the study-abroad participants’ proficiency gain was larger than those who did not have natural exposure to the language; that is, learners with a larger amount of exposure to content-based instruction have an advantage in the development of language proficiency over those who have less exposure. Their study also revealed that motivation led to proficiency in L2. Yamashita (2008) examined the effect of extensive reading on different aspects of foreign/second language ability. In this study, the development of general reading ability and lower-level linguistic ability was also examined. It was concluded that the effects of extensive reading might be evident more quickly in general reading skills than in L2 linguistic ability.
In addition, the relationship between language achievement/proficiency and some other variables has been studied by researchers. Tabasi (2000) investigated the relationship between age, sex, and level of proficiency and self-assessment. It was revealed that children and females were more dependable in self-assessment questionnaires for listening. Proficiency had no significant effect on self-assessment except for the intermediate group in their overestimation of their speaking. A significant interaction was noted between age and level of proficiency.
In a study conducted by Fijani (2005), an attempt was made to find out whether below-and above-average EFL students differed in terms of restoring short-range and long-range cloze items. The results of two paired samples t-tests showed that both low-and-high proficiency groups score significantly higher on short-range items. It was concluded that low-and high proficiency EFL learners perform differently on the two types of items. Short-range cloze items were easier for EFL learners to restore. It was also concluded that below average students did not have easy access to long-range constraint. In another study, Hassani (2005) investigated the role of EFL proficiency, gender, in the interaction of motivation type (intrinsic/extrinsic), gender, and levels of English achievement (high, mid, and low). Also, the role that major and educational year play on learners' type of motivation was studied. The outcomes of the study revealed that the learners were mostly intrinsically motivated. It was also made clear that there was no significant interaction among motivation, gender, and level of English proficiency. Besides, no significant relationship was found between motivation type and major or educational year. The studies reviewed demonstrated the fact that there is a gap in the related literature on the relationship between language and identity and the impact of identity aspects on language achievement in the EFL contexts including Iran.
1.3 The objectives of the study
The objectives of this study were two-folded. The primary objective of this study was to investigate the relationship between aspects of identity and language achievement among the Iranian language learners who study medicine at Yasuj University of Medical Sciences (YUMS). The second objective of this study was to explore what aspect(s) of identity predict(s) language achievement.
1.4 Research questions
Based on the objectives of the study, the present study aimed at answering the following research questions:
1. Was there any relationship between aspects of identity (personal, relational, social and collective) and language achievement among the Iranian EFL learners?
2. What type(s) of identity predicted language achievement among the Iranian EFL learners?
3. Was there any significant difference between male and female Iranian EFL learners regarding their categories of identities and language achievement?
2. Method
This section introduced the participants, specifying how, where and in what ways they were selected. Moreover, the instruments used for data collection including the achievement test, the questionnaire for determining the type of aspects of identity (personal, social, collective and relational) together with the reliability and validity of each, where necessary, were provided. In addition, the data analyses, along with the procedures made use of, were presented.
2.1 Participants
The participants of the study were 76 medical students who were selected on the basis of the purposive sampling from the School of Medicine of Yasuj University of Medical Sciences (YUMS), although the initial selection of the sample was based on their availability. The age factor is controlled in the way that most of the participants, around 96%, were 21.
2.2. Instruments
2.2.1 Language achievement test
A language achievement test was constructed. The test consisted of 70 items, 10 listening comprehension, 30 vocabulary and expressions, 20 grammar and 10 reading comprehension. The test was developed with reference to the objective of the course taught in that semester. The listening comprehension section of the test was administered in a language lab which was an appropriate place for the test to be administered.
2.2.2 Identity questionnaire
In the present study, the fourth version of Cheek, Smith and Tropp's (2002) scale (AIQ-IV) was used. The AIQ-IV measured four identity orientations- personal, relational, social, and collective - in individual’s self-concept .To avoid any confusion and enhance validity, the Persian version of the questionnaire was utilized and the back translation by three experts in the field confirmed the original concepts of the translated questionnaire. The numbers of items in the form of a Likert Scale in each category were as follows:
The AIQ-IV is a 45-item objective inventory that measured the importance of four identity orientations in individuals’ self-concept:
a. Personal identity: 10 items
b. Social identity: 7 items
c. Collective identity: 8 items
d. Relational identity: 10 items
e. Special items [not scored on scales]: 10 items
The items of the AIQ-IV were endorsed on a five-point Likert scale, anchored at the extreme values of 1 and 5. Respondents had to respond according to the extent that they agree or disagree with the statement made. The numerical five-point scale was divided as follows:
1= Not important to my sense of who I am
2= Slightly important to my sense of who I am
3= Somewhat important to my sense of who I am
4= Very important to my sense of who I am
5= Extremely important to my sense of who I am
2.2.2.1 Validity and reliability of the questionnaire
Cheek, Tropp and Chen (1994) ran a factor analysis whose results indicated that personal, social, and collective aspects of identity constituted three relatively distinct categories of identity attributes with adequate psychometric characteristics.
