Beyond the realm of consciousness
Examining subconscious effects of advertising on Facebook
Alina Kosoriga Student ID: 10875751
Master's Thesis
Graduate School of Communication Master's programme Communication Science
Supervisor: Sanne Kruikemeier 21 June 2016
Abstract
The pervasiveness of social networking sites (SNS) among different types of consumers makes such platforms very attractive for marketing purposes. Yet, even though a large number of people use SNS on a daily basis, many consumers ignore commercial content on such platforms. This raises concerns about effectiveness of advertising on SNS, as click-through rates for banner advertisements are decreasing. The present study focuses on the most popular SNS Facebook and investigates whether consumers, who do not pay directed attention to a banner advertisement, are affected by it on a subconscious level and
consequently develop positive implicit attitudes toward the brand. Moreover, consumers’ explicit and implicit attitudes are compared across the three conditions (directed attention vs. non-directed attention vs. control condition). Finally, it is examined whether implicit
attitudes can be an underlying mechanism in the relationship between both levels of attention to the banner advertisement and purchase intention of the brand. Although the experiment (N = 142) revealed no significant differences in attitudes and purchase intention across directed, non-directed and control conditions, results of the study show that implicit attitudes
significantly affect purchase intention of the brand. Such findings suggest that
subconsciously formed attitudes are able to predict consumers’ behavior. This study serves as a good starting point in the research of subconscious processing of advertisements on SNS. Directions for future studies, as well as theoretical and practical implications are given.
Beyond the realm of consciousness
Examining subconscious effects of advertising on Facebook
Today marketers increasingly employ social networking sites (SNS) in order to reach out to potential customers and to place their product in consumers’ consideration set
(Ramsaran-Fowdar & Fowdar, 2013). While there exist various interactive ways of promoting a brand, among which is creation of fan pages and development of various contests, display advertising is still frequently used in order to increase product sales. As a result, a large number of social media include commercial content, such as banner
advertisements and sponsored posts.
Despite the increasing popularity of social media for marketing, the majority of people use SNS to communicate with each other, which often results in annoyance toward many advertising tactics (Fournier & Avery, 2011). Consequently, people try to avoid
commercial content on social platforms. A number of studies found that web users developed a so-called “banner blindness”, which is a skill that helps them to easily ignore banner
advertisements that were made specifically to attract attention (Benway & Lane, 1998; Dreze & Hussherr, 2003; Burke, Hornof, Nilsen & Gorman, 2005; Barreto, 2013). This means that consumers do not cognitively encode banner advertisements, which can seriously hinder marketers’ attempts to persuade potential consumers.
Yet, even when consumers do not consciously pay attention to an advertisement, their implicit attitudes toward the brand can be influenced due to the subconscious registering of the banner. The key difference between explicit and implicit attitudes is that whereas the former refer to consumers’ deliberate evaluations of the stimulus, the latter refer to automatic evaluations that are formed outside of consumers’ awareness (Gawronski & Bodenhausen, 2006). Thus, unlike explicit attitudes, implicit attitudes do not require consumers’ cognitive
processing in order for them to develop certain evaluation of the stimulus. Correspondingly, according to the mere exposure effect, repeated exposure to a stimulus can increase one’s affective response towards it because of the subconscious familiarity with it (Bornstein & D’Agostino, 1992). Besides, it has been found that consumers’ attitudes toward the brand can increase even after a single exposure to an advertisement (Briggs & Hollis, 1997; Yoo, 2008).
Since a large number of people use such SNS as Facebook constantly in their everyday life (Duggan, Ellison Lampe, Lenhart & Madden, 2015, ‘Social Media Update 2014’), the aim of this thesis is to examine whether consumers can develop positive implicit brand attitudes without conscious attention to banner advertisements, as in such case they might have little opportunity to establish explicit attitudes. Level of attention is an important factor, as some consumers might be interested in the product and hence pay more attention to the advertisement (Maughan, Gutnikov & Stevens, 2007), while others can mainly direct their cognitive assets to reading Facebook posts (Hadija, Barnes & Hair, 2012). While in the former case consumers’ interest in the advertised product will make them have the directed attention to the advertisement, in the latter situation consumers will not examine the
advertisement closely and thus will have the non-directed attention to the banner. Moreover, in the latter case consumers will allocate most of their mental resources to the message processing and hence will not have enough cognitive capacity to consciously process banner advertisements (for more information, see Lang, 2000).
Yet, the end goal for every advertising campaign is to instigate consumers to make a purchase. Thus, it is valuable to examine whether consumers with the non-directed attention to the advertisement have higher purchase intentions of the brand than consumers who are not exposed to the advertisement at all, as such difference might be attributed to the change in consumers’ implicit attitudes toward the brand. Besides, since implicit attitudes were found to be predictive of purchase intentions even for consumers with established explicit attitudes
(Vantomme, Geuens, De Houwer & De Pelsmacker, 2005), it is also important to examine whether implicit attitudes predict purchase intentions of consumers, who paid direct attention to the advertisement. Therefore, this thesis will examine whether implicit attitudes toward the brand mediate the relationship between level of attention to the advertisement and purchase intention of the brand.
Users are expected to exert more effort to consciously avoid banner advertisements on SNS than on traditional websites due to the social nature of such platforms (Barreto, 2013). This study will contribute to the research field by collecting empirical data from the most popular social platform Facebook, as at the moment there is a lack of research on
subconscious effects of web advertising on SNS. Moreover, as the previous research almost did not investigate the role of implicit attitudes in formation of purchase intention, this study will shed the light on such relationship. Furthermore, this study will be of relevance for marketers, as it will provide evidence of whether consumers’ implicit attitudes toward the brand can be changed without their attention to the advertisement and whether such attitudes function as an underlying construct between level of attention and intention to buy product. Hence, the overarching research question of this study is:
RQ: What is the effect of different levels of attention to a banner advertisement on SNS on consumers’ explicit and implicit attitudes toward the brand and can implicit attitudes explain the relationship between level of attention to the advertisement and purchase intention of the brand?
Theoretical framework Explicit attitudes
The Internet has become one of the most used channels for marketing, as it delivers a number of benefits when compared to traditional advertising. One of the main advantages of
the Internet is that web advertising can stimulate consumers to take an instant action, thus making it possible to directly measure campaign effectiveness through click-through or conversion rates (Hoffman & Novak, 1996; Yoo, 2008).
In order to target advertisements to specific groups of consumers, marketers
increasingly employ various social networking sites (SNS). One of such platforms, namely Facebook, is especially popular for marketing purposes, as it has more than 1,5 billion monthly active users and is the most popular social network in the world (Statista, 2016). Marketers can make use of the website’s special advertising system that contains
demographic and psychographic information of every Facebook user and thus allows for precise consumer targeting (Maurer & Wiegmann, 2011). Therefore, the attractive opportunity to effectively reach selected target audience makes companies continuously invest their money in display advertising on this social platform (eMarketer, 2015). However, as click-through rates for banner advertisements are decreasing (Barreto, 2013), this raises question as to what extent display advertising on the Internet is still effective. One of the reasons that consumers often ignore commercial banners is because there is usually a large number of advertisements on the same page. Indeed, according to Nelson-Field, Riebe and Sharp (2013), Facebook banner ads are competing with several ads on the same page, which leads to advertising clutter.
