1 AN EXAMINATION OF MARKET INTELLIGENCE GAPS IN THE ADVERTISING INDUSTRY AND THEIR EFFECTS ON AGE N CY CLIENT RELATIONSHIPS: MEDIA PLANNERS PERSPECTIVE By JUN HEO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSI TY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Jun Heo
3 To my wife, Hyunmi
4 ACKNOWLEDGMENTS I have been very fortunate by having many people who supported and guided me during my years of doctoral work. I would like to say a special thanks to several people who have willingly offered their time, help, and invaluable advice. First of all, I am deeply indebted to Professor John C. Suther land for opportunities he provided for last four years and for the successful completion of my doctoral work. I could not have completed this work without his support. H e has been with me every moment that I need support, encouragement, and advice. I again sincerely appreciate everything he has done for me I also owe many thanks to Professor Jon D. Morris, Cynthia R. Morton, and Yong Jae Ko, who have been encouraging and supportive during my dissertation work I extend my gratitude t o Professor Chang Hoan Cho who guided me and provided me with the foundation for becoming a scholar when I changed my path from the industry to the academy. I would also like to offer my regards and blessings to Rev. Hee Y. Sohn and Minseok Sohn for their endless support, friendship, and prayer. My parents Sook and Youngja, receive my deepest gratitude and love for their dedication and the many years of support. Most im portantly, I deeply thank my wife, Hyunmi, for being my wonderful friend and supporter. Without her belief in m e, I would have fallen by the wayside. My special thanks also go to my kids, Yerin and Daniel, for being who they are! Last but not least, I must acknowledge the most heartfelt thanks to God, whose love always rise me up, and who gave me this promise: P e ace I leave with you; My peace I give to you; not as the world gives do I give to you. Do not let your heart be troubled, nor let it be fearful (John 14:27)
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 9 ABSTRACT ................................................................................................................... 10 CHAPT ER 1 INTRODUCTION .................................................................................................... 12 Significance of the Relationship between Advertising Agencies and Client Firms .. 12 Significance of Communication in Agency Client Relationships ............................. 14 Importance of Media Market Intelligence ................................................................ 15 Purpose of th e Study .............................................................................................. 17 2 LITERATURE REVIEW .......................................................................................... 21 Relationship Marketing in Professional Service Business ....................................... 21 Agency Client Relationships in the Advertising Industry ......................................... 23 Communications in the Context of Agency Client Relationships ............................ 27 Market Intelligence Gap .......................................................................................... 31 Media Market Intelligence ....................................................................................... 33 Structures of Advertising Agencies ......................................................................... 35 Relationship Quality Constructs .............................................................................. 38 Trust ................................................................................................................. 41 Commitment ..................................................................................................... 43 Satisfaction ....................................................................................................... 47 Cooperation ...................................................................................................... 48 Structural Aspects of Relationship Quality ........................................................ 50 Summary ................................................................................................................ 51 Summary of Research Questions and Hypotheses ................................................ 52 3 METHOD ................................................................................................................ 57 Participants and Procedures ................................................................................... 57 Instrumentation ....................................................................................................... 60 Independent Variables ..................................................................................... 60 Dependent Variables ........................................................................................ 61 Control Variables .............................................................................................. 63 Demog raphics .................................................................................................. 63 Expert Review ......................................................................................................... 63
6 Gap Analysis ........................................................................................................... 65 Statistical Tests ....................................................................................................... 66 4 RESULTS ............................................................................................................... 71 Description of Samples ........................................................................................... 71 Descriptive Statistics ............................................................................................... 72 Communication Variables ................................................................................. 72 Relationship Quality Variables .......................................................................... 73 C ontrol Variables .............................................................................................. 73 Research Question 1: Importance of Market Intelligence Items .............................. 74 Research Question 2: Delivery and Gap Scores of Market Intelligence Items ........ 76 Research Questions 3 & 4: Relationship between Gap Scores and Overall Relationship Quality ............................................................................................. 79 H ypotheses Tests ................................................................................................... 81 Hypothesis 1 ..................................................................................................... 81 Hypothesis 2 ..................................................................................................... 82 Hypothesis 3 ..................................................................................................... 82 Hypothesis 4 ..................................................................................................... 83 Hypothesis 5 ..................................................................................................... 84 Research Question 5: The Structure of Relationship Quality Constructs ................ 84 Measurement Model ......................................................................................... 85 Structural Model for the Modified Independent Factor Structural Model ........... 87 5 DISCUSSION ....................................................................................................... 104 Market Intelligence Gaps ...................................................................................... 104 Differen ces between Types of Advertising Agencies ............................................ 107 Relationships between Market Intelligence Gaps and Relationship Quality .......... 109 Implic ations of the Research ................................................................................. 111 Limitations and Future Directions ......................................................................... 113 APPENDIX A SURVEY INVITATION: Postcard .......................................................................... 117 B SURVEY INVITATION: SNOWBALLING .............................................................. 118 C SURVEY QUESTIONNAIRE ................................................................................ 119 D FOLLOW UP EMAIL FO R THE POSTCARD RECIPIENTS ................................. 127 LIST OF REFERENCES ............................................................................................. 128 BIOGRAPHICAL SKETCH .......................................................................................... 140
7 LIST OF TABLES Ta ble page 3 1 Market intelligence items to be included in the study .......................................... 69 3 2 Measurement items for communication variables ............................................... 69 3 3 Results of exploratory factor analysis (EFA) for measurement scales of relationship quality .............................................................................................. 70 3 4 Measurement items for existing conflict .............................................................. 70 4 1 Respondents profiles ......................................................................................... 90 4 2 Descriptive statistics for communication variables .............................................. 91 4 3 Descriptive statistics for relationship quality variables ........................................ 91 4 4 Descriptive statistics for control variable: existing conflict .................................. 92 4 5 Summary statistics of the perceived importance of market intelligence item ...... 92 4 6 Principle component factor analysis of the importance scores (N=117) ............. 93 4 7 Summary statistics of the perceived delivery of market intelligence item ........... 93 4 8 Importance/Delivery Market Intelligence Gaps (0 = no gap) ............................... 94 4 9 The results of principal component factor analysis of gap scores (N=96) ........... 94 4 10 Model sum mary of stepwise multiple regression model ..................................... 95 4 11 Effects of gap scores in the regression model on the overall relationship quality ................................................................................................................. 95 4 12 Effects of market intelligence gap factors on the overall relationship quality ...... 95 4 13 Importance score comparison between two structures of advertising agencies ............................................................................................................. 96 4 14 Delivery score comparison between two structures of advertising agencies ...... 96 4 15 ImportanceDelivery Gap score comparison between two str uctures of advertising agencies ........................................................................................... 97 4 16 Relationship score comparison between structures of advertising agencies ...... 97 4 17 Summary of Regression Analyses for the relationship quality constructs ........... 98
8 4 1 8 Goodness of fit indices and X 2 tests for the measurement models .................... 98 4 1 9 Correlation matrix for the modified independent factor structural model ............. 99 4 20 Summary of t otal/direct/indirect effects of the modified independent factor model .................................................................................................................. 99
9 LIST OF FIGURES Figure page 1 1 A conceptual framework of influences of market intelligence gaps on perceived relationship qual ity ............................................................................. 20 2 1 Information flow model for full service agencies ................................................. 53 2 2 Proposed i nformation flow model for media planning/buying agencies .............. 53 2 3 Proposed i nformation flow model for inhouse agencies .................................... 54 2 4 General relationship quality model ..................................................................... 54 2 5 I ndependent factor model of relationship constructs .......................................... 55 2 6 A second order hierarchical model of relationship constructs ............................ 55 2 7 The modified independent factor structural model (Full model) .......................... 56 4 1 Factor loadings of the general factor measurement model ............................... 100 4 2 Factor loadings of the independent factor measurement model ....................... 100 4 3 Factor loadings of the s econdorder measurement model ............................... 101 4 4 Modified independent factor s tructural model for market intelligence gap and relationship quality (Full model) ........................................................................ 102 4 5 Modified independent factor s tructural model for market intelligence gap and relationship quality (Reduced model) ............................................................... 103
10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial F ulfillment of the Requirements for the Degree of Doctor of Philosophy AN EXAMINATION OF MARKET INTELLIGENCE GAPS IN THE ADVERTISING INDUSTRY AND THEIR EFFECTS ON AG E N CY CLIENT RELATIONSHIPS: MEDIA PLANNERS PERSPECTIVE By Jun Heo December 2010 Chair: Jo hn C. Sutherland Major: Mass Communication As the landscape of the media industry rapidly changed, needs for advertising media planners to understand the market place has increased. Failing to understand marketers needs may influence overall performance of the advertising agency and in turn terminate the agency client relationship. The purpose of this online survey of U.S. advertising media planners was to determine the effects of market information sharing on the quality of relationships between clients and media planners. The quality of client media planner relationships was defined from the media planners perspective on trust, commitment, satisfaction, and cooperation with the client. Market intelligence gaps, the difference between the media planners perceived value of specific market information and how often clients assure media planners have access to the information, were used to define information sharing. A gap analysis confirmed media planners do not receive certain types of market information as frequently as they would like. The results of the study provided evidence that the perceived information gap has a negative effect on the quality of the client media planner relationship, as perceived by the media planner. In addition, the results
11 suggested that the relationship satisfaction may play a critical role in improving affective commitment to the client and in turn intention to cooperate. The current study also attempted to compare the findings between two types of advertising agencies: full service agencies and media specialized agencies. The results demonstrated that m edia planners in fullservice agencies are less likely to perceive the gaps and, thus, more likely to report higher relationship quality than those in media agencies. This study extends advertising literature by applying relationship marketing theory to explain agency client relationships. Specific suggestions for both clients and media planners to improve the quality of relationships through information sharing are provided. Rec ommendations for future research in this area are also provided.
12 CHAPTER 1 INTRODUCTION Significance of the Relationship between Advertising Agencies and Client Firms In a business to business (B2B) environment, establishing, nurturi ng, and maintaining successful relationships with partner firms have been considered a key driver to firms performance (Crosby, Evans, and Cowles 1990; Lages, Lancastre, and Lages 2008; Morgan and Hunt 1994). Building effective relationship is more import ant particularly to professional business services (e.g., insurance, private banking, commercial real estate, advertising) to which long term relationships with partner firms is critical than to other B2B sectors (Crosby Evans, and Cowles 1990; Gronroos 1990). The intangible nature of service business makes professional services more applicable to the notion of relationship marketing (Zeithaml, Parasuraman, and Berry 1985). The intangibility increases the level of ambiguity in the relationship between agency and client. Clients might experience a difficulty to evaluate agencies service quality because of the intangible nature of the service. In fact, evaluating agency is often a matter of opinion, rather than objective decisionmaking (Gummesson 1988). Th e advertising industry, one of the professional service sectors, has witnessed a big plummet in the average relationship length between client firms and advertising agencies. Enns (2009) reported that the average length of relationship has decreased 25% down to 5.3 years in 1997 from 7.2 years back in 1984, expecting to see 4 years in 2010. More and more, the changing nature of the advertising industry, such as dominance of the big fours, fragmentation of media, and proliferation of specialized agencies, requires advertising agencies to understand developing, maintaining, and enhancing the relationship with their clients.
13 Although agency client relationships are of a big concern to the advertising industry, not much academic attention has been taken on inter firm relationships in the industry. While relationship marketing theory has suggested mutual benefits and commitments are key aspects of building a long term business relationship (Andaleeb 1996; Berry 1995; Dwyer and Oh 1987; Frazier, Spekman, and ONeal 1988), the primary interest among agency client relationship studies has been largely on clients expectations, not on the agencys expectation. Likewise, research regarding relationships between client firms and advertising agencies has been limited to id entifying factors influencing agency selection (e.g., Cagley 1986; Dowling 1994; Henke 1995) and relationship vulnerability (e.g., Doyle, Corstjens, and Michell 1980; Michell 1986/1987; Mi chell, Cataquet, and Hague 1992), and thus dynamics of the agency cl ient relationship such as relationship development has largely been disregarded. The relationship studies would benefit advertising agencies as being able to build a long term relationship may mitigate possible uncertainty, save termination costs and develop a stable business plan. Client firms also would benefit from the studies in that building long term relationship with their agency may increase agencys overall perception of the relationship with the client firms and, in turn, improve their outcomes. Accordingly, the current study will employ relationship quality constructs (i.e., trust, commitment, satisfaction, and cooperation) as dependent variables and communication, one of the important indicators of business relationships, as independent variable. The following section will briefly discuss about the significance of communication in the relationship between the client and the advertising agency.
14 Significance of Communication in Agency Client Relationships Communication between the agency and the client has been suggested as one of the important aspects of maintaining successful relationships with the client for an advertising agency (Beltramini and Pitta 1991; Clements 1984; Hot z, Ryans, and Shanklin 1982; Mi chell 1986/1987 ; Mi chell Cataquet and Hague 1992; Sutherland, Duke, and Abernethy 2004; Weilbacher 1983). As many advertising scholars have agreed, poor communication between the two parties is a main reason of divorcing each other. It may imply that termination of the relationship is not onl y because of agencys techniques (e.g., creative and media skill) but also communication. Communication is particularly important to the agency client relationships because the relationship is defined by people intensity and human interaction (Halinen 1997). People intensity means that knowledge is embedded in persons who provide services. Given that knowledge is the most significant resource of an advertising agency, successful knowledge sharing (i.e., dissemination of marketing intelligence) from the client would improve competency of an advertising agency and, in turn, produce better performance. Also, advertising services are produced and consumed in an interaction between clients and agencies (Lovelock 1983). Clients participation is crucial in the pr ocess of advertising services. Clients provide critical information to agencies and decide the acceptance of final materials. Thus, well shared communication between the agency and the client is specifically important to their relationships. B2B relations hip marketing literature has identified communication as a critical determinant of good or bad interfirm relationships (e.g., Anderson and N arus 1990; Bleeke and Ernst 1993; Frazier and Rody 1991; Morgan and Hunt 1994; Sharma and Patterson 1999). Well shar ed communication may create the feeling of being an integral
15 part of a team and, thus, lead to better cooperation (Mohr, Fisher, and Nevin 1996). In other words, agency staffers would be interested in their clients success and put more efforts on the acco unt when they are treated as partners rather than mere suppliers (Huntley 2006). Despite the importance of communication between interfirms, research concerning business to business communication has mainly focused on measuring communication quality and i dentifying communication problems. Nevertheless, very little research has conducted to determine from where the communication problems stem (e.g., Sutherland Duke, and Abernethy 2004). In their market information flow study, Sutherland, Duke, and Abernethy (2004) found discrepancies between market information that creative people value and the delivery of the information. The current study, following the market information flow model (Sutherland, Duke, and Abernethy 2004), will focus on media market intell igence gaps between media planners perception about the value of market intelligence and the actual delivery of the information, as a source of communication problem. Furthermore, this study will examine whether market intelligence gaps affect agency staf fers perception of the relationship with their clients. Importance of Media Market Intelligence As one of the task interactive organizations (Beard 1996; Mills and Margulies 1980), advertising agencies need to obtain copious information from their client s. The focus of the task interactive organizations is on completing tasks via exchange of information between firms. In other words, dissemination of necessary information from client firms to agency staffers is crucial for the agency to complete its tasks in the market place (Korgaonkar, Moschis, and Bellenger 1984). Thus, client representatives should disseminate all the key information to agency staffers at least to assure that the
16 agency receives the key informationin order f or the agency to carry out its tasks successfully. However, the industry and the academy alike suggested that not all the key market intelligence is delivered to agency staffers, and this fact is one of the major sources of agencys dissatisfaction with their clients. Sutherland Duke, and Abernethy (2004) explained that failures of market information delivery may be because clients 1) may not have such information or 2) may decide not to share with the agency. In addition, Abratt and Cowan (1999) explicated that disagreement on the importance of market intelligence may be a source of the dissemination failure. In other words, clients may believe they share necessary information for campaign development, whereas agency staffers complain that they do not receive key information from the client. Such a poor dissemination of key information may cause poor performance of the agency and, in turn, termination of the relationship. Thus, this current study will use a gap analysis to identify discrepancies between agency staffers perceived i mportance of market intelligence and perceived delivery of the market intelligence. The current study is particularly interested in market intelligence that is necessary for better media planning. As Heo and Cho (2009) pointed out, the ever changing medi a environment has increased the importance of media planning, and media planners became even a focal point of IMC campaign (Cappo 2003). In fact, media planners today are often asked to provide solutions to marketing problems. Therefore, dissemination of t he key market intelligence from clients to media planners may be the key to success or failure of media campaigns (Sissors and Baron 2002). This phenomenon is relatively recent; thus, little academic attention has been paid to media
17 market intelligence ( e.g., Abratt and Cowan 1999). As far as media planners are concerned, three types of agencies are involved; full service agencies, inhouse agencies, and media planning/buying agencies (Horsky 2006). According to Sutherland, Duke, and Abernethy (2004), rel ay points during communication process may serve as failing points of communication. Thus, it is reasonable to assume that media planners who experience different relay points of communication may have different perceptions about both the quality of market intelligence and the relationship quality with their clients. Accordingly, the current study will attempt, first of all, to identify a set of comprehensive market intelligence items for media planners and explore the gaps between values and actual deliveries of the market intelligence items among media planners. Moreover, the gaps identified will be compared across three different structures of advertising agencies: full service agencies, inhouse agencies, and media planning/buying agencies. In addition, the relationship quality also will be compared among the three types of agencies. Purpose of the Study Overall, this dissertation examines the effects of market intelligence gaps on media planners perception of the relationship with their clients. The market intelligence gap specifically refers to as discrepancies between the perceived value of media intelligence and the actual delivery of the market intelligence from the client to media planners. The purposes of this dissertation are to 1) identify a s et of exhaustive market intelligence that agency staffers (i.e., media directors and media planners) would like to receive, 2) explore whether there are gaps between perceived value of market intelligence and perceived dissemination of the market intelligence, 3) examine whether
18 the market intelligence gap differ s by the structure of advertising agencies and thus affect s the perception of agency client relationships differently 4) examine the effects of the market intelligence gaps on media planners percei ved relationships with the client firm, 5) identify which market intelligence has more/less influences on the relationship quality, measured by trust, commitment, satisfaction, and coordination, and finally 6) identify the structure of relations hip quality constructs in the context of agency client relationships. A conceptual framework was developed in order to offer a clear foundation for this study (Figure 11). This framework describes market intelligence items selected for this study, structure of relationship constructs, and the relationship between market intelligence gaps and perceived relationship quality, and the role of potential control variables on the relationship between relationship quality constructs and market intelligence gaps. A gap analy sis was designed to identify the gaps between perceived value of each type of market intelligence and delivery of the market intelligence (i.e., market intelligence gaps = Vm), where F is the perceived frequency of the communication, c, and V is the value of the market intelligence, m.) Furthermore, this current study used various statistical techniques for addressing research question and testing proposed hypotheses, such as analysis of variance, multiple regression, factor analysis, and structural equation modeling. Theoretically, this research adds the growing body of knowledge on the relationship marketing in the context of professional service business. Especially, this study focuses on expectations and perception about the relationship from the perspective of service providers, rather than of buyers, which have rarely examined.
19 The current study also empirically indentifies communication problems between agencies and clients by defining market intelligence gaps. Managerially, the study provides an exhaustive checklist of important market intelligence both to client representatives and media planners. It further decreases clients role ambiguity and, in turn, improves advertising staffers perceived relationships (i.e., trust, commitment, satisfact ion, coordination) with their clients. In addition, this study evokes the importance of the notion of relationship marketing if not transaction marketing in the context of agency client relationships.
