\ WWW W ) 5&2; \HUIWWHHN‘UWHHWIWW This is to certify that the thesis entitled THE EFFECTS OF FAMILIARITY AND SIMILARITY ON PSYCHOLOGICAL DISTANCE TOWARD TRUSTWORTHINESS OF AN ONLINE COMPANY presented by LIBRARY chigan State University i Jin Kyun Lee M has been accepted towards fulfillment of the requirements for the Master of Arts degree in Advertising ’ Major Professor's Signature 7/7/2005 Date MSU is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 2/05 c:/ClRC/DateDue.tndd-p.15 THE EFFECTS OF FAMILIARITY AND SIMILARITY ON PSYCHOLOGICAL DISTANCE TOWARD TRUSTWORTHINESS OF AN ONLINE COMPANY By Jin Kyun Lee A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Advertising 2005 ABSTRACT THE EFFECTS OF FAMILIARITY AND SIMILARITY ON PSYCHOLOGICAL DISTANCE TOWARD TRUSTWORTHINESS OF AN ONLINE COMPANY By Jin Kyun Lee As online purchasing behavior continues to become more and more common these days, the perception of a company online that consumers interact with was extensively researched in the online advertising field. This study identified the role of psychological distance toward an online company as well as its preceding variables, such as company familiarity, brand familiarity, geographical distance, and congruent city images. This study applied the psychological distance concept that was studied in the interpersonal psychology and international trade field to the online advertising field, where this concept has not been explored. Additionally, the mediating effect of psychological distance between familiarity effects and the trustworthiness of an online company was examined. It was found that company familiarity and brand familiarity affect psychological distance. Even in the online enviromnent, consumers prefer to buy familiar brands from a highly recognized online company. However, it was found that the geographical distance and city of origin did not affect psychological distance. In addition, the mediating role of psychological distance between familiarity effects and trustworthiness was identified and proved its partial mediation. This thesis concludes with a discussion of limitations and directions for future research. ACKN OWLEGEMENT S I would like to thank the faculty and staff of the Department of Advertising at Michigan State University for providing me the educational seedbed to continue my research in such a fine setting with superior professors and their supports. Particularly, I would like to thank Dr. Edwards, my committee chairperson with . whom I worked through my master courses. His passion, attitude, and experience imbued my mind with invaluable inspirations to be a prospective researcher. Also, I would like to thank Dr. La Ferle, my academic advisor. Her guidance always gave me a motivation and encouragement when I need it. I would like to thank Dr. Lee for her support and help with last minute thesis drafts. Finally, I would like to thank my parents, family, Ji Eun, and friends. My parents have always supported me mentally and financially during my studies in the United States. Also, I would like to thank Ji Run for her love and patience for the past one- year. And the friends I have made while playing tennis were perfectly refreshing me. Thank you all. iii TABLE OF CONTENTS LIST OF TABLES ............................................................................... V LIST OF FIGURES ............................................................................. VI CHAPTER]: INTRODUCTION Introduction ............................................................................... 1 CHAPTER 2: PSYCHOLOGICAL DISTANCE IN ONLINE ADVERTISING Psychological Distance from the Interpersonal Perspective ........................ 3 Psychological Distance from the International Trade Perspective ................ 6 Psychological Distance in Advertising Online ....................................... 9 CHAPTER 3: THE ANTECEDENTS AND CONSEQUENCE OF PSYCHOLOGICAL DISTANCE Antecedents: Company Familiarity and Brand Familiarity ....................... 12 Antecedents: Similarity between Consumers and an Online Company. . . . l 5 Consequence: Trustworthiness of an Online Company ............................ 21 CHAPTER 4: METHODOLOGY Pretest ................................ , .................................................... 24 Data Collection .......................................................................... 24 Sample ................................................................................... 25 Experimental Design and Manipulations ............................................ 26 Dependent Variables ................................................................... 28 Covariate ................................................................................ 29 Analysis Methods ...................................................................... 30 CHAPTER 5: RESULTS Data Description ........................... ' ............................................. 31 Manipulation Checks ................................ , ................................... 3 1 Measurement Development ........................................................... 32 Hypothesis Testing ..................................................................... 35 Additional Analysis: Mediating Role of Psychological Distance ................ 38 CHAPTER 6: DISCUSSION Research Implications ................................................................. 40 Limitations and Future Research ..................................................... 43 BIBLIOGRAPHY .............................................................................. 48 LIST OF TABLES Table l: Manipulation Checks ................................................................. 32 Table 2: Factor Analysis for Company Familiarity, Congruent City Image, Brand Familiarity, and Geographical Distance .................................... 33 Table 3: Factor Analysis for Trustworthiness and Psychological Distance .............. 34 Table 4: Correlation Tests ...................................................................... 34 Table 5: Mediating Effects of Psychological Distance ..................................... 39 Table 6: ANCOVATest for Hypotheses Testing ....... 46 LIST OF FIGURES Figure l: Hypothetical Model ................................................................. 23 Figure 2: Interaction Effect with Geographical Distance .................................. 36 Figure 3: Interaction Effect with Brand Familiarity ......................................... 37 Figure 4: Stimuli Advertisements Sample ................................................... 47 vi I. Introduction 1. Introduction With the fast development of mass communication and transportation, consumer behavior has changed drastically. Widespread intemet use greatly impacts not only the development of local and global economy, but also individual consumption behavior. Online consumption behavior has received a great amount of attention from various fields of study. However, no attempts have been made to apply the psychological distance concept in the online advertising field. The concept of psychological distance has been defined numerous ways in many fields of study. The psychological distanCe concept is quite elusive to capture in specific terms given the different approaches of each field. In the international trade field, psychological distance was defined as the adaptive capability of a global company from the organizational perspective. Whereas, in the interpersonal . psychology field, this concept focused on being connectedness or psychOlogical bonds between people. In this study, psychological distance will be redefined to see if individual consumers feel distance toward an online company. Also, familiarity and similarity effects will be examined because these variables are expected to play an important role in affecting psychological distance toward an online company. The lack of consumer familiarity, such as company and brand familiarity prevents consumers from actively engaging in online buying participation. Unconsciously, consumers tend to buy products online without exactly knowing products’ tangible attributes and features in detail. All consumers can do is to believe product descriptions or refer to the previous buyers’ ratings about the product quality 1 or reputation of an online company. In terms of similarity, this research assumes that close geographical location of online company and congruent city image negatively affect psychological distance. So far, the issues of geographical distance and the congruent city image were not clearly discussed because most research overlooked the importance of location-based characteristics and underlying similarity effects. Even though consumers can access the intemet in an online space, they often forget that tangible products are manufactured and shipped from a certain city. Thus, this research will identify whether consumers consider the geographical distance and city of origin effects as long as price, delivery time, and other conditions are the same. Obviously, individual perception causes different evaluation of online companies. In this research, the role of familiarity and similarity effects on psychological distance toward companies will be identified. II. Psychological Distance in Online Advertising 1.Psychological Distance from the Interpersonal Perspective The concept of distance constitutes a geographical metaphor used to describe personal experience. Usually, this concept refers to a lack of connection between two people (Kreilkamp 1981) or a cognitive state of having weaker psychological bonds (Hess 2002). This concept has been identified as an important dynamic in human relationships in many areas, such as human communication, psychology, sociology, counseling and therapy, and family studies (Hess 2002; 2003). At first, the concept of distance gained its attention when Bogardus studied social distance among groups (Weinfurt and Moghaddam 2001). Bogardus introduced the social distance scale for use as an index of the social distance that people perceive between themselves and members of different groups defined by