Nevertheless, one of the most important factors that may hinder advertising tactics of marketers on SNS is the so-called banner blindness of consumers. This phenomenon can be described as users’ intuitive avoidance of web content that resembles advertising, due to former encounters with commercial banners, as well as knowledge of their location and dimensions (Barreto, 2013). According to Lapa (2007), knowledgeable Internet users
automatically recognize the structure of a webpage, which helps them to distinguish between useful information and unnecessary clutter. Indeed, the study conducted by Benway and Lane
(1998) revealed that visually salient and prominent banners, which were supposed to attract attention of consumers, were largely ignored. Similarly, by measuring participants’ eye movements, Dreze and Hussherr (2003) found that web surfers did not look at more than half of the commercial banners they were exposed to, as they purposefully tried to avoid such advertisements. This indicates that consumers tend to predict expected location of a banner advertisement, recognize it as irrelevant information with the help of peripheral vision and subsequently avoid it.
Since social networking platforms were developed for users to communicate online with their friends and acquaintances, commercial messages on such websites are often considered to be intrusive (Fournier & Avery, 2011). In their research of teenagers’ motives for advertising avoidance on SNS, Kelly, Kerr and Drennann (2010) found that banners’ irrelevance and credibility concerns were among the main reasons youngsters disregarded commercial messages. Similarly, Hadija, Barnes and Hair (2012) found in their study that normally users do not have negative opinions toward advertising on SNS, but they simply ignore it. Hence, banner blindness phenomenon can be especially prominent on Facebook, as consumers do not look for commercial information on this website and can repeatedly filter persuasive content, for it is incongruent with their goals. As a matter of fact, results of the study carried out by Barreto (2013) by means of measuring participants’ eye movements revealed that only 65 out of 254 displayed Facebook banners were seen and only 10 percent of participants clicked on a banner advertisement to get to the advertiser’s page. According to Barreto (2013), the main reasons for the weak advertising effectiveness on Facebook include lack of consumers’ interest toward commercial banners, as well as banner blindness
phenomenon. Similarly, in his research on priming effects of aggressive advertisements on Facebook, Buchanan (2015) discovered from participants’ feedback that they indeed disregarded banner advertisements during the experiment, as they try to avoid them in
everyday life. Since banners are usually located outside of users’ area of focus, it is rather easy for them to ignore such advertisements.
Another aspect that can explain why a large number of consumers fail to notice banner advertisements on SNS is because it is assumed that people possess selective
attention. This is due to the fact that a person’s mental capacity is limited (Kahneman, 1973; Lang, 2000). Researchers claim that the overall mental capacity to process all of the activities that a person is engaged in can be split in two parts: capacity used to complete a primary task and reserve capacity (Kahneman, 1973; Lang, 2000). Reserve capacity is usually spent on processing incidental changes in one’s surroundings, as well as small secondary tasks (Lee & Faber, 2013). Therefore, the larger the amount of mental capacity person devotes to the primary task, the lesser the amount of mental capacity is left for processing incidental changes.
Since the Web is over-saturated with content, consumers need to be able to focus their cognitive assets on a particular task they are performing or information they are looking for (Cho & Cheon, 2004). When people use their Facebook page to communicate with their friends or they get absorbed into reading a particular story, their cognitive assets are directed at performing this specific task. Hence, consumers might not be able to allocate their mental resources to conscious processing of banner advertisements in such situations. As a matter of fact, in their research on consumers’ attention to banner advertisements on different types of webpages, Hsieh and Chen (2011) found that people devote less attention to commercial banners on text-based and text-picture mixed webpages as compared to picture-based and video-based sites. This is due to the fact that text-based and text-picture mixed webpages require larger amount of consumers’ mental resources for processing (Hsieh & Chen, 2011). Since Facebook includes both text posts and picture posts, it can be considered as being
text-picture mixed webpage and thus banner advertisements are often overlooked there (Barreto, 2013).
Yet, users are not always focused on one particular task when they visit their Facebook page. Sometimes people do not have a specific goal for checking this SNS and merely spend their time browsing through available information. According to the
Elaboration Likelihood Model (ELM), people can process incoming information via central or peripheral route, depending on their level of involvement with the task. Petty and
Cacioppo (1986) claim that when people possess cognitive assets, find the information interesting and relevant to their goals, they are highly involved in the task and thus more likely to take a central route of information processing. As a result, consumers engage in active thinking and carefully evaluate arguments in the message. However, when consumers are not motivated to focus on the task at hand and/or do not have enough cognitive capacity to process incoming information, they are likely to take a peripheral route, which will make them more susceptible to secondary cues in the environment. In such case, people are easily distracted from paying attention to the content of the main message and more likely to take note of such distractors as banner advertisements (Hsieh & Chen, 2011). Therefore, when consumers do not direct their attention to a specific task or story on Facebook, their focus on banner advertisements is likely to be stronger.
Consumers’ attention to the advertisement is an important prerequisite for them to develop conscious attitudes toward the advertised brand (Clark, Brock & Stewart, 1994; Lee & Ahn, 2012). When consumers engage in cognitive elaboration of the advertisement, they create certain associations with this stimulus in their memory. Consequently, such effect as uncertainty reduction might take place (Lee, 2001). According to Obermiller (1985), uncertainty reduction describes an increase in liking of the stimulus due to the developed familiarity with it after cognitive encoding of the stimulus. Indeed, research by Lee (2001)
showed that participants had more positive attitudes toward the stimulus that was presented to them earlier, as compared to the stimulus that they have not seen before. Such increased affect is likely to be developed due to elimination of tension and mistrust toward the stimulus (Obermiller, 1985). Positive evaluation of the stimulus occurs because of consumers’
cognitive judgments and, according to Yoo (2008), effect of uncertainty reduction should be distinguished from the mere exposure effect, which takes place on a subconscious level1. Besides, such result can occur after a single exposure; repetition purely strengthens the effect of uncertainty reduction (Obermiller, 1985).
Based on the preceding discussion, it is expected that when people are involved in a primary task, they allocate their cognitive assets to the task at hand and do not consciously pay attention to the content that is incongruent with their goals. When, however, consumers do not have a definite goal for visiting a Facebook page, they are more likely to pay attention to the secondary information, such as banner advertisements. This increases familiarity with the advertised brand, which might lead to the increased affect toward the brand because of uncertainty reduction. In this research, level of attention to the banner will be manipulated by directing participants’ focus either to the written post or to the advertisement. Additionally, a control condition without a banner advertisement will be used as a baseline measure. As participants in the non-directed attention condition will be distracted from the advertisement, the following research sub-question is proposed2:
RQ1: Do participants in the directed attention condition have better explicit attitudes toward the brand than participants in the non-directed attention condition and do participants in the
1 The mere exposure effect will be elaborated upon in the following section. 2
As it was expected that two out of three conditions would not significantly differ on examined dependent variables, employment of research sub-questions was deemed more suitable in testing for the effects than
non-directed attention condition differ in their explicit attitudes toward the brand from participants in the control condition?
