|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 INVOLVEMENT: THE INDIVIDUAL VERSUS TEAM SPORT EXPERIENCE By THOMAS S. MUELLER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCT OR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Thomas S. Mueller
3 To my mother who taught me t o care for others and honor God
4 ACKNOWLEDGMENTS The journey to earn a doctor of philosophy degree could not have been accomplished without several key individuals in my academic path. Thanks go out to Greg V enne, my high school journalism teacher, who underst oo d a wayward young boy who needed direction and guidance. My undergraduate college years where sha ped in many ways by David Ford Hanson, my friend and photo professor. T he gap from M aster of Business A dministration to doctoral work would never have happened if not for the encouragement of Dr. Susan Fagan and Dr. Gail Arch at Otterbein College. Severa l academic scholars have been instrumental in my success at the University of Florida. I wish to acknowledge Dr. Debbie Treise and Dr. Marilyn Roberts for opening the door to the University of Florida ; Dr. Michael Weigold for modeling the academic life; D r. Dan Connaughton for being a friend who cares; Dr. John Sutherland, who made time for long discussio ns that formed my scholarly thoughts; and to Dr. Christina Staudhammer, who had patience and invested time in demonstrating how to apply statistical metho ds.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 8 LIST OF FIGURES .............................................................................................................................. 9 ABSTRACT ........................................................................................................................................ 10 1 INTRODUCTION ....................................................................................................................... 12 2 LITERATURE REVIEW ........................................................................................................... 17 History of Involvement Theory .................................................................................................. 17 Involvement as a Measureable Construct .................................................................................. 20 Involvement and Advertising ..................................................................................................... 23 Involvement as a Construct for Sport Fans and Spectators ...................................................... 25 Involvement Measures for Active Sport Participa nts ............................................................... 29 Participation Measures for Active Sport Participants ............................................................... 33 Social Identity, Stress and Cognition in Sport .......................................................................... 33 3 METHOD .................................................................................................................................... 40 Survey Analysis ........................................................................................................................... 40 Behavioral and Descriptive Statistics ................................................................................. 41 Unidimensional and Multidimensional Involvement Scales ............................................ 42 Respondent Development ........................................................................................................... 42 Data Analysis (Hypotheses Testing) .......................................................................................... 44 4 RESULTS .................................................................................................................................... 47 Main Study Results ..................................................................................................................... 47 Participant Analysis ............................................................................................................. 47 Dependent Variable: Participation ...................................................................................... 48 Independent Variable: Laurent and Kapferer Involvement Scale .................................... 49 Independent Variable: Zaichkowsky Involvement Scale ................................................. 5 1 Hypothesis 1: Dimensionality of the Zaichkowsky Scal e ................................................ 51 Retesting Reliability for the Two Factor Zaichkowsky Scale .......................................... 57 Hypothesis 2: Dimensionality of the Laurent and Kapferer Sca le .................................. 57 Hypothesis 3a: Relationship between the Zaichkowsky Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ................................................................................................................................. 60 Hypothesis 3b: Relationship between Zaichkowsky Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ..................................................................................................................... 60
6 Hypothesis 4a: Relationship between Laurent and Kapferer Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ................................................................................................................................. 61 Hypothesi s 4b: Relationship between Laurent and Kapferer Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ..................................................................................................................... 62 Hypothesis 5a: Laurent and Ka pferer Scale will be more predictive than the Zaichkowsky Scale for intent to participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) .......................................................................... 63 Hypothesis 5b: Laurent and Kapferer Scale will be more predictive than the Zaichkowsky Scale for behavioral participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ............................................................... 63 Hypothesis 6 : Cognitive Experience for Team auto racing (Team sport) will be greater than Cognitive Experience for Ultrarunning (Individual Sport) ...................... 63 Hypothesis 7: Affective Experience for Ultrarunning ( Individual Sport) will be greater than Affective Experience for Team auto racing (Team Sport). ...................... 64 Hypothesis 8: Group Identification for Team auto racing (Team Sport) will be greater than Grou p Identification for Ultrarunning (Individual sport). ........................ 64 5 DISCUSSION .............................................................................................................................. 66 Evaluation of the Hypotheses ..................................................................................................... 67 Hypothesis 1: Dimensionality of the Zaichkowsky Scale ................................................ 67 Hypothesis 2: Dimensionality of the Laurent and Kapferer Scale .................................. 68 Hypothesis 3a: Relationship between the Zaichkowsky Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ................................................................................................................................. 68 Hypothesis 3b: Relationship between the Zaichkowsky Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ..................................................................................................................... 68 Hypothesis 4a: Relationship between Laurent and Kapferer Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ................................................................................................................................. 69 Hypothesis 4b: Relationship between L aurent and Kapferer Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ..................................................................................................................... 69 Hypothesis 5a: Laurent and Kapferer Scale will be more pr edictive than the Zaichkowsky Scale for intent to participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) .......................................................................... 70 Hypothesis 5b: Laurent and Kapferer Scale will be more predictive than the Zaichkowsky Scale for behavioral participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) ............................................................... 70 Hypothesis 6: Cognitive Experience for Team auto racing (Team sport) will be greater than Cognitive Experience for Ultrarunning (Individual Sport) ...................... 71 Hypothesis 7: Affective Experience for Ultrarunning (Individual sport) will be gr eater than Affective Experience for Team auto racing (Team Sport). ...................... 71
7 Hypothesis 8: Group Identification for Team auto racing (Team Sport) will be greater than Group Identification for Ultraru nning (Individual sport) ......................... 72 Implications and Limitations ...................................................................................................... 72 General ................................................................................................................................. 72 Practical Implications .......................................................................................................... 73 Limitations ........................................................................................................................... 75 Future Research ................................................................................................................... 77 APPENDIX A INDIVIDUAL SPORT INVOLVEMENT SURVEY .............................................................. 80 B TEAM SPORT INVOLVEMENT SURVEY ........................................................................... 87 C HAVITZ AND DIMANCHES ADAPTATION OF LAURENT AND KAPFERERS INVOLVEMENT PROFILE (IP) SCALE (AS APPLIED TO ULTRARUNNING PARTICIPATION)* ................................................................................................................... 95 LIST OF REFERENCES ................................................................................................................... 96 BIOGRAP HICAL SKETCH ........................................................................................................... 103
8 LIST OF TABLES Table page 4 1 Descriptive indicators for individual and team sport athletes ............................................. 48 4 2 Mean and standard deviations for group participation variables ........................................ 49 4 3 Reliability testing for Laurent & Kapferer subscales on individual and team sport data .......................................................................................................................................... 51 4 4 Reliability testing for the Zaichkowsky scale on individual and team sport data .............. 51 4 5 Varimax rotation exploratory factor analysis for individual sport athletes ........................ 54 4 6 Reliability testing for the Zaichkowsky scale on cognitive and affective dimensions ...... 57 4 7 Regression of Zaichkowsky cognitive and affective subscales on intent/plan to participate for team sport athletes ......................................................................................... 60 4 8 Regression of Zaichkowsky cognitive and affective subscales on behavioral participation for individual sport athletes ............................................................................. 61 4 9 Regression of Laurent & Kapferer multidimensional subscales on intent/plan to participate for individual sport ath letes ................................................................................. 61 4 10 Regression of Laurent and Kapferer multidimensional subscales on behavioral participation for team sport athletes ...................................................................................... 62 4 11 Fishers z test of independent Rs for Zaichkowsky and Laurent and Kapferer scales ..... 63 4 12 ANOVA comparisons for cognitive, affective and group response ................................... 65 5 1 Involvement regressions for intention and behavior ............................................................ 70 5 2 Correlation matrix for involvement dimensions .................................................................. 77
9 LIST OF FIGURES Figure page 4 1 Individual sport unidimensional scale model ....................................................................... 52 4 2 Team sport unidimensional scale m odel ............................................................................... 53 4 3 Individual sport model modified with cognitive and affective subscales ........................... 55 4 4 Team sport model modified with cogni tive and affective subscales .................................. 56 4 5 Individual sport multidimensional scale model .................................................................... 58 4 6 Team sport multidimensional scale mo del ........................................................................... 59 5 1 Ultrarunning individual sport involvement (Zaichkowsky scale) ....................................... 76 5 2 Ultrarunning individual sport intent to part icipate ............................................................... 76
10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INVOLVEMENT: THE INDIVID UAL VERSUS TEAM SPORT EXPERIENCE By Thomas S. Mueller August 2009 Chair: John Sutherland Major: Mass Communication The purpose of this study was to compare and contrast the involvement profiles of individual and team sport athletes. Using social cogni tive and social identity theories, the study explored the cognitive and affective processing traits of participant athletes, where cognitive processing was predicted to be higher for athletes competing in a team environment. Ultradistance runners were used as the individual sport study group, while rally auto racing participants were introduced as team sport athletes. This is a fruitful area for research as little academic research exists on interactive team relationships in motorsport. Zaichkowskys 10i tem unidimensional involvement scale and Laurent and Kapferers 15 item multidimensional scale, were used for involvement measures. Confirmatory factor analysis indicated that the unidimensional scale did not fit the single factor solution for either res pondent group, but did hold construct validity when factored into cognitive and affective dimensions. The multidimensional scale retain ed its factor loadings for both individual and team sport data sets. The multidimensional scale proved to be a signific ant predictor of individual ultrarunning sport intent to participate, and rally team sport behavioral intent The unidimensional scale held a statistically significant relationship to rally team sport intent to participate and individual ultrarunning beha vioral intent Contrary to the hypotheses tested, cognition was higher for
11 individual sport athletes over team sport athletes in unidimensional involvement scale data, and team sport athletes held a lower group identification score than individual sport a thletes. However, in multidimensional scale testing, team athletes exhibited a lower affective response than individual sport athletes as predicted.