nationality, ethnicity, religion, or politics (Weinfurt and Moghaddam 2001). However, many scholars traced the ideas of anthropologist and social psychologist, Erving Goffman who concentrated on identifying human interactions, such as how people maintain faces and how social conventions govern behaviors. Based on his observation, people create a sense of psychological distance between themselves and others. Some of the more well known of these include avoidance, inattention, focusing away from the other, and cognitive disassociation (Hess 2002; 2003) People who feel psychological distance Show specific distancing behaviors that minimize potential conflicts with others because the actual space that people are trying to create is the key variable in human relationships (Hess 2002; 2003). Even 3 though it might be possible that actual distancing behaviors and psychological distance are different in the conceptual level, spatial distancing behaviors can actually be subsumed into psychological distance, because manipulation of physical distance is just one method by which people can regulate perception of distance in a relationship (Hess 2002; 2003). Thus, people who have psychological distance tend to regulate their behaviors in a way to reduce unnecessary conflicts with others. Hess (2002; 2003) studied distancing behaviors and identified that three approaches people use to remain separated from others: avoidance, disengagement, and cognitive dissociation. First, avoidance was found to be the most commonly employed tactic. Avoidance can be defined as behaviors to spend less time and have less interaction with other person. Avoidance could be accomplished either by eliminating interaction episodes or by avoiding interaction with other people. People try to prevent an interaction from happening by changing behaviors in an attempt to avoid the other person or ignoring the presence of others. Disengagement behavior was the second most commonly used. Disengagement can be defined as behaviors that reduce the amount of association or affiliation between each other during interactions (Hess 2002). Goffman (1967) suggests that disengagement occurs when a person might become psychologically alienated from an unfocused interaction, where the people are not attentive or where the interactions feel impersonal. In such situations, people disengage themselves by hiding information, using deception, and inattention to remain unengaged. Also, people can disengage themselves from others as a way to interact less personally. As mentioned above, avoidance and disengagement behaviors are overtly presented in behavioral dimensions. These two distancing behaviors are the easiest for people to do, and thus, are likely to be the first type of distancing behaviors used (Hess 2002). Third, people distance themselves from others by using cognitive dissociation. Cognitive dissociation can be defined as a cognitive process as a way to change perceptions, rather than outward behaviors (Hess 2002). Research on interpersonal psychology has Shown that the management of meaning, such as cognitive distancing, is also an important part of the relational experience. People might disregard messages or detach cognitively or emotionally from others to maintain attitudes that are consistent with their behaviors. The meaning one assigns to actions is as important as avoidance and disengagement behaviors for their cognitive consistency (Heider'195 8). For example, there is research studying reasons Jewish people buy German cars (Woodside and Chebat 2001). Even though Jewish consumers could have antipathy toward German people, they might have a favorable attitude toward the brand of German car. In that situation, Jewish consumers can balance their attitude either by beginning to like German people or not buying German car. Overt distancing behaviors tend to be accompanied with cognitive perceptions about relational partners who the people interact with in order to maintain consistency. When relational characteristics create conflicts, people are motivated to restore harmony by changing behaviors or by changing perceptions. When people find themselves desiring more distance in the relationship, it should be expected that they are likely to change their perceptions. . Cognitive psychologists contend that people streamline their cognitive process when possible (Tversky and Kahneman 1986). If the results are same, people prefer cancel their options to make their decisions easier. The rationale behind this is that it takes less effort to change the perception of the relationship than to conceive every possible way to enact alternative behavioral tactics. Thus, before going to the effort of engaging in distancing behaviors, people would be expected to try simple cognitive dissociations. In other words, people can perceive themselves as having weaker bonds without actually changing their behaviors. Any relationship between people has some levels of psychological distance. Despite the commOn assumption that distance is detrimental to interpersonal relations, many scholars argue that psychological distance is also needed for healthy close relationships. Although avoidance behavior is typically related to negative affects from people who are disliked, people also may distance themselves as possible protection from someone who is likely to deliver hurtful messages. Some of these people who make hurtfirl comments are loved ones, such as romantic partners or family members (Vangelisti, Corbin, Lucchetti, and Sprague 1999). Thus, in relationship dynamics, both closeness and distance are essential to balance out healthy relationships. Psychological distance prevents people from freely approaching what matters to them (Kreilkamp 1981). As psychological distance decreases, people are freer to discuss sensitive issues that are not mentioned with people we avoid. This phenomenon happens unconsciously without knowing exactly why it is done. Reduced psychological distance decreases the sense of danger, lower barriers and self-defense mechanisms as well as provides feelings which are real, genuine, more open, and trusting. Thus, from the viewpoint of interpersonal psychology, the psychological distance concept is related to how people create and regulate behaviors in order to maintain relationships with liked or disliked people. 2. Psychological Distance from the International hade Perspective Psychological distance has also been studied in the international business field in order to develop theory on the internationalization process of a firm. In this context, the psychological distance has been defined as factors preventing or disturbing the flow of information between a firm and foreign markets (Johanson and Wiedersheim-Paul 1975; J ohanson and Jan-erik Vahlne 1977; Nordstrom and Vahlne 1994, Evans and Mavondo 2002). These difficulties that companies face are related to the lack of direct communications with existing and potential customers or a lack of access to general market information. Psychological distances results from factors, such as differences in language, education, business practices, culture, religion, political systems, and industrial development (Carlson 1974; Johanson and Vahlne 1977; Johanson and Widersheim-Paul 1975; Dow 2000). In the international trade literature, there has been a strong tradition of using geographical distance as a surrogate indicator of all trade resistance factors (Dow 2000). Carlson (1974) provided a theoretical background for the use of geographical distance as a surrogate indicator of psychological distance. Geographical distance affects the collection and transmission cost of important management information. Many of the characteristics of information are tacit in nature and need face-to-face communications so that senior management time becomes a major component of collection costs. Nordstrom and Vahlne (1994) used geographical distance as a psychological distance indicator in that companies perform best in foreign markets most similar to their domestic market. The negative relationship between psychological distance and organizational performance is attributed to the fact that physically markets of close countries are easier to learn about and easily understood. In fact, Dow (2002) determined that geographical distance was a significant predictor of export market selection. As distance increases, export markets are less likely to be selected. At the same time, many researchers also have used scales based on Sethi’s clustering of world markets. Sethi (1971) analyzed data for 29 variables and 91 countries which resulted in four variable clusters: aggregate production and transportation, personal consumption, trade, and health and education. First, aggregated production and transportation dimension was macro indicators of the economic and business activity in a nation. This dimension includes electricity production, urbanization, the number of newspapers, air passengers, and air cargos. Second, personal consumption dimension represents micro indicators, such as per capita ownership of cars, radios, television sets, telephones, energy consumption, hospital beds, school enrollment, education per capita, and newspaper circulation. Third, trade dimension represents the interdependence of nations through trade and includes indicators, such as exports, imports, and consumer price index. Finally, health and education dimension includes indicators, such as life expectancy, physicians per capita, illiteracy among adults over age 15, and the percentage of population in agriculture. Based on the clustering of variables, Sethi categorized 91 countries into seven country clusters that were found to demonstrate homogeneous characteristics between countries. These country clusters were more accurate in predicting export market selection than just relying on geographical proximity. Relying on geographical proximity assumes that adjoining countries have similar characteristics such as culture, religion, socio-economics, politics, and marketing and distribution networks regardless of cultural idiosyncrasies. To add additional information to the examination of international trade, Hofstede’s (1983; 1991) cultural difference dimensions also have been used as surrogate indicators of psychological distance. These dimensions include power distance, individualism versus collectivism, masculinity versus femininity, uncertainty avoidance, and long-term orientation. A nation’s culture can be defined as collective programming of the mind which differentiates the members of one group or category of people from those of another (Hofstede 1983; 1991). Originally, Hofstede analyzed data collected by IBM and tried to find an explanation for the fact that some concepts of motivation did not work in all countries in the same way. Hofstede’s five dimensions explain why people and organizations in various countriesodiffer. Up until now, Hofstede’s dimensions have been found to be stable, even though socio-political changes in each country have had some effects on the cultures (De Mooij 1998). These dimensions describe how similar or different a country’s culture from another. This similarity is really akin to the psychological distance of countries’ cultures. Still, most researchers agreed that neither geographical distance and Sethi’s cluster analysis, norHofstede’s cultural dimensions completely captured the psychological distance concept in the international trade field. 