power of consumers using social media examining the ......as arousal, fear, and liking as well as...

Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/289122307

Powerofconsumersusingsocialmedia:

Examiningtheinfluencesofbrand-related

user-generatedcontentonFacebook

ArticleinComputersinHumanBehavior·May2016

DOI:10.1016/j.chb.2015.12.047

CITATION

1

READS

606

2authors:

AngellaJ.Kim

CaliforniaStatePolytechnicUniversity,Po…

13PUBLICATIONS239CITATIONS

SEEPROFILE

KimKPJohnson

UniversityofMinnesotaTwinCities

115PUBLICATIONS852CITATIONS

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,

lettingyouaccessandreadthemimmediately.

Availablefrom:KimKPJohnson

Retrievedon:03October2016

Full length article

Power of consumers using social media: Examining the influences ofbrand-related user-generated content on Facebook

Angella J. Kim a, *, Kim K.P. Johnson ba Department of Apparel Merchandising and Management, California State Polytechnic University Pomona, 3801 West Temple Avenue, Pomona, CA 91768,USAb Department of Design, Housing and Apparel, University of Minnesota, 240 McNeal Hall, 1985 Buford Avenue, St. Paul, MN 55108, USA

a r t i c l e i n f o

Article history:Received 18 June 2015Received in revised form16 December 2015Accepted 18 December 2015Available online xxx

Keywords:Brand engagementeWOMSeOeRUser-generated content

a b s t r a c t

This study examined the influences of positive brand-related user-generated content (UGC)1 shared viaFacebook on consumer response. The model tested was derived from the SeOeR consumer responsemodel (Mehrabian & Russell, 1974) that depicts the effects of environmental/informational stimuli onconsumer response. Specific research objectives were to investigate whether brand-related UGC acts as astimulus to activate consumer behavior in relation to brand and examine the processes by which brand-related UGC influences consumer behavior. Using the SeOeR model, brand-related UGC was treated asstimulus, pleasure and arousal as emotional responses, and perceived information quality as cognitiveresponse. Information pass-along, impulse buying, future-purchase intention, and brand engagementwere treated as behavioral responses. Participants (n ¼ 533) resided in the U.S. and had a Facebookaccount. Mock Facebook fan pages including brand-related UGC were developed as visual stimuli andpresented via an online self-administered questionnaire. SEM was used to analyze the data. Brand-related UGC activated consumers' emotional and cognitive responses. Emotional and cognitive re-sponses significantly influenced behavioral responses. Positive brand-related UGC exerts a significantinfluence on brand as it provokes consumers’ eWOM behavior, brand engagement, and potential brandsales.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Social media encompasses a broad range of online venues thatfacilitate interaction, collaboration, and the sharing of contentamong users (Tuten, 2008). Within retail environments, socialmedia accelerates the accessibility of brand content to consumers(Lipsman, Mudd, Rich, & Bruich, 2012). Social media empowersconsumers to share their views and exert their individual andcollective influence on other consumers as well as on brands.Because social media enables consumers to actively gather infor-mation and share opinions, consumers are no longer passive re-cipients of product information but active generators anddistributors of such information (Stewart& Pavlou, 2002) in a rangeof forms (e.g., videos, text, audio). Thus, consumers are able to in-fluence other consumers’ consumption activities on a level notpreviously seen (Accenture, 2013).

One of the ways consumer interaction happens via social mediais through user-generated content (UGC).1 UGC refers to mediacontent created by members of the general public and includes anyform of online content created, initiated, circulated, and consumedby users (Daugherty, Eastin, & Bright, 2008). UGC often includesbrand-related subject matter (Smith, Fischer, & Yongjian, 2012)driving product awareness and influencing consumers’ purchasedecisions (Blakley, 2013). Electronically delivered statements abouta product, service, or brand made by potential, actual, or formercustomers are called electronic word-of-mouth or eWOM (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). Although, UGC isbroader in its scope than eWOM, UGC and eWOM are often usedinterchangeably when UGC is brand-related (Smith et al., 2012).

Brand-related UGC shared via social media may have more in-fluence than other sources because it is transmitted by a trust-worthy information source embedded in a consumer's personalnetwork (Chu & Kim, 2011; Corrigan, 2013). In addition, the

* Corresponding author.E-mail addresses: [email protected] (A.J. Kim), [email protected] (K.K.P. Johnson). 1 UGC ¼ user-generated content.

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier .com/locate/comphumbeh

http://dx.doi.org/10.1016/j.chb.2015.12.0470747-5632/© 2015 Elsevier Ltd. All rights reserved.

Computers in Human Behavior 58 (2016) 98e108

influence of eWOM on social media may be greater than traditionalword-of-mouth (WOM) because eWOM messages can easily andquickly reach global audiences who share similar interests in aproduct or brand (Christodoulides, Michaelidou, & Argyriou, 2012).Therefore, this study investigated the influences of brand-relatedUGC on consumer decision-making within the context of Face-book. Facebook was selected because it is a dominant social mediaplatform that offers a range of features to enable brand-related UGCto be published and shared. Studying the influence of brand-relatedUGC is essential because UGC is a significant method of consumerinfluence within the marketplace (Riegner, 2007) and a challengeto retailers as they are limited in their ability to control it.

2. Literature review

2.1. Brand-related UGC via social media

Whether consumers share information about brands or prod-ucts in the form of online reviews or talk about their experiencewith brands or products on personal SNSs, brand-related UGCappearing in social media function as eWOM messages. Althoughnot specifically examined in Facebook context, previous researchershave explored behavioral consequences of eWOM in various con-texts (i.e., consumer review sites, brand websites, personal blogs).Researchers found that eWOM influences attitude toward andpurchase intention of a featured product (Christodoulides et al.,2012; Lee & Youn, 2009) along with willingness to recommend aproduct (Sun, Youn, Wu, & Kuntaraporn, 2006). Characteristics ofmessage content such as message valence (i.e., positive, negative)was found to influence the impact of eWOM on product attitudesuch that extremely positive reviews as well as moderately nega-tive reviews strengthened attitudes (Lee, Rodgers, & Kim, 2009)whereas negative reviews increased unfavorable product attitudes(Lee, Park, & Han, 2008).

2.2. Related research and hypotheses development

2.2.1. Stimulus-organism-response frameworkThe influence of brand-related UGC on consumer's attitudes and

behavior can be explained as the influence of stimulus on an or-ganism and the corresponding response according to the stimulus(S) e organism (O) e response (R) model (Mehrabian & Russell,1974). The relationship between these constructs is linear withorganism acting as mediator between stimulus and response(Eroglu, Machleit, & Davis, 2003; Kihlstrom, 1987). Applied toconsumer behavior research, SeOeR can be utilized as a structureto demonstrate the effect of external influences on consumers (S),the internal processes responding to that influence (O), and theresulting behaviors (R). The external influences (e.g., informationalinputs) can include managerially controllable factors such asadvertising, price, product design, or non-controlled environ-mental factors such as competition, social pressure, and economicconditions (Bagozzi, 1983). Representing internal processes, vari-ables for internal responses can include emotional responses suchas arousal, fear, and liking as well as cognitive responses such asperceived risk, dissonance, and expectations (Bagozzi, 1983). Last,intention to act, activities leading to choice, actual choices, out-comes, and reactions to choice can be included to representbehavioral responses (Bagozzi, 1983). In consumer behaviorresearch, conceptualizing individuals' reactions to environmentalinformation using the SeOeR framework presents an opportunityto capture elements of the complex process of consumer responseand decision-making (Bagozzi, 1983).

