The self as a cause and effect of liking brands on Facebook: An experimental study Anniek Eigenraam | 10001513
Supervisor: Dian de Vries
Research Master Communication Science University of Amsterdam
Abstract
This study combines previous research on impression management, self-presentation and identity shifting by the use of brands on the one hand and social networking sites on the other hand. If social networking sites, as well as brands, are used for impression management, than how are brands on social networking sites used for self-expression and self-enhancement? Based on two experimental studies it is concluded that liking brands on Facebook enhances the self-concept, specifically concerning the personality traits that are associated with the liked brand’s identity, and that the capacity of a brand to communicate a desired image is not a general concept that can improve the likeability of a brand to be liked, or enlarge the self-enhancing effect of liking a brand. The studies provide useful insights in causes for and effects of liking brands on Facebook in the light of impression management theory.
The self as a cause and effect of liking brands on Facebook: An experimental study Among both practicing marketers and marketing and communication researchers, online marketing on social networking sites (SNS) has caught attention. This marketing strategy has more and more become an important part of many companies’ marketing
activities (Cvijikj & Michahelles, 2013), especially in targeting youth (18-25) (Serazio, 2013), among whom SNS such as Facebook, Twitter and YouTube, are especially popular
(Subrahmanyam, Reich, Waechter & Espinoza, 2008). Companies can manage their own brand page (Cvijikj & Michahelles, 2013), through which interactions, online word-of-mouth, posts and advertisements spread costlessly through the Facebook users’ networks (Tucker, 2011). Additionally, word-of-mouth is shown, not just to be a costless way of spreading a brand, but also to be an effective way of persuading new consumers (Berger & Schwartz, 2011).
In order to effectively make use of these benefits of marketing on SNS, advertisers need to know how to have their consumers engage and positively identified with the brand(page). As a result, identifying the relevant factors that play a role in the decision making process of engaging in brands on SNS, for example through liking a brand page, is relevant for marketers (Berger & Schwartz, 2011). Therefore this study is focused on what type of brands are appealing for users to like on SNS Facebook, and what underlies this decision making process to like or not to like brands.
Prior research has shown that endorsing, using or talking about brands and products offline (e.g., Belk, 1988; Aaker, 1999; Berger & Heat, 2007; Park & Roedder-John, 2010; Reed, Forehand, Puntoni & Warlop, 2012) and online (Muntinga, Moorman & Smit, 2011; Belk, 2013) and sharing (content) on SNS (e.g., Ho & Dempsey, 2010; Schau & Gilly, 2014; Zhang, Jiang & Carroll, 2010; Zhao, Grasmuck & Martin, 2008) is done for two shared motivations, namely self-expression and self-enhancement. The self-expression motive refers
to the communication of one’s personality to the outside world, by the selection and use of brands and products that reflect the users’ personality (Reed, Forehand, Puntoni & Warlop, 2012). The self-enhancement motive refers to people’s search for activities that reinforce or reflect positively on the self-concept (Reed, Forehand, Puntoni &Warlop, 2012; Wojnicki & Godes, 2008). This indicates that brands that have a capacity to signal a desired identity would be preferred to use for self-presentation and self-enhancement motives than brands that don’t have this capacity. However, it remains unclear from the literature if the extent to which a brand identity has this capacity to signal one’s desired identity affects the likability of engagement with a brand. The first aim of this study is therefore to test if the capacity of a brand identity to communicate a desired identity influences the likeliness of a common form of online engagement, namely liking a brand on Facebook.
Besides identifying this potential factor that could make people like brands, it is interesting to see whether communicating a brand identity, in the case of this study by liking on SNS, affects the self-concept of the user, which the liking, seen the self-presentation and self-enhancement motivations of this communication is meant to do. If it seems a brand on Facebook has the capacity to make people shift in identity, this study would not only extent theory on brand identities and identity shifting, but might have practical implications for social media marketers as well. The second aim of this study therefore is to identify the effects of endorsing brands on SNS on the self-concept.
The self-expression and self-enhancement motives for using brands and using SNS provide us two aims for this study. In this research we’re interested in brand identity as a potential cause for liking brands, as well as enhancing the self-concept as a potential effect of liking brands on Facebook. In order to meet the formulated aims and to investigate this potential cause for, and potential effect of liking brands on Facebook, two experimental studies have been conducted.
