The notion of 'integrativeness' is a crucial element in the Gardnerian's socio-educational framework of attitudes and motivation. The model proposes that learners who have integrative orientations towards learning the target language (henceforth TL) possess favourable attitudes towards the language community, and an inherent interest in foreign languages (Gardner, 1985; Masgoret et al., 2001). These "integratively motivated" learners also exhibit different aspects of motivated behaviour such as effort, an expressed desire, or enjoyment of the TL learning process (ibid). Such inclinations also suggest that these learners desire to "come closer to the other language community" (Gardner, 2001a, p. 5).
The interest among researchers and educators in finding the link between achievement behaviour and success in language learning has generated considerable research in many second language (henceforth SL) and foreign language (henceforth FL) contexts (Lamb, 2004). More importantly, the interest in the area has witnessed a kind of "motivational renaissance" (Gardner and Tremblay, 1994a) with new concepts from educational psychology widening the scope of the motivational research agenda, resulting in the emergence of new understandings and conceptualizations in the study of motivation (Lamb, 2004). One of the most crucial being the reformulation of the notion of integrativeness itself.
Gardner (2001b) considers the amount of interest in the area, evident from the numerous studies, as an endorsement of the significance of integrativeness. This assumption is further encouraged by the fact that only minor changes had been made to the socio-educational model that he originally proposed. Yet, recent studies in the FL contexts have urged researchers to revisit the definition of integrativeness. Studies by Dörnyei and Clément (2001), Dörnyei and Csizér (2002), and Csizér and Dörnyei (2005a) in particular have revealed a major shortcoming in the traditional definition of integrativeness. While its predictive value has been confirmed in many language learning contexts, the researchers find that the original understanding of integrativeness is lacking in its explanatory power in FL settings. The conventional interpretation of this concept entails the existence of a recognizable group of native speakers in the learner's immediate environment with whom s/he wishes to interact and integrate in some way. This becomes problematic in the FL context because, as the researchers argue, the existence of native language speakers in the FL environment is not typical (Dörnyei and Csizér, 2002; Csizér and Dörnyei, 2005a). Given this reality and the argument that motivation is found to reside at the interaction of the individual and environmental dimensions (Hickey, 1997; McGroarty, 2001), Dörnyei and Csizér propose a reformulation of the concept of integrativeness. They suggest that FL learners are motivated to learn a TL to enact "possible selves" compatible with some social traditions and/or habits.
The whole notion of FL learners enacting "possible selves" is intriguing and one that requires investigation in different FL learning environments. Indeed this is the motivation for the current study which, based on the reformulation of integrativeness in FL settings as proposed by Dörnyei and Csizér (2002), and Csizér and Dörnyei (2005a), set out to investigate the nature of FL learners' possible selves in the Yemeni context. Yemen comprises a highly stratified society (Al-Bana'a and Al-Jabli, 2002), so the focus of the study is on the relationship between Yemeni EFL students' socio-economic background and their motivation to learn or use English to construct possible selves or an identity. In particular, the study is interested to find out if the subjects' motivation to learn English is influenced by the existence of a TL group which comprises Yemeni speakers of English who have a good socio-economic standing. The group comprises a remarkably small number of individuals but who have achieved a social status that is very much envied by others in the society (Saif, 1999; Al-Quyadi, 2000; Willis, 2007). The members of this group include those who are employed in the oil project in the Hadramout Province in Yemen.
The study
The current research partially replicates Lifrieri's (2005) study which was conducted on a group of Argentinean school boys. It investigates how Yemeni students' socio-economic status correlates with their English language learning motivational patterns. Given the recent emphasis on expanding the traditional social scope of motivational models to include macro-social variables (McGroarty, 1998; 2001; Spolsky, 2000; Clément and Gardner, 2001; Dörnyei, 2001b; Gardner, 2002), this study is timely as Yemeni EFL students live in a highly stratified society where English is associated generally with groups possessing more social, cultural, or financial resources (Nielsen, 2003). As established earlier, there exists a small number of socio-economically developed Yemeni English language speakers who have achieved an enviable social status in the Yemeni society (Kamali, 2000) and hold high-ranking, well-paid jobs in the oil industry as well as at foreign embassies and institutions. It is highly likely that Yemeni EFL students perceive practices such as speaking and using English by individuals in higher social positions as a symbol of status and prestige.