3. Psychological Distance in Advertising Online Any attempt to apply the psychological distance concept into the advertising field was not yet made. Given that the concept of psychological distance has never been explored in the online advertising field, it will be important to suggest how the psychological distance construct can be applied in online advertising and to identify how this concept can be interrelated with other concepts, such as familiarity, similarity, and trustworthiness. In this study, there are two main reasons why the perspective of interpersonal psychology will be used to examine the online consumer behavior. First, the focus of this research is within the boundary of individual consumer perception and behavior. In the international trade field, the psychological distance concept has been approached from the organizational perspective rather than focusing on the cognitive process of individual consumers. The concept of psychological distance indicates difficulties that international companies are facing in accessing market information in different countries due to geographical distance, cultural difference, language, education, business practices, religion, political systems, and industrial development. In other words, the psychological distance was approached from the macro perspective and was defined as one of impeding factors for the effective organizational operation in foreign markets. The locus of discussion is totally different and beyond our areas of interests because the research purpose is identifying the individual consumer’s perception and behavior online. Thus, the conceptual definition in interpersonal psychology is more relevant than that of the international trade field. Second, it is expected that individual consumers Show distancing behaviors toward online companies. As people who have psychological distance toward other people are likely to avoid, disengage, and cognitively dissociate themselves from disliked people in order to maintain cognitive consistency. In the same way, consumers will also have psychological distance toward online companies. Consumers who feel psychological distance toward an online company are expected to show specific behaviors that minimize potential conflicts with that online company. Consumers might try to avoid, disengage, and cognitively dissociate themselves from online companies based on their subjective evaluations. For example, consumers might trust SONY brand in general. However, they might not buy that product if the brand was sold at an online company that they had never heard of before (Woodsite and Chebat 2001). 10 Given the purpose of identifying the underlying mechanism of individual consumers’ cognitive perceptions and behaviors toward an online company, it will be more desirable to adapt the perspective of interpersonal psychology rather than the organizational perspective that was used in international trade field. It is predicted that consumers will have distance perceptions toward an online company depending on the levels of familiarity and similarity with the online company. Thus, the micro approach from interpersonal psychology provides this research with concrete theoretical and appliCable backgrounds. The main purpose of this research is to demonstrate the effects of familiarity and similarity on psychological distance. This research will contribute to the field of online consumer behavior for several reasons. First, this is the first research that applies the psychological distance to an online situation. In an online environment, consumers have quasi-physical relationships with companies. The absence of the physical existence of brick and mortar store and face-tO-face communicatiOns does not mean that there are no relationships between consumers and companies that sell products online. It is assumed that consumers will perceive different levels of psychological distance to the extent that how familiar they are with the online companies and what they have in common with. Second, the concept of psychological distance is approached from the consumers’ perspective in relation to the role of perceived familiarity and similarity. The research results will not only provide the reasons why consumers feel distance or closeness when they interact with online companies, but also suggest marketing implications on how consumers’ perceptions should be considered in developing and implementing marketing strategies and tactics. 11 III. The Antecedents and Consequence of Psychological Distance 1. Antecedents: Company Familiarity and Brand Familiarity The concept of familiarity has been studied extensively in various fields of social sciences. Knowing a person or an object means increased knowledge structure, thereby, affecting consumer information processing activities in several ways (Alba and Hutchinson 1987). The impact of familiarity on consumers’ information processing has been a feature of traditional and recent information processing theories of consumer choice (Howard and Sheth 1969; Bettrnan 1979). It has been argued that familiarity facilitates the acquisition of new information as well as the use of existing information (Park and Lessig 1981). V In terms of terminology, the term of familiarity has been used interchangeably with expertise and experience when refen'ing to prior knowledge (Rao and Monroe 1988). However, Alba and Hutchinson (1987) suggest that consumer knowledge has two components: familiarity and expertise. Familiarity is defined as the number of product-related experiences accumulated by consumers, and expertise is the ability to perform product-related tasks successfully (Rao and Monroe 1988). In general, product experience is a necessary, but insufficient condition for consumer expertise. In this research, consumers’ familiarity consists of brand familiarity and company familiarity. Theoretically, conSumers who are becoming more familiar with companies and brands are likely to undergo several cognition-related changes. In particular, increasing familiarity leads to a more elaborated cognitive structure due to increased knowledge (Alba and Hutchinson 1987) In this phenomenon, familiarity can be explained that the more often people are exposed to a certain stimulus, the 12 more positively they will evaluate it, and therefore, feel less psychological distance toward the stimulus. Specifically, attitude polarization theory supports how familiarity affects the evaluation of companies and brand online. Zajonc (1968) is generally acknowledged as having initiated a substantial amount of research on attitude polarization effects. He found that more exposure to a stimulus enhances attitudes toward the object as long as the previous exposure generates positive attitudes. If previous encounters have been negative, further exposure seems to produce increasingly negative evaluation (Burgess and Sales 1971; Suedfeld, Epstein, Buchanan and Landon 1971). Thus, it is also expected that when the outcome of the evaluation of the additional encounter is positive, the consumers who have spent relatively more time thinking about the company tend to be more positive than the consumers who have spent relatively less time thinking about the company. On the other hand, when the outcome of the evaluation of the additional encounter is negative, the consumers who have spent relatively more time thinking about the company tend to be more negative than the consumers who have spent relatively less time thinking about the company (Tesser 1978). Individuals with extensive networks of knowledge reflecting substantial prior thought, are likely to make inferences that are more polarized than individuals with a less elaborate knowledge structure. In contrast, a limited knowledge structure, which reflects little prior thought, produces moderate rather than extreme effect (Peracchio and Tybout 1996). As long as consumers are committed to form an attitude of companies and brands, a polarization pattern is expected because an individual feels a need to judge the additional stimulus so that the evaluation is consistent with prior evaluation. 