Theoretical framework Impression management offline
The idea of people actively trying to present themselves in a way they would like others to see them is not a new and online specific concept. According to Goffman (1959) people engage in self-presentation through communicating signs and symbols they’d like to be associated with. Goffman (1959) refers to this self-presentation as impression management. In the offline world people manage their impression for example by engaging in certain
activities and talking in a specific way about certain things or dressing up with certain clothes, so that desired personality traits can be associated with them.
Impression management online
Krämer and Winter (2008) point out the suitability of SNS for carefully engaging in accurate impression management, more than real life, for SNS such as Facebook facilitate its users to communicate symbols and signs, for example likes, posts or shares, on a personal place that is connected to the users’ friends personal online places (Schau & Gilly, 2014). SNS are pre-eminently a suitable platform for impression management.
Brands as attributes for impression management
Brands can be seen as symbols that are intended to communicate what the company or product stands for (Urde, 2013). People can assign different meanings to brands and have many associations with brands. These associations include personality traits (Aaker, 1997). The whole of human personality traits associated with the brand are referred to as brand identity (Aaker, 1997). Aaker (1997) has distinguished five personality dimensions to categorize brands. Because of its personality traits brands are well suited for the use of impression management for they can communicate personality traits of a consumer.
According to Reed, Forehand, Puntoni, and Warlop (2012) impression management is a motive for the use and endorsement of certain brands. Using a brands identity for
impression management means buying or using a brand because it reflects certain personality traits of the user (Kirmani, 2009). A brand related impression management activity can for example be the communication of personality traits through using deliberately selected brands or products (Reed, Forehand, Puntoni &Warlop, 2012; Park & Roedder-John, 2010). Brands, due to their associated personality traits, offer people the opportunity to signal a personality trait of their own, and in this manner manage their impression (Berger & Heath, 2007). Brands, SNS and impression management
It can be concluded that brands and SNS have in common that they’re both used for impression management. SNS facilitate a suitable platform and brands offer the attributes for managing impressions. Accordingly, Wallace, Buil & de Chernatony (2012) point out
impression management is an important motive of Facebook users to like brands on Facebook. As briefly mentioned in the introduction the potential effects of differences in capacity of brands to communicate a desired identity on likelihood of brands to be liked, and the potential effect of liking a brand on the self-concept will be addressed in respectively the first and second study.
Study 1:
The self as a cause for brand likes on Facebook
As Leary and Kowalski (1990) put it, impression management activities are driven by how people think they are, and how people desire to be, or not to be. In engaging in
impression management, people will avoid to communicate an undesired identity, or undesired personality traits, to the world around them for people intend to communicate a desired image (Berger & Heat, 2007). When impression management is the goal for endorsing a brand, this brand must have been considered to have the capacity to communicate a desired image.
It seems that brands that are associated with a very desired identity are more likely to be used for impression management purposes than are brands that are associated with
neutrally desired and/or undesired personality traits, for brands with neutrally desired or undesired personality traits can never fulfil identity goals. This is confirmed by Reed,
Forehand, Puntoni and Warlop (2012) as they point out that stimuli, which can for example be brands, become linked to identities when they “reflect a desirable self-image, or embody the type of person that the consumer aspires to think, feel and be like” (p. 135). This leads to the first hypothesis.
H1: Brands with a desirable brand identity (vs. neutrally desirable brand identity) are more likely to be liked on Facebook.
This hypothesized relation between the likelihood of a brand to be liked and the identity of the brand come together in a conceptual model (see figure 1) and will be tested in an experiment.
Method Design
This study aims to test the conceptual model presented in figure 1, in an experiment using a single factor between subjects design with two levels (desirability brand identity: desirable vs. neutrally desirable). Participants are exposed to a screenshot of a Facebook brand page of a brand with either a very desirable or a neutrally desirable brand identity and asked to what extent they would be willing to like it. The first hypothesis, namely that brands with a desirable brand identity are more likely to be liked on Facebook than brands with a neutrally desirable brand identity, can then be tested.
Sample
In this experiment 119 people participated. Participants ranged in the age of 18 to 35, with an average age of 23 (SD = 2.50). 59.00% were female and the majority of the sample
has had higher or academic education, namely 82.60%. A link to the online experiment was spread via the SNS Facebook and e-mail among students of the University of Amsterdam and other emerging adults in the personal network of the researcher. The researcher has
specifically approached emerging adults in her personal network to ask for participation in the pre-tests. Participants of the pre-tests were not allowed to participate in the experiment as well, because they have some sort of inside information and might suspect what the purpose of the experiment is. To make sure the participants in the pre-tests are not the same as in the
experiment, the recruited participants for the pre-tests are explicitly asked not to participate in the experiment. Their names were listed and the online experiment has not purposely been directed to them.
