What Drive Consumers to Spread the Word in Social Media ?

The on-line social networks or social media facilitated the interactions and the information sharing between the Internet users. The marketers aware of the importance of the word of mouth and recommendation, tried to use the social networks to contact consumers and encourage them to promote their brands by playing the role of market mavens. In this study we tried to identify the motivations or the determinants of the word of mouth on the on-line social networks. For that purpose, we presented an integrator model which includes psychological and social factors to understand this phenomenon. We conduct exploratory and confirmatory analysis to ascertain friability and validity of measures. We used PLS analysis to test hypothesis. Findings and managerial implication are also presented.


Introduction
Social Web constitutes an inevitable opportunity for companies to be closer to their target and to increase their visibility.Indeed, Web 2.0 or social Web allowed to merge the technology and the human being.Thanks to its structure easily accessible to the users, Web 2.0 allows any user of Web to discuss, to create and to share content.Web 2.0 indicates all the uses of www based on the contents generated by the user, the networking and the communities allowing the Internet users to appropriate the new features of Web (Mayol, 2009).
Web 2.0 possesses an open structure which allows a real time communication between the Internet users themselves and between users and the company.Social Web is not any more the privilege of the companies which control the transfer of the information but it is more and more accessible to consumers who use it to create and spread content on products, brands and services.Web 2.0 presents new opportunities for the word of mouth through blogs and social networks (Goldsmith and Horowitz, 2006).
The social media are platforms intended to facilitate the interaction between the individuals.They are based on internet applications which facilitate the creation and the information sharing by users (Khan and Khan, 2012).Neff (2009) make a link between the social platforms like facebook and the natural environments.They consider that these platforms are ecosystems which are creating and nourishing relations of exchange between the company and the consumer and between the consumer and the contents which he creates and shares.
The on-line social networks changed profoundly the way with which the information propagates.Indeed, the information sharing became easier and easier.Several companies wanted to benefit from this phenomenon by creating «fan pages «on the social networks.The creation of a facebook page in the effigy of a brand allows to act as lever of fame and interaction with the consumer.(Palmer and Koenig-Lewis, 2009).By sharing, the company can touch at the same time the consumer fan of the page who shares its contents and his circle of friends (Veghes and Carmen, 2009).
The technologies of Web 2.0 facilitate the co-creation of the value by the buyer and the seller.Indeed, the on-line communities allow the consumer to become a co-builder of experiences (Royo-Vela and Casamassima, on 2011).The interactive nature of the social media allows not only the companies to share and to exchange the information with the consumers but it also allows consumers to exchange information between them.Through the use of the social media, companies can tie relations with the existing and potential customers (Khan and Khan, 2012).
With the advent of Web 2 .0 and social networks, companies face a real dilemma.On one hand companies can benefit from the accessibility of the social networks to touch a large number of consumers with relatively low costs (Croft and Hiwise, on 2008).They can be also attracted by the availability of the information about lifestyles and preferences of the members of the social networks to target them better.Nevertheless, they have to take into account the risk of intervention felt by these consumers and which is widely condemned by these (Boyd and Ellison, on 2007).
The consumers who recommend voluntarily the products of the company offer a precious help to this one by avoiding them hurting the consumers by too intrusive commercials because the consumer who recommends or diffuses a word of mouth is going to act as intermediate relay between the company and his circle of friends.(Royo-Vela and Casamassima, 2011).Lee and al, (2006) postulates that the research on the word of mouth in social networks deserves to be deepened.According to Lee and al (2011) the previous researches focused on the technological factors such as the ease of use or the speed of navigation to explain the activity of information sharing on Web.William and al (2010) indicates that the recommendation on the social networks constitutes a new type of word of mouth and merit more deepening to understand what urges the consumers to recommend on the social networks.
So, except these this mentioned factors, we can consider that the literature does not inform us enough about the relevance of the explanatory factors of information sharing through the social networks and Web 2.0.As a consequence, it seems to us interesting to verify the possibility of adding new factors susceptible to explain the phenomenon of recommendation on the on-line social networks as the affect of the consumer and the situational factors.
It seems interesting to identify the links between, on one hand, the psychosocial factors and the social factors and on the other hand, the intention of recommendation on the social networks.The principal question research is: What makes consumers recommend on social networks?

