Running head: THE EFFECTS OF SNAPSHOT PHOTOS 1
MSc Communication Science Graduate School of Communication
Master Thesis
The Effects of the Fashion Brand’s Activity of Posting Snapshot Photographs on Instagram on Millennials’ Brand Attitudinal Loyalty. The Role of Customer-Brand Identification
Ioana Ene 11800305
Dr. J.M.F. (Annemarie) van Oosten University of Amsterdam
February 2019 7755 words
THE EFFECTS OF SNAPSHOT PHOTOS 2
Abstract
This study investigates the effects of the brand's activity of posting snapshot photos on Instagram on millennials' brand attitudinal loyalty, and whether costumer-brand identification mediates this relation. A between-subject online experiment was conducted in which 125 respondents (aged 18-54 years) were randomly exposed to either snapshot or traditional photos of created Instagram accounts of the fictitious fashion brand Stylo. Results showed that neither the direct nor the indirect effect of snapshot pictures are statistically significant. Therefore, the exposure to snapshot Instagram photos does not lead directly to higher brand attitudinal loyalty of millennials, neither throughout the mediation of customer-brand identification. These findings suggest that the aesthetics used by brands to communicate with users within Instagram should not be necessary in line with the informal, lightweight decorum of this platform.
Keywords: social media, Instagram, photo aesthetic, brand loyalty, brand identification
THE EFFECTS OF SNAPSHOT PHOTOS 3
The Effects of the Fashion Brand’s Activity of Posting Snapshot Photographs on Instagram on Millennials’ Brand Attitudinal Loyalty. The Role of Customer-Brand Identification
Social media represent an important tool for marketers and advertisers to promote their brands and products due to the benefits of offering high exposure and traffics to audience segments. As Instagram has become the world's most popular photo-sharing platform and users are more connected than ever before, brands have started to use it in order to get their messages across to consumers (Chua & Chang, 2016). A recent study done in the United States showed that 64% of Instagram’s users are aged between 18 and 29 years and 42% of them use this platform at least once per day (Pew Research Center, 2018). This Instagram routine of millennials is important information for marketers as currently young adults are considered to be the biggest consumer age group, making up 80 million of the USA’s consumers. Thus, the buying power of millennials is ever growing (Fromm, Butler, & Dickey, 2015).
As a result of these developments and considering that Instagram is a social networking platform for sharing visual content (Geurin-Eagleman & Burch, 2016), marketers have tried to find out what type of pictures are engaging to consumers on digital media in order to create appealing content. Recently, it has been demonstrated that snapshot aesthetics, such as an amateur spontaneous photography style, are more suitable for fashion brands’ communication on Instagram because these photos are seen as more personal, and thus, they are seen as ʻsocialʼ, rather than ʻcommercialʼ (Colliander & Marder, 2018). This style is herein referred to as ʻsnapshot photographsʼ, which are informal and spontaneous photographs, as opposed to pre-planned traditional studio photographs. In essence, authenticity and naturalness are the key aspects of the snapshot-like images (Sarvas & Frohlich, 2011; Schroeder, 2010). Therefore, in order to keep up with social media trends and consumers’ preferences, some brands started to use snapshot aesthetics on their Instagram
accounts. For instance, a number of haute couture brands are already using snapshot aesthetics
when engaging their Instagram followers (e.g.
Gucci, Balenciaga). However, at present, little empirical research has been done on the effects of digital snapshot aesthetics on the relationship between brands and consumers.
In terms of effects, this study will look at brand loyalty as an outcome variable. Never before the fashion industry has had a plethora of brand options for consumers, so the competition amongst companies is significantly intense. Thus, brands are eager to find out what strategies they should implement in order to create a connection between them and their consumers (Tansey, 2017), and in particular to make them loyal towards their brand. In order to contribute to this gap in the literature, this paper analyzes the influence of the brand's activity of posting fashion snapshot photographs on Instagram on millennials' brand loyalty.
The aim of this study is to measure the immediate effect of the brand’s activity of posting snapshot photos. Therefore, brand loyalty will be considered as attitudinal loyalty, based on the determinist approach which investigates “the psychological commitment of the consumer in the purchase, without necessarily taking the effective purchase behavior into account” (Odin, Odin, & Valette-Florence, 2001). In other words, attitudinal loyalty is the psychological process (e.g., the motive of repurchase) which antecedents and determines the behavioral loyalty, namely the repeated purchasing behavior. Furthermore, this paper proposes to investigate whether the influence of the brand's activity of posting snapshot photos on Instagram on millennials’ brand attitudinal loyalty is mediated by increased customer identification with the brand (CBI). This identification is conceptualized here as the process through which the perceived brand identity is integrated into self-identity. In other words, brand identification represents the degree to which consumers define themselves by the brand attributes (Hughes & Ahearne, 2010).
In order to investigate this particular topic, the following research question (RQ) was formulated:
RQ. Does the brand's activity of posting fashion snapshot photographs on Instagram influence the millennials' brand attitudinal loyalty, through increased identification with the brand?
The study consists of a one factorial between-subjects experiment with Instagram users and it contributes both to businesses and the academic environment. I expect that the results will help marketers to develop strategies for online advertising and to better understand how brand communication may be used to enhance clients’ identification with and subsequent loyalty towards the brand. Academically, existing studies into customer reactions to advertising images focused more on the non-social media setting (Fox, 2004) and therefore it is not demonstrated precisely whether the consumers’ preference for snapshot images in the non-digital setting apply as well in the social media context. Hence, this study contributes to a greater understanding of how social media users respond to various social media content, in particular, to snapshot photos, and how this content can fit their identity and influence their purchase attitudes.