Manipulation
To manipulate the desirability of the brand identities two pre-tests were done to check this desirability. In the first pre-test participants were shown a list of (brand) personalities, derived from a Dutch translated version of Aakers’ (1997) brand personality scale (Smit, Van den Berge, & Franzen, 2003) and were asked to indicate on a scale ranging from 1 to 10 to what extent they desire to have that personality trait for themselves, 1 meaning the trait is very undesirable and 10 meaning the trait is very desirable. The ratings of these (brand)
personalities provided the experiment with neutrally desired, undesired and desirable personality traits. Personality traits that scored an average rate between 8 and 10 were considered to be very desirable. Personality traits that scored between 5 and 6 were
considered to be neutrally desirable. Personality traits that scored below 5 were considered undesirable personality traits. The results of the first pre-test (N = 26) show that people want to be fair. In the first pre-test, being fair is shown to be the most desirable personality trait out of Aakers’ (1997) brand personality list (M = 9.04; SD = .84). Therefore fair is chosen as trait to manipulate the desirability of the brand in the manipulation of this experiment. For the
manipulation in this research, the smartphone brand Match Mobile has been developed. Match Mobile is a fictive brand for smartphones. The brand is fictive, which means
participants do not have any associations with the brand beforehand. The second pre-test is done to check this manipulation. In the second pre-test the logo of the brand was shown either with or without a Fairtrade logo, to manipulate the fair image of the brand. Participants were either shown the fair logo, or the neutral logo and were asked to indicate on a scale from 1 (doesn’t fit at all) to 5 (fits perfectly) to what extent they found that fair, and some other random traits, fit the brand. The results of the second pre-test (N = 29) showed that the fair version of the logo (N = 16) scored higher on the trait fair (M = 3.75; SD = .77) than the neutral version of the logo (N = 12; M = 2.75; SD = .87). Therefore this material was used for the manipulation of desirability of brand identity in the first experiment. The logo’s and (translated) descriptions of the brand as different manipulations are presented in figures 3, 4, 5 and 6.
Measures
Willingness to like. The willingness to like the brand has been measured by asking participants to what extent they felt like they would like the brand on Facebook. They were asked to indicate this chance in a percentage.
Demographics. For demographics of the sample, participants were asked to indicate their age, gender (0 = male, 1 = female), and educational level (1 = high school, 2 =
vocational education, 3 = higher education, 4 = academic education).
Manipulation checks. The manipulation was checked in the experiment by asking participants to what extent they found the brand fit the trait fair on a scale from 1 (doesn’t fit at all) to 5 (fits perfectly), and to what extent they desired to have the trait fair on a scale from 1 (not desirable) to 10 (very desirable).
Via an hyperlink participants could access the experiment online. Assigning of participants to one of two conditions in the experiment was done randomly by the online survey/experiment tool Qualtrics. When participants accessed the experiment they first had to read an informed consent form in which it was stated this experiment is about Facebook and personality, and declare they had read it and are 18 years or older. Then participants were asked to answer a few common questions such as what their age and gender is. Participants were then exposed to either a screenshot of a neutral version of the Facebook brand page of Match Mobile or the fair version. The willingness to like the brand has then been measured. Participants were then asked to fill in the manipulation check questions and some additional scales that were not included in this study. After this the participants were debriefed and then thanked for their participation.
Results Manipulation check
To check the manipulation of desirability of brand identity a t-test has been done to see if there is a difference between groups that had seen the brand that has been manipulated either to be desirable (fair) or neutrally desirable, in the extent to which participants found the brand they had seen fair. As intended, participants in the fair condition (N = 55; M = 3.98; SD = .80) found the brand significantly more fair than participants in the neutral condition (N = 39; M = 3.00; SD = .79), t (92) = 5.86, p < .001, 95% CI [.65, 1.31]. Additionally, participants indicated a high desirability of the trait fair (N = 94; M = 8.82; SD = 1.42). The manipulation has succeeded.