Implicit attitudes
Although there is a general tendency for many web users to avoid banner advertisements, which leads to a decrease of click-through rates (Yaveroglu & Donthu, 2008), it is of crucial importance to not underestimate the role of subconscious processing of such commercial information. Despite the fact that choice of a certain product is a conscious decision, such decision can be considerably influenced by subconsciously formed attitudes (Janiszewski, 1990).One process that can lead to change of attitude toward the brand on a subconscious level is preattentive processing of commercial information. According to Ryu, Lim, Tan and Han (2007), preattentive processing is subconscious monitoring of all sensory channels in a person’s environment during performance of a primary task, in order for him or her to recognize occurrences that may require a shift of attention to another action. Such inspection of one’s environment differs from attentive processing in that it does not require person’s conscious thinking and it is outside of one’s control (Ryu et al., 2007). Therefore, even though a person does not have to allocate his or her cognitive resources to preattentive processing, he or she can still register information in his or her peripheral field of view, such as banner advertisements, while being engrossed in a story or an activity on the webpage (Ryu et al., 2007; Chatterjee, 2008). Such analysis, however, is not accessible to a person’s consciousness and memory due to a small amount of cognitive assets allocated to preattentive processing.
A phenomenon that results from preattentive processing of commercial information and explains why consumers can develop subconscious positive brand attitudes without a memory of seeing an advertisement is the mere exposure effect (Janiszweski, 1993).
for a certain stimulus that they have been repeatedly exposed to, but of which they were not consciously aware. As consumers do not pay conscious attention to the stimulus, they do not recognize its influence and thus an increase in affect toward the brand can be attributed to the sheer familiarity with the stimulus. Such an influence can occur in situations when
consumers’ cognitive capacity is preoccupied with another primary task and they do not recognize even the presence of an advertisement (Lee & Ahn, 2012).
Despite the fact that people are not aware of involuntary perception of commercial information located in their peripheral visual field, several studies found that consumers can nonetheless be subconsciously influenced in their affect toward the brand. In his research on varying effects of different advertisement locations in newspapers, Janiszewski (1990) found that preattentive processing of an advertisement placed to the left of an attended passage led to an increase in participants’ affect toward the brand, as well as it increased the probability for participants to choose the brand from an array of other options. Results of the study confirmed that such attitudes were formed outside of participants’ conscious awareness (Janiszewski, 1990). Similarly, in their research, Shapiro, Heckler and MacInnis (1997a) found that even when participants were busy with completing a primary task on the webpage (i.e. reading an article) and did not remember seeing an advertisement, their attitude toward the brand was increased due to preattentive processing of the banner.
As stated previously, when conscious attention to the stimulus is paid, an increase in explicit affect toward such stimulus might take place due to uncertainty reduction (Lee, 2001). With regard to implicit evaluation of the stimulus, Menon and Raghubir (2003) claim that the ease with which previously encoded information comes to mind can be involuntary used as a cue in automatic evaluation of the stimulus. According to Menon and Raghubir (2003), mere accessibility of the recently processed information is used unintentionally in the formation of one’s judgment and can determine evaluation of the stimuli. As cognitive
elaboration of the previously seen advertisement creates associations with the stimulus in one’s memory and may reduce mistrust in the brand, such connection when easily retrieved, can automatically affect implicit evaluations of the brand. Hence, since in this study implicit attitudes are measured shortly after exposure to the advertisement, it is expected that the familiarity and mere accessibility of the recently encoded advertisement can lead to an increase in implicit affect toward the brand in the directed attention condition.
Although in the present research both directed attention and non-directed attention conditions contain the same banner advertisement, it is plausible to assume that a single exposure and a short exposure time might not be enough for participants in the non-directed attention condition to develop positive explicit attitudes toward the brand. Nevertheless, based on the theoretical notions and empirical findings presented earlier, it is probable that participants in the non-directed attention condition will preattentively process the banner advertisement, which might have a positive effect on their implicit brand attitudes. Correspondingly, cognitive encoding and mere accessibility of the advertisement in the directed attention condition is expected to influence participants’ implicit attitudes in a positive direction as well. Therefore, the following research sub-question is formulated:
RQ2: Do participants in the directed attention condition and participants in the non-directed attention condition differ in their implicit attitudes toward the brand and do participants in the non-directed attention condition have better implicit attitudes toward the brand than participants in the control condition?
Purchase intention
While it can be possible for a banner advertisement to influence consumers’ implicit attitudes toward the brand without their conscious awareness, the ultimate goal for every commercial campaign is to instigate consumers to make a purchase. It is, thus, important to
understand the role of implicit affect in guiding consumers’ purchase intention and whether implicit attitudes can act as underlying mechanism in the relationship between both directed and non-directed attention toward the advertisement and purchase intention.
According to the Motivation and Opportunity as Determinants (MODE) model, if a person has the intention and opportunity to make a certain deliberative decision, his or her actions will be influenced primarily by his or her explicit attitudes (Fazio & Towles-Schwen, 1999). Conversely, in a spontaneous decision-making, when both motivation and opportunity to consider available information are limited, implicit attitudes will be more predictive of the behavior. Fazio and Towles-Schwen (1999) claim that deliberative decision-making requires careful consideration of possible actions in a particular situation, which is only possible when a person has a motivation to engage in rational thinking. Consequently, if a person is
motivated to consider advantages and shortcomings of a particular action, by doing so he or she ensures that he or she will avoid costly mistakes. Besides, according to the MODE
model, to be able to perform effortful consideration, a person needs to have the opportunity to do so (Fazio & Towles-Schwen, 1999). However, when there is no motivation or opportunity to carefully consider possible benefits and disadvantages of certain behaviour, a person is expected to act in accordance with his or her implicit attitudes (Schuette & Fazio, 1995; Fazio & Towles-Schwen, 1999). Such attitudes are most likely to impact person’s
spontaneous behaviour, with him or her not necessarily being aware of their existence and their influence.