Since the SeOeR model was introduced, it has been used inconsumer research to understand consumer responses to various

consumption contexts. Initial work focused on variables evident intraditional brick and mortar stores including music, lighting, color,and scent (Chebat&Michon, 2003; Yalch & Spangenberg, 2000) ona range of consumer intentions and behaviors, including patronageintention (Babin, Hardesty, & Suter, 2003; Baker, Levy, & Grewal,1992; Wu et al., 2013), buying intention (Babin et al., 2003;Bellizzi & Hite, 1992), unplanned purchasing (Donovan, Rossiter,Marcoolyn, & Nesdale, 1994), and time spent in the store(Donovan & Rossiter, 1982; Donovan et al., 1994). Additionally, themodel has been adopted to investigate the impact of advertising.Olney, Holbrook, and Batra (1991) examined the impact of adcontent on viewing time (i.e., behavioral response) and foundpleasure, arousal, and attitude toward the ad mediated the rela-tionship between ad content and viewing time.

Researchers have also demonstrated the applicability of theSeOeR framework to an online shopping context examining howatmospheric attributes of online retail websites (i.e., product pre-sentation, design quality, music) impact consumer responsesincluding emotions and shopping intentions (Koo & Ju, 2010), ser-vice quality and satisfaction (Eroglu et al., 2003; Ha & Im, 2012;Wang, Hernandez, & Minor, 2010), arousal, purchase, and inten-tion to revisit the site and repurchase (Peng & Kim, 2014: Wang,Minor, & Wei, 2011), website patronage intention (Eroglu,Machleit, & Davis, 2001; Jeong, Fiore, Niehm, & Lorenz, 2009),amount of money and time spent (Eroglu et al., 2001), and intentionto engage in WOM activities (Ha & Im, 2012).

2.2.2. Application of the SeOeR modelThe SeOeR framework was employed because the constructs

included and the relationships among them illustrate the core in-terests of this research. It was proposed that brand-related UGC (S)evoked emotional and cognitive responses (O) within consumers,and these internal states influenced consumer's behavioral re-sponses (R). That is, when a consumer encounters brand-relatedUGC on Facebook, UGC (S) activates internal information process-ing by the consumer (O) and consequently behavioral actions (R)related to the brand as the consumer processes the information.

Researchers utilizing the SeOeR model in online settings havefound purchase intention (Eroglu et al., 2003; Koo & Ju, 2010; Peng& Kim, 2014; Wang et al., 2010, 2011) and intention to engage inWOM activities (Ha & Im, 2012) to be linked to UGC. Thus, thesewere deemed possible brand-related outcomes of consumer ac-tivities that could be tied to exposure to UGC in a social mediacontext and included as possible behavioral outcomes. In additionto the foundational constructs of the SeOeR model (i.e., informa-tional stimuli, emotional responses, cognitive responses, approachand avoidance behaviors) examined previously, our model includesa relationship-building variable (i.e., brand engagement) as acomponent of behavioral responses because important goals forcompanies engaged in social media activities are to increase brandawareness and to build and enhance relationships with existingand new customers (Hennig-Thurau et al., 2004; Kim & Ko, 2012).Brand-related content can trigger new customers’ interest in thebrand and motivate learning about brand activities that couldpotentially impact brand relationships. Also reflecting the instan-taneous aspect of online setting, our proposed model dividesbehavioral responses into immediate responses (i.e., informationpass-along via SNS platform, impulse purchase via e-commercewebsites) and latent responses (i.e., future-purchase intention,brand engagement) so the model can be applied to examine theinfluence of informational stimuli within an online specific context.

Within the SeOeR framework consumer responses may followdifferent response sequences (i.e., cognitive, affective, parallel)stemming from the reactions evoked within the organism afterexposure to a stimulus (Bagozzi, 1983). The cognitive response

A.J. Kim, K.K.P. Johnson / Computers in Human Behavior 58 (2016) 98e108 99

model holds that cognitions occur before affect (Bagozzi, 1983).Fiore and Kim (2007) noted cognitive response components withina SeOeR framework included beliefs, thoughts, and perceptionsconstructed via direct interaction with the stimulus or the pro-cessing of secondary sources of information (e.g., advertisements,WOM). When a consumer is exposed to stimuli such as eWOM thatprovides information (e.g., facts about a product, usage instruction),cognitive processes must occur to understand the factual contentpresented. After understanding the stimuli, the consumer mightdevelop an affective or emotional response toward the information.

In contrast, in the affective responsemodel a stimulus evokes anaffective state (e.g., joy, fear) prior to any cognitive response(Bagozzi, 1983). When a consumer is exposed to eWOM with acompelling positive message (e.g., best vacation ever), or acompelling negative one (e.g., it was awful), the stimuli may induceinitial emotional reactions. Cognitive activities that are believed tofollow are limited to finding ways to obtain the product andlocating the resources to do so (Bagozzi, 1983).

Finally, in the parallel response model, both cognitive and af-fective responses occur simultaneously. In this case, the affectivecontent within the stimulus is salient enough to evoke a responsebut not so high that it overwhelms the cognitive information(Bagozzi, 1983). When both cognitive and affective responses areevoked, each has independent effects on behavioral response.Brand-related UGC on Facebook often contains evaluation of con-sumption experiences, factual information, and affect related to abrand (Chen, Chen, Chen, Chen, & Yu, 2013; Smith et al., 2012). Asbrand-related UGC shared via Facebook contains both informa-tional and emotional messages, the sequence of responses antici-pated was exemplified by the parallel response sequence (see Fig. 1for the model tested).

2.2.3. Internal responses to brand-related UGC: SeO relationshipsUsing the SeOeR model, researchers have studied the effect of

website stimuli on emotional and cognitive responses. These re-searchers have found that website design elements (i.e., colors,images, interactive features) had a positive influence on consumers’emotions (e.g., pleasure) and cognitions such as perceived

information quality (Eroglu et al., 2003; Ha & Im, 2012; Park, Stoel,& Lennon, 2008). In addition, informational stimuli such asadvertising content was also found to influence pleasure as well asarousal (Olney et al., 1991). Because brand-related UGC (S) oftencontains information about brands, consumers who encounter itare expected to process the information (O) and determine itsquality. Thus, the following hypotheses were proposed.

Hypothesis 1. Brand-related UGC influences consumers’emotional response (a. pleasure, b. arousal).

Hypothesis 2. Brand-related UGC influences consumers’ cogni-tive response (i.e., perceived information quality).