Hypothesis
The first hypothesis, namely that brands with a desirable brand identity (vs. neutrally desirable brand identity) are more likely to be liked on Facebook, was tested using a t-test in which the mean scores of the willingness to like the brand is compared between the group that
had seen the brand with a desirable brand identity and the group that had seen the brand with the neutrally desirable brand identity. Although participants that had seen the brand with a desirable brand identity were more willing to like the brand (N = 46, M = 27.65, SD = 27,91) than were participants that had seen the brand with the neutrally desirable brand identity (N = 33, M = 25.85, SD = 27.40), no significant difference between the groups has been found, t (77) = -.29, p = .776, 95% CI [-14.39, 10.78]. This study can therefore not support the first hypothesis.
Discussion
This study cannot support the hypothesis that brands with a desirable brand identity (vs. neutrally desirable brand identity) are more likely to be liked on Facebook. It seems that the general capacity of a brand to communicate a desired image does not necessarily influence the likelihood of a brand to be liked. Perhaps, in addition to the capacity of a brand to
communicate a desired identity, there has to be a need of the liker to enhance their self-concept concerning the traits that are associated with a brand, in order for the brand to be more likely to be liked. These results do not necessarily implicate that liking brands on Facebook is not a decision making process in which impression management plays a role. Merely these results suggest that the capacity of a brand to present a desired image differs per person. This has not been taken into account in this experiment. What remains interesting is to see if liking a brand has an effect on how the likers perceive themselves after this form of self-presentation. This will be further discussed in the second study.
Study 2:
The self as an effect of liking brands on Facebook
Because liking a brand on Facebook is predicted to be a form of impression
management, as well as a form of self-enhancement we can expect people to feel better about themselves concerning the liked (brand) identity. Park and Roedder-John (2010) point out
research has shown that consumers have a preference to “choose brands with appealing personalities in an attempt to affirm and enhance their sense of self” (p. 655). Additionally, Reed, Forehand, Puntoni and Warlop (2012) point out that for self-enhancement purposes, people incorporate brands in their concept, for they reflect positively on their self-concept. There seems to be a certain use of brands, to improve one’s self-view. Gao, Wheeler and Shiv (2009) refer to this as self-view bolstering consumption. If we’d say liking on SNS, just as consuming, is a form of self-enhancement, we can expect that liking a brand on SNS with a desirable brand identity will help people enhance their sense of self.
Wallace, Buil, de Chernatony, and Hogan (2014) distinguish typologies of likers on Facebook. Among them the ‘self-expressives’, who like brands to express their sense of self. They suggest that the more a brand is self-expressive, the more it will generate brand
advocacy. As a result of deliberately connecting symbols (e.g. brand identities) to a personal Facebook page, Gonzales and Hancock (2010) confirm that an identity shift can occur from viewing one’s own Facebook page. This means, after viewing one’s own Facebook page, or in other words the desired image of the self, people have more esteem. Due to the self-expressiveness of a brands identity, a brand is expected to be able to enhance someone’s perceived self, when someone has liked the brand. We can expect a brand with greater self-expressive capacity, or in other words a desirable brand identity to reflect positively on the self-concept, when the brand is liked. Nonetheless, this identity shifting process has not been investigated for brands on Facebook. This brings us to the second and third hypotheses.
H2: After liking (vs. being exposed to) a brand page, a positive identity shift will occur concerning the, through the brand liking, signalled identity.
H3: After liking (vs. being exposed to) a brand page, a positive identity shift will occur concerning the, through the brand liking, signalled identity, only if the liked brand has a desirable brand identity (vs. neutrally desirable brand identity).
These predictions are related to the effect of liking brand pages on SNS and come together in a conceptual model (see figure 2) which is tested in an experiment.
Method Design
This study aims to test the conceptual model presented in figure 2 in an experiment with a 2 (brand engagement: liking vs. exposure) x 2 (desirability brand identity: very
desirable vs. neutrally desirable) between subjects design. The second hypothesis, namely that after liking (vs. being exposed to) a brand page, a positive identity shift will occur concerning the, through the brand liking, signalled identity, and the third hypotheses namely that this effect only occurs if the liked brand has a desirable brand identity (vs. neutrally desirable brand identity), are tested.
Sample
In this experiment 144 people participated. Participants ranged in the age of 18 to 29, with an average age of 21 (SD = 2.29). 68.10% were female and the majority of the sample has had academic education namely 84.70%. Two participants were excluded from the analysis due to unusual behaviour such as drug use which was reported in a logbook. Participants were recruited via a research website of the University of Amsterdam and were notified of the research through posters spread over the campus of the University of
Amsterdam. Participants were rewarded €5 or if they were students, credits for their bachelor degree. Participants of the pre-test were recruited in the same manner. The pre-test was embedded in another study. To be able to check for participation of these respondents in the experiment, participants in the experiment were asked whether they are familiar with the brand Match Mobile, and if so, how they know of it.