Intention of Recommendation in Social Media
The electronic word of mouth is defined as «an evaluative informal communication between two or several persons about the characteristics of a brand, a product or a service and which is spread on the Internet" (Carl, on 2006).In this study we are going to be interested in particular type of the word of mouth: the recommendation of products about the on-line social networks.
The recommendation of products / service on the social networks can take the form of sharing of links, photo, and videos on the profile of the user.Mak et al (2011) postulated that the concept of electronic word of mouth drew a lot of attention these last years because of the growth of the internet and the popularity of the e-commerce.Indeed, the electronic word of mouth is considered as an extension of the interpersonal communications at the new age.Therefore, Brown and al (2007) suggest that the existing theoretical frame can explain the electronic word of mouth.However it is necessary to enrich them to end in a relatively sufficient understanding of the phenomenon.
The consumers usually try to fill their lack of expertise in certain categories of products by the recourse to the recommendations transmitted by the other persons.The recommendation is an information source which supplies a personalized advice or not and which allows the consumer to make his process of purchase more effectively (Lopez and al, 2010).
So, the intention of recommendation on the social networks can be presented as the subjective probability which has an individual to undertake an activity of recommendation of products on the social networks (Okazaki, 2009).Boyd and Ellison (2008) define the social networks as «Web-based services which allow the individuals to build a public or semipublic profile articulating a list of users with whom the user interacts ".
According to Riegner (2007), the social networks can be considered as a platform or a space convenient to the development of word of mouth that they are based on the exchange between Internet users having several points in common.Lee and al (2011) argue that the on-line social networks profoundly changed the transmission and the exchange of the information.The word of mouth benefited from this improvement, indeed, the word of mouth and more exactly the recommendation of products is not any more transmitted to some persons whom know the consumer but it is passed on in a wider circle.Ajzen and Fishbein (1980) define behavioral intention as "a measure of the likelihood that a person will engage in a given behavior".Ajzen (1991) 'Intentions are assumed to capture the motivational factors that influence a behavior.According to the theory of reasoned action, a person's intention is a function of two basic determinants, one personal in nature (attitude toward behavior) and the other reflecting social influence (subjective norms).

Attitude toward Recommendation of Products in the Social Media
According to the theory of reasoned Action by Fishbein and Ajzen (1975), the behavioral intention is influenced by the attitude which is influenced by the subjective norms itself.Lassus, on 2003) defines the attitude as «a psychological tendency which expresses itself by the evaluation of a specific entity according to a certain degree of favor or disfavor».So, the attitude to the recommendation of product on the social networks is defined psychological reaction which expresses itself by the evaluation of the activity of recommendation according to a certain degree of favor or disfavor.Boss (1993) suggests that the attitude develops further to several positive experiences.This favorable attitude can result in a positive word of mouth which constitutes a logical reaction of a set of pleasant experiences.

Motives for Product Recommendation in Social Media
According to Kahle and Kennedy (1989) the cognitive social theory postulates that the personal factors (psychological factors, demographic factors) and social factors (influence of the family, the friends) interact to influence the behavior of the consumer.Cheema and Kaikati (2010) postulates that the consumers pass on a word of mouth for social and\or psychological motives.Indeed, these consumers can be motivated by the sense of the obligation to help others and probably by the pleasure inherent to the discussion about products or to reach certain social status.Walsh and al, (2004) underline that the recommendation of products on the social networks is has a Journal of Marketing Research & Case Studies 4 psychological nature, in fact it serves as a mean of expression of personality's traits.

