Tunisian Consumer's Cultural Identity Dynamics

The main objective of this research is to identify the major changes affecting consumer cultural identity. Due to globalization and multimedia development, cultural changes and cultural identity shifting became more and more possible especially with the development of the acculturation insitu (Hirshmann Touzani 2011). We developed in this paper a conceptual model reflecting cultural identification possibilities and their antecedents such as family cultural orientation and demographic characteristics. We tested this model with a sample of Tunisian students. The results of our research show a valid model where local consumer's cultural identity is bidirectional. This means that in North American markets (Laroche and al 1998, (Berry 1980, 2001, Penazola 1994, Laroche, Kim et Tomiuk 1998, Maldonado, Tansuhaj 2002) or in other emerging markets (Ustuner and Holt 2007, Touzani and Hirshman 2011), in front of a new hegemonic culture, consumers’ cultural identity develops two major parallel dimensions: identification to local culture and acculturation to occidental one. These results are interesting for exporting firms who desire to have a more precise idea on local consumer's cultural orientations; this would help them to develop more effective exporting strategies.

However, this problem has evolved until now.
Indeed changes caused by globalization have trained changes also at cultural identity level (Gupta 2012). Thus, in the postmodern perspective, the cultural identity of the consumer is constantly changing.
Thus, the rise of globalization due to open borders and the acceleration of economic and social exchanges has created the opportunity for consumers to build a more complex cultural identity (Gupta, 2012, and Ustuner holt, 2007, 2010.However, for any exporter identifying trends at the cultural affiliation to the consumer of the recipient countries will either adopt standardization strategy, an adaptation strategy or an intermediate one. This will allow these companies to target and optimize their efforts in terms of marketing efforts. In this context, our study is interested in the analysis of the cultural affiliation of the consumer, taking into consideration the two major cultural trends: identifying the origins and acculturation for the overall model represented by the western culture.
Thereby, through this research, we propose to develop and test a model for understanding the dynamics of cultural identity among local consumers paying attention to socio-demographic characteristics and cultural orientation of the parents. We first present in what follows a review of the literature on cultural identity and acculturation before addressing the research methodology and testing the conceptual model.

Conceptual frame
During the 90s, right in the process of globalization, Douglas and Craig (1992) found that culture is the most important variable to consider in business strategies at the domestic level and in the exporting strategies. Long afterwards, authors such as Morimoto and La Ferle (2008), Douglas and Craig (2005), and Butt and Run (2012) confirm the ubiquity of the influence of cultural identity and the individuals' relationships with it on most consumer behaviors.
As such, the local culture specific to a geographical area, country or ethnic group is not the main concern of businesses, but rather the degree of adherence to it or its reject in favor of a global cultural trend represented by Western culture (Hassan (2011), Üstüner and Holt (2007), Touzani and Hirshman (2011), Gupta (2012)).
For this purpose, the literature offers two main trends of cultural dynamics: identification to the local culture and acculturation, due to confrontation with a dominant second culture (Laroche and al 1998). Acculturation dynamics can be noticed in cases such as in the case of individuals experiencing immigration or local individuals undergoing the effects of globalization (Zmund and Arce (1992), Stayman and Deshpande (1989), Laroche and al (1998), Berry (2005), and Hirshmann Touzani (2012)). Thus, these two phenomena constitute two crucial variables to consider in the development of the marketing strategy in a given market.
Moreover, the literature review reveals that companies, countries and individuals around the world are moving towards the global cultural model, with different rhythms and different intensities (Ustuner andHolt (2007, 2010), Gosh (2011 )). Thus, in the same country, some individuals living in an acculturated or westernized environment will behave in a similar way to other individuals living in Western countries (Lange and Anders (1990), Nexhmi Russel and Kingshott (2001), while others are more focused on their original cultural identity. The cultural dynamic is so, the combination of two main trends (Usinier (1992) Indeed, Berry (1989Berry ( .1994Berry ( , 2001) speaks of cultural change in the context of acculturation as a bi-dimensional change where there are two main tendencies: the preservation of the original cultural features and the acquisition of new traits of new culture.
The research of Berry is the most adopted particularly in the context of studying cultural changes affecting consumers at the local level or insitu (Ustuner andHolt (2007, 2010), Ghosh (2011), and Hirshmann Touzani (2011)). This is due to the fact that it takes into account the two major trends in the dynamics of cultural identity i. e. the degree of cultural identification to the local cultural backgrounds and the degree of adoption of behaviors and attitudes specific to the dominant Western culture.