While some studies found evidence in support of the MODE model (Sanbonmatsu & Fazio, 1990; Fabrigar, Petty, Smith & Crites, 2006; Rhodes & Ewoldsen, 2009), other empirical findings demonstrate that implicit attitudes are able to guide not only spontaneous, but deliberative behavior as well. Specifically, several studies found that implicit stereotypes can strongly predict deliberative decision-making. For example, in their study Agerström and
Rooth (2011) found that hiring managers’ implicit attitudes toward obese people predicted the hiring discrimination in real corporate settings. According to Agerström and Rooth (2011), labour market discrimination was present despite managers’ differing explicit attitudes. Furthermore, in their study of drug and alcohol nurses’ desires to change the job, von Hippel, Brener and von Hippel (2008) found that nurses’ implicit prejudice toward injecting drug users mediated the relation between their self-reported stress and intention to leave drug and alcohol nursing. Explicit prejudice, however, did not mediate this relation. Therefore, empirical evidence proves that although implicit attitudes can impact spontaneous behaviors, they can play a significant role in predicting thoughtful and rational actions as well.
Purchase of a certain product can be either a deliberative or spontaneous action, depending on whether the purchase was planned and whether a consumer has certain prior experience with a product (Baumgartner, 2002). For product purchase to be a deliberative action, purchase intention is a necessary prerequisite, as it constitutes consumer’s deliberation and hope to buy a specific product (Rizwan, Qayyum & Qadeer, 2014). As a matter of fact, implicit attitudes were found to have an effect on such deliberative decision as intention to buy a product. In their research, Vantomme et al. (2005) found that participants, who intended to buy ecological products, had more positive implicit attitudes toward ecological brands than traditional products. Moreover, the findings of the study revealed that
participants’ positive implicit attitudes toward ecological products correlated significantly with their intention to purchase such products, whereas their explicit attitudes did not
correlate with purchase intention (Vantomme et al., 2005). Hence, empirical evidence shows that implicit attitudes can play an important role in formation of purchase intention of a certain product.
It is important to acknowledge that consumers’ final decisions about whether or not to purchase a certain product are dependent upon at least four basic brand offers, such as price, product, promotion and place (McCarthy, 1964). This means that aside from the clearly important role of promotion in motivating consumers to make a purchase, their behavioral reactions are also based on the perceived value of the product for the offered price (Kwon & Schumann, 2001). Therefore, as the focus is on brand promotion and there are a number of other variables that may influence the actual purchase behavior, the attention is paid to participants’ purchase intentions. The ultimate purchase, however, can be affected by a number of aforementioned factors.
In their research, Shapiro, MacInnis and Heckler (1997b) found that when participants were busy with reading a passage of text and did not cognitively process an advertisement, the advertised brand was still very likely to be included in participants’ consideration set. Such a result occurred despite the fact that participants were exposed to the advertisement only once and did not even recollect seeing it. The present study aims to extend the research by Shapiro et al. (1997b) by investigating whether the non-directed attention to the banner would be sufficient for consumers to develop purchase intention of the brand. Besides, this thesis will investigate how participants’ purchase intention in the non-directed attention condition differs from purchase intention of participants in the directed attention condition.
Furthermore, in their related research, Shapiro et al. (1997a) found that under similar experimental conditions participants unconsciously developed a liking for the brand, even though they were engrossed in another task and did not have a memory of the advertisement. Based on the preceding discussion, it is assumed that implicit attitudes have the ability to affect purchase intention. Besides, due to preattentive processing of the banner advertisement in the non-directed attention condition and automatic retrieval of the recently encoded
is expected to increase after exposure to the banner advertisement in both conditions. Therefore, it is important to investigate whether implicit attitudes are an underlying
mechanism in the relationship between both levels of attention and purchase intention of the brand. Given that implicit attitudes function as mediator between both directed and non-directed attention to the banner advertisement and purchase intention of the brand, it would indicate that cognitive encoding of the advertisement is not a necessary prerequisite for consumers to develop purchase intention of the brand. Therefore, the following research sub-questions are derived:
RQ3: Do participants in the directed attention condition and participants in the non-directed attention condition differ in their purchase intentions of the brand and do participants in both directed attention condition and non-directed attention condition have higher purchase intentions of the brand than participants in the control condition?
RQ4: Are implicit attitudes toward the brand positively associated with purchase intention of the brand?
RQ5: Is the effect of attention to the banner advertisement on purchase intention of the brand mediated by participants’ implicit attitudes toward the brand?
Method Design and sample
This study had a randomized between subjects design with a single factor Level of
attention with three levels: directed attention, non-directed attention and control condition.
participants were recruited through a personal Facebook account of the experimenter. The reason for recruiting people online is attributed to the fact that the focus of this research is on knowledgeable users of Facebook. Therefore, by recruiting people directly from this SNS it could be ensured that they are familiar with this platform and are active users of it.
A total of 200 participants took part in the study. 58 participants were excluded from the sample, as some of them indicated that they do not have a Facebook account, whereas others did not provide answers to several survey questions. The final sample consisted of 142 participants, 35% were male (SD = 0.48). Participants were between 18 and 56 years old (M = 27.87, SD = 8.02). Most participants indicated to come from Ukraine (37.3%), 17.6% from the Netherlands, 6.3% from Germany, 5.6% from Greece, 5.6% from Russia and 27.6% indicated to come from other 18 countries (SD = 8.97). The majority of the respondents were highly educated, with 41.5% having obtained Bachelor’s degree and 40.1% having obtained Master’s degree. 18.4% of participants were lower educated (SD = 1.36).
Stimulus
In order to develop stimulus material, two screenshots of Facebook page with and without banner advertisement were created (see Appendix A). Each screenshot featured a Facebook post (i.e. story) written by a student about the misfortune that happened to his cat. The post was located at the center of the page, thus simulating the actual appearance of the information posted by people on Facebook. While participants in the control condition were exposed to the Facebook page without the banner, participants in directed and non-directed attention condition saw the identical page with the banner advertisement.
The commercial banner was an advertisement for the unknown coffee brand Black
Swan. For two experimental conditions this banner was located in the right sidebar of the
Facebook page. Such placement was chosen due to the fact that this part of Facebook page is usually devoted to commercial banners and thus the location of Black Swan banner was
congruent with the location of real banner advertisements. Furthermore, it was expected that it would be easier for participants in the non-directed attention condition to overlook the advertisement in the sidebar, as their cognitive load would be directed to the information in the Facebook post and the banner would not be in their field of vision. In addition, according to Lapa (2007), Internet users become knowledgeable about the structure of the webpage and are able to automatically differentiate between important information and the unnecessary content for their goals, such as banner advertising.
Moreover, despite the fact that in 2012 new forms of advertising, such as ‘sponsored posts’ and ‘suggested posts’, were introduced on Facebook (Darwell, 2012, ‘Facebook tests page’), they are designed to look as standard Facebook posts and thus their commercial nature is less obvious. Therefore, such commercial posts are not defined as banner advertisements and they were not chosen as prototypes for the stimulus.
