1 ADDING AN AGENT : ROLE OF THE INTERNET AS A CONSUMER SOCIALIZATI ON MECHANISM AMONG THE MILLENNIALS By QINWEI VIVI XIE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF T HE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Qinwei Vivi Xie
3 To my parents uncle, Matt and Alex for their love, support and guidance throughout my life and career. I could not ha ve made it this far without you
4 ACKNOWLEDGMENTS I would like to extend my sincerest gratitude to all of my committee members for their mentorship and support throughout the dissertation process. Dr. Morris I cannot thank you enough fo r being my chair, as your guidance and commitment has been vital for the completion of my doctoral research and dissertation. Moreover, your expertise and training has been critical to my academic success and directed my career goals. My thanks also go t o Dr. Goodman Dr. Lutz and Dr. Sutherland You entrusted me with your precious time, wisdom, and resources, without which I would not have grown into the researcher I am today. I could not possibly forget to offer my gratitude to my friends and family. Dad, mom, uncle, Matt and Alex you have supported me throughout my life and during the past four years. I would not be where I am today without you.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHA PTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 2 REVIEW OF LITERATURE AND HYPOTHESIS DEVELOPMENT ........................ 19 The Millennial Gene ration ................................ ................................ ....................... 19 Consumer Socialization ................................ ................................ .......................... 21 Social Learning Theory ................................ ................................ ..................... 23 Tr aditional Consumer Socialization Agents ................................ ...................... 25 Mass media influence ................................ ................................ ................ 25 Interpersonal influence ................................ ................................ ............... 27 Socialization Outcomes ................................ ................................ .................... 29 Attitude toward online advertising ................................ .............................. 30 Materialism ................................ ................................ ................................ 32 Emotional responses toward shopping ................................ ...................... 33 Internet as Consumer Socialization Agents ................................ ............................ 37 The Internet as Mass Medium ................................ ................................ .......... 39 Cultivation Theory ................................ ................................ ............................ 40 The Internet as Interpersonal Medium ................................ .............................. 44 High interactive channels ................................ ................................ ........... 45 Low interactive channels ................................ ................................ ........... 47 Consumer Susceptibility to Influence on t he Internet ................................ ....... 48 Social Structural Variables ................................ ................................ ............... 52 3 METHOD ................................ ................................ ................................ ................ 56 Resear ch Design ................................ ................................ ................................ .... 56 Sampling ................................ ................................ ................................ ................. 56 Measures ................................ ................................ ................................ ................ 59 Dependent Variables (Consum er Learning Skills) ................................ ............ 59 Attitude towards Internet advertising ................................ .......................... 59 L evel of materialism ................................ ................................ ................... 60 Emotional responses towards consumption ................................ ............... 60
6 Independent Variables (Web based Socialization Agents) ............................... 62 Usa ge of online television, radio, newspaper and magazine ..................... 63 Interpersonal communication about consumption on the internet .............. 63 Moder ator (Normative/informative susceptibility to personal influence ) ............ 64 Social Structural Variables ................................ ................................ ............... 65 Regression Models ................................ ................................ ................................ 66 Model Summary ................................ ................................ ............................... 66 Model Specifications ................................ ................................ ........................ 68 Ordinal least squares regressi on ................................ ............................... 68 Ordered logit regressions ................................ ................................ ........... 69 4 RESULTS ................................ ................................ ................................ ............... 73 Bivariate Result s ................................ ................................ ................................ ..... 73 The Internet as Mass Medium ................................ ................................ .......... 73 Relationship between frequency of viewing online television and consumer socialization (CS) outcome variables ................................ ..... 73 Relationship between frequency of reading online newspaper and CS outcome variables ................................ ................................ ................... 74 Relationship between frequency of reading online magazines and CS outcome variables ................................ ................................ ................... 74 Relationship between frequency of listening to online radio and CS outcome variables ................................ ................................ ................... 75 The Internet as Interpersonal Medium ................................ .............................. 75 Relationship between consumption related communication via high interactive tools and CS outcome variables ................................ ............ 75 Relationship between consumption related communication via low interactive tools and CS outcome variables ................................ ............ 75 Social Structural Variables ................................ ................................ ............... 76 Relationship between socioeconomic status and CS outcome variables ... 76 Relationship between gender and CS outcome variables .......................... 76 Relationship between age and CS outcome variables ............................... 77 Relationship between ethnicity and CS outcome variables ........................ 77 Multivariate Results ................................ ................................ ................................ 77 OLS Assumptions and Multicollinearity ................................ ............................ 77 Linearity of the phe nomenon ................................ ................................ ...... 77 Normality of the error term distribution ................................ ....................... 78 Autocorrelation ................................ ................................ ........................... 79 Constant variance of the error term (homoscedasticity) ............................. 79 Muticollinearity ................................ ................................ ........................... 80 OLS Results ................................ ................................ ................................ ..... 80 The Internet as mass medium ................................ ................................ .... 81 The Internet as interpersonal medium ................................ ....................... 81 Moderating effects of consumer susceptibility to personal influence .......... 82 Social structural variables ................................ ................................ .......... 82 Ordered Logit Results ................................ ................................ ...................... 83 The Internet as mass medium ................................ ................................ .... 84
7 The Internet as interpersonal medium ................................ ....................... 84 Moderating e ffects of consumer susceptibility to personal influence .......... 85 Social structural variables ................................ ................................ .......... 85 Emotional Responses toward Shopping as a Socialization Outcome ..................... 86 5 CONCLUSION AND DISCUSSION ................................ ................................ ...... 101 Discussion of Results ................................ ................................ ............................ 102 Conclusion and Implications ................................ ................................ ................. 109 Limitations and Future Research ................................ ................................ .......... 112 LIST OF REFERENCES ................................ ................................ ............................. 115 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 134
8 LIST OF TABLES Table page 3 1 Scale d escriptions ................................ ................................ .............................. 70 3 2 Descriptive statistics for key v ariables ................................ ................................ 71 4 1 Bivariate relationship: consumer skills and socialization v ariables ..................... 91 4 2 OLS results of consumer s ki lls measures and socialization v ariables ................ 92 4 3 O rdered Logits of the ER toward online shopping m odels ................................ .. 93 4 4 Ordered Logits of the ER toward mall/store shopping m odels ............................ 94 4 5 Emotional responses to shopping online and shopping in a mall/s tore .............. 95 4 6 Group differences in s ociali zation variables (ER to online s hopping) ................. 96 4 7 Group differences in socialization variables (ER to m all /store s hopping) ........... 97
9 LIST OF FIGURES Figure page 3 1 The AdSAM visual s cale ................................ ................................ .................. 72 4 1 Normal probability plot for attitude toward online advertising m odel ................... 98 4 2 Normal p r obability plot for materialism m odel ................................ ..................... 98 4 3 AdSAM Groups emotional responses to online s hopping ............................. 99 4 4 AdSAM Groups emotional responses to mall s hopping .............................. 100
10 LIST OF ABBREVIATIONS CCH Co nsumption related Communication via High interactive Channels CCL Consumption related Communication via Low interactive Channels CS Consumer Socialization ER Emotional Responses OR Odds Ratio SES Socioeconomic Status
11 Abstract of Dissertat ion Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ADDING AN AGENT : ROLE OF THE INTERNET AS A CONSUMER SOCIALIZATI ON MECHANISM AMONG THE MILLE NNIALS By Qinwei Vivi Xie May 2013 Chair: Jon D. Morris Major: Mass Communication Consumer socialization (CS) is the process by which individuals develop consumer related skills, knowledge, and attitudes. Socialization agents are the persons, or inst itutions, that transmit norms, attitudes or behaviors through frequency on contact with the consumer. Traditionally, CS research has accepted three types of influence agents in the CS process: the mass media, family (particularly parents), and peers. How ever, there have been no studies to date that look at the Internet as a socialization mechanism. Using CS as a framework, this dissertation examined the effects of online media and online communication processed on various consumer skills among the Mille nnials. It conceptualizes attitude toward online advertising, materialism and emotional responses towards shopping as outcomes of the socialization process, all general consumer skills learned through interaction with the newly proposed online socializati on agents. In particular, it was hypothesized that the extent of mass media usage through the Internet (online television, radio, newspaper, etc.), as well as, the frequency of interpersonal communication about consumption matters via various online venue s (high vs low interaction), relate positively to the three socialization outcomes. O
12 susceptibility to personal influence (informational and normative) would moderate the relationship between online interpersonal communication, and the three socializ ation outcomes. This framework was tested with survey data from 693 young adults, between 21 and 31 years of age (i.e. ) A linear regression model predicted that high interactive communication about consumption matters was a strong pre dictor for both attitude towards online advertising and materialism. An ordered logit regression resulted in similar findings and provided further evidence of the strong relationship oward shopping However, the amount of online television viewing was not significantly associated with any of the outcome variables Overall, f indings in this dissertation demonstrated that t he Millennial consumers are turning away from passive learning (online mass media); instead, they develop consumer attitudes and values by actively participating in online conversation about consumpti on (high interactive channels) or by actively pursuing consumption related information (low interactive channels).
13 CHAPTER 1 INTRODUCTION Defined by Ward (1974 p. 2 individuals acquire skills, knowledge and attitudes relevant to their functioning as seventies, the majorit y of studies on consumer socialization (CS) have focused on the effects of traditional socializing agents consumer attitudes, skills and knowledge. The reason is because thes e three elements are what individuals encounter on a daily basis. However, the contemporary media environment has changed tremendously since early studies of consumer socialization (Moschis & Churchill, 1978). T he advent of the Internet represents one of the more significant changes. Young adults have completely integrated the Internet into their day to day lives (Moscardelli & Liston Heyes, 2005; Calenda & Meijer, 2009). They not only utilize the Internet as a form of medium (e.g. movies, news, music, sh opping, gaming etc.), but are also able to move interpersonal communication, whether that be with parents, peers or applications are also, more often than not, intertwined with social networking sites, email, IM (Instant Messaging) and even blogs. As a result, Internet greatly impacts their beliefs and expectations about consumption. In order to understand the group of individuals who grow up having Internet in their life as consumers, it becomes critical to consider Internet as a potential socializing agent. Naturally, the first step is to examine the level Internet use within the Millennials generation, as this is the first generation that experienced the rapid development and prevalence of the Internet.
14 Millennials (Gen Y) are identified as in 1981 to 1991 the first That is, at the time of writing (2012) they are aged between 21 an d 31 years old (e.g. Chu & Kamal, 2011; Pew Research, 2010). They are currently the largest generation in Rainer & Rainer 2011). Having grown up with access to the Internet, Millennials use the Internet for a wide variety of activities such as watching videos/movies, instant messaging, reading news, writing blogs, listening to music, and ( Myers & Sadaghiani, 2010; Pew Research, 2010), the Millennials tend to use ne w communication tools such as IM (instant messaging), email and social networking sites to obtain social support (LaRose, Eastine & Gregg, 2001); to reduce depression (Morgan & Cotton, 2003); and to maintain their personal offline relationships ( Subrahmany am, Reich, Waechter & Espinoza, 2008). As the Millennials spend so much time online working, playing and interacting, the Internet has become an important avenue for reaching this consumer group. The Web is different from traditional media (i.e. televisio n, radio, newspaper and magazine), because it creates an opportunity for users to not only interact with the cyber space, a machine (Lee, Conroy & Hii, 2003), but also connect to strangers from all over the world. Users use chat rooms, instant messaging an d SNSs (Social Networking Sites) to communicate with their friends and family members, on a daily basis ( Subrahmanyam et al., 2008 ; Steinfield, Ellison & Lampe, 2008). Therefore, as a hybrid of mass media and interpersonal communication channels, the ubiq uitous Internet has major potential
15 remain constantly in contact with each other. Surprisingly, only a few studies have looked at the Internet as a socialization me chanism (e.g. Moscardelli & Liston Heyes, 2005). Television, family members and peers are widely considered as the main agents of socialization (e.g. Martin & Bush, 2000; Moschis & Churchill, 1978; Roedder John, 1999). A question arises about how the Inter net plays a role in providing individuals with information used in constructing their consumer reality. Moreover, the few studies that have taken the Internet into account as a CS (consumer socialization) agent only consider it as a type of mass media. Pri or studies that have considered Internet as a possible agent typically used a total, or general, measure of Internet use (Chia, 2010; LaFerle & Edwards & Lee 2000; Park, Villar & Amador, 2010 ). None of these studies, however, have looked further into diff use patterns and what aspects of the Internet have an impact on CS. In addition, small sample size and the qualitative nature of these studies (e.g. Lee & Conroy, 2005; Lee et al., 2003) also preclude the accuracy of res ults. Moreover, most studies on CS consider the way that individuals learn to become consumers as a negative process (Lee & Conroy, 2005). Nevertheless, with people spending more time shopping, working, communicating and reading on the Internet, individua the consumer reality for the Millennials, the current study presents the following objectives: (1) to examine the influence o f the Internet (as mass media and interpersonal related attitudes
16 and values i.e. attitudes toward online advertisements, emotional responses towards shopping and level of materialism; (2) t o assess how online socialization agents influence young consumers differently with regards to the three consumer socialization outcomes; (3) to investigate the effects of socioeconomic and demographic variables as antecedents on consumer learning. The re sults of this research will further refine and extend the existing knowledge regarding CS in the following ways. Very few studies have looked into consumer socialization beyond adolescence ( e.g. Bush, Smith & Martin, 1999 ; Smith & Moschis, 1985; Singh, Kw on & Pereria, 2003). Moschis (1981; 1987) advocates that individuals learn different things at different times from different sources throughout the whole life span. Therefore, ycle Moving the focus beyond children and adolescents and studying the socialization effects on adults has long been called for in consumer socialization research ( e.g. Brim, ; Singh et al., 2003; Ward, Klees & Robertson, 1987 ) However, to date, socialization among young adults remains a relatively unexplored area. The present research fills this gap by extending the literature to the generation of Millennials, and exploring how Internet, as an emerging technology and possibly a consumers. The results of this research will provide insights regarding how the Millennials utilize Internet as socialization agents to seek market information. Other tha n testing two common outcome properties attitude towards advertising and materialism responses toward consumption as a possible outcome of the socialization process. That
17 is, does consume shopping? Major studies in the field of consumer behavior have demonstrated the ( Batra & Ray 1986; Bur ke & Edell 1989; Holbrook & Batra 1987; Morris, Woo, Geason & Kim, 2002) can also be found in psychology literature ( e.g. Esses, Haddock & Zanna, 1993; Stangor, Sullivan & Ford, 1991) Given that prior research on consumer socialization has not taken affective responses into account as a CS outcome this paper also contributes to the literature by potentially adding an additional outcome variable, that is the emotional response, to CS theory. As the Internet has become a critical avenue for reaching the Millennials, research on how the web impacts their socialization processes could also yield insights into how marketers can develop more effective new media campaigns directed at this group of young consumers. This research addresses questions such as whether Internet as mass media has the same cultivation effect as offline traditional media or whether social influence exerts a powerful effect on consumers via Internet channels. Answe rs to such questions will help marketers determine which online communication tools are most effective at reaching the Millennial generation, and how marketing messages could be effectively conveyed to consumers via the Internet. In addition, given that c onsumer skills and attitudes are expected to influence purchasing behavior, advertisers should constantly monitor consumers' level of materialism, attitude towards Internet advertising and shopping as well as how various online activities are related to th em (Yoon, 1995). Previous advertising research has
18 pointed out that public acceptance of advertising may continue to deteriorate as a result Zanot, 1981) This no t so new trend has been a concern for marketers because effectiveness (Bush et al., 1999; Yoon, 1995). However, these studies mainly focused television advertising. Given that marketing practices on the Internet are still a relatively new phenomenon, it becomes critical to new form of advertising and shopping before Internet advertising reach maturity. Chapter 2 presents background literature and a traditional consumer socialization framework, followed by a presentation of the new online consumer socialization model and hypotheses, as well as conceptual theor ies that support them. Chapter 3 covers research design and data analysis method. Chapter 4 presents the empirical results. Finally, conclusion, implications and future research recommendation are addressed in Chapter 5
19 CHAPTER 2 REVIEW OF LITERATURE AND HYPOTHESIS DEVELOPMENT The Millennial Generation Each generation has shared traits, attitudes and experiences that unite its members. the Millennials generation is portrayed ( Pew Research, 2010, p. 1). A primary characteristic of the Millennials is that they are very technology savvy. The Internet is their primary source of product and purchasing related information ( Nowak, Thach & Olsen, 2006; Nowak & Newton, 2008 ). Other research indicates that they use various online venues to discover or talk about products and services and, in turn, rely on others for purchase related information (eMarketer 2011; Tapscott 1998; Wi edman, Wals ch & Mitchell 2001). In his book up immersed in an Internet driven world. In addition to the large amount of time they spend online, the Millennials often use the Internet for social reasons ( Ellison, Steinfield & Lampe, 2007) Evidence suggests that the Millennials are very sensitive to peer pressure ( Temkin & Popoff Walker, 2007) Most major consumer product companies consider the Millennials market segment as a generation with high buying power ( Bruwer, Saliba & Miller, 2011; Nowak, Thach & Olsen, 2006 ). The Millennials are also the most educated generation in American history (Pew Research, 2010). With one third of the millennial generation being non Caucasia n (Key Findings, 2004), they are a more ethnically and racially diverse group than older generations (Pew Research, 2010). The major focus here is
20 the acquisition of consumer learning properties (outcome) among the Millenn ials. In order to adapt to the fast changing marketplace, consumers from different age groups must continuously learn and replace old attitudes, skills and behaviors toward nd Shrum (1997) suggested that socializing effects are present throughout the adult life cycle. Phillips and Sternthal (1977) advocated that consumer socialization research should expand beyond children and devote more attention to the aged consumer group. While most previous studies on consumer socialization have focused on children and adolescents (Ward, Klees & Wackman, 1990), the parameters of CS research have expanded to include adult age groups. Singh, Kwon and Pereria (2003) studied the CS process o f Generation X (age then: 22 30 years) from three ethnic groups within the United States, and found that young adults from these three ethnic groups differ significantly in their susceptibility to different socialization agents. Choi and Ferle (2004) compa red consumer socialization variables between Korean and American young adult consumers. Their findings indicate communication, especially by peers, followed by parents, an d then mass media. Although Koreans are found to be more motivated to comply with others than their American counterparts. Moore Shay and Berchmans (1996) examined intergenerational overlap and found positive correlation between young adults and their pa rents regarding different consumer variables. Smith and Moschis (1984) explored the effects of consumer socialization processes on attitudes toward advertising among the
21 elderly. Their evidence suggested that cognitive age was positively associated with a ttitudes toward advertising among the elderly. Bush et al. (1999) investigated the influence of consumer socialization variable on attitude towards advertising among African American and Caucasian young adults. Their findings indicated that interpersonal communication s mass media consumption, gender and race were all significantly related to attitude towards advertising. Meanwhile, African Americans presented more positive attitudes toward advertising than Caucasian Americans. Rindfleisch, Burroughs and Denton (1996, 1997) s ampled a group of young adults between the age 20 to 32, and found that compared to young adults reared in intact families, those who grew up in disrupted families exhibit higher levels of compulsive consumption and are more materiali stic. Benmoyal Bouzaglo and Moschis (2009) adopted a similar approach, with a sample of French college students. Their findings showed similar trends as those reported by Rindfleisch et al. (1997). Consumer Socialization The broad term socialization ref ers to the process by which an individual acquires, through interaction with others, certain cognitions and behaviors that help them participant effectively as members in a group or in the society in general (e.g. Brim, 1966; Goslin1969). As a sub concept of socialization, consumer socialization consumers (e.g. Moschis & Churchill, 19 78; Ward, 1974) As the theoretical framework used in the current study, the concept of consumer socialization (CS) is first addressed by Ward (1974). In the original model of CS, Moschis and Churchill (1978) proposed three components of the process: 1) social structural variables such as gender, age,
22 race, socio economic status, and other demographic factors as antecedents, 2) socialization process that includes both the CS agents, namely mass media, parents and peers, as well as the progression of agent learner interactions, and 3) the outcomes Moschis & Churchill, 1978). As the dependent variable throughout the current study, emotional responses towards consumpt ion, materialistic attitudes and attitudes towards online advertising are considered as the consumer learning properties. Both cognitive development model and social learning model each provide a fundamental theoretical framework for CS research (Moschis & Churchill, 1978; Churchill & Moschis, 1989). Cognitive development theory takes a developmental perspective, and represents the view of learning that focuses on how different f learning (Baran & Davis, 2009). CS researchers employing cognitive development theory believe that socialization is a function of development stages in cognitive progress through out childhood and adolescence (e.g. Ward 1974; Ward, Wackman & Wartella 19 77). In other words, the focus of cognitive development theories is the interaction process of personal and environmental factions, whereas learning theories emphasize change of attitudes and values as a function of forces applied to the individual (Ward, 1974). Scholars who adopted social learning theories attempt to explain formation of CS outcomes from sources of influence (agents) transmitting attitudes, motivations, and values to the individual (e.g., Bandura, 1977; 1986; Churchill & Moschis, 1979; M oschis & Churchill 1978; Moschis, Lawton & Stampfl 1968).