Furthermore, the alpha coefficients of reliability for each scale were: personal (.84), social (.86), and collective (.68). Moreover, Cheek, Smith and Tropp's (2002) psychometric analyses in a sample of 1999 college women yielded 10 items for the new relational scale. As such, the final version of the questionnaire comprises four categories. Furthermore, the researcher ran a confirmatory factor analysis. The results indicated that personal, relational, social, and collective aspects of identity constituted four relatively distinct categories of identity attributes with adequate psychometric characteristics. The validity of the third scale was checked by Jowkar and Latifian (2006) utilizing factor analysis which yielded three main factors; also all the items delineated acceptable go-togetherness. As for the reliability, Jowkar and Latifian (2006) ran the Cronbach's alpha coefficients of reliability and the indices for personal, social and collective identities were .63, 67 and .55, respectively. Moreover, the researcher calculated the reliability index for the 5 variables of the questionnaire using Cronbach's Alpha. The results were .61 (personal identity), .67 (social identity), .71 (collective identity), .73 (relational identity), .67 (special items) and for all the items of the questionnaire the reliability index was .79. Therefore, Cronbach's Alpha depicted that each of the factors by itself and the combination of all factors had acceptable reliability indices.
2.3 Procedures for data collection and analyses
All participants were given a brief overview of the study including the objectives, the questionnaire, the time required for completion and the directions for answering the items of the questionnaire. They were told that they remained anonymous. Moreover, the researcher had access to the participants' total scores on the achievement tests by the exam center of the SULC. To analyze the collected data a combination of descriptive and inferential statistical procedures such as correlation coefficient, multiple regression and independent samples t-test were run.
3. Results and discussion
As mentioned above in the present study the relationship between language achievement and identity aspects, the impact of identity aspects on the Iranian learners' academic success as well as the effects of demographic variables on both language achievement and aspects of identity were investigated.
Table 1
Basic Descriptive Statistics for the Participants' Language Achievement
Factor |
N. |
Minimum |
Maximum |
Mean |
Std Deviation |
Language Achievement |
76 |
43 |
97 |
75.61 |
10.69 |
As Table 1 shows, 76 participants took part in the study with the mean of 75.61 and the SD of 10.69. Further, the minimum and the maximum scores on the achievement success were 43 and 97, respectively. The data showed that achievement tests were positively skewed when they were administered at the end of the academic semester. Moreover, the SD showed that the participants were heterogeneous regarding their language achievements.
Table 2
Basic descriptive statistics for the participants' aspects of identity
Aspects of Identity |
N |
Minimum |
Maximum |
Mean |
Std. Deviation |
Personal |
76 |
3 |
5 |
4.33 |
.62 |
Relational |
76 |
3 |
5 |
4.19 |
.51 |
Social |
76 |
2 |
5 |
3.47 |
.73 |
Collective |
76 |
2 |
5 |
3.64 |
.64 |
Table 2 clarifies the idea that the Iranian students' personal and relational identities were higher than their social and collective ones. This is indicative of the idea that the Iranian educational context is an individualistic rather than a collectivist one. Brown (2007) believes that in collectivist societies education is a way of gaining prestige in one's social environment whereas in individualistic ones education is a way of improving one's economic worth and self-respect based on ability and competence. Another reason for the low social and collective identities among the Iranian learners was that approaches and methods such as Communicative Language Teaching (CLT), Community Language Learning (CLL), Cooperative Learning and Participatory Learning were missing to a great extent in the Iranian Context (Razmjoo & Rizai, 2006; Larsen-Freeman, 2000).
Table 3
Spearman Correlation Coefficient for Aspects of Identity,Language Achievement and Age
Variables |
Personal ID |
Relational ID |
Social ID |
Collective ID |
Achievement |
Personal ID |
1 |
.514* |
.289* |
.317* |
-.034 |
Relational ID |
.514* |
1 |
.393* |
.367* |
-.071 |
Social ID |
.289* |
.393* |
1 |
.443* |
-.028 |
Collective ID |
.317* |
.367* |
.443* |
1 |
-.076 |
Achievement |
-.034 |
-.071 |
-.028 |
-.076 |
1 |
* Significance level was set at .05
Vis-à-vis the results portrayed in Table 3, no significant relationship was found between language achievement and the aspects of identity. As regards as the significant correlation among the aspects of identity, it was justifiable since all of these four categories belong to the major category of identity and the confirmatory factor analysis proved it. However, the findings did not support Leibowitz et al.'s (2005) results that language both as proficiency in the dominant medium of communication and as discourse was a key component of identity in the higher education institution.