Procedure
An invitation to participate in the study was sent to participants via personal Facebook account of the experimenter. Once participants clicked on the link in the invitation, they were transferred to the Qualtrics page. Before taking part in the experiment, participants were presented with the factsheet and letter of consent, to which they had to agree in order to continue their participation. Next, participants were asked several socio-demographic questions (i.e. their age, gender, usage of Facebook and nationality), after which they were randomly assigned to one of the three conditions (directed attention vs. non-directed attention vs. control). Depending on the condition that participants were assigned to, they received instructions to either focus on the advertisement or on the Facebook post.
Participants could study the Facebook page for as long as they wanted to. When they were done, participants were asked three questions about either Facebook post or banner advertisement. Next, they were directed to the first part of the implicit test, during which they
had to rank five product brands in five product categories (i.e., milk, cookies, yogurt, coffee, chips) from the least preferred to the most preferred brand by dragging a picture of the brand to a desired place. Coffee category contained the target brand Black Swan. When participants were done, they moved to the second part of the implicit test, during which they had to write five product brands in their preferred order for five product categories. Product categories in the second part of the implicit test were identical to product categories in the first part (i.e., milk, cookies, yogurt, coffee, chips). After the implicit test, participants’ explicit attitudes and purchase intention were measured with the help of questionnaire, as well as they were asked whether they knew about Black Swan brand prior to the experiment. Finally,
participants were debriefed about the nature of the experiment and were thanked for taking part in the study.
Measures
Level of attention. This independent variable was manipulated by directing participants’ focus to either Facebook post or banner advertisement. Participants in the directed attention condition were asked to carefully study the banner advertisement in terms of its appearance and how it fits in the webpage as a whole. They were told to pay close attention to such aspects as banner size, its colors and location. This way participants’ attention in the directed attention condition was drawn to the advertisement. Participants in the non-directed attention condition were exposed to the identical page with the
advertisement, but this group was asked to carefully study the information in the written post as though the purpose of the study was to measure their understanding of the presented content. It is assumed that their cognitive assets were spent on studying information carefully and not on the advertisement (Shapiro et al., 1997b). Participants in the control condition were exposed to the same Facebook page, but without the banner and received the same instructions as participants in the non-directed attention condition. The choice of such
manipulation was adapted from the research by Yoo (2008), who studied the effects of unconscious processing of web advertising.
Implicit attitudes: stimulus-based brand choice. The measurement of this variable was adapted from the research by Vandeberg, Murre, Voorveld and Smit (2015), primarily due to the possibility of measuring this way implicit attitudes online. Participants were asked to organize five product brands for five product categories in their preferred order (1 = most
preferred – 5 = least preferred). Each product set was presented on a separate page and once
participants completed organizing the products in one category, they could move to the next one. Five categories covered such products as milk, cookies, yogurt, coffee and chips. Coffee category contained the target brand Black Swan. Participants could move every brand in each set by dragging and dropping the picture of the product in the desired place. The target brand was scored based on its position on the list and subsequently the values were recoded in the reverse order (1 = least preferred – 5 = most preferred), (M = 2.74, SD = 1.36).
Implicit attitudes: memory-based brand choice. The measurement of this variable was inspired by the measures used by Vandeberg et al. (2015), primarily due to the
possibility of measuring this way implicit attitudes online. Participants were asked to list five product brands in their preferred order for each of the five product categories. Five product names were presented on one page and together they covered such products as milk, cookies, yogurt, coffee and chips. If a participant entered Black Swan in the coffee category, such response was given a value of 2; otherwise it was given a value of 1 (M = 1.03, SD = 0.17).
Explicit attitudes. In order to strengthen the measurement of such latent construct as ‘explicit attitudes’, two scales were employed. This variable was measured by three seven-point Likert items derived from the study by Sengupta and Johar (2002), as well as it was measured by five dichotomous items derived from the research by Spears and Singh (2004). An example of the seven-point Likert item is: “I think Black Swan is a very good coffee” (1
= Strongly disagree – 7 = Strongly agree). An example of the dichotomous item is: “I think Black Swan coffee is” (1 = Unappealing – 2 = Appealing). One scale was created for the three seven-point Likert items measuring explicit attitudes. An exploratory factor analysis showed that the scale was unidimensional (EV = 2.21), explaining 73.53% of the variance. The reliability of the scale is good, Cronbach’s alpha = .82 (M = 12.33, SD = 2.55). A second scale was created for the five dichotomous items. The reliability of the scale is very good, the Kuder-Richardson 20 coefficient of reliability = .93 (M = 3.82, SD = 1.87). Therefore, two separate scales measuring participants’ explicit attitudes toward the brand were computed. Purchase intention. Participants’ purchase intention was measured by three seven-point Likert items adapted from Putrevu and Lord (1994). An example of the item is: “It is very likely that I will buy Black Swan coffee” (1 = Strongly disagree – 7 = Strongly agree). An exploratory factor analysis indicated that the scale was unidimensional (EV = 2.48), explaining 82.70% of the variance. The reliability of the scale is very good, Cronbach’s alpha =.90 (M = 10.54, SD = 4.35). Therefore, it is plausible to conclude that the scale measures participants’ purchase intention of the brand.
Socio-demographic variables. To ensure that participants did not significantly differ in their socio-demographic characteristics across the three conditions, such variables as gender, age, level of education and nationality were assessed (for specific information about
M and SD see p. 17).
Manipulation check. In order to check whether the manipulation was successful, after exposure to the stimulus participants were asked three questions about the page element that their attention was directed to. While participants in the control and the non-directed attention condition were asked questions about the written post (e.g.: “What was the name of the cat in the story?”), participants in the directed attention condition were asked questions about the banner advertisement (e.g.: “Do you think the colors of the banner advertisement
were pleasant?”). Moreover, to examine whether participants were aware of the Black Swan brand prior to the experiment, at the end of the questionnaire they were asked a question: “Have you heard of Black Swan coffee prior to the experiment?” The answer options were ‘Yes’ or ‘No’.
Analysis
In order to answer the first three research questions, several one-way ANOVAs were performed. This test was chosen due to the fact that it allows analyzing for the presence of effect of two or more levels of the independent variable on the continuous dependent variable. As the first three research questions are concerned with finding such effects, this statistical analysis was deemed as being the most suitable. In order to answer research questions 4 and 5, however, PROCESS SPSS macro developed by Hayes (2013) was employed, as it allows testing for the indirect effect of the independent variable on the dependent variable through the mediator, as well as it shows the direct effect of the mediator on the dependent variable.