12 CHAPTER 1 INTRODUCTION Sport as an entertainment product has become increasingly prominent in society. Th e fabled English football club Manchester United, a prominent global sports brand, touts mor e than 75 million fans (Hill & Vincent, 2006). Mega events such as World Cup soccer headline major sports stadiums. In the United States professional football, collegiate sports and motorsports are packing venues across the nation. Sport spending by marketers is also on the increase. One of the most common methods utilized to capitalize on sport is the sponsorship of athletes, events, teams, and facilities. U.S. sponsorship investment was projected to reach $14.93 billion in 2007, an 11.7 percent increase over 2006. Sports sponsorship was targeted for an increase of 10.8 percent, and attracted $9.9 billion (I nternational Events Group 2007). Television broadcast rights, Internet spending and equity investment in teams and facilities are also part of the new sport business model. Consulting firm Pricewaterhouse Coopers reports that spending related to large sporting events may boost revenues in the global market by 37 percent over the next five years to almost $141 billion. U.S. spending is projected to reach $69.1 billion by 2012 (Sport Business, 2008). Gibson (1998) classified consumers who partake in sport in three distinct categories: those who w atch sport e vents, visit sport related attractions, and those who participat e Many consumers have gone beyond the observation of sport and adhere to the credo of sport apparel manufacturer Nike and Just Do It. The National Center for Chronic Disease Prevention an d Health Promotion states that approximately 15 percent of U.S. adults engage in weekly vigorous physical activity, and approximately 22 percent of adults engage in weekly sustained physical activity ( Center for Disease Control 2008).
13 Physical participati on in sport occurs at different levels and in different environments. An example of a participant sport endeavor that offers individualism and independence is running. Running is on the increase among participants. In 2008, consumers spent $3. 1 6 billion on running products (Sporting Goods Manufacturers Association, 2009 ). And, a survey among American running clubs documented that 8.9 million runners competed in and finished road racing events in 2007 ( Road Runners Club of America 2008). Some athletes p ursue distances beyond the 26.2 mile marathon and complete solo 50 mile runs during the night then return home at dawn to assume work and family responsibilities (Karnazes, 2006) The sport of long -distance running is independent in nature and seldom rel ies on interaction or camaraderie with others. Most long-distance running competi tors focus inward, with the individual competing to achieve his or her best performance against the race course. During long races, you think about something for a while so metimes its a passing thought or random insight, other times an attempt to work out problems and then you move on. The thinking is not entirely linear. After a while, you accept this. After a while, you settle in. And then, the mind goes its own way slowi ng down, wandering more freely, giving itself over to the body, and finally, ultimately, to the heart (Toor, 2008, p. 8). Another select group of active athletes participat e in the more complex arena o f team competition. This study will introduce the tea m aspect of motorsport. Little research has been conducted on motorsport i nvolvement and the subject offers a fruitful area of exploration for the researcher. Lipsky (1979) developed a framework for understanding motor racing in a sport context, through t he impetus provided by the g rowing force of industrialism where t eams and sports
14 contribute to the cohesion of an industrialized society. Professional motorsport is controlled on a worldwide basis by the Federation Internationale de lAutomobile (FIA). I t is a multibillion dollar business that is comprised of diverse disciplines including gokarts, Formula 1 open wheel cars and closed (covered) wheel cars, such as NASCAR (Fuller & Myerscough, 2001). Motorsports attracts a participant base from varying socioeconomic levels of society. The National Speedway Directory lists 929 racing facilities in 50 states, but omits drag racing facilities, which would nearly double the total number. Motorsport r acing activity in the United States is representative of each states size and population is well distributed among the U.S. and is not restricted to regional interests (Barendse, 1983). An example of motorsport athletes competing in a team context is rally car auto racing. Rally racing is a timed motorsport competition on routes that include public or private roads, many times in secluded rural areas. Drivers, with the assistance of a navigator, pilot modified production or specially built road -legal cars. The decision -making process is highly integrated wi th crew members who offer assistance at designated check points on the race course. On many occasions crew members repair damage to the car, which impacts the overall time and performance of the team. The task of driving a rally car competitively is so complex that both the driver and navigator are necessary. Specific instructions pertaining to the course determine turns, terrain changes and the speeds required to stay on the proper pace throughout the event. Each member of a rally team can either cont ribute to, or minimize, anxiety or stress as part of racing competition ( Roberts & Kundrat, 1978) The focus in rally auto racing is on elapsed time and the overall finish against other teams. A personal best has little relevance in determining success. Eccles and Tenenbaums
15 (2004) social -cognitive psychological approach supports the premise that team members perform differently than individuals. Eccles and Tenenbaum determined that team competitors must acquire additional knowledge about teamwork and communicate within the team group, which places an additional load on cognitive demand as part of team decision -making and coordination. When competing in a sport under a team constraint, thoughts and actions are more group oriented. Social Identity Theor y posits that identification with a collective group can have a powerful impact on behavior. Organizational behavior is defined when a psychological group is more than the sum of its interpersonal relationships. Social identification is a mechanism that a llows individuals to conceive of and form loyalty to an organization ( Ashforth & Mael, 1989). Social identity theory supports the premise that individuals and groups (teams) will use different decision -making processes. In order to study and explore the t raits of individual and team competitors, the measure of involvement proves valuable. In their 1991 study on recreation and t ourist events, Dimanche, Havitz and Howard (1991) stated that further work is needed to conceptualize and refine the understandin g of the various dimensions purported to underlie the construct of involvement [especially with] other populations and other activities (p. 63). Involvement has been tested as an appropriate measure in sport participant research. For example, Havitz and Dimanche (1990) applied Laurent and Kapferers (1985) measurement scale to participants in an international track and field competition and confirmed both reliability and validity within the construct of leisure activities. Wiley, Shaw & Havitz (2000) examined the gender aspects of leisure involvement. Ajzen and Driver (1992) applied the theory of planned behavior to leisure intentions and included involvement in their model. McGehee, Yoon &
16 Cardenas (2003) utilized an adaptation of Zaichkowskys (1994) 10item semanticdifferential scale in a study that measured involvement among recreational runners in North Carolina. Sport scholars have operationalized involvement to measure the level of connectedness, yet little research exists that examines the validity of involvement scales in comparisons of individual and team sport. T his study will explore the relationship between involvement and participation in individual (ultrarunning) and team (rally car racing) sport utilizing Zaichkows k ys (1994) unidi mensional 10-item scale and Havitz and Dimanches (1990) sport adaptation of Laurent and Kapferers 15 item multidimensional scale. The construct validity of the Zaichkowsky and Laurent and Kapferer scales will be tested through confirmatory factor analysi s R egression analysis will determine if the scales are predictive of participation in each sport. A contribution will be made to the field of involvement study by exploring the construct of involvement within the individual versus t eam sport experience. With an increasing magnitude of investment in sport, understanding participant involvement is beneficial to advertisers, sport managers and investment brokers. The indicators of involvement may serve as a predictor in building effective marketing and adv ertising campaigns. From an individual and team point of view, event promoters and product advertisers may gain greater insights into the predicators for event participation Advertising campaigns targeting key involvement indicators may lead to increased event entries and higher sales for aftermarket products and services. Continuing study in the refinement of involvement models in sport presents fertile ground for academic research.