20 Figure 11. A conceptual framework of influences of market intelligence gaps on perceived relationship quality Item 1 Item 2 Item 3 Item 4 Item 5 Item 6 Item 7 It em 8 Item 9 Item 10 Item 11 Item 12 Item 13 Item 14 Market Intelligence Gaps Relationship Quality Trust Commitment S atisfaction Cooperation Trust 1 Trust 2 Trust 3 Trust 4 Commitment1 Commitment2 Commitment3 Commitment4 Commitment5 Commitment6 Satisfaction1 Satisfaction2 Satisfaction3 Satisfaction4 Cooperation 1 Cooperation 2 Cooperation 3 Cooperation 4 Satisfaction5
21 CHAPTER 2 LITERATURE REVIEW Relationship Marketing in Professional Service Business Relationship mark eting is a relatively recent phenomenon in the marketing discipline. While the idea of relational exchange in the marketing area has emerged back in the late 1960s, the concept of relationship marketing was formally introduced by Berry (1983) back in 1983. Since its inception, many scholars in the marketing arena have studied the phenomenon, and the landscape of marketing has enormously changed both in pr actices and academia. Berry (1983) explained that relationship marketing, unlike transaction marketing, is based on a mutually beneficial exchange between manufacturers and consumers and, accordingly, building long term relationships with consumers (Berry 1983). Gronroos (1990) further suggested that relationship marketing is not only for consumer marketing, but also for the service encounter (i.e., the seller buyer relationship in professional service sectors). Gronroos (1990) understood that practices of relationship marketing in the professional service sectors are completed by giving promises, fulfilling promises, and giving a new set of promises. In a study of relationships between retailers and dealers, Morgan and Hunt (1994) suggested that relationship marketing can be extended to all forms of relational exchanges in supplier, buyer, lateral, and inter nal partnerships (p.23). They specifically suggested ten types of relational partners: goods suppliers, service providers, competitors, nonprofit organizations, government, ultimate customers, intermediate customers, functional departments, employees, and business units. In the context of professional services such as advertising, service provider buyer relationships have been of interest. Crosby Evans, and Cowles (1990), particularly,
22 pointed out that professional business services need to focus on bui lding effective relationships with partner firms. Three characteristics of professional business services that make the concept of relationship marketing more applicable to the professional service area were discussed (Crosby Evans, and Cowles 1990, p. 69): 1) the service is complex, customized, and provided over a continuous transactions, 2) many buyers are relatively less knowledgeable about the service, and 3) the environment is dynamic and unclear about future supply and demand. In addition, the intang ible nature of service business makes the relationship concept more relevant to evaluating the quality of professional service providers such as financial counseling, real estate, and advertising (Zeithaml, Parasuraman, and Berry 1985). The concept of rela tionship marketing is also extended to interorganizational (e.g., employees and functional department) or channel relationships (e.g., business units) (Morgan and Hunt 1994). Bowen and Schneider (1988) specifically pointed out that consumers seem more like ly to contend with the service when employees of the service provider value their organizations. This implies that happy employees do things because they want to, rather than because they need to, and thus exert more efforts on customer satisfaction. The c hannel literature has suggested that relationships among channel members (e.g., distributors and manufacturers) are integral parts of success of each firm in a given marketing channel (Anderson and Narus 1990; Kalwani and Narayandas 1995; Keith, Jackson, and Crosby 1990). Thus, establishing, developing, and maintaining successful relationships with employees or channel members would facilitate success of organizations (Anderson and Narus 1990; Dwyer and Oh 1987;
23 Guiltinan, Rejab, and Rodgers 1980; Mathieu and Zajac 1990; Meyer et al. 1989; Mohr, Fisher, and Nevin 1996). Adopting the notion of relationship marketing contributes to success of an organization in many different ways. Kalwani and Narayandas (1995), for instance, compared the firms that employ a r elational approach to the firms that employ a transactional approach. They found that the firms of a relational approach outperformed their counterparts who employ a transactional approach on both level of sales growth and profitability. Other than the fin ancial benefits, many intangible merits have been reported: long term commitment (Crosby Evans, and Cowles 1990; Mohr, Fisher, and Nevin 1996; Morgan and Hunt 1994; Sharma and Patterson 1999), reduced uncertainty (Morgan and Hunt 1994), enhanced satisfact ion and intention to coordinate (Lages, Lancastre, and Lages 2008; Mohr, Fisher, and Nevin 1996), and anticipation of future interaction (Crosby Evans, and Cowles 1990), and better job performance (Meyer et al. 1989). Among all the r elational partners (M organ and H unt 1994), this study specifically focuses on service providers (i.e., advertising agencies) and ultimate consumers (i.e., clients). Specifically, the focus of this study is on relationship outcomes such as commitment, satisfaction, and cooperat ion from the perspective of professional service providers. Agency Client Relationships in the Advertising Industry The dynamics of agency client relationships in the advertising industry derives from a distinctive characteristic: intangibility (Parasurama n, Zeithaml, and Berry 1995). Consumers of tangible goods may employ a variety of cues to evaluate product quality, such as package, label, style, and etc. When it comes to advertising services, however,
24 tangible evidence is limited to agencys physical facilities, equipment, and personnel (Parasuraman Zeithaml, and Berry 1995). Therefore, client firms depend on other intangible cues, such as perceived relationship, to evaluate the service quality of advertising agency. The concept of relationship marketing is more important to the agency client relationship in the advertising industry than any other service encounter because the relationship between advertising agencies and client firms is especially vulnerable. A transactional approach to serve client fir ms may result in higher competition among agencies, offering reduction of cost and fees (Kalwani and Narayandas 1995). It is relatively easy for client firms to switch their advertising agencies because significance of intangible effects and risks that are involved in divorcing is hard to evaluate and easily ignored (Kulkarni, Vora, and Brown 2003). Thus, it is highly plausible that the average agency client relationship length plummets down to less than 4 years if advertising agencies employs a transactional approach, rather than a relational approach to serve client firms. Another factor that defines the dynamics of agency client relationships is peoplein tensity (Halinen 1997). Halinen (1997) explained that the evaluation of advertising services is mostly bound to individual s who provide with the services. The human element of the relationship may imply that the relationship between agencies and clients is mainly affected by peoplerelated reasons (Heekin 1983; cited in Wackman, Salmon, and Salmon 1986/ 1987), which include arrogance, an inability to listen, obsession with new accounts, frequent staff changes, and coordination problems. Studies have also indicated a variety of reasons of relationship termination between clients and advertising agencies. Doyle Corstjens, and Michell (1980) found
25 that dissatisfaction with agency performance was the main reason for termination. In a longitudinal study, Henke (1995) identified key predictors that influence agency termination and suggested agencys creative and media skills are the key aspects of agencys performance. In addition, many industry publications (e.g., Weilbacher 1983) suggested other factors influencing divorce between advertising agencies and client firms including: a change in the top management of the client firm; a sense that a client had outgrown the agency through internal development or merger; a change in client marketing philosophy or strategies; and a change in agency due to growth or acquisition of new, competing accounts. From an agency s perspective, Beard (1996) argued that clients role ambiguity, which results in ineffective performance of clients role, may be the source of poor outcomes of advertising agencies, clients dissatisfaction and, eventually, agency termination. In the m anagement literature, Huntley (2006) argued that one must make a distinction between partner and vendor to understand the atmosphere in the context of business to business market. She elaborated that service providers in a vendor atmosphere tend to be passive, whereas those in a partner atmosphere are likely to be motivated to work closely with customers for their collaborative goals. The partner atmosphere defines the highest excellence in the relationship based on the high level of bonding (Huntley 2006). In an agency client relationship study, Beverland, Farrelly, and Woodhatch (2007) suggested that agencys proactivity, which maybe a consequences of the partner atmosphere, is positively associated with clients satisfaction and therefore positive renewal decision. It implies that when an agency is perceived as a partner, the agency is more likely to get involved in clients business, satisfy the client,
26 and collaborate to achieve clients goals (Haytko 2004; Ulaga and Eggert 2006). Given that integrat ed marketing communication (IMC) increased the importance of collaborative efforts between agencies and client representatives (Beard 1996), the concept of relationships should be viewed as one of the imminent issues in the advertising industry. Although agency client relationships are of a major concern to the advertising industry, little research has focused on interfirm relationships in the industry. In his book, Relationship Marketing in Professional Services, Hanlinen (1997) pointed out that research regarding relationships between client firms and advertising agencies has been limited to identifying factors influencing agency selection (e.g., Cagley 1986; Dowling 1994; Henke 1995) and relationship vulnerability (e.g., Doyle, Corstjens, and Michell 198 0; Michell 1987; Michell, Cataquet, and Hague 1992), and thus dynamics of the agency client relationship such as relationship development has largely been disregarded. As a matter of fact, building successful agency client relationships has been of interes t to advertising agencies, rather than to clients firms. An advertising agency may face a devastating situation when loosing an account (Wackman Salmon, and Salmon 1986/ 1987), whereas clients firms easily ignore consequences of divorcing that are intangible (Kulkarni, Vora, and Brown 2003). However, based on the fact that the agency client relationship is usually initiated by client firms, much research on the relationship has focused on issues and benefits of client firms (Holmlund 2008; Kalwani and Naray andas 1995). Understanding agencies perception of relationships with their clients, however, is critical to understand the relationship dynamics because mutual benefit and commitment are key aspects of building a long term business relationship
27 (Berry 199 5; Dwyer and Oh 1987; Frazier, Spekman, and ONeal 1988). In addition, relationships perceived by an advertising agency may influence the performance of the agency and in turn building successful long term relationships with the client (Kalwani and Narayandas 1995). Thus, the current study attempts to explore advertising agencies perception of their relationship quality with the clients, as a dependent variable. Findings from the relationship study would benefit advertising agencies as being able to build a long term relationship may save termination costs and develop a stable business plan. Communications in the Context of Agency Client Relationships In the relationship marketing literature, communication has been identified as a critical determinant of good or bad interfirm relationships (e.g., Anderson and Narus 1990; Bleeke and Ernst 1993; Frazier and Rody 1991; Friman et al. 2002; Morgan and Hunt 1994). Bleeke and Ernst (1993) elaborated that communication between firms is the most important element to building and maintaining successful relationships. Sharma and Patterson (1999) reported that communication is the single powerful determinant of relationship commitment. In a manufacturer dealer relationship study, Mohr, Fisher, and Nevin (1996) specifical ly suggested that collaborative communication may make dealers feel like an integral part of the team (p. 103), and enhance dealers satisfaction with, coordination with, and commitment to the manufacturer. In addition, communications between service providers and customers often is referred to as an indicator of service providers performance (Beard 1996; John, R u e kert, and Churchill 1983) and outcomes of the firms (Anderson and Narus 1990; Anderson and Weitz 1989, 1992; Boyle et al. 1992; Frazier and Rody 1991; Guiltinan, Rejab, and Rodgers 1980).
28 On the other hand, communication difficulty has been viewed as a vital source of problems in interfirm relationships (Mohr and Nevin 1990; Mohr, Fisher, and Nevin 1996; Sharma and Patterson 1999). Sharma and P atterson reported that more than 50% of complaints against professional services are related to poor communication. Anderson and Narus (1990) also suggested that lack of communication between the partner firms results in a lack of cooperation and in turn poor performance. In the advertising agency client firm relationship literature, much research identified communication problem as a major reason for agency termination (Beltramini and P itta 1991; Clements 1984; Hotz Ryans, and Shanklin 1982; Hur ley, Gro pper, and Roma 1996; Mi chell 1986 /1987; Mi chell Cataquet and Hague 1992; Weilbacher 1983). As Wackman, Salmon, and Salmon ( 1986/ 1987) argued, an advertising agency may face a devastating situation when the agency client communications is not healthy, and accordingly the termination happens abruptly. Specifically, in the field of advertising, many researchers suggested that communication is one of the important aspects of maintaining successful relationships with client firms for an advertising agency (B eltramini and Pitta 1991; Clements 1984; Hotz Ryans, and Shanklin 1982; Michell 1986/1987; Mi chell Cataquet and Hague 1992; Sutherland, Duke, and Abernethy 2004; Weilbacher 1983). Shared communications may create the feeling of being an integral part of a team and, thus, lead to better cooperation, commitment, and satisfaction with the client (Mohr, Fisher, and Nevin 1996; Mohr and Nevin 1990). Beverland, Farrelly, and Woodhatch (2007) specifically pointed out that well shared client relevant information is a precondition of agencys proactivity, and it will in turn increase agencys involvement in clients
29 business. In fact, given that knowledge is the most significant resource of an advertising agency (Halinen 1997), successful knowledge sharing (i.e., dissemination of marketing intelligence) from the client would improve competency of an advertising agency and, in turn, produce better performance. Korgaonkar, Moschis, and Beilenger (1984), for instance, have found that obtaining necessary information fr om the client is associated with success of advertising campaigns. Research suggested that client representatives may be responsible for poor performance of service providers because of poor communication initiated by the client (Beard 1996; Mills and Morr is 1986; Weilbacher 1983). Sutherland, Duke, and Abernethy (2004) discussed possible reasons for the poor communication between agencies and clients; a lack of information to share, an intention not to share, poor communication processes, agencys internal communication problem. The relationship marketing literature concerning communication has attempted to measure communication quality and identify many facets of communication between firms and/or channel members (e.g., Malts and Kohli 1996; Mohr and Nevi n 1990; Mohr, Fisher, and Nevin 1996). In their model of channel communication, Mohr and Nevin (1990) included four communication strategies: frequency, modality, contents, and direction. Anderson and Narus (1990) suggested three dimensions of communicatio n quality: timely, reliable, and relevant. Also, many other indicators of interfim communication were used: formality of communication (Anderson, Lodish, and Weitz 1987; Howell 1987; Moorman, Deshpande, and Zaltman 1993; Sharma and Patterson 1999), bi dire ctionality (Anderson, Lodish, and Weitz 1987; Farace, Monge, and Russell 1977), and noncoersive contents (Frazier and Rody 1991). Among those indicators of
30 communication quality, frequency has been the most widely used indicator of the amount of communicat ion in the organizational communication literature (Farace, Monge, and Russell 1977; Malts and Kohli 1996; Mohr and Nevin 1990; Morgan and Hunt 1994). Malts and Kohli (1996) emphasized the importance of frequency with which market intelligence is dissemina ted. While many advertising scholars attempted to identify communication problems between agencies and clients (Beltramini and Pitta 1991; Clements 1984; Hotz Ryans, and Shanklin 1982; Hurley Gropper, and Roma 1996; Michell 1986 /1987; Michell, Cataquet, and Hague 1992; Weilbacher 1983), limited number of research has exerted efforts on determining from where the communication problems stem (e.g., Sutherland Duke, and Abernethy 2004). In their market information flow study, Sutherland, Duke, and Abernethy (2004) identified a set of market information that is communicated between the clients and the agencies and found discrepancies between market information that creative people value and the actual delivery of the information. Further, they pointed out tha t relay points during the communication process (e.g., account manager, media director, and client representative) may distort information and thus impede the dissemination of information. It implies that market intelligence may not be transferred properly if there are many relay points involved in the process of communication between the client and the media planner. The current study, following the market information flow model (Sutherland, Duke, and Abernethy 2004), will use a comprehensive set of marke t intelligence disseminated from clients to media planners and discrepancies between media planners perception
31 about the value of market intelligence and the actual delivery frequency of the information. Accordingly following research questions were devel oped; Research Question1: Do media planners consider certain types of market intelligence more important than other types market intelligence? Research Question 2: Are all the types of market intelligence delivered equally to, less than, or more than the v alue of the market intelligence, suggesting market intelligence gaps between the value of market intelligence and the delivery of the market intelligence? Market Intelligence Gap Market intelligence shared between clients and agencies has been viewed as an essential precondition for successful advertis ing campaign (Helgesen 1994; Mi ch ell 1986; Sissors and Baron 2002). According to Mills and Margulies (1980), interaction of service organizations are classified into three different types: maintenance interactive, task interactive, and personal interactive. Beard (1996) included the agency client relationship in the task interactive classification. The focus of these task interactive organizations is on the tasks to solve problems, suggesting that the interac tion between agency employees and client representatives requires an exchange of prolific information needed to complete tasks. From the task interaction perspective, dissemination of the necessary information from clients to agency staffers may be the key to better outcomes of the agency and, in turn, successful agency client relationships (Beard 1996; Korgaonkar, Moschis, and Bellenger 1984). Thus, client representatives should be sure that the agency regularly receives all the critical information and al so minor information of interest to the agency. Zack (1999, 2003), however, pointed out that research has mainly focused on knowledge management process, rather than dissemination of knowledge. Bierly, Kessler, and Christensen (2000) argued that one of the biggest impediments to transfer
32 knowledge is receivers perception of usefulness of the knowledge. Limited number of advertising literature has had interest in usefulness of market intelligence for advertising staffers. Sutherland, Duke, and Abernethy (2004) reported that there exists a discrepancy between importance of market information and how often the information was received, so called market intelligence gaps. They identified six categories of market information that are necessary for creative st affers to develop their creative: target audience demographic profile, a main selling point, product performance, marketing strategy, product usage, competitors product performance. Further, they found that certain types of market intelligence, such as target audience demographic profile and marketing strategy, are more valued by the receivers (i.e., creative directors, copy writers, and art directors), and each type of information is delivered equally to, less than, or overly than the value of the market information. Few studies, to the best of my knowledge, have examined consequences of market intelligence gaps. Korgaonkar Moschis, and Bellenger (1984), for instance, found that poorly downloaded information is related to unsuccessful advertising campai gn. Sutherland, Duke, and Abernethy (2004) suggested that not having certain types of information hinders creative staffers to develop better advertising. As Beard (1996) suggested a collative efforts for a better communication of both service providers and their customers facilitates outstanding outcomes of the service providers. Consequently, it is critical to understand which market intelligence is more/less useful for advertising agencies, whether the intelligence is properly downloaded to the agency, a s well as whether the information gaps influence agency staffers perception of the relationship with their client firms.
33 Thus, the current study will focus on identifying market intelligence gaps in the advertising industry and consequences of the gaps s uch as perceived relationships of agency staffers with their clients. Media Market Intelligence Media planning has once been largely neglected by advertising practitioners because their focus was on advertising messages rather than pipes or conduits o f the message (Schwartz 1983). Horsky (2006) said, by citing Lodish (1986), that the media tasks were even considered a necessary evil. Cagley (1986) however pointed out that agencies tend to underestimate the importance of media creativity, and it is a so urce of clients dissatisfaction. Stemming both from unprecedented proliferation of media platforms and the highly fragmented media audience, media planning today became a critical aspect of advertising campaign (Heo and Cho 2009, p.145). The radical ch ange in the media landscape has raised skepticism about conventional approaches to media planning among marketers, and thus the role of media planners has never been emphasized as it is now. In the context of integrated marketing communication, media planners increasingly play a leading role (Abratt and Cowan 1999; Cappo 2003). Cappo (2003) mentioned that media planners may be a focal point for future IMC campaigns, citing a presentation given by Heyer (2002), then CEO of CocaCola Ventures: it [being stra tegic partners with advertising agencies] starts with a better understanding of the media plan that ties to consumer touch points more than it does to the creative idea (p.152). Media planning is no longer behind the scenes science, but a critical indicator of agency client relationships (Henke 1995; Lloyd, Slater and Robbs 2000). In a longitudinal study, Henke (1995) reported that dissatisfaction with agencys media skill
34 is a great predictor of divorcing agencies, whereas creative skill is an important determinant of winning client. This changing perception of media planning strongly suggests the need for media planners to understand marketing situation of the client and provide with solutions to the marketing problems. However, despite the increasing i mportance of media planning, very few studies have explored information needs specifically for media planners (e.g., Abratt and Cowan 1999). In an exploratory study, Abratt and Cowan (1999) indentified 68 types of media market intelligence and categorized them into 13 information categories: 1) budget, 2) communication planning and strategy, 3) consumer behavior, 4) competitive, 5) creative, 6) historical, 7) market size, 8) media, 9) information concerning objectives, 10) product/brand, 11) sales, 12) segm entation and target market, and 13) timing. They further compared media planners perception of the importance of media market intelligence to clients perception and found that there are significant differences in importance measure between media planners and clients. These discrepancies would create impediments for media planners to acquire necessary information from their clients (Abratt and Cowan 1999; Sutherland, Duke, and Abernethy 2004). Another obstruction for media planners to successfully receive media market intelligence from the client is, as Sutherland, Duke, and Abernethy (2004) discussed, a failure point. Communication fails when many relay points (e.g., client gate keepers, account managers, and media directors) are present in the sharing of the market intelligence. As a result of the failure points, the market intelligence may be misunderstood by the media planners, resulting in lower job satisfaction, higher stress,
35 and consequently unsuccessful media c ampaigns (Sissors and Baron 2002 ; Suth erland, Duke, and Abernethy 2004). Accordingly, the current study will attempt to identify a comprehensive set of market intelligence that is necessary for media planners to create better media planning. Also, it will explore whether media market intellig ence gaps exist and the gaps influence relationship quality perceived by the service provider, media planners. This study also examines differences in dissemination of the media market intelligence among various structural characteristics of advertising agencies, in that communication failure points (Sutherland, Duke, and Abernethy 2004) exist differently depending on structures of advertising agencies. Structural characteristics of advertising agencies are discussed in the next section. Structures of Adv ertising Agencies Over the past four decades, the advertising industry has witnessed increasing number of mergers and acquisitions among agencies and changes in architecture of the industry (Ducoffe and Smith 1994). The changing nature of the advertising i ndustry, such as emergence of the big four (i.e., Omnicom, WPP, Interpublic, and Publicis Groupe) and fragmentation of media and audiences, encouraged large agencies to spin out their specialized departments as separate companies (Horsky 2006). In fact, the advertising industry is now highly concentrated with the four advertising conglomerates, and the big companies have led the way to diversification of advertising agencies. Due to many benefits that the large agencies may provide (i.e., economies of scale, bargaining power with the media, expertise in that particular industry, etc.), they have been able to attract many account, even conflicting accounts, into a single roof (Horsky 2006; Silk and Berndt 1993). However, account conflicts forced the communication
36 giants to split off its sections as a separate business (Villas Boas 1994). Also, the proliferation of digital media and cable channels have indeed increased chances for specialized media agencies to emerge. As a result of all those changes, marketers nowadays are enjoying the ever increasing number of agency options from full serv ice agencies to highly specialized marketing service agencies (e.g., creative boutique, media planning/buying agency). Advertising agencies differ by ownership and the range of services. In his book with regard to the Finland advertising industry, Halinen (1997) classified advertising agencies concerning agency ownership into three categories: local independent, inhouse, and international agencies. Local independent agencies are owned by local entrepreneurs, and inhouse agencies are owned by advertisers themselves. Meanwhile, international agencies are owned fully or partly by international chains. Those international chains are often called holding companies in that they con trol a number of agency brands all over the world. Horsky (2006), on the other hand, classified agencies regarding services provided into two options: bundling and unbundling options. While bundling includes full service and inhouse agencies that provide both creative and media services, unbundling contains of creative boutiques and media buying shops, which carry out specialized tasks suggesting different market information needs among different types of agencies Horsky (2006) argued that creative and r esearch tasks are remaining in full service agencies, but media tasks tend to be aggressively outsourced. This phenomenon derives from the facts that media planning/buying agencies offer economies of scale, such as buying power with the media, and mediane utrality, which advertising is placed in a medium that fits to clients
37 needs, not just the one that earns more money for the advertising agency. The medianeutrality particularly is possible because feebased compensation (i.e., prefixed amount or percentage of compensation) is largely accepted among media agencies, rather than commissionbased compensation (i.e., 15% of the media space they purchased). The feebased system, however, has threatened media agencies to take a cut of as little as 2 3 percent of billings and thus became a detrimental to stability of media agencies. As far as media planners presence, there are three types of advertising agencies: fullservice agency, inhouse agency, and media planning/buying agency (Horsky 2006). Those agenc ies have different structures (i.e., relay points) through which communicate with their clients. In explaining communications between clients and creative staffers in the fullservice agency, Sutherland, Duke, and Abernethy (2004) suggested that communicat ion relay points may serve as failing points of communication. They specifically pointed out that replay points include gatekeepers in client firms (e.g., client representatives) and in agencies (e.g., account managers). Clients gatekeepers impede the flow of market information by deciding how much information to share with the agency (Sutherland, Duke, and Abernethy 2004, p.42). In addition, internal communication within the agency has been viewed as a major impediment of the information flow ( Hurley Gro pper, and Roma 1996; Vanden Bergh, Smith, and Wicks 1986). Vanden Bergh, Smith, and Wicks (1986), for instance, found that 47% of agency staffers (e.g., creative team) felt that account managers, agency gatekeepers, did not share client information properl y.