Results Randomization check
To test if random assignment of participants to the three conditions was successful, first one-way ANOVA with level of attention as the independent variable and age as the dependent variable was carried out. The analysis revealed that there was no effect of the attention condition on age, F (2,139) = 0.12, p = .891. Hence, this test result shows that participants did not differ in their age across three conditions. Next, to test whether
participants were randomly distributed by gender, Chi-square test was conducted with level of attention variable and gender variable. The relation between these variables was non-significant, χ2 (2, N = 142) = 0.12, p = .942, which indicates that there were no differences across three conditions on gender. In order to examine whether participants were randomly
assigned to three conditions by regions of Europe, Chi-square test was carried out with level of attention variable and European regions variable (see Appendix C). The analysis revealed that the relation between these variables was non-significant, χ2 (8, N = 128) = 11.29, p = .186. This means that participants did not significantly differ in their national background across the three conditions. Lastly, to examine whether participants were randomly
distributed by level of education, one-way ANOVA with level of attention condition as the independent variable and education as the dependent variable was performed. The results showed that there was no effect of the attention condition on level of education, F (2,139) = 0.28, p = .754. This test result indicates that participants did not differ in their level of
education across conditions. Overall, it can be concluded that randomization was successful. Main results
To answer the first research sub-question relating to explicit attitudes, a one-way ANOVA with level of attention as the independent variable and explicit attitude (scale 1) as the dependent variable was performed. The results revealed a significant, weak main effect, F (2,139) = 3.07, p = .049, η2 = .042. However, despite a significance found in the main effect, a Bonferroni post-hoc test revealed that participants in the directed attention condition (M = 4.02, SD = 0.79, n = 54) do not hold significantly better explicit attitudes toward the brand than participants in the non-directed attention condition (M = 3.99, SD = 0.80, n = 49, p = .999) and participants in the non-directed attention condition do not hold significantly better explicit attitudes than participants in the control condition (M = 4.39, SD = 0.95, n = 39, p = .076). Another one-way ANOVA with level of attention as the independent variable and explicit attitude (scale 2) as the dependent variable was performed to examine for the possible differences in the results. The analysis indicated that none of the three conditions differ significantly in their explicit attitudes toward the brand, F (2,139) = 0.24, p = .788, η2 = .003. Table 1 shows the mean scores on explicit attitudes for all conditions. Therefore, based
on the aforementioned results for explicit attitudes for both scales, we can answer the research sub-question 1 and conclude that participants in the directed attention condition do not have better explicit attitudes toward the brand than participants in the non-directed attention condition and that participants in the non-directed attention condition do not differ in their explicit attitudes toward the brand from participants in the control condition.
Participants in all three conditions do not differ in their explicit attitudes toward the brand.
Table 1
Means and Standard Deviations of Explicit Attitude, Implicit Attitude and Purchase Intention by Level of Attention.
Condition Explicit attitudes (scale 1) Explicit attitudes (scale 2) Implicit attitudes Purchase intention M SD M SD M SD M SD Control 4.39 0.95 0.75 0.40 2.77 1.29 3.80 1.51 Non-directed 3.99 0.80 0.75 0.38 2.55 1.32 3.31 1.36 Directed 4.02 0.79 0.79 0.35 2.89 1.45 3.49 1.47
Since only four participants from the entire sample wrote down Black Swan brand for the memory-based brand choice variable, this measure of implicit attitudes was deemed unsuitable due to the lack of responses containing the target brand and hence it was not used in the analysis. Thus, only the stimulus-based brand choice measure was used in the analysis of participants’ implicit attitudes. Therefore, to answer the research sub-question 2 relating to implicit attitudes, a one-way ANOVA with level of attention as the independent variable and implicit attitudes as the dependent variable was carried out. The test revealed a
non-significant, weak effect, F (2,139) = .80, p = .451, η2 = .011. Table 1 shows the mean scores on implicit attitudes for all conditions. Hence, this result gives us the answer to the research sub-question 2 and we can conclude that various levels of attention to the banner
advertisement do not have an impact on participants’ implicit attitudes toward the brand. Participants in all three conditions do not differ in their implicit attitudes toward the brand.
To answer the research sub-question 3 relating to purchase intention, a one-way ANOVA with level of attention as the independent variable and purchase intention as the dependent variable was carried out. The test revealed that none of the three conditions differ significantly in their purchase intentions of the brand, F (2,139) = 1.30, p = .277, η2 = .018. The mean scores on purchase intention for all conditions are displayed in Table 1. Therefore, the results answer research sub-question 3 and indicate that various levels of attention to the banner advertisement do not have an effect on participants’ purchase intention of the brand. All three conditions do not differ in purchase intention of the brand.
Before conducting an analysis to answer the research sub-questions 4 and 5, it is important to point out that there exist various views on conditions that should be met prior to testing for a mediation relationship. According to the ‘causal steps approach’ by Baron and Kenny (1986), only when the direct effect of the independent variable on the dependent variable (path c), as well as the effect of the independent variable on the mediator (path a) and the effect of the mediator on the dependent variable (path b) are significant, one can test for the presence of mediation. Mediation is present when the effect of the independent variable on the dependent variable is significantly less with the controlled mediator (path c’) than without control of the mediator (Baron & Kenny, 1986).
However, according to Hayes (2009), mediation can be present even when some constituent paths in the mediation model are not significant. For example, in small samples path a and path b may be non-significant because of low statistical power (Preacher & Hayes, 2004). Therefore, series of separate tests that do not involve the actual indirect effect might be not powerful enough to detect mediation. Hayes (2009) claims that such procedure as bootstrapping is more effective in detecting indirect effects even without significant separate
paths in the model, as the original dataset is repeatedly resampled and the indirect effect is calculated in each resampled dataset (Preacher & Hayes, 2008). This method does not assume normal distribution in the sample, which makes it a very useful approach for testing
mediation in small samples (Preacher & Hayes, 2008). Therefore, the bootstrapping method is widely applied in behavioral sciences as it has greater statistical power (Morrison & Heimberg, 2013). Hence, as this study made use of the small sample and two separate paths (i.e. path a and path c) were not found to be significant, the bootstrapping statistical tool PROCESS developed by Hayes (2013) was used to answer the research sub-questions 4 and 5.
In order to carry out the mediation analysis, level of attention was used as the independent variable, purchase intention was used as the dependent variable and implicit attitudes was specified as mediator. Since the independent variable ‘level of attention’
consists of three categories (i.e., directed attention vs. non-directed attention vs. control), this variable was specified as being multicategorical. Thus, PROCESS automatically constructed two dummy variables D1 and D2 for the non-directed attention and directed attention
condition respectively, while keeping control condition as constant. Mediation analysis based on 1000 bootstrapped samples showed that implicit attitudes toward the brand do not mediate the effect of non-directed attention on purchase intention, b1 = - 0.41, t (138) = -1.42, p =
.158, CI [-0.99, 0.16] and that implicit attitudes do not mediate the effect of directed attention to the banner on purchase intention, b2 = -0.36, t (138) = -1.28, p = .204, CI [-0.92, 0.20].
Results revealed that there is neither a significant effect of non-directed attention on implicit attitudes, b1 = - 0.22, t (139) = - 0.75, p = .457, CI [-0.80, 0.36], nor a significant effect of directed attention on implicit attitudes, b2 = 0.12, t (139) = 0.42, p = .677, CI [-0.45, 0.69].