2.2.4. Behavioral responses to brand-related UGC: O-R relationships

2.2.4.1. Information pass-along. The unique features of Facebookfacilitate users sharing informationwith others by clicking “like” onposts made by other users, or “share” by posting information on afriend's wall (Chen et al., 2013). Passing along brand-related con-tent on Facebook is eWOM activity, as the content conveys opin-ions, facts, or user experiences with brands or products. Examiningthe emotional and the cognitive aspects of WOM, Ladhari (2007)found that pleasure and arousal (O) had significant influences onWOM intention (R). Kim and Niehm (2009) found perceived in-formation quality of a website (O) positively influenced recom-mendation intention (R). Ha and Im (2012) examined the influenceof website design on WOM activities. They found that perceivedquality of information as well as participant affect (O) was signifi-cantly related toWOM intention (R). Therefore, it was hypothesizedthat:

Hypothesis 3. Emotional response (a. pleasure, b. arousal) posi-tively influences information pass-along.

Hypothesis 4. Cognitive response positively influences informa-tion pass-along.

2.2.4.2. Impulse buying. Impulse buying refers to making a pur-chase based on a sudden urge to buy something immediately

Fig. 1. Model of hypothesized relationships.

A.J. Kim, K.K.P. Johnson / Computers in Human Behavior 58 (2016) 98e108100

(Adelaar, Chang, Lancendorfer, Lee, & Morimoto, 2003). Impulsebuying is different from purchase intention because decision-making time is very short and the purchase is unreflective(Weun, Jones, & Beatty, 1998). Conceptualizing impulsive buyingbehavior, Stern (1962) identified suggestion impulse buying as oneof four impulse buying categories (i.e., pure, reminder, suggestion,planned). According to Stern (1962), suggestion impulse buyingoccurs when a consumer sees a product for the first time and vi-sualizes a need for it. Applying this category of impulse buying to anexperience of encountering brand-related UGC, a consumerexposed to such contentmay also see the product featured, imaginewhat it would be like to use the product, and consequently want topurchase it. Due to the development of e-commerce, consumerscan easily act upon their impulses and immediately purchasefeatured products. In fact, features on Facebook fan pages allowconsumers to make instant purchases by providing links to onlinestores. Thus, UGC could motivate impulse purchases.

In addition, Adelaar et al. (2003) found that emotional responsewas positively related to impulse buying because sensory stimuli(i.e., audio, text, pictures) reduced self-control mechanisms. Thus, apositive emotional response (O) activated by brand-related UGC (S)could motivate impulse buying intention (R) because positiveemotions evoke approach behaviors (Baker et al., 1992). Therefore,the following hypotheses were developed.

Hypothesis 5. Emotional response (a. pleasure, b. arousal) posi-tively influences impulse buying.

Hypothesis 6. Cognitive response positively influences impulsebuying.

2.2.4.3. Future-purchase intention. In general, purchase intentionrefers to a future plan to buy a particular product or service(Adelaar et al., 2003). Within the SeOeR framework, purchaseintention represents an intention to act favorably (i.e., approachbehavior) in response to informational stimuli related to brands orproducts. Regarding relationships between emotional response andpurchase intention, researchers have demonstrated that intentionto purchase follows after positive internal states. Examining theinfluence of retail store environment on consumer response, Bakeret al. (1992) found that participants’ willingness to purchase wasenhanced as pleasure and arousal increased. Although previousresearchers did not examine the influence of eWOM on purchaseintention via cognitive response, findings on the consequences ofeWOM have shown that eWOM influences purchase intention(Christodoulides et al., 2012; Yu & Natalia, 2013).

Hung and Li (2007) suggested one possible explanation for therelationship between eWOM and purchase intention is that eWOMprovides opportunities for consumers to gain knowledge aboutbrands and to store the information into their consideration set. Aconsumer's consideration set contains all the brands a consumercan think of when making a purchase. Any brands included in aconsumer's consideration set may be recalled and purchased in thefuture. The process of comprehending and storing eWOMmessagesrepresents consumer's cognitive mental activity in response toeWOM. These findings led to the prediction that emotional andcognitive responses (O) stemming from brand-related UGC (S) ispositively related to future-purchase intention (R). Thus, thefollowing hypotheses were developed.

Hypothesis 7. Emotional response (a. pleasure, b. arousal) posi-tively influences future-purchase intention.

Hypothesis 8. Cognitive response positively influences future-purchase intention.

2.2.4.4. Brand engagement. Previous researchers have documentedthat UGC such as eWOM influences customer relationships withbrands. According to Kim and Ko (2012), brand-related content onFacebook influences relationship equity, that is, the tendency ofconsumers to stay in a relationship with a brand. Brand engage-ment is a key component in building relationships between brandsand customers (Keller, 2001). Brand engagement describes theemotional tie that connects customer to brands (Goldsmith, 2012).A brand engaged customer shows willingness to be involved with abrand and gather information about the brand, talk about it, andexhibit its use to others (Keller, 2001). Brand engagement (R) is aconsequence of emotional and cognitive states (O) evoked by thebrand (Allen, Fournier, & Miller, 2008; Goldsmith, 2012) and thatcould be evoked by brand-related UGC (S). Consequently, it washypothesized that:

Hypothesis 9. Emotional response (a. pleasure, b. arousal) posi-tively influences brand engagement.

Hypothesis 10. Cognitive response positively influences brandengagement.

3. Method

3.1. Visual stimuli

The design of this study was an online survey. Visual stimulisimulating Facebook fan pages were developed and included as apart of the self-administered questionnaire. A pretest was con-ducted to identify fashion brands with a Facebook fan page thatoffered unisex fashion products consumed by all ages, so the brandwould have some appeal regardless of participants' gender or age.The context for this study was consumers’ encountering brand-related UGC during casual Facebook browsing. Casual Facebookbrowsing is an exploration of Facebook without a planned objectiveor search strategy.

Brand awareness was included as a control variable because pre-existing knowledge and attitudes toward a brand can affect brand-related decision-making (Aaker, 1996). In order to control anypossible influence of brand awareness on the hypothesized re-lationships, two different brands were used. Sperry Top-Sider wasselected to represent a fashion brand with high brand awarenessand Sebago Docksides, a competing brand, was identified torepresent a brand with low brand awareness. Pilot testing revealedthe brand awareness level between Sperry Top-Sider and SebagoDocksides was significantly different (Sperry Top-Sider m ¼ 5.30,Sebago Docksides m ¼ 1.48; t ¼ 13.95, df ¼ 112, p ¼ .001).

Six postings were created as brand-related UGC to be includedin each brand's mock Facebook fan page. Brand-related UGCappearing on existing Facebook fan pages were reviewed carefullyand selected posts were modified to include product descriptions,product usage, recommendations, and product reviews (see Fig. 2for an example). The postings appearing in the mock fan pagescontained both informational and emotional messages and somepostings were accompanied with related-product images. Allmessages were positive in terms of valence. Mock fan pages pre-sented both informational and emotional messages together in onepage for the purpose of face validity of the stimuli.

3.2. Measurements

The questionnaire beganwith questions concerning participantsFacebook activity (i.e., frequency of Facebook use, reasons forFacebook use, fashion brands following, brand-related UGC post-ing). In the following sections, participants viewed one of the two


Fig. 2. Visual stimulus: Sperry Top-Sider Facebook fan page.


stimuli, responded to the measures of the investigated variables,and responded to demographic questions.