38 The current study adopted Sutherland, Duke, and Abernethy s (2004) information flow model i n order to explain communication flow from the client to the media planner (Figure 2 1). Media planning/buying agencies are believed to have minimal number of fa iling points because media planners (or media directors) directly communicate with the client representatives (Figure 2 2). In addition, inhouse agencies are assumed to have different relay points in that clients for the inhouse agency are internal employee within the same organization with the media planner (Figure 23) Based on the proposed information flow models, t h is study assumes that the less relay points that are involved during the communication process, the better communication between client representatives and media planners T herefore, this study explores whether media market intelligence gaps exist differently across the three agency types (i.e., full service agencies, media planning/buying agencies, or inhouse agencies) Accordingly, the following three hypotheses with regard to relationships between the agency structures and the market intelligence gaps were proposed; Hypothesis 1: Three agency structures, full service agencies, inhouse agencies, and media planning/buying agencies, will differ in their perception of importance of each type of market intelligence. Hypothesis 2: Media planning/buying agencies will perceive that each type of market intelligence items will be provided to them more often than will full service agencies and inhouse agencies. Hypothesis 3: Market intelligence gaps will be significantly smaller for media planning/buying agencies than for full service agencies and inhouse agencies. Relationship Quality Constructs Relationship marketing theory (Berry 1983; Dwyer and Oh 1987; Gronroos 1984; Zeithaml, Berry, and Parasuraman 1988) is considered to derive from social psychology, namely, social exchange theory. Accordingly, concepts/variables used in
39 relationship marketing theory share similar ideas of social exchang e theories. Social exchange theory assumes that exchange partners committed to each other tend to resist attractive short term alternatives, invest more in their relationships, and act less opportunistically (Morgan and Hunt 1994). Following these ideas, r elationship marketing theory uses commitment as one of the major concepts. Additionally, many other constructs have been used to explain the relationship among exchange partners, such as communication, trust, satisfaction, cooperation, relational dependenc y, intimacy, and love. Among many relationship constructs, four constructs are chosen to be included in this study: trust, commitment, satisfaction, and cooperation. There are several reasons why the four constructs are chosen. First, these indicators have appeared as integral constructs of relationships in many areas of study including interorganizational relationships, relationship marketing, and channel management (e.g., Anderson and Narus 1990; Anderson and Weitz 1989; Baker, Simpson, and Siguaw 1999; D wyer Schurr, and Oh 1987; Morgan and Hunt 1994; Rauyruen and Miller 2007). With regard to the relationship between relationship quality constructs and communication, many studies identified these dimensions as critical outcomes of communication quality. M ohr, Fisher, and Nevin (1996), for instance, suggested that collaborative communication is a precursor of commitment, satisfaction, and cooperative norms and, in turn, build a long term relationship. Friman et al. (2002) defined communication between firms as a prerequisite of commitment and trust. In addition, communication has been suggested to be closely related to mitigating conflict, increasing trust, and improving coordination (Anderson and Narus 1990; Anderson and Weitz 1989; Stem and El Ansary 1988)
40 Second, these constructs are considered key concepts that capture perceptions of the subordinators (e.g., dealers, franchises, and agency staffers) in the relationships. As a matter of fact, these constructs are predominantly applied to clients perspec tives. Holmlund (2008), however, argued that suppliers perspectives should not be disregarded so that one may understand the quality of an entire business relationship. In a study of relationships between computer manufacturers and dealers, Mohr, Fisher, and Nev i n (1996) explained that commitment, satisfaction, and cooperation are key channel outcomes that measure dealers perception of the manufacturer. In addition, Baker Simpson, and Siguaw (1999) identified four constructs (i.e., trust, commitment, sat isfaction, and cooperative norms) as key variables of relationships perceived by suppliers (e.g., media planners for this particular study). Although the focus of the current study is on relationship quality as outcomes of the communication flow, it should be noted that relationship quality is also viewed as an indicator of many outcomes. Many possible consequences of relationship quality constructs have been suggested as follows; service quality ( Woo and Ennew, 2004; Bennett and Barkensjo 2005); relationship strength (Storbacka Strandvik, and Grnroos 1994); relationship longevity (Storbacka, Strandvik, and Grnroos 1994; Scanlan a nd McPhail, 2000; Friman et al. 2002); customer retention (Hennig Thurau and Klee, 1997); relati onship enhancement (Selnes, 1993 ) and continuity (Selnes, 199 3 ); future intentions regarding the relationship (Garbarino and Johnson, 1999; Venetis and Ghauri, 2004; Ulaga and Eggert, 2006); propensity to leave the relationship (Morgan and Hunt, 1994; Ulaga and Eggert, 2006).
41 Based on t he literature review on the notion of market intelligence and the relationship quality constructs, following research questions were developed; Research Question 3: Is the level of market intelligence gaps related to the level of overall relationship quali ty? Research Question 4: Are certain types of market intelligence items related to overall relationship quality than are other types of market intelligences? In addition, i n order to examine the relationship between the communication flow that are differe nt across agency structures and the relationship quality constructs, following hypotheses were proposed; Hypothesis 4: The level of overall perceived relationship quality will be higher for media planning/buying agencies than for full service agencies and in house agencies. Hypothesis 5: The smaller the market intelligence gaps, (a) the greater the media planners trust in their clients, (b) the higher media planners perceived commitment to their clients, (c) the more satisfied media planners are with thei r clients, and (d) the greater media planners intention to cooperate with their clients. Trust Trust has broadly been defined as the belief about exchange partners that one is reliable and that the one will fulfill obligations in an exchange relationship (Andaleeb 1996; Shurr and Ozanne 1985). In a buyer seller context, trust has been viewed as beliefs about sellers in terms of expertise, reliability, and intentionality (Ganesan 1994; Moorman Deshpande, and Zaltman 1993; Morgan and hunt 1994) Garbarino and Johnson (1999) addressed that trust is an essential ingredient for successful long term relationships. Further, Andaleeb (1996) explained that trust is a key to an exchange relationship because low trust situation may increase complexity in the relationship, making it risky, costly, and difficult to continue (p.79).
42 Many studies concerning t rust have largely been interested in buyers trust in sellers (Schurr and Ozanne 1985). It is reasonable in that buyers in many cases have the power to enter or leave the relationship. However, in a study of the role of trust and dependence on relationships in marketing channels, Andaleeb (1996) emphasized the importance of understanding suppliers trust in buyers (i.e., advertising agencies trust in clients). Anda leeb (1996) convincingly argued that suppliers may face manipulation (opportunism) and potentially produce negative outcomes when a powerful party (such as clients) is perceived as untrustworthy, and have a dominant power over the relationship. Low level o f trust may generate dissatisfaction, mitigate commitment, and in turn produce negative outcomes (Andaleeb 1996). Schurr and Ozanne (1985) suggested that lower trust is likely to activate defensive behavior to conceal true attitude toward exchange partner. In other words, an advertising agency may behave opportunistically and produce not so good outcomes when they perceive the client as untrustworthy but need to maintain the relationship. Much scholarly work has included trust as one of the important dimens ions of relationship quality (e.g., Crosby Evans, and Cowles 1990; Dorsch, Swanson, and Kelly 199 8 ; Dwyer and Oh 1987; Han, Wilson, and Dant 1993; Jap, Manolis, ans Weitz 1999; Morgan and Hunt 1994; Wilson and Jantrania 1996). Along with commitment and sa tisfaction, trust has been one of the most frequently used dimensions in measuring relationship quality (Athanasopoulou 2006). In terms of determinants of trust, Anderson and Weitz (1989) reported that reputation, satisfaction, and experience are some of t he determinants of trust. Further, Ganesan (1967) added vendors investment in the relationship as one of the determinants of trust. Meanwhile, many scholars have
43 suggested that trust may be one of the preconditions for relationship quality, rather than a dimension of the relationship quality constructs (e.g., Friman et al. 2002; Hibbard, Kumar, and Stern 2001; Lagace, Dahlstorm, Gassenheimer 1991 ; Ndubisi 2006). Thus, this study examines service providers (agencies) trust in buyers (clients) and includes trust as one of the important dimensions of the relationship quality In addition, it examines whether this particular construct works as an antecedent of the relationship quality. Commitment Commitment, similarly to trust, has been viewed as an importan t dimension of relationship quality Commitment is broadly defined in the business to business market as a perceived importance of relationships between firms, which is underlying basis of business relationships (Friman et al. 2002; Hakansson and Snehota 1995). Anderson and Weitz (1992) specifically defined relationship commitment as the desire to develop a stable relationship, a willingness to make short term sacrifice to maintain the relationship, and a confidence in the stability of the relationship (p. 19). In their book Retailing Management Levy and Weitz (2000 ) mentioned that commitment is the major characteristic that differentiates relationship partnerships from functional relationships. Much relationship marketing literature has identified commitm ent as an important indicator of developing successful long term relationships (Anderson and Weitz 1992; Kumar, Scheer, and Steenkamp 1995; Morgan and Hunt 1994; Rauyruen and Miller 2007). For instance, Morgan and Hunt (1994), in their KMV (Key Mediating V ariable) model, proved that commitment and trust are to be key mediating constructs, rather than two more independent vari ables, of long term relationships.
44 In terms of the relationship between level of commitment and performance, many suggested positive relationships between the two constructs. Organizational studies concerning the commitment construct have specifically suggested that employees who are committed to the organization are likely to be motivated and get engaged in extra role that contributes to bet ter performance (Gronroos 1990; Schlesinger and Heskett 1991). Mowday, Porter, and Steers (1982) suggested that commitment is associated with motivation and involvement and, thus, makes partner firms to put considerable efforts. In a metaanalysis of organizational commitment studies, Mathieu and Zajac (1990) reported that employees commitment is highly correlated with job involvement and, in turn, creates a high level of performance. Additionally, in an agency clien t relationship study, Beverland, Farrelly, and Woodhatch (2007) suggested that agencys commitment enhances clients satisfaction through job performance and, thus, influences positively relationship renewal process. Although much research has suggested the importance of mutual commitment, focuses have been on clients commitment to the relationship because the relationship is usually initiated by the superior in the relationship. However, research should address subordinates perspective to understand dynamics of the relationship becaus e mutual commitment is a strong foundation of strong relationships (Berry and Parasuraman 1991). Thus, this dissertation focuses on agency staffers (i.e., media planners) commitment to their clients, by hoping that it shed light to understanding dynamics of agency client relationships. Various facets of commitment have been suggested. Chenet, Tynan, and Money (1999), for instance, summarized three dimensions of commitment: attitudinal,
45 behavioral, and normative. Attitudinal commitment includes identificat ion of the value of an organization, involvement in work activities or loyalty to the organization (Buchanan 1974). It involves an affective or emotional attachment to the partner firm (Allen and Meyer 1990). Behavioral commitment was described as individuals bond to an organization (Porter et al. 1974). This behavioral commitment is also referred to as calculative commitment because this commitment is formed by extrinsic interests such as benefits, rather than affects toward an organization (Becker 1960). Meanwhile, normative commitment is defined as the totality of pressures within the organization to act in a way that is consistent with organizations interests or goals (Wiener 1982) In the organizational commitment literature, o n the other hand, two di fferent conceptualizations about commitment have been popular: affective and continuance commitment. Porter et al. (1974) defined affective commitment as the strength of an individual's identification with and involvement in a particular organization (p. 604). This definition is in line with the definition of attitudinal commitment by Buchanan (1974) because it concerns organization identification and involvement. On the other hands, continuance commitment was defined as the aptitude to engage in consist ent lines of activity (Becker 1960, p. 33). Meyer and Allen (1984) explained that this tendency refers to anything of the value the individual would be lost when he or she were to leave the organization. At the interfirm level, when a firm has continuance commitment they would be engaged in partner firm because of the perceived costs associated with leaving the partner firm. Among many conceptualizations of commitment, a ffective commitment was found to be positively correlated with job performance, whereas continuance commitment was
46 negatively associated with job performance (Meyer et al. 1989). Meyer et al. (1989) argued that employees affectively committed to an organization do things because they want to, whereas those who have continuance commitment do things because they need to. It suggests that commitment is not simply an evaluation of the partner firm based on the perception of costs and benefits of the relationship (Anderson and Weitz 1992) but rather it is more likely to be a willingness to make s hort term sacrifice to gain long term benefits from the relationship (Dwyer Schurr, and Oh 1987). Furthermore, Robert s, Valki, and Brodie (2003) have reported that among three types of commitment: affective, continuance, and normative, only affective comm itment affected the degree to which customers want to retain the relationship with the firm. In their discussion, Mohr Fisher, and Nevin (1996) pointed out that it may generate mixed results of communication effects on commitment if a clear distinction between affective and continuance commitments was not made. Additionally, Young and Denize (1995) explained that strong emotional bonds, rather than economically based commitment, to the partner firm are likely to be linked with a high degree of cooperative understanding. C ommitment that an agency has to the client is much like involvement in clients business and psychological rather than physical --contract. Therefore, a gencys relationship orientation needs to be reconsidered if agencys commitment to the client is mostly based on continuance commitment. Because r elationship termination costs and risks associated with switching agency for the client are relatively small compared to other professional service sectors (Heide and John 1988), agencies that hold continuance commitment may face a devastating situation when the client abruptly leave the relationship. E conomically based commitment is highly vulnerable, and
47 affective commitment, which are psychological and sentiment attachments, is more important for a long term relationship between clients and agencies (Jarvis and Wilcox 1977; Schurr and Ozanne 1985; Young and Denize 1995). Accordingly, the current study focuses on agency staffers affective commitment to the client firm in that it may lead to client s satisfaction and eventually building long term relationships with the client (Schlesinger and Heskett 1991). Satisfaction Relationship satisfaction refers to as the cognitive and affective evaluation based on personal experience across all episodes wit hin the relationship (Roberts, Varki, and Brodie 2003, p. 17). This definition implies the importance of emotional evaluation in terms of overall experience between parties. The relational satisfaction differs from general customer satisfaction that is de fined based on transactionoriented evaluation (Olive r 1993). Many studies concerning relationship satisfaction conceptualized the construct as a cumulative affective status, rather than as the outcome of a transaction (Ab dul Muhmin 2005; Anderson, Lodish, and Weitz 19 8 7). Specifically, Abdul Muhmin (2005) emphasized the role of relationship satisfaction in developing agency client relationships in the advertising industry. For decades, s atisfaction has been the most important global construct to predict consumer behavior (Garbarino and Johnson 1999). However, along with the introduction of other constructs such as trust and commitment, the satisfaction construct was developed as one of the dimensions of the relationship quality construct (Garbarino & Johns on 1999; OdekerkenSchrder, D e Wulf, and Schumacher 2003; Palmatier, D ant and Evans 2006) Over the years studies concerning satisfaction have paid great attention to examine overall consumer satisfaction, including buyers satisfaction with
48 suppliers ( Lages, Lancastre, and Lages 2008) On the other hand, suppliers perception has been largely disregarded because interfirm relationships are mostly initiated by superiors ( i.e., buyers clients ) in the relationship (Holmlund 2008). However, agency staffers can be considered not only service providers but also employees of the client firm, considering the agency client relationship lies in a partner atmosphere, rather than vendorship (Huntley 2006). With this in mind, agency staffers satisfaction with the c lient would be critical because satisfied employees are likely to outperform those who are not satisfied and contribute to customers satisfaction (Brown and Lam 2008). In terms of outcomes of relationship satisfaction, much research has paid attention to tangible outcomes such as sales increase, market share, or profit (Crosby Evans, and Cowles 1990; Palmatier, D ant, and Evans 2006). For instance, Palmatier D ant, and Evans (2006) provided supportive evidence that relationship satisfaction is positively associated with sales, market share, and profit. On the other hand, many scholars have suggested that relational outcomes such as cooperation, long term orientation, or development of future relationship ( Ganesan 1994; Ro berts Varki, and Brodie 2003) Th e current study particularly defines relationship satisfaction as an affective status in the relationship and focuses on relational outcomes in contrast to tangible outcomes from the perspective of service providers (i.e., media planners) in the advertising industry. Cooperation Cooperation, in the business relationship context, is defined as coordinated actions taken by firms in interdependent relationships to achieve mutual outcomes or singular outcomes with expected reciprocation over time (Anderson and Narus 1990, p.45). Cooperative behavior in fact is a necessary condition for relationship (Morgan
49 and Hunt 1994). According to Chenet Tynan, and Money (1999), cooperation often takes the form of teamwork, which is one of the characteristics of partners hip ( Baker, Simpson, and Siguaw 1999; Mohr and Spekman 1994). The concept of cooperation is especially important in agency client relationships because agencys perceived partnership may result in positive atmosphere for better performance (Huntley 2006). In other words, agency staffers would perform better when they are willing to work together with the client as a partner than when they are treated as a vendor. Lusch and Brown (1996) and Hewett and Bearden ( 2001) elaborated as to the importance of inform ation exchange in increasing coopera tive norms in the context of buyer supplier relationships In a study of franchise relationships, Guiltinan, Rejab, and Rodgers (1980) also found that cooperation is related to perceived communication effectiveness, the degree of uncertainly reduction, and the extent to paticipative decision making. This suggest that communication between agencies and clients may improve agency s willingness to cooperate with the client by being perceived themselves as partners and, in tu rn, produce better outcomes. While there is general agreement as to the definition of cooperation there is continuing discussion over the role of the cooperation construct. Much research has viewed cooperation as an outcome of relational exchange (Anderso n, Hikansson, and Johanson 1994; Anderson and Narus 1990 ; Chenet, Tynan, and Money 1999; Dwyer Schurr, and Oh 1987; Lancastre and Lages 2006). For example, Anderson and Narus (1990) suggested that cooperation is one of the biggest outcomes of communicatio n between firms On the other hand, other scholars have suggested that cooperation may be one of the dimensions of relationship quality (Baker Simpson, and Siguaw 1999;
50 Holmlund 2007; Lages Lancastre, and Lages 2008; Mohr and Nevin 1990) T h us, this part icular study examines the role of cooperation in the context of agency client relationships Structural Aspects of Relationship Quality Main purpose of the present study is not verifying the structure of the relationship quality construct per se; however, it is worthwhile to understand dynamics of relationship quality in the context of agency client relationships It would facilitate understanding of how market intelligence gaps influence the perception of relationships. There may be several possible struct ures of the relationship quality construct. Relationship quality may be a global factor that collapses all the aspects of relationship quality. In this assumption, the individual dimensions of the relationship quality (i.e., trust, commitment, satisfaction and cooperation) do not work as dist inctive constructs (Figure 2 4 ). The second possibility is that the four dimensions of relationship quality (trust, commitment, satisfaction, and cooperation) are completely independent, suggesting each construct has unique aspect of the relationship quality (Figure 25 ). The third possibility is that relationship quality is a higher order construct manifested by the four dimensions ( De Wulf, OdekerkenSchrder, and Iacobucci (2001) ; Holmlund 2007; Lages Lancastre, and Lages 2008; Rauyreun and Miller 2007). Holmlund (2007) argued that relationship quality should be treated as perception, instead of an objective quality. In this perspective, she suggested relationship quality as a higher latent concept comprising of subdimensions that are different but related each other (Figure 26 ). The final possibility is that cooperation may be an output of trust mediated by commitment, as suggested by many scholars ( Garbarino and Johnson 1999; Moorman, Zaltman, and Deshpande 1992; Morgan and Hunt 1994) In addition, satisfaction is added as a
51 determinant of commitment in the modified independent factor structural model as suggested by Varona (1996) (Figure 2 7 ). By comparing proposed models, the current study attempts to understand underlying structure of relationship quality constructs in the context of agency client relationships. Accordingly, fifth research question was developed as follows; Research question 5: How are relationship quality constructs structured in the context o f agency client relationships? Summary In the business to business context, the notion of relationship marketing has grown enormously. H owever, advertising agency client relationships have received relatively little attention from business relationship s cholars and, particularly, from advertising scholars. Therefore, this study started from a brief discussion about the significance of the notion of relationship marketing in the agency client relationship and proposed communication problem as one of the most significant obstacles in building long term relationships between agencies and clients. Communication problem was operationalized as market intelligence gaps, discrepancies between perceived value of market intelligence and delivery of the market intell igence, from the perspective of service providers (i.e., media planners). Among many practical and academic issues with regard to relationship marketing, the current study focused on relationship quality. Four constructs were selected for inclusion in the study (trust, commitment, satisfaction, and cooperation) as important components of relationship quality. Three different structures of advertising agencies were discussed in terms of communication flow from clients to media planners and compared based on market intelligence gaps and effects of the gaps on perceived relationship quality. Furthermore, this study examined
52 underlying structure of the relationship quality construct I the context of agency client relationship. Summary of Research Questions and Hypotheses Research questions are presented based on literature review above; Research Question1: Do media planners consider certain types of market intelligence more important than other types market intelligence? Research Question 2: Are all the types of market intelligence delivered equally to, less than, or more than the value of the market intelligence, suggesting market intelligence gaps between the value of market intelligence and the delivery of the market intelligence? Research Question 3: Is the level of market intelligence gaps related to the level of overall relationship quality? Research Question 4: Are certain types of market intelligence items related to overall relationship quality than are other types of market intelligences? Research Que stion 5: How are relationship quality constructs structured in the context of agency client relationships? In addition, a set of hypotheses was proposed in order to 1) test whether the market intelligence gap is generated differently across three structures of advertising agencies and 2) whether agencies perceive the relationship with their clients differently based on the structure of advertising agencies as follows; Hypothesis 1: Three agency structures, full service agencies, inhouse agencies, and me dia planning/buying agencies, will differ in their perception of importance of each type of market intelligence. Hypothesis 2: Media planning/buying agencies will perceive that each type of market intelligence items will be provided to them more often than will fullservice agencies and inhouse agencies. Hypothesis 3: Market intelligence gaps will be significantly smaller for media planning/buying agencies than for full service agencies and inhouse agencies. Hypothesis 4: The level of overall perceived relationship quality will be higher for media planning/buying agencies than for full service agencies and inhouse agencies.