13 Grush (1976) insisted that when a stimulus is unfamiliar, it elicits only a few associations. With repeated exposures, subjects generate increased associations to the stimulus. Thus, the final evaluation of the stimulus is assumed to be a function of the accumulated evaluations of its associations. As exposure increases, the summed evaluations of positive stimuli become more positive, and the summed evaluation of negative stimuli become more negative (Séderlund 2002). The rationale behind this explanation is that consumers tend to avoid risks by preferring the familiar to the unknown (Bomstein 1989). Thus, consumers who have positive impressions of companies will continuously accumulate positive evaluations as the number of interactions increase online. Given the fact that most commercial online companies are consumer-oriented, they try to consistently convey positive images of their companies. In other words, none of online company hopes to build negative attitude of consumers that will lead to decrease their sales profit in the future. Thus, the more familiarity consumers have, the more they are likely to accumulate positive evaluation of companies and brands (Tessor 1978). The rationale is based on the premise that positive schema are activated given the positive direction of an additional stimulus. It means that positively accumulated experiences are used as a point of reference for an additional positive encounter. Thus, consumers who encountered companies with positive images will more likely have consistent positive images and will form a favorable attitude toward online companies. Consequently, consumers who are familiar with companies and brands online will feel less psychological distance. Based on arguments above, two hypotheses are forwarded as follows. 14 H I : Increased company fan: iliarity will reduce feelings of psychological distance toward the online company H2: Increased brand familiarity will reduce feelings of psychological distance toward the online company 2. Antecedents: Similarity between Consumers and an Online Company Similarity has been examined in a large number of empirical studies across the literatures in sales, marketing, and social psychology. The concept of similarity represents consumer judgments about whether they and the companies are alike. Companies that have similar attributes with consumers are more likely to build rapport, persuade buyers, and close sales (Lichtenthal and Tellesfsen 2001). It was reported that consumers in mature business markets give more weight to companies’ attributes than to other aspects of the offering, such as price, product, and delivery (Wilson and Lichtenthal 1985). It was also found that similarity was significantly related to higher degrees of trust and purchase intention (Bush and Wilson 1976; Capon 1975). Higher degree of similarity led to better evaluations of companies or more productive sales interactions (Bush and Wilson 1976; Brock 1965; Capon 1975; Mathews, Wilson, and Monoky 1972; Woodside and Davenport 1974). Companies had Significant degrees of similarity with sold prospects and a significant degree of dissimilarity with unsold prospects. Similarity between consumers and companies can be explained by similarity- attraction paradigm (Byme 1961). This theory explains that consumers’ similarity has strong and long-term effects on the consumers’ perceptions and evaluations of companies. If consumers perceive some degree of similarity with a company, they will feel positively toward that company. Similarity-attraction paradigm describes how perceived similarity may alter consumers’ feelings through two stages (Byme 1961; Byme and Neuman 1992). 15 In the consensual validation stage, a consumer learns that the company has analyzed a given issue and reached the same conclusion as consumers. Thereby, when the company expresses a position that is the same to the consumer’s own position, that company reinforces the consumers’ self-esteem and provides a psychological reward. In the attraction stage, consumers will be attracted to the company to the extent that the company provides some psychological rewards (Byme 1961; Byme and Griffitt 1966; Byme and Neuman 1992). Consequently, a positive evaluation increases consumers’ willingness to accept the company’s assertions and ultimately, to buy product of that company. In this research, it is expected that geographical distance and the city of origin will affect consumers’ similarity perception. In terms of geographical distance, it is very likely that consumers who live in a certain area will have a homogeneous community environment compared with those who live in different cities. Consumers are more likely to feel close with companies located in the same city or near their living cities because those companies are likely to have similar environmental attributes. Thus, geographical distance is expected to affect perceived similarity between consumers and online companies. Simultaneously, consumers will judge the location images of the online company. It is expected that consumers will consider the images of a city from which they order product compared with their current city as a reference point. Then, the perception of the city will affect either positively or negatively on the evaluation of an online company which warehouse is physically located in that city. Thus, consumers who have positive images of current living city will have positive city of origin effects if consumers’ current living city and the city from which product is shipped share similarities. 16 Whereas consumers who have negative images of current living city will have negative city of origin effects if two cities are similar. Eventually, the city of origin affects the evaluation of the online company that is located in that city. In the next part, the similarity effects on psychological distance toward online companies will be discussed. 1) Geographical Distance During 1980s many observers were arguing that technology and globalization were making geography irrelevant. However, research on inequality, on cultural identity, on social and economic change, and on the cultural changes brought about by electronic media and globalization has made that space and place matter (Griswold and Wright 2004). Additionally, sociologists and anthropologists have found that the culture of place, the influence of local social patterns, and interpretation are more robust than ever before (Griswold and Wright 2004). Even though telepresence of intemet enables consumers to access cyber stores and buy products without geographical and spatial constraints, it is still true that tangible products are physically transported to the delivery destinations. The notion that online buying and delivery procedure are not completely separated from the time and spatial limitations might prevent consumers from purchasing products online often. Consumers’ perception of geographical proximity is represented in the form of geographical distance. If offline stores are located in the consumers’ current living city or adjacent cities, consumers perceive that those stores share similar attributes and community environments. For example, a consumer in East Lansing will buy a product in East Lansing or relatively adjacent cities unless they can buy the product in 17 East Lansing. In the same context, even in an online situation, consumers will buy products from websites that have geographically closer physical stores in their current living city than buying from the websites that have physical stores which is geographically far away, as long as other conditions are the same. Additionally, geographical closeness indicates that people who have similar attributes congregate together and form a homogeneous community. Living in a certain city can be associated with status symbol and individualized identity (O’Cass and Frost 2002). Once people share a homogeneous community environment, they are likely to form similar collective identity that is differentiated from members of different societies and creating inter-group attitudes (Sharlin and Moin 2001 ). Within the same community environment, people are supposed to share similarities that guide their attitudes and behaviors. Given geographical closeness has positive relationship with the increase of similar attributes in a homogeneous community, consumers will have less psychological distance toward online companies which are physically close to consumers’ current living city. Based on arguments above, one hypothesis is forwarded aS follows. H3: Geographical distance will be positively related to feelings of psychological distance toward the online company 2) Congruent Images between Current Living City and City of Origin Perceived congruence is one of the most important factors in marketing and in the formulation of marketing strategy (Gregan-Paxton and John 1997). The surrounding environment tends to influence the way people will behave in that environment (Sharlin and Moin 2001). Consumers’ conception of the surrounding 18 sociocultural environment is based on their knowledge of that environment (Campbell 1979). By adjusting themselves into their surrounding environments, their prior knowledge is used as a way to construct reality (Hirchman and Walendorf 1982). Given the importance of living environment, the perception of the city where products are shipped from will affect consumers’ purchasing decision process. In an offline situation, consumers will have either positive or negative attitudes toward the city of origin compared with their current city as reference criteria, such as the public . school system, cultural opportunities, and appearance of the region, park system, and basic local government services, entertainment, restaurants, and housing availability (Glaser 1991). Congruent images between consumers’ current living city and the city from which products are shipped will play an essential role in evaluating online companies. In an online environment, consumers do not even consider directly to visit offline . stores because of the physical limitation, such as time and space. Thus, consumers in an online situation will either rely on external information, such as mass media or on analogy from a familiar current living city to a less familiar city where products are shipped from because the information that they obtained will affect the evaluation of online companies which warehouse is located in that city. The cognitive process how the knowledge of current living city is transferred to the evaluation of city of origin can be explained with knowledge transfer theory through four stages (Gregan-Paxton and John 1997). First, the access stage of the analogical learning process is concerned with the cognitive ability of consumers to access a base domain. The goal of this phase is to activate consumers’ mental representation of a current living city so that it can be considered as a potential source of information about the images of city from which they order products. 