17 CHAPTER 2 LITERATURE REVIEW History of Involvement Theory The concept of involvement is broad in scope, having been applied to individuals, a mass medium, social issues, consumer products, brands, political elections, or interactions within these categories. The concept has intrigued academics, marketers and advertising ex perts for several decades. The concept of involvement has gained increasing importance in the academic study of consumer response and mass communication. Salmon (1986) identified that involvement has been conceptualized as an individual state (the interes t of an individual in a stimulus); as a response to a stimulus (consequences of a stimulus for an individual); and as a characteristic of a product, issue or situation (arouses concern in most or all individuals). From its inception in post World War II re search, scholars have extolled the value of involvement measures in studies that explored individual self -identity, the mass media, social issues and politics, products and brands. Involvement has reached such heightened perception that one researcher stated involvement is the greatest thing since sliced bread (Rothschild, 1984, p. 216). The historical roots of involvement were articulated by Sherif and Cantril (1947) through their work in egoinvolvement. Later work examined the persuasive effects of m edia on attitude change. Krugman (1965, 1966, 1971) took a cognitive approach to involvement and argued that contrary to popular belief, individuals held a low personal involvement with media. Several theoretical approaches have been applied to involvem ent research. Ward (1965) used social judgment theory to examine ego involvement as an antecedent to new racial attitudes. Other scholars approached involvement through the elaboration likelihood model of
18 persuasion, which refers to the likelihood that individuals will engage high involvement and issue relevant thinking, rather than low involvement thinking where focus is on cues rather than message processing (Cacioppo & Petty, 1979; 1984). Ray (1973) elaborated that different levels of involvement led to varying hierarchies of effects in consumers. He proposed that three components led to these effects: a cognitive component, an affective component and a conative component (which referred to changes in behavioral intention or an actual change in beha vior). Zaichkowsky (1985) stated that involvement is defined as a persons perceived relevance of the object based on inherent needs, values and interests (p. 342). She developed unidimensional scales to measure consumer involvement for both product cat egories and advertising, entitled the Personal Involvement Inventory (PII). Zaichkowsky stated theorists proposed that consumers actively search for information to make informed choices. She noted that for some decisions, consumers may not have active i nformation processing, and may in fact exhibit low involvement (versus high involvement) consumer behavior. Under high involvement, an individual scrutinizes incoming information, while under low involvement, focuses on cues rather than the information pr esented. While Zaichkowsky pursued a unidimensional construct with a single measure of perceived importance, Rothschild (1979) concluded that no single indicator of involvement was sufficient as a predictor. Laurent and Kapferer (1985) opted to follow R othschilds presupposition and developed a multidimensional scale that included five facets for involvement. Their involvement scale encompassed perceived importance, perceived risk of negative consequences, perceived risk and probability of making a mist ake, symbolic sign value and hedonic value.
19 Laurent and Kapferer traced the genesis of product involvement to Sherif and Cantrils (1945) early work in the 1940s, which drove the definition of psychology of attitudes in relationship to standardized values norms or other criteria of conduct through perceptual and judgmental processes. Later Sherif and Cantril (1947) incorporated the definition and measurement of ego involvement as applied to social attitudes and identifications. The two authors argued that attitudes are consistent with ego, and when attitudes are called for, an individual becomes personally involved in the situation. The ego involvement model states that there is a direct association between a persons attitude related to self -esteem or central values and that individuals ego involvement in the situation. Funk, Ridinger & Moorman (2004) point to social judgment theory (Sherif & Hovland, 1961) as the precursor to involvement theory. Ward (1965) used social judgment theory and ego involve ment to study racial attitudes. Similar work was later applied to consumer behavior research to better understand purchase behavior related to consumer goods. Krugmans (1966) early research focused on the measurement of advertising involvement and is a mong the most frequently cited in advertising and consumer behavior research. In contrast to other scholars of his day, Krugman believed that involvement was measured between an individuals personal life and a stimulus, and was measured through the numbe r of bridging experiences, connections, or personal references made per minute. Krugman determined that consumer involvement with advertising was highest when attention was directed at editorial content; that advertising consistent with editorial content maintained the highest involvement; magazine advertising tends to hold a higher involvement over television with high involvement products; involvement is less predictable with low involvement products, while high involvement products share a similar level of interest.
20 Ray (1973) elaborated on Krugmans work and suggested that behavioral intention was classified within cognitive, affective and conative (purposeful) components. Though this hierarchies of effects model has been adapted to classify possible ch ange processes, it does not provide a conceptual definition of high or low involvement (Salmon, 1986). Involvement as a Measureable Construct Involvement literature indicates that there are varying and somewhat ambiguous views on the concept of involvement and the definition on how to operationalize and measure it. The development of measurement instruments for involvement spans several decades. During the past 40 years, at least 23 measures have been developed, incorporating nine manipulation checks. Sc ales are generally based on a semantic differential or Likert type formats. The number of items range from three to 33 with scale point ranging from five to seven (OCass, 2000). Researchers have explored the involvement construct and have developed sever al applied measurements. Sherif and Cantril (1947) explored ego involvement as related to social attitudes Sherif and Hovland (1961) studied involvement through social judgment theory. Krugman (19 6 6 1967) developed early research regarding the effect o f involvement in advertising Rothschild (1979) differentiated between situational involvement, such as purchase location; enduring involvement, which indicates an ongoing concern for the purchase situation; and response involvement, an indicator for the c onsequences related to the inner state of being involved. Bloch and Richins (1983) created a model for product importance, where enduring involvement prompted ongoing responses and situational involvement prompted task-related responses. Lastovicka and Ga rdner (1979) asked subjects to evaluate 14 products and measured importance, commitment and affect. The analysis revealed low involvement, high involvement, and enthusiast products (that express the individuals hobby or special interest). Research has
21 c onceptualized involvement as a multidimensiona l construct. Funk, Ridinger & Moorman (2004) point to the work of Laurent and Kapferer as highly influential, based on their conceptualization of involvement as a multidimensional construct. Laurent and Kapfe rer noted five antecedents for involvement that included perceived importance, perceived importance of negative consequences (risk) perceived probably of a poor purchase decision (risk probability) symbolic or sign value and the hedonic value or pleasur e provided. However, they concluded that an involvement profile should include the perceived importance of consequences in a mispurchase, the probability of a mispurchase, the hedonic (pleasure) value of a product and the perceived sign value of a product class. The research by Funk et al. indicated involvement does not in itself create expected differences in behavior. Individual measurement of the antecedents in the involvement profile led to a better defined view of behavioral outcome. A key distincti on in the semantics of involvement literature was presented by Traylor and Joseph (1984). Terminolo gies such as high involvement products, or low involvement mediums were disputed in the sense that stimuli are not inherently involving or uninvolving. It was argued only consumers can be involved, not the products or mediums that are being used as stimuli. In an effort to create a general instrument to measure involvement, Traylor and Joseph tested a 22 item scale, reduced it to ten items, and then factor ed the results into a six item scale that tested reliability and validity. The scale was developed with an emphasis on the relationship of a product to the consumer. Zaichkowskys (1985) Personal Involvement Inventory (PII) was initially developed to test consumer involvement with products and was later adapted (1994) to measure advertising response. Zaichkowsky stated that involvement is defined as a persons perceived relevance of the object based on inherent needs, values and interests (p. 342). Th e PII was developed
22 utilizing four data sets with undergraduate students, two data sets with Master of Business Administration students, and two additional data sets with clerical and administrative staff members. Zaichkowsky stated that Likert scale item s were problematic, based on a differentiation in response to frequently purchased goods versus durable goods. Her initial PII was developed using a 20 item semantic differential scale to measure product involvement; it was later tested for reliability an d validity and was reduced to a 10 item scale for advertising measurement. Jensen, Carlson & Tripp (1989) investigated involvement research by Bloch (1981) Lastovicka and Gardner (1979) Laurent and Kapferer (1985) Slama and Tashchian (1985) Traylor and Joseph (1984) and Zaichkowsky (1985). In order to test the dimensionality of involvement across multiple products, Jensen et al. utilized exploratory and confirmatory factor analysis for the variables related to Importance, Knowledge, Preference and Com mitment. They summarized that involvement may be multidimensional between products and when collapsing across products. However, dimensions may not be congruent and will not allow for meaningful comparisons between products or situations. Laurent and Kapferer (1985) noted the work of Houston and Rothschild (1977) who made a distinction between enduring and situational involvement. Rothschild (1979) stated that an individual may purchase low -priced brand alcohol based on low enduring involvement, but migh t employ high involvement with the purchase of a specific brand for an important gue st Other scholars have identified psychological involvement as a state of motivation, arousal or interest with regard to a product, an activity or an object (Rothschild, 1 984). Mitchell (1979) defined involvement as an internal variable that reflects the amount of arousal, interest or drive evoked by a particular stimuli or situation that mediates consumer behavior.