Bourdieu comments that "the principles of division [in any society], like volume and structure of capital, determine the structure of social space and are reinforced by principles of division relatively independent of economic or cultural properties" (1986, p.743). Thus, the socio-economic backgrounds of the sample in this study result from their perceptions of the occupational status of their parents (i.e. economic capital) and the accumulation of cultural capital in their families. According to the status-based approach to social stratification, students' internalised knowledge and perceived modes of status differences between them and the local TL group are shaped by their socialisation in given socio-economic positions (Bourdieu, 1985; 1986; 1989). On the basis of the students' knowledge towards the meaning of the representation and the social practices' symbols of the domestic TL group, their motivational patterns may then relate to their drive to identify or integrate with the social practices of that group.
The present research thus aims to establish a link between students' socio-economic backgrounds and motivation to learn English in the Yemeni EFL context based on a methodology that is informed by Gardner's (2001a) integrative motivation framework and Bourdieu's (1985, 1986, 1989) reproduction theory. It is hoped that the link between Gardner's framework of L2 motivation and Bourdieu's approach in sociology will reveal an accurate picture of the relationships between the components and/or subcomponents of FL motivation and socio-economic factors.
The study set out to answer two research questions as follows: 1) To what extent is there a reliable relationship between the economic capital (i.e. parental occupation) and the motivation of Yemeni university EFL students? 2) To what extent is there a reliable relationship between the cultural capital (i.e. parents' educational level) and the motivation of Yemeni university EFL students? The answers to these questions will contribute new insights into the ongoing research on FL learning motivation and have an impact on the future of FL instruction and educational planning and policy in Yemen.
Methodology
The current research is framed theoretically and conceptually within Gardner's (2001a) integrative motivation framework and Bourdieu's (1985; 1986; 1989) status-based approach to social stratification. As well, the study is guided by the hypothesis that if there are individuals who are characteristically associated with English and who are also socio-economically established in the learners' social environment, their integrative tendencies are most likely associated with those individuals, even if they are not native English speakers. Given the concept of integrativeness as proposed by Csizér and Dörnyei (2005a) and Dörnyei and Csizér (2002) and the perception of English in Yemen, that is, knowing English is regarded as a prestigious practice associated with the high society (Al-Quyadi, 2000), the present study assumes that the Yemeni EFL students' integrative motives towards the local TL group are crucial to the relationship between their socio-economic status and motivation to learn English.
The motivation questionnaire and the socio-economic questionnaire were administered sequentially to 142 students who were randomly selected from a target population of 250 students. They were fourth-year students at the English Department of the College of Arts and Education, the Hadramout University of Science and Technology. The responses from the 142 questionnaires were tabulated and then calculated to obtain the average score for the motivational and socio-economic factors. This was followed by a statistical analysis using the SPSS 11.5 for Windows to find out the correlation result between the variables.
An examination of the results given in the table above shows that there are considerably high levels of Overall Motivation (M= 5.6) which exist among 113 (79.3%) students in the sample. Likewise, the average mean scores of the three main motivational components are similarly high. The average mean scores of the three motivational components tend to be close to the highest score value in the motivation questionnaire, i.e. 7. This is evident from the two average mean scores of 5.6 equally obtained for Integrativeness and Motivation. In both cases 116 students (82%) positively rated all the items under these two motivational components. Also, the average mean score corresponding to Orientations is 5.5 with 108 students (76%) in the sample rating the items under this motivational component positively. The results in Table 1 are presented graphically in Figure 8.
Figure 8: The distribution of mean scores of the motivational components.
