Young people’s identity & Facebook behaviour: The role of gender and ethnicity

Abstract The aim was to investigate the effects of gender and ethnicity on Facebook visit and identity among young people (14–25 years old) living in Mauritius. According to the results obtained, males were shown to visit more Facebook and had a stronger Facebook identity than did females. However, females compared to males considered themselves to be persons that are more similar online as offline, and their Facebook activity represented more who they were than it did for males. Hindu participants were shown to most infrequently visit Facebook. They were also the group with the weakest Facebook identity. Creole and Muslim groups were reported to have the strongest Facebook identity followed by the Mixed participants. This study concludes that both gender and ethnicity might have a significant impact on Facebook activity and identification among young people.


Introduction
This article is based on a case-study from Mauritius. Mauritius is a small island state, which is situated towards the north east of Madagascar and covers an area of 1,865 square kilometres with a ABOUT THE AUTHORS Komalsingh Rambaree (PhD) is currently an associate professor of social work at the University of Gävle, Sweden. He has a PhD in social work and social policy from University of Manchester, UK. He is originally from Mauritius, where he has been working as a youth worker for more than 10 years, and as a lecturer in social work and social policy at the University of Mauritius for more than five years. His research areas include eco-social work, youth, health, internet, sexuality and qualitative data analysis.
Igor Knez (PhD) is currently a full professor of psychology at the University of Gävle, Sweden. He has a PhD in psychology from University of Uppsala, Sweden. His research areas include cognitive, emotion, and environmental psychology. He is currently engaged in a research on autobiographical memory and the self, related to the physical places of our lives.