Identification with the local culture
The dual aspect of individual and social concept of "identity" allows us to understand the concept of cultural identity defined by Zmund and Arce (1992) as the basic identity that aims to differentiate one group from another. According to Marshall (2002), this variable is also called ethnicity or ethnic/cultural identity; it is strictly linked to race by extending the social facts, cultural and historical distinguishing ethnic groups.
As part of our research, we will adopt the description of Keefe and Padilla (1987), Keefe (1990) and Phinney (1990), who are interested in the psychological aspect of cultural identification that is often described by the term "ethnic loyalty "or the degree of commitment and identification with the original culture that characterizes the social group.
According to Jung, Ball and Gentry (1993), Zmun and Arce (1992) and Phinney (1996), Ethnicity influences consumer behavior through specific cultural identity. Weber (1961) identified ethnic identity as a set of common dimensions of individuals from the same ethnicity sharing the same customs, language, religion, values, morals and etiquette. These dimensions are used also in psychology (Bhurga D (1999), Osei Appiah (2001)).
Given the multiplicity of dimensions that have been cited, it's worth noting that measuring the degree of identification with the culture of origin (as opposed to acculturation trends) is very complex. Hence, to study identifying trends in native culture by marketers, authors like Rexhma and Kingschott (2001), Mendoza (1995), Laroche et al (1998) have focused on two main factors that are: • The Interactions with the local culture: This variable was identified by Keef and Padilla (1987); it reflects the level of preference of association with members of the ethnic group. This preference and this interaction produce among others a normative influence that guides consumer choices. The latter is done through the search for gratification of the group and through the adoption of the group's values (Faircloth and Webster (1994). Kallini and Hausman (2007) add that this preference for ethnicity hinders the adoption of innovations when there is risk of rejection by the group.

•
The Religiosity: Religion is considered as an important marker of identity of culture (Eid (2003)). The literature review allows us to clarify the meaning of the word religiosity that has been defined by Worthington et al (2003) as the degree to which the individual adheres to its religious values, beliefs and practices and specific uses in his life daily (P 85).

Acculturation
Penaloza ( The process of acculturation is often an asymmetrical process of struggle between the dominant culture and the culture of origin. In our research framework, we are interested in the dynamics of cultural identity and specifically in its twodimensional dynamic as stated by Berry

The factors influencing the degree of identification with the origins and the foreign culture
Through the literary review, we found two major factors that could influence and affect the dynamics of cultural identity among consumers, which are the demographic variables (such as income, rurality and gender) and the degree of identification of parents to local culture. H1.c.2: Consumers living in rural areas have a higher level of identification and interaction with the local cultural group than Consumers living in urban areas.

H3:
The demographic variables influence the degree of acculturation.

The degree of identification with parents
Attitudes and values characteristics of a culture are transmitted through parents, from one generation to another by the process of enculturation. This phenomenon is a cultural adaptation process that allows the individual from an early age to adapt the characteristics of the surrounding culture With these theoretical results, we propose the following research hypotheses: H2: The parental orientation in relation to the local culture influences the orientation of the offspring to the same native culture.
H2.a: The more the parents identify themselves to the local culture; the higher will be the identification and integration into the local cultural group of descendants.
H2.b: The more the parents identify themselves to the local culture; the greater will be the degree of religiosity of descendants.