Theoretical Background Snapshot Aesthetics and Brand Attitudinal Loyalty
Academic research has investigated the role that digital image aesthetics could play in the users’ engagement with a different type of photos. Previous studies have demonstrated that consumers prefer snapshot aesthetics because these photos are more fluently to translate, and in general, the more fluently people can process a picture, the more positive their aesthetic response (Reber, Schwarz, & Winkielman, 2004). Snapshot aesthetics started to be used in social media by several fashion brands since nowadays this type of images are the norm in social photo sharing (Diderich, 2017). Investigating the effects of a clothing brand using either snapshot aesthetics, or traditional studio aesthetics in their social media images, a study
conducted by Colliander and Marder (2018) demonstrated that the use of snapshots by a fashion brand is favored by users because these images hold greater congruence with the
custom of the
medium and they might lead to positive cognitive effects and finally, to an increased preference for them. These facts are in line with the findings of Reber et al. (2004) that social
media users
are more familiar to snapshot aesthetic because they are more frequently exposed to it within social media platforms and thus, they consider this style as being congruent within the context of digital platforms. Moreover, Colliander and Marder (2018) also showed that in social media, snapshot photos produce higher positive brand attitudes and intentions to recommend others to follow the Instagram account, while these effects are mediated by increased liking of the images and high source credibility.
Even though limited research has focused on the effects of snapshot aesthetics on consumers, this study proposes that the brand’s activity of posting snapshot photos leads to higher attitudinal loyalty of millennials due to the character of these photos to be meaningful, easy to translate and to determine positive cognitive effects. Considering that there is no clear theoretical model to specifically explain this direct effect, this paper proposes a part from The Differential Susceptibility to Media Effects Model (Valkenburg & Peter, 2013) as explaining similar mechanisms to the processing of congruent content. This part is The Disposition-Content Congruency Hypothesis which states that individuals seek out media which is similar to their beliefs, emotions, and attitudes. The authors (2013, p. 12) affirm that “congruent content is processed faster and more efficiently because it can be related to more existing mental schemata of the media use”. Moreover, congruent content enhances media users’ experience of familiarity and thus, individuals learn more from this content because it is easier to process.
Hence, this paper proposes that a similar process occurs for snapshot images which are authentic and natural and they can be congruent with millennials’ beliefs that marketing communication should be transparent and real (Black Bear Design, 2016). As Valkenburg and Peter (2013) state that media content which is congruent with consumers' dispositions is more likely to lead to positive media effects, it can be considered for the context of this study that the positive cognitive effect of snapshot images is the consumers’ higher brand attitudinal loyalty. This concept is referred to as the immediate outcome of consumers’ emotional and cognitive
responses (Hwang & Kandampully, 2012). Therefore, the next hypothesis was formulated: H1. The brand’s activity of posting snapshot photos on Instagram leads to higher brand attitudinal loyalty of millennials compared to traditional studio photos.
Snapshot Aesthetics and Customer-Brand Identification
Some previous studies investigating the most efficient practices for brands to engage with millennial consumers on Instagram demonstrated that brands need to get on the same level as their audience, showing them that “there is a human on the other side of the screen” (Black Bear Design, 2016, p. 1). Within this age period, also called the “me-moment” because it is the most self-focused age of life, millennials are investing highly on themselves both personally and professionally (Arnett, 2007). Regarding their purchase preferences, they are looking for brands and products which fit their personality (Gurău, 2012). Therefore, brands can have success on Instagram by reaching them authentically and openly because millennials want to see transparency in companies’ online communication, or at least a human feeling in their photos (Black Bear Design, 2016). According to Kim, Han, and Park (2001), when individuals perceive this human touch in the brand’s personality, they will use these human characteristics to express their own personality which leads to the process of consumer identification with the brand.
The theoretical framework which explains the consumer identification phenomenon is the Social Identity approach (Hornsey, 2008) which comprises both the Identification Theory (Stets & Burke, 2000) and Self-Categorization Theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). These theories explain the motivations that encourage individuals to relate to companies. This happens through a cognitive categorization process in which individuals psychologically perceive, feel, and value their belongingness with a company or a brand, by enhancing similarities with it (Lam, Ahearne, Mullins, Hayati, & Schillewaert, 2013). Moreover, the Social Identification and the Categorization Theory have been used specifically as a basis for understanding individuals’ psychological attachment to a certain organization (Hughes & Ahearne, 2010). According to these theories, organizational identification appears when individuals' beliefs about an organization become self-defining (Pratt, 1998, as cited in Bhattacharya & Sen, 2003, p. 2). As a result of this process, individuals will psychologically accept the organization as part of their personal identity (Martínez & del Bosque, 2013).
Some marketing research extended this organizational identification logic to a more micro level research domain, namely brands. This extension is possible because brands can signify self-relevant social categories with which customers can identify and they can transfer meaning between themselves and the brands (Lam, Ahearne, Hu, & Schillewaert, 2010). These ‘brand meanings’ determine to what degree the brand is perceived as being congruent with the self-concepts of consumers (Heath & Scott, 1998). Therefore, similar to the organizational identification, brand identification can be explained as “the degree to which a person defines him- or herself by the same attributes that he or she believes define a brand”. This social construction involves that the perceived brand identity is integrated into the self-identity of the clients (Hughes & Ahearne, 2010).
Taking into account that snapshot images are natural and spontaneous photos, there is a reason to believe that they could reflect a human touch which may fulfill the millennials’
preference for a human feeling in the marketing photos. Based on the Social Identity approach which states that consumers are attracted to brands that have a similar symbolic image as their own self-concept (Lam et al., 2010), it can be considered that millennials may interpret these human characteristics of snapshot images as self-defining. This fit between the brand image and consumers’ self-image can lead to millennials’ self-identification in these photos. Hence, the following hypothesis was formulated:
H2. The brand’s activity of posting snapshot photos on Instagram leads to higher brand identification of millennials compared to traditional studio photos. Customer-Brand Identification and Brand Attitudinal Loyalty
The author Kapferer (2008) affirms that consumers often choose certain brands that might reflect their identity. Some studies showed that consumer-brand identification is a strong predictor of consumer behavior (e.g., repurchase intention, word-of-mouth, and symbol passing) (Donavan, Janda, & Suh, 2006; Kuenzel & Vaux Halliday, 2008). Moreover, based on the prediction of Social Identification Theory, consumers with increased brand identification are more willing to engage in brand-related activities, such as protecting company’s reputation, supporting its products and brand loyalty (Bhattacharya & Sen, 2003). In line with this idea, Heath and Scott (1998, p. 2) explain that “consumers who perceive the product image to be consistent with their actual self-concept are likely to feel motivated to purchase and consume that product”.