53 Hypothesis 5: The smaller the market intelligence gaps, (a) the greater the media planners trust in their clients, (b) the higher media planners perceived commitment to their clients, (c) the more satisfied media planners are with their clients, and (d) the greater media planners intention to cooperate with their clients. Figure 21. Information flow model for full service agencies Note. adopted from Sutherland, Duke, and Abernethy (2004) Figure 22. Proposed i nformation flow model for media planning/buying agencies Knowledge of market intelligence Client Gatekeeper Agency Gatekeeper Media planner External communication Interna l communication Knowledge of market intelligence Client Gatekeeper Media planner External communication
54 Figure 23. Proposed i nformation flow model for inh ouse agencies Figure 2 4 General relationship quality model Knowledge of market intelligence Client Gatekeeper Agency Gatekeeper Media planner In ternal communication Internal communication Relationship Quality V1 V2 V3 V4 V5 V6 V7 V8 V9 V1 0 V11 V12
55 Figure 2 5 I ndependent factor model of relationship constructs Figure 2 6 A secondorder hierarchical model of relationship constructs T 1 T 2 T 3 C1 C2 C3 S1 S2 S3 P1 P2 P3 Trust Commitment Satisfaction Cooperation T 1 T 2 T 3 C1 C2 C3 S1 S2 S3 P1 P2 P3 P4 Trust Commitment Satisfaction Cooperation Relationsh ip Quality Relationship Quality
56 Figure 27. T h e modified independent factor structural model ( Full model) Trust Satisfaction Commitment Cooperation
57 CHAPTER 3 METHOD The purpose of this study is to identify media market intelligence gaps in the advertising industry and examine their effects on media planners perceived relationships with their clients. A survey method was employed to obtain empirical data from media planners on the gap between importance and delivery of media market intellige nce, relationship quality, and demographics. Among many data collecting methods, this study specifically employed an online survey method because of its strengths such as flexibility, speed, convenience, and easy of data entry and analysis (Evans and Mathur 2005). Data analyses were performed using various statistical techniques such as independent samples T test, Analysis of Variance (ANOVA), Confirmatory Factor Analysis (CFA), Multiple Regression, and Structural Equation Modeling. This chapter describes t he method used in this study in the following order: (1) Participants and procedures (2) Instrumentation (3) Expert review (4) Gap analysis, and (5) Statistical tests. Participants and Procedures Media planners and media directors in the U.S. advertising f irms were invited to participate in a self administered online survey. The potential respondents were initially approached by a postcard that contained an invitation to the online survey and a link to a website on which the survey questionnaire was posted. Mailing lists were downloaded from the LexisNexis advertising agency database, called The Advertising Redbooks. The database covers close to 15,000 U.S. and international advertising agencies and provides agency profiles that include annual billing, accounts, number of employees, contact information for key personnel, and company email. In order to follow up the
58 postcard invitation, an email containing the survey link was sent to corporate email addresses asking to be forwarded to media planners and me dia directors in the firm. The postcard was sent to a total of 1,430 media planners and media directors with individual name on it. Of these 1,430 postcards, a total of 62 surveys were responded, for a response rate of 4.3%. Because the response rate was insuffi cient for the subsequent analyses, second round of data collection was conducted by snowball sampling. The snowball sampling is a technique that asks existing respondents to a survey to help recruit future respondents. This particular technique is useful when the focus of study is on personal matter or knowledge among a homogeneous group (Biernacki and Waldorf 1981). Hackathorn (1997) specifically explained that because snowball sampling relies on referral of acquaintances, it suits well for identify ing subjects that share the same traits and knowledgeable individuals in the field. Thus, snowball sampling can be used to gather data from experts. In order to count the exact number of responses from the snowball sampli ng, a different survey link with th e identical questionnaire from the first round of the survey was cr eated. Initially, a total of 20 media planners and directors were contacted by email with openended invitation and agreed to forward the survey link to their colleague media pl anners and directors. Within 20 days of snowballing, 71 responses were collected. As the result of the two rounds of data collection, total 133 responses were collected. However, not all particip ants completed the 58 Likert scale items in the survey 4 6 participant s stopped the survey at a point and left the remaining questions unanswered. It should be noted that out of 133 surveys collected, only 87 respondents completed all survey questions.
59 The first part of the online questionnaire included the informed consent form that contained of the purpose of the study, request of voluntary participation, confidentiality agreement, and the right to withdraw from the survey in accordance with the Code of Federal Regulation. Participants were asked to click the agree button if they agreed to participate in the survey. The data was stored on a server provided by an online survey company, Qualtrics.com, and the data was downloaded at the end of data collection. The first section of the survey questionnaire asked respondents t o indicate their perceived importance of each type of market intelligence to a better media planning. After completion of the first section, respondents were asked to think of one of the clients for which they have most recently completed a media plan. In order to ensure unbiased variance Koslow, Sasser, and Riordan (2003) have asked respondents to have last three campaigns in mind to fill out the questionnaire. However, as they admitted, it caused a complicated questionnaire and required a long time to com plete the survey. The current study, therefore, asked participants to have one specific client that they have recently finished a media plan. Having the client in mind, participants proceeded to the second section that indicates the frequency of the market intelligence delivery from the client. The importance and the actual delivery of the market intelligence items were computed for creating market intelligence gaps. The third section asked participants to rate their perceived communication quality with their clients. Finally, participants were asked to indicate their perceived relationship with the client. The perceived relationship was composed of four categories of items: trust, commitment, satisfaction, and cooperation. At the end of the survey, they wer e asked if
60 they wanted to receive the results of the study and write down email address to which the result can be sent. Instrumentation The questionnaire was comprised of four main parts: independent variables (market intelligence constructs and communic ation quality constructs), dependent variables (relationship quality constructs), control variables, and demographics. Items in each section were randomly placed in order to avoid response bias from order effect. This study used a sevenpoint scale for the entire set of questions, following Krosnick and Fabrigar (1997)s suggestion that sevenpoint scale is the best compromise between too less scales and too many scales. Coefficient alpha (Cronbach 1951) was used to measure internal consistency of the measures and reported for each set of measures. The instrumentation process involved item selection, modification, and expert review. Independent Variables The media market intelligence items were selected both from an analysis of media briefs from major advertising agencies such as Universal McCann, Digitas, and Ogilvy & Mather and from a study of Abratt and Cowan (1999). Initially, a set of 70 media market intelligence items were identified, and many of the items that were believed to represent the same information were collapsed and categorized. A total of 14 market intelligence items finally were chosen to be used for this study (Table 31) The importance of market intelligence and frequency of the market intelligence delivery were measured by a sevenpoin t Likert type scale, anchored at 1=not at all important and at 7=extremely important and at 1=never and at 7=very frequently, respectively. The importance scores and the frequency scores were transformed into an independent variable, gap scores.
61 As less ma rket intelligence gaps should lead to better perception of information exchange with the client, the researcher included information exchange scales (Lusch and Brown 1996) as an indicator of market intelligence gaps (Cronbachs alpha=.86) Also, perceived utility of the market in telligence was measured by the market intelligence use scale adopted from Maltz and Kohli ( 1996) (Chronbachs alpha=.7 7 ) The measurement items for communication variables are presented in Table 32. Because no measurement scale on communication between clients and agency staffers had been developed at the inception of our study, all the items were modified to suit the agency client relationship context. Dependent Variables Four r elationship quality constructs (trust, commitment, s atisfaction, and cooperation) were include d as dependent variables. Initially, f our items were borrowed from Andaleep (1996) and modified to measure trust of media planners on their clients. The items were originally adopted from the previous study conduct ed by Crosby, Evans, and Cowles (1990) to measure interpersonal influence on service selling. To measure affective commitment to the client, six items were chosen from Ganesan and Weitz (1996), which are originally adopted from Mowday, Steers, and Porter ( 1979) that used 15 sevenpoint Likert type response scales. Five items were adopted from Dwyer and Oh (1987) and modified to measure perceived satisfaction with the client. Finally, intention to cooperate was measured by 4 items that were borrowed from Crosby, Evans, and Cowles (1990). The items were reworded in order to reflect media planners intention to cooperate with the client. All the sevenpoint scales measuring dependent variables were anchored at 1=strongly disagree and at 7=strongly agree.
62 Looki ng for parsimonious measures, a n exploratory factor analysis (EFA) was conducted to validate internal consistency for scales for the relationship quality construct s and decide which items to retain and which to eliminate among the initial 19 items. The dec isions were made by based on the following statistical criteria (Tabachnick and Fidell 1996): (1) an average corrected item to total correlation below .35, (2) an average interitem correlation below .2, (3) factor loadings below .45, and (4) items with cro ssloading greater than .4 on more than one factor. Also, a standardized alpha for each factor with and without each suspect variable was calculated. If a coefficient alpha of a factor could be improved by eliminating any variable, then the item was removed and a new alpha score was computed. Table 33 displayed the results of the item purification, including factor loadings and Cronbachs alpha coefficients. It is proposed that each item should, for acceptable construct validity, have a minimum factor loading of ".60" for its hypothesized construct (Nunnally 1978). The factor loadings were acceptable, ranging from .68 for one of the commitment items to .90 for one of the items of satisfaction. All subscales showed good internal consistency with Cronbachs alpha coefficients ranging from .81 for trust scales to .96 for satisfaction scales. In order to ensure construct validity of the scales, a CFA was also conducted with the selected items. After allowing error of one of the items for commitment to be correla ted with cooperation measure, the goodness of fit for t he measurement model significantly improved and 2 test for the measurement model 2(47)=60.19, p=.0 9 ), suggesting that the measurement model does fit well with the observed data. Other goodness of fit mea sures were reported as follows ; RMSEA = .06, CFI = .9 9, SRMR = .04 NNFI = .9 5 Accordingly a total of 12 items for
63 relationship quality constructs 3 items per each construct were included in the study for t he subsequent analyses. Control V ariables Following suggestions of Mohr Fisher, and Nevin (1996), this study included three variables as covariates: relationship length, clients size, and prior hist ory of conflict. First, as Mohr Fisher, and Nevin (1996) explained, prior history of the relationship with the client may be related to the current perception about the relationship with the client. Thus, inclusion of the length of relationship may provide alternative explanation for findings from this study. Second, the client size was reflected by annual media expenditure of the client. Large clients may induce higher communication effort and thus enhance the relationship quality. They reported that only conflict measures had a significant and negative relat ionship with the relationship quality measures. F our items used by Mohr Fisher, and Nevin (1996) were used to measure existing conflict between agencies and clients which were originally adopted from Anderson and Narus (1990), Brown and Day (1981), and S tern and El Ansary (1992). Cronbachs alpha coefficient was the highest ( =.78) when three among the four items were included in the analysis The items included in the analysis are presented in Table 34. Demographics Items measuring demographics of participants were included in the study. The questions asked about gender, position in the company, and years in the advertising industry. Expert Review The initial questionnaire was reviewed by four scholars who have expertise on methodological issues on survey research. They agreed that participants, a cohort
64 group from the same industry, should be familiar with market intelligence items presented in the questionnaire. However, they pointed out that lengthy list of items may cause survey fatigue and thus increase nonresponse rate (Porter, Whitcomb, and Weitzer 2004). Given that media planners lack disposable time during work hours, shortening the list of market intelligence would facilitate increasing response rate. Accordingly, the items were collapsed and categorized as suggested by Abratt and Cowan (1999) and media briefs coll ected from the industry. Final ly, all market intelligence items were reviewed by seven media experts in the advertising industry in order to ensure both face and content validity. The reviewers agreed that the list of intelligence items were comprehensive; however, a couple of suggestions were made to improve the market intelligence items. As a result of the expert review, following items were added in the survey: criteria for measuring success of the campaign, mandatories (e.g., what plan must/must not inc lude, media to be used for the campaign), and company related background information (e.g., thought leadership articles, news releases, white papers, and corporate blogs). A total of 14 market intelligence items were finally included in the questionnaire. In addition, all the questions were closely investigated and rephrased as suggested by Graesser et al. (2000). They suggested 12 problems with regard to questions in a survey questionnaire: (1) complex syntax, (2) working memory overload, (3) vague or am biguous noun phrases, (4) unfamiliar technical terms, (5) vague or imprecise predicate or relative term, (6) misleading or incorrect presupposition, (7) unclear question category, (8) amalgamation of more than one question category, (9) mismatch between the question category and the answer options, (10) difficult to recall
65 information, (11) respondent unlikely to know answer, and (12) unclear question purpose. Finally, I split the 14 market intelligence items onto two separate pages in order to minimize num ber of items in one computer screen. As the results of the expert review and the revision, the survey questionnaire was ready for a main study. Gap Analysis Expectationperformance gap theory (Parasuraman, Zeithaml, and Berry 1985) suggests that discrepancy between consumers expectations about the performance and their assessment of the actual performance drives the perception of service quality. Research on expectation and performance has used mathematical approaches to explain perceptional gaps between quality and expectations (Cronin and Taylor 1992 ; Gronoos 1982; Oliver 1980, 1997; Parasuraman, Zeithaml, and Berry 1985). This dissertation operationalizes market intelligence gaps as discrepancy between the value of market intelligence and the communicat ion quality of the market intelligence. Hill and McCrory (1997) suggested that expectation and performance studies may infer expectations from importance ratings. That is, in the context of the present study, if a media planner perceives a type of market i ntelligence is important, the media planner should expect quality communication for that type of market intelligence. Mathematically, this premise is expressed as: market intelligence gap = Vm), where F is the perceived frequency of the communication, c, and V is the value of the market intelligence, m (Sutherland, Duke, and Abernethy 2004). A gap is simply a media planners perceived frequency of the communication (F) minus the value of market intelligence item (V). However, Sutherland, Duke, and Abernethy (2004) argued that the F V a pproach generates minus values del ivery of the information likely to hardly exceed the value of
66 the informationwhich make the interpretation of the data difficult. Therefore, in order to enhance the interpretation of the data the present study uses more common approach, the V F approach: Fc) (Parasuraman, Zeirhaml, and Berry 1985; Sutherland, Duke, and Abernethy 2004). With the V F approach, the communication quality of the market intelligence will be assessed to be equal to, better than, or worse than the perceived value of the market intelligence (Parasuraman, Zeirhaml, and Berry 1985). In this regards, plus values in the gap analysis indicate that communication quality has not met the importance of the market intelligence, and the larger degree of the discrepancy between V and F indicates that the lower communication quality than the value of the information. On the other hand, minus values in the gap analysis indicates that communication quality has exceeded the value of the information, which may imply that the information is overly or unnecessarily delivered. When the communication quality of market intelligence meet the value of the market intelligence, media planners may perceive their relationships with the client satisfactory, comm it to the client, and coordinate with the client (Brown and Swartz 1989). Statistical Tests In order to address research questions and the proposed hypotheses, a variety of statistical analyses was conducted, including both exploratory and confirmatory factor analyses, analysis of variance (ANOVA), multiple regression analysis, and structural equation modeling. Level of statistical significance used by this study was 0.5, which means the study takes the risk that a true null hypothesis would be rejected fi ve times out of 100 samples (Polit and Hungler 1999). To address research question 1 (Do media planners consider certain types of market intelligence more important than other types market intelligence?), 14 market
67 intelligence items were compared by a oneway repeated measures ANOVA. Also, gap scores were computed, and the gap scores for each type of market intelligence were compared by the same repeated measures ANOVA to address research question 2 (Are all the types of market intelligence delivered equal ly to, less than, or overly than the value of the market intelligence, suggesting market intelligence gaps between the value of market intelligence and the delivery of the market intelligence?). In addition, a series of factor analyses was conducted to 1) explore the dimensionality of the importance score and the gap score and to 2) control the effect of multicollinearity for the subsequent analyses. A multiple regression was conducted to address research question 3 (Is the level of market intelligence gaps related to the level of overall relationship quality?) and research question 4 (Are certain types of market intelligence items related to overall relationship quality than are other types of market intelligences?), by regressing the gap scores of each mar ket intelligence with the overall relationship quality score. Research question 5 (How are relationship quality constructs structured in the context of agency client relationships?) was addressed, using structural equation analysis by the method of maximum likelihood. LISREL 8.8 was used for performing data analyses. T o ensure construct validity of the measurement model of relationship constructs a confirmatory factor analysis was conducted. Second, several path analyses for different structural models wer e conducted to identify the structure of the latent variable, relationship quality, in the context of agency client relationships. The structural equation modeling demonstrated causal relationships among relationship quality constructs, as well as exogenous variables: market intelligence gap factors
68 Due to limited number of responses from inhouse agencies (n=5), the proposed hypotheses that compare among three types of agencies were needed to change to compare two types of agencies (full service vs. media agencies). Thus, hypotheses 1 through 4 were tested by using independent samples T test The first hypothesis (Three agency structures, full service agencies, inhouse agencies, and media pla nning/buying agencies, will differ in their perception of import ance of each type of market intelligence.) was tested by comparing media planners ratings of the importance of the market intelligence items across the two advertising agency structures. The second hypothesis (Media planning/buying agencies will perceive that each type of market intelligence items will be provided to them more often than will full service agencies and in house agencies.) was also tested by independent sample T test Delivery frequency scores for each type of market i ntelligence were compar ed between the two different agency structures. The third hypothesis (Market intelligence gaps will be significantly smaller for media planning/buying agencies than for full service agencies and inhouse agencies.) compared the gap score between the two ag ency structures. And, the fourth hypothesis (The level of overall perceived relationship quality will be higher for media planning/buying agencies than for full service agencies and inhouse agencies.) was addressed by comparing the overall relationship sc ore among the two structures of advertising agencies. Finally, the fifth hypothesis (The smaller the market intelligence gaps, (a) the greater the media planners trust in their clients, (b) the higher media planners perceived commitment to their clients, (c) the more satisfied media planners are with their clients, and (d) the greater media planners intention to cooperate with
69 their clients.) was tested by regressing the overall gap score of market intelligence with each relationship quality construct: t rust, commitment, satisfaction, and cooperation Table 31. Market intelligence items to be included in the study 1. Key marketing/communication assignment/challenges to be addressed. 2. Marketing objectives/goals 3. Budget for the campaign 4. Communication obje ctives/goals 5. Brand situation and strategies 6. Information about other marketing communication activities 7. Criteria for measuring success of the campaign 8. Target audience profile 9. Creative specification 10. Competitive brand situation and strategies 11. Ma n datori es 12. Current general market situation 13. Timing considerations 14. Company related background information Table 32. Measurement items for communication variables Information Exchange ( adopted from Lusch and Brown 1996) 1. The client provides us information frequently and informally 2. The client provides us with propriety information if it can help 3. This client keeps us informed about events or changes that may affect media planning Information U tility ( adopted from Maltz and Kohli 1996) 1. Information I received h elped shape the media plan 2. Information I received i mproved the quality of the media plan 3. Information I received i mproved my understanding of the market place
70 Table 33 Results of exploratory factor analysis (E FA) for measurement scales of relationship quality Factors and items (standardized) Trust (adopted from Andaleep 1996) .8 12 This client is not sincere about keeping its commitment [R] .8 5 This client cannot be counted on to be helpful [R] .8 3 This client is not very relia ble [R] .8 0 Commitment ( adopted from Ganesan and Weitz 1996) .8 65 Im proud to be part of this client. 69 I enjoy discussing this client with people outside it. .8 6 Im glad that I was chosen to work for this client. .6 8 Satisfaction ( adopted fr om Dwyer and Oh 1987) .9 60 Overall, this client is a good company to do business with. 87 Overall, this client treats me fairly. 90 All in all, my relationship with this client is very satisfactory. 86 Cooperation (adopted from Crosby, Evans, an d Cowles 1990) .87 1 Im willing to help this client to succeed in the marketplace even if it requires more time and effort. .8 5 Im willing to take the time to prepare formal proposals for this client to understand my media planning. 81 I have a desi re to develop a long term relationship with this client. 74 Note. [R] indicates reverse code. Table 34. Measurement items for existing conflict Conflict ( adopted from Mohr, Fisher and Nevin 1996) 1. We argue frequently with this client about busines s issues. 2. Our argument with this client are very heated 3. We disagree with this client ab out how we can best achieve our respective goals.