To check the effectiveness of the manipulation of the desirability of the brand personalities a pre-test has been done. Four new descriptions of the previously used fictive brand Match Mobile have been made, all presenting Match Mobile with a different brand personality trait. Based on the results of the first pre-test in study 1, two desirable and two neutrally desirable traits have been chosen for Match Mobile namely, fair, creative, non-conformist and sound. In the pre-test participants were exposed to one of the four descriptions and the logo of Match Mobile and were shown a list of (brand) personalities, derived from a Dutch translated version of Aakers’ (1997) brand personality scale (Smit, Van den Berge, & Franzen, 2003). They were asked to indicate on a 1 (totally disagree) to 5 (totally agree) scale, to what extent they find the traits fit Match Mobile. The results of this pre-test (N = 147) showed that the descriptions of the fair and non-conformist version of Match Mobile were accurate in presenting Match Mobile as a fair brand (N = 51; M = 4.08; SD = .98) and a non-conformist brand (N = 31; M = 4.03; SD = 1.17). Also these descriptions did not present Match Mobile as having the traits that it was not intended to present, meaning the brand described as fair is not perceived as very non-conformist (M = 3.10; SD = 1.05), and the brand described as non-conformist is not perceived as very fair (M = 2.97; SD = 1.08). The
descriptions that intended to present Match Mobile as creative (N = 31; M = 3.32; SD = 1.33) and as sound (N = 34; M = 3.50; SD = 1.08) were not very accurate in presenting Match Mobile conform the intended trait. Additionally, these two descriptions scored more or less equal on every one of the four traits. The descriptions of Match Mobile that present the brand as fair and non-conformist are therefore used in study 2. Looking back at pre-test 1 in study 1, fair presented a desirable personality trait (M = 9.04; SD = .82) and non-conformist presented a neutrally desirable trait (M = 5.00; SD = 1.80). The logo, (translated) descriptions and screenshots of the brand pages that have been created for Match Mobile are presented in
figures 3, 7, 8, and 9. The brand experience manipulation has been manipulated by asking participants to either view Match Mobile’s brand page, or like it.
Measures
Identity shift. To measure the effect of liking of and exposure to a brand page on Facebook on the perceived self we are interested in the difference between the measurements of the extent to which participants perceive themselves fair or non-conformist before and after liking or being exposed to a brand page of Match Mobile. Participants were presented a list of 10 personality traits derived from the Dutch translated version of Aakers’ (1997) brand personality scale (Smit, Van den Berge, & Franzen, 2003) and were asked to indicate in a percentage to what extent they feel like they have this trait. The traits fair and non-conformist were repeated measures, but the other eight traits were different in the first and second
measurement of the perceived self to prevent suspicion. The difference score of the first and second measurement of fair or non-conformist served as the identity shift measurement.
Demographics. For demographics of the sample, participants were asked to indicate their age, gender (1 = female, 2 = male), and educational level (1 = high school, 2 =
vocational education, 3 = higher education, 4 = academic education).
Manipulation checks. The brand identity manipulation was checked in the experiment, first by asking to what extent they found traits among which fair and
non-conformist desirable on a 1 (not desirable) to 10 (very desirable) scale, and second by asking participants to what extent they found the brand they had seen fair and non-conformist on a 1 (not at all) to 10 (completely) scale. Checking the brand experience manipulation was done by asking participants whether they had liked the brand, and by checking the likes on the brand page.
Assigning of participants to one of four conditions in the experiment was done randomly by the online survey/experiment tool Qualtrics. When participants accessed the experiment in a research lab at the University of Amsterdam and randomly had been assigned to one of four conditions they first had to read an informed consent form in which it was stated that this experiment is about Facebook and personality, and declare they have read it and are 18 years or older. Then participants were asked about their perceived self on certain personality traits, then asked to like or view a brand page on Facebook of a brand that either has a desired or neutrally desired brand personality. Afterwards they were asked about their perceived self again and they had to ask some questions for checking the manipulation. After this the participants were debriefed and then thanked for their participation.