Empirically, this relationship between brand identification and brand loyalty was investigated by many researchers in different areas. For instance, the results of Carlson, Todd Donavan, and Cumiskey (2009) demonstrated that there is a direct connection between brand identification and brand loyalty in the car industry while Kim et al. (2001) showed this relationship to be insignificant in mobile phone brands. In the hotel industry, one study demonstrated that customer hotel brand identification (CHBI) influences indirectly brand
loyalty via brand love (Alnawas & Altarifi, 2016), while another study concluded that even though strong customer brand identification (CBI) is not sufficient to establish hotel brand loyalty, it does have an indirect influence on brand loyalty via customers’ brand evaluation (So,
King, Sparks, & Wang, 2013).
Moreover, previous research also investigated the possible mediation role of costumer-brand identification on costumer-brand loyalty in different areas. For example, one study in the Corporate Social Responsibility field (CSR) showed that increased levels of consumer CSR initiatives are linked to stronger loyalty behavior due to consumers’ increased identity with
THE EFFECTS OF SNAPSHOT PHOTOS 11
(Marin, Ruiz, & Rubio, 2009). Another study investigating the context of pharmaceutical sales demonstrated that the effect of perceived company characteristics on both product utilization and extra-role behaviors is mediated by consumer-company identification (Ahearne, Bhattacharya, & Gruen, 2005). However, the mediation relationship of CBI is little researched in relation to the influence of the brand's social media photos on consumers’ brand loyalty.
In line with these results, one way that snapshot photographs could increase costumers’ loyalty towards the brand is by increasing costumers’ identification with the brand. The fit between brand personality and customers’ brand loyalty is determined by the degree to which the brand expresses and enhances the customers’ identity. When this fit happens, consumers may attach human characteristics to the brands which are used to express their own image or personality (Kim et al., 2001). Based on the informal character of snapshot photos, people can translate these photos more fluently and associate them human characteristics which we assume that can lead to customers’ identification with this kind of photos.
Therefore, considering the results presented above from antecedent studies and considering the supposition made based on the Social Identification Theory, there is a reason to believe that the influence of the brand's activity of posting snapshot photos on Instagram on millennials' brand attitudinal loyalty is mediated by increased customer identification with the brand. Thus, the following hypothesis was proposed:
H3. The brand’s activity of posting snapshot photos on Instagram has a positive indirect effect on millennials’ attitudinal loyalty via increased customer-brand identification.
Methods Research Design
This study benefits from the empirical-analytical approach, in order to obtain quantitative results. The experiment employed a between-subjects design with two levels of
the independent variable of image type (snapshot versus traditional studio photo), where
customer-brand identification was the mediator and the customer-brand attitudinal loyalty was the dependent variable. To test the hypotheses, the answers of users exposed to snapshot aesthetic posts on Instagram were compared to the other respondents who were exposed to traditional studio aesthetic posts in an online experiment.
The benefit of the experimental method is that they permit to evaluate stimuli in a controlled environment keeping other factors which may influence the manipulation under control. Therefore, this research method seemed to be suitable to test this particular research question since the purpose of this study is to isolate the effects of two particular types of photo aesthetics (snapshot and traditional studio images), thus increasing the internal validity of the study. I chose fashion as the research area for this study in order to capture processes that actually can be generalized to real situations and the population investigated. More and more fashion brands started to use social media as a marketing tool specifically, Instagram, which has been called “one of the most important platforms a [fashion] blogger can be on” (Veselinovic, 2014). Due to its importance, this domain was studied in previous research about the communication effects within blogs (Colliander & Dahlén, 2011) and within Instagram (Colliander & Marder, 2018). This study proposes that the current findings can serve as insights for marketers and brands in order to create appealing fashion content for the millennials.
Participants
The target group of this study was millennials, both females, and males, aged 20-30 years. The research benefited from the convenience sample strategy in order to ensure collecting
THE EFFECTS OF SNAPSHOT PHOTOS 13
the required amount of data in a short time span. The participants were recruited through the University of Amsterdam’s Facebook groups where the link to the online survey was published within a post with information about the research study. The other participants were recruited from the principal researcher’s social environment and were contacted directly either
on their
personal Facebook page, other social media channels or on their email addresses by sending the study’s URL.
A total of 125 participants, between 18 and 54 years of age, took part in this experiment, most of them being students at Bachelor or Master tracks within the University of Amsterdam. From the total number of participants, eight of them were dropped from the study because they did not meet the required age limit, while eleven cases were reported as missing values due to the exit of the respondents from the survey.
In the final sample of 114 participants, there were 80 women (64%) and 34 men (27.2%), while the mean age of the participants was 25.07 (SD = 2.19) years. The respondents were randomly assigned to the experimental or control condition, by selecting the feature on Qualtrics to randomly and evenly present conditions. Hence, 53 respondents were exposed to the snapshot images, while 54 respondents were exposed to traditional studio images. In order to check the randomization and to verify whether the respondents in the two conditions were similar regarding their gender, age and Instagram usage, a cross-tabulation and respectively, two separate independent t-tests were conducted.
The results of the cross-tab showed that the distribution of gender within the two conditions was almost similar, x² (1) = .48, p = .489. The independent t-test run with the variable age showed that the age of the respondents was almost similar in both conditions, t(99) = .98, p = .463, 95% CI [-.43, 1.30]. The other independent t-test run with the variable
Instagram usage showed that time spent daily on Instagram is almost similar for the respondents in both conditions, t(97) = .22, p = .466, 95% CI [-.35, .44].