23 Traylor (1981) stated that product involvement involves t he measurability of how a product is more or less central to an individuals life, personal attitudes, sense of identity, and relationship to the rest of the world. Traylor differentiated between brand and product involvement and stated that brand involve ment may not vary directly based on whether the product involvement is high or low. Involvement with a product class and commitment to a brand is more often observed in a purchase context. Involvement has also been defined as a state of mental readiness t hat influences the allocation of cognitive resources to a consumption obje ct, decision or action (Park & Mittal, 1985). Andrews, Durvasula & Akhter (2001) suggested that involvement indicates a state of arousal with intensity, direction and persistence pr operties. In this model, the state of arousal determines how a consumer responds to stimuli, such as advertisements or products. Involvement intensity is defined through the degree of arousal or preparedness the consumer generates, with respect to a goal related object. This intensity refers to the arousal level and not the actual processing or behaviors that are exhibited through the involvement. Consumers utilize involvement to conceptualize through attention and processing strategies, personal and si tuational involvement, audience and process involvement, and enduring product involvement. Changes in cognitive responses and attitude change may be related to involvement, but an increase in these dependent variables does not always indicate that involvement has been positively manipulated. Involvement and A dvertising Consumer involvement as it applies to products is now widely recognized as a major component and variable in a dvertising strategy (Laurent & Kapferer, 1985). Krugman (1966) stated that invo lvement was defined as the factor of direct personal experience (p. 583). Researchers have further defined involvement to refer to the relationship between a consumer
24 and a product. The premise of involvement is that certain products are more or less c entral to an individuals life, his/her attitudes, sense of identity, and relationship to the world (Traylor, 1981). In her review of the construct of involvement, Zaichkowsky (1985) stated the focus of product involvement is based on how relevant or impor tant the person perceives the product category. Three variables precede involvement : the characteristics of the individual, physical characteristics of the stimulus, and the varying situation under which the involvement is engaged. A conceptual view of product involvement focuses on the individual and stimulus of communication, as it relates to the importance of the product class, differences in product attributes, and preference to a particular brand. The two key factors that influence high and low inv olvement in product interaction are personal importance or relevance, and differentiation of alternatives. OCass (2000) posited an alternative view and suggested that product importance and relevance are a subset in the construct of involvement. They a re part of the mechanism, but do not constitute involvement alone or involvement per se. Advertising or purchase decisions may be deemed important and relevant by consumers, without an involvement in product. OCass posited that a more enduring approach will focus on products, purchase decisions, advertising and consumption. Celsi and Olson (1988) created implications for marketers when he stated that levels of involvement can affect consumer attention and comprehension processes, with relevance identifie d as a primary indicator of the level of involvement. Relevance determines how salient the concept is perceived to be and how it supports the pursuit of achieving personal goals and values. Situational sources of personal relevance (SSPR) and i ntrinsic so urces of personal relevance (ISPR) can be factored into the measurement of involvement with a message or
25 concept. These factors can contribute as a joint function to a felt involvement in attention and comprehension processes. Felt involvement is the se lf relevance of the goals and values that are activated in a situation and by the strength of their association with salient objects and actions in that situation (p. 223). Celsis research indicated that self relevant knowledge, which is activated and ex perienced, leads consumers to devote more attention to advertisements and to exert more cognitive function over the advertising content. Motivated consumers those with more felt involvement devote more attention to relevant information. Such involvemen t influences the amount and focus of the processing effort, making consumers domain knowledge more influential. Automatic processes are overtaken by more controlled comprehension processes that enhance personal relevance, importance, interest and physiol ogical arousal. Within the advertising environment, engagement represents measurement of the contextual relevance within which the brands messages are framed; a subsidiary affect of engagement is the measurement of involvement. There is a mediating effe ct between message involvement and message believability. When a consumer engages a message, it has a strong influence on the level of involvement that subsequently occurs (Wang, 2006). Involvement as a C onstruct for S port F ans and S pectators Select academic research has focused on sport fans and their involvement. Kerstetter and Kovich (1997) used involvement profile items from a scale devised by Havitz and Samdahl (1994) for a study of spectators at Division I Womens Basketball events. Fisher and Wake field (1998) measured the domain involvement of professional hockey fans with a 5item adaptation of Zaichkowskys scale. Gwinner and Swanson (2003) continued to work in domain involvement and utilized a 3 item adaptation of Fisher and Wakefields scale t o measure the response of fans at a NCAA Division 1 football game.
26 A pertinent extension of involvement in sport includes the study of sponsorship and its inherent value to advertisers. Pham (1992) measured the felt involvement of soccer fans by modifying Zaich k owskys 20 item advertising scale, inserting short sentences within the original semantic differential word pairings. McDaniel (1999) studied the effects of advertising in sport events and included Zaichkowskys 10-item product involvement inventor y to measure product category and sporting event response. Alternative research by sport academics brought forward new scales that attempt to measure fan motivation (Wann, 1995) and points of attachment (Trail et al., 2003). Wann developed his scale to measure motivations in involvement (p. 380), while Trail et al. (2003) devised the Motivation Scale for Sport Consumption (MSSC) and the Point of Attachment Index (PAI). Kerstetter and Kovich (1997) continued the research of Dimanche Havits & Howard (19 91) and tested a womens basketball audience with a scale that included all five of Laurent and Kapferers original factors. Once again, the IP scale demonstrated validity and reliability. The results revealed two facets of involvement enjoyment (which included items related to importance and pleasure, considered to be synonymous in leisure based activity) and risk probability (items related to sign value). Self -expression and sign value are of high importance to individuals engaged in leisure based act ivities. Shank and Beasley (1998) developed measurements that explored the psychological aspects of fan involvement. These measures included media habits, exercise habits as related to a specific sport and participatio n and demographics. An eight item s ports involvement inventory was utilized across a 7 -point Likert scale. The inventory had respondents rank their interest in sports from boring to exciting (affective) interesting to uninteresting (affective)
27 valuable to worthless (cognitive) appealing to unappealing (affective) useless to useful (cognitive) not needed to needed (cognitive) irrelevant to relevant (cognitive) and important to unimportant (cognitive). The profile of the highly involved sports fan was young, single and male. Age, mar ital status, and income levels did not significantly affect involvement levels. Capella (2001) identified emotion as a key measurement, where a team loss can bring emotional pain to an engaged sports fan. His Fan Behavior Questionnaire was based on a cong lomeration of literature regarding sport spectator research and consisted of 31 items based on a 5 point Likert scale, with options ranging from never to always. Funk, Ridinger & Moorman (2004) combined fan behavior and involvement literature to constr uct a Team Sports Involvement Model. It was developed to assess the relationships a mong 18 antecedents and four facets of involvement : attraction, risk, centrality and self -expression. A review of more than 50 sport involvement studies indicated that sp ort involvement, approached as a multidimensional construct, resulted in stronger content and face validity. Team sport involvement was defined as a psychological state of motivation, arousal or interest in a team and related activities. Armstrong (2007) used a modified version of Traylor and Josephs (1984) sixitem product scale to measure for involvement among fans at a minor league hockey game. Armstrong determined that symbolic communicative properties exist among sport consumption F irst time spe ctators at a live sporting event were more likely to infer a meaning to other attendees based on the type of sport attended R epeat attendees were more likely to construe symbolic meanings to those who regularly attended the minor league hockey games. Sy mbolic interactionism was seen as an individuals involvement with a select sports group, and this interactionism affected sport consumption frequency. Factors influencing consumer attendance