Psychological Factors of Product Recommendation in Social Media
Self Esteem Darley (1999) defines the self-esteem as «a previous attitude which consists in saying to itself that we have some value ".Darley (1999) argue that the individuals having high self esteem tend to look for more information about products from several sources.These individuals also, use their knowledge to persuade the others of the superiority of their own choice.
According to Wojnicki (2006), the consumers can propagate the information on online community with the aim of gaining the attention and the gratitude of the members by assuming an expert's role.Fuller (2006) adds that these persons share their know-how with other members to obtain gratification and embellish their ego.Hence, we posit: H1a: self esteem exerts a positive effect on attitude toward recommendation on social networks.
H1b: self esteem exerts a positive effect on intention of recommendation on social networks.

Intrinsic Motivation
The intrinsic motivation is an interest, a curiosity, a tendency to investigate several sources of stimulations felt about an activity.The consequences of intrinsic motivation can be a satisfaction or a positive attitude towards the activity.The activity is considered as intrinsically pleasant without taking into account the effort to achieve it or the profits which ensue from it (Fagan and al, 2008).Hennig-Thurau and al. (2004) indicates that the intention to share information on on-line networks can be partially influenced by the pleasure ensuing from this activity which can be considered as pleasant and funny.Workman and Studak (2005) stipulate that perceived pleasure is a state of mental stimulation characterized by a high level of concentration and a big interest in the execution of the activity.Walsh and al (2004) considers that the information sharing possesses at the same time a social value and an entertaining value.The consumers, who share information with the others, want to express the feeling of enjoyment associated with a positive experience.Therefore, our next two hypothesis are: H2a: intrinsic motivation exerts a positive effect on attitude toward recommendation on social networks.
H2b: intrinsic motivation a positive effect on intention of recommendation on social networks.

Product Involvement
Mitchell (2002) defines involvement as "an internal states variable that indicates the amount of arousal, interest or drive evoked by a particular stimulus or situation".It results in certain forms of research for product, processing of information and decision-making.Higie and Feick, (1989) postulate that involved consumer tends to inquire regularly, to look actively for information about products, to pass on the information.According to Elliot and Speck (2005), the product implication influences the attitude to the Web site.Bertrandias and Goldsmith, (2006), postulate that one of the most important motivations which can explain the information sharing is product implication.
Indeed, the involved consumers are interested in the product even if they do not envisage the purchase.Okazaki (2009) demonstrated that the product involvement (mobile phones) influences positively the attitude toward participation in promotion Company and the intention to recommend the product.
H3a: Product involvement a positive effect on attitude toward recommendation on social networks.
H3b: Product involvement exerts a positive effect on intention of recommendation on social networks.

Interpersonal Connectivity
Interpersonal connectivity defined as the "social benefits derived from establishing and maintaining contact with other people such as social support, friendship, and intimacy" (Dholakia et al, 2004).
According to, Hung and al (2010), the contribution of the users is an essential action for the development of sites 2.0.With the aim of developing the contribution of the users, companies are brought to facilitate «the socialization of these.The socialization is a process intended to improve the skills and the knowledge of consumers (Churchill, 1978).
Within the framework of Web 2.0, the socialization of the consumers is the process by which the consumers learn rules, standards and skills which help them to participate in 2.0 web sites.The authors specify that companies can launch the process of the socialization but the most important part falls especially to users (Hung and al, 2010).
The consumers can be inclined to participate in virtual communities or social networks with the aim of staying in touch with their circle of acquaintances.So the consumers can post comments on products and brands to strengthen their social links and assert their membership in their reference group (Lee et al, 2006).
The consumers can be inclined to participate in virtual communities or social networks with the aim of staying in touch with their circle of acquaintances.So the consumers can post comments on products and brands on social networks to strengthen their social links and assert their membership in their reference group (Pelling et al, 2009).Accordingly, we hypothesize: H4a: interpersonal connectivity exerts a positive effect on attitude toward recommendation on social networks.
H4b: interpersonal connectivity exerts a positive effect on intention of recommendation on social networks.