19 Second, in the mapping stage, knowledge of current living city and city of origin is aligned such that knowledge associated with the current living city can transfer to the city where products are shipped from. This stage can be accomplished through a process that constructs one-to-one correspondences between the elements of the current living city and city of origin. These correspondences indicate paths between the familiar one to less familiar one across which knowledge can be transported. Third, in the knowledge transfer stage, actual knowledge is transferred from current living city to the city from which products are shipped. Underlying the knowledge transfer decision is the belief that the domain known to be Similar in certain attributes are likely to be similar in other attributes as well. For example, the knowledge of current living city in terms of major industry, city size, housing, environmental issues, crime rate; educational support, and ethnic and cultural diversity will be transported to the evaluations of lesser-known city. Finally, a more abstract knowledge structure, such as schema, will be created as a byproduct of the internal learning process. This accumulated knowledge structure will be used as a base in the future learning process. It is expected that increased congruent image between current living city and city of origin will affect the psychological distance toward online companies. Differentially familiar consumers have differentially developed schema, and thus would use different information in objective evaluations (Park and Lessig 1981). Thus, if consumers created positive images toward the city from which products are shipped and the city in which they currently reside, they are likely to have less psychological distance. Whereas, if consumers have negative images of the city from which they order products compared with their current city, they are likely to have high psychological distance. Based on arguments above, one hypothesis is forwarded. 20 H4: Psychological distance is minimized when the image of a city where a product is shipped from matches the image of city in which a consumer resides 3. Consequence: Thustworthiness of an Online Company The concept of trust has long been discussed by marketers within the discipline of sociology, psychology, and economics (Delgado-Ballester and Munuera-Aleman 2001). The research of the trust concept originally comes from the analysis of personal relationships in the field of social psychology because it is considered an inherent characteristic of any valuable social interaction. Recently, this concept has become a popular issue in marketing literature, due to the relational orientation emerging in marketing field (Delgado-Ballester and Munuera-Aleman 2001). Trustworthiness became one of important factors that enable the relationship between consumers and companies not only in traditional forms of business transactions, but also in online business environment. Trustworthiness is suggested to be the most critical construct that focus for online medium currently. Trust has been conceptualized as a crucial antecedent and outcome of buyer-seller relationship development (McCole and Pahner 2002). Yoon (2002) defined trust as the ability to accept or approve of something without investigation or evidence. Trust also can be defined as a governance mechanism in exchange relationships that are characterized by uncertainty, vulnerability, and dependence. Trustworthiness could be defined as the reliability and dependability of the vendor offering products or services (Chen and Dhillon 2003). Recently, there has been much discussion about the role of trustworthiness in an online environments where the absence of face-to-face contact and other tangible cues. make consumers hesitate to participate ecommerce activity (McCole and Palmer 2002). Relationship between consumers and online companies that have had no prior 21 association is expected to emerge online and to start small actions that initially require little reliance on trust. Developmentally, if actions are reciprocated by partners, trustworthiness will be increased (Jarvenpaa, Tractinsky, and Vitale 2000). Thus, the trust becomes more important in an online environment where evaluation of trustworthiness solely depends on the interactive communication (Schlosser, Shavitt, and Kanfer 1999). Even though trustworthiness assumes variety of forms, there is a commonality in its core in that trustworthiness is the trustor’s expectation of good will on the part of a particular, and specifiable trusted party (Koehn 2003). Trust can be broken if the vendor engages in illegitimate businesses or when there is a general disregard for privacy and security. Since trustworthiness, has been shown to exert a significant and direct influence on purchase intention, increasing concerns related to trust over a period of time can negatively influence purchasing behavior. In relation to psychological distance toward online companies, when consumers feel the lack of connection or weaker bonds between themselves and online companies, they will hesitate to trust those companies in the end. Based on argument, it is expected that the psychological distance toward online companies will negatively affects trustworthiness. H5: Psychological distance will be negatively related to the trustworthiness of an online company 22 Based on hypotheses above, hypothetical model is suggested as follows; see Figure l. < Figure l > Hypothetical Model Familiarity 1)Cornpan Familiarity 2) Brand amiliarity Similarity 3) Geographical Distance 4) Congruent. City Image 23 IV. Methodology 1. Pretest A pretest was conducted to find out the perception of specific cities that were selected to be compared with East Lansing in terms of geographical distance and city image. First, 15 cities were randomly selected across the United States. Half of them were quite similar to the images of East Lansing, whereas, the rest of them were opposite to the image of East Lansing. Thirty undergraduate students in a research class were asked to rate how much those cities are similar to East Lansing with a 7 point Likert scale (1: Very dissimilar ~ 7: Very similar). Then, the most similar and dissimilar cities were selected with the combination of geographical distance from East Lansing. AS a result, it was found that Detroit was closest and opposite image city (—X_ = 2.55) and New York was farthest and opposite image city (Y = 1.45). Meantime, Stamford, CT was chosen as farthest and similar image city (:T = 3.89). 2. Data Collection Given the purpose of identifying structural relationship between each variable, this research used the experimental method to verify hypotheses in a quantitative manner. The data for this research were collected via an online survey from the lSt of May till the 30th of May, 2005 that consists of nine websites, beginning with an informed consent form, advertising stimuli, and main questionnaire (http://www.msu.edu/~steve/jin/consent.htm; see Figure 4). Subjects were asked to 24 view the assigned advertisements and then fill out a self-administered questionnaire online. Online questionnaires were published by the WebSurveyor Desktop 4.1 program. Participants could join the survey after they saw stimuli advertising online. When subjects went to the survey website and clicked the launch button, they were randomly shown one stimulus advertisement. Each page of the survey consisted of approximately 15 to 20 questions, to help prevent fatigue as respondents proceeded through the survey. Data were stored in the advertising department server safely. There are three main reasons why the online survey method was chosen for this research. First, given the research studying online product purchases, consumers have to experience a real an online environment. Thus, it was necessary to make a real online purchasing environment. Online stimuli advertisements were created and participants directly responded to questionnaires after carefully observing stimuli advertisements. Second, online survey was selected because of its convenience. Compared with a paper and pencil questionnaire, online survey is not constrained by physical limitations, such as time and space. Respondents can freely join the survey as long as they are connected to intemet. Third, researchers not only can collect data until the last minute, but also they do not have to code each number manually because the data are saved digitally. Due to these various advantages, this research used online data collection method. 3. Sample Considering young and educated consumers are more likely to prefer online buying, the sample population for this study is university students including masters 25 and Ph.D. students of Michigan State University (MSU). Most importantly, students who live around Lansing and East Lansing area were selected because one of our research purposes is to identify geographical distance and city of origin effects. Thus, it was important to select those who reside around the Lansing and East Lansing area. In this research, simple random sampling and convenience sampling was used among those who registered undergraduate classes during spring and fall semester of 2005 in the advertising and PR department. Also, participants including master and Ph.D. students were randomly selected in MSU directory and were solicited to join via email. For inducing participation, a SO-dollar drawing was given to those who completed the survey. Additionally, advertising and PR major students who registered for summer sessions were given credit incentives in return for their voluntary participations. 