28 were composition of the game audience, purchases made at game events and general consumption of other leisure related activities. There appeared to be a significant difference in the motivations of repeat and first time attendees. Repeat attendees view ed the event as more meaningful, a more symbolic self -expressio n, an identity affirming activity, and give a higher symbolic meaning as part of the sport consumption experience. Thus, indicating that there is a symbolic communicative role in the social psyche of sport fans. Other sport academics have created alternat ive multidimensional scales under the construct of motivations in involvement. The Sport Fan Motivation Scale (SFMS) was developed by Wann (1995). It measured eight dissimilar motives : eustress (desirable stress that can bring contentment), self -esteem, escape, entertainment, economic, aesthetic, group affiliation and family needs. These motives were divided into 38 items, which were incorporated into 8 point Likert -scale semantic differential questions. Wann initially developed and tested the SFMS thr ough two individual studies, constructed a 23item Likert scale measure, and later collaborated with Schrader and Wilson (1999) to expand on SFMS research, where and a three study project was developed. Wann (1995) indicated that the highest levels of mot ivation were found on the entertainment subscale, while the lowest levels were found on the economic subscale; fandom was not related to economic and family motivation. Individuals with a preference for an individual sport reported higher levels of ae sthetic motivation. These sports allow the fan to focus on the performance of a single individual, increasing the fans involvement in beauty, grace and artistic expression of the athlete. Fans of team sports appear to report higher levels of eustress ( desirable stress) and escape motivation (attractive to individuals seeking stimulation and diversion). Of particular interest is
29 the suggestion that sport fans appear to exhibit similar responses to that of athletes (i.e. anxiety, affect and biased predic tions ) Sport fans may be both intrinsically motivated ( prioritizing enjoyment of sport over potential rewards and less upset about poor team or athlete performance ) and extrinsically motivated ( distraught by poor performance, which may hinder extrinsic r ewards such as economic gain and self -esteem ) (Wann, 1995) An important differentiation between fans and spectators was identified by Sloan (1989) and Trail et al. (2003). Spectators merely watch and observe, while fans are enthusiastic devotees of a giv en diversion. To differentiate between these two respondent groups, a Point of Attachment Index (PAI) included seven subscales that focused on identification with players, coaches, the community, the sport, the university, the team, and the level of the s port (e.g. college versus professional). Each subscale was represented by three items, measured through a 7 point semantic differential questionnaire. Conversely, Trail et al. created a Motivation Scale for Sport Consumption (MSSC) that measured motivation for achievement, aesthetics, drama, escape, knowledge, physical skills and social interaction. These motives were segmented into three categories, for successful teams (achievement), for unsuccessful teams (aesthetics, physical skill of the athletes, eu stress/drama, and knowledge) and for both spectator and fan alike (escape and social interaction). Vicarious achievement was associated with team, coach, community or university A esthetic or dramatic qualities were associated with sport in general, or level of the sport. Social support and escape were associated with both spectators and fans. Involvement M easures for A ctive S port P articipants Early leisure research focused on the behavioral consequences of leisure involvement, as recreational activiti es and hobbies tend to stimulate a wide variety of participation types and levels. McIntyre (1989) incorporated Laurent and Kapferers (1985) factors, to develop an
30 instrument to measure involvement in a recreation context at a beach camping setting in so utheast Queensland, Australia. The factors utilized were importance, enjoyment (which corresponded with Laurent and Kapferers factor called pleasure), and self -expression (which corresponded with Laurent and Kapferers sign). Risk elements were not included, but centrality to lifestyle was added. McIntyres study indicated that attraction, self -expression and centrality best represented involvement in a recreational setting. Attraction refers to the perceived importance of an activity or product and p leasure (hedonic value) derived from participation or use S elf -expression represents the unspoken statement that purchase or participation conveys about the individual Centrality encompasses interaction with friends and family, and the central role of a n activity in an individuals life. McIntyre identified the three subscales as a predictor of enduring involvement, where centrality was a key indicator of involvement in a recreational setting. Dimanche et al. (1991) introduced the Laurent and Kapferer in volvement profile (IP) scale to the sport and leisure industry and tested its reliability and validity traits within recreation and tourism (Appendix A). Athletes in track and field competition were surveyed, and the Cronbachs Alpha reliability coefficients ranged from .80 to .96 within each of the five subscales within the IP. A factor analysis of the 15 item scale clustered into four subscales that were similar, but not identical, to the original Laurent and Kapferer subscales: sign, importance pleasure risk probability and risk consequence. The Dimanche research led to a new definition for involvement in sport, which they defined as a psychological state of motivation, arousal, or interest between an individual and recreational activities, tourist de stinations, or related equipment in one point in time,
31 characterized by the perceptions of the following elements: importance, pleasure value, sign value, risk probability, and risk consequences (p. 237238). Other scholars incorporated involvement indi cators within research directed at leisure activities and recreation. Ajzen and Driver (1992) explored behavioral intention among college students who participated in beach activities, jogging, mountain climbing, boating and bicycling. The study included an involvement scale that included questions on how much time participants spent obtaining information about leisure travel destinations how upset they would be if unable to participate in a given activity, how often they read publications related to the ir activity and what they thought their skill level was as related to the activity. High involvement reflected positive attitudes toward the leisure activity, which was correlated with increased performance in the activity. Ajzen and Driver concluded th at leisure sport participants evaluate an activity based on instrumental costs and benefits and the positive or negative feelings the activity engenders. Findings suggest that perceived cost may be offset by positive feelings, and negative effects may be offset by perceived benefits. The complexity of the involvement construct within individual sport participants was demonstrated through a market segmentation study conducted by Kyle, Kerstetter & Guadagnolo (2002) Athletes competing in a 10kilometer roa d race were studied as prototypical clusters based on involvement profiles and behavioral indicators The authors used Laurent and Kapferers 15 item involvement scale and tested the dataset using confirmatory factory analysis. Goodness of fit statistics indicated that the five subscales in the instrument did not work within the model. A subsequent exploratory factor analysis was conducted. An oblique rotation indicated three factors One, entitled enjoyment captured eight of the scale items. The second factor entitled risk consequence included three items. The third factor sign value, captured
32 three items. One item measuring risk probability was deleted due to low item to total -correlation value. The clusters were reported, segmenting partici pants into enthusiast, performer and recreationist groupings. Enthusiasts were reported to be high involvement with heavy loadings on the sign value items. Performers measured high on enjoyment, risk consequence and sign value facets. Recreationists hav e low responses to enjoyment and sign value facets and associated low risk outcomes within their participation group. The study concluded in stating that the risk indicators in the scale might be refined to include risk of physical injury, failure of achi eving personal goals, and performing poorly in front of peers. McGehee, Yoon & Cardenas (2003) utilized Josiam, Smeaton & Clements (1999) adaptation of Zaichkowskys 10 item unidimensional scale to measure involvement in recreational runners in North Carol ina. The unidimensional scale was utilized based on ease of administration, abbreviated length, and the opportunity to replicate results to previous measurements of involvement, and to future studies. Zaichkowskys premise that measurement is at the prod uct level rather than the brand level was applied to survey questions. Travel to a non -descript road race was considered, rather than a particular named (brand) road race. Based on the fact the study was exploratory in nature, McGehee, Yoon & Cardenas ar gued that a unidimensional scale provided the best fit in the context of an under -studied group. A response bias was also eliminated, where the answers given by respondents d id not reflect their true beliefs. Survey participants stated they lost interest in answering questions in a multi dimensional scale, or said the questions were redundant. The McGehee et al. (2003) study suggested that runners with a high level of involvement spent more money on running and travel to events and process advertisements more thoroughly than individuals with low involvement. It was posited event marketers should provide explicit
33 information for high involvement runners, but also serve medium involvement runners with highly accessible event information, as medium involveme nt runners will process information more passively. Participation Measures for Active Sport Participants In their study with tenkilometer running participants, Kyle, et al. (2002) stated that conceptualizations in both social psychological and behavioral measures were important in examining involvement among sport athletes. Behavioral measures included annual participation in the number of running events and previous participation in the ten kilometer running event used for data collection. Behavioral ch aracteristics were also measured and utilized in the McGehee, et al. (2003) research with North Carolina runners. Behavioral variables tested were frequency of participation, travel and expenditures on events. The test instrument included descriptive sta tistics to profile demographic information. Included were years as a running participant, years participating in running competition, number of days invested in running each week and number of miles run each week. Ajzen and Driver (1992) investigated the application of the theory or planned behavior to leisure choice. To capture specific beliefs regarding leisure behavior, the authors collected data on intent to participate. Two items were used, one soliciting information on respondents intended plan to engage in specific leisure activity in the next six months. The second item questioned the respondents attempt to try and engage in the activity in the next six months. Social I dentity, S tress and C ognition in S port The solitude of the endurance runn er allows for introspection and autonomy in decision making. A n all -night 50 mile training run presents solitude and a reprieve from a n ormal interactive lifestyle (Karnazes, 2006) It had been hours since I last had contact with humanity,