Table 4
Multiple Regressions for Aspects of Identity and Language Achievement
Independent Variables |
Dependant Variable |
Beta (β) |
t |
Sig. |
Personal ID |
Language Achievement |
.031 |
.259 |
.693 |
Relational ID |
-.049 |
-.597 |
.617 |
|
Social ID |
.022 |
.275 |
.739 |
|
Collective ID |
-.064 |
-.879 |
.354 |
To see whether the aspects of identity could predict language achievement had to obtain the standardized regression coefficient (β) and put it in the predictive formula:
Y = β₁ X₁ + β₂ X₂ + β₃ X₃ + β₄ X₄ + …+ βn Xn
Y = language achievement
X₁ = personal ID
X₂ = relational ID
X₃ = Social ID
X₄ = Collective ID
With reference to the findings of Table 3, Table 4 showed that none of the aspects of identity could predict language achievement in the Iranian educational context in the academic setting, which was in line with the findings of Razmjoo (2010) in the Iranian EFL institutional context. Despite the fact that aspects of identity were strong predictors for many variables such self-consciousness, sociability, institutional and altruistic selves, guilt, need for uniqueness etc (Cheek and Hogan, 1983; Luhtanen and Crocker, 1992), the present study showed that they could not predict language achievement in the Iranian academic context.
Table 5
Independent Sample t-test for the Male and Female's Aspects of Identity and Language Achievement
Variables |
Gender |
N |
Mean |
Std.Deviation |
t |
Sig. |
Personal ID |
Female |
44 |
38.64 |
4.72 |
2.96 |
.002* |
Male |
32 |
36.38 |
4.61 |
|||
Relational ID |
Female |
44 |
40.63 |
5.21 |
3.11 |
.001* |
Male |
32 |
38.51 |
4.34 |
|||
Social ID |
Female |
44 |
27.33 |
3.95 |
.462 |
.472 |
Male |
32 |
27.09 |
4.62 |
|||
Collective ID |
Female |
44 |
27.93 |
4.76 |
.512 |
.566 |
Male |
32 |
28.19 |
4.94 |
|||
Language Achievement |
Female |
44 |
75.13 |
11.16 |
-.785 |
.649 |
Male |
32 |
76.09 |
10.59 |
* Significant at .05
Five independent sample t-tests were run to determine the difference between male and female participants' performance on language achievement and aspects of identity. As it can be perceived, Table 5 indicated that significant and meaningful differences regarding personal identity and relational identity were found .Nevertheless, no significant difference was found between males and females concerning collective and social identities as well as language achievement. In general males’ and females’ social and collective identities as well as their academic success were to a great extent the same.
4. Conclusions and implications of the study
To summarize, the following findings were drawn from the present study:
1. The correlational analysis in the present research presented no significant relationship between language achievement and any type of aspects of identity.
2. Multiple regressions indicated that language achievement of the Iranian EFL learners could not be predicted by any aspects of identity.
3. The results of independent samples t-tests showed that male and female are only different regarding their personal and relational aspects of identities.
The findings can be justified in different ways. Due to the fact that it was for the first time that such a study was conducted in the Iranian academic context, the study can be replicated in other different university contexts of the country such as Humanities, engineering, other medical contexts in order to explore the relationship between students' academic achievements and their aspects of identities. Moreover, regarding the idea that the questionnaire included some items which hinder the participants to express their ideas freely and as a result they did not cooperate with the researcher, it is recommended that a new questionnaire on aspects of identity including items tapping the Iranian and local identity be designed and used for the Iranian context because most of the time the questionnaires validated for the inner circles (English as a native language) are not suitable for the expanding circles (EFL). Although in the present study, gender could not account for the differences among the participants regarding course achievement and aspects of identity, there might be other factors such as age, different major of studies, parents' literacy, social class, family economic level, etc which might be the determining variables for the two independent variables. As such in further research, such factors should be taken into account. Despite such unexpected findings, it is recommended that for any educational system, policy makers, materials designers, educators, teachers and in a nutshell all decision-makers have an eagle view of the learners' characteristics and consider learners' perceptions of themselves because the educational milieu is definitely affected by the learners' values, identities, as well personality, social, economic and cultural factors. Moreover, EFL teachers in Iran and similar contexts should keep in sight not only language, but also learner, the person. The data in the present study were confined to Iran’s ELT medical context, and its implications for other contexts remain to be examined in future research.