PUBLIC INTEREST STATEMENT
During the past decades, Mauritius has experienced some socio-economic and cultural changes resulting from its advancing towards a technology-based society. Within this particular context, this article investigates the effects of gender and ethnicity on young people's (14-25 years old) identity and Facebook behaviour. According to the results obtained, males were shown to visit Facebook more often than females. Males were also shown to have stronger online identity than females. However, females compared to males considered themselves to be persons that are more similar online as offline, and their Facebook activity represented more who they were than it did for males. Concerning ethnicity, Hindu participants were shown to most infrequently visit Facebook. They were also the group with the weakest Facebook identity. Creole and Muslim groups were reported to have the strongest Facebook identity followed by the Mixed ethnic group participants. This study concludes that both gender and ethnicity might have a significant impact on Facebook activity and identification among young people. population of about 1.2 million. The island has no indigenous population. During the French and British colonization period, migration from various parts of Africa and Asia led Mauritius to become a multi-ethnic, multi-religious and multi-lingual society. The country is often described as a 'rainbow nation' in local and international media, where different ethnic groups preserve and maintain their ancestral culture, language, tradition, and religion.
Ethnicity is usually considered as a group classification, identification, and relationships, through which members of a particular ethnic cluster share a common identity and tend to regard themselves as being different from other ethnic groups based on characteristics such as decent, history, language, religion, culture and tradition (Rambaree, 2017). The ethnic composition of Mauritius is often reported as: Indo-Mauritian 68%, Creole 27%, Sino-Mauritian 3%, Franco-Mauritian 2% (USAID, 2014). Eisenlohr (2006, p. 397) argues: "Mauritian state institutions encourage the cultivation of diasporic 'ancestral cultures' in the context of a wider Mauritian political ideal of 'unity in diversity' and ethnoreligious even-handedness". This diversity based on race, religion, language and places of origin makes Mauritian society and culture very complex. In particular, Eriksen (1993) opines that a concern to reproduce ethnic boundaries and an urge to remain ethno-religiously 'pure' is typical of Mauritian society. Furthermore, religious leaders of Hinduism, Christianity and Islam preach tolerance and simultaneously stress the importance of having one faith (Eriksen, 1997(Eriksen, , 1998. As such, ethno-religious factors play a more central role than class in defining the choices of friends, dating and marriage partners in Mauritius (Eriksen, 1993;Nave, 2000;Ng Tseung-Wong, 2013;Rambaree, 2008). Family members, community leaders, some schools and most of the ethno-religious organizations are the main agencies through which ethno-religious 'purity' is maintained. As Nave (2000) puts it: "Parents are also deeply concerned about passing on their ethnic identity and the corresponding corpus of cultural beliefs, particularly religion, to their children and grandchildren" (p. 342). Ethnicity therefore plays important role in the identity construction, lifestyles and behaviour of the young Mauritian (Ng Tseung-Wong, 2013; Rambaree & Knez, 2016).
Mauritius is a highly patriarchal society and gender shapes the opportunities and experiences of many young people in various domains of life (Gokulsing & Tandrayen-Ragoobur, 2014;Ramtohul, 2010). The term gender is a social construct based on biological determinism. It refers to the social differences such as norms, roles and relations between men and women, which are learned, and vary widely among societies, cultures. In particular, gendered obstacles based on conservative norms limits Mauritian girl's freedom and access to participate public domains such as sports and leisure (Rambaree & Knez, 2016;Ramtohul, 2010). Young girls in Mauritius are conditioned by various agents and institutions of the society to abide by the dogmas of their cultural systems based on patriarchal moral norms and values (Aumeerally, 2005). It is very common to see moral discourses especially targeted at the identity of young girls in Mauritius. For instance, girls are supposed to have sexual identity, which reflects purity such as being virgin and sexually submissive and docile (Rambaree, 2011). At the same time, boys are tolerated in projecting their affirmative masculine identity both in public and private spheres of the Mauritian society (Bhautoo-Dewnarain, 2014).
During the past decades, Mauritius has experienced some major socio-economic changes resulting from its advancement towards a technology-based society. Such socio-economic changes are believed to be having an impact on other aspects of young people's life. In particular, social media is claimed to be breaking down cultural barriers by bringing more information, communication and opportunities to the young people in Mauritius (Rambaree, 2008(Rambaree, , 2009(Rambaree, , 2011. A multi-ethnic and patriarchal society like Mauritius, which is transiting from being an agrarian to a technology-based society, therefore represents an interesting case study to explore role of gender and ethnicity on young people's identity and Facebook behaviour. Although Mauritius -as a multicultural and highly gendered society -represents an excellent case study for researching the effects gender and ethnic identity of young people's behaviour on social media, very few scientific studies exist on such aspects (Khedo, Suntoo, Elaheebocus, & Mocktoolah, 2013). Moreover, empirical evidence on young people's cyber behaviour and identity from emerging economies like Mauritius is still under-represented in international academic discourses. Parker (2004) opines that in an increasingly globalised era, most research activities are from economically developed countries. Youth in emerging economic contexts, like in many established economic contexts, are also at the forefront of technologies that are transforming social interactions in many ways; and their interactions with new technologies have yet to be fully understood (Manago, Graham, Greenfield, & Salimkhan, 2008). In particular, scientific studies on how SNSs' users from different countries identify themselves is crucial for inquiring into the dynamics of culture and youth in a globalising world (Boz, Uhls, & Greenfield, 2016).
For some social media anthropologists, identity switching is a classic example of the fluidity of identity in the modern world (Lewis, 2004). Despite popular platforms such as Facebook explicitly encourage users to provide truthful personal information, young people can still manipulate their profiles to create impressions of their self that they consider favourable through various means such as pictures, statements and interests (Herring & Kapidzic, 2015). Such fluidity therefore allows some young people not only to understand, but also, to question and challenge existing roles, expectations, and relations that affect aspects such as gender and ethnic in society.
What about fluidity of identity on Facebook among young people in a country like Mauritius? What are the effects of gender and ethnicity on young people's identity and Facebook behaviour in such a context? These are the main questions of this study.