23 Given that cognitive development model focuses on the developmental process that proceeds through a series of stages as children mature into adult consumers, while social learning theory appears to attitudes and values as consequences, which is in line with the current purpose ( Moschis & Churchill 1978; Roedder John, 1999). T his dissertation bases solely on the social learning mechanism. Moreover, p revious evidence showed that consumer socialization is more of a social learning interaction than a cognitive development process ( Moschis & Churchill 1978) The cross sectional nature of this study also does not allow for studying the processes themselve s. Cross sectional designs, however, are suitable for studying the extent and outcomes of agent learner interactions (Churchill & Moschis, 1979). Social Learning T heory Social learning theory holds that within a social structural context, individuals int eract with and learn from socialization agents through modeling, reinforcing and social interaction processes (McLeod & O'Keefe, 1972). Contemporarily, social learning theory is also known as social cognitive theory (Baran& Davis, 2009). As with previous CS research with a cross sectional design (e.g. Bush, Smith & Martin, 1999; Moschis & Churchill, 1978; Ward, 1974), the current study is not suitable for studying agent learner interaction processes themselves. Therefore, although exploring the types of processes associated with CS is critical, the focus here is only on the potential impact of online CS agents on the outcome of socialization. In an agent learner relationship, the socialization agent is any person or entity cognition and behavior through a series of direct contacts with the learner (Brim, 1966). T
24 attitudes, values and behaviors through a combination of modeling and reinforcement activities (Moschis & Churchill, 1978). Modeling is learning through observation and either rewards as positive reinforcement or punishments as negative reinforcement (Moschis & Churchill, 1978; McLeo operate jointly as a social interaction mechanism in the learning process (Bush et al., 1999; Moschis & Moore, 1979) Individuals develop similar attitudes and values as they interact frequently with a consi stent group (Lee & Convoy, 2005). In addition, individuals tend to interact with those who hold similar values and attitudes as them (Fiske & Taylor, 1991). As stated previously however, this dissertation does not intend to focus on the type of learning nor the processes per se, but rather to explore the impact that online socialization agents might have on materialistic attitudes, emotional responses towards consumption and attitudes toward online advertising. As a matter of fact, most CS studies did n ot subdivide socialization learning into modeling and reinforcement categories because of the nature of a cross sectional study. In most of the past socialization liter ature, learning theories assume that the recipient of socialization influence is a pass ive process (Lee, et al., 2003; Moschis & Churchill, 1978; Villani, 2001). Contemporary research of media effects suggests the opposite that people are active and motivated explorers of media ( Valkenburg & Cantor 2001 ). Particularly, the Internet has cr eated a new learning environment where learners are able to actively learn and participant. The modeling and reinforcement components discussed above, for instance, may no longer hold entirely, when young
25 adults today also learn through a process of disco very and participation (Tapscott, 1998). Traditional Consumer Socialization Agents Socialization agents, as the sources of influence, transmit norms, attitudes, motivations and behaviors to the other end of the learning process, the learner (Moschis & Chu rchill, 1978). Most prior CS studies accept two types of influence agents in CS process: mass media and interpersonal influence. Media influence concerns the effect of materialistic aspects of media and the effect of advertising, whereas social influence includes the impact of family and peers or friends (e.g. Chia, 2006, 2010; Martin & Bush, 2000; Moschis & Churchill, 1978; Roedder John, 1999). As pointed out in on the influence forces of consumer reality: It is hard for us to realize how little of our information comes from direct experience with the physical environment, and how much of it comes only indirectly, Mass media influence Mass media in general function as the transmitter of norms and values of a society (Ward, 1974). Individuals gather and sift information about consumption matters from all types of mass media. Numerous studies have recognized the influence different types of mass media, particularly televis ion, consumer skills ( e.g. Englis, Solomon & Olofsson, 1993; Mangleburg & Bristol, 1998; Moschis and Churchill (1978) argued that the amount of TV viewing is a major predicto r of materialistic values and motivations for consumption. Richins ( 1987) also found the correlation between
26 television viewing and material values among adult consumers, although attention paid to commercials was not related to materialism. Although pre vious CS research considered television as the most influential in constructing consumer reality, other forms of mass communication such as radio and print media also play a role in this learning process. Moore and Stephens (1975) investigated adolescents radio exposure on the formation of consumer skills and attitudes. They found that print media are more important predictor variables among younger adolescents, whereas television exposure time is a more powerful predictor among the older adolescents. Ward and Wackman (1971) examined relationships between television viewership and materialistic values as well as attitudes towards advertising. Their results suggested that social utility reasons for television and advertising exposure strongly associated with materialism and advertising attitudes among younger adolescents, while vicarious consumption reasons and communication about consumption with others were stronger predictor of attitudes towards advertising among older adolescents. Moschis and Moore preferences. In addition, both the editorial content and the advertising that mass media carry, employ roles in cul tivating consumer reality (Englis et al., 1993). Editorial content (of newspapers and magazines) or programming (of television and radio) may impact consumer learning directly or indirectly. For instance, a person may aspire to have the material goods of characters in a television program or a model from a magazine
27 (Moschis & Churchill, 1978), w hereas advertisements can be view ed as more of a Bandura (1971, 1986) suggested that through ad vertising exposure, people learn to attach social meaning to material goods. Ward, Wackman and Wartella (1977) also (1997) conducted two studies to examine the role of televi sion programming in consumer socialization. Their results stressed that television exposure is a key cause for the increase of materialism in young adults. However, a shortcoming of these studies is that it tried to separate the commercial contents of mas s media from actual editorial/program content (McNeal & Mindy, 1998). In real life, one could hardly distinguish the influence of these two media elements. Interpersonal influence Interpersonal influence occurs when an individual adapts his/her attitudes values and beliefs of other individuals (Trusov, Bodapati & Bucklin, 2010). Consumer researchers have long recognized interpersonal influence as a major determinant of consumer behavior ( Calder & Burnkrant, 1977). Two major traditional (offline) interpe consumer skills. Sociologists claimed that children first learn the rational aspects of consumption from their pa rents (Reisman & Roseborough, 1955). Consumer socialization literature has suggested that communication about consumption with quality relationships ( Ward & Wackman, 1973), social a nd economic motivations for consumption (Moschis & Churchill, 1978) and attitude towards advertising (Bush et al.,
28 socialization process, Rindfleisch et al., (1996, 1997) found that young adults from disrupted families are more materialistic and exhibit higher levels of compulsive consumption than those from intact families. However, the Millennials who are between the age of 21 and 31 are at a stage where they have started to detach themselves from parents and attach themselves to other social groups (Chia, 2010). With this transformation of personal relationships, eases (Chia, 2010; Feltham, 1998) acquisition of consumer skills seem to process more through subtle influence by media or other people than through direct and purposive consumer education or training by parents and schools (Ward, 1974). People also learn expressive aspects as well as styles and moods of consumption from their peers (Moschis & Churchill, 1978; Reisman & Roseborough, 1955;). They seek information regarding product evaluations, brands and store choices from their peers or friends in order to enhance their sense of belonging (Mangleburg, Doney & Bristol, 2004). Consumer socialization studies have shown that acquisition of consumer skills. puffery from fact positively relates to the frequency of communication with their peers. (Moschis & Churchil, 1978), product evaluation (Moschis & Moore, 1979) and attitudes toward advertising (Bush et al., 1999).
29 As the Millennials spend increasing amounts of time on the Internet, the circle of influence has become wider and has been shifted from offline to online (Roe dder John, 1999). Not only do individuals consume greater amounts of electronic media and spend less time in traditional way of using mass media, they also communicate with each other more on the digital world rather than face to face (German & Lally, 200 7). Moreover, by allowing young consumers to reach beyond geographic and time constraints and facilitating their access to wider and richer information, the Internet may enhance the learning process (Lee et al., 2003). Socialization Outcomes The learning of consumer reality involves the acquisition of a wide range of (Moschis & Churchill, 1978; Moore & Stephens, 1975). Socialization outcomes used in previous research includ e skepticism toward advertising (Mangleburg & Bristol, 1998), attitudes towards advertising in general (Moore & Stephens, 1975; Rose, Bush & Kahle, 1998; Ward & Wackman, 1971), attitudes toward prices (Moschis & Churchill, 1978), ability to filter advertis ing puffery (Moschis & Moore, 1979), social and economic motivations for consumption (Carlson, Walsh, Laczniak & Grossbart, 1994; Moschis & Churchill, 1978) and materialism (Churchill & Moschis, 1979; Rindfleisch et al., 1997; Rindfleisch, 1994; Ward & Wac kman, 1971). In order to cover both cognitive and affective consumer values, the learning properties under current investigation are attitudes towards online advertising, materialistic attitudes and emotional responses towards shopping. As indicated abov most CS studies are general rather than specific knowledge, attitudes, or skills. One
30 could argue that consumers absorb general attitudes and values from mass media rather than specific inform ation for consumption related skills (Lee, 1989). Ward (1974) distinguished skills, knowledge, and attitudes that are directly and indirectly related to consumption behavior. Direct consumer skills are those that are necessary for enactment of the consum er role. These include skills at budgeting and bargaining, knowledge of brand and price, attitudes toward specific product, or stores and sales persons. However, not all consumption behavior occurs as a function of direct consumer skills. Ward (1974) fu rther argued that indirect consumer skills that are not closely related to purchase decision or transaction itself (e.g. materialistic values, attitude towards the marketplace and business activity in general, knowledge concerning social norms and expectat ions) are more important in influencing shopping behavior. All three CS outcomes, as consumer skills involved in the current research, are indirectly relevant properties. Attitude toward online advertising As an important consumer skill, attitude toward advertising is one of the outcome variables in consideration. Advertising research has discussed extensively the attitudes towards advertising construct. This is due to the notion that attitude toward advertising is important in determining attitudes tow ards specific advertising (Lutz, 1985), which is in turn, a predictor of attitude toward the brand (Mckenzie & Lutz, 1989) as well as purchase intention ( MacKenzie, Lutz & Belch 1986). In the same vein, general attitude towards online advertising should s advertising (Schlosser, Shavitt & Kanfer, 1999) The impact of overall attitudes toward advertising on consumer behavior variables has always been of interest to advertising and marketing researchers. Sha vitt,
31 favored attitude towards advertising in general than what previous research suggested. Lutz (1985) suggested that attitudes toward advertising in general have an influenc e on that attitude toward advertising in general relates positively with involvement with individual advertisements. Yoon (1995) suggested that materialism is associated wi th positive attitudes toward advertising. Numerous consumer studies have measured attitudes toward online advertising attitude toward Internet advertising in general and its antecedents. He found that the Internet users considered online advertising to be informative and entertaining. Schlosser et al., (1999) compared attitudes toward online advertising in general and advertising in general with a national sample. T heir results show that more users perceived online advertising to be informative and trustworthy, than a demographically similar sample group found general advertising. Attitude toward Internet advertising is defined as a learned predisposition to respond consistently in a favorable or unfavorable manner toward online advertising in attitude toward advertising in general. Online advertising includes a wide variety of for mats from ads that are similar to traditional advertisements (e.g., billboards, banner ads) to new forms such as social ads or sponsored stories appearing on social networking sites (Xie, Zhang & Morris, 2011), hence consumers may idiosyncratically diffe r in their perceptions of what constitutes online advertising (Schlosser, Shavitt &
32 general consumer skills as the consequence of socialization online advertising is desc ribed broadly as any type of commercial content available on the Internet that is financially sponsored by companies (Schlosser, Shavitt & Kanfer, 1999). Materialism materialis tic values. Marketing and media scholars have documented the joint effects between family communication patterns, social influence, motives for media consumption, and the amount of media consumption as predictors of materialism (Chia, 2010; Moschis, 1978, 1987; Moore & Moschis, 1981; Moschis & Moore, 1979; Ward & Wackman, 1971). When used as a philosophical concept, materialism refers to the idea that nothing matters except for material possessions (Lange, 1865, 1925). In economic psychology and consumer views of material goods and money as an important path to personal happiness and social progress (Ward & Wackman, 1971), as well as the extent to which one attaches to worldly possessions (Chia, 20 10; Belk, 1985; Richins & Dawson, 1992). material goods, and see them as more of instrumental in achieving social goals and status, rather than simply fulfilling a functional need (Ward, 1974). High materialism consumers tend to emphasize pursuing unique consumer products (Lynn and Harris, indicated that materialism leads to compulsive buying (Ri ndfleisch et al., 1997; Rose, 2007), social consumption motivation (Fitzmaurice & Comegys 2006), self doubt and insecurity (Chang & Arkin 2002) etc.
33 Multiple studies have demonstrated the critical role of consumer socialization in the development of mater ialistic values (e.g. Ahuvia & Wong, 2002; Churchill & Moschis, 1979; Rindfleisch et al., 1997; Rindfleisch, 1994). Churchill and Moschis (1979) are among the first researchers to integrate materialism into a structural model of consumer socialization. Th e results of their model showed that materialistic values increase with the amount of television viewing and with the frequency of peer communication. Interestingly, opposite to most studies, males displayed higher materialistic orientation than females in their research (Churchill & Moschis, 1979). Moore shay and Berchmans (1996) investigated intergenerational overlap and found that conflicts between parents and young adults about consumption matters are a key indicant of m. Other studies also suggest that peer influence associates strongly with materialism (Achenreiner, 1997; Schroeder & Dugal, 1995). the exposure to television is highly related to perceptions of an affluent society. Shrum, Burroughs and Rindfleisch (2005) provide further evidence that the effects of television viewing extend beyond merely perceiving that society is full of high status goods and services, but to also having an effect on Emotional responses toward shopping In a summary of social learning theory, Bandura concluded that television towards persons, or things that have been associated with mod eled emotional
34 peers and mass media (Parson, Bales & Shils 1953; Reisman & Roseborough, 1955). have a direct influence on behavior that is not captured or summed up by attitude lso ample evidence that feelings or emotional responses (ER) have substantial effects on attitudes, intentions, and actual behavior (Batra & Ray 1986; Burke & Edell 1989; Edell & Burke 1987; Holbrook & Batra 1987; Stayman & Batra 1991). However, emotional response, despite its importance in consumer decision making and behavior, has not drawn attention from CS scholars as a possible socialization outcome. messages and found that ER dominates over cognition for predicting cognitive attitude and purchase intention. Holbrook and Gardner (1998) investigated the effects of Moreover, Batra and Ray (1986) argue that emotional responses should sup plement the cognitive responses, such as attitudes, more often studied in advertising research. They empirically demonstrated that the relevant chain of effects is Emotional responses Attitude towards the ads Attitude towards the brand Purchase inten tion (Batra & Ray 1986). Researchers have investigated different aspects of consumer affective responses such as using pre production measurements to predict emotional responses to finished commercials (Morris & Waine, 1993); emotive aspects of hedonic con sumption (Hirschman & Holbrook, 1982); emotional responses towards advertising/marketing messages (Morris, et al., 2002) ; TV commercials ( Morris & McMullen, 1994; Morris, Klahr, Shen, Villegas, Wright, He & Liu 2008); affective
35 response to the consumption of specific product categories (Holbrook, 1980; Levy, 1999 ); and the effects of emotional responses to media context (particularly television programming) on advertising evaluations (Coulter, 1998). The consumer ER system that has probably received the g reatest attention from consumer behavior researchers thus far is the three dimensional Pleasure Arousal Dominance ( PAD) theory (Mehrabian & Russell, 1974). Russell and Mehrabian (1977) classify the full spectrum of human emotions into three relatively ind ependent and bipolar dimensions: pleasure displeasure, arousal calm and dominance submissiveness. A s an emotional state pleasure is distinguished from "preference, liking, positive reinforcement, or approach avoidance...since the latter responses are als o determined by the arousing quality of a stimulus" (Mehrabian and Russell, 1974, p. 18). Pleasure is a combination of feelings such as happiness, contentment and satisfaction. Arousal activation. Dominance represents the extent to which one feels in control over a situation. Since then, numerous researchers have employed this theory for their study of emotion in the context of consumer behavior (Christ, 1985; Christ & Biggers 1984; Holbrook & Batra 1987; M orris, Woo & Chao, 2003; Morris et al. 2008 ). Based on its ability to characterize diverse emotional responses, and to determine behavior, the PAD approach is used here as theoretical base for the measure of emotional responses PAD theory was originally designed as emotional responses to situational stimuli, of the situation on actual behavior (Lutz & Kakkar, 1975). Moreover, Lutz and Kakkar (1975) claimed that PAD shoul d be able to explain general consumption situations
36 because the three dimensions of emotion are regarded as generic dimensions of situations. Although there are few studies that have looked at emotional responses toward consumption situation in general (e .g. shopping), many have examined Moschis 1987) Donthu and general. Karande and Ganesh (2000) explored outlet mall sh shopping in general. attitudes towards shopping in Internet stores. Yang and Lester (2004) compared attitudes towards Internet shopping between online shoppers and non shoppers. 1973). This emotional response then would guide subsequent response such as attitudes, intentions and behaviors ( Batra & Ray 1986; Holbrook & Gardner, 19 98). Therefore, this study considers two shopping situations (i.g. shopping on the Internet, shopping in a mall) as the me of consumer socialization. Most measures of attitude in the field of consumer behavior are verbal scales, verbal measures, individuals are required to have advanced verba l skills and cognitively analyze the information (Morris et al., 2002). In other words, the underlying assumption of these scales is that subjects are able to access various attitude components, and interpret their thoughts or feelings into responses acco rding to the verbal scales (Morris et al., 2002). Given that verbal measure requires cognitive processing, it does a poor
37 job differentiating emotional attitudes from cognitive attitudes (Morris et al., 2002). In the case of emotional responses however, nonverbal measures, such as those that consists of graphic materials, are able to avoid this cognitive processing ( Edell & Burke, 1987). Additionally, by not using verbal responses, it eliminates the problem that individuals comprehend the meaning of emot ional words differently. Therefore, this paper adopts a nonverbal scale AdSAM based on Self Assessment Manikin [SAM] (Lang, 1980) The method section illustrates this scale in more detail. Internet as Consumer So cialization Agents As is reflective of previous research on the effects of mass media and personal influence, the Millennials incorporate and sift consumption information from various digital sources (Lee & Conroy, 2005). Therefore, the Internet should al so represent an agent of socialization, as it provides people with a new resource to obtain knowledge about the world (LaFerle, Edwards & Lee, 2000). This underscores the need for CS scholars to expand the focus well beyond the confines of traditional pri nt or broadcast mass media (Englis et al., 1993; Solomon & Englis, 1993). Over the last two decades Internet has transformed the way we obtain and communicate information. Different from the era when mass media controlled the production and dissemination of news, marketing messages, product reviews and other information, the digital era makes it possible for the masses to be a major part of the information process. Words previously used to describe the masses such as (Xie, 2012). The question that arises is: is the Web a mass media channel or an
38 interpersonal communication channel? impossible to precisely define both this med ium and its users (Morris & Ogan, 1996). As a hybrid of all possible configurations of communication, on one hand, the Internet has the ability to disperse traditional media content to mass users. Sites operated by major networks, radio stations, newspap er and magazine companies are typical examples; on the other hand, the Web allows consumers to provide real time or immediate feedback to the individual originating the communication. Internet based applications such as email, instant messaging (IM), chat rooms and social networking Computer mediated communication researchers advocate that the inquiry for interactivity should be based on a continuum (e.g. Coyle & Thorson, 2001; Morris & Ogan, 1996; Rafaeli, 1988 ; Rafaeli & Sudweeks, 1997; Richards, 2006 ). At the least interactive end is one way declarative communication, whose m ain purpose is to disseminate media messages ( Rafaeli, 1988 ; Rafaeli & Sudweeks, 1997) ; at the most interactive end is fully interactive communication, the purpose of wh ich is to create and maintain personal contacts and social networks ( Rafaeli & Sudweeks, 1997) ; quasi interactive (or reactive) communication lies in the middle of the spectrum ( Rafaeli, 1988; Thompson, 1995). This concept helps to consider online interac tivity to be variable in nature, increasing or decreasing with particular Internet platforms in question (Morris & Ogan, 1996). As the virtual world offers users a wide variety of venues for interpersonal communication (e.g. social networking sites, review sites, blogs, Instant messaging, emails etc.), this paper divides the Internet into three types of communication medium
39 according to level of interaction as well as primary goal of its users. The following classification is also an update of Morris and O INTERNET AS MASS MEDIA A ll one way main objective is to receive information. This type of communication is asynchronous and largely non interactive, where whether users contribute t o the communication process does not impact the content itself. Online based television program, online radio, newspaper and magazines are amongst this category. INTERNET AS INTERPERSONAL COMMUNICATION (HIGH INTERACTION). S ites imary goal is to create and maintain relationship with friends, family or professional associates through on going conversations (Xie, Zhang & Morris, 2011). Instant messaging, Social networking sites (e.g. Facebook), chat rooms and email are typical high interaction communication tools. INTERNET AS INTERPERSONAL COMMUNICATION (LOW INTERACTION ). W ebsites such as review sites, blogs and video sharing sites (e.g. YouTube). Although at their most interactive, these types of sites encourage anyone to make c omments on the content and contribute to the community. The nature of blogs and video sites indicates that they are primarily vehicles for disseminating information generated by creator(s) of the content. In other words, an on going conversation is not a necessity for the existence or success of the third type. The Internet as Mass Medium Mass communication research generally accepts newspapers, radio, magazines and television as its objects of study (Morris, 1996). online world is an inevitable trend. In the current paper, online mass media entities include internet based broadcast programs, newspaper and magazine websites. Despite the pervasive online information systems, established media and publishing companies are still in charge of producing most mainstream news, public affairs and entertainment content on the Internet (Mansell, 2007). Digitalization of broadcast programming, newspaper and magazine content is the key of this phenomenon. Advances in technology, such as mobile wireless and 3G devices, made it common for people to listen to the radio, watch television, or read newspapers and magazines on
40 the Internet. The question arises as to what extent the mass media impact young r skills through the Internet. Other than social learning theory, a traditional framework in media studies that analyzes mass media effects is cultivation theory ( Gerbner, Gross, Elley, Jacksonbeeck, Jeffries Fox & Signorielli, 1977), which argues that fr equent media exposure cultivates a bias view of consumer reality. Cultivation Theory Developed by Gerbner team in the Violence Index, and later the Cultural Indicators Project during the 1970s and 1980s, cultivation analysis addresses broad questions abou & Davis, 2009). The underlying assumption of it is that mass media content, particularly that of television, is mostly stereotypical and inconsistent with the "real" world reality (Gerbner, Gross, Morgan & Signorielli This discrepancy between mass media content and the real world creates two different versions of reality (Lee, person is exposed to (Lee, 1989). In other words, heavier television viewers have beliefs abo ut the social reality that are more consistent with televised world. As a consequence, they live in a different reality and appeal to different sets of values and of past literature is that heavy viewers of mass media would receive more information 1977).