3.2.1. Brand-related UGCQuestions initially developed to measure eWOM message ap-

peals (Wu&Wang, 2011) served as the foundation for developing ameasure of brand-related UGC content. The original measure con-tained two items measuring rational and emotional eWOM mes-sage appeals (i.e., the message mainly describes the productfunction, benefit, and value; the message mainly spreads a certainatmosphere, emotion, and feeling). These two items were modifiedto develop six items to assess the informational and the emotionalbrand-related UGC content (see Table 1 for the measurementitems). Participants responded to each of these items using seven-point Likert scales (1 ¼ strongly disagree; 7 ¼ strongly agree).

3.2.2. Emotional responses: pleasure and arousalParticipants’ emotional responses were assessed using

measures of pleasure and arousal developed by Mehrabian andRussell (1974). Early researchers in environmental psychologyfocused on pleasure, arousal, and dominance (PAD) as theemotional responses to environmental stimuli. However, since theinitial development of the model, in numerous instances, only thepleasure and arousal variables have been included as emotionalresponse because the dominance dimension received limitedempirical support (Donovan & Rossiter, 1982). All items measuringpleasure and arousal were presented on seven-point semanticdifferential scales. Other researchers who have used these mea-sures reported the pleasure scale had a reliability of a ¼ .85 and thearousal scale had a reliability of a ¼ .80 (Eroglu et al., 2003).

3.2.3. Cognitive response: perceived information qualityPerceived information quality was measured using a scale

developed by Yang, Cai, Zhou, and Zhou (2005). The items used toassess information quality were: relevant information to thecustomer, up-to-date information, valuable tips on products, and

Table 1Summary of items comprising final measures.

Measures Number offinal items

Factor loading(min. e max.)

Cronbach'sa

UGC message 6 .70e.87 .92RM1: The postings that appear on the Facebook fan page describe functions of the featured brand and product.RM2: The postings that appear on the Facebook fan page describe values of the featured brand and product.RM3: The postings that appear on the Facebook fan page describe benefits of the featured brand and product.EM1: The postings that appear on the Facebook fan page create a positive atmosphere about the featured brand and product.EM2: The postings that appear on the Facebook fan page create positive emotions about the featured brand and product.EM3: The postings that appear on the Facebook fan page create positive feelings about the featured brand and product.

Pleasure 6 .80e.88 .92PL1: Unhappy-happyPL2: Annoyed-pleasedPL3: Dissatisfied-satisfiedPL4: Melancholic-contentedPL5: Despairing-hopefulPL6: Bored-relaxed

Arousal 2 .93e.93 .84*AR1: Sluggish-frenzied*AR2: Dull-jittery*AR3: Unaroused-arousedAR4: Relaxed-stimulatedAR5: Calm-excited*AR6: Sleepy-wide awake

Information quality 3 .85e.87 .81IQ1: The information contained in the postings is relevant.IQ2: The information contained in the postings is up-to-date.IQ3: The information contained in the postings provides valuable tips on the featured brand and products.*IQ4: The information contained in the postings is unique.

Information pass-along 4 .91e.95 .95*PA1: I would click “like” on the some of the postings.PA2: I would share the postings on my own timeline.PA3: I would share the postings on a friend's timeline.PA4: I would pass along the postings to contacts on my Facebook friends list.PA5: I would pass on the information along using other forms of social media.

Impulse buying 2 .95e.95 .89IB1: I will visit the brand's online store to purchase the product appear on this fan page right away.IB2: I intend to purchase the product featured on this fan page immediately.

Future-purchase intention 3 .93e.96 .94*FP1: The likelihood of purchasing the product featured on the fan page is high.*FP2: If I were going to buy the style of shoes featured on this fan page, I would consider buying them from this brand.FP3: I would consider buying the product featured on this fan page.FP4: The probability that I would consider buying products from this brand is high.FP5: My willingness to buy the product featured on this fan page is high.

Brand engagement 5 .89e.92 .94BE1: I would like to talk about this brand with others.BE2: I am interested in learning more about this brand.BE3: I would be interested in other products offered by this brand.BE4: I would be proud to have others know that I use this brand.BE5: I like to visit the website for this brand.*BE6: I would closely follow news about this brand.

* Dropped items.


unique content. Participants indicated their degree of agreementwith each item using seven-point Likert scales (1 ¼ stronglydisagree; 7 ¼ strongly agree). The measure had a reported reli-ability of a ¼ .84.

3.2.4. Behavioral responses: information pass-along, impulsebuying, future-purchase intention, and brand engagement

Items to measure information pass-along on Facebook weredeveloped by modifying existing measures of eWOM used by Chuand Choi (2011) to reflect a general SNS context. Five itemsmeasuring eWOM activities were reworded to include specifictypes of pass-along features provided by the Facebook platform(see Table 1 for the measurement items). Impulse buying wasmeasured using two items developed by Adelaar et al. (2003). Theoriginal items were revised to reflect the context of this study. Theinitial reported reliability of the measurewas a¼ .59 (Adelaar et al.,2003). Future-purchase intentionwasmeasured by adopting a scaleused by Dodds, Monroe, and Grewal (1991). Future-purchaseintention assesses the possibility and likelihood that a consumerwill purchase a certain product or brand in the future. Reportedreliability of the scale items ranged from a ¼ .93 to a ¼ .96 (Doddset al., 1991). Brand engagement was measured by adopting a scaledeveloped by Keller (2001) that assesses the level of an individual'sbrand-related emotional, cognitive, and behavioral activities. Par-ticipants responded to each of these measurement items usingseven-point Likert scales (1 ¼ strongly disagree; 7 ¼ stronglyagree).

3.2.5. Control variable: brand awarenessA control variable, brand awareness, was assessed by four items

adopted from Aaker (1996). These items were: I have heard of thisbrand, I know what this brand stands for, I have an opinion aboutthe brand, and this brand is likely to be one of the fashion brands Ican recall. Items were presented on seven-point Likert scales (1 ¼strongly disagree; 7 ¼ strongly agree).

3.3. Sampling and data collection

Individuals who were 18 years old and older with a Facebookaccount were recruited from online panel members obtained froma marketing research company specializing in consumer surveys(i.e., Survey Sampling International). An e-mail invitation thatprovided a link to access the consent form and the research ques-tionnaire posted online was sent to potential participants. Potentialparticipants were randomly assigned to one of the two online URLlinks (i.e., Sperry Top-Sider fan page, Sebago Docksides fan page).Data collection was completed in three business days and onaverage, the questionnaire took 7 minutes for participants tocomplete.

Participants who completed the questionnaire were identifiedby a unique ID generated by the online questionnaire system andwere compensated for their participation through a point systemwhereby points were credited to their accounts. Points credited toparticipants’ accounts can be redeemed for cash (e.g., via PayPal)and gift cards (e.g., Amazon.com gift card).

4. Results

4.1. Participant characteristics

The purposive sample consisted of 533 participants. Threehundred and forty were women (63.8%) and 193 were men (36.2%).The age of more than half of the participants (60.1%) fell between 18and 34 years old, a percentage that is very similar to the age esti-mates of Facebook users (Wildrich, 2013). Concerning participants'

Facebook activity, the majority were frequent users of Facebook,45% visited Facebook multiple times a day, and 28.9% continuouslyused Facebook. Most of the participants (90.4%) indicated that theyused Facebook to keep in touch with people and more than halfindicated they used Facebook to get information (59.7%). About halfof the participants (46.9%) were following fashion brands onFacebook with 39.0% of these individuals following from 1 to 10fashion brands. In regards to brand-related UGC posting activities,participants (36.6%) indicated that they posted brand-related UGCon Facebook. Participants (22.5%) posted brand-related UGCequally on brand fan pages and timelines. Participants’ most oftencontributed brand-related UGC was “liking” a brand (24.2%), fol-lowed by writing comments (21.4%), and posting photos (20.8%).