Social Influence
The social influence was declined in the form of several concepts, Ajzen (1991) in its theory of the reasoned action, proposed the concept of "subjective norms ".The concept of subjective norms is defined as "an individual's perception of whether people important to the individual think the behavior should be performed".Thompson and al (1991) speak about social factors which reflect the appropriation by the individual of the standards of his reference group conditioning his behavior.Internet advertising, as other forms of mass media channels, conveys and promotes cultural and lifestyle messages through associated status, depiction of ideal users, stimulate positively social response and attitude toward advertising and online products.
In the case of using social networks to recommend products, a person may believe that most people who are important to him think he should recommend or he may believe that they think he should not (Pelling, 2009).Li (2011) argue that the social influence, the concept proposed in the theory of the reasoned action by Ajzen and Fischbein (1975) influence positively the intention to share music video on social networks.Hence, we propose that: H5a: Social influence exerts a positive effect on attitude toward recommendation on social networks.
H5b: Social influence exerts a positive effect on intention of recommendation on social networks.

The Relation between Attitude toward Product Recommendation in Social Media and Intention of Recommendation
Attitudes refer to the degree to which a person has a favorable or unfavorableevaluation or appraisal of the behavior in question (Ajzen, 1991).Bressoud (2008) mentionned that the theory of the reasoned action is a reference to model the link between the attitude and the intention.The author adds that the essential interest of the theory of the reasoned action is its capacity to be empirically confirmed by several researches.
Seock and Norton, (2007) also found that attitudes toward the clothing web sites had a direct, positive effect on the intentions to search for information at this web sites as well as intentions to purchase clothing items from this web sites after finding the items there.
The likelihood of buying a brand or purchase intention has been shown to be influenced by attitudes to wards advertising and attitudes towards brands.According to Okazaki (2009), the Internet users having a favorable attitude to the information sharing on the social networks, are more inclined to emit a positive word of mouth.Okazaki (2009) demonstrates that the attitude to the participation to the viral campaign influences positively the intention of participation and willingness to make referrals.Based on the above arguments, we propose the following hypotheses: H6: Attitude toward recommendation exerts a positive effect on intention of recommendation on social networks.

Research Design and Methodology
We made a survey by questionnaire with 150 women who use social networks whose age varies between 18 and 35 years (58 % between 18 and 25 years 42 % between 26 and 35 years).The majority of the respondents are students (65 %), 20 % are active and 15 % are in search of a work.
The questionnaire was administered online and face to face.We focused on women who are members of groups or pages of social networking (facebook) specialized in fashion.

Measures
The four-item intention of recommendation measure was adopted from the work of Nefzi (2008).Respondents were asked to indicate the degree of their agreement or disagreement with each item, using a five-point Likerttype scale (5=totally agree, 1=totally disagree).
Attitude toward recommendation was operationalized on the basis of s3 items adopted from a scale developed by Okazaki (2009).Each of the items was accompanied by a 5 point response format, ranging from 5=totally agree to 1=totally disagree.The measures of interpersonal connectivity, product involvement, intrinsic motivation, self esteem and social influence were respectively adopted from the works of (Okazaki, 2009), Derbaix (2008) and Fuller (2006).

Ascertainments of Fiability and Validity
Cronbach's alpha statistic was used to assess the reliability of the scales while exploratory factor analysis with Varimax rotation was used to check unidimensionnality of the scales (Churchill, 1979).The discriminant validity and the convergent were tested via confirmatory factor analysis (CFA).Fit indices for all scales were higher than the commonly accepted values (appendix).

Data Analysis Process
Partial Least Squares (PLS) (Chin, 1998a(Chin, , b, 2001) ) was employed to estimate the model.PLS is a second generation structural equation modeling (SEM) technique (Chin, 1998).PLS does not rely on normality assumptions.PLS employs bootstrapping to test the significance of relationships so it work well with nonnormal data (Bontis et al, 2007).