41 Cultivation scholars have provided correlational evidence for a cultivation effect of broadcast programming on a wide range of social perceptions, values and attitudes. These researchers found that heavy television viewing correlates positively with higher 1993; 1993). In regards to the cultivation effect of television exposure on consumer reality, Potter (1991) found a significant relation between the amount of television v iewing and perceptions of affluence. Shrum, Burroughs and Rindfleisch (2005) found that the amount of television viewing is positively related to level of materialism, and that attention paid to television programs moderates the cultivation effect of TV v iewing on materialism. Not only do individuals develop a perception of societal affluence by consuming mass media, they also unconsciously conduct social comparison between their own lives and the reality presented in mass media (Richins, 1995). This cou ld advertising in general. The growing body of research based on cultivation theory produces consistent results that support the assumptions of cultivation of mass media. Mass media (particularly television) have been considered one of the main agents in t he process of CS (source). The new trend we are witnessing is that more and more people watch television shows, read news and magazines, and even listen to radio programs on the Internet. As more information is disseminated through the Internet, the theory of cultivation may shift toward a direction in which audiences or
42 users are taking a more active role in determining how they are cultivated. In this case 2011). The curre nt paper focuses mainly on exposure to actual editorial content rather than advertising. Prior CS studies have typically operationalized mass communication as only advertising, however, some scholars have suggested that in order to better understand consu mer socialization, more focus needs to be placed on the influence of media content and programming itself (Roedder Internet based television content is defined as television programming that users stream at or download from the Internet as opposed to watch through a traditional TV set. Almost all television stations or networks in the U.S. have long started cross promoting their broadcast content online and offline (Chan Olmsted & Ha, 2003). Media consumers are able t o access various television content through numerous platforms, main websites to third party services such as Hulu.com. Even though the current interest is on television programs delivered via the Internet rather than network TV because the content is the same, not to mention that most television programs online have embedded commercials now (in stream advertising). Cultivation research usually uses the resulting relationships between amount of Signorielli, 1990). Therefore, cultivation t heory, along with Bandura's speculation about consumer learning of consumption related skills from media suggest Hypothesis 1
43 Hypothesis 1 : Frequency of watching Internet based television is positively related to (a) attitude toward online advertising, (b ) materialism and (c) emot ional attitude toward shopping (i.e. pleasure, arousal and dominance). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale o f materialistic attitudes Emotional attitude toward shopping is measured by the non verbal AdSAM PAD scale. One of the weaknesses of cultivation theory is that it assumes homogeneity of media content (Baran & Davis, 2009). Critics have called for the n eed to assess how 1993; Sotirovic, 2001; Segrin & Nabi, 2002). Therefore, other than using simple measures of total exposure, online television viewing will be broken d own into exposure (2005) (e.g. sports, news, entertainment, soaps, music video etc.) Research Question 1: How do online television program genres relate to a) attitude toward online advertising, b) materialism and c) emotional responses toward shopping? Although previous CS research considered television as the most influential in constructing consumer reality, other forms of mass media still play a role in this learni ng process. Past research has shown that portrayals in the print media also shape conceptions of social reality (Bandura, 2001; Heath, 1984; Siegel, 1958). Online text based media are operationalized as the counterpart of offline print media such as newsp aper and magazine, whereas Internet radio is web based radio station or programming.
44 Hypothesis 2 : Frequency of reading online newspaper is positively related to (a) attitude toward online advertising, (b) materialism and (c) emotional attitude toward sho pping (i.e. pleasure, arousal and dominance). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude towa rd shopping is measured by the non verbal AdSAM PAD scale. Hypothesis 3 : Frequency of reading online magazine is positively related to (a) attitude toward online advertising, (b) materialism and (c) emotional attitude toward shopping (i.e. pleasure, aro usal and dominance). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude toward shopping is measured b y the non verbal AdSAM PAD scale. Hypothesis 4 : Frequency of listening to online radio is positively related to (a) attitude toward online advertising, (b) materialism and (c) emotional attitude toward shopping (i.e. pleasure, arousal and dominance). At titude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude toward shopping is measured by the non verbal AdSAM PAD scale. The Internet as Interpersonal Medium Previous studies blame mass media, particularly television, for the erosion of interpersonal communication. Because time is considered to be finite, it has been suggested that spending time on TV watching means having less time to spend on interpersonal relationships such as getting involved with friends and family (Putnam,
45 1993, 1995, 2000). This phenomenon has, however, shifted due to the emergence of the Internet. The aforementioned evidence has demonstrated the effects of reference Lius, Davies & Terry, 2007). Prior to the Internet era, these reference groups impacted an individual mainly through face to face, telephone and mail. As more people began using the Internet as a major vehicle for communication, entertainment and information sharing, the reference groups have expanded beyond family, friends and other personal acquaintances to potential strangers who may impact on different ways (e.g. blogs, online review sites). Needless to say, face to face interaction is no longer an essential component for interpersonal influence (Robins, Pattison & Elliott, 2001). This is because in the online worl d, various reference agents can now exert personal influence in the form of digital content (Robins et al., 2001). As reasoned above, this paper classifies the online personal communication channels into the two categories based on their level of interac tivity (Rafaeli & Sudweeks, 1997). The most often used messaging services (IM), Social networking sites (e.g. Facebook), chat rooms or forums and email are typical high interactive communication tools. Popular interactive user generated video sharing sites (e.g. YouTube). High interactive c hannels Instant messaging is a form of synchronous and text based computer medi ated programs such as Instant Messenger (from AOL) and Windows Messenger (from Microsoft) (
46 daily chat with friends; to keep in touch with those whom one is not able to see in person; to offer and receive information; or simply because it is a more convenient way of communication ( Ramirez, Lin & Dimmick, 2004). Chat forums or chat rooms are real time, tex t based and synchronous means of communication between users of a community with a specific theme (Kaye & Johnson 2004). Different to IM system that is based on independent software or programs and is private in nature, chat forum functions in the format of a website. Motivations for using online chat forums include information seeking, entertainment, social utility and convenience (Kaye & Johnson 2004). Social networking websites (SNSs) (e.g. Facebook, MySpace, Twitter, and LinkedIn) are one of the m ost popular online activities for young consumers. The main purpose of social networking sites is to allow users to create and maintain a network of Murchu Breslin & Decker 2 004 ). As a relatively new phenomenon, SNS communities have facilitated and enhanced the diffusion of information by online users (Xie, 2011). Electronic mailing (email) is the Internet based system that facilitates message exchange and storage Just as w ith the traditional mailing, individuals use email to communicate with family, friends, and colleagues while marketers use it to reach customers ( Phelps, Lewis, Mobilio, Perry & Raman 2004). The most popular email applications are Mac Mail and Outlook. Add itionally, popular third party email services include Gmail, Hotmail, and Yahoo Mail. Hypothesis 5: Frequency of communicating with others about consumption related subjects via high interactive online platforms (i.e. chat forums, social networking sites, email and instant messages) is positively related to (a) attitude toward online
47 advertising, (b) materialism and (c) emotional attitude toward shopping (i.e. pleasure, arousal and dominance). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude toward shopping is measured by the non verbal AdSAM PAD scale. Low interactive c hannels Weblog (blog) is a web ba sed medium through which users publish online entries that are chronologically maintained in an individualized journal writing format ( Kim, 2009). Characteristics of blogs include their dynamism, single theme focus, reverse chronological presentation and d ominant use of the first person ( Jung, Youn & Mcclung, 2007). Although blog readers are highly encouraged to make comments on blog posts and interact with the blog owner as well as other people, they typically act as receivers of information but do not co ntribute to site content. User generated video sharing websites (VSW) are sites where individuals are able to upload, watch and share video clips (Pasche, 2008). Different to Internet based TV content sites that are generated and controlled by media compa nies or third party services, VSWs allow everyone to produce and post videos to share with other individuals. The prevalence of user generated video sharing sites has an extraordinary impact on Internet traffic. The most dominant video sharing site, YouT ube along has 800 million unique visitors each month, and over 3 billion YouTube videos are viewed daily (YouTube Statistics, 2012). In addition to blogs and user generated video sites, online reviews also have a great impact on consumer socialization. V arious online review sites create a venue for
48 users to post and read opinions or recommendations about an item listed there. These recommendations may consist of textual reviews and/or star ratings (Heath, Motta & Petre, 2006). Chatterjee (2001) claimed that online consumer reviews are an important source of electronic word of mouth communication, as these reviews have a substantial Meyer, 2009). In fact, many consumers s earch for and read online reviews as the first step in shopping (Park & Lee, 2009). Hypothesis 6 : Frequency of communicating with others about consumption related subjects via low interactive online platforms (e.g. online reviews and blogs) is positively related to (a) attitude toward online advertising, (b) materialism and (c) emotional attitude toward shopping (i.e. pleasure, arousal and dominance). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in ge neral. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude toward shopping is measured by the non verbal AdSAM PAD scale. Consumer Susceptibility to Influence on the Internet The magnitude of the influence varies ac interpersonal influence. Specifically, such reference group impacts are greater for those with internal control orientations (O rth & Kahle, 2008). This particular individual Bearden, Netemeyer and Teel (1989) defined consumer susceptibility to interpersonal influence (CSII) as: 1) and willingness to conform to the expectations of other individuals through the purchase of market goods and brands, and/or 2) the tendency to gain knowledge of the marketplace by observing others and/or
49 seeking information from others. Consumers that ar e highly susceptible to personal influence are generally more likely to be influenced by others in making marketplace decisions than those who are less susceptible (Schroeder, 1996; Lalwani, 2002; Yang & Cho, 2000). This paper employs the CSII lens to det susceptibility to personal influence on the Internet, and whether susceptibility to influence moderate the impact of various Internet use s on consumer socialization. Given that online communication tools such as Facebook are specifically designed to facilitate and enhance social or personal communications (Stefanone & This paper proposes that CSII acts as an interaction variable in the r elationship between susceptibility to social influence. The hypotheses are formally articulated after the explanation of two types of personal influence. In a study of social influence upon individual judgment, Deutsch and Gerard (1955) first split interpersonal influence into two dimensions, namely, the normative influence and informati ve influence. Susceptibility to normative influence (SNI) reflects the tendency to conform to others expectations, while susceptibility to informational influence (SII) reflects the tendency to trust and internalize information obtained from others (Beard en, Netemeyer & Teel, 1989, 1990; Childers & Rao, 1992; Deutsch & Gerard 1955; Mangleburg & Bristol, 1998). In other words, one may be susceptible to the influence from others either because of the information others provide, or because others may reward positive behaviors and help one gain positive self identities.
50 recipients search for info rmation from knowledgeable others or when they learn and internalize marketplace information (Park & Lessig 1977). Previous research has found that informational influence may affect consumer decision regardi ng product evaluations (Burnkrant & Cousineau 1975; Pincus & Waters 1977), product/brand choices (Park& Lessig 1977), and attitudes towards advertising ( Mangleburg & Bristol, 1998). Copious literature on compliance and conformity research indicated that i ndividuals have a tendency to change their attitude and behavior in order to conform to the expectations of significant others (Burnkrant & Cousineau, 1975; Cialdini & Brist ol, 1998). Contrary to informative influence, which happens through a process of internalization of obtained information as objective reality, normative influence occurs when one form attitudes or make decisions simply to conform to the expectations of o thers in order to gain approval and avoid punishments (Bearden et al., 1989, 1990; Burnkrant & Alain, 1975; Childers & Rao, 1992; Deutsch & Gerard 1955). Hypothesis 7 : Consumer susceptibility to informative influence moderates the relationship between the frequency of online communication about consumption matters and (a) attitude toward online advertising, (b) materialism as well as (c) emotional attitude toward shopping (i.e. p leasure, arousal and dominance). In other words, the strength of these relati onships influence on the Internet. These relationship s should be more salient amo ng those who have high relative to low levels of consumer susceptibility to informative influence. Hypothesis 8 : Consumer susceptibility to normative influence moderates the relationship between the frequency of online communication about consumption matters and (a) attitude toward online advertising, (b) materialism as well as (c) emotional attitude toward shopping (i.e. pleasure, arousal and dominance). In other words, the
51 strength of these relationships normative influence on the Internet. These relationship s should be more salient amo ng those who have high relative to low levels of consumer susceptibility to normative influence. With regards to comparing the effect of informative influence versus normative influence, in a study using privacy protecting behavi ors, Moscardelli and Divine (2007) found that informative influence has a stronger impact than normative influence. With a sample of young adults, Zhou (2009) also found that people received more informative influence from four socialization agents than n ormative influence. Moschis (1987) argues that with increasing age, individuals acquire consumption skills more through observation rather than compliance to others. In other words, they rely more on informative sources of influence when it comes to learn ing. Moreover, the Millennial generation, as young adults, is currently in a life stage where they focus the most on personal independence and self development as they step in the frontier of adulthood ( Arnett, 2003; Rumbaut, 2005; Zhou, 2009). In this ph independence and to become self sufficient are the most key aspects of self cognition (Arnett, 2007). Therefore, as they become more individualistic in the early adulthood, young adults may be mor e likely to learn and internalize information as evidence about reality from knowledgeable others (informative influence), than to form attitudes and decisions just because they want to conform to the expectations of others (normative influence). Hypothes is 9: the Millennials are more susceptible to informative influence than to normative influence, measured with the verbal scales of susceptibility to personal influence.
52 Social Structural Variable s While it seems reasonable to assume that living in the so cial environment, everyone should acquire such consumer skills, knowledge, and attitudes, their level may vary across individuals as a result of differences in some social constructs (Moore & Stephens, 1975). The social learning theory considers social st ructural composition such as age, gender, ethnicity, social class, income etc. as antecedent variables of social learning. Previous studies have suggested that these variables may both directly and ill, 1978; Moore & Stephens, takes place, these social structural variables are critical in explaining learning outcomes zation studies (e.g. Bush et al., 1999; socialization that may occur as a direct or indirect consequence, of gender, age, ethnicity and socioeconomic status These str uctural variables have been used in previous CS literature (Bush, Smith & Martin, 1999; Moore & Stephens, 1975; may differ with respect to family income and socioeconomic back ground (Katz, 1964). consumer learning (Rindfleisch, 1994; Rindfleisch et al., 1997). From a learning theory standpoint, CS scholars seem to agree that income level is posi tively related to the amount of experience that people have with money and affluent goods (Moschis and turn, the amount of experience with their own wealth may impact their perc eptions of
53 socioeconomic backgrounds may have greater awareness of, and preference for, commercial stimuli in their life than their lower class counterparts (Moschis & Moore, 1979). Following this line of reasoning, this dissertation proposes that socioeconomic assume that individuals who have more money to spend (disposable income) will h ave more experience and hence more knowledge in the marketplace (Moscardelli & Liston Heyes, 2005). Hypothesis 10 socioeconomic status is positively related to (a) attitude toward online advertising, (b) materialism and (c) emotional attitude towar d shopping (i.e. pleasure, arousal and dominance). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude toward shopping is measured by the non verbal AdSAM PAD scale. Moschis (1987) has long recognized the variation between men and women in regards to their communication styles and mass media usage patterns. For instance, they may have different preferenc es for television or print media. Another difference lies in their social communication objectives (Daugherty & Zhang, 2009). Sociolinguistics researcher Tannen (1990, 1995) stressed men's communication objectives of problem solving and social standing, and women's goals of achieving emotional support. One would expect that these differences carry over into online communities. In a study of gender difference in electronic word of mouth, Awad and Ragowsky (2008) found that females primarily use the Inter net to gain a sense of social
54 support and that the effect of online trust, in intention to shop online is stronger for women than for men. Therefore, female consumers should be more susceptible to impact of Internet as interpersonal medium on various socialization outcomes may be more pronounced mavens, the results showed that market mavens were more likely than nonmav ens to be female, and that market mavens tend to use mass media more than nonmavens. about many kinds of products, places to shop, and other facts of markets, and initiate d iscussions with consumers and respond to requests from consumers for market possess higher consumer skills, because they frequently communicate about consumption matters w ith others and are more exposed to mass media. Hypothesis 11 : Female respondents score high er than male respondents on (a) attitude toward online advertising, (b) materialism and (c) emotional attitude toward shopping (i.e. pleasure, arousal and dominanc e). Attitude toward online advertising is measured with a verbal scale of attitude toward advertising in general. Materialism is measured with a verbal scale of materialistic attitudes Emotional attitude toward shopping is measured by the non verbal AdSAM PAD scale. associates positively with greater skill in using various consumption related information sources (Moschis, 1978), more accurate evaluation of product (Ward et al. 1977), stronger brand preferences (Guest 1944 ) and higher tendency to evaluate a product by brand name (Moschis & Moore, 1979). The
55 specific lifetime span examined here is the current Millennials generation (Gen Y). At the time of writing (2011), this group of young consumers is aged between 21 an d 31 years old (e.g. Chu & Kamal, 2011; Pew Research, 2010). This dissertation uses age As one grows older, he/she becomes a more sophisticated consumer. But does being a attitudes and higher emotional responses towards shopping, or vice verse? Although the above studies suggest a positive relationship between age and direct consumer ski lls, some evidence instead finds a negative relationship when applied to indirect consumer skills, such as general attitude towards advertising (Moore & Stephens, 1975). Therefore, this dissertation posts the following research question: RQ 2 : How is age related to (a) attitude toward onlin e advertising, (b) materialistic values and (c) emotional attitude toward shopping ? 2007). In the general field of consumer research, ethnici ty has been recognized as an important segmentation variable (e.g. Appiah, 2001; Cui, 1997; Hirschman, 1981, Kern Foxworth, 1991; Kim & Kang, 2001). However, e thnicity or race as a social construct A few CS studies that examined ethnicity as a social structural variable have proved that ethnicity or race is a major determinant of susceptibility to personal influences (Kang & Kim, 2001; Singh, Kwon & Pereira, 2003) and attitudes toward advertising ( Bush, Smith, & Martin, 1999). To build on many of the demographic factors used in previous CS literature, this research also consider ethnicity as a potential antecedent variables. RQ 3. How is ethnicity (a) attitude toward onlin e advertising, (b) materi alistic values and (c) emotional attitude toward shopping ?