4.2. Preliminary data analyses

Prior to checking assumptions for the main data analyses, in-dependent sample t-tests were conducted on 40 items comprisingthe original measurement set assessing the constructs underinvestigation to see whether differences existed between the twosample brands (i.e., Sperry Top-Sider, Sebago Docksides). The re-sults of independent t-tests showed that responses to the mea-surement items were not significantly different across the twobrands. Therefore, the two data sets were combined to test thehypothesized model.

4.2.1. UnidimensionalityThree different analyses were conducted for measurement pu-

rification. First, skewness and kurtosis values of all scale items wereexamined and all values were in acceptable ranges as identified byHair, Anderson, Tatham, and Black (1998). Second, exploratoryfactor analysis with varimax rotationwas conducted on the originalitem set. Factor loadings exceeding .50 with eigenvalues greaterthan 1.0 were considered evidence for construct validity (Hair,Black, Babin, Anderson, & Tatham, 2006). This process resulted inthe elimination of two arousal items as they demonstrated lowfactor loadings. As a final step, confirmatory factor analysis wasconducted using maximum likelihood estimation on the item cor-relation matrices. The magnitude of item error variances, largemodification indices (MI), and standardized residuals values werecarefully examined. In the process, an additional seven items wereremoved. As a result of these preliminary analyses, 31 of the orig-inal 40 items were retained for hypotheses testing. The final mea-sures used for the data analysis are organized by construct inTable 1.

4.2.2. Measurement modelTo test the measurement model, confirmatory factor analysis

(CFA) with maximum likelihood was conducted on the 31 in-dicators of the eight latent variables for the measurement model.The results of CFA indicated that the measurement model hadacceptable construct validity. The model exhibited an excellent fitwith the data: c2 ¼ 1144.90 with 406 df, c2/df ¼ 2.82, p ¼ .001;comparative fit index (CFI) ¼ .95; non-normed fit index(NNFI) ¼ .95; standardized root mean square residual(SRMR) ¼ .04; root mean square error of approximation(RMSEA) ¼ .05. In general, CFI, NNFI values of .95 or higher andRMSEA and SRMR of .06 or lower indicate a satisfactory model fit(Hu & Bentler, 1999). All the factor loadings on their respectedconstructs were higher than .70. Internal consistency was testedusing Cronbah's a and all latent variables had Cronbah's a over .70,indicating adequate construct reliability. Convergent validity wassupported by all the factor loadings being significant (p < .001) andthe composite reliability for each construct greater than .81,exceeding the recommended level of .60 (Hair et al., 1998).


Variances extracted were also computed to assess the variances inthe indicators accounted for by the latent construct. The averagevariance extracted (AVE) for each construct was greater than .59,fulfilling the recommended benchmark of .50 (Hair et al., 1998).Discriminant validity was assessed by performing chi-square dif-ference tests between an unconstrained model estimating thecorrelation between a pair of constructs and a constrained modelwith the correlation between that pair of constructs fixed to 1.0. Atotal of 28 rival models fixing the correlation between each pair ofconstructs to unity were compared to the measurement model onepair at a time. Discriminant validity of the measures was achievedas the results of chi-square difference tests all showed significantdifferences (p ¼ .001) indicating the two constructs in a pair aresignificantly different constructs (Ping, 1994). Descriptive statisticsfor the research variables are summarized in Table 2.

4.3. Structural model evaluation and hypotheses testing

The hypothesized relationships in the structural equationmodelwere examined by the maximum likelihood estimation method.The structural model exhibited a good fit with the data(c2 ¼ 1182.26 with 413 df, c2/df ¼ 2.86, p ¼ .001, CFI ¼ .95,NNFI ¼ .94, SRMR ¼ .05, and RMSEA ¼ .05). All paths related toHypotheses 1 through 10 were statistically significant exceptHypothesis 1-b. Brand-related UGC message positively influencedpleasure (b ¼ .60, t ¼ 13.31, p < .001) and perceived informationquality (b¼ .89, t¼ 18.18, p < .001), supporting Hypotheses 1-a and2. No significant relationship between brand-related UGC andarousal was found. Therefore, Hypothesis 1-b was rejected. Therewere positive and significant influences of pleasure (b ¼ .31,t ¼ 5.91, p < .001) and arousal (b ¼ .73, t ¼ 2.71, p < .01) on pass-along intention. Perceived information quality also positivelyinfluenced pass-along intention (b ¼ .26, t ¼ 2.30, p < .05) sup-porting Hypotheses 3-a, 3-b, and 4. These findings are consistentwith previous researchers (Ha & Im, 2012; Kim & Niehm, 2009;Ladhari, 2007) who also found that emotional responses (i.e.,pleasure, arousal) and perceived information quality were signifi-cant factors influencing WOM intention. As hypothesized, pleasure(b ¼ .38, t ¼ 7.27, p < .001), arousal (b ¼ .79, t ¼ 2.72, p < .01), andperceived information quality (b ¼ .33, t ¼ 2.70, p < .01) positivelyinfluenced impulse buying intention, supporting Hypotheses 5-a,5-b, and 6. Participants who responded to the brand-related UGCwith positive emotions and perceived the information to be usefulindicated they were more likely to purchase the product featuredon the fan page immediately by visiting an online store. The in-fluences of emotional and cognitive responses on future-purchaseintention were significant as well. Pleasure (b ¼ .30, t ¼ 26.68,

p < .001), arousal (b ¼ .70, t ¼ 2.71, p < .01), and perceived infor-mation quality (b ¼ .55, t ¼ 4.95, p < .001) positively influencedfuture-purchase intention, supporting Hypotheses 7-a, 7-b, and 8.Hypotheses 9 and 10 predicted that emotional and cognitive re-sponses positively influence brand engagement. Pleasure (b ¼ .30,t ¼ 6.68, p < .001), arousal (b ¼ .70, t ¼ 2.72, p < .01), and perceivedinformation quality (b¼ .60, t¼ 5.33, p< .001) positively influencedbrand engagement, supporting Hypotheses 9-a, 9-b, and 10. Par-ticipants who responded to the brand-related UGC with pleasureand perceived the information to be useful indicated they werelikely to associate with the brand. This finding provided empiricalsupport for the conceptual definition of brand engagement, that is,brand engagement is a result of emotional and cognitive responsesstimulated by the brand (Allen et al., 2008; Goldsmith, 2012). Fig. 3illustrates the final model and provides a visual summary of theresults of hypotheses testing.