Results Manipulation check
The manipulation check confirmed the intended desirability of the traits used to manipulate a neutrally desirable brand identity, namely non-conformist (N = 144, M = 5.98, SD = 2.06) and a desirable brand identity, namely fair (N = 144, M = 8.52, SD = 1.59). Additionally, participants from the non-conformist condition found the brand they had seen more non-conformist (N = 73, M = 4.32, SD = 1.31) than participants in the fair condition (N = 71, M = 3.94, SD = 1.26). The difference reached a one-tailed significance, t (142) = 1.73, p = .086, 95% CI [-.05, .80]. Participants from the fair condition found the brand they had seen more fair (N = 73, M = 4.79, SD = 1.12) than participants in the non-conformist condition (N = 71, M = 3.97, SD = 1.19), t (142) = -4.24, p < .001, 95% CI [-1.20, -.44]. Manipulation of desirability of brand identity has succeeded. In the analyses 19 participants were excluded for they did or did not like the brand page, as they were not supposed to do in their condition. Hypotheses
The second hypothesis, namely that after liking (vs. being exposed to) a brand page, a positive identity shift will occur concerning the, through brand liking, signalled identity, and the third hypothesis namely that this effect only occurs if the liked brand has a desirable brand identity (vs. neutrally desirable brand identity) was tested using a one-way analysis of
variance (ANOVA) with a difference score of, dependent on condition, perceived fairness or non-conformism before and after the exposure or liking of a brand as dependent variable, and brand experience (liking vs. exposure) and brand identity desirability (desirable vs. neutrally desirable) as independent variables.
There was no interacting effect of brand experience and brand identity found on identity shift, F (1, 140) = 1.31, p = .254, η2 = 01. Excluding the interacting effect from the analysis provides us with a weak significant main effect only for brand experience, F (1, 141) = 7.51, p = .007, η2 = 05, and no effect for brand identity, F (1, 141) = .03, p = .855, η2 = 00. Participants that had liked the brand with a desirable brand identity (N = 31, M = 1.61, SD = 7.07), as well as the brand with a neutrally desirable brand identity (N = 40, M = .25, SD = 7.37) perceived themselves more positive concerning the brand personality trait of their condition, then participants that had been exposed to the brand with a desirable brand identity (N = 34, M = -3.91, SD = 10.42), as well as the brand with a neutrally desirable brand identity (N = 39, M = -2.10, SD = 7.89). A t-test confirms this difference in identity shift between the participants that had liked a brand (N = 71, M = .85, SD = 7.22) and had been exposed to a brand (N = 73, M = -2.95, SD = 9.14), t (142) = -2.76, p = .007, 95% CI [-6.51, -1.07]. These results provide support only for the second hypothesis. The third hypothesis cannot be supported.
Discussion
Not only does this study support the second hypothesis namely that after liking (vs. being exposed to) a brand page, a positive identity shift will occur concerning the, through
liking, signalled identity, interestingly it also presents us this effect is negative for exposure. It can be concluded that liking a brand on Facebook is a way to include brand identities in the self-concept and enhance one’s self-view. Additionally it seems exposure to a brand identity without associating it with one’s personal Facebook profile is confronting for people, and the brand identity reflects negatively on the self-concept.
This study cannot support the third hypothesis namely that the effect of liking of (vs. exposure to) a brand only occurs if the liked brand has a desirable brand identity (vs. neutrally desirable brand identity). It seems that the general capacity of a brand to communicate a desired image does not influence the effect of liking on the self-concept. These results do not necessarily implicate that there are no differences between brand identities in the effects that liking these brand identities have, for the same principle suggested in study 1 can be applied here. The capacity of self-expressiveness of a brand cannot be generalized because it might be dependent on the need of likers to self-enhance.
General discussion Theoretical implications
The two studies combined provide useful theoretical insights in causes and effects of liking brands on Facebook in the light of impression management theory. In view of the first aim which is to test if the capacity of a brand identity to communicate a desired identity influences the likeliness of a brand to be liked on Facebook, the lack in support for the first and third hypotheses provide us with the potential implication there is no such thing as a general capacity of a brand identity to communicate a desired identity. A general desirability of brand identity has not been found to affect the likelihood of a brand to be liked in the first study, and it has not been found to affect the effect of liking a brand on self-perception in the second study. If the need to enhance the self-concept differs per person, this would mean
desirability of a brand identity is different per person and cannot be seen as a general capacity of a brand.