71 CHAPTER 4 RESULTS The purpose of this study is to identify media market intelligence gaps in the advertising indust ry and examine their effects on media planners perceived relationships with their clients. A survey method was employed to obtain empirical data on the gap between importance and delivery of media market intelligence, relationship quality, and demographic s. Among many data collecting methods, this study specifically employed an online survey method because of its strengths such as flexibility, speed, convenience, and easy of data entry and analysis (Evans and Mathur 2005). Data analyses were performed using various statistical techniques such as e xploratory factor a nalysis (EFA), independent samples T test, Analysis of Variance (ANOVA), confirmatory factor analysis (CFA), multiple regression analysis and covariance structural analysis Description of Sampl e s A total of 133 surveys were received, and 87 respondents completed the survey. As shown in Table 41 among the 87 respondents who completed the survey 5 8 (66. 7 %) were female and 2 9 (33. 3 %) were male. In terms of job title, it demonstrated a greater participation from media directors (N=38 44. 2%). A total of 73 respondents (85.0%) were media directors, media managers, or media planners, suggesting majority of the respondents were experts of media planning practic es. Other job titles (N=13, 15.0%) inclu ded account executives, campaign managers, strategic planners, or CEOs of advertising agencies, which demonstrate their capability of understanding media planning practices. In addition, years of their experience in the advertising industry were various fr om minimum 1 year to maximum 43 years. The average year of experience
72 was 14.4 years ( SD= 11.89 N =82 ), demonstrating that th ey are capable enough to participate in an advertising professional survey. Among the 88 respondents who provided the type of agency for which they work number of employees in full service agencies (N= 44, 50.0%) and media planning/buying agencies (N=35 39. 8 %) were closely matched. However, employees in inhouse agencies (N=5, 5.7 %) were not enough for separate analyses. Thus, inhous e agencies will be analyzed and discussed along with full service agencies in that most inhouse agencies provide with full services. Also, product categories for which the respondents have recently completed media plans appeared to be various, and demonst rated that the product category is not skewed toward any specific category. Descriptive Statistics Means and standard deviations were reported for the purpose of descriptive statistics because all the measurements were continuously measured (=numeric). A ll the means and standard deviations are shown in Table 4 2 for communication variables, Table 43 for relationship quality variables, and Table 44 for control variables. Means and standard deviations for market intelligence items were ranordered and presented in the section of research questions. As previously stated, sevenpoint Likert scales ranging from (1) strongly disagree to (7) strongly agree were used in all measures. Communication Variables Descriptive statistics for communication are presented in Table 42. The means of the information exchange items was 5.10 (SD=1.35, N=92). Among the items for information exchange, the client provides us with propriety information if it can help had the highest mean (M = 5.41 SD = 1.42) and the item the client keeps us informed
73 about events or changes that may affect media planning had the lowest mean (M = 4.91 SD = 1. 57). The utility of marke t intelligence items showed mean score of 5.4 1 (SD=1.06, N=92) with the highest score for the item information I received from this client h elped shape the media plan (M=5.78, SD=1.14) Relationship Quality Variables The mean score for the total relationship quality was 6.04 (SD=0.7 6, N=88) Among the four relationship quality measures, the respondents rated coo peration highest (M=6.42 SD=0.6 6) and trust lowest (M=5.81 SD=1.2 6 ). The respondents moderately agreed that they are trusting in (M=5.81 SD=1.2 6 ), committed to (M=5.97, SD= 1 11), satisfied with (M=5.85 SD=1. 21), and willing to cooperate with (M=6.42 S D=0.6 6 ) the client. A s shown in Table 43, among the trust items, the item this client is not very reliable (reversed) was rated highest (M=6.0 2 SD=1.2 8 ), and the item this client can be counted on to be helpful (reversed) was rated lowest (M=5. 59, S D=1. 59). Among the commitment items, the item Im glad that I was chosen to work for this client was rated higher (M=6. 18, SD= 1.21) than other items. The lowest item among the commitment items was I enjoy discussing this client with people outside it (M=5. 63, SD=1. 53). In term of satisfaction, mean scores among the items were close, ranging from 5.83 to 5. 89. Among the items, the item Overall, this client is a good company to work with was rated highest (M=5. 89, SD=1.2 7 ). The items of cooperation were also close amon g each others, ranging from 6.40 to 6.44 Control Variables The current study considered three control v ariables: existing conflict, years of the relationship, and clients advertising expenditure. Table 44 demonstrates means and
74 standar d deviations of the conflict measurement items. The conflict measures demonstrated that the respondents felt low level of conflict with client (M=2.13, SD=1.05). Among the conflict items, the item Our argument with this client are very heated was rated l owest (M=1.53, SD=0.94). Average year of the relationship was 4. 4 years (SD=5.00), ranging from 3 months to 25 years. Average advertising expenditure of the clients was 17.6 million dollar ( SD=29.85) with 150 million dollar s indicated as being the highes t and 30,000 dollar s as the lowest. Before discussing the findings for the research questions and specific hypotheses, the effects of control variables need to be mentioned. Controlling for the covariates increases internal validity providing confidence that changes in dependent variables are caused by the independent variable. In terms of covariates, existing conflict showed a strong, negative relationship with the overall relationship quality as was expected. On the other hands, contrary to Mohr, Fisher, and Nevin (1996) s findings, the length of relationship was strongly, positively related to the overall relationship quality. Clients size, measured by annual billing of the client, was not significantly related to the relationship outcome. T he inclusion of covariates d id not affect the sign or magnitude of the coefficient of the independent variable, overall gap scores on the dependent variable, overall relationship quality. Research Question 1: Importance of Market Intelligence Items To address research question 1 (Do media planners consider certain types of market intelligence more important than other types of market intelligence?), oneway repeated measure ANOVA was used on the importance scores of market intelligence items. Mauchlys test indicated that the assumption of sphericity had been violated,
75 2 (90)=317. 14 p <.05; therefore, degrees of freedom were corrected using GreenhouseGe isser estimates of sphericity ( =.71 ) as suggested by Field (2009) The results show that there is significant dif ference in perceived importance of market intelligence items, F( 9.21 1,068. 74)= 47. 47, p <.05 n =117 These results suggested that certain market intelligence items were significantly perceived as more important than other mark et intelligence items. Table 45 shows the summary statistics, including means, standard deviations and 95% confidence interval for each market intelligence item. The highest rated item was information about marketing challenges (M=6.66, SD=.75, n =117), whereas company background information was the least rated item (M=4.72, SD=1.33, n =117). In order to explore the underlying structure of the subjective evaluation o f market intelligence items, 14 items were factor analyzed using principal component analysis with Varimax rotation. Kais er M eyer Olkin (K MO) measure of .849 indicated a high sampling adequacy for the factor analysis. Bartlett's test of sphericity, which tests whether the correlation matrix is an identity matrix, is significant; 2(91)=798.28 ( p <.000). This indicates that the set of variables are adequately related for factor analysis. Table 46 displayed factor loadings, communality, Eigen values, means, and standard deviations. The number of factors extracted was three, and the three factor solution all together explained 61.07% of the total variance. As suggested by Tabachnick and Fidell ( 1996 ), items with cross loading greater than 0 .4 on more than one factor were removed from the factor analysis and a series of factor analyses were conducted until no items were heavily cross loaded on more than one factor The results of Varimax rotation produced threefactor solutions with 9 items that are clearly loaded
76 on each of the three factors The three factor solution explained 67.21% of the total variance. Retained factors are: Factor I Information about objective s & situation : marketing objectives, brand situation, competitive situation, general market situation, company rela ted, and background information; Factor 2 Information abo ut a ccountability: budget information and criteria measuring success of campaign ; Factor 3. Information about marketers concerns : target profile and key marketing challenges The result of a repeated measures ANOVA suggested that those three factors are significantly dif ferent from each other, F(2,231)=10 2.97 p <.05. Information about marketers concerns (M=6. 5 8, SD=. 67 N=117) was significantly perceived as the most important intelligence, followed by accountability related information (M=6.38, SD=.7 9 ). I nformation about objectives & situation (M=5.54 SD=.89) was perceived relatively less important than other factors. The results indicated that media planners do perceive certain types of market intelligence as more important than other types of intelligence. Research Question 2: Delivery and Gap Scores of Market Intelligence Items Re search question 2 (Are all the types of market intelligence delivered equally to, less than, or overly than the value of the market intelligence, suggesting market intelligence gaps between the value of market intelligence and the delivery of the market intelligence?) was addressed by oneway repeated measure ANOVA for perceived frequency of the market intelligence items and in turn by a gap analysis between the importance scores and the delivery scores. The results of oneway repeated measure ANOVA demonstrated that there is significant difference in perceived delivery of m arket intelligence items, F(8. 86, 842.11)= 2 2 .1 4 p <.05, after correcting degree of freedom by GreenhouseGeisser
77 estimates of sphericity. The result suggested that certain market intelligence items were significantly more frequently delivered than other market intelligence items. Table 47 shows the summary statistics, including means, standard deviations, and 95% confidence interval for each market intelligence item based on perceived frequency of information delivery. The most frequently delivered information item s were budget information (M=5.93, SD=1.31, N=96) and timing considerations (M=5.92, SD=1.40). And, the least frequently delivered items were competitive situation (M=4.36, SD=1.62), general market situation (M=4.31, SD=1.76), and company related background information (M=4.10, SD=1.79) T he results of a gap analysis were reported in Table 48 The gap score, as discussed above, was computed as F c), where Vm is perceived value of market intelligence item and F c is perceived frequency of the intelligence delivery All the items appeared to have positive gap scores, indicat ing that all types of market information are delivered less than the value of the information, as perceived by media planners According to the repeated measures ANOVA for gap scores, there is significa nt differences among the gap score across market intelligence items; F(8.88, 843. 48)= 9.84 p <.05. The biggest gap was reported on criteria for measuring success of the campaign (Gap score = 1 92 ) whereas the smallest gap was found in the budget information item (Gap score = 0.40) and timing consideration (Ga p score=0.29). It sug gested that media planners do not receive information about criteria measuring success of the campaign as frequently as the perceived importance of the information, but do receive budget information and timing consideration as frequently as the value of th e information.
78 To control for the effects of multicollinearity and identify the underlying structure of the gap scores, factor analysis was conducted on the gap score. The resulting factors were utilized as independent variables in a multiple regression whose dependent variables are overall relationship quality and the measured indicators of the relationship quality: trust, commitment, sati sfaction, and cooperation. Kaiser Meyer Olkin (KMO) measure of .82 indicated a high sampling adequacy for the factor analysis. Bartlett's test of sphericity, which tests whether the correlation matrix is an identity matrix, is significant; 2 (66 )= 448 44 ( p <.000). This indicated that the factor model is appropriate. The results of t he factor analysis suggested a threefactor solution. The three factor s all together explained 62 9 2 % of the total variance. After Varimax rotation, two variables were left ambiguous with cross loadings; thus, the two variables (communiation objectives and target profile) were excluded from the factor model. The resulting factors are shown in Table 49. Factors retained are: Factor 1. Information about objectives & situation : company related information, brand strategies competiti ve strategies, general market situation, other marketing communication activities, creative specification; Factor 2. Information about m arket ers concerns : marketing objectives, key mark eting challenges, criteria for measuring success of the campaign Factor 3. Information about restriction : budget information, timing consideration, mandatories A repeated measures ANOVA revealed that the three gap factors are significantly different among each other ; F(2,190)=24.62 ( p<.001). Information about marketer s concerns and accountability showed the biggest gap score (M=1.55, SD=1.38, N=96), followed by information about objectives & situation ( M=1.01, SD=1.33) Information about restriction, on t he other hands, demonstrated the smallest gap score ( M=0.52, SD=1.17) indicating that restriction related information is given as frequently as the
79 importance of the type of market intelligence. The results demonstrated that media planners think that cert ain types of market intelligence are delivered more frequently than other type of market intelligence. Research Questions 3 & 4: Relationship between Gap S c ores and Overall Relationship Q ual i ty A stepwise m ultiple regression analysis was conducted to address both research question 3 (Is the level of market intelligence gaps related to the level of overall relationship quality?) and research question 4 (Are certain types of market intelligence items related to overall relationship quality than are other types of market intelligences?). The analysis specifically examined the relationship between market intelligence gaps and the over all relationship quality score: a combination of all the relationship quality constructs. Market intelligence gap factors are util ized as multiple independent variables, and the relationship quality score was used as a dependent variable. The effect of the level of intelligence gap on the overall relationship quality was tested using a linear regression analysis. The regression of overall gap score on the overall relationship quality showed that gap score significantly negatively affects t he overall relationship quality; t he smaller the gap score, the higher the relationship quality ( t = 3.5 0 p <.05). 1 2.4 % of the variability of rela tion ship quality was explained by the single independent variable (i.e., market intelligence gap) T he results of stepwise multiple regression analysis showed contributions of individual market intelligence item to the overall model fit ( Table 410 ) The two variables added to the equation made substantial contributions to the overall model fit with substantive increase in the R2 and adjusted R2 while decreasing the standard error of the estimat e. With only the first variable (informat ion about market challenges),
80 2 3. 7 % of the var iation in relationship quality wa s explained. A dditional variable ( company related background information) was added to reach the final model, but this variable, although statistically significant, made much smaller contribution. The stepwise procedure highlig hted the importance of the t wo variables in assessing overall model fit. The rest of the market intelligence items did not make contribution to the overall model. Table 41 1 showed the relative impact of each variable included in the regression model. M o del 2 showed that the two variables are negatively related to overall relationship quality score ( = .43, p <.05 for the market challenge item; = .2 5 p <.05 for the company background information item) indicating that the respo ndents with lower gap scores on these variables were expected to have higher scores in perceived relationship quality In addition, in order to explore the effects of underlying structure of the market intelligence gap scores on overall relationship quali ty, the market intelligence gap factors were regressed with the overall relationship quali ty score (Table 41 2 ). The multip le regression model explained 1 6 9 % of the total variance in relationship quality. Among the three gap factors identified the first factor, information about objectives & situation ( t = 2.14, p <.05) and second factor information about marketer s concerns ( t = 2.14, p <.05) w ere significant in their effect s on the relationship quality and showed similar impact in their magnitudes ( = .2 5 for the first factor; = .25 for the second factor) The two gap factors w ere negatively associated with overall relationship quality suggesting that those with lower scores on th e se gap factor s tend to have hig h er scores in perceived relationship quality On the other hands, the third gap factor (i.e., information about restrictions) was not significant in predicting relationship quality.
81 Hypotheses Tests H ypotheses 1, 2, 3, and 4 were proposed to test differences among the three structures of advertising agencies. However, due to limited number of respondents from in house agencies (n=5), responses of inhouse agencies were combined with those of fullservice agencies. A series of independent samples T test s was used to compare 1) importance scores, 2) delivery scores, 3) gap scores, and 4) overall relationship quality between the two types of advertising agencies: full service agencies (including in house agencies ) and media planning/buying agencies. The fifth hypothesis was tested by regressing the overall gap score of market intelligence with each of the relationship quality constructs: trust, commitment, satisfaction, and cooperation. Hypothesis 1 As stated, due to the insufficient number of in house agency data, the hypothesis was modified as follows ; t wo a gency structures (i.e., fullservice agencies including in house agencies and media planning/buying agencies ) will differ in their perception of importance of each type of market intelligence. A series of independent samples T test s was conducted to test the modified hypothesis The analyses compared media planners ratings of the importance of each market intelligence item between the two independent samples. The results of T tests are presented in Table 413. As shown in Table 41 3 perceived import ance for market intelligence items is not differ between the t wo types of agencies (Mfull=6.0 vs. Mmedia=5.9, t ( 82 ) = .72 p = 47) Also, n one of the individual market intelligence item showed significant difference between the two agency structures, indi cating that media planners do not perceive the importance of market intelligence items differently depending on the type of agencies in which they work. This comparison did not support hyp othesis 1.
82 Hypothesis 2 The second hypothesis (Media planning/buying agencies will perceive that each type of market intelligence items will be provided to them more often than will full service agencies and inhouse agencies.) was also tested by the same independent samples T test The delivery frequency score for each t ype of market intelligence was compared across the two different agency structures. As depicted in Table 41 4 the mean difference for the perceived delivery of market intelligence between the two agency types was 0.46 and was significant (Mfull=5.12 vs. Mmedia=4.66, t ( 82) = 2.06, p < 05) By individual item, three market intelligence items were significantly different between full service and media agencies at the 95% confidence level : information about other marketing communication activities (Mfull=4. 84 vs. Mmedia=4.03 ) general market situation (Mfull=4. 76 vs. Mmedia=3. 49) and company related background (Mfull=4.4 5 vs. Mmedia=3. 46) Although significant differences were found, the result was against the proposed hypothesis in that fullservice agencies rated higher on these perceived delivery items of market intelligence than media planning/buying agencies did. Also, it should be noted that although statistically insignificant 12 out of the 14 information delivery items appeared to be higher for full ser vice agencies than for media planning/buying agencies. Thus, hypothesis 2 was not supported, suggesting alternative interpretation of the result T he implication of this finding will be discussed in the discussion section. Hypothesis 3 The gap scores of fullservice and in house (M= .8 8 SD= .9 4, N=49 ) and media planning/buying (M= 1.24 ,SD= 1.1 5, N=35 ) agencies were compared to test the hypothesis that market intelligence gaps will be significantly smaller for media
83 planning/buying agencies than for full serv ice agencies and inhouse agencies. The result of independent samples T test showed that overall gap score between full service and media agencies was not significantly different ( t (82) = 1.59, p =.12). As shown in Table 41 5 o nly two market intelligence it ems appeared to be significantly different between the two types of agencies at the 0.5 level of significance: information about other marketing communication activities (Mfull=1.12 vs. Mmedia=1.91) and general market situation (Mfull=.90 vs. Mmedia=1.83) Although statistically insignificant, perceived gap scores for 12 scores out of the 14 gap scores were smaller for full service agencies than for media agencies. That is, media planners in full service agencies may perceive less market intelligence gaps than those in media agencies. This result is against the proposed hypothesis. Hypothesis 4 The fourth hypothesis was proposed to test differences in the perception of media planners relationship with their clients between full service agencies and media agencies ( i.e., two agency structures will significantly differ in their perception of relationship with their clients) As shown in Table 41 6 the overall relationship quality differs by the type of agencies ( Mfull=6. 22 vs. Mmedia=5. 83, t =2. 43 p <.0 5 ). By each construct, all the relationship quality constructs, with exception of commitment, were significantly different between the two types of agencies: trust (Mfull=6. 1 0 vs. Mmedia=5. 46, t =2.5 0 p <.05) ; satisfaction (Mfull=6.1 4 vs. Mmedia=5. 69 t =1.9 8 p =.0 5 ) ; cooperation (Mfull=6.56 vs. Mmedia=6.28, t =2.00, p =.05). The results indicated that respondents in full service agencies rated higher on all the relationship quality constructs although insignificant for commitment than those in media agencies. That is, media planners in full service agencies may perceive their
84 relationship with the client as better than those in media agencies do This finding is consistent with the r esults of hypotheses 2 and 3 that media planners in full service agencies showed higher perceived delivery of market information and less market intelligence gaps than those in media agencies. Hypothesis 5 The fifth hypothesis (The smaller the market intelligence gaps, (a) the greater the media planners trust in their clients, (b) t he higher media planners perceived commitment to their clients, (c) the more satisfied media planners are with their clients, and (d) the greater media planners intention to cooperate with their clients.) was tested by regressing the overall market intel ligence gap score with each construct of the relationship quality. The gap score was utilized as a single predictor, and each of the relationship quality constructs w as utilized as dependent variable. The summary table of the regression analyses of market intelligence gaps with the relationship quality constructs is depicted in Table 41 7 The overall market intelligence gap score is significantly related to all the relationship quality constructs with the exception of cooperation: trust (F(1, 88)=7 .2 9 p < .0 5 ), commitment (F(1, 8 8)=10. 13, p <.0 5 ) and satisfaction (F(1, 8 8)=13. 20, p <. 05 ) The three significant ly related constructs were negatively associated with the overall gap score, meaning that levels of trust .28 ) commitment .32 ) and satisfaction .36 ) will be decreas ed as market intelligence gap increases H ypotheses a), b), and c) were supported, but hypothesis d) was not supported. Research Question 5: The Structure of Relationship Quality Constructs Finally, the fifth research question (How are relationship quality constructs structured in the context of agency client relationships?) was addressed by confirmatory
85 factor analysis (CFA) and covariance structure analysis As proposed in the struc tural aspects of relationship quality section, relationship quality construc ts may have different structure in different context. The proposed structural models were 1) general relationship quality model, 2) independent factor model, 3) a second order hier archical model, and 4) mediated structural model. Before testing the proposed models, in accordance with the twostep procedure (Anderson and Gerbing 1988 ; Kline 2005) confirmatory factor analyses (CFA) were first of all, performed to assess construct va lidity of the multiitem construct measures and determine inclusion of a measurement model in to the structural model In the second stage of the twostep modeling, only CFA models that fit the data in the first stage included into the structural models (Kl ine 1995). Measurement Model Three separate confirmatory factor analyses were performed for the first, second, and third possible measurement model using LISREL 8.8 The fourth model shares the assumption of complete independence for each dimension with the second model; therefore, no separate CFA was needed for the mediated structural model. Goodness of fit indices used to evaluate overall fit of the CFA models were the comparative fit indices (CFI) in conjunction with the standard root mean squared resid ual (SRMR) as sugg ested by Hu and Bentler (1999). According to Hu and Bentler (1999), a cutoff value is recommended as .95 or higher for CFI in combination with a cutoff value close to or preferably smaller than .09 for SRMR. Additionally, following a sugg estion from Brown and Cudeck (1992), the r oot m ean square e rror of a pproximation (RMSEA), which is thought to reduce problems with incremental fit indices (e.g., CFI) and absolute fit indices (e.g., GFI), was also used. Hu and Bentler (1999) suggested RMSE A values of smaller than .06 as an indication of good fit.
86 Furthermore, the non normed f it i ndex (NNFI), which penalize for adding parameters, was also used (Bollen and Long 1993). A cutoff value for NNFI is recommended as .95 or higher. Model fit informat ion for the three proposed models is provided in Table 418. The general relationship quality model that specified a global relationship quality on which loaded all 12 items showed inadequate fit ( 2(54 )= 309.29, p=.0 00, RMSEA = .2 6, CFI = 80, SRMR = .15 NNFI = .7 5 ). Figure 41 depicted factor loadings for the model. 7 items out of 12 relationship quality items did not meet the minimum criterion of factor loading (i.e., .60, Nunnally 1978). The in dependent factor model that set the relationship quality dimensions to be completely uncorrelated with each other fitted to the data adequately ( 2(48 )= 76.14 p=.0 1, RMSEA = .08 CFI = .98, SRMR = .05, NNFI = .97). 2 test was signif 2 / df ratio was 1.57 which is lower than suggested threshold of 3.0 (Kiline 2005). Figure 42 showed factor loading of the 12 items for relationship quality All items exceeded the minimum criterion for factor loading, ranging from .70 for one of the items of commitment to .96 for one of the items of satisfaction. This specific model was confirmed to be included in the structural model. The secondorder hierarchical factor model w as tested by loading for the first order fac tors (i.e., trust, commitment, satisfaction, and cooperation) on the second order factor (relationship quality). With the exception of Trust ( ), all standardized loadings exceeded 6 0 suggested minimum factor loading, ranging from .74 for cooperation to .89 for commitment (Figure 4 3) T he goodness of fit indices suggested that the secondorder hierarchical model can be preceded to the s tructural model 0 )= 87. 50, p=.00, RMSEA = .09, CFI = .9 7 SRMR = .0 8 NNFI = .9 6 ).