Results
According to the boards which appear in the appendix 2, the global indices of the global adjustment GoF what is of the order of 0,986, what testifies of the good adequacy of the model (Espozito and al, on 2010).Besides the technique of bootstrap demonstrates that this fit indices is stable.
To establish the significance and the stability of the parameters we used the technique of bootstrap (200 bootstrap with the size of the sample = 200).R2 of the attitude is of the order of 44, 3 % what suggests that the model explains almost 45 % of the variance of this endogenous variable.Also R2 of the intention of recommendation is of the order of 60% what implies that the model explains almost 60 % of the variability of this construct.According to Tenenhaus and al. (2005) a value of R2 superior to 0,4 is considered as relatively good.So we can consider that the results suggest a good predictive value of the model.
The results also show that the self-esteem influences positively the attitude toward recommendation and it at the 5 % level (t = 2,530 p 0,012).Also, the intrinsic motivation impacts positively and significantly the attitude to the recommendation at the 5 % level (t = 5,535; p = 0,000).Besides we did not find significant links between the social influence, the implication and the connectivity on one hand and the attitude toward recommendation on the other hand (Appendix).
Besides, the hypothesis which stipulates a positive link between the interpersonal connectivity and the intention of recommendation, is confirmed (t = 5,967: p = 0,000).Also for the hypothesis which stipulates a positive link between the implication in the category of the product and the intention of recommendation (t = 2,318; p = 0.022).The results also show that the social influence impacts positively on the intention of recommendation (t = 2,899; p=0,004).Finally, the results of the analysis PLS show that the attitude towards the recommendation impacts positively and significantly the intention of recommendation at the 5 % level (t = 3,396; p = 0,001) (Appendix).

Discussion
The PLS analysis demonstrates that the attitude to the recommendation is strongly influenced by the intrinsic motivation and the self-esteem.The favorable attitude toward recommendation leans on affective reactions resulting from an appreciated experience on the social networks.Indeed, according to Mehrabain and Russel (1974), the environment generates feelings which are going to act on the evaluation of the experience.The experience on the social networks is considered as hedonist experience associating positive feelings which can influence positively the attitude to the recommendation.Walsh and al (2004) asserts that the recommendation possesses a social value and entertainment value.Indeed, the consumers pass on information to the others because they want to share their enjoyment with these.They also want to express their positive feelings associated with a successful experience of purchase.So these consumers seem to be motivated by the pleasure which they feel by sharing information.
The self-esteem makes is the favorable feeling resulting from the good opinion which we have of one's value.According to Cally (2011), certain consumers do not wish to be as everybody, for it they decide for example to buy luxury items because they are in search of prestige.On the social networks, the activity of recommendation can be considered as a means of asserting one's personality.
An activity which allows the social media user to be admired by his circle of friends.Bertrandias and Goldsmith, (2006) stipulates that the ready-to-wear clothing is a product which possesses a social and symbolic value as far as it exposes consumer other judgments making it both a socially and emotionally risky product and it generates easily a word of mouth (speaking about fashion trends).