In view of the second aim of this study which is to identify the effects of endorsing brands on SNS on the self-concept, the support for the second hypothesis provides us with the insight that liking brands can function as a form of self-view bolstering consumption, as expected considering the existing literature (Gao, Wheeler & Shiv, 2009). In addition to the previously found positive effects of brands (Park & Roedder-John, 2010) and viewing one’s own Facebook profile (Gonzales & Hancock, 2010) on the self-concept, this study illustrates the positive effect on the self-concept of the combined causes, namely liking brands on Facebook.
Limitations and future research
The results show us that we cannot conceptualize in terms of a generalized desirability of brand identities. Perhaps desirability of a brands identity does affect the likelihood of a brand to be liked and the effect of liking a brand on self-perception, but not as manipulated in these studies. Future research might replicate the studies but manipulate the capacity of a brands identity to communicate a desired image as a personal factor, in order to confirm these theoretical assumptions, and to be able to elaborate practical managerial implications for the development of appropriate brand identities.
It has to be noted that the effect found in study 2 in identity shifting is small although significant. At this point, concluding that brands or SNS have the capacity to entirely change our identities is really not appropriate. It is probably fortunate, mankind’s identities are not entirely a product of commercial expressions. Future research could give us more insights in the magnitude of the identity shift effect. Replicating the test with existing brands might enlarger the difference, for associations with existing brands are far more established. Additionally, it has to be noted that the scores on identity shifting was a difference score
between two points in time that were very close to each other. More time between the first measurement and the manipulation might make the difference bigger, and more time between the liking and the second measurement might shed light on potential long-term effects.
Considering brands, identity and social media, it would be interesting for future research to look for differences in identity shifts for different types of likes or other brand related activities on Facebook, for example likes provoked by the use of win promotions. It would also be interesting to find the effects of (these potential different) identity shifts on brand attitude and purchase intentions, in order to be able to give explicit managerial recommendations.
References
Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, XXXIV, 347-356. doi:10.2307/3151897
Aaker, J. L. (1999). The malleable self: the role of self-expression in persuasion. Journal of Marketing Research, XXXVI, 45-57. doi:10.2139/ssrn.945453
Belk, R. W. (1988). Possessions and the extended self. The Journal of Consumer Research, 15, 139-168. doi:10.1002/9781444316568.wiem03037
Belk, R. W. (2013). Extended self in a digital world. Journal of Consumer Research, 40, 477- 500. doi:10.1086/671052
Berger, J., & Schwartz, M. (2011). What drives immediate and ongoing word-of-mouth? Journal of Marketing Research, XLVIII, 869-880. doi: 10.1509/jmkr.48.5.869
Berger, J., & Heat, C. (2007). Where consumers diverge from others: Identity signalling and product domains. Journal of Consumer Research, 34, 121-134. doi:10.1086/519142 Cvijikj, I. P,. & Michahelles, F. (2013). Online engagement factors on Facebook brand pages. Social Network Analysis and Mining, 3, 843-861. doi: 10.1007/s13278-013-0098-8 Gao, L., Wheeler, S. C., Shiv, B. (2009). The “shaken self”: Product choices as a means of restoring self-view confidence. Journal of Consumer Research, 36, 29-38.
doi:10.1086/596028
Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday. Gonzales, M. A., & Hancock, J. T. (2010). Mirror, mirror on my Facebook wall: Effects of exposure to Facebook on self-esteem. Cyberpsychology, Behavior, and Social Networking, 14, 1-5. doi:10.1089/cyber.2009.0411
Ho, J., & Dempsey, M. (2010). Viral marketing: Motivations to forward online content. Journal of Business Research, 63, 1000-1006. doi: 10.1016/j.jbusres.2008.08.010
Kirmani, A. (2009). The self and the brand. Journal of Consumer Psychology, 19, 271-275. doi:10.1016/j.jcps.2009.05.011
Krämer, N., & Winter, S. (2008). Impression management 2.0. Journal of Media Psychology, 20, 106-116. doi:10.1027/1864-1105.20.3.106
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review and two-component model. Psychological Bulletin, 107, 34-47.
doi:10.1037/0033-2909.107.1.34
Muntinga, D., Moorman, M., & Smit, E. G. (2011). Introducing COBRA’s: Exploring motivations for brand-related social media use. International Journal of Advertising, 30, 1-13. doi:10.2501/IJA-30-1-013-046
Park, J. K., & Roedder-John, D. (2010). Got to get you into my life: Do brand personalities rub off on consumers? Journal of Consumer Research, 37, 655-669.