87 Between the two measurement models that fitted to the data adequately, the independent factor measurement model was selected to proceed to the structural analysis. Altho ugh the result of CFA for the secondorder hierarchical model confirmed that relationship quality is a higher order concept manifested by four relationship constructs, not much implication about relationships among the four constructs was found. Therefore, i n an attempt to explore relationships among the relationship quality constructs, the independent factor measurement model was selected to test the proposed modified independent structural model (see Figure 27) Structural Model for the Modified Independent Factor Structural Model The modified independent structural model was tested using LISREL 8.8. Before testing the proposed model, i t should be noted that with such a small number of samples the result of this covariance structural analysis may be unstable (For n ell 1983). Therefore, the result should be treated as a preliminary examination. Table 419 showed correlation matrix for the proposed model. The result of testing the full model that tested the effects of the predictor (i.e., market intelligence g ap) on t h e criterion variables (i.e., trust, commitment, satisfaction, and cooperat i on) was depicted in Figure 44 The modified independent factor structural model showed adequate fit to the data ( X2 (57) = 77.60 ( p =.04 ) RMSEA = 06, CFI = 97, SRMR = 0 5 NNFI = .97 ) The purpose of structural equation modeling is to develop a parsimonious model with fewer estimated parameters ; therefore, the reduced model was also tested. The result of the test was depicted in Figure 45. The reduced model also fitted t o the data adequately ( X2 (61) = 82.24 ( p =.04) RMSEA = 06, CFI = 97 SRMR = 06, NNFI = .97 ) The X2 comparison between the two model s was performed The 2difference between the two models was 4.64 ( 2 reduced 2 full = 82.24 77.60 ), and the degrees o f
8 8 freedom were df = dfreduced dffull= 61 57 = 4 Since the critical value of 2.05, 4 is 9.49 ( 2 difference < 2 critical), the reduced model was confirmed that it achieved the parsimony without hurting goodness of fit. Thus, the reduced model was select ed to interpret The effect of the market intelligence on each of relationship quality factors was summarized in Table 4 20. According to t he reduced independent factor model trust (standardized = 29, z = 2.52) was directly affected by the market int elligence gap. Also, satisfaction was affected directly and indirectly by the market intelligence gap. The total effect of market intelligence gaps on satisfaction was .37 (z = 3.60) and the effect was primarily due to the direct effect of .25 (z = 2. 47). Commitment was indirectly influenced by the market intelligence gap through trust and in turn satisfaction. The total effect of the market intelligence gap on commitment was .28 (z = 3.36). In terms of cooperation, the market intelligence gap indire ctly influenced the intention to cooperate, with a total effect of .20 (z= 3.12). The findings suggested that all four dimensions of relationship quality are negatively affected by the market intelligence gap directly and/or indirectly indicating the smaller market intelligence gaps, the better relationship quality media planners perceive. The model specified that cooperation is directly affected by commitment and indirectly by trust and satisfaction. The total effect s of trust and satisfaction on cooperation were .23 (z = 3.12) and .55 (z = 5.77), respectively. Specifically, satisfaction directly influenced on commitment and indirectly on cooperation, with relatively large standardizes coefficient (.75 and .55, respectively) Although all relationship constructs in this study showed significant impact on the cooperation factor, a direct effect on the intention to cooperate was by commitment (.73). Unlike Morgan and Hunts (1994)
89 findings, the result suggested that trust may have no direct effect on commitm ent or cooperation; thus, the role of trust in the agency client context may need alternative expl anation. Also, satisfaction appeared as one of the important mediator between the market intelligence gap and commitment and, in turn, cooperation. Alternativ e explanation is discussed in the discussion section.
90 T able 41. Respondents profiles Demographics N % Gender Male 2 9 33. 3 Female 5 8 66. 7 Total 87 100.0 Job t itle Media director 3 8 44. 2 Media manager Media planner Media buyer Others 14 20 1 13 16.3 23.3 1.2 1 5 1 Total 86 100.0 Agency t ype Full service In house Media planning/buying 44 5 35 50 0 5.7 39 8 Others 4 4 5 Total 80 100.0 Product category Consumer pack age goods 8 7.8 Automotives 4 3.9 Consumer durables (excl. autos) 3 2.9 Retail stores 14 13.6 Restaurants/Food services 8 7.8 Financial services/Banking 10 9.7 Other services 11 10.7 Business to business 6 5.8 Tele comm./Technology 2 1.9 Other 37 35.9 Total 103 100.0 N Min Max Mean SD Years in the industry 82 1 43 1 4.4 11. 55
91 Table 42. Descriptive statistics for communication variables Variables: Communication (N=92 ) M SD Informat ion Exchange The client provides us information frequently and informally The client provides us with propriety information if it can help 5.10 4.97 5.4 1 1.35 1.57 1. 42 This client keeps us informed about events or changes that may af fect media planning. 4. 91 1.5 7 Information Utility (Information I received) 5.4 1 1.06 Helped shape the media plan 5.7 8 1.1 4 Improved the quality of the media plan 5.4 2 1. 29 Improved my understanding of the market place 5.0 3 1. 39 T able 43. Descriptive statistics for relationship quality variables Variables: Relationship Quality M SD Trust (N=90) This client is not sincere about keeping its commitment [R] 5.81 5.8 2 1.26 1.53 This client cannot be counted on to be hel pful [R] 5. 59 1.5 9 This client is not very reliable [R] 6.0 2 1.2 8 Commitment (N=90) 5.97 1 11 Im proud to be part of this client 6. 09 1.0 4 I enjoy discussing this client with people outside it 5.6 3 1.5 3 Im glad that I was chosen to work for this client 6. 18 1. 21 Satisfaction (N=89) 5.8 5 1. 21 Overall, this client is a good company to work with 5. 89 1.2 7 Overall, this client treats me fairly 5.8 3 1.2 1 All in all, my relationship with this client is very satisfacto ry 5.8 4 1.2 8 Cooperation (N=88) 6.4 2 0.6 6 Im willing to help this client to succeed in the market place even if it requires me more time and effort 6. 42 0.7 5 Im willing to take the time to prepare formal proposal for this client t o understand my media planning 6.4 4 0.7 1 I have a desire to develop a long term relationship with this client 6. 40 0.8 8 Note. Note. [R] indicates reverse code.
92 Table 44. Descriptive statistics for control variable: existing conflict Control Variables (N=92 ) M SD Conflict We argue frequently with this client about business issues Our argument with this client are very heated 2.13 2.36 1.5 3 1.05 1.39 0.94 We disagree with this client ab out how we can best achieve our r es pective goals 2. 51 1.3 9 N Min Max Mean SD Years in relationship 86 0.3 25 4. 39 5. 00 N Min Max Mean SD Clients billing ($MM) 65 0.03 150 1 7 6 2 9 85 Table 45 Summary statistics of the perceived importance of market intelligence item M 95 % Confidence Interval Market intelligence items (N= 117) SD Lower Upper Key marketing challenge s 6.6 6 0 .7 5 6.5 2 6.79 T arget profile 6. 50 0 .8 8 6.3 4 6.6 6 B udget information 6.3 9 0 .9 6 6.21 6.5 6 C riteria measuring success of the campaign 6.3 7 0 .8 7 6.2 1 6. 5 3 C omm unication obj ectives 6.3 6 0 .8 4 6.2 1 6.51 T iming consideration 6.21 1.03 6.0 3 6.40 Marketing objectives 6.2 1 1.0 4 6.0 2 6. 40 M andatories 6.1 3 1.22 5.9 1 6.35 Other m arcomm activities 5.9 5 0 .97 5.77 6.1 3 Creativity specification 5.8 3 1.0 8 5.63 6.0 3 Brand situation 5.6 8 1.0 6 5.48 5.8 7 G eneral m arket sit uations 5.59 1.18 5.37 5.8 1 C ompetitive situation 5.4 9 1.08 5.2 9 5.6 9 Company related background information 4.71 1.3 3 4.4 8 4.96
93 Table 46 Principle component factor analysis of the im portance scores (N=117) Market intelligence items F1 F2 F3 Com m u nality Factor 1. Info. about objectives and situation Company related background information .76 .61 General market situation .71 .72 Competitive situation .80 77 Brand situation .76 .67 Marketing objectives .61 .57 Factor 2. Info. about imminent concerns Criteria measuring success of the campaign .72 .62 Timing consideration .65 .64 Budget information .86 .79 Factor 3. Info. about accountability Target profile .78 .70 Key marketing challenges .77 .64 Eigen Value 4.56 1.12 1.04 % of Variance 45.60 11.21 10.40 Cumulative % 45.60 56.81 67.21 Mean (SD) 5.54 (.89) 6.38 (.7 9 ) 6.58 (.67) T able 47 Summary statistics of the perceived delivery of market intelligence item M 95% Confidence Interval Market intelligence items (N=96) SD Lower Upper B udget information 5.93 1.31 5.66 6.19 Timing consideration 5.92 1.40 5.63 6.20 Key marketing c hallenge s 5.43 1.30 5.16 5.69 T arget profile 5.41 1.58 5.09 5.73 C omm unication ob jectives 5.31 1.41 5.03 5.60 M andatories 5.22 1.62 4.89 5.55 Creativity specification 4.92 1.61 4.59 5.24 B rand s ituation 4.79 1.59 4.47 5.11 Mar k eting obj ectives 4.76 1.72 4.41 5.11 Other marcomm activities 4.56 1.70 4.22 4.91 C riteria measuring success of the campaign 4.45 1.76 4.09 4.80 C ompetitive situation 4.36 1.62 4.04 4.69 G eneral m ar k et sit uation 4.31 1.76 3.96 4.67 C ompany related background information 4 .10 1.79 3.74 4.47
94 Table 48 Importance/Delivery Market Intelligence Gaps (0 = no gap) Market intelligence items Importance score Delivery score Gap score Rank Key marketing challenges 6.66 5.43 1.23 5 Target profile 6.50 5.41 1.09 7 Budget info rmation 6.39 5.93 0.46 13 Criteria measuring success of the campaign 6.37 4.45 1.92 1 Communication objectives 6.36 5.31 1.05 8 Timing consideration 6.21 5.92 0.29 14 Marketing objectives 6.21 4.76 1.45 2 Mandatories 6.13 5.22 0.91 9 Other marcomm ac tivities 5.95 4.56 1.39 3 Creativity specification 5.83 4.92 0.91 9 Brand situation 5.68 4.79 0.89 11 General market situations 5.59 4.31 1.28 4 Competitive situation 5.49 4.36 1.13 6 Company related background information 4.71 4.10 0.61 12 Ta ble 49 The results of principal component factor analysis of gap s cores ( N =96) Market intelligence items F1 F2 F3 Commu nality Factor 1. Info. about objectives and situation Company related background information .80 .65 Other marcomm activities .72 .73 General market situation .72 .73 Competitive situation .67 .68 Brand situation .65 .59 Creativity specification .63 .54 Factor 2. Info. about imminent concerns & accountability Criteria measu ring success of the campaign .87 .77 Marketing objectives .78 .70 Key marketing challenges .63 .44 Factor 3. Info. about restrictions Timing consideration .78 .64 Budget information .74 .65 Mandatories .64 .44 Eigen Value 4.70 1.58 1.28 % of Variance 39.16 13.13 10.63 Cumulative % 39.16 52.29 62.92 Mean (SD) 1.01 (1.33) 1.55 (1.38) .52 (1.17)
95 Table 41 0 Model summary of stepwise multiple regression model Overall Model Fit R2 Change Statistics St ep R R2 Adj.R2 S.E R 2 Change F Change d f 1 d f 2 Sig.F Change 1 48 .2 4 .2 3 .6 8 .2 4 2 6 79 1 86 .000 2 .5 5 .3 0 .2 8 .6 6 .0 6 7. 20 1 85 .0 0 9 Step 1: market challenge Step 2: market challenge, company related background information Note. D ependent variable= t he overall relationship quality Table 41 1 Effects of gap scores in the regression model on the overall relationship quality Model B S.E. 1. Constant 6. 3 8 .10 Market challenge 28 .0 6 4 9 *** 2. Constant Market challenge 6. 4 1 .2 5 .1 0 .0 5 .4 3 *** Company related background info .1 1 .0 4 .2 5 Note. p <.05, ** p <.01, ***p <.001 Table 41 2 Effects of market intellig ence gap factors on the overall relationship quality Gap factors B S.E. Constant Gap f actor 1 info about objectives & situation 6. 37 .1 4 .12 .0 7 2 5 Gap f actor 2 info about m arketer s concerns .1 4 .07 25 Gap f actor 3 info about restric tions .0 7 .0 7 11 R = .4 1 R 2 = .1 7 Adjusted R 2 = .1 4 Note. p <.05, ** p <.01, ***p <.001
96 Table 41 3 Importance score comparison between two structures of advertising agencies Items Fullservice (n= 49) Media Agencies (n=3 5 ) t Sig. Key marketi ng challenges 6. 69 (. 51 ) 6. 60 (. 85 ) 63 53 Marketing objectives 6.22 (. 92 ) 6.2 6 (.92 ) 16 87 Budget information 6.4 5 (.84 ) 6.2 6 (1.07 ) 92 36 Communication objectives 6.33 (.80 ) 6.46 (.66) 79 43 Brand situation 5.73 (1.0 2 ) 5.51 (1.07 ) 96 34 O ther marcomm activities 5.96 (.89 ) 5.94 (1.08 ) .08 94 Criteria for measuring success 6.35 (.83 ) 6.40 (.9 1) .28 78 Target profile 6.47 (.84 ) 6.43 (.95 ) 21 84 Creativity specification 5.59 (1.1 0 ) 5.83 (1.10 ) .97 33 Competitive situation 5.47 (.94 ) 5.20 (1.26 ) 1.13 26 Mandatories 6.10 (1.2 5 ) 6.00 (1.31 ) .36 72 General m arket situation 5.65 (1.15 ) 5.31 (1.21 ) 1 31 20 Timing consideration 6.22 (1.07 ) 6.03 (1.10) 82 41 Company related background information 4.80 (1.27 ) 4.40 (1.38 ) 1 36 18 T otal 6.00 ( .6 4 ) 5. 90 ( .6 1 ) 72 47 Note. ( )=standard deviation Table 41 4 Deliv ery score comparison between two structures of advertising agencies Items Fullservice (n= 49) Media Agencies (n=3 5 ) t Sig. Key marketing challenges 5.57 ( 1.26 ) 5. 23 ( 1 3 3 ) 1 20 23 Marketing objectives 4.86 ( 1 53 ) 4.34 ( 1 98 ) 1 34 1 8 Budget information 5.96 ( 1.22 ) 6.20 (1. 11 ) 93 36 Communication objectives 5.47 ( 1.37 ) 5 03 ( 1 51 ) 1. 40 1 7 Brand situation 4.92 (1. 50 ) 4.40 (1. 63 ) 1 .5 1 1 4 Other marcomm activ ities 4.84 ( 1 65 ) 4.03 (1. 72 ) 2 17 03 Criteria for measuring success 4 65 ( 1.59 ) 4.00 ( 1 85 ) 1 74 09 Target profile 5.57 ( 1 44 ) 5.09 ( 1 77 ) 1 38 1 7 Creativity specification 4.86 (1. 57 ) 5.00 (1. 65 ) 4 0 6 9 Competitive situation 4.53 ( 1 46 ) 3.91 (1. 77 ) 1 75 09 Mandatories 5.29 (1. 67 ) 5 .06 (1. 66 ) 62 54 General m arket situation 4 .76 (1. 60 ) 3 49 (1. 65 ) 3.54 00 ** Timing consideration 5 98 (1. 33 ) 5 97 (1. 36 ) .03 98 Company related background information 4.45 (1. 71 ) 3 46 (1. 79 ) 2 57 01 Total 5.12 ( 1 03 ) 4.66 ( 1 0 0 ) 2 06 0 4* Note. p <.05, ** p <.01, ***p <.001 ( )=standard deviation
97 Table 41 5 Importance Delivery Gap score comparison between two structures of advertising agencies Items Fullservice (n= 49) Media Agencies (n=3 5 ) t Sig. Key marketing challenges 1. 12 ( 1.32 ) 1.3 7 ( 1 37 ) 84 40 Marketing objectives 1.37 ( 1 51 ) 1.91 ( 2 20 ) 1. 35 1 8 Budget information .49 (1.42 ) .06 (1. 06 ) 1 53 13 Communication objectives .86 (1.24 ) 1.4 3 (1.58) 1.86 07 Brand situation .82 (1. 52 ) 1 11 ( 2 .01 ) .77 44 Other marcomm activities 1.12 (1.6 4 ) 1.9 1 (1. 67 ) 2. 1 7 .0 3 Criteria for measuring success 1.69 (1. 79 ) 2.40 (1. 94 ) 1. 72 09 Target profile .90 (1. 26 ) 1.34 (1.7 5 ) 1. 36 .1 8 Creativity specification .73 (1. 38 ) .83 (2.09 ) 25 81 Competit ive situation .94 (1. 6 3 ) 1.29 ( 2 27 ) 82 4 2 Mandatories .82 (1.7 6 ) .94 (1. 98 ) 31 76 Market situation .90 (1. 69 ) 1.83 ( 1.98 ) 2.32 .0 2 Timing consideration .24 (1.3 2 ) .06 (1. 76 ) 56 58 Company related background information .35 (1.6 0 ) .94 ( 2 20 ) 1. 44 1 5 Total .8 8 ( .9 4 ) 1 2 4 ( 1. 1 5 ) 1. 5 9 12 Note. p <.05, ** p <.01, *** p <.001 ( )=standard deviation Tab le 4 16. Relationship score comparison between structures of advertising agencies Relationship Quality Fullservice (n= 49) Media Agencies (n=3 5 ) t Sig. Trust 6. 1 0 ( 93 ) 5.46 ( 1.43 ) 2.5 0 01 Commitment 6.08 ( .96 ) 5.90 ( 1.22 ) .74 .4 6 Satisfaction 6.14 ( 1.00 ) 5 69 (1. 10 ) 1. 9 8 0 5* Cooperation 6.56 ( .59 ) 6 28 ( .71 ) 2.00 05* Overall relationship quality 6.22 ( .75 ) 5 .8 3 ( .69 ) 2 43 02* Note. p <.05, ** p <.01, ***p <.001 ( )=standard deviation
98 Table 417. Summary of Regression Analyses for the relationship quality constructs Dependent Variables Trust Commitment Satisfaction Cooperation Market Intelligence Gap .28 ** .3 2 ** .3 6 ** 09 F (1,78) 7 .2 9 10.13 13.20 .65 Significance .008 .002 .000 .42 t value 2.70 3.18 3.63 .81 R 2 Adjusted R 2 .08 .07 .10 .09 .13 .12 .01 .00 Note. The regression coefficients are standardized regression coefficients p <.05, ** p <.01, *** p <.001 Table 41 8 Goodness of fit indices and X 2 tests for the measurement models Model RMSEA CFI SRMR NNFI X 2 / df General 26 80 15 75 309 29/54=5.73 Independent 08 98 05 97 76 14/48=1.59 Second Order .09 .97 .08 .96 87.50/50=1.75
99 Table 41 9 Correlation matrix for the modified independent factor structural model Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 1. gap 1 2. trus1 -.265 1 3. trus2 -.265 .600 ** 1 4. trsu3 .171 .643 ** .583 ** 1 5. comit1 .304 ** .244 .159 .321 ** 1 6. comit2 .318 ** .264 .214 .170 .666 ** 1 7. comit3 .225 .175 .062 .194 .762 ** .614 ** 1 8. satis1 .391 ** .351 ** .313 ** .382 ** .602 ** .410 ** .526 ** 1 9. satis2 .338 ** .343 ** .373 ** .452 ** .652 ** .466 ** .581 ** .890 ** 1 10. satis3 -.347 ** .337 ** .279 ** .417 ** .696 ** .491 ** .643 ** .868 ** .917 ** 1 11. coop1 -.055 .089 .026 .114 522 ** .303 ** .568 ** .415 ** .510 ** .479 ** 1 12. coop2 -.077 .077 -.036 .068 .485 ** .264 .515 ** .404 ** .464 ** .485 ** .723 ** 1 13. coop3 .071 .134 .140 .193 .557 ** .394 ** .672 ** .420 ** .555 ** .505 ** .717 ** .637 ** 1 Note. Gap=market intelligence gap; trus=trust; comit=affective commitment; satis=relationship satisfaction; coop=intention to cooperate *= p<.05, **=p<.01 Table 420. Summary of t otal/direct/indirect effects of the modified independent factor model Variable Effect Gap Trust Commitm ent Satisfaction Trust Total .29 Direct .29 Indirect Satisfaction Total .37 .42 ** Direct .25 .42 ** Indirect .12 Commitment Total -.28 ** .32 ** .75 ** Direct .75 ** Indirect .28 ** .32 ** C ooperation Total -.20 ** .23 ** .73 ** .55 ** Direct .73 ** Indirect -.20 ** .23 ** .55 ** Note. *= p<.05, **=p<.01
100 Figure 41 Factor loadings of the general factor measurement model Figure 42 Factor loadings of the independent factor measurement model .82 80 8 7 96 .9 1 95 .8 6 70 90 82 73 .7 9 52 55 58 9 6 .8 9 .9 5 66 52 73 .46 33 37 Relationship Quality V1 V2 V3 V4 V5 V6 V7 V8 V9 V1 0 V11 V12 T 1 T 2 T 3 C1 C2 C3 S1 S2 S3 P1 P2 P3 Trust Commitment Satisfaction Cooperation
101 Figure 43 Factor loadings of the s econdorder measurement model 89 74 .8 4 42 Trust Commitment Satisfaction Cooperation Relationship Quality
102 Goodness of fit indices: X2/df =77.60/57=1.36, RMSEA=.06, SRMR=.05, CFI=.97, NNFI=.96 Note. p <.05, ** p <.01, *** p <.001 Figure 44 Modified independent factor s tructural model for market intelligence gap and relationship quality (Full model) Trust Commitment Satisfaction Cooperation Market Intell igence Gaps 06 30 ** 24* .7 6 ** 18 08 43** 29 55 **
103 Goodness of fit indices: X2/df =82.24/61=1.35, RMSEA=.06, SRMR=.06, CFI=.97, NNFI=.97 Note. p <.05, ** p <.01, *** p <.001 Figure 45. Modified independent factor s tructural model for market intelligence gap and relationship quality (Reduce d model) 73 ** Market Intelligence Gaps Trust Satisfaction Commitment Cooperation 29 ** 25 ** 42 ** 75 **
104 CHAPTER 5 DISCUSSION The primary purpose of this study was to examine the effects of market intelligence gaps on perceived relationship quality from the perspective of professional service providers (i.e., media planners). To achieve this pur pose the current study attempted to 1) identify market intelligence gaps that are operationalized as discrepancies between the perceived value of market intelligence and the delivery of the market intelligence, 2) compare perceptions and gaps of market in telligence between two different types of advertising agencies : full service and media planning/buying agencies, 3) examine the effects of market intelligence gaps on overall relationship quality as well as four dimensions of the relationship quality const ruct: trust, commitment, satisfaction, and cooperation, and 4) find the underlying structure of the relationship quality construct in the context of agency client relationships A total of 5 research questions and 5 hypotheses were addressed and tested by various statistical analyses such as EFA, CFA, repeated measures ANOVA, independent samples T test, Multiple Regression, and covariance structural analysis. This section will begin with the discussion on the results. Next, managerial and academic contribut ions of the research findings from this study will be discussed. Lastly, limitations of the study will be addressed along with suggestions for future research. Market Intelligence Gaps Fourteen items that measured the importance of market intelligence was factor analyzed into three categories: objectives & situation, marketers concerns, and accountability related information. The results confirmed that media planners value each of the market intelligence categories differently Information about marketers
105 concerns appeared to be the most important information for better media planning, followed by accountability related information and objectives & situation. It was unexpected that information regarding objectives and situation was rated significantly les s important than others because the information is considered universal market intelligence in marketing practices. M edia planners rather, gave more credits to information regarding problems with which marketers are faced. I t may imply that media planner s today are proactive in solving marketing problems and think themselves as marketing partners, rather than mere behindscene scientists. By market intelligence item, they rated higher for information about key marketing challenges, target profile, budget information, and criteria measuring success of the campaign than other information types These findings are different from the finding s of Sutherland, Duke, and Abernethy (2004) in that they found target profile, marketing strategy, and product usage inf ormation are preferred by creative staffer T he differences may suggest that clients should write the media brief differently from the creative brief in considering what media planners would like to receive The study s results also showed that media planners do not get the information as frequently as they would like. The market information gap was significantly higher for marketers concerns, which comprise of marketing objectives, key marketing challenges, and criteria measuring success of the campaign, than other gap factors. T his finding may suggest that clients do not share their marketing concerns with media planners as frequently as media planners value the information. I n this rapidly changing marketing environment, however, sharing such information is critical for media planners to keep up with the market place and produce upto dated media plans On the other hand, the gap
106 was found to be smaller for information regarding restrictions which comprise of timing, budget, and mandatories. It is not sur prising because such restrictions are always given to advertising staffer to limit the scope of media plans According to Sutherland, Duke, and Abernethy (2004), t he purpose of this type of analysis is to provide the market intelligence that media planner s value most, rather than to reduce the information gap. A large information gap as a matter of fact, is not necessarily a bad indicator per se because a large gap may be preferred for the information that is less valued. Therefore, this study proposes th e need for clients to provide media planners certain types of market intelligence that are valued but not frequently shared. Those types of market intelligence, according to the findings, include key marketing challenges, criteria measuring success of the campaign, target profile, and marketing objectives. T here are several possible explanations for the communication flaw, market intelligence gap. According to Sutherland, Duke, and Abernethy (2004), client may not have the information or decide not to shar e certain types of information with the agency. For example, clients may have no clear idea about how to measure success of the campaign and thus they leave media planners uninformed. Another explanation is that clients may assume that all the information with which they provide account manager will transfer to the media planner. Nonetheless, there are many evidences of interagency or betweenagency communication problems. Thus, it is critical for clients to ensure that media planners receive the necessary information as intended. Most importantly, c lients should be aware of the fact that sharing critical information with agency staffer is important especially in improving chances of success of the campaign
107 (Helgesen 1994; Mi chell 1986/1987 ; Sissors and Bar on 2002) In a task oriented relationship such as advertising agency client relationships, exchange of information is a precondition for successful completion of tasks (i.e., media planning) (Beard 1996) Differences between Types of Advertising A gencies The study initially proposed to compare market intelligence gaps and perceived relationship quality among three types of advertising agencies: full service agency, inhouse agency, and media planning/buying agency. However, due to the limited number of res ponses from media planners in in house agencies the researcher had to combine responses from in house agencies with those from fullservice agencies Thus, the comparison was made between the two types of agencies (i.e., fullservice vs. media specialized agencies) The initial assumption of this comparison was based on number of relay points during communication procedure from the client to the media planner. Sutherland, Duke, and Abernethy s (2004) information flow model was adopted to explain communica tion flow from the client to the media planner. M edia planning/buying agencies were believed to have minimal number of failing points in that media planners have a direct route to communicate with the client representatives. Accordingly, it was hypothesized that media planners in media agencies may perceive better communication quality (i.e., smaller market intelligence gap) and thus higher relationship quality with the clients than those in full service agencies. However, t he results revealed opposite find ings from the expectation. Media planners in media agencies reported significantly less perceived delivery of market intelligence than those in full service agencies. Accordingly, less information gap although not significant was found in full service agencies than in media agencies.