Methodology
In our investigation we concentrate on cultural characteristics; this was relatively tricky especially for the part where we studied religiousness, due to the political context of Tunisia pre and post Revolution. So, we made a first approach with semidirective interviews (10 individuals). This allowed us to identify recommendations pertaining in particular to the fact that the interviewees were feeling embarrassed when they were asked questions about their religiosity and degree of involvement in religious practice. This contravened us to choose a relatively simple measurement scale of religiosity in the case of the scale developed by Laroche et al (1998). Furthermore, measurement scales adapted in our research are respectively as follows: who advocate Likert scales to 10 points, ranging from "Extreme disagreement" to "Extreme agreement." -Interaction with the group and Adoption of foreign media scales: Scale (Laroche et al, 1998, 2004) and who advocate Likert scales to 10 points, ranging from "Extremely disagree" to "Extremely to agreement ".
-Parents identification to local culture scale: The scales of Laroche et al (1991) and Rexha and Kingshott (2001), Likert 10 points, ranging from "Extremely disagree" to "Extremely Okay ".
For the adaptation of the measurement scales to the Tunisian context, we used the retro-translation method to achieve a questionnaire in French.
For our sample, we chose the convenience method; its size is 310 students. It was consisting of students aged over 20 years at different levels, from the Faculty of Economic Sciences and Management of Tunis and the Higher Institute of Business Administration of Gafsa. This age has a specific characteristic, according to Arnett (2000) Holbrook and Schindler (1994), it is the age of the emergence of adulthood during which the individual develops their own identity choices through exploration and experimentation of other identities and other cultures (Erikson 1968). Moreover, as rurality and urbanity are exogenous variables of our model we performed the first part of our investigation in Gafsa, mining town in the south west of Tunisia in the Tunisian Higher Institute of Business Administration. The second part was conducted in the Faculty of Economic and Management Sciences of the country capital Tunis.

Presentation of the exploratory phase results
As the measurement scales were developed in the North American and French markets, a phase of adaptation to the Tunisian cultural context and the particular context of research is required. This adaptation phase was performed according to the following main steps (Nyeck S, S Paradise, JM Xuereb, Chebat JC (1996)): • The retro-translation of items

Construct validity of measurement scales
Construct validity; help to check whether the measuring instrument allows a good representation of the studied phenomenon.
To check this type of validity, marketing literature adopts two instruments namely the convergent validity and discriminant validity.

The results of convergent validity
The convergent validity is verified by the values of Rho of convergent validity and t of Student associated with each factor contribution for each of the variables in our model. The values obtained show that measurement tools are convergent (all Rho VC> 0.5) and the t value associated with each factor contribution (loading> 1.96).(See appendix 1

Discriminant validity
Discriminant validity verifies that a measure is not correlated with another measure with which it is supposed to be different (Peter 1981). For Akrout (2010), we just need to verify that the square root of Rho of convergent validity of each construct exceeds the correlations shared with other constructs. For our research, we will conduct verification of this type of validity for variables that have more than one dimension, i. e. the two dimensions of identification with local culture (Interaction with local culture and religiosity). The results chow that the measurement scales of the two dimensions of identification to local culture (Interaction with local culture and Religiousness) are not correlated.

Validation of the conceptual model
The model fit test phase allows the verification of the quality of adjustment of the model to the empirical data. Thus, we proceed with the verification of reconciliation between the theoretical model proposed during the theoretical phase and the resulting model sample data. To this end, we used structural equations method to verify the significance and the simultaneity of relationships between different variables in the model. The software used in this phase is 22 Amos, the estimation method is the Maximum Likelihood method and the sample size is 320 individuals. By the analysis of measurements of the quality of the structural model adjustment (See appendix 3), we can conclude that: -The structural model we proposed has good overall qualities of adjustment (GFI> 0.9 and AFM> 0.8).
In addition, the RMSEA and RMR, that measure the deviation from a perfect fit, are acceptable for our model (RMSEA <0.05).
-The Value of CFI (Comparative fit index) and TLI (Tucker-Lewis index) is> 0.9, it means that we compare our model to a relative model where all correlations between variables are null; the model tested is acceptable.
-Finally for the level of parsimonious indicators that assess the quality of estimation of coefficients for the relationships in the model, we can see that the ratio chi2 / df <2 and the PCFI > 0.5.
For all these findings, we can say that our model has a good fit.