4.4. Mediating effect of organism between stimulus and response

In order to verify the mediating effect of organism in the SeOeRmodel, a mediation test was conducted following Baron and Kenny(1986). To establish mediation, the following conditions weretested via regression analyses: (1) brand-related UGC affectsbehavioral responses; (2) brand-related UGC affects emotional andcognitive responses; (3) emotional and cognitive responses affectbehavioral responses; (4) the effects of brand-related UGC onbehavioral responses declines or disappears when the effects ofemotional and cognitive responses are statistically controlled for inexplaining behavioral responses (via multiple regression withbrand-related UGC and emotional and cognitive responses explainbehavioral responses). Arousal was excluded from the mediationtest since there was no significant relationship between brand-related UGC and arousal.

In the four-step approach testing for mediation (Baron & Kenny,1986), mediation is supported if the effect of mediating variableremains significant after controlling for independent variable in thestep four. Full mediation can be inferred if independent variable isno longer significant when the mediator is controlled, and partialmediation can be inferred if independent variable is still significant.Both emotional and cognitive responses (i.e., pleasure, perceivedinformation quality respectively) mediated relationships betweenbrand-related UGC and all of the behavioral responses examined(see Table 3).

5. Discussion and implications

Amodel examining the influences of positive brand-related UGC

Table 2Descriptive statistics of the research variables.

Construct 1 2 3 4 5 6 7 8

1. UGC Message 1.002. Pleasure .53 1.003. Arousal .15 .25 1.004. Information Quality .75 .56 .19 1.005. Information Pass-along .25 .39 .28 .40 1.006. Impulse Buying .33 .46 .26 .45 .78 1.007. Future-purchase Intention .49 .54 .21 .60 .66 .78 1.008. Brand Engagement .49 .54 .24 .62 .72 .78 .89 1.00Mean 5.29 5.18 4.10 5.13 3.83 3.94 4.43 4.44SD .93 1.05 1.498 .98 1.64 1.53 1.47 1.42Composite Reliabilitya .92 .92 .92 .81 .95 .88 .94 .94Variance Extractedb .66 .66 .86 .59 .82 .79 .84 .77

a Composite Reliability ¼ (S standardized loading)2/(S standardized loading)2 þ S measurement error.b Variance Extracted ¼ S(standardized loading)2/S (standardized loading)2 þ S measurement error.


shared via Facebook on consumer response was tested. Brand-related UGC including information and emotional content was

positively related to pleasure and perceived information quality.Brand-related UGC acted as informational stimuli to activate

Fig. 3. Results of testing the conceptual model of consumer response to brand-related UGC.

Table 3Results of mediation test.

Step Independent Variable Dependent Variable b R2 Mediation

1 UGC message Information pass-along .30*** .092 UGC message Pleasure .53*** .28

UGC message Information quality .77*** .603 Pleasure Information pass-along .43*** .19

Information quality Information pass-along .50*** .254 UGC message (a)

Pleasure (b)Information pass-along .09 (a) .23 Full

.38*** (b)UGC message (a)Information quality (b)

Information pass-along #.24***(a) .28 Partial.69***(b)

1 UGC message Impulse buying .33*** .112 UGC message Pleasure .53*** .28

UGC message Information quality .77*** .603 Pleasure Impulse buying .47*** .22

Information quality Impulse buying .51*** .264 UGC message (a)

Pleasure (b)Impulse buying .12** (a) .23 Partial


Impulse buying #.15* (a) .25 Partial.62*** (b)

1 UGC message Future-purchase intention .54*** .292 UGC message Pleasure .53*** .28

UGC message Information quality .77*** .603 Pleasure Future-purchase intention .56*** .31

Information quality Future-purchase intention .65*** .434 UGC message (a)

Pleasure (b)Future-purchase intention .33*** (a) .39 Partial


Future-purchase intention .07 (a) .43 Full.60*** (b)

1 UGC message Brand engagement .47*** .222 UGC message Pleasure .53*** .28

UGC message Information quality .77*** .603 Pleasure Brand engagement .54*** .29

Information quality Brand engagement .63*** .044 UGC message (a)

Pleasure (b)Brand engagement .26*** (a) .34 Partial


Brand engagement #.40 (a) .40 Full.66*** (b)

*p < .05, ***p < .001.


consumer's emotional and cognitive responses when participantsencountered brand-related UGC during Facebook browsing. Thesefindings confirm those of previous researchers (i.e., Eroglu et al.,2003; Ha & Im, 2012) who also documented that pleasure andperceived information quality are emotional and cognitive re-sponses to visual stimuli.

Contrastingly, findings revealed that arousal, another emotionalresponse component within the SeOeR framework, was notsignificantly influenced by brand-related UGC. This insignificantrelationship could have been influenced by the study context. Thisstudy was concerned with consumer response to brand-relatedUGC during Facebook browsing. In this situation, participantswere simply asked to review the postings included in Facebook fanpages and respond to questionnaire assessing their responses to theprovided brand-related UGC. In this case, it is possible that par-ticipants were in a generally low involvement situation comparedto when they were asked to complete a task after reviewing brand-related UGC (e.g., evaluate product, make purchase) or activelysought information about a particular product. It is possible thatthe participants were not aroused by the stimulus because notenough attention was paid to the brand-related UGC. Perhapsbrand-related UGCwould influence arousal if the circumstancewasa high involvement situation.

Supporting the O-R relationship proposed in the SeOeRframework (Mehrabian & Russell, 1974), the results clearly showedthat emotional and cognitive responses examined (i.e., pleasure,arousal, perceived information quality) significantly influenced allbehavioral outcomes in relation to the brand identified. Thus, forthese participants, information pass-along, impulse buying, future-purchase intention, and brand engagement are behavioral re-sponses to brand-related UGC.

The study contributes to extending prior work in the area ofconsumer response to environmental/informational stimuli andconsumer behavior within a SNS context. As the proposed modeldemonstrated a significant influence of brand-related UGC onbrand engagement, the model suggests a place for relationshipbuilding variables (i.e., brand engagement) as possible behavioraloutcomes leading to brand choice in addition to the outcome var-iables identified by previous researchers (e.g., Bagozzi, 1983; Fiore& Kim, 2007). The new model may be used to explain the influ-ence of physical atmospheric factors or informational stimuli onbehavioral responses including both sales related and relationshipbuilding variables via emotional and cognitive reactions of con-sumers in various consumption contexts. The model holds greatpotential for further application in emerging areas of consumerbehaviors such as new shopping environments (e.g., video shop-ping, virtual shopping, augmented reality retailing).

The findings also provide practical implication for social mediamarketing practitioners by helping them to understand the con-sequences of consumers engaging in Facebook activities. Con-sumers’ behavioral responses to brand-related UGC were related tobrand sales (i.e., impulse buying, future-purchase intention), rela-tionship building (i.e., brand engagement), and eWOM (i.e., infor-mation pass-along). Thus, providing new venues for consumers toconnect and talk about brands on social media can contribute toincreases in brand sales and both initiate and sustain brand-customer relationships. However, consumers should be aware ofthe drawbacks of unintended impulse purchases while browsingFacebook because negative consequences of impulse buyinginclude financial problems, disappointment with the product, orfeelings of guilt (Park & Choi, 2013).