56 CHAPTER 3 METHOD Research Design In line with most empirical research on consumer socialization that utilized survey data to capture the influence of socialization agents (e.g., Bush et al., 1999; Churchill & Moschis, 1979; Mangleburg & Bristol, 1998; Moscardelli & Liston Heyes, 2005; Moore Shay & Berchmans, 1996; Rindfleisch, Burroughs & Denton, 1997; Shrum et al., 2005), this study also employs survey method to collect data. In a study of the im pact of advertising and media on material values, Richins (1987) pointed out that answers to causal questions in the area of media effects remain elusive because of the inadequacy of experimental methods for assessing long term causal effects (Pollay, 1986 ). She further recommended that correlational survey studies would be more appropriate. Sampling According to Kerlinger (1978, 1986), the ratio of sample size to number of items should be at least 10:1. Thus, the author determined the sample size for th is dissertation by multiplying observed variables to be used in the current model by 10. As a result, with 58 items or observed variables in the survey, the sample size should be at least 580 As the data base for the current study, 693 qualified questi onnaires were collected via the following two approaches between August 1st and November 15th, 2012 student sample and Amazon Mechanical Turk sample. Prior studies have used different recruitment methods to collect survey data ( Rindfleisch et al., 1997; Tang, Wu, Yeung & Yan, 2009; Valentova, Rieger, Havlicek, Linsenmeier & Bailey 2011 ). The data for this dissertation initially came from a sample of undergraduate and range 21 and 31 years old (e.g. Ch u
57 & Kamal, 2011; Pew Research, 2010) These students are currently enrolled in various majors from three colleges of a large southern state university. Subjects received extra credit in their undergraduate and graduate level classes as incentive to volunta rily participant. This dissertation used college students for a number of reasons. First, consumer behavior research has found students to be valuable subjects for conceptual model building (Bush et al., 1999; Calder, Phillips & Tybout, 1981; Petty & Cac ioppo, 1996). Additionally, current undergraduate and graduate students represent a significant portion of the Millennial generation, which is the consumer group of interest for this dissertation. Finally, university students have a high adoption rate of Internet usage (Xie, 2011, 2012). P rospective respondents were screened for Internet usage because the present research is only interested in Internet users. Given that the initial survey distributed among college students resulted in a low number of re spondents from upper range of the targeted age group (i.e. 26 31), the author conducted a second sampling of individuals by using a paid crowdsourcing tool the potential cont ributions of AMT to social science research, particularly its values to survey recruitment and data collection. Their findings indicate d that a) participants recruited from AMT are significantly more demographically diverse than standard college student s amples; b) AMT respondents are at least as diverse and more representative of U.S. population than those of typical Internet samples; and c) the data obtained are at least as reliable as those obtained via traditional approaches. In sum, they recommended the Amazon Mechanical Turk service to be used to obtain high quality data in a low cost way (Buhrmester et al., 2011). Recent marketing research
58 studies have also adopted AMT service as a mean to recruit respondents (e.g. Loewenstein, Raghunathan & Heath, 2011; Sussman & Olivola, 2011). A total of 1,200 participants were recruited through AMT service, among which, 395 of them were qualified, and hence received payment for their participation. To examine the potential sampling method related differences be tween the respondents recruited via two different sampling approaches (i.e. respondents recruited from campus and respondents recruited from AMT service) the author first compared the means and standard deviations of each dependent and independent variabl e between the two sample groups with the Independent Samples T tests Overall, the respondent profiles appear ed to be remarkably similar. Differences between means of each variable across two sample methods are not statistically significant with almost a ll p values greater than 0.05. The only significant sample method related difference was the mean level of communication via low interactive tools such as blogs and review sites (College = 3.8, AMT = 4.3, t (691) = 4.69, p < .001). This alone however, does not justify the heterogeneity of the two sample methods. In addition, a multivariate analysis of variance (MANOVA) was conducted to examine the potential effects of using two different sampling approaches on three dependent variables (i.e. materiali sm, attitude toward advertising and emotional responses toward shopping). This method was adopted f rom Rindfleisch et al. (1997) to determine whether sampling approaches ma ke a difference in key measures. The e hypothesis of homogeneity (i.e. variance covariance matrix for the dependent variables was equal across two groups) cannot be M = 15.29, p = .125. Hair et al. (2009) suggested that p values greater
59 than .01 indicate no differences betwe en groups. Therefore, there did not appear to be any significant response differences between the two sampling approaches. G iven that no significant response differences were found between the student sample and the AMT sample, the author collapsed acros s them and conducted the following results based on the combined sample. Of the 693 respondents, female represented 62.9% of the sample (n=436) and male participants represented 37.1% (n=257) The average age of participants was 24.4 ( Standard Deviation =3.27). More than half of the respondents were Caucasian ( 71.1%) followed by Hispanic or Latino (15.4%), Asian (10%), African American or Black (9.2%) etc. Measures This research used a web based survey because it was justified as one of the most efficie nt and convenient ways to reach Internet users today (Xie, 2012). Existing scales was used and slightly adapted according to the particular hypotheses and research questions of the current study. In order to refi ne the questionnaire, a pretest was conduct ed using a similar group of young adults (n = 60) enrolled in a large southwest university The survey instrument was then modified based on the pretest response. This survey utilizes a 58 item questionnaire to collect data. Table 3 1 presents a summary of the operationalization and measurement of the major variabl es used in the present research. Dependent Variables (Consumer Learning Skills) Attitude towards Internet advertising This scale consisted of three, seven point semantic differential pairs ( favorable/unfavorable, good/ bad, and pleasant /unpleasant). Participants were asked
60 to evaluate their overall attitude toward Internet advertising based on the three items used by MacKenzie and Lutz (1989) L evel of materialism This scale was indexed b y averaging responses to nine items like I admire people who own expensive homes, cars, and clothes ." The responses range d from strongly disagree (=1) to strongly agree (=7). This scale was adopted from the nine item scale (MVS) (2004). Although previous research item version and six item version scales (Shrum et al., 2005; Yang & Oliver, 2010), Richin s (2004) cautioned against using the three and six item version of the MVS, while claimed th at the nine item scale correlates as strongly with criterion variables, on average, as the longer versions of the scale. In regards to internal and extern al criteria, the scale possessed satisfactory levels of validity (r=.36) and reliability, with a coef ficient alpha of 0.84 ( Richin s 2004). Emotional responses towards consumption This scale was measured using AdSAM (Morris et al., 2002), an attitude self assessment manikin that was developed based on the SAM scale. SAM has been used successfully in a number of previous studies as a well validated self report dimensional assessment device ( Bradley & Lang, 1994; Lang, Greenwald, Bradley & Hamm 1993; Morris et al., 2008). This is a visual measure of emotional response based on the three dimensional PAD ( Pleasure, Arousal and Dominance) theory developed by Mehrabian and Russell (1974) as mentioned in the literature review. Prior studies have suggested that compared to verbal techniques, SAM is a better tool in that it does not require respondents to cogni tively translate their feelings and subconscious thoughts into words (Lang 1985; Morris & Waine 1993; Morris et al., 2008). The correlations between SAM
61 and Mehrabian and .660 for dom inance (Morris, Bradley, Sutherland, & Wei, l993; Morris, Bradley, Lang, & Waine, 1992; Morris & Waine, l994). This indicates that SAM is a reliable measure for emotional responses. AdSAM a research tool that utilizes a database of 165 emotional adject ives, was developed to measure emotions that consumers may experience in certain situations (Morris et. al. 2002; Stewart, Morris & Grover, 2007). AdSAM has been used in a great amount of studies and has been proven to be a valid and reliable tool to ch aracterize emotional responses to various situations ( Goodman, Morris & Sutherland, 2008; Greenwald, Cook & Lang, 1989; Lang et al., 1993; Morris et. al., 2003, 2008; Morris & Waine, 1993; Morris & Boone, 1998; Xie, 2011, 2012). The measurement instrumen t for ER that researchers most widely use is based on statements that are either product specific or situation specific (Havlena & Holbrook, nal context, this paper measured emo tional responses towards consumption by providing consumers two shopping environment (online shopping and mall shopping) to increase the specificity. Like its predecessor SAM, the AdSAM measure (see Figure 3 1 ) consisted of graphic characters arrayed alon g three continuous nine point scales: Pleasure (measures how positive/negative the subject feels towards the situation ), Arousal (measures involved in the feeling the subject is) and Dominance (measures how empowered the respondents feel) Participants re spond ed
62 by rating their levels of Pleasure, Arousal, and Dominance along the graphic continuum. Specifically, they click ed on a dot below or b etween the manikins in each scale. Independent Variables (Web based Socialization Agents) Due to the cross sectional nature, almost all consumer socialization studies capture d the effect of socialization agents by examining the amount of exposure to diffe rent media as well as the frequency of personal communication regarding consumption matters (e.g. Bush, Smith & Martin, 1999; Lueg & Finney, 2007; Moschis & social learnin g theory had demonstrated t he theoretical justification for using frequency of interaction to capture this complex process (e.g., Churchill & Moschis, 1979; Gerwitz, 1969; Moschis &Moore, 1979 ) For instance, theory of instrumental learning or o perant co nditioning suggested that learning resulting from television commercials is a function of frequent pairing of pleasant outcomes with purchasing advertised products (or unpleasant outcomes with failure to use advertised products) (Churchill & Moschis, 1979; Bandura, 1971; Skinner). In line with the literature, the present research conceptualized this process in other traditional media consumpt ions, the questionnaire included questions related to consumption of online media content (i.e. Internet based television programs, radio, newspaper and magazine sites), as well as social interaction on the Internet (i.e. i nstant messaging, social networking sites, email, blogs, review sites and video sharing sites).
63 Usage of online television, radio, newspaper and magazine The general use of the Internet as mass media were based on responses to how frequently the participa nts use particular medium on the Internet (i.e. Internet radio, magazine, newspaper and web based television). The responses ranged from 1 = never through 5 = 2 3 times a month to 7 = daily ( stman, 2012). In order to capture the effect of various telev ision genre types, the respondents are also asked to report the frequency with which they watch specific programs on the Internet (i.e. soap operas, reality shows, news programs, sports, documentary, entertain ment). These categories were adapted from Tigge genre types. Interpersonal communication about consumption on the i nternet This was operationally defined as personal interaction on the Internet concerning goods and services (Bush, et al., 1999; Moore et al. 1975; 1977; 1978; M oore & Moschis 1978; Moschis & Moore 1978). The scales were a modification of previous scales for peer and family communication about consumption (Bush, et al., 1999 ; Lueg & Finney, 2007 ; Moschis, 1981 ). Based on the three items used in the literature, p articipants evaluate d their level of interpersonal communication via different online applications. The response options range d from strongly disagree (=1) to strongly agree (=7). As indicated in Chapter 2 the six online communicati on methods included h igh interactive channels (i.e. instant messaging, social networking s ites, email chat forums) and low interactive channels (i.e. blogs, user generated video sharing and review sites ) Example items are S
64 were used in order to capture more completely the richness of onl ine interpersonal communication construct. Individuals nowadays not only interact with their circle of friends but also by referent others, and even strangers (e.g. on review sites) ( Lueg & Finney, 2007). Moderator (Normative/informative susceptibility t o personal influence ) Regarding consumer susceptibility to personal influence, t his measurement followed previous CSII scales (Bearden et al., 1989; Schroeder, 1996). According to the theory, t he scale was divided i nto two separate elements informationa l dimension Susceptibility to nor mative influence scale consisted of eight items (e.g. I rarely purchase the latest fashion styles until I am sure my friends approve of them), whe reas susceptibility to info rmative influence scale consisted of four items (e.g. I frequently gather information from friends or family about a product before I buy). The two dimensions correlated differently with other factors such as self esteem and att ention to social comparison ( Schroeder, 1996). Participants respond ed to a seven point scale anchored with strongly disagree and strongly agree. The authors reported test retest reliability as 0.75 and 0.79 for the informational and normative scales, res pectively (Bearden et al., 1989). The inter correlation between the two factors was.44, and alpha reliabilities were 0.82 and 0.88 for the informational and normative factors (Bearden et al., 1989). Before the tests of the moderating effects of consumer susceptibility to interpersonal influence (normative and informative) on the relationships between frequency of consumption related communication and all dependent variables, both the independent variables (i.e. frequency of online communications about con sumption
65 matters) and the moderator were mean centered in order to minimize effects of multicollinearity in the presence of interaction terms (Cronbach, 1987). Social Structural Variables is me asured based on a com bination of three dimensions: (a) family annual income level, (b) father's education, and (c) mother's education. This measure was adopted from prior consumer research (Ahuvia & Wong, 2002; Rindfleisch et al 1997). Family income was ascertained by aski ng respondents to choose a category representing the total family income they had last year. The categories were $20,000; $20,001 $40,000; to $140,001 educ ational attainment was was used here. The three responses were then com bined into an overall measure of SES by averaging subjects' summed standardized scores for these measures (Rindfleisch et al., 1997). For the sample of young adults s ocial class is seen as the best indicator of their socioeconomic background. Prior studies have also demonstrated that using parental education is a reasonable approach 2002; Rindflei sch et al 1997). Age is measured with an open ended question that asked respondents to report their age at last birthday. Gender is a dichotomous variable (Male =1; Female=2). Cronbach's alpha is used to assess the reliability of these scales. As re alpha
66 value higher than 0.60 is acceptable. Ranging from lowest to highest, t he reliability coefficients are 0.64 for socioeconomic status 0.78 for consumer susceptibility to informative influence, 0.86 for consumption related communication on chat forums, .87 for consumption related communication via email, 0.87 for materialism, 0.89 for consumption related communication on product review sites, 0.90 for both consumption rela ted communication through instant messaging and through social networking sites, 0.92 for consumer susceptibility to normative influence, 0.93 for attitude toward online advertising, 0.94 for consumption related communication on blogs and 0.95 for communic ation about consumption via online video sites. Table 3 2 presents descriptive statistics for all major variables, including means, minimum, maximum and standard deviations. Regression Models Model Summary R egression analysis was performed to test the h ypothesized effects. Researchers use regression models to analyze the relationship between several factors and a single outcome variable. In a regression model, one or more independent variables are used to predict a single dependent variable (Hair et al ., 2009). Moreover, t his follows the procedures that hav e been used in previous studies (e.g. Moschis & Churchill, 1978; Moschis & Moore, 1979; Moscardelli & Liston Heyes, 2005; Moore Shay & Berchmans, 1996 ) and ng theoretical consumer socialization framework According to the consumer socialization literature (e.g. Churchill & Moschis, 1979; Mangleburg & Bristol, 1998; Moore Shay & Berchmans, 1996; Rindfleisch, Burroughs & Denton, 1997; Shrum et al., 2005 ), the
67 main elements in the CS models should include social structural variables socialization processes (interaction with CS agen ts) and socialization outcomes. This dissertation applies two multivariate model s to assess its primary research hypotheses : ordina l least squares regression (OLS) and an ordered logistic regression. They address ed whether and to what extent the proposed online socialization agents influence d related attitudes and values. More specifically, both OLS and ordered log it models regress ed the depende nt variables (i.e. materialism, emotional responses towards shopping and attitude towards online advertising) on online mass media usage and communication about consumption matters (high interactive and low interactive), in addition to antecedent demographi c or the social structural variables Both attitudes and materialism were run using an OLS or linear regression models given that they are continuous dependent variables. To determine whether an OLS model is appropriate, t his dissertation first examined and corrected all five assumptions as dictated by the literature (Hair et al., 2009) These include (a) linearity, (b) normality of the erro r term, (c) homoscedasticity and (d ) non autocorrelation Multicollinearity was also conducted before the actual analysis. The remaining six dependent variables emotional responses toward online shopping (pleasure, arousal and dominance) and emotional responses toward mall shopping (pleasure, arousal and dominance) were r un using an ordered logit model. B ecause the emotional response to shopping measure is a discr ete and ordered scale ordered logit analysis is considered more appropriate than ordinary least squares regression ( e.g. Hair et al., 2009; Kennedy, 1998 ; Rego, Billett & Morgan, 2009 ). To
68 ensure confidence, tests for all four OLS assumptions were run for th ese six models ; all of them lacked the goodness of fit met by the ordered logit model. fluence and normative influence will moderate the relationship between the frequency of online interpersonal communication about consumption matters and three learning outcomes ( DVs). Thus, there are two sets of interactive terms namely: consumption rela ted online communication X susceptibility to informative influence; consumption related online communication X susceptibility to normative influence. For each hypothesized interaction effect, the analysis regress ed all DVs separately on online media usage and frequency of communication about consumption via high/low interactive online media, while include d the interaction terms between consumption related communication variables and each of the two expected moderators. In addition to compare usceptibility to informative influence and their susceptibility to normative influence (Hypotheses 9), a t test will be conducted. Model Specifications The following include the regression equations of this dissertation to assess its primary research obje ctives: Ordinal least squares regression DV1: Attitudes towards online advertising DV2: Materialism
69 IV1: Amount of online television viewed IV2: Online radio listenership IV3: Online newspaper readership IV4: Online magazi ne readership IV5: Consumption related communication via high interactive media IV6: Consumption related communication via low interactive media IV (ma): Consumer susceptibility to normative influence (moderator a) IV (mb): Consumer susceptibility to infor mative influence (moderator b) Ordered logit r egressions DV 3 = Pleasure online DV 4 = Arousal online DV 5 = Dominance online DV 6 = Pleasure mall DV 7 = Arousal mall DV 8 = Dominance mall
70 Table 3 1. Scale d escriptions Index Operational Definitions M easure Source Materialism Tendency of emphasizing possessions and money for personal happiness and social progress 9 item, 7 point agree disagree scales. expensive homes, cars, and clothes Richin (2004) Attitude towards Online Advertising Cognitive orientations concerning liking of Internet advertising in general 3 item, 7 point semantic differential scale (favorable/unfavorable, good/ bad, and pleasant /unpleasant) MacKenzie & Lutz (1989) Emotional Responses towa rds Shopping Affective orientations concerning feelings about shopping situations Non verbal SAM scale measuring levels of Pleasure, arousal and dominance Morris et al., 2002 Use of Internet as Mass Media A. Online Television Viewed Self reported avera ge time spend watching television program categories on the Internet in a typical day Open end responses to questions about time spent online viewing specific (1997) B. Online New spaper Readership Self reported average time spend on newspaper websites in a typical day Open end responses Moschis & Churchill, 1978; Moore & Stephens, 1975 C. Online Magazine Readership Self reported average time spend on magazine websites in a typica l day Open end responses Moschis & Churchill, 1978; Moore & Stephens, 1975 D. Online Radio Use Self reported average time spend listening to radio programming on the Internet in a day Open end responses Consumption related Communication via High Inter active Channels Frequency of interactions about brands, products and services through high interaction tools on the Internet 3 item, 7 point agree disagree scales. Example Item Networking Sites and I often talk about product we wan Bush, Smith & Martin, 1999; Lueg & Finney, 2007 ; Moschis, 1981 Consumption related Communication via Low Interactive Channels Frequency of interactions about brands, products and services through low interaction tools on the Internet 3 item, 7 point agree disagree scales. Example Item I often seek out the advice of people regarding which brand to buy via Bush, Smith & Martin, 1999; Lueg & Finney, 2007 ; Moschis, 1981 Consumer Susceptibility to Personal Influence A. Susceptibility to Informative influence marketplace by observing others and/or seeking information from others. 4 item, 7 point agree disagree scales. Example Item information from friends or family about a Bearden et al., 1989 B. Susceptibility to Normative influence expectations of other individuals through the purchase of market goods and brands. 8 item, 7 point agree disagree scales. Example Item fashion styles until I am sure my friends Bearden et al., 1989
71 Table 3 2. Descriptive statistics for key v ariables Variables Mean Minimum Maximum Standard Deviation Attitude towards Online Advertising 3.42 1.00 7.00 1.55 Materialism 4.18 1.00 7.00 1.29 Emotional Response towards Online Shopping Pleasure online 6.76 1.00 9.00 1.44 Arousal online 5.35 1.00 9.00 1.99 Dominance online 6.49 1.00 9.00 1.91 Emotional Response towards Mal l Shopping Pleasure mall 6.06 1.00 9.00 2.11 Arousal mall 5.92 1.00 9.00 2.11 Dominance mall 5.60 1.00 9.00 2.13 Online Television Viewed 4.95 1.00 7.00 1.83 Online Newspaper Readership 2.96 1.00 7.00 1.43 Online Magazine Readership 2.39 1.00 6.78 1.37 Online Radio Use 3.76 1.00 7.00 2.30 CCH 3.31 1.00 6.67 1.31 CCL 4.11 1.00 7.00 1.47 Consumer Susceptibility to Personal Influence CSII 5.11 1.00 7.00 1.14 CSNI 3.01 1.00 6.88 1.29 SES 3.48 1.00 7.00 1.45 CCH: Frequency of consumption related communication via high interactive platforms CCL : Frequency of consumption related communication via low interactive platforms CSII: Consumer Susceptibility to Informative Influence CSNI: Consumer Susceptibility to Normative Influence SES Socioe conomic Status
72 Figure 3 1. The AdSAM visual s cale Pleasure Arousal Dominance
73 CHAPTER 4 RESULTS Bivariate Results Following previous research (e.g. Bush et al., 1999; Moschis & Churchill, 1978), this paper first examined individual relationships between three selected con sumer skills and the socialization agents to assess each Hypotheses 1 through 11 Other than zero order correlations, partial correlations were also estimated in order to control statistically the effects of other independent variables (Hair, 2009; Bush et al., 1999; Moschis & Churchill, 1978) The relationship between consumer skills under consideration and the socialization variables are reported in Table 4 1 The Internet as Mass Medium Relationship between frequency of viewing online television and consumer socialization (CS) outcome variables The relationships between online television viewing and three dependent variables (attitudes toward online advertising, materialism, emotional responses to shopping) were almost all not si gnificant. Although t here seemed to be a significant but weak relationship between online television viewing and arousal toward online shopping (r = .08, p < .05), the correlation between the two variables became not significant after controlling the explanatory variables, sug gesting that H1a, H1b and H1c were not supported. To answer the Research Question one, this study correlated each type television genre viewing with all dependent variables. Only three genre types have positive correlation with attitude toward online adv ertising, namely, documentaries (r = .095, p < .05), drama (r = .086, p < .05) and home/leisure (r = .077, p < .05). The other
74 socialization outcome variables (materialism and emotional responses to shopping) however, did not have any significant relati onship with television genre types. Relationship between frequency of reading online newspaper and CS outcome variables Although the zero order correlations between online newspaper readership and three dependent variables were not significant the relati onship between newspaper readership and attitude to advertising became significant after removing the effect of other variables. Nevertheless, this significant relationship was not positive as hypothesized, suggesting that the data did not support H2a, H2 b and H2c. Relationship between frequency of reading online magazines and CS outcome variables H3a and H3b were both supported by significant and positive correlations between online magazine readership and (a) attitude toward online advertising (r = .20 p < .001) and (b) materialistic values (r = .13, p < .001). Although slightly dec reased, the significance remained after controlling the impacts of other independent variables. Regarding the relationship between online magazine readership and emotional responses toward online shopping and mall shopping, the data indicate d that only pleasure had a positive relationship with online magazine readership (r = .16, p < .001 for pleasure to online shopping and r = .13, p <. 01 for pleasure to mall/store shoppi ng). Both the zero order correlations and the partial correlations indicate d that online magazine readership did not have a significant relationship with emotions of arousal and dominance toward shopping, and thus H3c was only partially supported.
75 Relat ionship between frequency of listening to online radio and CS outcome variables Only the emotions of pleasure to online and mall shopping had statistically significant relationship with online radio listening. However, contrary to the h ypotheses the data suggest ed a negative relatio nship between the two variables As a result H4a, H4b and H4c were disconfirmed The Internet as Interpersonal Medium R elationship between consumption related communication via high interactive tools and CS outcome variables The correlations between CCH and (a) attitude toward online advertising (r = .23, p < .001) and (b) materialistic values (r = .25, p < .001) were significantly positive. Although slightly decreased, the relationships remain ed after controlling the impact s of other independent variables, suggesting H5a and H5b were both supported. H5c was partially confirmed given that both emotions of pleasure and arousal to online shopping were positively and significantly related to CCH (r = .23, p <.001 and r = .18, p < .001 respectively). R elationship between consumption related communication via low interactive tools and CS outcome variables The zer o order correlations between CCL and (a) attitude toward online advertising (r = .12, p < .001) and (b) materialistic values (r = .18, p < .001) were both significant. However, the partial co rrelations between CCL and the two dependent variables were not statistically significant to confirm H6a and H6b. Similar to communication about consumption via high interactive ch annels, CCL had a positive and significant relationship with emotions of pleasure and arousal to online shopping (r = .22, p <.001 and r = .17, p < .001 respectively). Even though these relationships were
76 slightly reduced when hold constant the effects of other independent variables, the result still pr ovided partial support for H6c. Hypothesis 9 stated that the Millennials were more susceptible to informative personal influence (CSII) than to normative personal influence (CSNI) In other words, the pape r proposed that the respondents would have higher s usceptibility to informative influence than to normative influence. A paired sample t test (t (692) = 36.85, p < .001) revealed that participates in this study scored significantly higher in CSII (Mean=5. 11, SD=1.13) than in CSNI (Mean = 3.01, SD = 1.29). Therefore, H9 was supported. The moderating effects of CSII and CSNI will be tested in the following section ( Multivariate Model Testing). Social Structural Variables Relationship between s ocioeconomi c status and CS outcome variables did not seem to have significant relationship with any of the dependent variables. Hence, H10a, H10b and H10c were all rejected. Relationship between gender and CS outcome variables Neith er attitude to online advertising n or materialistic values exhibited significant relationshi p with gender, and thus suggested that H11a and H11b were not supported. The data show ed a significant positive relationship between gender and pleasure to online shopping (r = .13, p<.01). Further, the correlation between gender and emotional responses to shopping in a mall or store was positive and significant (r = .25, p < .001 for pleasure, r = .15, p<.001 for arousal and r = .12, p<.001 for dominance), suggest ing partial support of H11c, which predicted tha t female respondents (coded as 2 ) exhibited higher score th an male counterparts (coded as 1 ) on emotions of pleasure, arousal and dominance to shopping.