However, there is a significant effect of implicit attitudes on purchase intention3, b = 0.39, t (138) = 4.63, p < .001, CI [0.22, 0.56] (see Figure 1). This result answers the research sub-question 4, as it means that implicit attitudes significantly affect purchase intention. However, since the analysis revealed that there is no significant effect of two levels of attention on implicit attitudes toward the brand and that there is no significant indirect effect through a mediator, it gives us an answer to the research sub-question 5 and indicates that implicit attitudes do not mediate the effect of level of attention toward the advertisement on purchase intention of the brand.
Figure 1. The Mediating Role of Implicit Attitude in the Effect of Level of Attention on Purchase Intention.
Note: D1 – non-directed attention, D2 – directed attention. ** = p < .01
Discussion and conclusion
The aim of this thesis was to examine the role of attention to a banner advertisement on the most popular SNS, Facebook, in consumers’ affect toward the brand and purchase intention of the brand. The effects of different levels of attention to a banner advertisement
3
To confirm the significant effect of implicit attitudes on purchase intention, a Spearman’s rank-order correlation was carried out. There was a strong, positive correlation between implicit attitudes and purchase
on consumers’ explicit and implicit attitudes toward the brand were explored. Moreover, it was analyzed whether implicit attitudes can lead to purchase intention of the brand, as well as whether they are an underlying mechanism between different levels of attention and purchase intention. The purpose of this research was to explore whether conscious attention to a banner advertisement on SNS is a necessary prerequisite for consumers to develop positive brand evaluation and intention to purchase the brand.
First of all, the results of two analyses conducted with two different scales revealed that while participants in the non-directed attention condition and control condition did not differ in their explicit attitudes, participants in the directed attention condition did not hold significantly better explicit attitudes either. These findings do not correspond with results of the research by Lee (2001), who found that participants’ affect toward the stimulus increases after the cognitive encoding of it due to activated familiarity with the stimulus. One of the possible reasons that no significant relationship between directed attention to the banner and explicit attitudes toward the brand was found is because a single exposure to an unfamiliar brand could have been not enough to evoke positive affect. Although previous research provided evidence that one exposure to an advertisement can be sufficient to influence brand attitudes (Briggs & Hollis, 1997; Yoo, 2008), the findings of the current study do not support this view.
The results also indicated that neither directed attention, nor non-directed attention to the banner significantly influenced participants’ implicit attitudes toward the brand. Although participants in both directed attention condition and non-directed attention condition did not differ in their implicit attitudes, they also did not have better implicit attitudes than
participants in the control condition. These results are inconsistent with the predicted increase in implicit attitudes in the directed attention condition due to the mere accessibility of
attention condition, the results contradict empirical findings of Janiszewski (1990) and Shapiro et al. (1997a), who found an increase in consumers’ evaluation of a brand due to preattentive processing of an advertisement. Failure to find significant effects on participants’ implicit attitudes could be ascribed to the replication of real banner placement on Facebook page. Whereas the actual Facebook advertisements are located in the right sidebar of the page, Janiszewski (1990) claims that consumers are more likely to evaluate brand positively when a pictorial advertisement in placed on the left of the attended text. Hence, Facebook authorities should consider changing location of banner advertisements, as consumers’ subconscious processing of commercial information can be limited simply because of improper placement of banners.
In addition, this study aimed to extend the research by Shapiro et al. (1997b) and examine whether the non-directed attention to the advertisement would be enough for participants to develop purchase intention of the brand and whether it would differ from purchase intention of participants in the directed attention condition. The results showed that neither participants in the non-directed attention condition, nor participants in the directed attention condition had higher purchase intention of the brand than participants, who were not exposed to the advertisement. Such findings do not correspond with the previous research, which found that consumers are more likely to include brand into their consideration set after the unconscious processing of an advertisement (Shapiro, et al., 1997b; Yoo, 2008). The insignificant findings could be ascribed to the fact that intention to purchase a product is a more serious decision than inclusion of a brand into one’s consideration set, as it
characterizes a preference for a specific brand among the set of alternatives (Chand, n.d., ‘Understanding Purchase decision’). Since participants’ purchase intentions were measured right after the exposure to the advertisement, the very short time between exposure and measurement of purchase intentions could have been not enough for them to develop such
decisions. Therefore, future research should focus on analyzing purchase intentions after a certain period of time, when associations with the advertised brand are strengthened in consumers’ minds.
Nevertheless, the analysis revealed a significant effect of implicit attitudes toward the brand on purchase intention, which indicates that increase in implicit attitudes led to the increase in intention to purchase this brand. Such results are in line with the research by Vantomme et al. (2005), who found that implicit attitudes towards ecological brands significantly predicted participants’ purchase intention of these products. Moreover, the discovered effect is in conflict with the MODE model (Fazio & Towles-Schwen, 1999), which posits that implicit attitudes affect spontaneous behaviors, while consciously formed attitudes are predictive of deliberative actions. However, the findings of the present study suggest that in order to develop purchase intention of the brand, consumers do not necessarily have to hold positive explicit attitudes, as their decision about whether to purchase the
product or not can be influenced by attitudes formed outside of their awareness. Finally, the results revealed that implicit attitudes did not mediate the effect of directed and non-directed attention on purchase intention. This finding could be ascribed to the fact that although the original sample was repeatedly resampled by means of
bootstrapping procedure, the relationship between both directed and non-directed attention and implicit attitudes remained non-significant. This shows that neither participants in the non-directed attention condition, nor participants in the directed attention condition had significantly better implicit attitudes than participants in the control condition, which led to the insignificant mediation effect of implicit attitudes. Future research should examine whether consumers’ implicit attitudes increase after several exposures to an advertisement and whether a different location of a banner affects implicit attitudes as well. Subsequently, it
would be valuable to further investigate the mediating effect of implicit attitudes, as the current research revealed that implicit attitudes do affect purchase intention.
Limitations and future research
Several limitations should be taken into account when interpreting results of the current study. One of the limitations can be ascribed to the measurement of participants’ implicit attitudes. Generally, implicit attitudes are measured with the help of the Implicit Association Test (IAT), which provides a very accurate depiction of consumers’
subconscious attitudes without interference of their explicit evaluations (Gibson, 2008). As the present research was conducted online, the IAT could not be employed due to the absence of available tool for remote measures. Thus, reported implicit attitudes could have been influenced by participants’ explicit judgments, as the measurement method used in this study was less advanced. Therefore, future research should adopt the IAT in order to detect
variations in consumers’ implicit attitudes more accurately.
Furthermore, in the present research participants could study the Facebook page for as long as they wanted to. This could result in additional attention to the advertisement from participants in the non-directed attention condition, as their eyes could wander around the page after finishing the reading task. Thus, in order to achieve more effective manipulation in the future, it is suggested to establish a fixed timeframe for stimulus exposure. In addition, an eye-tracking technology could be used in order to understand the pattern of participants’ eye movements during exposure to an advertisement.