108 There are some p ossible explanations of these contradicting results First, the basic assumption that number of communication relay points may influence market intelligence gaps may be wrong in the first place. Second, however, if the assumption is correct, the communication flow this present study proposed may be incorrect. For example, there is a possibility that media planners in the media agenc y have more, not less, communication relay points than those in the fullservice a genc y. In fact, m any media agencies are liaised by external agencies (i.e., creative boutiques or agencies without media function) i n communicating with client representatives. The results of the current study would gain a validity if it was confirmed that the client download critical information to media agencies through other agencies. Although identifying the communication flow is not the scope of this study, this contradicting result suggests the needs for future studies to explore the flow of market information between the client and the agency depending on types of agencies. In addition, t he different perception of relationship quality between full service and media agencies was reported and this finding may be explained by the number of communication relay points, as proposed. However, there may be a n alternative explanation for the difference. It is plausible that the difference in the relationship perception may be because of the way the agency is being treated. For example, the fullservice agency may be considered a marketing partner, whereas the media agency is treated as a vendor that is specialized only in buying time and space for the advertising campaign. H untley ( 2006) argued that the relationship atmosphere may influence on their perceived relationship such as commitment. This alternative
109 explanation suggests for the future study to examine the effects of the p erception of the level of partnership with the client on perceived relationship quality. In sum, the result s of the comparison between the two types of agencies confirmed that there exist differences in perception of information gaps and the relationship quality between the two agency types. But, the contradicting results from the initial expectation provided the needs for alternative approaches to broaden understanding of the differences among different organizations. Relationships between Market I ntelligence Gaps and Relationship Quality Findings from the present study strongly suggest ed t hat the market intelligence gap is negatively related to the relationship quality suggesting that media planners who think they receive necessary market intelligence as frequently as they need are more likely to 1) be satisfied with, 2) trust on, 3) commit to, and 4) cooperate with the client. Only exception was with the information regarding restrictions suggesting the level of the relationship quality is not affected by the restriction information It would be imperative for the client to assure that media planners receive all the necessary inform ation in order to improve the relationship quality as perceived by the media planner. By individual information item, two items (i.e., key marketing challenges and company rel ated background information) wer e salient in predicting the relationship quality It is interesting to note that the company s background information is significantly less important in planning media, but i s significantly important in predicting the relationship quality. It implies that media planners may want to receive information th at makes them better understand the client and thus feel like a part of the client organization. From this point of view it would be worthwhile for the client to share company related information with the media planner. Such information includes, as in
110 th e survey, news releases, downloadables, corporate blogs, or thought leadership articles. In terms of the relationship among the constructs of relationship quality, a multiple regression analysis initially provided evidence that the market intelligence gap directly influence on trust, commitment, and satisfaction, but not on cooperation. Among those influenced by the intelligence gap, satisfaction was the most influenced construct. Further a covariance structural analysis provided a richer explanation for t he relationship among the relational outcomes though extra care should be taken because of the nature of sample The results implied evidences of direct and/or indirect effects of the market intelligence gap on e ach of the four constructs. The summary of t he results is as follows; both trust and satisfaction are directly influenced by the market intelligence gap, whereas commitment and cooperation are indirectly related to the gap. A lso, it was found that trust is not a direct indicator of commitment or cooperatio n, rather an indirect indictor through satisfaction. T h ese f indings are consistent with the previous literature for the communication effects on satisfaction and cooperation ( Anderson, Hikansson, and Johanson 1994; Anderson and Narus 1990; Chenet, Tynan, and Money 1999; Dwyer, Schurr, and Oh 1987; Lancastre and Lages 2006) For example, as suggested by Mohr, Fisher, and Nevin (996) satisfaction was the most influenced relati onal outcome by communication, operationalized in this study as the market intellig ence gap. Also, cooperation appeared to be an outcome of relational exchanges, rather than one of the dimensions that is directly influenced by the market intelligence gap. For trust, however, the results are somewhat different from the previous fi ndings. According t o Morgan and Hunt s (1994)
111 KMV model trust and commitment mediate between communication and cooperation, and commitment is influenced by communication through trust. The present study, however, suggested that commitment is influenced di rectly by satisfaction and indirectly by trust. Also, the magnitude of satisfaction in predicting both commitment and cooperation appeared to be larger than the one of trust. One possible explanation of this finding may be that in the advertising industry the service provider s trust on the client may not be the most important indicator of affective commitment to their clients. In other words, service providers (i.e., media planners ) commit affectively to the client because they are satisfied with the relat ionship, rather than they trust on the client. As reviewed previously, c lients may value trust on agencies more than satisfaction with agencies in that they seek for competency of an advertising agency. The service provider s perspective, however, suggested that the relationship satisfaction may be the most important factor in increasing service providers commitment to the client. The practical implication of this finding is discussed in terms of agency client relationships in the next section. Implication s of the Research This study provides implications both for practitioners and academicians. Practically, both client representatives and media planners would benefit from this study. This present study, first of all, created a comprehensive set of market i ntelligence items for clients to disseminate to media planners. It provides a checklist of important market intelligence items for better media planning and thus reduces clients role ambiguity (Beard 1996). The client should consider providing market int elligence that media planners really value, such as key marketing challenges, criteria measuring success of the campaign, and other imminent marketing concerns, for better media
112 planning. Second, f or client representatives to maximize media planners commi tment as well as cooperation, the client should frequently provide market intelligence items that make the media planner feel like they are a partner, rather than a vendor, such as key marketing challenges and company related background info rmation. T h ird, m edia planners should include market intelligence items that they think important in the media brief and make sure to receive critical information. As suggested by Beard (1996), having critical market intelligence may help media planners to enhance their performance and thus improve their relationship with the client. Fourth, agency managers should provide both client s and media planners with chances to increase the relational satisfaction in that it may help develop more commitment of the media planner to the client and, in turn, more intention to cooperate. Finally this study may evoke the import ance of the relationship notion in agency client relationships In fact, as competition among agencies gets severe, they tend to build their relationships with clients based on transactionoriented approaches. For example, they cut off their remuneration by 2%~3% of the total advertising expenditure, which was 15% commission back in the day. It will eventually be detrimental to the industry overall; thus, this stu dy is expected to increase the notion of relationship marketing in the context of agency client relationship. Academic contributions are made in many ways. First, this particular study has an important implication in that not much research in the advertis ing field has examined the agency client relationship. M o st studies regarding the agency client relationships have focused their attention on clients perception about the relationship and possible reasons of termination of the relationship. T his study ho wever, e xpands the notion of
113 relationship marketing to the professional service sector, advertising and chang es viewpoints from client to service providers (i.e., media planners) Second, t his study provides an empirical examination of the communication problem between client representatives and media planners I t specifically uses market intelligence gaps to identify from where the communication problems stem during the communication process Next, Not much research in the advertising field, if any, has b een interested in the relationship quality construct in the agency client relationship. This study investigates the effects of the market intelligence gap on the relationship quality perceived by service providers in the advertising industry. The development of the conceptual model of relationship constructs may help researchers understand the nature of the agency client relationship. All in all, this study expends advertising literature by applying relationship marketing theories to the agency client relat ionship research. Limitations and Future Direction s As with other studies, this dissertation has limitations The first lim itation is related to the nature of sample collected for this study. The researcher strived to give a chance to be a survey particip ant to as many individuals as possible in the population of interest media planners. T herefore, a total of 1,430 media planners nationwide in the U.S. was contacted by a postcard invitation. However, only 62 persons (4.3%) participated in the online survey despite a follow up that was sent through corporate emails. Although 71 more respondents were further recruited through the second wave of the survey, a total of 133 respondents, with some incompletes, may limit this study s generalizability. One possib le explanation of such a low response rate for the first round of the survey may be because of the invitation method. Research has suggested that a p ostcard is effective
114 when it is used to improve response rate for mail survey s as a primer or a reminder (C hiu and Brennan 1990) Not much research, however has reported the effectiveness of a postcard as a main survey invitation tool Another explanation maybe that media planners at work usually are tied to a schedule and thus do not have time for a study par ticipation. Meanwhile, i t should be noted that an email would be a better method recruiting participants for an online survey, however personal email addresses for media planners were restricted from disclosing. C onsidering the limited response rate f r om this particular group of interest, future studies should contact more potential participants to collect reasonably enough number of responses. Second, the context of the study might limit the generalizability of the findings. Although media planning now is one of the most important functions of advertising agencies, the findings might be different wh en the same questions were administered to other disciplines such as account managers or creative staffer. I n order to embrace a broad scope of agency client relationships, researchers should attempt to understand perceptions of various disciplines in an advertising agency. Third, the present study did not clearly indentify the reasons why the comparison between the two agency types produced opposite findings from the expectation. Future studies would benefit if they, as discussed earlier, identify the flow of information from clients to agency staffer. N umber of communication relay points f or example, may be measured and compared to the quality of communicati on as well as the relationship quality. Further, it would be worthwhile to explore the role of the level of perceived partnership in predicting the relationship quality with regard to the comparison of the two types of agencies.
115 Fourth, the present study f ocused only on service suppliers perspective. T he relationship theory literature, however, suggested that relationships may be better understood when a researcher scrutinize both sides of relational partners. Despite the importance of service provider s p erspective in building and maintaining relationships, further studies sh ould investigate both parties in the relationship simultaneously in order to provide richer understanding s about the agency client relationship. Fif th, the scope of the study might li mit the contri bution of this study in that this study focused only on communication problems. A dvertising studies regarding agency client relationships have suggested various factors that might influence on the relationship. For example, the agency client relationship may be depending on humanrelated issues such as client s role ambiguity or changes of advertising managers ( Beard 1996; Wackman, Salmon, and Salmon 1986/1987). Thus, it would be worthwhile for fu ture stud ies to examine the effects of all poss ible factors on the relationship quality. Six th this study employed four r elationship quality constructs which may not be the full range of possible components. Hence, further empirical testing with a more comprehensive set of distinct relationship qual ity constructs will help fully understand the concept of relationship quality Other relationship quality constr ucts appeared in the literature include : relat ional dependency, control mutuality, intimacy, or love Finally, the present study used the relati onship quality as an outcome of the market intelligence gap, one of the communication problems there may be many other outcomes that are associated with the communication problem. The inclusion of other relationship outcomes as dependent variables such a s relationship longevity,
116 relationship strength, or propensity to leave the relationship, would facilitate understanding agency client relationships with a greater depth.
117 APPENDIX A SURVEY INVITATION: P OSTCARD Front side of the postcard Back side of t he postcard
118 APPENDIX B SURVEY INVITATION: S NOWBALLING Hello, You are cordially invited to participate in an online survey, conducted by the Department of Advertising at the University of Florida. Our goal is to gather perceptions of media planners on the flow of information between you and your clients in the campaign planning process. Please take part in this survey and help us identify the optimal flow of information between clients and media planners. The survey is brief and should take only about 10 minutes of your time. Your survey answers will be used for research purposes only. If you agree to participate in the survey, please click on the link below. https://ufljour.qualtrics.com/SE?SID=SV_098K78YaUVGrvU0&SVID= We would encourage you to forward this email or the survey link to media planners that you know. If you have any question, please do not hesitate to contact firstname.lastname@example.org Sincerely yours, Jun Heo, Ph.D. candidate John C. Sutherland, Ph.D., Professor
119 APPENDIX C SURVEY QUESTIONNAIRE
127 APPENDIX D FOLLOW UP EMAIL FOR THE POS TC ARD RECIPIENTS Hello, I believe you have received a survey invitation sent by the Department of Advertising at the University of Florida. This is a friendly reminder of the invitation. It is ok if you have not received the postcard yet. You can simply go ahead and click on the link below. https://ufljour.qualtrics.com/SE?SID=SV_bro1WVlwP0CF7MM Our goal is to gather perceptions of media planners on the flow of informati on between you and your clients in the campaign planning process. Please take part in this survey and help us identify the optimal flow of information between clients and media planners. The survey is brief and should take only about 10 minutes of your ti me. Your survey answers will be extremely valuable and used for research purposes only. We sincerely appreciate your taking the time and contribution to understanding the market place better. Please contact the researcher if you have any question about the survey. Sincerely, Jun Heo, Ph.D. candidate John C. Sutherland, Ph.D., professor
128 LIST OF REFERENCES Abdul Muhmin Alhassan G. (2005), Instrumental and Interpersonal Determinants of R elationship Satisfaction and Commitment in Industrial Market s, Journal of B usiness R esearch 58(5), 619628 Abratt, Russell, and Deanna Cowan (1999), Client Agency Perspectives of Information Needs for Media Planning, Journal of Advertising Research, 39(6), 3752. Allen, Natalie J., and Meyer P. John (1990), The Measurement and Antecedents of Affective, Continuance, and Normative Commitment to the Organization, Journal of Occupational and Organizational Psychology 63 1 8. Andaleeb, Syed S. (1996), An Experimental Investigation of Satisfaction and Commitment in Marketing Channels: The Role of Trust and Dependence, Journal of Retailing 72(1), 7793. Anderson, James C., and James A. Narus (1990), A Model of Distributor Firm and Manufacturer Firm Working Partnership, Journal of Marketing, 54(1), 4258. ------, and David W. Gerbing ( 1988), Structural Equation Modeling in Practice: A Review and Recommended TwoStep Approac h, Psychological Bulletin 103(3), 411 423. ------, Hakan H a kansso n, and Jan Johanson ( 1994), Dyadic Business Relationships within a Business Network Context Journal of Marketing, 58 ( 4 ), 1 15. Anderson, Erin, and Barton Weitz (1992), The Use of Pledges to Build and Sustain Commitment in Distribution Channels, Journal of Marketing Research, 29(1), 18 34. ------, Leonard M. Lodish, and Barton A. Weitz ( 1987) Resource Allocation Beh avior in Conventional Channels Journal of Marketing Research, 24 ( February), 8597. Athanasopoulou, Pinelopi ( 2006), Relationship Quality: A Critical Literature Review and Research Agenda, European Journal of Marketing, 43(5 / 6), 583610. Baker, Thomas L., Penny M. Simpson and Judy A. Siguaw (1999), The Impact of Sup pliers' Perceptions of Reseller Market Orientation on Key Relationship Constructs, Journal of the Academy of Marketing Science, 27(1), 50 57. Beard, Fred (1996), Client Role Ambiguity and Satisfaction in Client Ad Agency Relationships, Journal of Advertising Research, 39(2), 69 78. Becker, Howard S. (1960), Notes on the concept of commitment, American Journal of Sociology 66, 32 40.
129 Beltramini, Richa rd and Dennis Pitta (1991), Underlying D imensions and C ommunications S trategies of the A dvertising A gency C lient R elationship, International Journal of Advertising, 10, 151 159. Bennett Roger, and Anna Barkensjo ( 2005), Relationship Q uality, R elationship M arketing, and C lient P erceptions of the L evels of S ervice Q uality of C haritable O rganisations International Journal of Service Industry Management 16(1), 81106. Berry Leonard L. (1995), Relationship Marketing of Services: Growing Interest, Emer ging Perspectives, Journal of the Academy of Marketing Science, 23(4), 236245. ----(1983), Relationship Marketing, in Emerging Perspectives on Servic es Marketing, Leonard. L. Berry, Lynn G. Shostack Gregory D. Upah eds., Chicago, IL: American Marketing Association, 2528. ------, and A. Parasuraman (1991), Marketing Service: Competing Through Quality New York: The Free Press. Beverland, Michael, F rancis Farrelly, and Zeb Woodhatch (2007), Exploring the Dimensions of Proactivity within Advertising Agency Client Relationships, Journal of Advertising, 36(4), 4960. Bierly, Paul E., Eric H. Kessler, and Edward W. Christensen (2000) Organizational Learning, Knowledge, and Wisdom, Journal of Organizational Change Management 13(6), 595618. Biernacki, Patrick, and Dan Waldorf (1981), Snowball Sampling: Problems and Techniques of Chain Referral Sampling, Sociological Methods & Research 10(2), 141163. Bleek e Joel, and David Ernst (1993), Collaborating to Compete New York, NY: Wiley. Bollen Kenneth A., and Scott J. Long ( 1993), Testing Structural Equation Models Newberry Park: Sage Publications. Bowen, David E., and Benjamin Schneider (1 988) Services Marketing and Management: Implications for Organizational Behavior in Research in Organizational Behavior Greenwich, CT : JAI Press Inc., 43 80. Boyle, Brett, Robert F. Dwyer, Robert A. Robicheaux, and James T. Simpson (1992), Influence Strategies in Marketing Channels: Measures and Use in Different Relationship Structures, Journal of Marketing Research, 24(Nov.), 462473.
130 Brown James R., and Ralph L. Day (1981) Measures of Manifest Conflict in Distribution Channels Journal of Marketing Research, 18(3), 263274. Brown Michael W., and Robert Cudeck (1992 ), Alternative Ways of Assessing Model Fit Sociological Methods & Research 21 ( 2 ) 230 258 Brown Stephen W., and Teresa A. Swartz ( 1989), A Gap Analysis of Professional Se rvice Quality Jo urnal of Marketing, 53 ( 2 ), 9298. Brown, Steven P., and Son K. Lam (2008) A Meta Analysis of Relationships Linking Employee Satisfaction to Customer Responses Journal of Retailing 84(3) 243255. Buchanan, Bruce (1974), Building Organizational Commitment: the Socialization of Managers in Work Organizations, Administrative Science Quarterly 19, 533546. Cagley, James W. (1986), A Comparison of Advertising Agency Selection Factors: Advertisers and Agency Perceptions, Journal of Advertising Research, 26(June/July), 3944. Cappo, Joe (2003) The Future of Advertising: New media, New Clients, New Consumers in the Post Television Age, McGraw Hill, 152. Chenet, Pierre, Caroline Tynan, and Arthur Money (1999), Service Performance G ap: Re evaluation and Redevelopment, Journal of Business Research, 46(2), 133147. Chiu, Irene, and Mike Brennan (1990), The Effectiveness of Some Techniques for Improving Mail Survey Response Rates: A Metaanalysis, Marketing Bulletin, 1, 1318. Cle ments, Michael (1984), Who Is to Blame for the Low Standards of Industrial Advertising? International Journal of Advertising, 3, 239 244. Cronbach, Lee J. (1951), Coefficient Alpha and the Internal Structure of Tests, Psychometrika ,16 297 334. Croni n Joseph J., and Steven A. Taylor ( 1992) Measuring Service Quality: A Reexamination and Extension, Jo urnal of Marketing, 56( 3 ), 5568 Crosby, Lawrence A., Kenneth R. Evans, and Deborah Cowles (1990), Relationship Quality in Service Selling: An Inte rpersonal Influence Perspective, Journal of Marketing, 54, 6881.