77 Relationship between age and CS outcome variables Age w as negatively associated with almost all dependent variables, among which materialism (r = .12, p<.001), arousal to online shopping (r = .20, p<.001) as well as emotional responses to mall shopping (r = .16, p < .001 for pleasure, r = .12, p<.01 for ar ousal and r = .16, p<.001 for dominance) have statistically significant relationship with it. However, attitude to online advertising and pleasure to online shopping did not associate positively with age. Relationship between ethnicity and CS outcome va riables Finally, r egarding the research question addressing how ethnicity was related to three consumer skills under investigation, a one way ANOVA revealed that none of the dependent variables, including attitude towards online advertising (F=.458, p=.808 ), materialistic values (F=.556, p=.734) and the emotions toward shopping (e.g. F pleasure to online shopping =1.228, p= 294 ; F pleasure to mall shopping =1.228, p= 294 ) differ ed significantly across all ethnicity groups. Multivariate Results OLS Assumption s and Multicollinearity Prior to conducting an ordinal least squared regression for attitudes and materialism, four assumptions were tested and, if necessary corrected. All tests are adopted from Kennedy (1998) and Wooldridge (2009 ). Specific results o f these assu mptions are listed as followed: Linearity of the p henomenon Both a Pregibon Link test ( Pregibon 1980 ) and a Hosmer Lemeshow test were run to assess linearity ( Hosmer & Lemeshow 1989 ; Kennedy, 1998) For the attitudes model, the Preg est was not statistically significant (yhat^2 p = .89) and the yhat
78 coefficient was .98. The Hosmer Lemeshow test was also statistically insignificance through its F value of .27, though the five coefficients did not alternate between positive and negativ not statistically significant (yhat^2 p = .92) and the yhat coefficient was 1.01. The Hosmer Lemeshow test was statistically insignificance through its F value of .52, and all five coeffic ient categories alternated in a positive and negative pattern. In a perfectly l inear model, yhat^2 should not be significant, yhat should be 1, the F value should be insignificant, and the coefficients should alternate between positive and negative (Kenne dy, 1998; Wooldridge, 2006 ). Both models met these criteria, thus indicating that the linearity assumption holds. Normality of the error term distribution To test for normality of the error terms, this study first ran a normal probability plot (Hair et a l., 2009) and then a Jarque Bera test ( Jarque & Bera, 1980) For the attitudes model, the normal probability plot indicated that the model has a normal error term (see Figure 4 1). If skewness (right or left tailed) or kurtosis (length of that tail) were issues, the dark line would have varied from the 45 degree line significantly. Other than slight deviation between x axis .00 .25, the model appeared normal. The Jarque Bera test also indicated skewness to be .08 and kurtosis to be 2.45, where ideally the skewness would equal 0 and kurtosis would equal 3 (Hair et al., 2009) Despite finding a chi squared of 16.75, which is higher than the critical value of 5.991, the att itudes model primarily indicated a normal error term. For the materialism model, t he normal probability plot also indicated that our model has a normal error term (see Figure 4 2). Other than slight deviation between x axis .50 .75, the model appeared normal. The Jarque Bera test also indicated
79 skewness to be .14 and kurtosis to be 2.64. Despite also finding a chi squared of 7.32, which is higher than the critical value of 5.991, the materialism m odel also indicated a normal error term. Autocorrelation Upon examining the variables and survey respondents, autocorrelation was not con sidered a potential threat for this dissertation A utocorrelation occurs when it is possible to say something about one error term based on another error term. To put it differently, the current observation of a variable is highly related or dependent up on the value of the preceding period (Dupagne, 1997). This typical l y occurs for time series data, rather than cross sectional data. Constant variance of the error term (homoscedasticity) Finally, the study conducted tests for heteroskedasticity for each independent variable and across both models more broadly. Heteroskedasticity implies that the variance between error terms is different for any given independent variable; for example, the variance in Y may increase or decrease as X increases. For both e ( White, 1980 ) and a Breusch Pagan ( Breusch & Pagan, 1979 ) test were conducted, with both chi square tests (attitude and materialism models) indicating statistical significance of p < .05 and the presence o f heteroskedasticity across both models. Visual residual plots were then run for each independent variable, and the outcomes found heteroskedasticity across only four modeli ng procedure, which inflates the standard errors and allows for more conservative and robust tests ( Drudi & Massa, 2005; White, 1980)
80 Muticollinearity As the last step prior to the main analysis, the tolerance values and the variance inflation factor (VI F) were computed to assess multicollinearity (Hair et al., 2009). The suggested cut off threshold for a multicollinearity problem is a tolerance value of .10 (or a corresponding VIF of 10.0) ( Cooper & Schindler, 2006; Hair et al., 2009). All independent variables in the equations have low VIF scores ranging from 1.009 to 1.563 and high tolerance values ranging from .627 to .991, revealing that multicollinearity should not be a problem. Hence, the absence of multicollinearity indicates that the following regression coefficients can be precisely estimated for the relationships depicted. OLS Results Having accounted for all linear regression assumptions, multivariate models for attitudes toward online advertising and materialism were conducted and the resu lts presented in Table 4 2 Table 4 2 shows relationships between the independent variables and two dependent consumer skill variables. This paper utilized standardized rather than unstandardized coefficients in order to directly compare the relative cont ribution of each socialization variables. Several of the standardized regression coefficients were significant. However, only one of the independent variables ( consumption related communication via high interactive channels) had a significant coefficient across both regression models. Moreover, the variances explained by the 13 predictor variables considered in the two linear regression equations were relatively small Specifically, the materialism equation explained approximately 9.8 % of the variance i n materialism and the poorer equation involved attitude to online advertising for which the predictors
81 account ed for only 8.6% of the variance. Therefore one should interpret the strength of the relationship between the explanatory variables and the out co me variables reported in this table with a proper degree of caution: The Internet as mass m edium online television viewing was measured to compare with the impact of television viewing used in traditio nal socia lization studies (e.g. Englis, Solomon & Olofsson, 1993; Mangleburg & Bristol, ). A ddition ally this paper also considered other forms of online mass media such as online radio use, online newspaper and magazine readership. Con trary to prior research (e.g. Moschis & Churchill, 1978; Richins, 1987 ), the amount of online television viewing did not relate significantly to either attitude to online advertising or materialism (b= .007, p= .856; b= .027, p= .469 respectively). Simil arly, online radio listening did not show impact on the dependent variables. Online newspaper readership had a significant but negative relationship with attitude to online advertising (b= .101, p< .05). Lastly, as hypothesized, online magazine readershi p had a positive and significant relationship with attitude toward online advertising (b= .185, p < .001). However, it did not relate significantly to materialism (b= .087, p = .055). The Internet as interpersonal m edium As proposed in this study, consum ption related communication via high interactive online platforms (CCH ) was a strong predictor of both attit ude toward online advertising (b = 149 p < .001) and materialism (b = 214 p < .001). The finding suggested that the more an individual discuss es with others about consumption matters via high interactive tools such as email, instant messaging, forums and social networking sites, the more materialistic that person tends to be; and the more favorable
82 g becomes. The consumption related communication via low interactive online platforms (CCL) variable however, did not relate significantly with either one of the outcome variables (b= .078, p= .094 for attitude to online advertising; b= .072, p= .119 for materialism). Moderating e ffects of consumer susceptibility to personal i nfluence H 7 and H 8 put forth that the effects of the online consumption related communication on socialization outcome variables should be more salient among those who have high er ( v ersus low er ) levels of consumer susceptibility to interpersonal influence (including informative influence H 7 and normative influence H 8 ). The regression results reveal ed a not significant moderating effect. The paper used two moderators consumer s usceptibility to informative influence (CSII) and consumer susceptibility to normative influence (CSNI) to form interactions with two communication about consumption variables, namely CCH x CSNI, CCH x CSII, CCL x CSNI and CCL x CSII. However, none of t he four interaction terms appear ed to have a significant effect on the dependent variables. This indicates that the effect of consumption related communication on attitude to online ads and materialism did not o personal influence. Social structural v ariables Among turational development performed as a predictor of the level of materialism (b = .136, p<.001), suggesting that older participants are less materialistic than younger people. Sex and socioeconomic status however, did not appear to affect the acquisition of these two consumer skills (attitude to online advertising and materialism).
83 Ordered Logit Results The remai ning six dependent variables e motional r esponses (ER) toward online shopping (pleasure P online arousal A online dominance D online ) and emotional responses toward mall shopping (pleasure P mall arousal A mall and dominance D mall ) were run using an or dered logit regression. Table 4 3 and Table 4 4 five present the maximum likelihood estimates of the ordered logit equations. In order to interpret an ordered logit model, one should first look at the statistical significance of its coefficients as well as the direction of the signs (i .e. + positive or negative) (Liao, 1994). More specifically, a positive and statistically significant coefficient in an ordered logit regression indicates that the probability of increased emotional responses (e.g. pleasure, arousal or dominance) increa ses along with the explanatory variable that the coefficient represents. Conversely, a significant, negative coefficient suggests that an inverse relationship exists between emotional responses and a given independent variable (Liao, 1994). Both the co efficients and the odds ratio are reported to represent the effects of each explanatory variable in the model, as one should interpret the odds ratio (OR) rather than an ordered logit coefficient directly (Field, 2009). Odds ratios (OR) are obtained by ex ponentiating the beta coefficient (e ). The odds ratio represents the relative amount by which the odds of the outcome variable increase (OR > 1) or decrease (OR < 1), resulting from a one unit change in the predictor variable. Detailed interpretations of the ordered logit models are presented below. Among the three models representing emotional responses to online shopping, the Pleasure online 2 (13) = 103.93, p < 0.001) with approximately 14.5% of variance in the outcome (Pleasure online ) is explained by the
84 model (Nagelkerke Pseudo R square). The Arousal online model was statistically 2 (13) = 78.25, p < 0.001) with approximately 10.9% of variance explained by the model (Nagelkerke Pseudo R square). The Dominance online model however, was 2 (13) = 21.57, p =.062) with only approximately 3.1% of variance explained by the model (Nagelkerke Pseudo R square). Given that the Dominance online model does not fit the data well, the following interpretation only focuses on Pleasure online and Arousal online models. The Internet as mass m edium The results indicate d that neith er online television viewing and newspaper readership was a predictor of pleasure or arousal to online shopping. Online magazine readership had a weak but significant impact on pleasure (b=. 157 p<.05). People who read online magazines more frequently are more likely to score high on pleasure to online shopping. The odds ratio (OR) indicated that for every one unit increase in online magazine readership (i.e., going from 1 to 2), the odds of having a greater pleasure level of online shopping increase d by 17%. Contrary to what was proposed in the model, online radio had a significant but negative impact on pleasure (b= .085, p<.01). The OR suggested that for every one unit increase in online radio listening, people are 1.09 times less likely to gain plea as mass arousal to online shopping The Internet as interpersonal m edium Both of the frequency of consumption related communication variables were sufficient fact ors to explain the emotional responses to online shopping. The more frequently one communicates with others about consumption via high interactive online platforms (CCH), the higher pleasure (b=.286, p<.001) and arousal (b=.181,
85 p<.001). For each u nit increase in CCH, individuals were 1.33 and 1.20 times more likely to have higher pleasure and arousal responses to online shopping, respectively Likewise, respondents who communicate d with others about consumption via low interactive online platforms (CCH) more frequently scored higher on pleasure (b=.165, p<.01) and arousal (b=.151, p<.01). For each one unit increase in CCL, individuals were 1.20 and 1.16 times more likely to have higher pleasure and arousal responses to online shopping, respectivel y. Moderating e ffects of c onsu mer susceptibility to personal i nfluence Similar to the linear models of attitude toward online advertising and materialism the results here reveal ed a not significant moderating effect. Four interaction terms were formed with two consumption related communication variables and two moderators (i.e. consumer susceptibility to informative influence (CSII) and consumer susceptibility to normative influence (CSNI) namely CCH x CSNI, CCH x CSII, CCL x CSNI and CCL x CSII ) Ho wever, none of the four interaction terms appear ed to have a significant effect on the outcome va riables. This indicated that the effect of communication about consumption on emotional responses toward online shopping did susceptibility to personal influence. Social structural v ariables Among the social structural variables, age had online shopping (b = .115, p<.001). For each year increase in age, an individual is 1.12 times more likely to have a higher arousal response to online shopping. It did not however have a significant relationship with pleasure to online shopping. Gender was a .001). The odds ratio s uggested that women are 1.67 times more likely to have a higher
86 pleasure response to online shopping. Women did not appear to be significantly more aroused about online shopping than men (b= .117, p = .418). Similar to prior attitude and materialism mode ls, SES did not have a significant effect on the models under consideration. Emotional Responses toward S hopping as a S ocialization O utcome One of the purposes of this study was to add emotional responses toward shopping as a possible outcome of consumer s ocialization, and to test if or to what Therefore, it was of interest to examine further into the two ER variables emotional responses to shopping on the Internet and em otional responses to shopping in a mall/store. This study first compared responses to shopping on the Internet and emotional responses to shopping in a mall or a store. Table 4 5 indicates that on average, resp ondents experienced significantly greater pleasure (t (692) = 8.18, p < .001) as well as greater dominance (t (692) = 8.65, p < .001) to online shopping than to mall shopping. However, they felt more aroused or engaged about shopping in a mall or in a sto re than shopping online t (692) = 6.26, p < .001. Prior research has used AdSAM Emotional Groups to categorize respondents into groups based on the frequency of respondents who expressed similar emotional reactions (Goodman et al., 2008; Morris, 1995). Using the same method, the author classified the emotional responses into 9 groups to further investigate how people differ in their socialization var iables across groups. Figure 4 3 and Figure 4 4 illustrate the AdSAM Emotional Groups More specifica lly, respondents were divided into 9 groups in a Pleasure X Arousal space according to their pleasure and arou sal to online
87 shopping (Figure 4 3 ) as well as pleasure and arousal to sh opping in a mall/store (Figure 4 4 ). Dominance ratings were not included because the Dominance online model did not statistically fit the data, and none of the main socialization agents had a significant impact on Dominance mall The vertical y axis represents pleasure level, ranging from negative (1 bottom) to positive (9 top). The horizontal x axis represents left) to engaged (0 right). This study first used AdSAM Emotions Adjectives to help compare the emotions of each group ( see Figure 4 3 and Figure 4 4) The sele ction of these adjectives was in the AdSAM Emotions Database (Morris et al., 2002). AdSAM Emotion Groups' adjectives helped describe the emotions the respondents have when they ima gine shopping online or shopping in a mal l. For example, Figure 4 3 revealed that most people were in Group 6 (N=201). They felt warmed, mature and somewhat dignified when they imagine that they are shopping on the Internet. Participants in Group 9 (N= 187) felt enthusiastic, exuberant and stimulated toward online shopping. People in Group 5 (N=125) felt ambivalent, stoic and sensitive about online shopping. Group 3 felt comfortable, relaxed and untroubled, while Group 2, who shared the same level of engagement, was between indifferent, uninterested and unconcerned. Very few respondents held negative emotions toward online shopping, as Group 1 (N=5), Group 4 (N=8) and Group 7 (N=4) had less than 10 respondents. For emotional responses to mall shoppin g, respondents were more spread out across groups according to Figure 4 4 More people had negative reactions to mall shopping than online shopp ing, which aligned
88 with the t test result discussed above. Most people were in Group 9 (N=208) and they felt e nthusiastic, exuberant and stimulated toward online shopping. The differences between these nine AdSAM Emotional Groups were further analyzed using a multivariate analysis of variance (MANOVA). For ER to online shopping, the results indicat e d a statisti cally significant difference between the emotional groups, ( F P < .001); Tests of between subjects effects confirm ed the ordered logit regr ession results above and revealed that some of the variables differ significantly across emotional groups, including online magazine readership ( F (8, 684) = 2.39; P < .05), consumption related communication via high interactive channels (CCH) ( F (8, 684) = 5.84; P < .001), consumption related communication via low interactive channels (CCL) ( F (8, 684) = 5.47; P < .001), age ( F (8, 684) = 4.07; P < 001) and gender ( F (8, 684) = 3.23; P < .01). Th is study followed up these significant ANOVAs with Tukey's HSD post hoc tests, as shown below in Table 4 6 Only groups with significant differences were presented: The multiple comparison results indicate d that people in Group nine (high pleasure X high arousal) were significantly younger than people from Group 2 (Mean difference = 1.89, p < .01), Group 3 (mean difference = 2.11, p <.001) and Group 6 (mean difference = 1.09, p < .05). There were significant ly more women in Group 6 and Group 9 than in Group 5. Individuals from Group 9 read significantly more online magazine than those from Group 2 (mean difference = .64, p < .05). Individuals from different emotion groups appeared to differ the most in thei r frequency of communication about cons umption this further supported the regression results. More specifically, people from Group 2 tended to communicate significantly less about
89 consumption matters via high interactive platforms than those from Group 3 (mean difference = .74, p < .05), Group 6 (mean difference = .73, p < .01) and Group 9 (mean difference = .97, p <.001). People from Group 5 interact ed less frequently with other via high interactive tools on consumption matters than those from Group 6 (mean difference = .52, p < .05) and Group 9 (mean difference = .76, p < .001). Similar results obtained for CCL people from Group 6 and Group 9 rated significantly higher on CCL than those from Group 2 and Group 5. For ER to mall/store shopping, th e results also indicated a statistically significant difference between the emotional groups ( F P < .001). Tests of between subjects effects revealed that none of the online media usage variables differed significantly across the emotion groups. Consumption related communication via high interactive channels (CCH) ( F (8, 684) = 4.86; P < .01), Consumption related communication via low interactive channels (CCL) ( F (8, 684) = 11.94; P < .001), age ( F (8, 684) = 4.76; P < .001) and gender ( F (8, 684) = 7.24; P < .001). The study followed up these significant ANOVAs with Tukey's HSD post hoc tests, as shown below in Table 4 7 Only groups with significant differences were presented: The results indicate d that people in Group nine (high pleasure X high arousal) were significantly younger than people from Group 3 (Mean diff erence = 1.97, p < .05), Group 7 (mean difference = 2.52, p <.001) and Group 8 (mean difference = 1.59, p < .01); while Group 7 was significantly older than Group 6. There were significantly more women in Group 6 (mean difference = .43, p < .001) and Grou p 9 (mean difference = .36, p < .001) than Group 2. Similarly, significantly more women were found in Group 6
90 and Group 9 than in Group 4 an d Group 5, while Group 6 appeared to have more female participants than Group 8. Individuals in Group 1 tend ed to communicate significantly less about consumption matters via high interactive platforms than those from Group 6 (mean difference = .89, p < .05) and Group 8 (mean difference = .94, p < .01). Similarly, people from Group 6 and Group 8 rated significantly higher on CCL than those from Group 1. Group 8 had higher CCL than Group 5. Finally, both Group 6 (mean difference = .64, p < .05) and Group 8 (mean difference = .91, p < .001) had significantly greater CCL than Group 9.
91 Table 4 1. Bivariate r elation ship: consumer skills and socialization v ariables Dependent Consumer Skill Variables Emotional Response to Online Shopping Emotional Response to Mall/Store Shopping Independent Socialization Variables Attitude Materialism Pleasure Arousal Dominan ce Pleasure Arousal Dominance Socialization Agent Online TV .03 (.01) .01 (.01) .03 (.07) .08* (.06) .04 (.02) .07 ( .04) .03 ( .02) .06 (.04) Online Radio .02 (.01) .06 ( .06) .11 ( .10)** .03 ( .02) .05 (.04) .13 ( .10)* .03 ( .01) 03 (.02) Online Newspaper .07 ( .08)* .04 ( .05) .04 ( .07) .02 ( .07) .00 (.00) .04 ( .06) .00 (.00) .01 (.04) Online Magazine .20 (.15)*** .13*** (.06) .16*** (.10)* .08* (.05) .04 ( .06) .13 (.13)** .04 (.01) .06 (.05) CCH .23 (.13)*** .25 (.1 7)*** .23 (.22)*** .18*** (.10)* .04 (.01) .07 (.05) .05 (.01) .01 (.04) CCL .12 ***(.07) .18*** (.07) .22 (.20)*** .17*** (.11)* .08* (.07) .01 (.00) .06 (.06) .08 ( .1) Social Structural Age .01 ( .02) .12 (.13)*** 0.1 ( .06) .20 ( .20) *** .06 ( .07) .16 ( .14)*** .12 ( .11)** .16 ( .16)*** Gender .01 (.00) 0.0 ( .02) .13 (.11)** .05 (.03) .07 ( .06) .25 (.21)*** .15 (.14)*** .12 (.10)** SES .02 (.03) .01 (.00) .02 ( .01) .00 ( .01) .04 (.03) .01 ( .02) .00 ( .01) .0 3 ( .05) Note: Table entries are zero order correlations. Figures in parentheses are partial correlations. CCH Frequency of consumption related communication via high interactive platforms CCL Frequency of consumption related communication via low interactive platforms SES Socioeconomic Status ***Significant at .001 level; ** Significant at .01 level; Significant at .05 level.