As brand-related UGC impacts important behavioral outcomes,it is clear brands need to monitor what consumers are saying aswell as if they are saying anything at all about them. MonitoringUGC provides opportunities for brand management that what

consumers are saying about brand and how does that compare towhat they are saying about the brand's competitors could provideinsight into the possible product/service development opportu-nities, important areas to focus on for reputation management, andcustomer-brand relationships building and management. If con-sumers are not talking about your brand, perhaps it is a sign thatyour brand is no longer relevant to them. Clearly, consumers havethe power to shape the opinions and behaviors of fellow consumersin the marketplace. And while brands cannot control the content ofbrand-related UGC, they can use it and respond to it as part of theiroverall strategy of brand management.

6. Limitations and recommendations for future research

This study has limitations that could be addressed in futureresearch. The proposed model was developed and tested under aspecific context, that is, the casual encounter of brand-related UGCon fashion brands’ Facebook fan pages. Thus, the findings cannot besimply generalized to other consumption contexts. Future researchcould be directed at different situational contexts (e.g., intentionalvisit to brand fan pages) as well as a range of product categories(e.g., electronics, automobiles, healthcare, entertainment) that varyin terms of product attributes (e.g., hedonic, utilitarian) so that themodel of consumer response to brand-related UGC can be validatedand modified as needed. Examining diverse situational contextsand/or product categories could provide useful implications forsocial media marketing strategies applicable to a broader range ofconsumer products and services.

This study focused on the influence of positive brand-relatedUGC. This is a limitation to the external validity of the studybecause in reality, brand-related UGC is positive as well as negativein terms of its valence. Examining the influence of negative brand-related UGC is important because the impact of negative eWOM isgreater than positive eWOM (Park & Lee, 2009), especially whenthe brand-related UGC can easily reach a great number of con-sumers all over the world. Thus, researchers may want to examinethe influence of negative eWOM in social media contexts as well asvarious combinations of positive and negative eWOM so theensuing results provide practical implications for companies uti-lizing social media as communication platforms.

Self-report measures were employed in this research and thereare several common problems linked to such measures includingwhether or not participants were truthful in their responses orwhether there was a response bias present in some participants’responses. While it is possible that participants were not beinghonest in their responses, the questions asked were not addressingsensitive topics so we assume participants would have little moti-vation to protect their self-presentations. Techniques were alsoemployed to counter typical problems with the use of self-reportmeasures. For example, several of the questions were reversed toprevent participants from responding in the same way (e.g., posi-tively) to all items. Recognition was given to the use of ordinalmeasures in data collection so as to not over-interpret the signifi-cance of findings. We also eliminated data from speeders (i.e., re-spondents who completed the questionnaire in an unrealisticallyshort amount of time) and straight liners (i.e., respondents whogive the same response to every item) for analyses.

References

Aaker, D. A. (1996). Measuring brand equity across products and markets. CaliforniaManagement Review, 38(3), 102e120.

Accenture. (2013). Retail technology vision 2013. Retrieved from http://www.accenture.com/SiteCollectionDocuments/PDF/Accenture-Retail-Technology-Vision-2013.PDF.

Adelaar, T., Chang, S., Lancendorfer, K. M., Lee, B., & Morimoto, M. (2003). Effects of


media formats on emotions and impulse buying intent. Journal of InformationTechnology, 18(4), 247e266.

Allen, C. T., Fournier, S., & Miller, F. (2008). Brands and their marketing makers. InC. P. Haughtvedt, P. M. Herr, & F. R. Kardes (Eds.), Handbook of consumer psy-chology (pp. 781e821). New York, NY: Psychology Press.

Babin, B., Hardesty, D. M., & Suter, T. A. (2003). Color and shopping intentions: theintervening effect of price fairness and perceived affect. Journal of BusinessResearch, 56(7), 541e551.

Bagozzi, R. P. (1983). A holistic methodology for modeling consumer response toinnovation. Operations Research, 31(1), 128e176.

Baker, J., Levy, M., & Grewal, M. (1992). An experimental approach to making retailstore environmental decisions. Journal of Retailing, 68(4), 445e460.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction insocial psychological research: conceptual, strategic and statistical consider-ations. Journal of Personality and Social Psychology, 51(6), 1173e1182.

Bellizzi, J., & Hite, R. (1992). Environmental color, consumer feelings, and purchaselikelihood. Psychology and Marketing, 9(5), 347e363.

Blakley, J. (2013, May 2). Brands as publishers and how it's changing marketing [Weblog post]. Retrieved from http://www.postano.com/blog/brands-as-publishers-and-how-its-changing-marketing.

Chebat, J. C., & Michon, R. (2003). Impact of ambient odors on mall shoppers'emotion, cognition, and spending. Journal of Business Research, 56(7), 529e539.

Chen, C. Y., Chen, T. H., Chen, Y. H., Chen, C. L., & Yu, S. E. (2013). The spatio-temporaldistribution of different types of messages and personality traits affecting theeWOM of Facebook. Natural Hazards, 65(3), 2077e2103.

Christodoulides, G., Michaelidou, N., & Argyriou, E. (2012). Cross-national differ-ences in e-WOM influence. European Journal of Marketing, 46(11/12),1689e1707.

Chu, S. C., & Choi, S. M. (2011). Electronic word-of-mouth in social networking sites:a cross-cultural study of the United States and China. Journal of Global Mar-keting, 24(3), 263e281.

Chu, S. C., & Kim, Y. (2011). Determinants of consumer engagement in electronicword-of-mouth (eWOM) in social networking sites. International Journal ofAdvertising, 30(1), 47e75.

Corrigan, J. (2013, July 19). The benefits of user-generated content [Web log post].Retrieved from http://raventools.com/blog/benefits-user-generated-content/.

Daugherty, T., Eastin, M. S., & Bright, L. (2008). Exploring consumer motivations forcreating user-generated content. Journal of Interactive Advertising, 8(2), 1e24.

Dodds, W. B., Monroe, K. B., & Grewal, D. (1991). Effects of price, brand, and storeinformation on buyers' product evaluations. Journal of Marketing Research,28(3), 307e319.

Donovan, R. J., & Rossiter, J. R. (1982). Store atmosphere: an environmental psy-chology approach. Journal of Retailing, 58(spring), 34e57.

Donovan, R. J., Rossiter, J. R., Marcoolyn, G., & Nesdale, A. (1994). Store atmosphereand purchasing behavior. Journal of Retailing, 70(3), 283e294.

Eroglu, S. A., Machleit, K. A., & Davis, L. M. (2001). Atmospheric qualities of onlineretailing: a conceptual model and implications. Journal of Business Research, 54,177e184.

Eroglu, S. A., Machleit, K. A., & Davis, L. M. (2003). Empirical testing of a model ofonline store atmospherics and shopper responses. Psychology & Marketing,20(2), 139e150.

Fiore, A. M., & Kim, J. (2007). An integrative framework capturing experiential andutilitarian shopping experience. International Journal of Retail & DistributionManagement, 35(6), 421e442.

Goldsmith, R. E. (2012). Brand engagement and brand loyalty. In A. Kapoor, &C. Kulshrestha (Eds.), Branding and sustainable competitive advantage: Buildingvirtual presence (pp. 121e135). http://dx.doi.org/10.4018/978-1-61350-171-9.ch008.

Ha, Y., & Im, H. (2012). Role of web site design quality in satisfaction and word ofmouth generation. Journal of Service Management, 23(1), 79e96.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate dataanalysis. Upper Saddle River, NJ: Prentice Hall.