131 De Wulf, Kristof, Gab y OdekerkenSchrder, and Dawn I acobucci (2001), Investments in Consumer Relationships: A Cross country and Cross industry Exploration, Journal of Marketing, 61, 3 5 51. Dorsch, Michael J., Scott R. Swanson, and Scott W. Kelley (1998), T he R ole of R elationship Q uality in the S tratification of V endors as P erceived by C ustomers Journal of the Academy of Mar keting Science 26 (2), 128 42. Dowling, Grahame R. (1994) Searching for a New Advertising Agency: A Clients Perspective, International Journal of Advertising, 13(3), 229242. Doyle, Peter, Marcel Corsjens, and Paul Mi chell (1980), Signals of Vulnerability in Agency Client Relations, Journal of Marketing. 44 (Fall), 1823. Ducoffe Robert H., and Sandra J. Smith ( 1994 ), Mergers and Acquisitions and the Structure of the Advertising Agency Industry, Journal of Current Issues and Research in Advertising, 16 (1), 15 27. Dwyer, Robert F., and Sejo Oh (1987), Output Sector Munificence Effects on the Internal Political Economy of Marketing Channels, Journal of Marketing Research, 24(4), 347358. ------, Paul H. Schurr, and Sejo Oh (1987), Developing Buyer Seller Relationships, Journal of Marketing, 51, 11 27 Enns, Blair (2009), Pitches, Search Consultants & Hissing Cockroaches, available at http://www.winwithoutpitching.com/pitches and roaches (accessed July 29, 2009). Evans, Joel R., and Anil Mathur (2005), The Value of Online Surveys, Internet Research, 13, 195 219. Farace, Richard V., Peter R. Monge, and Hamish M. Russell ( 1977), Communicating and O rganizing, Reading, MA: AddisonWesley Pub. Co. Field, Andy (2009), Discovering St atistics Using SPSS: and Sex and Drugs and Rock n Roll, 3rd ed., London: Sage. Frazier, Gary L., and Raymond C. Rody (1991), The Use of Influence Strategies in Interfirm Relationships in Industrial Product Channels, Journal of Marketing, 55(1), 5269. ------, Robert E. Spekman, and Charles R. O'Neal (1988), Just In Time Exchange Relationships in Industrial Markets, Journal of Marketing 52(4), 5267. For n ell, Claes (1983) Issues in the Application of Covariance Structure Analysis: A Comment, Jo urnal of Consumer Research, 9 (March), 443 448
132 Friman, Margareta, Tommy Garling, Bruce Millett, Jan Mattsson, and Robert Johnston (2002), An Analysis of International Business to Business Relationships Based on the Commitment Trust Theory, Industrial Marketing Management 31(5), 403409. Ganesan, Shankar (1994), Determinants of Long Term Orientation in Buyer Seller Relationships, Journal of Marketing, 58(April), 1 19. ------, and Barton Weitz (1996) The Impact of Staffing Policies on Retail Buyer Job Attitudes and Behaviors Journal of Retailing 72(Spring ) 3156. Garbarino, Ellen, and Mark S. Johnson (1999), The Different Roles of Satisfaction, Trust, and Commitment in Customer Relationships, Journal of Marketing, 63(2), 7087. Graesser, Arthur C., Katja Wiemer Hastings, Roger Kreuz, Peter Wiemer Hastings, and Kent Marquis (2000), QUAID: A Questionnaire Evaluation Aid for Survey Methodologists, Behavior Research Methods, Instruments, & Computers 32 (2), 254262. Gronroos, Christian (1990), Relationship Approach to Marketing in Service Contexts: The Marketing and Organizational Behavior Interface, Journal of Business Research, 20(1), 311. Guiltinan, Joseph P., Ismai l B. Rejab, and William C. Rodgers (1980),Factors Influencing Coordination in a Franchise Channel, Journal of Retailing, 56, 41 58. Gummesson, Evert (1988), Service Quality and Product Quality Combined, Review of Business St. John's University, Winter. Hair, Joseph F., Rolph E. Anderson, Ronald L. Tatham, and William C. Black (1998), Multivariate Data Analysis 5 th e d ., Upper Saddle River, NJ: Prentice Hall. Hakansson Hakan, and Ivan Sne hota (1995) Developing Relationships in Business Networks London: Routledge. Halinen, Aino (1997), Relationship Marketing in Professional Services: A Study of Agency Client Dynamics in the Advertising Sector London and New York: Routledge. Han, Sang L in, David Wilson, and Shirish P. Dant ( 1993), Buyer S upplier R elationships T oday, Ind ustrial Mark eting Manage ment 22(4), 331 338. Haytko, Diana L. (2004), Firm to Firm and Interpersonal Relationships: Perspectives from Advertising Agency Account Manag ers, Journal of the Academy of Marketing Science 32(3), 312328.
133 Heckathorn, D.D. (1997). "Respondent Driven Sampling: A New Approach to the Study of Hi dden Populations". Social Problems. 44: 174199. Heide Jan B., and George John (1988) The Role of Dependence Balancing in Safeguarding TransactionSpecific Assets in Conventional Channels Journal of Marketing, 52( 1 ), 2035 Helgesen, Thorolf ( 1994), A dvertising Awards and Advertising Agency Performance Criteria, Journal of Advertising Research, 34(4), 4353. Henke, Lucy L. (1995), "A Longitudinal Analysis of the Ad Agency Client Relationship: Predictors of an Agency Switch," Journal of Advertising Research, 35 (March/April), 2430. Hennig Thurau, Thorsten and Alexander Klee (1997), The I mpact of C ustomer S atisfaction and R elationship Q uality on C ustomer R eten tion: A C ritical R eassessment and M odel D evelopment, Psychology and Marketing 14(8), 7377 64. Heo, Jun and Chang Hoan Cho (2009), A New Approach to Target Segmentation: MediaUsage Segmentation in the Multi Media Environment, Journal of Targeting, Mea surement, and Analysis for Marketing, 17(3), 145155. Hibbard, Jonathan D., Nirmalya Kumar, and Louis W. Stern (2001), Examining the Impact of Destructive Acts in Marketing Channel Relationships, Journal of Marketing Research, 38 (February), 45 61. Hill, Frances M., and Marlena L. M cC rory (1997) An A ttempt to M easure S ervice Q uality at a Belfast M aternity H ospital: Some M ethodological I ssues and S ome R esults Total Quality Management & Business Excellence, 8 (5), 229 242 Ho lmlund, Maria (2008), A Definition, Model, and Empirical Analysis of Business to Business Relationship Quality, International Journal of Service Industry Management 19(1), 3262. Horsky, Sharon (2006), The Changing Architecture of Advertising Agencies, Marketing Science 25(4), 367383. Hotz, Mary R., John K. Ryans, and Will iam L. Shanklin (1982), Agency / Client Relationships as Seen by Influentials on both Sides, Journal of Advertising, 11(1), 3744. Hu, Li tze and Peter M. Bentler (1999), Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Cri teria Versus New Alternatives Structural Equation Modeling, 6 (1), 1 55.
134 Huntley, Julie K. (2006), Conceptualization and Measurement of Relationship Quality: Linking Relationship Quality to Actual Sales and Recommendation Intention, Industrial Marketing Management 35, 703 714. Hurley, Robert F, Melissa Thau Gropper, and Gianpaolo Roma (1996), The Role of TQM in Advertising: A Concep t ualization and a Framework for Ap plication Journal of Marketing Theory and Practice, 4(3), 11 24. Jap, Sandy, C. Manolis, an d Barton Weitz ( 1999), Relationship Quality and Buyer Seller Interactions in Channels of Distribution Journal of Business Research, 46 303313 Jarvis Lance, and Jam es Wilcox (1977) True Vendor Loyalty or Simply Repeat Purchase Behavior?, Industrial Marketing Management 6 9 14. John, George, Robert Ruekert, and Gil bert A. Churchill, Jr. (1983), The Interrelationships of Role Percepti ons in Channels of Distribut ion, Working Paper 1 831 Graduate School of Business, University of WisconsinMadison. Kalwani, Manohar U. and Narakesari Narayanda (1995), Long Term Orientation in Manufacturer Supplier Relationships: Do They Pay Off for Supplier Firms? Journal of Marketing, 59(January), 116. Keith, Janet E., Donald W. Jackson, Jr., and Lawrence A. Crosby (1990), Effects of Alternative Types of Influence Strategies under Different Channel Dependence Structures, Journal of Marketing, 54(3), 3041. Kline, Rex B. (2005), Principle and Practice of Structural Equation Modeling 2nd, ed.,New York, NY: The Guilford Press. Korgaonkar, Pradeep K., George P. Moschis, and Danny Bellenger (1984), Correlates of Successful Advertising Campaigns, Journal of Advertising Research, 24(February/March), 4753. Koslow, Scott, Sheila L. Sasser, and Edward A. Riordan (2003) What Is Creative to Whom and Why? Perceptions in Advertising Agencies, Journal of Advertising Research 43 (1), 96 110. Krosnick, Jon A., and Le andre R. Fabrigar (1997) Designing R ating S cales for E ffective M easurement in S urveys in Survey Measurement and Process Quality L ars E Lyberg, P aul Biemer, M artin Collins, E dith D. DeLeeuw, Cathryn Dippo, Norbert Schwarz, and D ennis Trewin eds., New York NY : Wiley Interscience. Kulkarni, Mukund S., Premal P. Vora, and Terence A. Brown (2003), Firing Advertising Agencies: Possible Reasons and Managerial Implications, Journal of Advertising, 32(3), 7786.
135 Kumar, Nirmalya, Lisa K Scheer, and JanB enedict E M Steenkamp (1995) The E ffects of S upplier F airness on V ulnerable R esellers Journal of Marketing Research, 32 (1), 5465. Lagace, Rosemary R., Robert Dahlstrom, and Jule B. Gassenheimer (1991), The Relevance of Ethical Salesperson Behavior on Relationship Quality: The Pharmaceutical Industry The Journal of Perso nal Selling & Sales Management 11 (4), 3947. Lages, Luis F., Andrew Lancastre, and Carmen Lages (2008), The B2B RELPERF scale and scorecard: Bringing relationship marketing theory into business to business practice, Industrial Marketing Management 37, 686 697. Levy Michael, and Barton Weitz (2000 ) Retailing Management McGraw Hill College. Lloyd, Carla V., Jan Slater, and Brett Robbs (2000), The Advertising Marketplace and the Media Planning Course, Journalism and Mass Communication Educator 55(3), 413. Lovelock, Christopher H. (1983), Classifying Services to Gain Strategic Marketing Insights, Journal of Marketing, 47(3), 920. Lusch Robert F., and James R. Brown ( 1996), Interdependency, Contracting, and Relational Behavior in Marketing Channels Journal of Marketing, 60 ( 4 ), 1938 Malts, Elliot and Ajay K. Kohli (1996), Market Intelligence Dissemination across Functional Boundaries, Journal of Marketing Research 33(1), 4761 Mathieu, John E., and Dennis M. Zajac (1990) A Review and MetaAnalysis of the Antecedents, Correlates, and Consequences or Organizational Commitment Psychological Bulletin 108, 171 194. Meyer John P., and Natalie J. Allen (1984) Testing the S ide B et T heory of O rganizational C ommitment: S ome M ethodological C onsiderations Journal of Applied Psychology 69 372 378. ------, Sampo V. Paunonen, Ian R. Gellatly, Richard D. Goffin, and Douglas N. Jackson (1989), Or ganizational Commitment and Job Performance: It's the Nature of the Commitment That Counts, Journal of Applied Psychology 74(1), 152156. Mills, Peter K., and Newton Margulies (1980), Towards a Core Typology of Service Organizations, Academy of Management Review 5 255 265. Mills Peter K., and James H. Morris ( 1986), Clients as Partial Employees of Service Organizations: Role Development in Client Participation, The Academy of Management Review 11(4), 726735.
136 Michell, Paul C. (1986/1987), A uditing of Agency Client Relations Journal of Advertising Research 26 (December/January), 2941. ------, Harold Cataquet, and Stephen Hague (1992), Establishing the Causes of Disaffection in Agency Client Relations, Journal of Advertising Research, 3 2 (March/April), 41 48. Mohr, Jakki J., Robert J. Fisher, and John R. Nevin (1996), Collaborative Communication in Interfirm Relationships: Moderating Effects of Integration and Control, Journal of Marketing, 60(3), 103115. Moorman, Christine, Rohit D eshpande, and Gerald Zaltman (1993) Factors A ffecting T rust in M arketing R elationships Journal of Marketing, 57, 81101. ------, Gerald Z altman Rohit Deshpande (1992), Relationships between P roviders and U sers of M arket R esearch: the D ynamics of T ru st within and between O rganizations Journal of Marketing Research, 29 ( 3 ) 314 28. Morgan, Robert M., and Shelby D. Hunt (1994), The Commitment Trust Theory of Relationship Marketing, Journal of Marketing, 58(3), 2038. Mowday, Richard T., Lyman W. P orter, and Richard M. Steers (1982), EmployeeO rganization L inkages: The P sychology of C ommitment, A bsenteeism, and T urnover New York NY : Academic Press. ------, Richard M. Steers, and Lyman W. Porter (1979) The M easurement of O rganizational C ommitmen t Journal of Vocational Behavior 14(2), 224247 Ndubisi Nelson O. (2006) A structural E quation M odelling of the A ntecedents of R elationship Q uality in the Malaysia B anking S ector Journal of Financial Services Marketing, 11(2), 13 1 141. Nunnally Jum C. ( 1978 ), Psychometric T heory New York, NY: McGraw Hill. OdekerkenSchrder, G aby Kristof D e Wulf, and Patrick Schumacher (2003), Strengthening Outcomes of Retailer Consu mer Relationships: The Dual Impact of Relationship Marketing Tactics and Consumer Personality, Journal of Business Research, 56, 177 190. Oliver Richard L. ( 1980 ), A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions Journal of Marketing Research, 17 ( 4 ), 460 469. Palmatier Robert W., Rajiv P. Dant, and Kenneth R. Evans (2006), Factors Influencing the Effectiveness of Relationship Marketing: A Meta Analysis, Journal of Marketing, 70 (October), 136 53.
137 Parasuraman, A., Valerie A. Zeithaml, and Leonard L. Berry (1985), A Conceptual Model of Service Quality and Its Implications for Future Research, Journal of Marketing, 49, 4150. Patterson, Paul G., and Richard A. Spreng (1997) Modeling the R elationship between P erceived V alue, S atisfaction and R epurchase I ntentions in a B usinessto B usiness, S ervices C ontext: A n E mpirical E xamination International Journal of Service Industry Management 8(5), 414434. Polit Denise F., and Bernadette P. Hungler ( 1999), Nursing Research: Principles and Methods Philadelphia: Lippincott. Porter, Stephen R., Michael E. Whitcomb, and William H. Weitzer ( 2004), Multiple Surveys of Students and Survey Fat igue N ew D irections for I nstitutional R esearch, 121 Porter, Lyman W., Richard M. Steer, Richard T. Mowday, and Paul V. Boulian (1974) Organizational C ommitment, J ob S atisfaction, and T urnover a mong P sychiatric T echicians Journal of Applied Psychology 59, 603 609. Rauyruen, Papassapa and Kenneth E. Miller (2007) Relationship Q uality as a P redictor of B2B C ustomer L oyalty Jo urnal of Business Research, 60(1), 2131. Roberts, Keith, Sajeev Varki, and Rod Brodie (2003), Measuring the Q uality of R elationships in C onsumer S ervices: A n E mpirical S tudy European Journal of Marketing, 37(1), 16996. Scanlan, Laurel, and Janelle McPhail ( 2000), Forming Service Relationships with Hotel Business Travelers: The Critical Attributes to Improve Retention, Journal of Hospitality & Tourism Research, 24, 491 513. Schlesinger Leonard A. and James L. Heskett (1991) The ServiceDriven Service Company Harvard Business Review 69(5), 7181. Schumacker Randall E., and Richard G. Lomax ( 2004), A B eginners G uide to S tructural E quation M odeling, Mahwah, NJ : Lawrence Erlbaum Schurr Paul H., and Julie L. Ozanne (1985) Influences on Exchange Processes: Buyers' Preconceptions of a Seller's Trustworthiness and Bargaining Toughness Journal of Consumer Research 11(4), 939 953. Schwartz Tony ( 1983), Media: The Second God, Doubleday. Selnes, Fred (1993), An E xamination of the E ffect of P roduct P erformance on B rand R eputation S atisfaction and L oyalty, European Journal of Marketing, 27(9), 19 35.
138 Sharma, Neeru and Paul G. Patterson (1999), The Impact of Communication Effectiveness and Service Quality on Relationship Commitment in Consumer, Professional Services, Journal of Services Marketing, 13(2), 151170. Silk, Alvin J. and Ernst R. Berndt (1993), S cale and Scope Effects on Advertising Agency Costs, Marketing Science, 12(1), 5372. Sissors, Jack Z., and Roger B. Baron (2002), Advertising Media Planning Chicago, IL: McGraw Hill. Ste rn, Louis W., and Adel I. El Ansary (1988) Marketing Channels Englewood Cliffs, NJ: Prentice Hall. Storbacka, Kaj, Tore Strandvik, and Christian Grnroos (1994), Managing Customer Relationships for Profit: The Dynamics of Relationship Quality, International Journal of Service Industry Management 5(5), 2138. Su therland, John C., Lisa Duke, and Avery Abernethy (2004), A Model of Marketing Information Flow: What Creatives Obtain and Want to Know from Clients, Journal of Advertising, 33(4), 5952. Tabachnick Barbara G., and Linda S. Fidell ( 1996), Using Multiva riate Statistics 3rd ed., New York, NY: HarperCollins College Publishers. Ulaga, Wolfgang, and Andreas Eggert (2006), ValueBased Differentiation in Business Relationships: Gaining and Sustaining Key Supplier Status, Journal of Marketing, 70(1), 11913 6. Vanden Bergh, Bruce G., Sandra J. Sm ith, and Jan L. Wicks (1986), I nternal Agency Relationships: Account Services and Creative Personnel, Journal of Advertising, 15(2), 55 61. Varona, Federico (1996), Relationship between Communication Satisfactio n and Organizational Commitment in Three Guatemalan Organizations, The Journal of Business Communication, 33(2), 111140. Venetis Karin A., and Pervez N. Ghauri ( 2004), Service Quality and Customer Retention: Building Long Term Relationships, European Journal of Marketing, 38(11/12), 15771598. Villas Boas, J. Miguel (1994), Sleeping with the Enemy: Should Competitors Share the Same Advertising Agency? Marketing Science, 13 (2), 190 202. Wackman, Daniel B., Charles T. Salmon, and Caryn C. Salmon (1 986/1987), Developing an Advertising Agency Client Relationship", Journal of Advertising Research, 26 (December/January), 2128.
139 Wiener Yoash (1982) Commitment in O rganizations: A N orma tive V iew Academy of Management Review 7, 418 428. Weilbacher, William M. (1983), Choosing an Advertising Agency, Chicago: Crain Books. Wilson Koehler, and Swati Jantrania (1996) Understanding the V alue of a R elationship, Asia Australia Marketing Journal 2 ( 1 ) 5566 Woo, K a shing and Christine T. Ennew (2004) Business to B usiness R elationship Q uality: A n IMP I nteractionB ased C onceptualisation and M easurement, European Journal of Marketing, 38(9/10), 1252 1271. Young Louise and Sara Denize (1995) A C oncept of C ommitment: A lternative V iews of R elational C ontinuity in B usiness S ervice R elationships Journal of Business & Industrial Marketing 10 (5), 3546. Zack Michael H. (1999), Managing Codified K nowledge, Sloan Management Review 40(s ummer), 45 58. Zack M ichael H. (2003), Rethinking the Knowledgebased Organization, MIT Sloan Management Review Summer. Zeithaml, Valarie A., A. Parasuraman, and Leonard L. Berry (1985), Problems and Strategies in Services Marketing, Journal of Marketing, 49(2), 3346
140 BIOGRAPHICAL SKETCH Jun Heo earned his Ph.D. in m ass communication with an emphasis in a dvertising from the University of Florida at Gainesville in August 2010. He received his masters degree in a dvertising from Michigan State University at East Lansing in December 2003. He received his bachelors degree in journalism and m ass communication from Hanyang University in South Korea back in 1995. During his doctoral studies, he worked as an instructo r of record for 12 semesters and taught Advertising Media Planning and Principle of Advertising. Before joining the Ph.D. program he has worked as a media planner in multiple global advertising agencies such as Dentsu Young & Rubicam, Ogilvy & Mather, and Universal McCann. He has developed and implemented media plans for client firms including Chanel, Kimberly Clark, Citibank, Motolora, IBM, and INTEL. The goal of his research is to improve agency client relationships and to build a bridge between academi a and the advertising industry Beyond his relationship marketing research agenda, he has been pursuing research projects on understanding advertising media audiences and how media involvement influences advertising processing In addition, he is interested i n new media advertising business model such as mobile advertising His accomplishments in the research areas above include (1) two referred publications, (2) three manuscripts that are currently under review in the most prestigious journals and (3) fiv e research presentations The papers have been presented at conferences for the Association for Education in Journalism and Mass Communication (AEJMC), the American Academy of Advertising (AAA), and the World Media Economics and Management Conference. Begi nning fall 2010, he will serve as Assistant Professor of Advertising at the University of Southern Mississippi.