92 Table 4 2. OLS results of consumer skills measures and socialization v ariables Dependent Consumer Skill Variables Independent Socialization Variables Model 1: Attitude to Online Ads Model 2: Materialism Socialization Agent Online TV .007 .027 Online Radio .012 .080 Online Newspaper .101* .050 Online Magazine .185*** .087 CCH .149*** .214*** CCL .078 .072 CCH*CSNI .030 .003 CCH*CSII .005 .043 CCL*CSNI .023 .018 CCL*CSII .068 .032 Social Structural Age .023 .136** Gender .002 .018 SES .028 .004 R 2 (R) .086 (.293) .098 (.313) Note: Table entries are standardized (linear) regression coeffi cients (beta weights) between the independent variables and two dependent variables. For attitude to online advertising regression model, F = 4.91, p< .0001; for materialism regression model, F = 5.68, p<.0001. ***Significant at .001 level; ** Significan t at .01 level; Significant at .05 level. CCH Frequency of consumption related communication via high interactive platforms ; CCL Frequency of consumption related communication via low interactive platforms ; SES Socioeconomic Status
93 Table 4 3 O rdered Logit results of the ER toward online shopping m odels Emotional Responses to Online Shopping Models Independent Socialization Variables Pleasure online Arousal online Dominance online Socialization Agent B(SE) Odds Ratio B(SE) Odds Ratio B(SE) O dds Ratio Online TV Viewing .009(.040) 1.01 .059 (.038) 1.06 .026 (.038) 1.03 Online Radio .085 (.031)** .92(1.09 ) .017 (.030) .98 .038 (.030) 1.04 Online Newspaper .082 (.062) .92 .113 (.060) .89 .016 (.060) 1.02 Online Magazine .157 (.06 3)* 1.17 .065 (.061) 1.07 .083 (.061) .92 CCH .286 (.068)*** 1.33 .181 (.065)** 1.20 .064 (.065) 1.07 CCL .185 (.060)** 1.20 .151 (.058)** 1.16 .088 (.058) 1.09 CCH*CSNI .075 (.050) 1.08 .004 (.049) 1.00 .027 (.049) .97 CCH*CSII .022 (.054 ) 1.02 .023 (.052) .98 .060 (.052) 1.06 CCL*CSNI .052 (.046) .86 .042 (.044) .96 .061 (.044) .94 CCL*CSII .005 (.049) 1.00 .093 (.047)* .91 .041 (.047) .96 Social Structural Age .045 (.022) .96 .115 (.022)*** .89 (1.12) .039 (.021) .96 Gender .511 (.150)*** 1.67 .117 (.144) 1.12 .153 (.144) .86 SES .003 (.049) 1.00 .014 (.047) 1.00 .035 (.047) 1.04 Model Fit 2 Log Likelihood 2 (13) = 103.93, p < .001 2 (13) = 78.25, p < 0.001 2 (13) = 21.57, p =.062 R 2 .145 .109 .031 Note: ***Significant at .001 level; ** Significant at .01 level; Significant at .05 level. CCH Frequency of consumption related communicat ion via high interactive platforms ; CCL Frequency of consumption related communication via low interactive platforms ; SES Socioeconomic Status
94 Table 4 4 Ordered Logit results of the ER toward mall/store shopping m odels Emotional Responses to Mall Shopping Models Independent Socialization Variables Pleasure mall Arousal mall Dominance mall Socialization Agent B(SE) Odds Ratio B(SE) Odds Ratio B(SE) Odds Ratio Online TV Viewing .058 (.038) .94 .021 (.293) .98 .066 (.038) .94 Online Radio 092 (.030)* .91 (1.10) .011 (.030) .99 .024 (.030) .98 Online Newspaper .059 (.060) .94 .013 (.060) 1.01 .096 (.060) 1.10 Online Magazine .200 (.061)** 1.22 .025 (.061) 1.03 .052 (.061) 1.05 CCH .085 (.065) 1.09 .023 (.064) 1.02 .087 (.064 ) 1.09 CCL .001 (.058) 1.00 .066 (.057) 1.07 .051 (.058) .95 CCH*CSNI .084 (.049) .92 .086 (.048) .92 .048 (.048) .95 CCH*CSII .029 (.052) 1.03 .011 (.052) .99 .077 (.052) 1.08 CCL*CSNI .022 (.044) 1.02 .070 (.044) .93 .021 (.044) 1 .02 CCL*CSII .002 (.047) 1.00 .002 (.047) 1.00 .069 (.047) .93 Social Structural Age .087 (.021)*** .92 (1.09) .064 (.021)** .94 (1.06) .194 (.021)*** .82 (1.22) Gender .791 (.146)*** 2.21 .496 (.144)** 1.64 .334 (.143)* 1.40 SES .016 (.047) .98 .014 (.047) 1.00 .068 (.047) .93 Model Fit 2 Log Likelihood 2 (13) = 88.74, p < .001 2 (13) = 40.24, p < 0.001 2 (13) = 45.27, p < 0.001 R 2 .123 .058 .064 Note: ***Significant at .001 level; ** Significant at .01 level; Significant at .05 level. CCH Frequency of consumption related communica tion via high interactive platforms ; CCL Frequency of consumption related communication via low interactive platforms ; SES Socioeconomic Status
95 Table 4 5 Emotional responses to shopping online and shopping in a mall/s tore ER to Online Shopping ER to Mall/Store Shopping Dependent t test Pleasure (M=6.76, SD=1.4) Pleasure (M=6.06, SE=2.1) t (692) = 8.18, p < .001 Arousal (M=5.35, SD=1.9) Arousal (M=5.92, SE=2.1) t (692) = 6.26, p < .001 Dominance (M=6.49, SD=1.9) Dominance (M=5.60, SE=2.1) t (69 2) = 8.65, p < .001
96 Table 4 6. Group differences in socialization variables (ER to online s hopping) Socialization Variables Pairs of Groups Difference of Mean Std. Error Sig. Age 2 9 1.89 ** .468 .002 3 9 2.11 *** .439 .000 6 9 1.09 .328 .026 Gender 5 6 .21 ** .054 .003 5 9 .18 .055 .023 Magazine 2 9 .64 .198 .034 CCH 2 3 .74 .217 .019 2 6 .73 ** .183 .003 2 9 .97 *** .185 .000 5 6 .52 .144 .011 5 9 .76 *** .146 .000 CCL 2 6 .96 *** .207 .000 2 9 1.12 *** .209 .000 5 6 .52 .163 .042 5 9 .68 ** .165 .001 Note: ***Significant at .001 level; ** Significant at .01 level; Significant at .05 level. CCH Frequency of consumption related communication via high interactive platforms ; CCL Frequency of consumption rel ated communication via low interactive platforms ;
97 Table 4 7. Group differences in socialization variables (ER to mall/store s hopping) Socialization Variables Pairs of Groups Difference of Mean Std. Error Sig. Age 3 9 1.97 .617 .038 6 7 2.0 6 .622 .027 7 9 2.52 *** .572 .000 8 9 1.59 ** .418 .005 Gender 2 6 .43 *** .083 .000 2 9 .36 *** .075 .000 4 6 .34 .096 .012 4 9 .28 .089 .049 5 6 .31 *** .064 .000 5 9 .25 *** .053 .000 6 8 .22 .071 .043 CCH 1 6 .89 .253 .014 1 8 .94 ** .258 .009 CCL 1 6 1.23 *** .288 .000 1 8 1.50 *** .286 .000 5 8 .84 ** .203 .001 6 9 .64 .178 .010 8 9 .91 *** .187 .000 Note: ***Significant at .001 level; ** Significant at .01 level; Significant at .05 level. CCH F requency of consumption related communication via high interactive platforms ; CCL Frequency of consumption related communication via low interactive platforms ;
98 Figure 4 1 Normal probability plot for attitude toward online advertising m odel Figu re 4 2 Normal p r obability plot for materialism m odel
99 Figure 4 3. AdSAM Groups emotional responses to online s hopping
100 Figure 4 4. AdSAM Groups emotional r espo nses to mall s hopping
101 CHAPTER 5 CONCLUSION AND DISCUSSION Consumer socializatio n (CS) refers to the process by which an individual acquires through interaction with various socialization agents, certain attitudes and values relevant to their functioning as consumers in the marketplace ( Moschis & Churchill, 1978; Ward, 1974 ) While television, family and peer communicati on are widely considered to be influence agent s in the consumer socialization process, no one has yet examined the Internet as a potential consumer socialization agent. Guided by the consumer socialization framework, t his dissertation serves as an initial investigation of the role the Internet may play in the consumer learning process. A s a hybrid of mass media and interpersonal communications, the Internet provides an environment where young adults may develop atti tudes and values relevant to their functioning as consumers, not only through passive learning such as heavy exposure to television programming online, but also through active learning, such as frequent interac tion with people online Therefore, the pre sent study extends a classic question of the Internet age posed by Morr i s and Ogan ( 1996 ): is the Web a mass media channel or an interpersonal communication channel ? and asks : d oes the Internet function more as a mass media vehicle or as an interpersonal communication consumer attitudes and values? To address these issues, this dissertation surveyed 693 people representing the Millennials generation. Having grown up with access to computers, this is the first generat ion that experienced the rapid development of the Internet technology ( Myers & Sadaghiani, 2010) The survey responses suggested that the Internet as interpersona l medium is a stronger CS agent than the Internet as mass medium. Chapter 5 will
102 summarize a nd discuss the survey results in a more broad perspective. It will also present the implication of the results as well as recommendations for marketers. Finally, some limitations of the current dissertation and a future research agenda will be highlighte d. Discussion of Results The tests of hypotheses showed no significant relationship between online television viewing and any of the so cialization outcome variables. Similarly, online newspaper and online radio usage exhibited no impact on consumer valu es and attitudes. Among the Internet as mass medium variables, only online magazine readership has significant and positive relationship with three socialization outcome variables, including attitude toward online advertising and pleasure to both online a nd mall shopping. In regards to the two Internet as interpersonal medium variables, consumption related communication via high interactive venues (CCH) exhibited strong correlation with the CS outcome variables, whereas consumption related communication t hrough low interactive channels (CCL) only had significant relationship with emotional attitude toward online shopping. Among the social structural variables, socioeconomic status and ethnicity did not have significant relationship with the CS outcome va riables, indicating that these two factors may not play a role in the online consumer socialization process. The findings in dicated that age is negatively related to materialism and emotional responses to shopping. Although gender did not yield any signi ficant impact on attitude to online advert ising and materialistic values, it significantly related to pleasure to online shopping and emotional attitudes to mall shopping. Specifically, the data demonstrate d that
103 women are more positive to both online and mall shopping, as well as more engaged and more empowered in a mall shopping situation. As for the regression results, m ost models fit the data and were statistically supported and as expected, t he amount of variance explained by the regression equation s is relatively small (ranging from 9.8% for materialism model to 14.5% for pleasure to online shopping model ). T he low variance explained is not surprising, and is consistent with representative CS studies. In one of the pioneer studies within the CS p aradigm, Churchill and Moschis (1979) found that television viewing, family communication about consumption and peer communication about consumption explained 6% of the variance for materialism and 5% of the variance for motivations for consumption. The v ariance results in this study were higher. (1997) investigated the role of television in the construction of consumer reality, and found that television viewing and other social structural variables accounted for merely 10% of the varian ce in the affluence estimates. Bush et al., (1999) found that socialization agents only explained 14% of the variance of attitude toward advertising. Both are comparable to the findings in the current study. An important contribution of this dissertatio n to the CS theory is finding that consumption related communications via online tools are more relevant for c onsumer socialization process, especially when tools are highly interactive (b=.149, p<.001 for attitude toward online advertising model; b=.214, p<.001 for materialism; Odds Ratio = 1.33 for Pleasure online ; Odds Ratio=1.20 for Arousal online ) This is in line with CS literature where the level of interpersonal communication regarding consumption
104 matters has been found to affect a variety of consume r attitudes and values (e.g. Bush et al., 1999; Moschis 1978; Moschis & Churchil, 1978) The high interactive communication tools (i.e. email, instant messaging, social networking sites and forums) appear to be the most important sources of gaining consu mer skills or values, because these tools contribute positive attitude toward advertising, a more materialistic values and better emotional responses to online shopping in a consistent manner Although attitude toward online a dvertising and materialism relate d only with high interactive communication and not low interactive communication the data indicate line shopping and consumption related communication through both low interact ive and high interactive tools. A closer look at the emotional responses toward consumption gave more insights into how consumption related online communication is related to this newly proposed socialization outcome This dissertation also demonstrated that the AdSAM Emotional Groups could be used as an additional way to interpret data. Specifically, t his study divided respondents into nine groups based on their affective reaction s toward shopping. The group comparison reinforced the regr ession results It suggested that people who frequently communicated with others online had significantly greater pleasure and arousal than those who communicated with others less frequently moving from unimpressed, unconcerned or a mbivalent about shoppi ng to excited, warmed or comfortable about shopping. An implication of this is that companies could induce these feelings by encouraging people to talk about their products or brands with others
105 on the Internet, particularly through high interactive appli cations such as chat forums and social networking sites. For the demographic factors, it seemed that younger respondents feel more enthusiastic, exuberant and stimulated toward shopping (both online and mall) than the older Millennials, while female are m ore engaged and pleased about shopping than male respondents. Furthermore, in regards to shopping on the Internet the majority of respondents (56% ) demonstrated high positive and high engagement affections They felt enthusiastic, excited, warmed or di gnified about online shopping The number re duced to less than half (44%) when the participants were asked about their feelings towards shopping in a mall. Instead, people (30%) shifted to feel more ambivalent, stoic, anxious or ap prehensive when it com es to mall shopping I n comparison, only 22% of respondents share d these feelings about online shopping Finally, 15% of respondents had strong negative feelings when they imagine d themselves shopping in a mall ; they felt sullen and weary (5%), displease d and distrustful (5%) and even irritated and startled (5%). However, these negative feelings were not intimately related to the experience of online shopping, due to only 2% of respondents feeling displeased about online shopping As a side comparison the Millennials seemed more engaged in shopping in a mall or a store than shopping on the Internet The reason for this may be that when people shop in a mall or store, as opposed to a virtual shopping environment, they are surr ounded by tangible sights a nd sounds and are thus more physically engaged Nevertheless, greater arousal or engagement does not necessarily imply greater
106 pleasure or dominance as the finding suggested that online shopping induced more pleasure and dominance in participants. An ob vious reason for higher dominance in an online shopping environment is that consumers have greater control over where to go and what to buy ( Gretzel, Yuan & Fesenmaier, 2000; Schlosser, Shavitt, Kanfer, 1999) Another supported finding was that f emale pa rticipants exhibited stronger emotional responses toward shopping than did their male counterparts. This is in line with those of other studies: women generally have a more positive attitude toward shopping (e.g. Campbell, 2000; Raajpoot, Sharma & Chebat 2008). Additionally, a ge appeared to be significantly related to materialism, arousal to online shopping as well as pleasure, arousal and dominance to mall shopping. All these associations give a negative answer to the questio n raised in Chapter 1 : D oes and higher emotional responses towards shopping, or vice versa Neither gender nor age has an effect in attitude toward online advertising. This is somewhat consistent with generally do not relate to attitude toward advertising. Finally, the present research found no differences in regards to ethnic differences in consume r socialization. This is contrary to previous studies (e.g. Bush et al., 1999; Singh et al., 2003) The results fail to support the proposed models with respect to several key relationships. In this regard, p erhaps one of the most unexpected aspects invol ve s the finding that other than online magazine readership, none of the Internet as mass medium variables contributed significantly to the criterion variables. Although past research within the CS paradigm has recognized the influence that mass media,
107 pa values and attitudes (e.g. Englis, Solomon & Olofsson, 1993; Mangleburg & Bristo 1989), the survey data suggests that heavy exposure to online television does not lead to higher materialistic values, more favorable attitude to online advertising or stronger emotional responses to shopping among this audience This finding indicate s that the cultivation theory ( Gerbner & Gross, 1976 ) which was the theoretical foundation f or the media effects component of this study, may not be the most appropriate framework for the online media context Nevertheless, this disconfirmed role of online television as a socialization agent is still a significant contribution to unexplored area of consumer socialization. The emergence of online media has raised the need to examine cultivation effect in this new media environment. Past research on mass media effect s has traditionally revealed weak but persistent cultivation effect (e.g. Gerbner Gross, Morgan & Signorielli, 1986; Potter, 1993) Indeed, G erbner et al., (1986) justified this small effect by arguing that the messages are so stable, the med ium is so ubiquitous, and accumulated total exposure that counts, then almost everyone shou ld be affected However, the di versification of media choices, particularly on the Int ernet, has made the mass media online television viewers now have the ability to control the amount of exposure to advertising compared to their counterpart in the offline world (Lee, 2007). In addition, while watching traditional television means decreased time interacting with other people, people tend to be more multitasking whe n they are on the Internet ( Prensky, 2001). Therefore,
108 through exposure to mass media, particularly through television and advertising (e.g. Moschis & Churchill, 1978), the findings evidently cultivation effects do not hold true for the online environment Providing that the present study took into account the potential difference in impact of television genre types on consumer learning, it had a more extended a nd fine grained measure of television program viewing compared to previous CS studies However, similar to general television viewing, genre types did not have substantial impact on consumer skills, nor did they make a significant difference on the effec ts. This provided further evidence to the disconfirmed relationship between overall online television viewing and the CS outcomes. Considering prior CS studies that included socioeconomic status as a factor have produced mixed results (John, 1999), it is reasonable to find that socioeconomic status is not related to any of the three socialization outcome variables. The data suggest that socialization practices presented in higher socioeconomic classes (e.g., more information and opportunities for consum ption) ( Moschis & Churchill, 1978) may not where they found no significant re lationship between SES and attitude toward advertising. In addition, susceptibility to personal influence, as an individual difference factor, did not have a moderating effect on any of the socialization outcome variables Although this is inconsistent w ith prior research (e.g. Achenreiner, 1997; Mangleburg & Bristol, 1998
109 obtained similar results. Perhaps the reason lies in the fact that an offline scale was used to measure an online phenomenon Bailey (2005) investigated the relationship between consumer susceptibility to informational influence review website, the author found that whether a person is more or less susceptible to personal influence does not dissertation is that both studies used the traditional scale of consumer susceptibility to personal influence ( Bear den et al., 1989) influenced by reference groups on the Internet This scale however, may be out of date people that the subject know in real life such as family and peers ( Bearden et al., 1989). As the Millennials spend more time on the Internet, their circle of influence has expanded beyond acquaintance and has shifted from offline to online Therefore, it may increase its validity if future research develops a new scale that can assess the susceptibility to online interpersonal influence. A recommendation would be to take into susceptibility to o nline personal influence. Conclusion and I mplications In conclusion, this study revealed that Consumer Socialization manifests differently online than it does in a traditional setting. T consumer attitudes and values was exhi bited mainly through fostering consumption related communication via interactive channels such as email and social networking sites rather than through disseminating consumption related information via media channels such as online television. The Millen nial consumers are turning away from
110 passive learning (online mass media); instead, they develop consumer attitudes and values by actively participating in online conversation about consumption (high interactive channels), or by actively pursuing consumpti on related information (low interactive channels). Potentially extending the concept of Consumer Socialization (CS) is the theoretical contribution of studying both media and social influence on the Internet. As stated previously, an increasing amount o f such communications is conducted via the Internet. This paper represents an update on the CS framework by identifying how the Internet, as a type of media and interpersonal means of communication, plays a role in the process of consumer socialization am ong the Millennials In this regard, the present research serves as a response to the call of Roedder John (1999), who in a review of twenty five years of CS research indicated the need to address the Internet as a socialization agent. The model results provide d a new angle and new evidence for understanding socialization effectiveness of computer mediated communications. People learn different things at different times from different sources throughout their whole life span ( Moschis, 1981; 1987 ) Numer ous researchers have called for CS studies that focus beyond children and adolescents ( e.g. 1997 ; Singh et al., 2003; Ward, Klees & Robertson, 1987 ). However, to date, socialization among young adults remains a relatively unexp lored area. The present research fills this gap by extending the literature to the generation of Millennials, and exploring how Internet, as an emerging technology and possibly a dominant medium n as consumers.