doi:10.1086/655807
Reed, A., Forehand, M. R., Puntoni. S., & Warlop, L. (2012). Identity-based consumer behavior. International Journal of Research in Marketing, 29, 310-321.
doi:10/1016/j.ijresmar.2012.08.002
Schau, M., & Gilly, H. J. (2014). We are what we post? Self-presentation in personal web space. Journal of Consumer Research, 30, 385-404. doi:10.1086/378616
Serazio, M. (2013). Selling (digital) millennials: The social construct and technological bias of a consumer generation. Television & New Media, 1-17.
doi:10.1177/1527476413491015
Smit, E. G., Van den Berge, E., & Franzen, G. (2003). Brands are just like real
Subrahmanyam, S., Reich, N., Waechter, N., & Espinoza, G. (2008). Online and offline social networks: Use of social networking sites by emerging adults. Journal of Applied Developmental Psychology, 29, 420-433. doi:10.1016/j.appdev.2008.07.003 Tucker, C. (2011). Ad virality and ad persuasiveness. Available at SSRN 1952746. doi:10.2139/ssrn.1952746
Urde, M. (2013). The corporate identity matrix. Journal of Brand Management, 20, 742- 761.doi:10.1057/bm.2013.12
Wallace, E., Buil, I., & de Chernatony, L. (2012). Facebook ‘friendship’ and brand advocacy. Journal of Brand Management, 20, 128-146. doi:10.1057/bm.2012.45;
Wallace, E., Buil, I., de Chernatony, L., & Hogan, M. (2014). Who “likes” you… and why? A typology of Facebook fans. Journal of Advertising Research, 54, 92-109.
doi:10.2501/JAR-54-1-092-109
Wojnicki, A., & Godes, D. (2008). Word-of-mouth as self-enhancement. HBS Marketing Research Paper, (06-01).doi: 10.2139/ssrn.908999
Zhang, S., Jiang, H., & Carroll, J. M. (2010). Social identity in Facebook community life. International Journal of Virtual Communities and Social Networking, 2, 66-78. doi:10.4018/jvcsn.2010100105
Zhao, S., Grasmuck, S., & Martin, J. (2008). Identity construction on Facebook: Digital empowerment in anchored relationships. Computers in Human Behavior, 24, 1816- 1836. doi:10.1016/j.chb.2008.02.012
Appendix A: Conceptual models
Figure 1. Conceptual model of cause for liking brands on Facebook.
Appendix B: Stimulus materials
Figure 3. Logo Match Mobile.
Figure 4. Logo Match Mobile Fairtrade.
Figure 5. Description Match Mobile experiment 1, neutral condition.
There’s no arguing taste. That’s why we match your needs to our capabilities and make your smartphone.
You can compose your smartphone to your taste and preferences in our stores or online. Choose from different colours and designs or upload your own photo or design. Choose also the
specifications that best suit you. What is the best place for your menu button? Do you often makes calls abroad? Perhaps a dual sim is a good option for you.
This is how we match your wishes to our capabilities and provide a unique smartphone that takes your preferences into account.
Figure 6. Description Match Mobile experiment 1, fair condition.
Figure 7. Description Match Mobile experiment 2, non-conformist condition.
Figure 8. Description Match Mobile experiment 2, fair condition.
Figure 9. Screenshot Facebook brand page Match Mobile experiment 2.
Match Mobile stands for fair smartphones. At Match Mobile raw materials are won under good working conditions. Match Mobile treats people well. Match Mobile takes everyone into account. Match Mobile stands for smartphones that are not compliant with the standard. At Match Mobile they do not account for the trends. Match Mobile differs from the mainstream. Match Mobile just does his own thing.
There’s no arguing taste. There is to argue fair smartphone production though. That’s why we match your needs and our capabilities with those of the labourers who produce your smartphone, and those of nature, and make your fair smartphone.
All the raw materials in the smartphone are won under fair working conditions, the smartphone is environmentally friendly produced and all components in the smartphone are recyclable.
You can compose your smartphone to your taste and preferences in our stores or online. Choose from different colours and designs or upload your own photo or design. Choose also the
specifications that best suit you. What is the best place for your menu button? Do you often makes calls abroad? Perhaps a dual sim is a good option for you.
This is how we match your wishes to our capabilities and the needs of the producer and nature, and provide a unique smartphone that takes everyone into account.