Journal of Marketing Research & Case Studies 8
So the Internet user, who perceives a positive image of himself, develops a favorable attitude toward recommendation because she allows him to be appreciated by her circle of friends on the social networks.Indeed, the activity of recommendation represents a symbolic function besides its utilitarian function (Cardoso, on 2004).
Thanks to recommendation on social media, the user can assert himself as an active member of the community; he can assert also his membership of his reference group (in our case the modern women and the elegant woman who are interested in the fashion.reference groups can also exert direct and indirect forms of influence through shared experiences between members and non-members who want to join the group in the future (Lawrence, 2006) We can illustrate this by some pages or groups on facebook which are dedicated to women clothes: "Mode et beauté " and "100 % feminine" (we contacted certain members of these groups to answer the questionnaire).In the same vein, Bertrandias and Goldsmith (2006) stipulate that the consumers who recommend products need to be «individuated ".In fact, individuation is willingness be unique or different the others.The recommendation allows these consumers to distance themselves from the others and it allows them to satisfy their need of individuation.
The results also show that the intention of recommendation on the social networks is influenced by interpersonal connectivity, product involvement, social influence, and the attitude toward to recommendation.Indeed, the attitude toward recommendation is considered as an important determinant of the intention of recommendation.What confirms the results of the previous researches stipulating a significant link between the attitude and the behavioral intention (Ahrholdt, 2011).Brown and al, (2007) stipulate that the Internet users who share information on the social networks want to act as opinion leaders; they possess a certain expertise in a specific domain gained by their involvement.Walsh and al (2004) also adds that these Internet users feel obliged to share the information, the sharing of the information gets them a big pleasure and they tend to be altruistic and to want to help the other consumers.Laughlin and McDonald (2010) indicate that the information sharing for these consumers seems to be also motivated by their desire to be appreciated by their acquaintance.The results also show that the social influence impacts positively on the intention of recommendation.Indeed, the consumers trying to obtain the recognition others and to make good impression are more inclined to share information on the social networks.
According to Wang and al, (2010), the search for "the social approval ", urges the consumers to supply more time and effort in the activity of recommendation.Bertrandias and Goldsmith, (2006) indicate that the propensity of the consumers to use "the social signs» to define the behavior influences their intention to share the information.So the consumers who think that social influence is important, pay attention on the reactions and on the opinions of the others and care strongly about what the others think of them.These consumers use the recommendation to become better integrated into their social group.
The absence of links between the intrinsic motivation and the self-esteem on one hand and the intention of recommendation on the other hand could be explained by the mediating effect that could practice the attitude to the recommendation between these variables and the intention of recommendation (Cardoso and al, on 2008).Also, the important influence of the interpersonal connectivity on the intention of recommendation is explained by the importance of the process of socialization which is the willingness of the Internet users to maintain narrow social links with their acquaintances through the interaction on the social networks (Okazaki and al, on 2007).According to Laughlin and McDonald (2010), the consumers can execute several activities on the social networks, they can look for information, comment on the publications of their friends but those who pass on the information want to maintain narrow links with friends.

Managerial Implications
In this study we tried to identify the factors which influence the intention of recommendation on the electronic social networks.The theoretical frame which we proposed incorporates psychological and social factors.This research also presents a managerial interest.Indeed, with the distribution of the contents generated by the users, the multiplication of blogs, the increase of the number of the members of the social networks, the extent of the word of mouth widened.Okazaki (2009) asserts that the social networking became one of the most important activities on Web.
Companies which can use social networks to implement new promotion techniques of their products.The electronic word of mouth displays as a promising medium allowing the companies to increase their visibility on Web.Rigner (2007) postulates that the social networking can be considerate as a real lifestyle for those who use it mainly to exchange information on their center of interest.This is could be an an opportunity for the companies which want to promote their products and services on these social networks by appealing to these "consumers 2.0".Marketers are brought to identify «the opinion leaders» who are characterized by high self esteem, a strong involvement and a favorable attitude to the recommendation.These consumers tend to be «active social networks users», they can help companies to increase their visibility on these networks (Park and Feinberg, on 2010).
These consumers can help the companies by development of new products, the diffusion of the information on products and informing the company about the feedback of the other consumer (Thomas and al, on 2007).

Limitations and Future Research
Although our study makes important theoretical contributions to the understanding of the motivations of intention of recommendation on social networks, it has limitations.
Fist, the category of the product, (woman clothing) directed us to the constitution of a sample consisted of women who use social networks (the majority are students) what reduces the external validity of the study.The second limitation relates to the use of a non-probability sampling (convenience sample) procedure and a relatively small sample.
Besides, it would thus be necessary to lead a study with diverse categories of products with a more representative sample of the population of the Internet users.Besides, it would be relevant to introduce other variables such as the extrinsic motivation, to increase the explanatory power of the model.