Effect of demographic variables on religiosity
-The test of our model shows that the gender (CR 3477> 1.96, P=0.000) has a significant effect on the degree of religiosity. Thus, the female individuals have a greater degree of religiosity than males with a coefficient = 0.943, so the hypothesis H.1.a.1 was validated. This result was also found in the United States in a study on electoral intensions in presence of a female candidate (Marsden 2012).
-On the other hand, the hypotheses regarding the effect of urban or rural origins on religiosity H.1.c.1 and the hypothesis regarding the effect of family income on religiosity H.1.b.1 were not validated (P> 0.05).

The effect of demographic variables on the degree of interaction with the local culture
-Family income has a significant negative effect on the degree of interaction with the local culture (|CR| = 2.

The effect of demographic variables on acculturation
Acculturation, as a phenomenon taking place by a contact with a dominant culture, has interested marketing researchers notably at countries deeply affected by immigration such as the United States, United Kingdom and Canada (Berry (1990(Berry ( , 1997(Berry ( , 2001; Penazola (1993); Laroche and al (1998) Gbadamodi (2012)).
This interest in the phenomenon of acculturation was also evident at the local level through the contact of the local cultures of developing countries with occidental ones (Ustuner and Holt (2007) (2010), and Hirshmann Touzani (2011), Gupta (2012)).
In both cases, research has shown that the demographic variables have a strong influence on the propensity to adopt attitudes and behaviors (including consumption) typical of the foreign culture. So it seems that men are more likely to adopt the foreign culture than women. Individuals from rural areas are less receptive to foreign cultural facts as those, which originate in urban areas.
-As part of our research, we found that the higher the income the more significant would the trend of acculturation. For our sample, family income has the only significant effect (CR = 3. -Gender influences the degree of acculturation. In fact, in our sample men have a more favorable orientation to acculturation than women (ICRI = 2.625 <1.96, P = 0.009> 0.05), so H.3.a is validated.
The effect of the degree of identification of the parents to the local culture on the cultural affiliation of offspring: The study of the effect of the extent of parents' identification with the ethnic culture on the cultural affiliation of descendants was done both at the local level (Patel, Bhavnagar and Power ((1996)) and at the level of the country of immigration (Jun, Ball and Gentry (1993); JM Farver, Eppe S, D Ball (2006)). In our research, for the effect of the degree of identification of the parents to the origins on both children's identification to local culture and acculturation, we found the following results: -Hypothesis H.2.a regarding parental influence on religiosity of their descendants (CR = 8968> 1.96, P <= 0.000 <0.05) and hypothesis H2.b about the influence of the parents on the degree of interaction with the local culture (CR = 8.3> 1.96, P <= 0.000 <0.05) were validated.
-Hypothesis H4 relative to the effect of the parental identification to the origins on the degree of acculturation of descendants was validated (|CR| = 3.9091> 1.96, P = 0.000> 0.05).
These results confirm the outcomes obtained in other North American studies. Thus we found that the more parents highlight their ancestral identity, the stronger will be the identification of progeny to their original cultural identity (Harwood and al 2002;Himmelfarb, 1977Himmelfarb, , 1979Greeley and Rossi, 1966;Greeley, 1976). Conversely, the more the parental identification to the origins is low and acculturation is strong, the more acculturation and the integration of descendants in the new culture will be easier and faster (Carter and Midlemiss 2002: Sussman, Nan 2000, Chabot 2004).

Conclusion
In the present context of globalization, our research has enabled us to validate the model proposed from the literary review. Thus, the proposed model has verified the fact that consumers in a country such as Tunisia have a two-dimensional cultural identity that varies in one direction or another depending on demographic variables and according to the parental cultural orientation. This is particularly interesting especially in international marketing where knowledge of the nature of local cultural affiliation and the different factors that might influence it are considered to be primary for a better adaptation of marketing strategies.
Indeed, our research explains dynamics of the local cultural identity varying mainly according to two different trends that are: Acculturation in favor of Western culture and identification with local culture. Moreover, this is the case for most of the developing and emerging markets. So, companies have great interest in focusing on the influence of acculturation and mainly on the degree of adoption of target consumers of exporting countries' cultural traits. This is particularly important for the case of European and North American exporting firms. This interest should be present in their strategies, including communication. In fact communication in Western media may have an impact on local importing countries consumers. Indeed, one of the acculturation dimensions adopted in our research is the degree of adoption of foreign media.