34 and the night a ir was silent and warmId watched the signal indicator on my cell phone diminish to the point of no reception, severing my contact with the outside worldI was able to enjoy the tranquility of my surroundings in an otherwise frenetic life. At times I fou nd myself mesmerized by the full moon illuminating the hillsides. (p. 3 4). The individual sport of endurance running heightens the affective indicators of involvement. Affect is the emotional and motivational constructs that contrast with cognitive pro cessing such as emotion, mood, arousal, incentives, needs, motives, conditioni ng and reinforcement (Eagley & C haiken, 1993). In endurance sport, long periods of time are spent alone, outside the confines of a performance based on the actions of other comp etitors or team members. E ndurance running removes a competitor from interaction with humanity and sharpens the senses towards affective states such as the tranquility of surroundings and attraction to the lure of a full moon (Karnazes, 2006) Unlike th e affective processing promulgated by individual sport, team sport introduces complex cognitive processing requirements through multiple interactions with team members. Eagley and Chaiken (1993) define cognitive involvement as a stimulus response link... mediated by a sequence of mental operations or cognitive processes (e.g. encoding, interpretation, storage, retrieval) (p. 389). The team sport experience in rally car competition is complex. A team driver, course navigator and mechanical support crew m ust coordinate their actions over a course that traverses hundreds of miles, over multiple days. Logistics and communications are paramount in a contest where several elapsed minutes can constitute the difference between victory and defeat. The team spor t environment demands a multiplicity of decisions. Organizational stress, social
35 cognition and social identity are prevalent in the cognitive thought process in team sport and are predictors of a high cognition load in involvement processing. Woodman and Hardy (2001) identified organizational stress in sport as the interaction between the individual and sport organization and the individuals cognitive appraisal of the situation within the work environment that is central to (this) organizational stress process (p. 208). Woodman and Hardys study operationalized organizational stress to include team atmosphere, which included tension between athletes. New team members can experience heightened operational stress when attempting to integrate into a tea m that has been competing for a long period of time. One athlete interviewed as part of a qualitative survey stated, it was just that our mind wasnt on what we were supposed to be concentrating onwe were too busy thinking about things that were going o n amongst ourselveswe werent focusing 100 percent (p. 225). The subjective experiences of the individual are complicated by the broader organizational, social, political and cultural environment of team sport. Social cognition within a team context inc orporates coordination, communication and organization among team members to extend beyond the sum of the cognitive properties of the constituent members (Eccles & Tenenbaum, 2004). While knowledge and cognition clearly affect the individual performer, te am performance must incorporate coordination, which introduces additional cognition requirements for team members. Rally car entrants share knowledge and teamwork through communication and an increased cognitive demand is required to achieve team coordina tion. Cognitive stress in team sport is also magnified through a competitors social identity to the team. Social identity theory states that individuals tend to classify themselves and others into various social categories. According to Ashforth and Mae l (1989) social classification
36 cognitively segments and orders the social environment, which allows the individual a systematic means of definition among the social group or team. Perceived organizational prestige affects self -esteem and is related to or ganizational identification. T he personal identity of the individual does not always find compatibility within the personalities of other team members. A rally car team may attempt integration of such identities, but it does not entirely resolve conflict s. Team competitors undergo a cognition process where they must order, separate or buffer their own identities from that of their teammates. Turner (1982) stated that social identity is the cognitive mechanism which makes group behavior possible (p. 21). While individual competitors are free to compete within their own constraints, team competitors must adjust to an increased cognitive load to perform within the team construct. Hypothesis and R esearch Q uestions Zaichkowsky (1985) cited the previous research of Bloch and Richins (1983) in defining the antecedent factors of involvement as (1) the characteristics of the person, (2) the characteristics of the stimulus and (3) the characteristics of the situation. Zaichkowskys semantic -differential sca le was designed to measure the state of involvement, unlike Laurent and Kapferers Likert -type scale, which measures the antecedents of involvement. The Zaichkowsky scale was successfully tested for content validity (range of meanings within a concept), r eliability (stability of the scale over time), criterion -based validity (how the measures relate to external criterion) and construct validity (how a measure related to other variables in the scale within theoretical relationships).
37 Laurent and Kapferer (1 985) based their conceptualization of involvement on a multidimensional construct, citing the work of Rothschild (1979) and stating that no single indicator of involvement could satisfactorily describe or explain, or predict the antecedents to, involvement Laurent and Kapferer suggested that an involvement profile would more fully represent the relationship between a consumer and product category, and noted five antecedents for involvement: perceived importance, perceived sign value, perceived pleasure va lue, perceived risk of negative consequences and perceived risk of making a mistake. This study will test (1) the dimensionality and internal reliability of the Zaichkowsky and Laurent and Kapferer scales within participatory sport at the individual and te am levels and (2) the relationship between involvement and participation within the individual and team sport experience. Therefore, H1: The Zaichkowsky scale will be unidimensional for individual and team sports H2: The Laurent and Kapferer scale will be multidimensional (perceived importance, perceived sign value, perceived pleasure value, perceived risk of negative consequences and perceived risk of making a mistake) for individual and team sports H3a: A relationship exists between the Zaichkowsky S cale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H3b: A relationship exists between the Zaichkowsky Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto ra cing (Team sport). H4a: A relationship exists between the Laurent and Kapferer Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H4b: A relationship exists between the Laurent and Kapferer Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H5a: The Laurent and Kapferer Scale will be more predictive than the Zaichkowsky Scale for intent to participate in (1) Ultrarunning (Individual sp ort) and (2) Team auto racing (Team sport).
38 H5b: The Laurent and Kapferer Scale will be more predictive than the Zaichkowsky Scale for behavioral participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). When Zaichkowsky initially factored her scale, it indicated one general factor that captured interesting, appealing, fascinating, exciting and involving. This general factor was associated with relevant emotions. A minor component captured important, relevant, valuable, means a lot and needed. This represented a rational or cognitive theme. Zaichkowsky stated her scale represents cognitive and affective dimensions. The two components are not independent but may be useful in determining the type of involvement demonstr ated by consumers responding to advertisements. The scale was projected to easily relate across product categories and can be appropriate to other domains (p. 342). Zaichkowskys scale tests for the state of involvement without antecedent influence and also demonstrates validity across the affective and cognitive subscales. Factor analysis of the data generated from the Laurent and Kapferer scale proved discriminant validity and indicated four specific factors : Imporisk (the perceived importance of th e product and the perceived importance of the consequences of a mispurchase) (p. 44); the subjective probability of a mispurchase; the hedonic (pleasure) value of the product class; and the perceived sign value of the product class. It was determined tha t no single facet alone catches the richness of the relationship between a consumer and a product class (p. 45). The think -feel dimension (p. 52) of the scale was related to the weight of the sign and pleasure value subscales. When sign or pleasure s ubscales show a strong positive or negative response, respondents are operating in a higher affective mode. A social -cognitive conceptual framework states that team members must participate in a high level of group communication and must cope with a highe r cognitive demand. The demand
39 for higher cognition among team members may lessen the involvement in affective measures. Social Identity Theory states social identification within a collective group is a conceptual cognitive construct and is not necessa rily associated with affective states. Therefore, H6: Cognitive Experience for Team auto racing (Team sport) will be greater than Cognitive Experience for Ultrarunning (Individual Sport). H7: Affective Experience for Ultrarunning (Individual sport) wil l be greater for than Affective Experience for Team auto racing (Team Sport). H8: Group Identification for Team auto racing (Team Sport) will be greater than Group Identification for Ultrarunning (Individual sport).