Procedure
The experiment took place online over the course of two weeks. Data were collected using online self-completion surveys which were hosted on https://uvacommscience.eu.qualtrics.com/. This method was chosen because online surveys can
be administered quickly, inexpensive, and asynchronous (Fowler Jr, 2013). The participants received the study URL which redirected them to the landing page where they were made aware of the purpose of the study from the factsheet and they gave their consent for participating in the experiment. After filling in the basic questions about their gender and age (see Measures), the respondents were asked to mention the amount of time they spend daily on Instagram: “How much time, on average, do you spend on Instagram each day?” (response options ranged from less than 1 hour; 1-2 hours; 3-4 hours; 5-6 hours; 7-8 hours; 9-10 hours; more than 10 hours; I don’t use Instagram). Then, the respondents were asked to reflect on their fashion product purchase behavior, as a warm-up question. After this, the respondents were randomly assigned to either snapshot photos or traditional studio photos. After the exposure to one of these conditions, the respondents filled out the survey about their identification with the brand in the pictures (i.e., the fictitious brand “Stylo”) and about their attitudinal loyalty towards this brand. The last set of questions represented the manipulation check after which the participants were debriefed regarding the real purpose of the research. Stimuli
The experimental manipulation was the image type and two sets of stimulus materials were constructed for the study, traditional studio and snapshot images. In order to eliminate any potential effects of brand recognition, a fictitious fashion brand, “Stylo”, was created and
also two separate Instagram accounts were set up for the purpose of this study. One of these accounts displayed the snapshot photos, while the other one showed the traditional studio images (see Appendix A). The style of the photos was the main difference between the
conditions, while a
second difference was that in each picture were different people, not the same models. Other aspects (such as the number of posts, followers, and the number of people in the picture and their gender) were kept identical. Both the snapshot and traditional studio photos were collected from Pinterest (www.pinterest.com), a social network for browsing a variety of
photos, GIFs,
and videos from different themes. The criteria for selecting the photos was to represent the same style of clothing which was fashionable in the period of the experiment (see Appendix B). The photos contained similar colors and no logo was shown in order to eliminate the possibility that the respondents could be able to recognize the actual brand of the clothes. Moreover, respondents could perceive both images from the two aesthetic styles as being clothes from the same brand and hence, avoid that respondents could be aware of the manipulation.
Measures
The demographic details of the participants were asked in two separate questions. Age was measured on an open-ended question (“Please indicate your age”) in order to obtain the exact age of each participant, while gender was measured on a close-ended question (“Please indicate your gender”), having two answer options: 1 – female; 2 – male.
Brand attitudinal loyalty was measured on an interval scale in order to investigate the respondents’ degree of loyalty. The scale constructed by (Quester & Lin Lim, 2003) was taken as a baseline for this concept. This scale sums up 16 pre-existing items from previous studies in order to create a comprehensive measure of attitudinal loyalty, by combining all of
the three components of attitude (cognitive, affective and conative). All the subscales are highly reliable (coefficient alpha ranging between 0.63 and 0.90) and they were measured on a seven-point Likert scale (ranging from 1 – strongly disagree to 7 – strongly agree). Within this research, only eight subitems were used from this scale and they were adapted in order to measure the immediate attitudinal loyalty of consumers after they are exposed to the stimuli within the experiment. Some of the adapted subitems are: “In the next few months/years, I
will consider
to buy this particular brand because I really liked the brand”; “I felt excited about this particular brand of clothes”, and “I felt good about this particular brand of clothes”.
In order to indicate whether the latent construct brand attitudinal loyalty has multiple dimensions or it is uni-dimensional, a Principal Axis Factoring Analysis (PAF) was run with the eight manifest items measured on a seven-point Likert scale. Based on the “7+ rule”, all the items could be considered as continuous items. Before running the factor analysis, value "99" was reported as a missing value for all these items. The PAF showed that all the items formed a single uni-dimensional scale: only one component had an eigenvalue above 1 (eigenvalue 5.89) which explains 73.62% of the total variance and there is a clear point of inflection after this component in the scree plot. The Correlation Matrix table showed that all items correlated positively, meaning that there are clusters of items that belong to each other. The KMO measure of sampling adequacy was .90 which is higher than the Kaiser’s bare minimum of .50 score, so the sample size was adequate for factor analysis. In addition, Bartlett's test of sphericity is statistically significant, x² (28) = 737.39, p < .001, indicating that the correlations between items were sufficiently large for PAF. Moreover, the Factor Matrix showed that all the items were correlated positively, while the variable measuring the respondents’ excitement towards the brand Stylo had the strongest association (factor loading was .91). After this, the reliability
of the scale was checked in order to see whether a new variable can be created. The results showed that the reliability of the scale is very good, Cronbach’s alpha = .94, meaning that the scale measured brand attitudinal loyalty. Therefore, a new variable called ´Brand Attitudinal Loyalty’ was computed.
Customer-brand identification was measured on a seven-point Likert scale (1 – strongly disagree; 7 – strongly agree), while only one item was measure with the Venn-diagram item constructed by Bergami and Bagozzi (2000). The scale items were taken from existing scales that Lam et al. (2013) have put together in order to measure the dynamics of consumer-brand identification (CBI) and its antecedents in the context of the launch of a new brand (all the subscales are highly reliable, coefficient alpha ranging between 0.70 and 0.90). Some of these adapted scale items are: “Imagine the situation in which you already started
to wear
clothes from the brand Stylo within the last few months. For the following set of questions, please answer with your own thoughts/perception about yourself within this situation: I believe others would respect me for my association with this particular brand; I would consider myself a valuable partner of this particular brand”.
The Venn-diagram measures the overlap between consumer identity and brand identity: “In the below picture, imagine that the circle at the left in each row represents your own personal identity and the other circle, at the right, represents the brand Stylo’s identity. Please indicate which case (A, B, C, D, E, F, G, or H) best describes the level of overlap between your identity and this particular brand’s identity (Choose the appropriate letter)”. Both the diagram item and the scale items were adapted for the context of the present study in order to measure the immediate respondents’ identification with the fictitious brand, Stylo.
multiple dimensions or it is uni-dimensional, a Principal Axis Factoring Analysis (PAF) was run with the three manifest items measured on a seven-point Likert scale. Based on the “7+ rule”, all the items could be considered as continuous items. The Venn-diagram item was not included in the factor analysis because it was measured on an eight-point Likert scale, and thus it was used separately in the research, as a continuous variable for this construct. Before running the factor analysis, value "99" was reported as a missing value for all these items.
The PAF showed that all the three items formed a single uni-dimensional scale: only one component had an eigenvalue above 1 (eigenvalue 1.96) which explains 65.51 of the total variance and there is a clear point of inflection after this component in the scree plot. The Correlation Matrix table showed that all items correlated positively, meaning that there are clusters of items that belong to each other. The KMO measure of sampling adequacy was .66 which is higher than the Kaiser’s bare minimum of .50
score, so the
sample size was adequate for factor analysis. In addition, Bartlett's test of sphericity is statistically significant, x² (3) = 65.54, p < .001, indicating that the correlations between items were sufficiently large for PAF. Moreover, the Factor Matrix showed that all the items were correlated positively, while the variable measuring the thoughts of the respondents if they would already start to wear clothes from the brand Stylo had the strongest association (factor loading was .80). After this, the reliability of the scale was checked in order to see whether a new variable can be created. The results showed that the reliability of the scale is good, Cronbach’s alpha = .73, meaning that the scale measured customer-brand identification. Therefore, a new variable called “Customer-Brand Identification” was computed.
worked, the respondents were asked to answer the following question: “In your opinion, do you think that the brand Stylo’s photos are more like spontaneous photos or more like pre-planned photos?”, having two answer options: Spontaneous (snapshot) photos and Pre-planned (studio) photos.