Theoretical frame: Identity and social media
Identity is a complex and fuzzy concept. Rattansi and Phoenix (2005) argue that the term identity, "has become a matter of key theoretical and empirical concern in the social sciences, and considerable debate has developed as to its proper conceptualization and the research strategies most appropriate to its investigation" (p. 97). In its broad conceptualization, identity defines a person by reference to various aspects such period of age, gender, ethnicity, nationality. In addition, a person can have different identities based on traits and characteristics, social relations, roles, and social group, grounded in autobiographical memory (Knez, 2014(Knez, , 2016Knez & Eliasson, 2017). In this sense, memberships define who one is or want to be through a self-definition and self-concept (Oyserman, Elmore, & Smith, 2012;Waterman, 2008). A broad definition of identity therefore encompasses a theorization of self as a person (Michikyan, Dennis, & Subrahmanyam, 2014;Michikyan & Subrahmanyam, 2012;Subrahmanyam & Smahel, 2011).
Online self-presentation on Facebook takes place primarily through profiles where users create their own identity and visually display connections to their social network (Herring & Kapidzic, 2015). Some studies indicate that the use of social media (such as Facebook) through online interactions affect an individual's self-concept, especially their own gender and ethnicity (Grasmuck, Martin, & Zhao, 2009;Guzzetti, 2006). Particularly, social identity theory posits that group membership(s) to which an individual belongs partly affects one's self-concept (McKinley, Mastro, & Warber, 2014;Turner & Oakes, 1986). However, social media give people means to explore their self-expression and allow users to create multiple selves (Davis, 2012;Michikyan & Subrahmanyam, 2012;Michikyan et al., 2014). Consequently, young people have the opportunity to take an online identity that is different from their offline one.
In essence, social media provide what Erickson (as referenced in Turkle, 1999, p. 284) would have called a "psychosocial moratorium" a central element in how he thought about identity development in adolescence. From this perspective, some researchers argue that social media provide an environment where young people present their real personalities on online sites such as Facebook rather than their idealized self (Gosling, Augustine, Vazire, Holtzman, & Gaddis, 2011;Michikyan & Subrahmanyam, 2012;Michikyan et al., 2014). However, Michikyan and Subrahmanyam (2012) argue that young people presents their self, on Facebook, in a variety of ways such as their real self (aspects that are authentic) and ideal self (who one wishes/desires to be), but they also presented their false self (aspects that are not fully truthful).

Sample
Data for this article are from a study on "Youth issues in Mauritius", which were collected from all around the island by volunteers -working either as teachers in schools or youth officers in youth centres -under the supervision of the main author. The research team followed social research ethical guidelines such as voluntary participation, anonymity in data collection and reporting; and permissions of all the stakeholders involved in the study were sought before the data collection.
A stratified sample by age, sex and location (urban or rural) was designed. A pilot test with 20 questionnaires was undertaken with minor changes made in the final self-administered questionnaire. The data collection were completed towards the end of the year 2015. In total 517 answered the questionnaire. The participants were aged between 14 and 18 (69.2%), 19 and 25 (30.8%). 77.6% came from urban areas and 93.2% were students. Females represented 67.5% of the total sample. Only 2.3% were married, and 94.5% lived with their parents.

Measures
Using the above-considered theoretical frame on identity and social media, this study therefore hypothesized differences in general terms between males and females, and between different ethnic groups in Facebook visit and identity estimations. In particular questions measuring Self-Presentation on Facebook Questionnaire (SPFBQ), that is Facebook visit and identity based on developmental theory of the self (Michikyan et al., 2014) were included. For instance, participants were asked to rate how often they visit Facebook on a six-point scale from 1 "have it open all the time" to 6 "less than once a week". The objective, here, was to find out how Facebook use varies by gender and ethnic groups. The respondents' Facebook identity fluidity was measured on a five-point scale, ranging from (0) "never" to (4) "always", through 21 items/statements looking at online and offline self-presentation. The 21 items/statements were used for the objective to find out the role of gender and ethnicity on young Mauritian identity and Facebook behaviour.

Design
The questionnaire included three genders (female, male, and other) and eight ethnicity groups [Sino-Mauritian, Creole, Franco-Mauritian, Hindu, Muslim, Mixed (of mix ethnicity), other, don't know]. Due to too few participants in each cell of gender by ethnicity grouping, only combinations of female and male by Creole, Hindu, Muslim, and Mixed were used for the statistical analyses. Furthermore, due to too few participants in each cell of gender by ethnicity grouping answering the Facebook identity items, variance analyses were performed for Gender and Ethnicity respectively. Accordingly, 322 participants were included in Gender-analysis (32% males and 68% females) and 297 in Ethnicity-analysis (Creole 12%, Hindu 66%, Muslim 18%, Mixed 4%).
IBM SPSS Statistics 22 software was used for the analyses. Two MANOVAs (multivariate analyses of variance) were calculated, for each Gender and Ethnicity variable, due to several items included in Facebook identity measure, and two ANOVAs for the measure "How often do you visit Facebook?"