Hair, J. F., Black, W., Babin, B., Anderson, R., & Tatham, R. (2006). Multivariate dataanalysis (6th ed.). Upper Saddle River, NJ: Prentice Hall.

Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronicword-of-mouth via consumer opinion platforms: what motivates consumers toarticulate themselves on the Internet? Journal of Interactive Marketing, 18(1),38e52.

Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structureanalysis: conventional criteria versus new alternatives. Structural EquationModeling, 6, 1e55.

Hung, K. H., & Li, S. Y. (2007). The influence of eWOM on virtual consumer com-munities: social capital, consumer learning, and behavioral outcomes. Journal ofAdvertising Research, 47(4), 485e495.

Jeong, S. W., Fiore, A. M., Niehm, L. S., & Lorenz, F. O. (2009). The role of experientialvalue in online shopping: the impacts of product presentation on consumerresponses towards an apparel web site. Internet Research, 19(1), 105e124.

Keller, K. L. (2001). Building customer based brand equity: A blueprint for creatingstrong brands (pp. 01e107). MSI Report.

Kihlstrom, J. F. (1987). The cognitive unconscious. Science, 237(4821), 1445e1452.Kim, A. J., & Ko, E. (2012). Do social media marketing activities enhance customer

equity? an empirical study of luxury fashion brand. Journal of Business Research,65(10), 1480e1486.

Kim, H., & Niehm, L. S. (2009). The impact of website quality on information quality,

value, and loyalty intentions in apparel retailing. Journal of Interactive Market-ing, 23(3), 221e233.

Koo, D. M., & Ju, S. H. (2010). The interactional effects of atmospherics andperceptual curiosity on emotions and online shopping intention. Computers inHuman Behavior, 26(3), 377e388.

Ladhari, R. (2007). The effect of consumption emotions on satisfaction and word-of-mouth communications. Psychology & Marketing, 24(12), 1085e1108.

Lee, J., Park, D. H., & Han, I. (2008). The effect of negative online consumer reviewson product attitude: an information processing view. Electronic CommerceResearch and Applications, 7(3), 341e352.

Lee, M., Rodgers, S., & Kim, M. (2009). Effects of valence and extremity of eWOM onattitude toward the brand and website. Journal of Current Issues & Research inAdvertising, 31(2), 1e11.

Lee, M., & Youn, S. (2009). Electronic word of mouth (eWOM): how eWOM plat-forms influence consumer product judgement. International Journal of Adver-tising, 28(3), 473e499.

Lipsman, A., Mudd, G., Rich, M., & Bruich, S. (2012). The power of “Like”: how brandsreach (and influence) fans through social-media marketing. Journal of Adver-tising research, 52(1), 40e52.

Mehrabian, A., & Russell, J. A. (1974). An approach to environmental psychology.Cambridge, MA: The MIT Press.

Olney, T. J., Holbrook, M. B., & Batra, R. (1991). Consumer responses to advertising:the effects of ad content, emotions, and attitude toward the ad on viewing time.Journal of Consumer Research, 17(March), 440e453.

Park, J. E., & Choi, E. J. (2013). Consequences of impulse buying cross-culturally: aqualitative study. International Journal of Software Engineering & Its Applications,7(1), 247e260.

Park, C., & Lee, T. M. (2009). Information direction, website reputation and eWOMeffect: a moderating role of product type. Journal of Business Research, 62(1),61e67.

Park, J., Stoel, L., & Lennon, S. J. (2008). Cognitive, affective and conative responsesto visual simulation: the effects of rotation in online product presentation.Journal of Consumer Behaviour, 7(1), 72e87.

Peng, G., & Kim, Y. (2014). Application of the stimulus-organism-response (SOR)framework to online shopping behavior. Journal of Internet Commerce, 13(3/4),159e176.

Ping, R. A. (1994). Does satisfaction moderate the association between alternativeattractiveness and exit intention in a marketing channel? Journal of the Acad-emy of Marketing Science, 22(4), 364e371.

Riegner, C. (2007). Word of mouth on the web: the impact of web 2.0 on consumerpurchase decisions. Journal of Advertising, 47(4), 436e447.

Smith, A. N., Fischer, E., & Yongjian, C. (2012). How does brand-related user-generated content differ across YouTube, Facebook, and Twitter? Journal ofInteractive Marketing, 26(2), 102e113.

Stern, H. (1962). The significance of impulse buying today. The Journal of Marketing,26(April), 59e62.

Stewart, D. W., & Pavlou, P. A. (2002). From consumer response to active consumer:measuring the effectiveness of interactive media. Journal of the Academy ofMarketing Science, 30(4), 376e396.

Sun, T., Youn, S., Wu, G., & Kuntaraporn, M. (2006). Online word-of-mouth (ormouse): an exploration of its antecedents and consequences. Journal ofComputer-Mediated Communication, 11(4), 1104e1127.

Tuten, T. L. (2008). Advertising 2.0: Social media marketing in a web 2.0 world.Westport, CT: Greenwood Publishing Group.

Wang, Y. J., Hernandez, M. D., & Minor, M. S. (2010). Web aesthetics effects onperceived online service quality and satisfaction in an e-tail environment: themoderating role of purchase task. Journal of Business Research, 63(9), 935e942.

Wang, Y. J., Minor, M. S., & Wei, J. (2011). Aesthetics and the online shoppingenvironment: understanding consumer responses. Journal of Retailing, 87(1),46e58.

Weun, S., Jones, M. A., & Beatty, S. E. (1998). Development and validation of theimpulse buying tendency scale. Psychological Reports, 82, 1123e1133.

Wildrich, L. (2013, May 2). Social media in 2013: User demographics for Twitter,Facebook, Pintrest and Instagram [Web log post]. Retrieved from http://blog.bufferapp.com/social-media-in-2013-user-demographics-for-twitter-facebook-pinterest-and-instagram.

Wu, J., Ju, H. W., Kim, J., Damminga, C., Kim, H. Y., & Johnson, K. K. P. (2013). Fashionproduct display: an experiment with Mockshop investigating colour, visualtexture, and style coordination. International Journal of Retail & DistributionManagement, 41(10), 765e789.

Wu, P. C., & Wang, Y. C. (2011). The influences of electronic word-of-mouth messageappeal and message source credibility on brand attitude. Asia Pacific Journal ofMarketing and Logistics, 23(4), 448e472.

Yalch, R., & Spangenberg, E. (2000). The effects of music in a retail setting on realand perceived shopping times. Journal of Business Research, 46(2), 139e147.

Yang, Z., Cai, S., Zhou, Z., & Zhou, N. (2005). Development and validation of an in-strument to measure user perceived service quality of information presentingweb portals. Information & Management, 42(4), 575e589.

Yu, Y. W., & Natalia, Y. (2013, July). The effect of user generated video reviews onconsumer purchase intention. In Paper presented at the Seventh InternationalConference on Innovative Mobile and Internet services in ubiquitous computing,Taichung, Taiwan. Abstract retrieved from http://ieeexplore.ieee.org/xpl/login.jsp?tp¼&arnumber¼6603780&url¼http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6603780.


power of consumers using social media examining the ......as arousal, fear, and liking as well as...

Documents

Transcript of power of consumers using social media examining the ......as arousal, fear, and liking as well as...