Finally, participants’ explicit and implicit attitudes, as well as purchase intention were measured very shortly after exposure to the advertisement. The absence of a time break between exposure to the stimulus and brand evaluation could have contributed to the fact that participants’ attitudes and intention to purchase the product were not significantly influenced by the recently encoded information. Therefore, future research should measure consumers’
attitudes, as well as purchase intention in a certain period of time after exposure to an advertisement (e.g. 24 hours), as empirical research found presence of unconscious effects after such period of time (Yoo, 2008).
Theoretical and practical implications
This research contributed to the field by studying the effects of subconscious
processing of advertisements specifically on SNS. Despite the widespread use of such SNS as Facebook for marketing purposes (Statista, 2014), at the moment there is a lack of research addressing subconscious processing of banners in the context of social media. Even though there were no significant effects of a single banner exposure on consumers’ brand evaluations and purchase intention, this research provided the evidence of implicit attitudes affecting purchase intention of the brand. As previous studies mainly found the effect of implicit attitudes on spontaneous actions (Sanbonmatsu & Fazio, 1990; Fabrigar, Petty, Smith & Crites, 2006; Rhodes & Ewoldsen, 2009), results of the present research show that subconsciously formed attitudes can also affect deliberative behaviors. Hence, this study demonstrates the importance of investigating the effects of subconscious brand evaluations on consumer behavior and serves as a starting point in this important direction in the communication field.
This thesis offers some valuable insights for practitioners as well. Since
non-significant effect of the banner advertisement on participants’ implicit attitudes is inconsistent with previous research (Janiszewski, 1990; Shapiro et al., 1997a), such results indicate that placement of advertisements in the right sidebar might diminish their influence on
consumers’ implicit brand evaluations. Therefore, as subconsciously formed attitudes do have an influence on consumers’ purchase intention, Facebook executives should consider changing location of banner advertisements to the left side of the page. As Facebook provides its services for free due to commercial content (Ostrow, 2011, ‘Facebook Will Never
Charge’)and click-through rates for banners are decreasing, it is in Facebook authorities’ best interest to facilitate effectiveness of banners on their website in order for marketers to continue investing in it.
Finally, advertisers should acknowledge the impact of implicit brand evaluations on intention to purchase products and consider adjusting their promotional tactics on social media accordingly. Perhaps, instead of trying to attract as much attention to their
advertisements as possible, advertisers could invest in more frequent, but short exposures. This might increase chances for consumers to process banner advertisements preattentively and change their implicit attitudes, instead of becoming irritated by the attention-grabbing banners. Despite the fact that consumers might not remember seeing the advertisements, it is possible that they will be influenced in their purchase intentions nonetheless.
References
Agerström, J., & Rooth, D. O. (2011). The role of automatic obesity stereotypes in real hiring discrimination. Journal of Applied Psychology,96(4), 790.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of
personality and social psychology, 51(6), 1173.
Barreto, M.A. (2013). Do users look at banner ads on Facebook?.Journal of Research in
Interactive Marketing, 7(2), 119-139.
Baumgartner, H. (2002). Toward a personology of the consumer. Journal of Consumer
Research, 29(2), 286-292.
Benway, J.P. & Lane, D.M. (1998) "Banner blindness: Web searchers often miss 'obvious' links". internetworking, December 5
Bornstein, R. F., & D'Agostino, P. R. (1992). Stimulus recognition and the mere exposure effect. Journal of personality and social psychology, 63(4), 545.
Briggs, R., & Hollis, N. (1997). Advertising on the Web: Is there response before click-through? Journal of Advertising Research, 37, 33–46.
Buchanan, T. (2015). Aggressive priming online: Facebook adverts can prime aggressive cognitions. Computers in human behavior, 48, 323-330.
Burke, M., Hornof, A., Nilsen, E., & Gorman, N. (2005). High-cost banner blindness: Ads increase perceived workload, hinder visual search, and are forgotten. ACM Transactions
on Computer-Human Interaction (TOCHI),12(4), 423-445.
Chand, S. (n.d.). Understanding Purchase Decision Process with Purchase Decision Model. Retrieved from http://www.yourarticlelibrary.com/decision-making/understanding-purchase-decision-process-with-purchase-decision-model/22167/
Chatterjee, P. (2008). Are unclicked ads wasted? Enduring effects of banner and pop-up ad exposures on brand memory and attitudes. Journal of electronic commerce Research,
9(1), 51.
Cho, C.H., & Cheon, H.J. (2004). Why do people avoid advertising on the Internet? Journal
of Advertising, 33, 4, 89–97.
Clark, E. M., Brock, T. C., & Stewart, D. W. (1994). Attention, attitude, and affect in
Darwell, B. (2012). Facebook tests page post ads in News Feed to reach users beyond fans
and friends. Retrieved from http://www.adweek.com/socialtimes/facebook-tests-page-post-ads-in-news-feed-to-reach-users-beyond-fans-and-friends/283322?red=if
Drèze, X., & Hussherr, F. X. (2003). Internet advertising: Is anybody watching?. Journal of
interactive marketing, 17(4), 8-23.
Duggan, M., Ellison, N.B., Lampe, C., Lenhart, A. & Madden, M. (2015). Social Media
Update 2014. Retrieved from http://www.pewinternet.org/2015/01/09/social-media-update-2014/
eMarketer, (2015). Social Network Ad Spending to Hit $23.68 Billion Worldwide in 2015. Retrieved from http://www.emarketer.com/Article/Social-Network-Ad-Spending-Hit-2368-Billion-Worldwide-2015/1012357#sthash.7F247WwF.dpuf
Fabrigar, L. R., Petty, R. E., Smith, S. M., & Crites Jr, S. L. (2006). Understanding
knowledge effects on attitude-behavior consistency: the role of relevance, complexity, and amount of knowledge. Journal of Personality and Social Psychology, 90(4), 556. Fazio, R. H., & Towles-Schwen, T. (1999). The MODE model of attitude-behavior
processes. Dual-process theories in social psychology, 97-116.
Fournier, S., & Avery, J. (2011). The uninvited brand. Business horizons,54(3), 193-207. Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in
evaluation: an integrative review of implicit and explicit attitude change. Psychological
bulletin, 132(5), 692.
Gibson, B. (2008). Can evaluative conditioning change attitudes toward mature brands? New evidence from the Implicit Association Test. Journal of Consumer Research, 35(1), 178-188.
Hadija, Z., Barnes, S. B., & Hair, N. (2012). Why we ignore social networking
advertising. Qualitative Market Research: An International Journal, 15(1), 19-32. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new
millennium. Communication monographs, 76(4), 408-420.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. Guilford Press.
Hoffman, D. L., & Novak, T. P. (1996). Marketing in the hyper- media computer-mediated environments: Conceptual foundation. Journal of Marketing, 60, 50–68.
Hsieh, Y. C., & Chen, K. H. (2011). How different information types affect viewer’s attention on internet advertising. Computers in Human Behavior,27(2), 935-945.