The high average mean scores and frequencies suggest that the majority of the Yemeni EFL students in the sample are highly motivated to learn English. The students in this study are mature enough to be able to provide responses on their actual socio-economic backgrounds and are regularly engaged in the English language to reflect on their attitudes and motivation towards learning the TL. In the individual semi-structured interviews, a majority of them (i.e. 19 students) expressed that they had become more interested in learning English and believed that it would offer an enjoyable challenge for them. They also felt satisfied with the possible future opportunities which seemed to have motivated them to study the language. The students also stated that their increased interest in English was accompanied by favourable attitudes towards the TL group members. In other words, the students expressed that there is an inherent desire to learn English in order to form ties with the Yemeni speakers of English. So, it can be said that the students in the sample are motivated and that they partake in language learning because of the satisfaction they feel and the interest they have gradually developed in the TL and the local TL group.
As shown in Table 2, all of the correlations that included the motivational factors show interesting degrees of positive and negative relationships with the socio-economic variables. Clearly, the relationships between the motivational subcomponents and Mother's Past/Present Occupations, as well as with Father's Past/Present Occupations, are negative whilst Cultural Capital correlates positively with all motivational subcomponents. All of these relationships are significant at the 0.01 confidence level. Mother's Past/Present Occupation has a moderately negative correlation with all of the motivational subcomponents. The coefficient values for: Attitudes towards the TL Group, Interest in Foreign Languages, Integrative Orientation, Motivational Intensity, Attitudes towards Learning the TL group, Desire to Learn the TL, and Instrumental Orientation with Mother's Past/Present Occupation are r = -0.48; r = -0.52; r = -0.57; r = -0.55; r = -0.51; r = -0.50; and r = -0.53 respectively. All of these relationships are also moderate except for the one with Attitudes towards the TL Group, which has a weak correlation coefficient. All of these relationships might result from the absence of any influential role of the sample's mothers, given that the majority are illiterate, and those who work positions at the lower end of the scale.
In contrast, Father's Past/Present Occupations have a strong, significant, negative relationship with all of the motivational subscales. It is evident from the results that the role of fathers in the students' families outperforms the mother's role. The majority of students in the interviews expressed that their fathers played a more influential role in their families than the mothers. They stated that their fathers encouraged and supported their efforts to learn English language. Hence, the correlation coefficients of Father's Past/Present Occupations with the motivational subcomponents are: r = -0.71 with Attitude towards the TL Group; r =-0.74 with Interest in Foreign Languages; r=-0.71 with Integrative Orientation; r=-0.73 with Motivational Intensity; r=-0.71 with Attitudes towards Learning the TL; r=-0.71 with Desire to Learn the TL; and r=-0.75 with Instrumental Orientation. Finally, the Cultural Capital is strongly correlated with the motivational subcomponents. The correlation coefficients for this socio-economic component with the motivational subcomponent variables are: r= 0.84 with Attitudes towards the TL Group; r = 0.83 with Interest in Foreign Languages; r = 0.88 with Integrative Orientation; r =0.83 with Motivational Intensity; r =0.82 with Attitudes towards learning the TL; r = 0.84 with Desire to Learn the TL; and r =0.80 Instrumental Orientations.
The results on the relationship between parental past/present occupations and all the motivational subcomponents are in keeping with the results presented in the previous section. It can generally be observed that the less skilled occupation a student's father or mother had/has, the more integrative is the student attitudes towards the Yemeni speakers of English. This finding accords the point raised by Gardner (1985). He draws our attention to some evidence of stronger parental influence on the learners' opinions when the families tend to belong to the lower socio-economic stratum. The learners have a more pragmatic attitude towards English language learning.