4. Experimental Design and Manipulations MSU students (n=412) were randomly assigned to view one of 16 experimental online advertisements. The study used a 2 (low company familiarity vs. high company familiarity) x 2 (low brand familiarity vs. high brand familiarity) X 2 (short distance city vs. long distance city) X 2 (incongruent city image vs. congruent city image) full- factorial, between subjects design. Sixteen different online advertisements were devised to manipulate the different levels of company familiarity, brand familiarity, geographical distance, and congruent city images. The format of the stimuli advertisements was standardized except for the manipulated variables of interest. Each ad contains the company name at the top of the page with a large picture of the same digital camera. The reason the digital camera was selected is that technology-based products are not usually affected by specific 26 cultural factors and digital cameras are purchased frequently among university students. The advertisements also contain brand name and product shipping location. Product shipping location represents long and short geographical distance from East Lansing and similar and dissimilar city images compared with East Lansing. To manipulate different levels of company familiarity, an existing online company (Amazon. com) was used for the high company familiarity condition because Amazon.com is a nationally known online company. Whereas, a fictitious company name (CanogaCamera.com) was created for the low company familiarity condition because CanogaCamera.com is unknown to consumers compared to Amazon.com. To manipulate different levels of brand familiarity, existing brand name (SONY) was used for the high brand familiarity condition because subjects have extensive prior knowledge. Also, made-up brand name (PASOCON) was used for the low familiarity condition because made-up brand is not associated with prior knowledge of consumers in anyways and excludes consumers’ predisposition of a certain existing brand. In the meantime, similarity was manipulated according to the pretest results. East Lansing was selected as the closest and the most similar image city, while Detroit was the close, but dissimilar image city. Stamford, CT was the farthest and similar image city, while New York is the farthest and dissimilar image city. Thus, these four cities were selected for the manipulation of geographical distance and congruent city image. The main questionnaire contained items measuring company familiarity, brand familiarity, geographical distance, and congruent city images between the current living city and city of origin. Also, psychological distance toward online companies and trustworthiness were included. For additional analysis, a few auxiliary questions were asked, such as perception of East Lansing and demographic information. 27 5. Dependent Variables In terms of dependent variables, psychological distance can be conceptually defined as the lack of connection or weaker psychological bonds between consumers and an online company. Psychological distance was measured with four items in positive directions, ‘This company would be understanding about delivery problems’, ‘This company will be open in dealing with me’, ‘This company cares about my problems’, and ‘This company seems knowledgeable about its products.’ These items were created after modifying the relational distance index (Hess 2003). Another dependent variable is trustworthiness of a company. Trustworthiness is conceptually defined as the ability to rely on or have conviction about the quality or attributes of a company. The trustworthiness concept was measured with six items, ‘I feel confidant in purchasing a product from this company’, ‘I would be comfortable buying products from this company’, ‘This company seems trustworthy’, ‘Compared with other companies, I trust this company more’, ‘This company has reliable service’, and ‘This company is easy to contact if I have a problem.’ Additional variables were measured in order to check manipulations and assess the degree to which psychological distance mediates the relationships between familiarity and trustworthiness as well as the relationships. Company familiarity and brand familiarity can be conceptually defined as how well you know a company and a brand (Strauss, Barrick, and Connerley 2001). Company familiarity was measured with four items, ‘I am familiar with this company’, ‘I have heard of this company’, ‘I have previously purchased from this company’, and ‘I don’t know this company very well.’ Brand familiarity was measured with three items, ‘I have heard of this camera’, ‘I know a lot, about this camera’, and ‘I know this brand of camera very well.’ 28 Also, similarity can be conceptually defined as a consumer’s judgment about whether consumers and online companies are alike in some way or sharing similar attributes (Lichtenthal and Tellesfsen 2001). Similarity was measured in terms of actual geographical distance and congruent city images between current living city and city of origin. Geographical distance was measured with four items, ‘The camera I ordered will be shipped from a location near East Lansing’, ‘The company is located near East Lansing’, ‘The camera will be shipped from a city which is similar to East Lansing’, and ‘The' company is located in a city which reminds me. of East Lasing.’ Congruent city image was asked with four items, ‘People working in this company see the world like me’, ‘I have Similar beliefs to the people working in this company’, ‘I am similar to the people working in this company’, and ‘People working in this company have similar values to my own.’ All dependent and additional variables were measured with seven point Likert scale (1: Strongly Disagree ~ 7: Strongly Agree). 6. Covariate Items regarding perception of East Lansing were included in the questionnaire becausejit would affect the geographical distance and other city images. Perception of East Lansing was measured with five items, ‘People in East Lansing are kind’, ‘East Lansing is attractive’, ‘East Lansing is culturally open’, ‘People in East Lansing are ethical’, and ‘People in East Lansing are sophisticated (or = .80).’ The perception of East Lansing as covariate significantly affects the psychological distance, F (1,393) = 20.83, p < .01. Given the fact that the attitude toward East Lansing will be different individually, it was necessary to include this variable into our research. Thus, perception of East Lansing was used as a covariate throughout all subsequent analysis. 29 7. Analysis Methods Collected data was analyzed with the SPSS 10.0 statistical package. Factor analysis and reliability tests were used for the measurement development. Correlation analysis, ANCOVA, and simple regression tests were used as major analysis methods. First of all, factor analysis and reliability test were conducted to extract relevant and reliable items so that the structure of each variable could be identified. Then, the effects of independent variables were examined with AN COVA and correlation analysis. Additionally, the mediating role of psychological distance was testified by using regression analysis. 30 V. Results 1. Data Description Four hundred and twenty two questionnaires were collected and 10 questionnaires were screened out because they showed pattemized responses or insincere responses. A total of 412 questionnaires were analyzed. From the sample population, it was found that the ratio of gender was almost the same because the number of male subjects was 190 (46.1%) and the number of female subjects was 217 (52.7%). Ages from 20 to 24 totaled 211 (51.2%) andthe majority of subjects were in their early twenties. In relation to gender, the average purchasing frequency of males was 4.11 times, whereas the average purchasing frequency of females was 3.47 times. Statistically, there was no significant difference on purchasing frequencies, t (401) = - 1.47,p=.l4. 2. Manipulation Checks Independent T-tests were used to assess the manipulation of each condition. Participants in the Amazon.com condition (E = 4.65) reported having significantly more familiarity with the company than those in the CanogaCamera.com condition (E = 2.79), t (408) = -9.78, p < .01; see Table 1. In terms of brand familiarity, participants in the SONY condition (f = 3.59) reported having significantly more familiarity than those in the PASOCON condition (.27 = 1.84), t (385.35) = -12.39, p < .01. Participants also correctly perceived geographical distance difference. Participants in the East Lansing and Detroit condition (2? = 3.86) reported that they 31 felt significantly closer than those in the New York and Stamford condition (I)? = 2.77), t (408) = 8.20, p < .01. However, the congruent city image manipulation was not successful because the East Lansing and Stamford condition (:1; = 3.55) did not Show significant differences from the Detroit and New York condition (E = 3.46), t (407) = - .77, p = .44. Consistent with our manipulations, participants correctly perceived all conditions except congruent city image. < Table 1 > Manipulation Checks Variable Group X t df P Company Amazon.com 4.65 . . . - .7 4 8 . ** Familiarity CanogaCamera.com 2,79 9 8 O 00 Brand SONY 3.59 12 39 385 32 00" Familiarity PASOCON 1.84 ' ' ' ' Geographical EL, Detroit 3.86 - ' *II! Distance NY, Stamford 2.77 8.20 408 .00 Congruent EL, Stamford 3,55 City Image Detroit, NY 3.46 "77 407 '44 ”1 p < .01, *2 p < .05 3. Measurement Development . 1) Additional Variables Fifteen items were put into factor analysis to examine company familiarity, brand familiarity, geographical distance, and congruent city image; see Table 2. The maximum likelihood method and direct oblirnin rotation was used to find out relatedness and direction between factors. Four factors were identified and named as ‘company familiarity’, ‘ congruent city image’, ‘brand familiarity’, and ‘geographical distance.’ Four items of company familiarity (or = .90) and congruent city image (or = .91) tumed out to be reliable respectively. Also, three items of brand familiarity were reliable (or = .82). Finally, 32 four geographical distance items were reliable, too (or = .79). Company familiarity has a significant correlation with brand familiarity (r = .41, p < .01) and congruent city image (r = .22, p < .01); see Table 4. Also, brand familiarity has a significant correlation with geographical distance (r = .14, p < .01) and congruent city image (r = .17, p < .01). < Table 2 > Factor Analysis for Company Familiarity, Congruent City Image, Brand Familiarity, and Geographical Distance Factors Company Coragiruent Brand Geographical Familiarity ty Familiarity Distance Image Familiarity .97 .43 Heardcompany .86 .39 Prepurchase .82 .39 Knowcompany .66 . .33 Samewordview .92 Samebelief .89 .35 Similarpeople .83 Similarvalue .81 .38 Heardbrand ' .37 .88 Knowl .34 .82 Know2 .41 .69 ShipfromEL .76 SimilarEL .70 RemindEL .34 .68 NearEL .63 Eigenvalues 3.95 2.86 1.59 1.27 % of Variance 26.36 19.05 10.59 8.49 2) Dependent Variables A Similar process was followed to evaluate 12 items measuring trustworthiness and psychological distance. The maximum likelihood method and direct oblimin rotation was used to find out relatedness and direction between factors. Additionally, two items regarding credibility and reliability were deleted because they were loaded across factors. 33 < Table 3 > shows the factor loadings. Two factors were identified and named as ‘trustworthiness’ and ‘psychological distance.’ Six items of trustworthiness were reliable (or = .91) and four items of psychological distance were also reliable (or = .82) Trustworthiness has a strong negative correlation with psychological distance (r = -.71, p < .01); see Table 4. < Table 3 > Factor Analysis for Trustworthiness and Psychological Distance Factors Trustworthiness Psych ologr cal Distance Confidence .88 .65 Comfort] .87 .56 Trustworthy .84 .67 Trustrnore .83 .43 Relaible2 .79 .54 Easycontact .75 .44 Open .61 - .81 Careproblem .47 .83 Understand .68 .79 Knowledgeable .47 .78 Eigenvalues 5.86 .98 % of Variance 58.56 9.79 < Table 4 > Correlation Tests Company Brand Geo- Congruent Psy- Trust- Familiarfly Familiarity Distance City Image Distance worthiness C°mPim¥ 1 000 41" - 06 22M - 52** 72** Familiarity ' ' ' ' ' ' Brand an: an: an: an): Familiarity 1.000 .14 .17 —.31 .35 g?“ 1.00 ' .29" -.11* .09 istance Congruent _ H a... City Image 1.00 .37 .45 Psy- , u Distance 1 '00 '71 Trust- worthiness ! '00 “‘2 p < .01, *z p < .05 34 4. Hypothesis Testing Psychological distance and trustworthiness factors were used as dependent variables. Items of each factOr were averaged out to create indices and these indices were used to test five hypotheses. The ANCOVA test examined the effects of different conditions. Additionally, the mediating role of psychological distance was identified using Baron and Kenny’s mediating test methods (1986). l) Hypothesis 1 Hypothesis 1 predicts that increased company familiarity will-reduce feelings of psychological distance toward the online company. As expected, there is a significant simple main effect of company familiarity. Subjects felt greater psychological distance when viewing CanogaCamera.com (X = 4.02) than Amazon.com (2? = 3.76), F (1.393) = 4.78, p < .05. Given the significant main effect of company familiarity, hypothesis 1 is accepted. Familiarity with an online company reduces feelings of psychological distance. 2) Hypothesis 2 Hypothesis 2 predicts that increased brand familiarity will reduce the feelings of psychological distance toward the online company. As expected, there is a significant simple main effect of brand familiarity. Subjects felt greater psychological distance when viewing PASOCON (X? = 4.18) than SONY (3?: 3.61), F(1.393)= 24.02, p < .01. It is also identified that there is a significant interaction effect with geographical distance, F (1.393) = 3.76, p = .05; see Figure 2. Interaction results are driven by the simple main effect of brand familiarity within the short distance condition, where the low brand familiarity (27 = 4.30) leads to higher 35 psychological distance than the high brand familiarity (Y = 3.50), t (226) = 5.02, p < .01. However, there is no significant interaction effect of brand familiarity within the long distance condition, t (180) = .1.46, p = .15. Also, there are no significant main effects of geographical distance within low brand familiarity, t(198)= 1.67, p = .10, and high brand familiarity, t (208) = -1.64, p = .10. Given the significant main effect of brand familiarity and the interaction effect with geographical distance, hypothesis 2 is accepted. Familiarity with a brand reduces feelings of psychological distance. < Figure 2 > Interaction Effect with Geographical Distance Psychological Disxmce 4.30 ow Brand Familiarity 4.07 350) gh Brand Familiarity j 1 ._:__, FarGeographical Distance Near 3) Hypothesis 3 Hypothesis 3 proposes that geographical distance will be positively related to feelings of psychological distance toward the online company. It is found that geographical distance does not affect the psychological distance, F (l. 393) = .00, p = .99. However, there is an interaction effect with the brand familiarity, which is significant, F (1,393) = 3.76, p = .05; see Figure 3. The significant main effect of the brand familiarity is identified within the close distance, where the low brand familiarity condition ( X = 4.30) leads to higher 36 psychological distance than the high brand familiarity condition (X = 3.50), t (226) = 5.02, p < .01. However, there is no main effect of brand familiarity within long distance, t (180) = 1.46, p. = .15. In addition, it is found that there are no significant interaction effect within the low brand familiarity condition, t (198) = 1.66, p = .10, and the high brand familiarity condition t (208) = -1.64, p = .10. Opposite of our expectations, when consumers are unsure of the brand, they might prefer to buy products from New York or Stamford because they might think that those cities are more likely to import foreign brands that they never heard of before. Even though the geographical distance affects psychological distance depending on the different levels of the brand familiarity, there is no significant main effect of geographical distance. Thus, hypothesis 3 is rejected. < Figure 3 > Interaction Effects with Brand Familiarity Psychological Disiimce PASOC 0N SONY grand Familiarity 4) Hypothesis 4 Hypothesis 4 proposes that psychological distance is minimized when the image of a city where a product is shipped from matches the image of city in which a consumer resides. It is found that there is no significant main effect of congruent city 37 image, F (1, 393) = .54, p = .46. Thus, hypothesis 4 is rejected. 5) Hypothesis 5 Hypothesis 5 proposes that psychological distance will be negatively related to the trustworthiness of an online company. According to correlation tests, it is found that psychological distance Shows a significant negative correlation with trustworthiness (r = -.71, p < .01). This result supports the idea that increased psychological distance negatively related to trustworthiness of an online company. Thus, hypothesis 5 is accepted. 5. Additional Analysis: Mediating Role of Psychological Distance One of our research purposes is to identify the mediating effects of psychological distance between the familiarity and the trustworthiness of an online company. To demonstrate mediation, four relationships must hold (Baron and Kenny 1986).1 First, familiarity must have a significant positive effect on trustworthiness. To examine this relation, company familiarity and brand familiarity are averaged out to create a new index: overall familiarity. It is found that familiarity has a significant effect on trustworthiness, F (1,408) = 316.77, p < .01, B = .66. Next, familiarity must have a significant effect on psychological distance. We find that there is a significant negative effect on psychological distance, F (1,408) = 140.72, p < .01, B = -.51; see Table 5. Third, it must be established that psychological distance significantly affects trustworthiness. The estimated regression model Shows that there is a significant negative effect on trustworthiness, F (1. 408) = 403.75, p < .01 , B = -.71. Finally, when 1 Similarity effects are not being examined because it was not significantly related to psychological distance. 38 the psychological distance is included in the trustworthiness model, the formerly significant effect of familiarity becomes insignificant or reduced. It is shown that the effect of familiarity is significant, but reduced, F (2.407) = 334.35, p < .01, B = .41, while the effect of psychological distance on trustworthiness remains significant, F (2. 407) = 334.35, p < .01, B = -.50. This result means that overall familiarity not only has a direct effect on trustworthiness, but also has an indirect effect via psychological distance. Thus, it is concluded that psychological distance partially mediates the effect of familiarity on trustworthiness. < Table 5 > Mediating Effects of Psychological Distance Independent Var. Dependent Var. [3 F P Psychological Distance Trustworthiness -.71 403.75 .00** Overall Familiarity Psychological Distance -.51 140.72 .00" Overall Familiarity Trustworthiness .66 316.77 .00** 0’6”" Fam‘l’anty Trustworthiness .41 334.35 .00M Psychological Distance **: p< .01, *: p< .05 39 VI. Discussion 1. Research Implications Our results show the effects of familiarity and similarity on psychological distance and the trustworthiness of an online company. Given that the psychological distance concept has not been explored in the online advertising field, our research contributes to this field by identifying underlying relationships between preceding variables and psychological distance. Also, the series of mediation tests supports the assertion that familiarity effects evoke psychological distance, which negatively related to trustworthiness. This research provides replication of previous findings of familiarity effects. One benefit of creating a strong company familiarity and brand familiarity is to increase communication effectiveness (Campbell and Keller 2003). Because familiarity is important component of brand equity, these research findings provide some empirical supports for the beneficial effect of company familiarity and brand familiarity on communications in an online situation. These study findings can be interpreted as showing that high familiar online company and brands are able to make consumers perceive less psychological distance. In relation to psychological