Gender on Facebook visit and identity
ANOVA showed a main effect of Gender on "How often do you visit Facebook?", F(1, 457) = 18.31, p < .001, η 2 = .04. As can be seen in Figure 1, males were reported to visit more Facebook than did females. This part of the analysis therefore shows that the patriarchy maps onto the use of Facebook.
MANOVA showed a main effect of Gender, Wilks λ = .78 (21, 300) = 4.09, p < .001, η 2 = .22, associated with following statements: Who I want to be is often reflected in the things I do on my Facebook profile (e.g. status posts, comments, photos, etc.) (p < .01); I post information about myself on my Facebook profile that is not true (p < .01); I try to impress others with the photos I post of myself on my Facebook profile (p < .01); On Facebook I can try-out many aspects of who I am much more than I can in real life (p < .01); Who I am online is similar to who I am offline (p < .01); I don't have a clear idea of what I want to be so I use Facebook to figure it out (p < .01); I try-out different looks or attitudes on Facebook to see how people react (p < .01); I am a completely different person online than I am offline (p < .01); I change my photos on my Facebook profile to show people the different aspects of who I am (p < .01); I feel like I don't really know what I am all about or who I am so I use Facebook to try-out different things that can help me figure it out (p < .01); I compare myself to others on Facebook (p < .01); Sometimes I feel like I keep up a front on Facebook (p < .01); I only show the aspects of myself on Facebook that I know people would like (p < .01); I post things on Facebook to show aspects of who I want to be (p < .01); I feel like I have many sides myself and I show it on my Facebook profile (p < .01).
As can be seen in Figure 2a, females compared to males felt more that they are similar persons online as offline; as well as that their Facebook activity represented who they are. On all other dimensions/items, males were however shown to have a stronger Facebook identity than did females (see Figure 2b). The gender differences prevailing on the society therefore also maps onto the social media such as Facebook.

Ethnicity on Facebook visit and identity
For ANOVA we calculated Levene's test to assess the equality of variances, due to unequal number of respondents in Ethnicity groups. The test showed following statistics F(3, 424) = 2,11, p = .10. Given the non-significant p-value we can conclude that there is no difference between the variances in Ethnicity groups. For MANOVA we used Pillai's criterion instead of Wilk's lambda (λ), because it is the most robust criterion when cell sizes are unequal.
ANOVA showed a main effect of Ethnicity on "How often do you visit Facebook?", F(3, 427) = 11.59, p < .001, η 2 = .04. As can be seen in Figure 3, Mixed were reported to visit most Facebook followed by Muslim, Creole and Hindu respondents. I am all about or who I am so I use Facebook to try-out different things that can help me figure it out (p < .05); I compare myself to others on Facebook (p < .05); Sometimes I feel like I keep up a front on Facebook (p < .01); I post things on Facebook to show aspects of who I want to be (p < .05); I feel like I have many sides myself and I show it on my Facebook profile (p < .01); I feel like using Facebook has allowed me to question who I am whereas before I didn't really care (p < .05).
Both Figures 4a and 4b show clearly that Hindu participants had the weakest Facebook identity (on all dimensions/items except for "Who I want to be is often reflected in the things I do on my Facebook profile"). They were also the ones who most infrequently visited Facebook (see Figure 3). As can be seen in Figure 4a, Creole and Muslim groups had the strongest Facebook identity. However, on five dimensions/items Creole, Muslim and Mixed participants showed similar Facebook identity (see Figure 4b). This part of the results clearly show that there is a significant difference in ethnic identification on Facebook among different ethnic group members, which needs to be further explored in future studies.