111 This dissertation also extended the literature by testing responses toward consumption as a possible outcom e of the socialization process. The findings suggested that among the criterion variables, emotional response towar d online shopping is able to capture the greatest influence. This serves as another contribution of the current study; that is, it demonstrated that emo tional responses to consumption should be considered as an i mportant CS outcome variable. For marketers, if reaching this generation segment is of focal concern, high interactive communication tools seem to be the most appropriate medium. This group of consumers is not only heavy users of social networking sites email, chat forums and instant messaging, they are also highly influenced by consumption related communications with others using these tools. Therefore, marketers should encourage positive reviews, comments, or discussion about the brand or product acr oss these online communication tools. The result indicated that the Millennials are more prone to receive informative influence from others on the Internet than normative influence. This may be evidence that the consumption related communication impact c onsumer values and attitudes more through the informational route than the conformity route. The implication here is that when targeting the Millennials consumers, marketers should construct informational messages that encourage consumers to internalize t he marketplace information, rather than using messages that aim to motivate the desire to conform to social norms. For instance, instead of focusing on appeals that convey social norms or approved behavior to induce conformity to the communicated behavior ( Nolan, Schultz, Cialdini, Goldstein
112 & Griskevicius, 2008), marketers should emphasize on marketing communications that include unique selling proposition comparative or factual messages Finally when comparing between feelings towards shopping on the Internet and in a mall, this study found that the respondents experienced higher pleasure and felt more in control when they shop on the Internet. A greater pleasure to online shopping is good news for marketers in the electronic marketplace. Because online shopping is substantial and continually growing, online marketers should put more effort into increasing the overall shopping experience. Limitations and Future R esearch While this study has provided the first step toward investigating the Internet as a Consumer Socialization agent, it still leaves some unanswered questions. For example, given the cross sectional nature of this study, the causal direction remains uncertain. In addition, a single survey design would not be able to capture th e long term development or the process of consumer socialization. Therefore, further research is encouraged to adopt a longitudinal design, as it is more suited to examine the psychological or sociological processes (Moschis & Churchill, 1978). For insta nce, future CS research could obtain longitudinal data to compare one' s levels of materialism or attitudes toward consumption before and after adulthood. The sample of this dissertation consisted of a unique group of consumers the Millennials. The rea son for this being, that the y are the first generation to experience the rapid development of the Internet. Future research should also explore the consumer learning of younger generations or individuals who are still in their adol escent adol escent s were born after the Internet revolution ( Montgomery & Chester, 2009), the Internet may play a greater role in their consumer
113 learning. Another drawback of this study is that the results were likely to be skewed by gender, as there were about 20 p ercent more female than male respondents. Following the consumer socialization framework, this study tested three online socialization agents online mass media, communication through high interactive tools, and communication through low interactive tool s. These may not be the only factors that might impact an individual s consumer learning. As a wide variety of digital tools and online venues future researchers should expect that there may be other reference groups Although this study confirmed the important role of peer communication through high interactive tools, it did not specify who the reference groups might be; in other words, who are they communicating with on the Internet? For instance, are online opinion leaders more influential or people with whom one has strong ties more groups on the Internet may be able to answer this question. In add ition, t his study did not differentiate among the types of applications such as email, social networking sites and instant messaging. Wang, Yu and Wei (2012) took the initiative and explored peer communication on social media as the consumer socialization agent. Similar studies should be performed to compare socialization effects between the various interactive formats. Future research can also differentiate these communication venues by the levels of the sales funnel For instance, consumers may use so cial networking sites earlier in the demand generation stage. Then they may use product review sites when they are getting closer to purchase.
114 This study used two traditional variables to repres ent the socialization outcomes: attitude towards online adv ertising and materialistic values. It was also able to present and demonstrate emotional responses toward consumption as an important consumer socialization outcome variable. Future studies should extend the list of CS outcome variables to include other consumer knowledge or attitudes related variables such as consumer affairs knowledge (Moschis & Churchill, 1978), skepticism toward advertisi ng (Mangleburg & Bristol, 1998), brand sensitivity (Lachanc e, Beaudoin & Robitaille, 2003), and attitude towards p ri ce (Moschis & Churchill, 1978). Consumption motive related variables such as impulsive consumption (Rindfleisch et al.,1997), social motivations and economic motivations for consumption (Churchill & Moschis, 1979) should also be extended to the list of CS oucome variables While some CS researchers have looked at cross cultural differences in the consumer socialization process (e.g. Rose, 1999; Singh et al., 2003), most studies in this area still use data from only one country (John, 1999). Further res earch should investigate consumer samples from other cultural background s particularly addressing how the Internet and other new technology such as mobile devic es play a role in the socialization process. This would help gain great insights to understand i ng cultur al differences, as they relate to the relative influence o f various socialization agents ( suc h as those in this dissertation) on consumer socialization outcomes. For instance, it seems reasonable to expect that in a collectivist society, such as China or Japan, interactive communications about consumption may play an even bigger role in this process.
115 LIST OF REFERENCES Achenreiner, G. B. (1997). Materialistic values and susceptibility to influence in children. Advances in consumer research 24 ( 1), 82 88. Ahuvia, A. C., & Wong, N. Y. (2002). Personality and values based materialism: Their relationship and origins. Journal of Consumer Psychology 12 (4), 389 402. Allen, C.T., Machleit, K. A. & Kleine, S. (1992). A Comparison of Attitudes and Emoti ons as Predictors of Behavior at Diverse Levels of Behavioral Experience. Journal of Consumer Research, 18, 493 504. advertisements. Journal of Advertising Research 41, 7 22 Arnett, J. J. (2003). Conceptions of the transition to adulthood among emerging adults in American ethnic groups. New directions for child and adolescent development 2003 (100), 63 76. Arnett, J. J. (2007). Socialization in Emerging Adulthood. in Handbo ok of Socialization: Theory and Research, Joan E. Grusec and Paul D. Hastings (eds.). New York: The Guilford Press. Austin, W. A. & Totaro, M. W. (2011). Gender differences in the effects of Internet usage on high school absenteeism. Journal of Socio Econo mics 40, 192 198. Awad, N. E. & Ragowsky, A. (2008). Establishing trust in electronic commerce through online word of mouth: An examination across genders. Journal of Management Information Systems. 24 (4), 101 121. Bandura, A (1971). Modeling influenc es on children. Testimony to the Federal Trade Commission. Bandura, A (1977). Social Learning Theory Englewood Cliffs, NJ: Prentice Hall. Bandura, A (1986). Social Foundations of Thought and Action Englewood Cliffs, NJ: Prentice Hall, Inc. Bandura, A (1994). Social Cognitive Theory of Mass Communication. In J Bryant and D. Zillman, (Eds.), Media Effects: Advances in Theory and Research. Hillsdale NJ: Erlbaum. Bandura, A (2001). Social Cognitive Theory of Mass Communication. Media Psychology 3, 3, 265 299 Baran S & Davis D (2009). Mass Communication Theory: Foundations, Ferment, and. Future (5th Ed. ). Belmont, CA: Wadsworth.
116 Batra, R & Michael L. R. (1986). Affective Responses Mediating Acceptance of Advertising. Journal of Consumer Research, 13, 234 249. Bearden, W. O., Netemeyer, R. G., & Teel, J. E. (1989). Measurement of consumer susceptibility to interpersonal influence. Journal of consumer research 473 481. Bearden, W. O., Netemeyer, R. G., & Teel, J. E. (1990). Further validation of t he consumer susceptibility to interpersonal influence scale. Advances in consumer research 17 (1), 770 6. Belk, R. W. (1985). Materialism: trait aspects of living in the material world. Journal of Consumer Research 12, 265 280. Benmoyal Bouzaglo, S., & Moschis, G. P. (2009). The effects of family structure and socialization influences on compulsive consumption: a life course study in France. International Journal of Consumer Studies 33 (1), 49 57. Bollen, K. A. (1989). Structural equatio n with latent variables New York: Wiley. Bradley, M. M. & Lang, P. J. (1994). Measuring emotion: The self assessment manikin and the semantic differential. Journal of Behavioral Therapy and Experimental Psychiatry, 25, 49 59. Brim, O. G. (1966). Socializa tion through the life cycle. In O. Brim and S. Wheeler, (Eds.), Socialization after childhood (3 49). New York: John Wiley & Sons, Inc. Breusch, T. S., & Pagan, A. R. (1979). A simple test for heteroscedasticity and random coefficient variation. Econometri ca: Journal of the Econometric Society 1287 1294. Bush, A. J., Smith, R., & Martin, C. (1999). The influence of consumer socialization variables on attitude toward advertising: A comparison of African Americans and Caucasians. Journal of Advertising, 28 13 24. Burke, M. C. & Edell, J. A. (1989). The Impact of Feelings on Ad Based Affect and Cognition. Journal of Marketing Research 26, 69 83. Burnkrant, R. E., & Cousineau, A. (1975). Informational and normative social influence in buyer behavior. Journa l of Consumer Research 206 215. Byrne, B. M. (1994). Structural equation modeling with EQS and EQS/Windows: Basic concepts, applications, and programming Thousand Oaks, CA: Sage. Calenda, D., & Meijer, A. (2009). Young people, the Internet and political participation. Information, Communication & Society, 12, 879 898.
117 Carlson, L., Walsh, A., Laczniak, R. N., & Grossbart, S. (1994). Family communication patterns and marketplace motivations, attitudes, and behaviors of children and mothers. Journal of Consu mer Affairs 28 (1), 25 53. Calder, B. J. & Burnkrant, R. E. (1977). Interpersonal Influence on Consumer Behavior: An Attribution Theory Approach. Journal of Consumer Research 4, 1, 29 38 Calder, B. J., Phillips, L. W. & Tybout, A (1981). Designing resear ch for applications. Journal of Consumer Research 8, 197 207. Campbell, C. (2000). Shopaholics, spendaholics, and the question of gender. I Shop, Therefore I Am: Compulsive Buying and the Search for Self. Edited by Benson AL. Northvale, NJ, Jason Aronson 57 75. Chang, L & Arkin R. M. (2002). Materialism as an Att empt to Cope with Uncertainty. Psychology & Marketing 19 (5), 389 406. Christ, W. G. (1985). Voter Preference and Emotion: Using Emotional Response to Classify Decided and Undecided Voters. Jo urnal of Applied Social Psychology 15, 237 254. Christ, W. G. & Biggers, T. (1984). An Exploratory Investigation into the Relationship between Television Program Preference and Emotion eliciting Qualities a New Theoretical Perspective. Western Journal of Speech Communication 48, 293 307. Choi, S. M., & La Ferle, C. (2004). Convergence across American and Korean young adults: socialisation variables indicate the verdict is still out. International Journal of Advertising 23 (4), 479 506. Childers, T. L. & Rao, A. R. (1992). The Influence of Familial and Peer Based Reference Groups on Consumer Decisions. Journal of Consumer Research, 19(2), 198 211 Chia, S. C. (2006). How media contributes to misperception of the social norm. Mass Communication and Society 9, 301 320 Materialism. Communication Research 37 (3), 400 419. Chiu, C. M., Hsu, M. H., & Wang, E. T. (2006). Understanding knowledge sharing in virtual communities: An integ ration of social capital and social cognitive theories. Decision support systems 42 (3), 1872 1888. Cialdini, R. B. & Goldstein, N. J. (2004). Social Influence: Compliance and Conformity. Annual Review of Psychology 55, 591 621.
118 Clark, R. A., & Goldsmit h, R. E. (2005). Market mavens: psychological influences. Psychology and Marketing 22 (4), 289 312. Coyle, J. R. & Thorson, E (2001). The Effects of Progressive Levels of Interactivity and Vividness in Web Marketing Sites Journal of Advertising, 30, 3, 6 5 77 Coulter, K. S. (1998). The effects of affective responses to media context on advertising evaluations. Journal of Advertising, 27, 4, 41 51 Cooper, D. R., & Schindler, P. S. (2006). Marketing research New York: McGraw Hill/Irwin. Cohen, J. B. & Go lden, E (1972). Informational social influence and product evaluation. Journal of Applied Psychology 56(1), 54 59 Cronbach, L. J. (1987). Statistical tests for moderator variables: Flaws in analyses recently proposed. Cui, G. (1997). Marketing strategies in a multi ethnic environment. Journal of Marketing Theory and Practice 5, 122 134. Daugherty, T. & Zhang, J (2009). Third person Effect and social networking: implications for online marketing and word of mouth communication. American Journal of Busine ss, 24, 2, 53 70. Deutsch, M. & Gerard, H. B. (1955). A study of normative and informational social influences upon individual judgment. Journal of Abnormal and Social Psychology, 51(3), 629 636. Delener, N., & Neelankavil, J. P. (1990). Information sourc es and media usage: A comparison between Asian and Hispanic subcultures. Journal of Advertising Research 30, 42 52. D. & Choudhury, P. K. (2000). Effect of Assimilation on Consumer Susceptibility to Interpersonal Influence. Journal of Consumer Marketing, 17 (4), 290 307. Drudi, F., & Massa, M. (2005). Price Manipulation in Parallel Markets with Different Transparency*. The Journal of Business 78 (5), 1625 1658. Dupagne, M. (1997). Beyond the Principle of Relative Constancy: determinants of cons umer mass media expenditures in Belgium. Journal of Media Economics 10 (2), 3 19. Edell, J. A. & Burke, M. C. (1987). The Power of Feelings in Understanding Advertising Effects. Journal of Consumer Research 14, 421 33.
119 Englis, B. G., Solomon, M. R. & Olof sson A (1993). Consumption Imagery in Music Television: A Bi Cultural Perspective. Journal of Advertising 22(4), 21 33 Social capital and college students' use of online social network sites. Journal of Computer Mediated Communication Esses, V. M., Haddock, G. & Zanna, M. P. (1993). Values, stereotypes,and emotions as determinants of intergroup attitudes. In D. M. Mackie & D. L. Hamilton (Eds), Affect, Cognition, and stereotyping: Interactive Processes on Group Percepti on, pp. 137 166. San Diego: Academic Press. Feltham, T. S. (1998). Leaving home: Brand Purchase influences on young adults. Journal of Consumer Marketing, 15, 372 385. Fiske, S. T. & Taylor, S. E. (1991). Social Cognition (2rd Edition). New York: McGra w Hill Field, A. P. (2009). Discovering statistics using SPSS for Windows (3rd edition). London: Sage. Fitzmaurice, J. & Comegys, C. ( (2006). Materialism and Social Consumption. Journal of Marketing Theory and Practice, 14 (4), 287 299. Flurry, L (2005) Children's influence in family decision making: Examining the impact of the changing American family Journal of Business Research 60, 4, 322 330 Fournier, F. & Richins, M. L. (1991). Some theoretical and popular notions concerning materialism. In F. W. Rudmin (Ed.), To have possessions: a handbook on ownership and property (special issue). Journal of Social Behavior Personality, 6, 6, 403 414. depth interviewing by instant messaging. Social Research Update 53 ( 2), 1 4. Gerbner, G., & Gross, L. (1976). Living with television: the violence profile. Journal of Communication 26, 173 199 Gerbner, G., Gross, L., Eleey, M., Jackson Beeck, M., Jeffries Fox, S., & Signorielli, N. (1977). TV Violence Profile, No. 8: T he Highlights. Journal of Communication, 27, 2, 171 180. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1980). America: Violence Profile No. 11. Journal of Communication 30 3 10 29 Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1986). Living with television: The dynamics of the cultivation process. Perspectives on media effects 17 40.
120 Gerwitz, J. L. (1969). Me chanisms of social learning. In D. A. Goslin (Eds.) Handbook of Socialization Theory and Research Chicago: Rand McNally. Goodman, J. R., Morris, J. D., & Sutherland, J. C. (2008). Is beauty a joy forever? Young women's emotional responses to varying types of beautiful advertising models. Journalism & Mass Communication Quarterly 85 (1), 147 168. Goslin, D. A. (1969), Handbook of Socialization Theory and Research Chicago: Rand McNally and Co. Greene, W.H., (2003). Econometric Analysis. New Jersey: Prentice Hall. Granovetter, M. S. (1973). The strength of weak ties. American journal of sociology 1360 1380. Gretzel, U., Yuan, Y. L., & Fesenmaier, D. R. (2000). Preparing for the new economy: advertising strategies and change in destination marketing organiza tions. Journal of travel Research 39 (2), 146 156. Guest, L. (1944). A Study of Brand Loyalty. Journal of Applied Psychology 28, 16 27. Hair, J F., Black, W. C., Babin, B. J. & Anderson, R. E. (2009). Multivariate Data Analysis (7th ed.), Englewood Clif fs, NJ: Prentice Hall Havlena, W. J. & Holbrook, M. B. (1986). The Varieties of Consumption Experience: Comparing Two Typologies of Emotion in Consumer Behavior. Journal of Consumer Research. 13 3, 394 404 Heath, L. (1984). Impact of newspaper crime repor ts on fear of crime: Multimethodological investigation. Journal of Personality and Social Psychology, 47, 263 276. Heath, T. ; Motta, E. & Petre, M. (2006). Person to person trust factors in word of mouth recommendation. In: Conference on Human Factors in C omputing Systems 27 Apr 2006, Montreal, Quebec, Canada. Hirschman, E. C. & Holbrook, M. B. (1982). Hedonic Consumption: Emerging Concepts, Methods and Propositions Journal of Marketing 46, 3, 92 101 Hirschman, E. C. (1999). The role of emoti on in consumer research. Advances in Consumer Research 26, 4 11. Holbrook, M. B. (1980). Some Preliminary Notes on Research in Consumer Esthetics. In Jerry C. Olson (Eds.) Advances in Consumer Research, Vol. VII, (104 108). Ann Arbor: Association for Con sumer Research.
121 Holbrook, M. B., & Batra, R. (1987). Assessing the role of emotions as mediators of consumer responses to advertising. Journal of Consumer Research, 14 404 420. Hosmer, D. W., & Lemeshow, S. (1989). Applied regression analysis. New York, J ohn Willey Hutcheson, G. D., & Sofroniou, N. (1999). The multivariate social scientist: Introductory statistics using generalized linear models Sage Publications Limited. Karande, K. & Ganesh, J (2000). Who Shops at Factory Outlets and Why? An Explorat ory Study. Journal of marketing Theory and Practice, 8, 4, 29 42 personal influences. Journal of Current Issues and Research in Advertising 23, 43 53. Kaye, B. K. & J ohnson, T. J. (2004). A Web for all reasons: uses and gratifications of Internet components for political information. Telematics and Informatics 21, 197 223 Katz, F. M. (1964). The Meaning of Success: Some Differences in Value Systems of Social Classes. Journal of Social Psychology 62, 141 148. Kennedy, P. E. (2003). A guide to econometrics MIT press. Kerlinger, F. N., & Lee, H. B. (1986). Foundations of behavioral research. Kim, R. H. (2009). Evaluating Open Source Applications to Support Online Lear ning. Proceedings of the Southern Association for Information Systems Conference, Charleston, SC Kim, Y. K., & Kang, J. (2001). The effects of ethnicity and product on purchase decision making. Journal of Advertising Research, 41, 39 48. Koyuncu, C. & Li en, D (2003). E commerce and consumer's purchasing behaviour. Applied Econo mics, 35, 6, 721 727. Ittelson, W. H. (1973). Environment and Cognition New York: Seminar Press. James, W. G. & Kover, A. J. (1992). Observations: do overall attitudes toward ad vertising affect involvement with Specific ads? Journal of Advertising Research 32, 5, 78 83. Jarque, C. M., & Bera, A. K. (1980). Efficient tests for normality, homoscedasticity and serial independence of regression residuals. Economics Letters 6 (3), 2 55 259.
122 Jarvenpaa, S. L., Tractinsky, N. & Saarinen, L (1999). Consumer Trust in an Internet Store: A Cross Cultural Validation. Journal of Computer Mediated Communication 5, 2. John, D. R. (1999). Consumer socialization of children: A retrospective look at twenty five years of research. Journal of consumer research 26 (3), 183 213. Jung, T., Youn, H., & Mcclung, S. (2007). Motivations and self presentation strategies on Korean based" Cyworld" weblog format personal homepages. CyberPsychology & Behavior 10 (1), 24 31. Junk, V. W. & Young, M. H. (1999). Retention of Older In Migrants: Role of Family, Friends, and Community Services. Family and Consumer Sciences Research Journal 27, 3, 267 292. Lachance, M. J., Beaudoin, P., & Robitaille, J. (2003). Adole apparel: influence of three socialization agents. International Journal of Consumer Studies 27 (1), 47 57. Lang, P. J., Greenwald, M. K., Bradley, M. M., & Hamm, A. O. (1993). Looking at pictures: Affective, facial, visceral, a nd behavioral reactions. Psychophysiology, 30, 261 273. Lang, P. J. (1980). Behavioral Treatment and Bio Behavioral Assessment: Computer Applications in Technology. In J. B. Sidowski, J. H. John son, & T. A. Williams (Eds). Mental Health Care Delievery S ystems Norwood, NJ: Ablex, LaRose, R., Eastin, M. S., & Gregg, J. (2001). Reformulating the Internet paradox: Social cognitive explanations of Internet use and depression. Journal of Online Behavior 1 (2), 1092 4790. LaFerle, C., Edwards, S. M. & Lee, W. the Internet. Journal of Advertising Research 40, 3, 55 65. Lalwani, A. K. (2002). Interpersonal orientation of spouses and household purchase decisions: the case of restaurants. Service Industries Journal 22 (1), 184 200. Leung, L. (2010). Effects of Internet Connectedness and Information Literacy on Quality of Life. Social indicators Research 98 (2), 273 290. Lee, A. L. (2009). Changing Effects of Direct to Consumer Broadcast Drug Advertising Information Sources on Prescription Drug Requests. Health Communication 24: 361 376 Lee, C. K., Conroy, D. M., & Hii, C. (2003, December). The internet: a consumer socialization agent for teenagers. In ANZMAC Conference Proceedings (pp. 1708 1715).
123 Lee, C. K. C., & C onroy, D. M. (2005). Socialisation through consumption: Teenagers and the internet. Australasian Marketing Journal 13(1), 8 19. Lee, W. N. (1989). The mass mediated consumption realities of three cultural groups. Advances in consumer research 16 (1), 771 778. Lee, J. K. (2007). The effect of the Internet on homogeneity of the media agenda: A test of the fragmentation thesis. Journalism & Mass Communication Quarterly 84 (4), 745 760. Levy, S. J. (1999). Symbols for sale. reprinted in: Brands, Consumers, Sym bols & Research 203 212. Liao, T. F. (1994). Interpreting probability models: Logit, probit, and other generalized linear models (Vol. 101). Sage Publications, Incorporated. Loewenstein, J., Raghunathan, R., & Heath, C. (2011). The Repetition Break Plot Structure Makes Effective Television Advertisements. Journal of Marketing 75 (5), 105 119. Louis, W., Davies, S., Smith, J., & Terry, D. (2007). Pizza and pop and the student identity: The role of referent group norms in healthy and unhealthy eating. The J ournal of social psychology 147 (1), 57 74. Long, B. C. & Schutz, R. W. (1995). Temporal Stability and Replicability of a Workplace Stress and Coping Model for Managerial Women: A Multiwave Panel Study Journal of Counseling Psychology 42, 3, 266 278. L utz, R. J. (1985). Affective and Cognitive Antecedents of Attitude Toward the Ad: A Conceptual Frame work. In L. F. Alwitt & A. A. Mitchell (Eds.) Psychological Processes and Advertising Effects: Theory, Research and Application ( 45 63). Hillsdale, NJ: L awrence Erlbaum Associates, Lynn, M. & Harris, J. (1997). The desire for unique consumer products: a new individual differences scale. Psychology and Marketing 14, 601 616. MacKenzie, S. B. & Lutz, R. J. (1989). An Empirical Examination of the Structura l Antecedents of Attitude toward the Ad in an Advertising Pretesting Context. Journal of Marketing 53(2), 48 65 MacKenzie, S. B., Lutz, R. J. & Belch, G. E. (1986). The Role of Attitude toward the Ad as a Mediator of Advertising Effectiveness: A Test of C ompeting Explanations. Journal of Marketing Research 23 (2), 130 143 and behavior? Journal of Consumer Marketing, 17, 441 454.