40 CHAPTER THREE METHOD S urvey A nalysis S urvey research is frequently used as a tool for data collection and analysis in the social sciences, and proves beneficial for descriptive, explanatory and exploratory studies. It has been deemed the best method of data collection when collecting data fro m a population too large for direct observation. Variables to be studied are operationalized through written survey questions, inclusive of descriptive statistics, behavioral characteristics, explanatory items and exploratory items. Standardized surveys are generally weak on validity, as respondents must conform to Likert scale or semantic differential questions, which are approximate indicators of the actual opinions. However, surveys generally hold a high degree of reliability, through a standardized s timulus that reduces variability and produces similar data throughout repeated observations. (Babbie, 2007). New technologies have propagated opportunities for additional channels of survey research, specifically the use of the Internet for online surveys. One contingent of modern researchers contends that online surveys will replace traditional survey methods, such as personal interviews, mail surveys and telephone surveys. Others remain cautious, based on issues with the representativeness of online s urvey sample groups (Couper, 2001). Online survey research allows for access to a broad database through email solicitation. It offers data collection within a short period of time and has low implementation costs. Respondents are afforded anonymity whil e the online interactive process eliminates interviewer error and bias (Sheehan & Hoy, 1999). Certain populations are ideally suited for online survey research, particularly those who visit specific websites and those holding interests associated with a sp ecific activity group.
41 Conversely, general population groups may be more difficult to analyze using online surveys, such as online polling for political elections (Wilson, 1999 in Babbie, 2007). Survey Development The questionnaire for this study was diss eminated using a web -based survey tool. The survey contained an informed consent protocol, which was approved as part of the Institutional Review Board process Respondents were required to prompt an acknowledgement of the informed consent protocol befor e proceeding to the survey. Two questionnaires were developed for this study, one directed to endurance sport athletes and the other directed to motorsport athletes. Each survey was identical in design, with the exception of the introductory wording and wording specific to behavioral indicators where required. The surveys were divided into seven sections, which included a general introduction to the study (ultrarunning or rally car racing), Informed Consent Protocol language and opt -in option, intention to participate items, behavioral indicator items demographic questions, multidimensional scale items and unidimensional scale items ( see Appendix A and B ). Behavioral and Descriptive Statistics Behavioral questions that incorporated years participating in the sport, level of commitment to the sport, intent to continue to participate in the sport, number of annual events the respondent participated in, the distance traveled to events and annual financial commitment to the sport. Three questions pertained t o intent to participate in future events. Ten demographic questions captured information on age, gender, income, marital status, education, state of residence, distance traveled to events, print media usage related to the sport, Internet usage related to the sport and perceived skill level in the sport.
42 Unidimensional and Multidimensional Involvement Scales Jamrozy, Backman & Backman (1996) conducted research on naturebased tourism and compared the application of multidimensional and unidimensional involvement scales. Their findings indicated that both scales were equally reliable and valid and were acceptable to measure the involvement construct in a leisure environment. In order to further contribute to the field of involvement within a leisu re contex t, Jamrosy, Backman & Backman chose Zaichkowskys Personal Involvement Inventory (PII) and Laurent and Kapferers Involvement Profile (IP) in an effort to further explore each scales multidimensionality. Following forward and bringing an individual and te am sport context to leisure activities, this study incorporated Havitz and Dimanches (1990) adaptation of Laurent and Kapferers multidimensional Involvement Profile ( Appendix C), and Zaichkowskys tenitem unidimensional Personal Involvement Inventory. Each of the 15 items in the Laurent and Kapferer scale were administered with a 5 point Likert type response option, where 1 was coded to strongly disagree and 5 was coded to strongly agree. The Zaichkowsky scale utilized a 1 to 7 measurement on semantic differential word pairs. Some questions were reverse -scored (positive word to 7 and negative word to 1) and were manipulated in the Statistical Package for the Social Sciences ( SPSS) prior to analysis. Respondent D evelopment The study population is compri sed through a non -probability sampling group of available subjects in the endurance sport (individual) and rally car (team) communities. Respondents were solicited through email databases provided by organizers for ultrarunning and rally car auto racing e vents. Sample groups characterized a broad spectrum of participants, from entry -level participants to accomplished athletes.
43 Ultrarunning, the sport of foot racing in events surpassing the traditional 26.2 mile marathon distance, has increased over 100 per cent in the past 20 years. In 2007, 14,251 individuals competed in ultradistance events, amassing a total of 25,841 finishes; 29 percent of the entrants were females, and the median age was 42.8 years (Medinger, 2008). Most ultrarunners perform continuously for more than four hours, to more than 24 hours and beyond when multi -day races are entered. Muscular degeneration does occur over these extended race distances, while mental and physiological exhaustion also comes into play. Ultrarunner s must constan tly monitor fluid intake, electrolyte levels and food consumption, while also monitoring emotional and cognitive decision -making abilities. The ultrarunning experience can be somewhat mystic. Jones (2005) conducted research that captured ultrarunners ex periences in endurance events that were 100 miles in distance and beyond. Jones identified a hypothetical mystical state of consciousness he termed Absolute Unitary Being (p. 40). Unlike the solitary environment of the ultrarunner, the sport of rally ca r racing is built on team cooperation and overall finishing results. Rally car auto racing is a timed motorsport competition where cars perform to get from one place to another in the shortest possible time. Rallies are organized in stages, on routes tha t include public or private roads, many times in secluded rural areas. Several surfaces, such as tarmac or gravel, may be incorporated into the course. Special stages are included in the regular stage segments, where elapsed time through the special stage is cumulated to attain the total stage time. Liaison stages connect the overall course, but are not included in the overall time (rallycars.com)
44 Drivers, with the assistance of a navigator, pilot modified production or specially built road -legal cars. Reconnaissance of the rally course is allowed a week prior to an official event, to allow navigators and teams the opportunity to create notes for race day. Rally races are extremely popular in Europe, South America and Asia, with an audience emerging in t he United States. Major events in Europe regularly attract hundreds of thousands of fans, who line the borders of twisty mountain roads. Respondent databases for endurance sport were provided by the editor of UltraRunning Magazine, the leading publication for ultrarunning endurance sport athletes. The promoter of the Umstead 100 mile endurance run, a premier event in North Carolina, also provided his database of entrants for this survey. The database for rally car auto racing was provided by Rally Ameri ca, the U.S. sanctioning body for rally sport in North America. Data Analysis (Hypotheses Testing) H1: The Zaichkowsky scale will be unidimensional for individual and team sports H2: The Laurent and Kapferer scale will be multidimensional (perceived imp ortance, perceived sign value, perceived pleasure value, perceived risk of negative consequences and perceived risk of making a mistake) for individual and team sports Hypotheses 1 and 2 will be tested for statistical significance using a confirmatory fa ctor analysis (CFA) which explores patterns of relationships among variables. Structural equation modeling provides a confirmatory role, as the indicators are specified for each construct. Three statistical tests within CFA were chosen to determine the v alid ity of the Zaichkowsky and Laurent and Kapferer scales, as applied to individual and team sport respondents A test for goodness -of -fit (GFI) of the proposed factor solution will determine how covariances and correlations of the observed model are pre dicted by the estimated model. (Ha ir Anderson, Tatham & Black, 1998). GFI offers an advantage in model testing in that it is not affected by sample size. A test for the maximum likelihood fitting function (FF) was the basis
45 for a Chi square test statis tic, an are two indexes frequently used in CFA. The third test was the root -mean -square residual (RMR) which is bounded by 0 and 1. The test examines the square root of the mean of the squared residuals in the observed and expected elements of the models (Marsh, Balla & McDonald, 1988). H3a: A relationship exists between the Zaichkowsky Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H3b: A relationship exists between the Zaichkowsky Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H4a: A relationship exists between the Laurent and Kapferer Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H4b: A relationship exists between the Laurent and Kapferer Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). Hypothese s 3 and 4 will be tested regarding the contribution each involvement scale makes as related to participation among individual and team sport athletes. Multiple regression analysis will be used, which allows researchers to determine if a dependent variable is affected simultaneously by several independent variables (Babbie, 2007). A correlation coefficient ( R) will indicate the correlation between variables and the positive and negative associatio ns between participation and involvement. H5a: The Laurent a nd Kapferer Scale will be more predictive than the Zaichkowsky Scale for intent to participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). H5b: The Laurent and Kapferer Scale will be more predictive than the Zaichkowsky S cale for behavioral participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport). Hypothesis 5 will be tested using Fishers z test for independent correlation coefficients which is a transformation of correlation coefficien ts to a normally distributed variable. Fishers z test is used to compare sample and means to determine if there is a significant difference between independent groups. The Fisher test is most used when two sample groups have one
46 variable in common ( Meng Rosenthal & Rubin, 1992) In this study the common variables among individual and team sport groups are participation intent and behavioral participation. H6: Cognitive Experience for Team auto racing (Team sport) will be greater than Cognitive Experie nce for Ultrarunning (Individual Sport). H7: Affective Experience for Ultrarunning (Individual sport) will be greater for than Affective Experience for Team auto racing (Team Sport). H8: Group Identification for Team auto racing (Team Sport) will be grea ter than Group Identification for Ultrarunning (Individual sport). Hypotheses 6, 7 and 8 will be tested with one -way analysis of variance (ANOVA), where it can be determined if the observed variance of the data sets have statistical significance as relate d to a common dependent variable. The ANOVA test allow s for examination of group means, with those variables in each scale that represent cognitive processing, affective processing and the antecedent for group affiliation (the sign value subscale in the La urent and Kapferer scale).