Discussion
The aim of this study was to investigate the role of gender and ethnicity on Facebook visit and identity among young people living in Mauritius. In almost all societies, gender and ethnic differences, as well as some similarities, are apparent in Social Media (Herring & Kapidzic, 2015;Michikyan, Subrahmanyam, & Dennis, 2015;Schradie, 2012). In this study, males were shown to visit more Facebook and had a stronger Facebook identity than did females. Similarly, from her study on the usage of SNSs in Mauritius, Gokulsing (2014) argues that gender maps the virtual space in terms of self-identity and social interactions and activities. The finding on the dominance of male users of Facebook in Mauritius is somewhat similar to the study undertaken in Turkey by Mazman and Usluel (2011). However, it contradicts that from the American context, where it was found that more females than males use Facebook (Herring & Kapidzic, 2015). Within a patriarchal context like Mauritius and Turkey, it is not surprising to find out the use of Facebook is gendered. New technologies that are being integrated into existing household relations and youth activities, reflect the highly gendered relations that are already operating within different social structures of a society.
Unlike the study undertaken by Michikyan et al. (2014), the case study from Mauritius found a gender difference on online and offline identity, where females compared to males considered themselves to be more similar as persons -online as offline. Why do more boys present different online and offline self-identity in Mauritius? This question requires a further study. Some internet anthropologists argue that young people change identity online from offline for benign reasons, such as for fun, pleasure, curiosity and play (Lewis, 2004). However, some development psychologists posit that there are some profound psychological reasons behind such act; such as personality development and motivational psychological factors that push young people towards online and offline identity change (Manago et al., 2008;Michikyan et al., 2014;Salimkhan, Manago, & Greenfield, 2010). It is also known from some previous studies, in context outside Mauritius, that girls tend selfdisclose (the sharing of personal thoughts, feelings, and experiences) more than boys do (Lenhart, Madden, Macgill, & Smith, 2007;Valkenburg, Sumter, & Peter, 2011).
Regarding ethnicity, the frequency of Facebook visit of Hindu participants were shown to be lower than Mixed participants. This is surprising as the Hindu ethnic group is a not only a demographically dominant, but is also a group that is very visible in various domains of the Mauritian society (such as social, cultural, economic and political).
Given that the vast majority of the research participants were still in their formal education years, one possible explanation could be that, the Hindu participants consider Facebook usage as a hindrance to their study. This hypothetical possibility needs to be tested through a further study. Indeed, Mauritius has a competitive education system, where the students and parents from the Hindu ethnic group invest a lot to top the competition (Maurer, 2014;Ng Tseung-Wong, 2013). However, studies from different country contexts, have found a mixed relationship between student's academic performance and Facebook usage. In a study from Netherlands, Kirschner and Karpinski (2010) report that Facebook users reported having lower GPAs and spend fewer hours per week studying than nonusers. Whereas, Mahmood and Farooq (2014) from Pakistan and Michikyan et al. (2015) from America, found that there was no relationship between student's academic performance and Facebook usage. Such mixed findings therefore call for more research on such particular issues from more a variety of different country contexts, in order to have a better understanding on the dynamic relationship between the variables.  Through a further study in Mauritius, it would be also be interesting to explore why research participants from Hindu ethnic group showed weakest identity. Some identity theorists argue that identities make up one's self-concept that is variously described as what comes to mind when one thinks of oneself in relation to others (Oyserman et al., 2012). From this particular perspective, it could be argued that the Non-Hindu ethnic groups (Creole, Muslim and Mixed) feel the need to affirm a strong ethnic identity online because of the demographic and political dominance of the Hindu ethnic in Mauritius. Several studies carried out from different parts of the world have shown ethnic identification is stronger amongst ethnic minority than majority group members (Nazroo & Karlsen, 2003;Ng Tseung-Wong, 2013;Pickard, Barry, Wallace, & Zeigler-Hill, 2013;Verkuyten, 2005). In particular, Jones and Galliher (2007) opine that identity development for young people from minority groups may be complicated by the additional task of incorporating ethnic identity into one's overall sense of self.

Conclusion
In line with the general hypotheses of this study and as a conclusion, it can be said that both ethnicity and gender have noticeable effects on young people's Facebook visit and identity. This is mainly so, due to the role differences in gender and ethnicity in a society (Bauer & McAdams, 2004;Boz et al., 2016;Brody, 1993;Cohn, 1991;Fischer & Manstead, 2000;Lewis, 2004;Subrahmanyam & Smahel, 2011;Tani, Peterson, & Smorti, 2014;Turkle, 1999), like Mauritius. In the digital era, identity as self-presentation is therefore no longer confined to face-to-face interactions, but also to online interactions through Social Media, such as Facebook (Yang & Brown, 2016). Countries with emerging economies, like Mauritius where large-scale investment on new internet-based technologies are happening, need continuing research to examine young people's cyber behaviour, culture and identity. New generation of young people have more possibilities through the digital media to negotiate identities, and perhaps more opportunities challenge certain conservative moral discourses on gender and ethnicity. The central question is therefore, how long can certain key agents of the Mauritian society continue with their discourses on 'pure' identity based on gender and ethnicity, particularly in this technology-driven era.