Filler questions. In order to eliminate the possibility that the respondents didn’t judge only the style of the images, they were asked to give their opinion about the clothes and about the fashion models in the pictures to see if these factors influenced their answers: “On a scale from 0 - 10 (0 = the lowest grade; 10 = the highest grade), how much do you like the models who appear in Stylo’s photos on Instagram?” and “On a scale from 0 - 10 (0 = the lowest grade; 10 = the highest grade), how much do you like the clothes which appear in Stylo's photos on Instagram?”.
Analytic Approach
Before running specific analyses in order to test the hypotheses, the missing values were checked and reported. They represented the non-answer from the respondents who dropped out from the survey and only in the case of the variable age,
eight values were
dropped from the study because they did not meet the age requirements. For each scale item, measures of centrality and dispersion have been calculated and the scale reliability was checked. Moreover, the outliers were verified for every variable used in the tests and there were few outliers for the variable customer-brand identification and Instagram usage. Conducting Descriptive Statistics, I compared the median values with the corresponding mean values, and it can be concluded that for all variables the mean values were not biased by outliers. Finally, the normality assumptions were checked and the results showed that the sampling distribution of the mean was normal and all the variables used in the statistical tests had a symmetric bell-shaped curve.
In order to test Hypothesis 1, independent t-test was the preferential option because brand attitudinal loyalty was continuous variable and the experimental conditions were categorical variables and hence, the assumption of variable type was met. Moreover, the assumption for using the t-distribution as the theoretical probability distribution was met (N > 30). Therefore, considering that the assumptions were met, an independent t-test analysis with brand attitudinal loyalty as the dependent variable and the type of image as the independent variable was run.
To test Hypothesis 2, a MANOVA was conducted with both variables used to measure customer-brand identification as dependent variables and type of images as the independent variable. Hence, the assumption of variable type was met because the dependent variable was continuous and the independent variable was categorical, having two groups. The results of normality tests showed that only the diagram variable was significant (p < .05) for
both experimental groups, meaning that it didn't meet the normality assumption. However, the histograms showed that the overall responses were quite normally distributed, so I considered that the dependent variables are multivariate normally distributed for each population. The Covariance test was not significant (p > .05), so the assumption of covariance was met. The Levene’s test was also not significant for both the dependent variables (p > .05), so the test of equality of variance was met. Thus, the population variances and covariances among the dependent variables were the same across all levels of the independent variable. Hence, considering that the assumptions were met, I could go further with the analyses.
In order to test hypothesis 3 and the model in Figure 1, two separate PROCESS models (model 4) were run with the brand-attitudinal loyalty as the dependent variable, the type of image as the independent variable, and either the scale variable measuring CBI or the
diagram item measuring CBI as a mediator. The assumption of variable type was met and therefore, it was determined that the mediation was appropriate to be tested. Regarding linearity, a series of regressions were run. For the sake of space, there will be presented only the results of the X and M predicting Y regression, but all other relationships respected the assumptions. For both variables measuring CBI (the diagram and scale variables), the regression appears linear since the Loess curve center is close to zero along the entire X-axis. Moreover, from the same scatterplot, it could be noticed that the data spread on the Y-axis quite consistently and slightly equally throughout the plot. The estimation error was normally distributed because the data fitted well with the diagonal line. Therefore, considering that the assumptions were met, I could go further with the mediation analysis via PROCESS model 4.
Finally, to check whether the manipulation went correctly, a cross-tab was conducted with the type of image variable and the categorical variable used to verify whether the respondents named correctly the images they saw during the experiment.
Figure 1. The Conceptual model of this study. Results
Hypothesis 1 stated that the brand’s activity of posting snapshot photos on Instagram leads to higher brand attitudinal loyalty of millennials compared to traditional studio photos. To test this hypothesis, an independent t-test was conducted with the type of images as
the independent variable and brand attitudinal loyalty as the dependent variable. The Levene’s test showed an F-value of .01 and a p-value of .947 which is above the critical p-value < 0.05 for a 95% Confidence Interval. Hence, the homogeneity of variance was assumed. However, the results showed that on average, respondents from the traditional studio condition (M = 3.63; SD = 1.47) scored similar on brand attitudinal loyalty as respondents in the snapshot condition (M = 3.62; SD = 1.40): t(96) = .04, p = .972, 95% CI [-.56, .58]. Therefore, hypothesis 1 was rejected.
Hypothesis 2 stated that the brand’s activity of posting snapshot photos on Instagram leads to higher brand identification of millennials compared to traditional studio photos. To test the direct effect of the experimental conditions on customer-brand identification, a MANOVA test was conducted with both variables used to measure customer-brand identification as dependent variables and the type of images as the independent variable. It should be noted that the assumption of equal variances in the
population has been met for
both dependent variables (the diagram variable, p = .978, and the scale variable p = .724). There was an insignificant difference between respondents in the traditional studio image condition and the respondents in the snapshot image condition when considered jointly on the variables measuring costumer-brand identification, Wilk’s Λ = .98, F(2, 96) = .81, p = .446, partial η2
= .02. A separate ANOVA was run for each dependent variable, with each ANOVA evaluated at an alpha level of .025. There was not a significant difference between respondents in the traditional image condition and the respondents in the snapshot condition on the diagram variable measuring CBI, F(1, 97) = .46, p = .497, partial η2 = .01, or the scale variable measuring CBI, F(1, 97) = .03, p = .856, partial η2 = .01. Therefore, hypothesis 2 was rejected.