124 Mangleburg, T. F., Grewal, D. & Bristol, T. (1997). Socialization, Gender, and Reports of Their Generalized Use of Product Labels. Journal of Consumer Affairs 31, 2, 255 279. Mangleburg, T. F. & Bristol, T Advertising. Journal of Advertising 27, 3, 11 21. Seminar on New Media. LIRNEasia, Colombo, Sri Lanka (Vol. 19, pp. 1 16). Marsh, H. W. (1993). Stability of individual differences in multi wa ve panel studies: Comparison of simplex models and one factor models. Journal of Educational Measurement, 30, 157 183. Mauldin, C (1976). Communication and the Aging Consumer. in H.J. Oyer and E.J. Oyer (eds.). Aging and Communication Baltimore: Univ ersity Park Press. Mascarenhas, O. A. & M.A. Higby (1993), "Peer, Parent, and Media Influences in Teen Apparel Shopping," Journal of the Academy of Marketing Science, 21, 53 58. McLeod, J. M. & Chaffee, S. R. (1972). The Construction of Social Reality. In J. T, Tedeschi, (Eds.) The Social Influence Process, Chicago, IL: Aldine Atherton. Mehrabian, A., & Russell, J.A. (1974). An approach to environmental psychology Cambridge: t.I.T. Press. McLeod, J. M. & Garret J. O., (1972). The Socialization Perspective a nd Communication Behavior. In G. Kline and P. Tichenor (Eds.) Current Perspectives in Mass Communication Research (121 168). Beverly Hills, CA: Sage, McNeal, J. U., & Ji, M. F. (1998). The role of mass media in the consumer socialization of Chinese child ren. Asia Pacific Advances in Consumer Research. Provo, UT: Association for Consumer Research 6 12. Morgan, C., & Cotten, S. R. (2003). The relationship between Internet activities and depressive symptoms in a sample of college freshman. CyberPsychology & Behavior, Morgan, M. & Signorielli, N. (1990). Cultivation Analysis: Conceptualization and Methodology. In N. Signorielli and M. Morgan (Eds.) Cultivation Analysis: New Directions in Media Effects Research. Newbury Park, CA: Sage. Morris, J. D., Woo, C. M., Geason, J. A. & Kim, J. (2002). The p ower of a ffect: Predicting i ntention. Journal of Advertising Research, 42 7 17. Morris, J. D., Woo, C, & Cho, C.H. (2003). Internet Measures of Advertising Effects: A Global Issue. Journal of Current Issues and Research in Advertising 25, 25 43.
125 Morris, J, Klahr, N., Shen, F., Villegas, J., Wright, P., He, G., & Liu, Y. (2008). Mapping a multidimensional emotion in response to television commercials. Human Brain Mapping, 3, 789 796. Morris, J. D. & W aine, C. (1993), Managing the Creative Effeort: Pre production and Post production Measures of Emotional Response. In the Proceedings of the Conference of the American Academy of Advertising, American Academy of Advertising. Morris, J. D. & McMullen, J. S (1994). Measuring Multiple Emotional Responses to a Single Television Commercial. In Chris T. Allen and Deborah Roedder John (Eds), Advances in Consumer Research Volume 21, (175 180). Association for Consumer Research. Provo, UT Morris, M & Ogan, C (19 96). The Internet as Mass Medium. Journal of Communication, 46, 39 50. Moschis, G. P. & Churchill, G. A., Jr. (1978). Consumer socialization: A theoretical and empirical analysis. Journal of Marketing Research 15, 500 609. Moschis, G. P. (1978). Teenage Journal of Retailing 54 (4), 80 93. Moschis, G. P., & Churchill Jr, G. A. (1979). An analysis of the adolescent consumer. The Journal of Marketing 40 48. Moschis G. P., & Moore, R. L. (1979). Decision making among the young: A socialization perspective. Journal of Consumer Research 6, 101 112. Moschis, G. P. (1981). Patterns of consumer learning. Journal of Academy of Marketing Science, 9, 110 117. Moschis, G. P. (1987). Consumer socialization: A life cycle perspective Lexington, MA: Lexington. Moore R L, & Moschis, G. P. (1981). The role of family communication in consumer learning. Journal of Communication 31(4), 42 51. Moore, R. L., & Stephens, L. F. (1975). Some communication and demographic determinants of adol escent consumer learning. Journal of Consumer Research 80 92. Moore Shay, E. S. & Berchmans, B. M. (1996). The Role of the Family Environment in the Development of Shared Consumption Values: An Intergenerational Study. Advances in Consumer Research 23 1, 484 490
126 Moscardelli, D., & Liston Heyes, C. (2005). Consumer socialization in a wired world: the effects of internet use and parental communication on the development of skepticism to advertising. Journal of Marketing Theory and Practice 62 75. Mosc ardelli, D. & Divine, R (2007). Internet: An Empirical Analysis of Predictors and Relationships With Privacy Protecting Behaviors. Family and Consumer Sciences Research Journal 35, 3, 232 252. Montgomery, K. C., & Chester, J. (2009). Interactive food and beverage marketing: targeting adolescents in the digital age. Journal of Adolescent Health 45 (3), S18 S29. Myers, K. K., & Sadaghiani, K. (2010). Millennials in the workplace: A communication perspective o Journal of business and psychology 25 (2), 225 238. Nowak, L., Thach, L., & Olsen, J. E. (2006). Wowing the millennials: creating brand equity in the wine industry. Journal of Product & Brand Ma nagement 15 (5), 316 323. Nowak, L. I., & Newton, S. (2008). Using winery web sites to launch relationships with Millennials. International Journal of Wine Business Research 20 (1), 53 67. Nolan, J. M., Schultz, P. W., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2008). Normative social influence is underdetected. Personality and social psychology bulletin 34 (7), 913 923. Nunnally, J.C. (1978). Psychometric Theory (2nd ed.) McGraw Hill: New York. O'Donohoe, S. (1995). Attitudes to advertising : a review of British and American research. International Journal of Advertising 14 245 245. O'Guinn, T. C. & Ronald J. F. (1989 ). Compulsive Buying: A Phenomenological Exploration. Journal of Consumer Research 16, 2, 147 157 O'Guinn, T. C., Faber, R. J., & Rice, M. (1985). Popular film and television as consumer acculturation agents: America 1900 to present. Historical Perspective in Consumer Research: National and International Perspectives 297 301. O'Guinn, T. C. & Shrum, L. J. (1997). The Role of Television in the Construction of Consumer Reality. Journal of Consumer Research 23, 278 294. stman, J (2012). Information, expression, participation: How involvement in user generated content relates to democratic engagement among young people. New M edia & Society 14: 1004 1021
127 Ownbey, S. F., & Horridge, P. E. (1997). Acculturation levels and shopping orientations of Asian American consumers. Psychology & Marketing, 14, 1 18. Park, C., & Lee, T. M. (2009). Antecedents of online reviews' usage and pu rchase influence: an empirical comparison of US and Korean consumers. Journal of Interactive Marketing 23 (4), 332 340. Park, C. W., & Lessig, V. P. (1977). Students and housewives: Differences in susceptibility to reference group influence. Journal of Co nsumer Research 102 110. Park, D.J., Villar, M.E., & Amador, J. (2010). TV and Internet Usage Among African Americans and Their Effects on Material Values and Socialization Processes. Paper presented at the Communication and Technology Division for the 2010 annual conference of the International Communication Association, Singapore. Parsons, T., Bales, R. F., & Shils, E. (1981). Working papers in the theory of action Greenwood Press. Pasche, F (2008). Some methodological reflections about the study of religions on video sharing websites. Marburg Journal of Religion 13, 1, 1 10. Pew Research. (2010). Millennials, a portrait of generation next. Retrieved from http://pewsocialtrends.org/files/2010/10/millennials confident connected open to change.pdf Petty, R. E. & Cacioppo, J. T. (1996). Addressing Disturbing and Disturbed Consumer Behavior: Is It Necessary to Change the Way We Conduct Behavioral Science. Journal of Marketing Research 33, (February), 1 8. Phelps, J. E., Lewis, R., Mobilio, L., Perry, D., & Raman, N. (2004). Viral marketing or electronic word of mouth advertising: Examining consumer responses and motivations to pass along email. J ournal of advertising research 44 (04), 333 348. Phillips, L. W. & Sternthal, B (1977). Age differences in information processing: a perspective on the aged consumer. Journal of Marketing Research 14, 4, 444 457. Pincus, S. & Waters, L.K. (1977). I nformational Social Influence and Product Quality Judgments. Journal of Applied Psychology 62 (5), 615 619. Potter, W. J. (1991). Examining cultivation from a psychological perspective. Communication Research 18, 92 113 Potter, W. J. (1993). Cultivation theory and research. Human Communication Research 19 (4), 564 601.
128 Pollay, R. W. (1986). The Distorted Mirror: Reflections on the Unintended Consequences of Advertising. Journal of Marketing 50, 2, 18 36. Pollay, R. W. & Mittal B (1993) Here's the bee f: factors, determinants, and segments in consumer criticism of advertising. ]ournal of Marketing, 57, 99 114. Prensky, M. (2001). Digital natives, digital immigrants Part 2: Do they really think differently? On the horizon 9 (6), 1 6. Pregibon, D. (1980). Goodness of link tests for generalized linear models. Applied Statistics Putnam, R. D. (1993). Making democracy work: Civic traditions in modern Italy Princeton, NJ: Princeton University Press. cial capital. Journal of Democracy, 6 65 78. Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community New York: Simon & Schuster. Raajpoot, N. A., Sharma, A., & Chebat, J. C. (2008). The role of gender and work status in shoppi ng center patronage. Journal of Business Research 61 (8), 825 833. Ramirez A. R., Lin, S & Dimmick, J. (2004). Revisiting Media Competition : the Gratification Niches of Instant Messaging, Email and the Telephone', paper presented at the Annual Meeting of the International Communication Association, New Orleans, LA, 27 31 May Rafaeli, S (1988). Interactivity: From new media to communication. In R. Hawkins et al. (Eds.), Advancing communication science: Merging mass and interpersonal processes (110 134). N ewbury Park, CA: Sage. Rafaeli, S & Sudweeks, F (1997). Networked Interactivity. Journal of Computer Mediated Communication 2, 4, 1 17 Rainer T. S. & Rainer J. (2011). The Millennials: Connecting to America s Largest Generation. B & H Books: Nashville, T N. Reddy, S. K. (1992). Effects of ignoring correlated measurement error in structural equation models. Educational and Psychological Measurement, 52, 549 570. Rego, L. L., Billett, M. T., & Morgan, N. A. (2009). Consumer based brand equity and firm risk. Journal of Marketing 73 (6), 47 60. Riesman, D., & Roseborough, H. (1955). Careers in Consumer Behavior. In L. Clark (Eds.) Consumer Behavior, Vol. 2. New York: New York University Press.
129 Richins, M. L. (1987). Media, Materialism, and Human Happiness. In M elanie Wallendorf & Paul Anderson (Eds.), Advances in Consumer Research Volume 14 (352 356 ). Provo, UT : Association for Consumer Research. Richins, M. L. (1995). Social comparison, advertising, and consumer discontent. American Behavioral Scientist 38 (4 ), 593 607. Richins, M. L. (2004). The material values scale: measurement properties and development of a short form. Journal of Consumer Research, 31, June, 209 219 Rindfleisch, A., Burroughs, J. E., & Denton, F. (1996). Family disruption and consumer att itudes and behavior: an exploratory investigation. Advances in Consumer Research, 23, 83 90. Rindfleisch, A., Burroughs, J. E., & Denton, F. (1997). Family structure, materialism, and compulsive consumption. Journal of consumer research 312 325. Rindfle isch, A (1994). Cohort Generational Influences of Consumer Socialization. Advances in Consumer Research 21, 470 476. Richards, R (2006). Users, interactivity and generation. New Media & Society, 8, 531 550. Robins, G., Pattison, P., & Elliott, P. (200 1). Network models for social influence processes. Psychometrika 66 (2), 161 189. Ross, G. M. (1999). Consumer Socialization, Parental Style, and Developmental Timetables in the United States and Japan. Journal of Marketing 63(3), 105 119 Rose, G. M., Bu sh, V. D. & Kahle, L (1998). The influence of family communication patterns on parental reactions toward advertising: a cross national examination. Journal of Advertising 27(4), 72 85. Rose P. (2007). Mediators of the association between narcissism an d compulsive buying: the roles of materialism and impulse control. Psychology of Addictive Behaviors, 21, 576 581. Russell, J., & Mehrabian, A. (1977). Evidence For A Three Factor Theory of Emotions, Journal of Research in Personality, 11, 273 294. Rumbau t, R. G. (2005) On the Frontier of Adulthood: Theory, Research, and Public Policy. Chicago: University of Chicago Press. Schroeder, J. (1996). An analysis of the Consumer Susceptibility to Interpersonal Influence Scale. Journal of Social Behavior and Perso nality 1 (3), 585 600. Schlosser, A. E., Shavitt, S., & Kanfer, A. (1999). Survey of Internet users' attitudes toward Internet advertising. Journal of Interactive Marketing 13 (3), 34 54.
130 Segrin, C., & Nabi, R. L. (2002). Does Television viewing cultivat e unrealistic expectations about marriage? Journal of Communication 52, 247 263. Shankar, V & Meyer J (2009). Internet and International Marketing. In Kotabe Masaki, Helsen Chritiann (Eds). Handbook of International Marketing, Sage. P. 451 467. Shrum, L J. (1996). Psychological processes underlying cultivation effects: Further tests of construct accessibility. Human Communication Research 22, 482 509. social reality : accessibility as an explanatory variable. Communication Research 20, 436 471. Shrum, L. J., Burroughs J. E. & Rindfleisch, A. (2005). Material Values. Journal of Consumer Research 32, 3, 473 479 Siegel, A. E. (1958). The i expectation. Child Development, 29, 35 56. Singh, N Kwon, I & Pereria, A ( 2003). Cross Cultural Consumer Socialization: An Exploratory Study of Socialization Influences across Three Ethnic Group s. Psychology & Marketing, 20, 10, 867 881 Skinner, B.F. (1953), Scientific and Human Behaviour New York, Macmillan. In East, R. (1990), Changing Consumer Behaviour Cassell Educational Limited Smith, R. B. & Moschis G. P. (1984). Consumer Socialization on the Elderly: An exploratory study. In Thomas C. Kinnear (Eds.) Advances in Consumer Research Volume 11 (548 552) Provo, UT : Association for Consumer Research, Solomon, M R. & Basil G. E. (1993). Consumption constellations: implications for integrated communication strategies. In E. Thorson and J. Moore (Eds .) Interated marketing communications: synergy of persuasive voices Hillsdale, NJ: Lawrence Erlbaum Sotirovic, M. (2001). Media use and perceptions of welfare. Journal of Communication, 51, 750 77 4. Stangor, C., Sullivan, L. A. & Ford, T. E. (1991). Affective and cognitive determinants of prejudice. Social Cognition, 9,359 381. Steven H. C & Miriam J. M (2001). The End of Mass Communication?, Mass Communication and Society 4 (4), 365 379 Stainfiel d, C., Ellison, N. & Lampe, c. (2008). Social Capital, self esteem, and use of online social network sites: a longitudinal analysis. Journal of applied developmental psychology 29, 434 445.
131 Stayman, D. & Batra, R. (1991). Encoding and Retrieval of Ad Aff ect in Memory. Journal of Marketing Research 28. 232 39. Stefanone, M A. & Jang, C (2008). Writing for Friends and Family: The Interpersonal Nature of Blogs Journal of Computer Mediated Communication 13, 123 140 Subrahmanyam, K Reich, S. M., Waechter N. & Espinoza, G (2008). Online and offline social networks: Use of social networking sites by emerging adults Journal of Applied Developmental Psychology 29 (6), 420 433 Sussman, A B., & Olivola, C. Y. (2011). Axe the tax: Taxes are disliked more than equivalent costs. Journal of Marketing Research 48 (SPL), 91 101. Tannen, D (1990). New York, NY: William Morrow & Co. Tann en, D (1995). The Power of Talk: Who gets heard and why. Harvard Business Review 73 (5), 138 148. Tapscott, D. (1998). Growing up Digital: The Rise of the Net Generation New York: McGraw Hill. Tang, C. S. K., Wu, A. M., Yeung, D., & Yan, E. (2009). At titudes and intention toward old age home placement: A study of young adult, middle aged, and older Chinese. Ageing International 34 (4), 237 251. Temkin, B.D. & Popoff Walker, R. (2007). The Gen Y Design Guide Forrester Research: Cambridge, MA. Thompson J.B. (1995 ) The Media and Modernity: A Social Theory of the Media. Cambridge: Polity Press. Thornton, A. (1989). Reciprocal influences of family and religion in a changing world. Journal of Marriage Family 47, 381 394. Tiggemann, M (2005). Television and adolescent body image : The role of program content and viewing motivation. Journal of Social and Clinical Psychology 24, 3, 361 381. Trusov, M Bodapati, A. V. & Bucklin, R. E. (2010). Determining Influential Users in Internet Social Networks. Journ al of Marketing Research 47, 643 658 Valentova, J., Rieger, G., Havlicek, J., Linsenmeier, J. A., & Bailey, J. M. (2011). Judgments of Sexual Orientation and Masculinity Femininity Based on Thin Slices of Behavior: A Cross Cultural Comparison. Archives of sexual behavior 40 (6), 1145 1152.
132 Villani, S. (2001). Impact of media on children and adolescent: a ten year review of the research Journal of the American Academy of Child and Adolscent Psychiatry, 40, 4, 392 401. Ward, S (1974). Consumer socializat ion. Journal of Consumer Research 1 (2), 1 14. Ward, S & Wackman, D (1971). Family and Media Influence on adolescent consumer learning. American Behavioral Scientist 14 (4). 415 427. Ward, S & Wackman, D (1973). Effects of Television Advertising on Cons umer Socialization. Cambridge, Mass.: Marketing Science Institut. Ward, S. Klees, Donna M. & Wackman, D. B. (1990), Consumer socialization research: Content analysis of post 1980 studies and some implications for future work. In Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay (Eds.) Advances in Consumer Research Volume 17 (798 803.), Provo, UT : Association for Consumer Research Ward, S ., Klees, W., Daniel B. & Wartella, E. (1977). How children learn to buy: the development of consumer information processing skills Berverly Hills, CA: Sage. Ward, S ., Klees, Donna M. & Robertson, T. S. (1987). Consumer Socialization in Different Settings : An International Perspective. Advances in Consumer Research 14, 468 472 White, H. (1980). A heteroskedasticity consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica: Journal of the Econometric Society 817 838. Wooldridge, J. M. (2009). Introductory econometrics: A modern approach South Western Pub. Wright, P. (1974). The u se of phased, noncompensatory strategies in decisions between multiattribute products Unpublished working paper, Stanford University, California Xie, Q, Zhang, M & Morris, J. (2011). Find us on Facebook: Emotional and Cognitive Responses Towards Faceboo the Association for Business Communication Division of the National Communication Association Annual Conference New Orleans, Louisiana, Novermber 2011. Xie, Q (2012). Share it on Facebook: Examining Viral Mar keting and Motives to Pass Along Electronic Content on Social Networking Sites. Paper accepted by 2012 Annual Conference of American Academy of Advertising Myrtle Beach SC, March 2012.
133 Yang B & Lester, D (2004). Attitudes toward buying online. CyberPsyc hology & Behavior, 7(1), 85 91. Yang, H & Oliver M. B (2010). Exploring the Effects of Television Viewing on Perceived Life Quality: A Combined Perspective of Material Value and Upward Social Comparison. Mass Communication and Society, 13:118 138 Yang, Y & Cho, M. J. (2000). Effect of Word of Attitude Change. Korean Journal of Advertising 11 (3), 7 34. Yoon, K. (1995). Comparison of beliefs about advertising attitude toward advertising and materialism held by Africa n Americans and Caucasians. Psychological Report 77, 455 466. YouTube. (2012) Press Room (Statistics). Retrieved from http://www.youtube.com/t/press_statistics Zanot, E (1984). Public Attitudes T oward Advertising: The American Experience. International Journal of Advertising 3 (1), 3 15. Zhou, X (2009). Culture and Consumer Socialization. Retrieved from ProQuest Digital Dissertations. (AAT 3383240 )
134 BIOGRAPHICAL SKETCH Qinwei Vivi Xie was born in Guiyang China to Ying Xie and Guihui Zhong After completing her Bachelor s of Science in Business Administration at Guangdong University of Foreign Studies in 2007 Vivi came to the Unite d States to pursue her Masters of Arts in International Business at the University of Florida She returned home to work for the Guangdong Television Network for six months before pursuing her Ph.D. in Mass Communication. Vivi specializes in cross cultural advertising, emotional responses to marketing communications and new media advertising. During her final semester in the Ph.D. program she received a dissertation award from the University of Florida to support this research.