distance, it is found that both company familiarity and brand familiarity are important preceding variables of psychological distance. As consumers become more familiar with a company and brand, they feel less psychological distance toward the online company. If consumers perceive that the company that they interact is more open, cares about consumers’ concerns, and tries to understand them, they feel less psychological distance. Therefore, it is expected that individual consumers will show less distancing behaviors toward the online 40 company. Whereas, consumers who have high psychological distance toward an online company are more likely to avoid, disengage, and cognitively dissociate themselves from that company in order to maintain cognitive consistency. Thus, this research Shows the possibility that consumers not only have psychological distance toward other people (Hess 2002; 2003), but also they have it toward an online company that they interact with. Therefore, the research results prove the applicability of psychological distance from the solid theoretical background of interpersonal psychology to advertising online. In the meantime, it is identified that perceived psychological distance is negatively related to trustworthiness. As expected, the research results clearly Show that psychological distance is one of important preceding factors that affect trustworthiness. Consumers would not trust an online company if they think that the company is not communicative, not care consumers’ concerns, and indifferent to them. Thus, there is a causal relationship between psychological distance and trustworthiness. However, it is found that geographical distance does not affect psychological distance. Previous research findings in international trade converged that geographical distance is one of the trade resistance factors (Dow 2000). Carlson (1974) insisted that geographical distance affects the collection and transmission cost of important management information. Many of the characteristics of information are tacit in nature and need face-to-face communications so that the geographical distance negatively affects international trade. However, individual consumption behavior in an online situation might be interpreted differently. First of all, geographical distance itself is not a high priority 41 for consumer decision because of individually-customized mass transportation and. reduced delivery time. Consumers think that the delivery time is a more important cue than geographical distance. Regardless of geographical difference, one to two delivery days might be faster than they have expected. Thus, relatively fast delivery days might negate the effects of geographical distance. Also, it is found that congruent city image does not affect psychological distance. This means that participants do not perceive fhat the images of East Lansing and Stamford are different from the images of Detroit and New York Also participants do not think that the images of a city affect the evaluation of an online company. Previous similarity studies are mostly done in the offline situation between consumers and sellers, where relational dynamics affect the perception of both consumers and marketers. Specifically, previous research asserted that similarity effect is significantly related to higher degrees of trust and purchase intention (Bush and Wilson 1976; Capon 1975). Also, the higher degree of similarity led to better evaluations of companies or more productive sales interactions (Bush and Wilson 1976; Brock 1965; Capon 1975; Mathews, Wilson, and Monoky 1972; Woodside and Davenport 1974). However, one thing that previous similarity study has overlooked is that online purchasing behavior hardly requires instant and dynamic human interactions compared with conventional transaction in the offline environment in which two parties meet together. Online buyers do not need to rely on peripheral information, such as city images, external, and internal characteristics of the people who work in an online company. This research results shows that in an online environment, consumers do not consider images of a city from which products are Shipped nor they believe that city images affect the perception of an online company that is located in 42 that city. Thus, the environmental differences between an offline situation and online situation might have led to inconsistent results in our research. Given the previous arguments, these results provide implications for online marketers and practitioners. They must keep in mind that consumers evaluate not only the value of the company name, but also consider the brand name in an online Situation. Thus, marketers need to enhance their company image as well as brand values. Marketers of unfamiliar company and brands need to build familiarity to compete better with more familiar company and brands. Even though geographical distance does not affect psychological distance by itself, its effect will be different depending on the delivery time. Consumers would not willing to buy' from a city which is far away if the online company do not offer fast delivery. Thus, marketers need to provide services that can compensate geographical distance. 2. Limitations and Future Research There are several limitations regarding discriminant validity of psychological distance, and construct validity of congruent city image, uncontrolled environmental factors, online data collection, and generalizability issues. First of all, our research is interested in identifying the role of psychological distance in an online situation. In terms of discriminant validity, the psychological distance concept is not clearly distinguished from trustworthiness in our research. Psychological distance could be the same concept or subconcept of trustworthiness because these two concepts partially overlap. In the future research, it will be important to conceptually separate psychological distance from trustworthiness. Second, in terms of the construct validity of congruent city image, city images 43 are manipulated by choosing four different cities. However, this manipulation raises questions whether the city images are manipulated as intended. It might be possible for the subjects to regard city image differences as geographical differences. Also, it is expected that some subjects do not have previous knowledge about the cities they saw in stimulus advertisements so that they might be unable to compare city image differences. Therefore, it will be necessary to devise more elaborate manipulation of congruent city image in the future research. By doing this, the effects of geographical distance and congruent city image will be distinguished from each other. Third, this study was unable to control environmental changes during data collection periods. The data collection was conducted between May lst and May 30th. During that time, summer vacation began so that the subjects were no longer staying in East Lansing. Their reference point of geographical distance was not standardized due to their heterogeneous locations. This uncontrolled environmental change might affect location sensitive items and lead to insignificant results. Fourth, this research used an online data collection program called WebSurveyor Desktop 4.1. Using this program could be justified by two different reasons. Given that our focus is on online consumer attitude and behavior, it seems quite practical to capture consumer response right after the subjects see stimuli advertisement. Another merit is that there are no physical restrictions for the subjects to participate in the survey as long as they are connected to intemet. These advantages provide a flexible research schedule and convenient data collection. Despite the major advantages mentioned above, online data collection has potentially dangerous aspects. Most importantly, it is difficult to control survey participants. In other words, it is hard to know who actually participated. Also, there might be duplications because no procedures to check individual identity are invented. 44 Thus, it might have been possible for a participant to join the survey several times intentionally or by mistake. Thus, these issues need to be addressed in the future research. Finally, there is a generalizability issue because research participants are limited to MSU students. In addition, this research used convenient and simple random sampling among MSU students so that it is difficult to generalize across ages, regions, and other demographic characteristics. 45 < Table 6 > AN COVA for Hypotheses Testing Independent Variable df 3:32:16 F P Condition 2 . . . Amazon.com 3.76 *4! Company Familiarity 1 6.41 4.78 .03 CanogaCamera.com 4.02 . . . SONY 3.61 *4! Brand Famlllarlty l 32.16 24.02 .00 P AS 0 C ON 4.] 8 . . 4.31E - Short 3.89 Geographical Dlstance 1 O4 .00 .99 Long 3 . 89 . Similar 3.85 Congruent Clty Image 1 .73 .54 .46 Dissimilar 3.94 . Short 3.50 Brand Familiarity X ,, SONY Log 3.72 . . l 5.04 3.76 .05 Geographlcal Dlstance Short 4.30 PASOCON Long 4.07 Perception of EL 1 27.89 20.83 .00" Company Familiarity X Brand F amiliarigl 1 '89 '67 '42 Company Familiarity X Geographical Distance 1 '95 '71 '40 Company Familiarity X Congruent City Image 1 '23 '17 '68 Brand Familiarity X Congruent City Image 1 1'29 '96 '33 Geographical Distance X Congruent City Image 1 '23 '17 '68 Company Familiarity X Brand Familiarity X 1 .19 .14 .71 Geographical Distance Company Familiarity X Brand Familiarity X l .41 .31 .58 Congruent City Image Company Familiarity X 1 514B Geographical Distance X 1 '_03 .00 .97 Congruent City Image Brand Familiarity X Geographical Distance X 1 1.60 1.19 .28 Congruent City Image Company Familiarity X Brand Familiarity X Geographical Distance X l 3'82 2'85 '09 Congruent City Image Error 393 1.34 Dependent Variable: Psychological Distance R Squared = .14 (Adjusted R Squared = .10) "z p < .01, *z p < .05 < Figure 4 > Stimuli Advertisements Sample amazon‘com' limo“ 1 mum] DEE-£53 | to: 14:9. 5112116 Rm . —»_ ,.—\,— form ma fljfipnkmsgfifimt ifaf 03g," - .. .. lit-'1). Illlllllhrllfi ll I'er‘u, Aptinhl. (4:1 WWW -. W”? . . _. ~a¢m7a