Instagram has a positive indirect effect on millennials’ attitudinal loyalty via increased customer-brand identification. The results of the first PROCESS model showed that the relationship between the snapshot condition and the diagram variable measuring CBI is not statistically significant, b = .26, t(1, 96) = .76, p = .446, CI [-.41, .93]. The direct relationship between CBI and brand loyalty was statistically significant, b = .56, t(2, 95) = 8.50, p < .001, CI [43, .70]. The results of the indirect effect showed that there was a positive, but insignificant, indirect effect of the snapshot images on brand attitudinal loyalty via the mediator CBI, B = .15, SE = .19, 95% BCI [-.22, .53]. The other PROCESS model showed that the relationship between the snapshot condition and the scale variable measuring CBI is not statistically significant, b = -.02, t(1, 96) = -.07, p = .942, CI [-.45, .42]. The direct effect of CBI (scale variable) on brand loyalty is statistically significant, b = 1.06, t(2, 95) = 13.28, p < .001, CI [.91, 1.22]. Overall, there was a negative, non-significant, indirect effect of the snapshot images on brand attitudinal loyalty via the mediator CBI (measured with the scale variable), B = -.02, SE = .22,93%, 95% BCI [-.47, .44]. Hence, hypothesis 3 was rejected.
Regarding the manipulation check, a cross-tab was run with the condition variable and the categorical variable used to verify whether the respondents named correctly the images they saw during the experiment. The results showed that for the respondents in the traditional studio condition, the manipulation went very well because only 3 respondents (6.3%) answered wrongly, while 45 respondents (93.8%) identified correctly the traditional studio images. In the case of the snapshot condition, the manipulation was not good enough because there were more respondents who identified wrongly the images than the ones who identified them correctly: only 23 respondents (46.9%) named well the snapshot images, while 26 respondents (53.1%) answered wrongly that the images are traditional studio pictures. The results of the Chi-Square showed that the association between these two variables was statistically significant, x² (1) = 20.46, p < .001.
Discussion
Summary of the Research Findings and Theoretical implications
This study is one of the first to investigate the influence of the brand’s activity of posting snapshot photos on Instagram on millennials’ attitudinal brand loyalty, considering the mediating role of consumer-brand identification. It was expected that millennials identify themselves with the informal vibe of the fashion snapshot-like images and subsequently, the immediate effect of this identification would lead to brand attitudinal loyalty. Statistical analysis of the data yielded several findings that will be summarized and discussed in more detail in this section.
The research results provided interesting information. One of the major implications of the current study showed that the use of snapshot aesthetic by a fashion brand on Instagram does not lead to millennials' attitudinal loyalty. These findings are opposed to the results found in the existing study (Colliander & Marder, 2018) on the effects of exposure to snapshot-like images on brand attitudes and intentions to enact word-of-mouth. The latter study showed that within Instagram, a snapshot aesthetic has greater
meaning for the
users because they can translate it more fluently. Thus, users liked more the usage of snapshot images within Instagram and this aesthetic was associated with the increased positive attitude of the brand and higher intention to enact word-of-mouth. Possible reasons for this difference in results are first, that the previous study investigated the longer effects of snapshot images (respondents were exposed to the conditions over a period of one week) and second, that the brand attitude and word-of-mouth intention were the outcome variables. The present research investigated the immediate effects of snapshot images on the brand attitudinal loyalty, as the outcome variable.
Another implication of the results is that they showed the notion that the relation between the brand's activity of posting snapshot images on Instagram and brand attitudinal loyalty is not mediated by customer-brand identification. This finding does not sustain the prediction that millennials would identify themselves in the brand’s snapshot photos, despite the previous research which states that millennials want to see a human touch in the brand photos which they can use to express their own personality and subsequently, lead to consumer identification with the brand (Black Bear Design, 2016; Kim et al., 2001). This non-identification could be explained by the fact that the process of identification requires a longer time to be created. This assumption is supported by previous literature which affirms that brand identification is a dynamic process that develops over time (Bhattacharya & Sen, 2003).
Moreover, the notion upheld by previous studies that costumer-brand identification has a mediation role on costumers’ brand loyalty in different fields: Corporate Social Responsibility (Marin et al., 2009) and pharmaceutical sales (Ahearne et al., 2005) is not valid for the fashion context of the current study. The findings showed that posting snapshot-like images by fashion brands on Instagram does not make millennials to identify with these pictures and therefore, this process does not have an immediate indirect
effect on millennials' brand
attitudinal loyalty. This discovery contradicts the previous theory that individuals who perceive the brand image to be consistent with their self-image are likely to engage in brand-related activities, such as brand loyalty (Bhattacharya & Sen, 2003; Heath & Scott, 1998). This difference in results can be explained also by the sample groups investigated because previous studies on CSR and pharmaceutical sales focused on adults (aged over 27 years) and not on millennials. Perhaps the mediation role of customer-brand identification in the present relation could be significant if older adults would be the target group because this age group
takes into account the emotional goals in the brand attachment more than young adults (Jahn, Gaus, & Kiessling, 2012), so they could identify quicker in the brand images.
Despite this non-significant indirect effect, this paper found a positive relationship between customer-brand identification and brand attitudinal loyalty. This finding is in line with the conclusion of previous studies which state that customer-brand identification is a strong predictor of customer behavior, such as brand loyalty (Donavan et al., 2006; Kuenzel & Vaux Halliday, 2008). Moreover, the current study strengthens the Social Identification Theory that consumers with increased brand identification are more willing to engage in brand-related activities, such as brand loyalty (Bhattacharya & Sen, 2003). Therefore, this paper adds the fashion industry next to the fields (such as car and hotel industry) in which the positive connection between customer-brand identification and brand loyalty was demonstrated by antecedent studies (Alnawas & Altarifi, 2016; Kim et al., 2001).
Managerial implications
The study provides several managerial implications. Fashion brands engaging users on Instagram should not consider the norms of this medium in the decisions regarding the aesthetic of image content created because the snapshot aesthetic which is congruent with these norms did not outperform the traditional studio aesthetic in the context of this research. Despite the recommendation of the previous study (Colliander &
Marder, 2018)
which states that fashion brands should tailor the content aesthetic to suit the particular medium they use, the current paper advice fashion brands to tailor their image aesthetic based on other factors, such as the product type categories. As Schroeder (2010) affirms that snapshot aesthetics represents a casual image of brands, there is a reason to consider that this aesthetic is more suitable for the casual product lines of the brand, while for the
haute couture products, brands could use a more professional style, such as traditional studio images.
