The Relationships among Iranian EFL Learners’ Perfectionism, Anxiety and English Language Achievement: The Contribution of L2 Selves

Individual differences are known to play a crucial role in the acquisition of a second language (L2). An especially important variable of this type is perfectionism. Recent studies (e.g., Erozkan, 2016; Flett, Hewitt., Su., &n Flett., 2016) have sought to explain the relationship between this variable and anxiety. However, this task is complicated by the fact that the mechanism or process which underlies this relationship is believed to implicate various mediator variables. Among these are the roles of the self (Mercer & Williams, 2014). In recent years, the key role of the self in second language acquisition (SLA) has also  increasingly been gaining recognition from SLA scholars and practitioners (Mercer & Williams, 2014), since being human implies the reflective consciousness of having a selfand the nature of the self is the very essence of being human (Lewis, 1990).
       It is worth noting that previous studies have investigated only some, but not all, of the variables under scrutiny in the present study. Accordingly, the overriding goal of the current project was to investigate the relationships among perfectionism, anxiety, and language achievement in the L2 context, considering the mediating roles of self. Another key feature of this investigation was that Iranian learners of English (ILEs) will be the L2 group of interest. To date, relatively little work has been done on this group, regardless of the specific topic.
       In addition to fulfilling the research goal stated above, this study achieves several significant advances in the field of SLA; for example, it will increase the meagre amount of research that has been done on ILEs to date. By extension, it willalso help L2 learners in general to be more aware of the factors that affect learning. In addition, it will provide teachers, teacher educators, curriculum designers and researchers with a better understanding of the above factors.
            The target population in this study were 400 EFL learners within the age range of 15-30 who enrolled in private language learning institutes in Iran. The participants’ proficiency levels fell within the range of upper-intermediate to advanced sub-levels of ACTFL Proficiency Guidelines as determined by a placement test already administered by the institutes. To estimate perfectionism, Multi-dimensional Perfectionism Scale MPS was used (Hewitt & Flett, 1991). For anxiety factor, Foreign Language Classroom Anxiety Scale (FLCAS) was utilized (Horwitz et al., 1986). Learners’ L2 Self was measured by application of Taylor’s Quadripolar Identity Questionnaire (2010) structured into 4 primary scales representing the four self-categories (private self, ideal self, public selves, imposed selves). Learners’ language achievement was obtained from their final exam scores. The participants were asked to fill out the related questionnaires. To draw the sample, probability sampling was used. The sampling method in the present study was stratified sampling: five institutes were selected randomly as the accessible population. All the instruments were administered by regular EFL teachers, with the support of bilingual aides when needed. All study measures were administered to the participants in the form of a take-home survey to be returned in the next session of the class.
        Using a descriptive correlational design, the predictive independent variables in this study are types of perfectionism, FLA and L2 achievement are the dependent variables, and L2 selves is mediating variable. The participants’ scores on the final exam were collected. Initially, Cronbach’s alpha coefficient was computed to estimate the internal consistency estimate of the reliability of the four questionnaires. Then the data collected by means of questionnaires was explored using correlation analysis and structural equation modeling (SEM). The software package SPSS 24.0 and Amos 8. were used for the descriptive statistics and the correlation analyses, respectively. Reflecting on the research findings, it is assumed that there exists a structural relationship from perfectionism to FLCA and English language achievement via the mediating role of L2 Selves.
Two models were considered to test the hypothesis. The first model evaluated the effect of perfectionism on achieved score through selves and anxiety constructs. The second model assessed the effect of perfectionism on achieved score through selves’ sub-factors and anxiety constructs.
       In the light of the results, values of fit indices show that all the models have a good fit and for the each of two models it is in good and the least acceptable level considering the value of goodness of fit indices, proportation of chi-square to degree of freedom, RMSEA, CFI, NNFI, GFI, and AGFI. In other words, they are in the least acceptable limit for accepting the model.
            Overall, the findings of this study are important for researchers, university teaching staff, and the students themselves. It is hoped that increasing and detailed knowledge about the dynamics of foreign language perfectionism, anxiety, and self-will enlighten foreign language educators and researchers. Thus, these people can be in a better position to help to reduce the perfectionistic tendencies and as a result anxiety in the English classroom that will better help the students to benefit from English language learning