Clearly, the parents' occupations are strongly related to the formation of students' attitudes towards the local TL group. Hence, it is suggested that the TL group for the sample of the present study is represented by Yemeni speakers of English, not native English speakers. This is evident from the students' responses in the interviews with regard to their perception of the members of the local TL group. The group members' distinguished social position creates within the students a desire to identify with their habits and practices. For students whose fathers or mothers have lower-ranked occupations, English learning tends to be associated with such characteristics. This supports the assumption of the study that English represents a prestigious practice typical of high-status groups, at least in the minds of more deprived students. Whilst this is the case, there is nothing conclusive about how fathers or mothers' involvements in language learning affect students' motivation, and how these parents with less prestigious jobs may impact the attitudes and motivation of these students. It is difficult to comment critically on the nature of the parental influence from the results of the study, but it seems that the fathers believe more strongly in the value of English for the attainment of high social status. It should also be noted that two of the twenty students interviewed, who were socio-economically better-off than the others, pointed out that English has been a much more widespread phenomenon in Yemen, and that being able to speak the language is not necessarily the privilege of only socially advantaged groups.
Although correlation is not causation, all of the aforementioned significant relationships between motivational component factors and the socio-economic factors are reflective of the students' integrative tendencies towards the local TL group and their high level of motivation towards learning English in general. Interestingly, two significant correlations are negative, that is, when either one of the parents have or had lower-ranked occupations. So, socio-economic factors are powerful predictors of the students' motivation to learn English. The results also show that cultural capital has a strong, positive correlation with all of the motivational variables. This is despite the fact that the majority of the students in the sample are from families with a parent with a low-ranked occupation, and have smaller amounts of cultural capital. As expressed by the students in the interviews, this may be due to the sample's knowledge of the importance of cultural capital, and the prevailing culture and linguistic characteristics of the parents.
Conclusion
The results of the study reveal that the Yemeni EFL students in the study developed strong integrative tendencies towards the local TL group and high level of motivation to learn English. Since language use is inextricably linked to social class and status, the students' motivation to learn English is influenced by membership in a certain socio-economic strata. It is evident from the study that that social stratification promotes a particular attitudinal identity construction/formation in the sample of this study. Social stratification actually influences all the sample of this study to learn English, observable via all their motivational levels. Socio-economic factors including parental occupation and cultural capital seem to be powerful factors contributing to the high levels of the students' motivation to learn English. The results of the study show the Yemeni EFL students with poorer economic backgrounds tend to be more optimistic and motivated. They exhibit consistent intra-group patterns of TL motivation, which points to the reproductive effects of Bourdieu's habitus.
Csizér and Dörnyei (2005a, p. 29) place the source of learners' identification/integration process in the internal representations of the self. Given this, the L2 learner devoid of his/her social contextualization, seems to be depicted as randomly desiring to seize certain attributes, which would make him/her agreeable or professionally successful in the society (ibid). This view of the L2 learner constructing an ideal image and dissociating him/her from the social reality has been challenged in the current study by the incorporation of some elements from the learners' social space. In this study, the sample's experiences in and with their social reality may have influenced their motivation, expectations and construction of new identities/images. The complex interactions between these dimensions are clearly reflected in the encompassing theoretical and conceptual framework of the current study. Unlike the concept of "possible selves" proposed by Csizér and Dörnyei (2005a), Bourdieu's habitus which is adopted in the current study, offers a psychosocially grounded construct to express the situation in which the sample internalizes ways of understanding their social positions and thinking about and reacting to their social world. Bourdieu's theory contributes well to the current study as it is based on a mental-social relational perspective. Ultimately, the concept of social stratification clarifies aspects of the cognitive formation of Yemeni students' attitudes towards the domestic TL group and their motivation to learn English.
It can be stated that socio-economic prestige is a crucial factor behind the strong integrative tendencies and the high motivation levels of the Yemeni EFL students in the sample. The students consider the use of English in their context as the prerogative of members in high-status groups, where knowledge of English allows them the access to better employment opportunities and the potential to move up the social scale. They perceive the ability to use English language as a means to social distinction and closer to the TL group.
In conclusion, the results of the study provide further confirmation of the impact of learners' socio-economic factors on their integrative tendencies in particular and on motivational patterns in general in FL learning. Integrativeness, the study shows, is particularly motivated by social aspirations. And the significant negative correlations between parental occupations and integrativeness that emerged in the present study deserve attention and exploration in other FL contexts.