Furthermore, considering the notion stated by the latter author (2010) that snapshot-like photos often incorporate the fine line between formal photography features (such as posing and editing) and spontaneity features (such as photography as experience and randomness), it can be premediated that the marketers can also use a mix of the snapshot and traditional studio aesthetics when promoting fashion brands on Instagram, in order to better engage with the customers. Given that snapshot images look like amateur generated photos, perhaps marketers should consider to use them only as consumer-generated ads, letting consumers become both subjects and producers of the brand’s image content, while the brand-generated ads can contain a more professional aesthetic, such as traditional studio photos.
Limitations and Future Research
The present study had at least three limitations which can be useful to be addressed for future research. First, the aim of this paper was to measure the immediate effect of the type of aesthetics, respondents were exposed only once to the stimuli and having only 10 seconds per image to analyze it. However, as I mentioned above, it can be considered that the formation of the consumers’ identification process with the image aesthetic might form differently over a longer time period with repeated exposure of the respondents to the stimuli. Therefore, on one hand, my recommendation for future
researchers is to replicate
this study or conduct new studies of this nature over a long time period (longitudinal design), considering repeated exposure to the stimuli, in order to investigate the prolonged effect of snapshot aesthetics. On the other hand, in order to further investigate the immediate effect of image aesthetic, researchers could focus on other mediators and
outcome measures that are more easily and quickly affected at the moment, such as liking of the images or preferences.
Second, this study did not make a distinction between heavy users and non-heavy users of fashion accounts on Instagram. Prior research (Colliander & Marder, 2018) revealed that due to the congruence with the informal decorum of social media, snapshot images are processed by consumers with increased meaning. Based on this statement, there is a reason to believe that heavy users, accustomed to the features of Instagram and fashion images, might identify details in the pictures that non-heavy users cannot. Hence, these two groups might respond differently to the image aesthetics and therefore, future research should investigate and compare their reactions.
Lastly, this study benefited from the convenience sampling method in order to collect a high amount of data in a short time span. However, this method has the disadvantage that the sample is not chosen at random, so it is unlikely to be representative of the population from which it was taken. Therefore, the findings of this study cannot be generalized to the millennial population. My recommendation for future studies is to consider probability sampling methods, such as the simple random sample. Conclusion
Overall, the current study showed that the brand’s activity of posting snapshot images on Instagram does not have the potential to immediately stimulate consumers’ brand attitudinal loyalty. In addition, snapshot images do not immediately lead to consumer-brand identification and therefore, this type of images do not have an indirect effect on millennials’ attitudinal loyalty, through consumer identification with the brand. Therefore, in the context of Instagram, the findings of the current study do not support the use of snapshot-like images by clothing brands to determine millennials’ brand attitudinal loyalty, at least in the context of a one-time exposure. Future research among
millennials is therefore needed to deepen this knowledge about the influence of snapshot aesthetic on brand loyalty and the mediating role of customer-brand identification.
THE EFFECTS OF SNAPSHOT PHOTOS 30
References
Ahearne, M., Bhattacharya, C. B., & Gruen, T. (2005). Antecedents and consequences of customer-company identification: Expanding the role of relationship marketing. Journal of Applied Psychology, 90(3), 574. doi:10.1037/0021-9010.90.3.574
Alnawas, I., & Altarifi, S. (2016). Exploring the role of brand identification and brand love in generating higher levels of brand loyalty. Journal of Vacation Marketing, 22(2), 111– 128. doi:10.1177/1356766715604663
Arnett, J. J. (2007). Emerging adulthood: What is it, and what is it good for? Child
Development Perspectives, 1(2), 68–73. doi.org/10.1111/j.1750-8606.200700016.x Bergami, M., & Bagozzi, R. P. (2000). Self‐categorization, affective commitment, and group
self‐esteem as distinct aspects of social identity in the organization. British Journal of Social Psychology, 39(4), 555–577. doi/abs/10.1348/014466600164633
Bhattacharya, C. B., & Sen, S. (2003). Consumer-company identification: A framework for understanding consumers’ relationships with companies. Journal of Marketing, 67(2), 76–88. doi/10.1509/jmkg.67.2.76.18609
Black Bear Design. (2016). How Instagram is Becoming a Millennial Marketing Treasure Trove. Retrieved from https://www.blackbeardesign.com/blog/millennials-and-instagram/
Carlson, B. D., Todd Donavan, D., & Cumiskey, K. J. (2009). Consumer-brand relationships in sport: brand personality and identification. International Journal of Retail & Distribution Management, 37(4), 370–384. doi.org/10.1108/09590550910948592 Chua, T. H. H., & Chang, L. (2016). Follow me and like my beautiful selfies: Singapore
teenage girls’ engagement in self-presentation and peer comparison on social media. Computers in Human Behavior, 55, 190–197. doi.org/10.1016/j.chb.2015.09.011
THE EFFECTS OF SNAPSHOT PHOTOS 31
Colliander, J., & Dahlén, M. (2011). Following the Fashionable Friend: The Power of Social Media: Weighing Publicity Effectiveness of Blogs versus Online Magazines. Journal of Advertising Research, 51(1), 313–320. doi:10.2501/JAR-51-1-313-320
Colliander, J., & Marder, B. (2018). ‘Snap happy’ brands: Increasing publicity effectiveness through a snapshot aesthetic when marketing a brand on Instagram. Computers in Human Behavior, 78, 34–43. doi.org/10.1016/j.chb.2017.09.015
Diderich, J. (2017). Balenciaga Subverts Logo in Understated Fall Ads. Retrieved from https://wwd.com/business-news/media/balenciaga-fall-advertising-campaign-demna-gvasalia-10894221/
Donavan, D. T., Janda, S., & Suh, J. (2006). Environmental influences in corporate brand identification and outcomes. Journal of Brand Management, 14(1–2), 125–136. Retrieved from https://link.springer.com/article/10.1057/palgrave.bm.2550057 Fowler Jr, F. J. (2013). Survey research methods. Sage publications.
Fox, B. (2004). Rethinking the amateur. Spectator, 24(1), 65-79. Retrieved from https://cinema.usc.edu/archivedassets/098/15844.pdf
Fromm, J., Butler, C., & Dickey, C. (2015). How to engage Millennials: Re-imagining the consumer as a partner, not a target audience, to increase engagement. Journal of Brand Strategy, 4(1), 27-36. Retrieved from
https://www.ingentaconnect.com/content/hsp/jbs/2015/00000004/00000001/art00004 Geurin-Eagleman, A. N., & Burch, L. M. (2016). Communicating via photographs: A
gendered analysis of Olympic athletes’ visual self-presentation on Instagram. Sport Management Review, 19(2), 133–145. doi.org/10.1016/j.smr.2015.03.002
Gurău, C. (2012). A life-stage analysis of consumer loyalty profile: comparing Generation X and Millennial consumers. Journal of Consumer Marketing, 29(2), 103–113.
THE EFFECTS OF SNAPSHOT PHOTOS 32
Heath, A. P., & Scott, D. (1998). The self-concept and image congruence hypothesis: An empirical evaluation in the motor vehicle market. European Journal of Marketing 32(11/12), 1110-1123. doi:10.1108/03090569810243749
Hornsey, M. J. (2008). Social identity theory and self‐categorization theory: A historical review. Social and Personality Psychology Compass, 2(1), 204–222.
doi.org/10.1111/j.1751-9004.2007.00066.x
Hughes, D. E., & Ahearne, M. (2010). Energizing the reseller’s sales force: The power of brand identification. Journal of Marketing, 74(4), 81–96. Retrieved from
https://www/jstor.org/stable/27800827
Hwang, J., & Kandampully, J. (2012). The role of emotional aspects in younger consumer-brand relationships. Journal of Product & Brand Management, 21(2), 98–108. doi:10.1108/10610421211215517
Jahn, S., Gaus, H., & Kiessling, T. (2012). Trust, commitment, and older women: Exploring brand attachment differences in the elderly segment. Psychology & Marketing, 29(6), 445-457. doi:10.1002/mar.20533
Kim, C. K., Han, D., & Park, S.-B. (2001). The effect of brand personality and brand
identification on brand loyalty: Applying the theory of social identification. Japanese Psychological Research, 43(4), 195–206. doi.org/10.1111/1468-5884.00177
Kuenzel, S., & Vaux Halliday, S. (2008). Investigating antecedents and consequences of brand identification. Journal of Product & Brand Management, 17(5), 293–304. doi:10.1108/10610420810896059
Lam, S. K., Ahearne, M., Hu, Y., & Schillewaert, N. (2010). Resistance to brand switching when a radically new brand is introduced: A social identity theory perspective. Journal of Marketing, 74(6), 128–146. Retrieved from https://www.jstor.org/stable/25764287
THE EFFECTS OF SNAPSHOT PHOTOS 33
Lam, S. K., Ahearne, M., Mullins, R., Hayati, B., & Schillewaert, N. (2013). Exploring the dynamics of antecedents to consumer–brand identification with a new brand. Journal of the Academy of Marketing Science, 41(2), 234–252. doi:10.1007/s11747-012-0301-x
Marin, L., Ruiz, S., & Rubio, A. (2009). The role of identity salience in the effects of corporate social responsibility on consumer behavior. Journal of Business Ethics, 84(1), 65–78. doi:10.1007/s10551-008-9673-8
Martínez, P., & del Bosque, I. R. (2013). CSR and customer loyalty: The roles of trust, customer identification with the company and satisfaction. International Journal of Hospitality Management, 35, 89–99. doi:10.1016/j.ijhm.2013.05.009
Odin, Y., Odin, N., & Valette-Florence, P. (2001). Conceptual and operational aspects of brand loyalty: An empirical investigation. Journal of Business Research, 53(2), 75–84. doi:10.1016/S0148-2963(99)00076-4
Pew Research Center. (2018). Demographics of Social Media Users and Adoption in the United States. Retrieved from http://www.pewinternet.org/fact-sheet/social-media/ Quester, P., & Lin Lim, A. (2003). Product involvement/brand loyalty: is there a link? Journal
of Product & Brand Management, 12(1), 22–38. doi:10.1108/10610420310463117 Reber, R., Schwarz, N., & Winkielman, P. (2004). Processing fluency and aesthetic pleasure:
Is beauty in the perceiver’s processing experience?. Personality and Social Psychology Review, 8(4), 364–382. doi:10.1207/s15327957pspr0804_3
Sarvas, R., & Frohlich, D. M. (2011). From snapshots to social media-the changing picture of domestic photography. Springer Science & Business Media.
doi:10.1007/978-0-85729-247-6
Schroeder, J. E. (2013). Style and Strategy: Snapshot Aesthetics in Brand Culture. In Imagining Organizations (pp.139-161): Routledge.
THE EFFECTS OF SNAPSHOT PHOTOS 34
So, K. K. F., King, C., Sparks, B. A., & Wang, Y. (2013). The influence of customer brand identification on hotel brand evaluation and loyalty development. International Journal of Hospitality Management, 34, 31–41. doi:10.1016/j.ijhm.2013.02.002 Stets, J. E., & Burke, P. J. (2000). Identity theory and social identity theory. Social
Psychology Quarterly, 224–237. Retrieved from https://www.jstor.org/stable/2695870 Tansey, M. (2017). Fashion & Loyalty: Analysis of Consumer Loyalty in the Clothing
Industry (Master's thesis). Retrieved from
https://repository.cardiffmet.ac.uk/handle/10369/8665
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Basil Blackwell. Valkenburg, P. M., & Peter, J. (2013). The differential susceptibility to media effects model.
Journal of Communication, 63(2), 221-243. doi:10.1111/jcom.12024
Veselinovic, M. (2014). Cash a la Mode: How Style Bloggers Are Turning Social Savvy into Six-Figure Salaries. Retrieved from
THE EFFECTS OF SNAPSHOT PHOTOS 35
Appendix
Appendix A – The created Instagram accounts
For the snapshot photos: For the traditional studio photos:
THE EFFECTS OF SNAPSHOT PHOTOS 36
Appendix B – The Stimuli Used Snapshot images:
THE EFFECTS OF SNAPSHOT PHOTOS 37
Traditional studio images:
THE EFFECTS OF SNAPSHOT PHOTOS 38