47 CHAPTER FOUR RESULTS Main Study Results Participant Analysis Data w ere collected from t wo sample groups for this study. The individual sport sample, comprised of ultrarunning endurance sport athletes, consisted of 424 usable r esponses. This sample was captured from a distribution to 2400 athletes (2000 in the UltraRunning magazine email database and 400 in the Umstead 100 Mile Endurance Run database ) which constitutes a 17.7% response rate. Participant age for individual athl etes ranged from 20 79 years, M=46.74, Mdn=46.00, SD=10.72. 71% (N=301) were male and 27.4% (N=116) were female, with 1.7% (N=7) not reporting their gender (Table 4.1) The proportion of female competitors was similar to that of the general ultrarunning population, which is 29.4 percent (Medinger, 2008). The most frequently reported level of competition was Intermediate (N=178) followed by Accomplished (N=131). The team sport sample, comprised of rally car auto racing teams, consisted of 2 91 useable resp onses. This sample was captured from a distribution to 850 online participants who are members of Rally America the U.S. sport sanctio ning body for rally auto racing, and constituted a 34.2% response rate. Participant age for team sport athletes ranged from 16 67 years, M=37.88, Mdn=36.00 SD=11.09. 85.9% (N=250) were male and 12.4% (N=36) were female, with 1.7% (N=5) not reporting their age (Table 4.1) The most frequently reported position on the team was driver (N=144) followed by navigator (N=93).
48 Table 4 1. Descriptive i ndicators for i ndividual and t eam s port a thletes Individual sport Team sport Mean age 46.74 37.88 Male % 71.00 85.90 Female % 27.40 12.40 Years competing (M) 7.98 9.35 Dependent Variable: Participation This study mea sure d participation in two ways. One was an intention measure and the other was a behavioral measure. The items used to measure intent to participate were (1) It is my intention to participate in a rally car (ultrarunning) event in the next year and (2) In the next year I plan to participate in a rally car (ultrarunning) event. (K. Kaplanidou, personal communication, June 10, 2008). The items used endpoints strongly disagree and strongly agree within a 5 -point scale The mean scores for intent ion were 4.64 for ultrarunning athletes and 4.35 for rally car racers (Table 4 2) The Cronbachs Alpha for participation indicators in the individual (ultrarunning) sport data was .89. The Cronbachs Alpha for the participation indicators in the team (r ally) sport data was .86. Two items related to behavioral participation were incorporated into the study Team rally participants were asked to input the average number of hours committed to rally cars each week and the average number of days committed to rally cars each week. Hours per week were divided by the days per week to create an average hours per day variable. Ultrarunning individual sport participants were asked to input the average number of miles run each week and the average number of days r unning each week. Miles run per week was divided by average number of days run each week to create an average miles per day variable The mean scores for behavioral participation were 9.82 (miles per day) for ultrarunning athletes and 5.34 (hours p er day committed) for rally car racers (Table 4 2) Correlations
49 between the two items identified for behavioral characteristics were statistically significant for both ultrarunning and rally car data. Table 4 2. Mean and standard deviations for group participat ion variables Intent Behavior M SD M SD Ultrarunning 4.64 1.02 9.8 3.78 Rally car racing 4.35 1.27 5.3 4.09 Independent Variable: Laurent and Kapferer Involvement Scale A consumer involvement profile (CIP) developed by Laurent and Kapferer (1985) and later adapted to leisure activity (Havitz & Dimanche, 1996) was used to incorporate a multidimensional facet. The 5 -point, 15 item scale used the endpoints of strongly disagree and strongly agree. The antecedents of involvement, each represented as a 3 -item subscale, are perceived importance, perceived consequence of risk, perceived probability of risk, symbolic or sign value and hedonic pleasure (Appendix C). Reliability was tested to determine internal consistency within subscales. The multidimensionality of a scale can also affect Alpha. Cortina (1993) posited that if a scale has one common factor, Alpha represents the strength of that factor. However, literature does not provide a clear understanding of the extent to which Alpha is affected by dimensionality. Schmitt (1996) stated that Cronbachs Alpha was an understatement of reliability, defined as the communality of items when multidimensionality is present. Cronbachs Alpha was tested for the five subscales within the multidi mensional scale for individual sport (ultrarunning) athletes (Table 4.3). The 3 subscale items representing importance had a Cronbachs Alpha of .64, with inter item correlations ranging from f=.33 to f=.50. The 3 subscale items representing pleasure had a Cronbachs Alpha of .64, with inter item correlations ranging from r=.35 to r=.41. The 3 subscale items representing risk
50 consequence had a Cronbachs Alpha of .68, with inter item correlations ranging from r=.35 to r=.55. The 3 subscale items represe nting risk probability had a Cronbachs Alpha of .52, with inter item correlations ranging from r=.25 to r=.37. The 3 subscale items representing sign value had a Cronbachs Alpha of .60, with inter -item correlations ranging from r=.24 to r=.68. Cronbach s Alpha was tested for the five subscales within the multidimensional scale for team sport (rally car) athletes (Table 4.3). The 3 subscale items representing importance had a Cronbachs Alpha of .90, with inter item correlations ranging from r=.68 to r=. 88. The 3 subscale items representing pleasure had a Cronbachs Alpha of .95, with inter item correlations ranging from r=.81 to r=.95. Kapferer and Laurent (1993) believed it was mandatory to include the antecedent of risk in the development of the CIP s cale and to make a distinction between risk probability and risk consequence. However, the authors found that risk probability was the least related to the remaining four subscales. Kapferer and Laurent believed that risk probability should not be remove d from the scale, as it represented the input of others, decision making and information gathering. Kyle, Graefe, Manning & Bacon (2003) stated that when the Laurent and Kapferer scale is applied to leisure activity, the subscales of risk probability and r isk consequence have performed least consistently. In the team sport context of rally, the 3 subscale items representing risk consequence had a Cronbachs Alpha of .69, with inter item correlations ranging from r=.31 to r=.55. The 3 subscale items repres enting risk probability had a Cronbachs Alpha of .66, with inter item correlations ranging from r=.16 to r=.86. The 3 subscale items representing sign value had a Cronbachs Alpha of .70, with inter -item correlations ranging from r=.22 to r=.81.
51 Table 4 3 Reliability testing for Laurent & Kapferer sub scales on individual and team sport data Items tested Alpha ultrarunning Alpha rally car # of items importance 0.64 0.9 0 3 pleasure 0.64 0.95 3 risk consequence 0.68 0.69 3 risk probability 0.53 0.66 3 sign value 0.6 0 0.7 0 3 Independent Variable: Zaichkowsky Involvement Scale The i nternal consistency of the Zaichkowsky scale w as tested through reliability analysis The 10 semantic differential items in the Zaichkowsky scale (1994) were tested. It ems comprised in the scale were: important unimportant, boring interesting, relevant -irrelevant, exciting unexciting, means nothing-means a lot to me, appealing unappealing, fascinatingmundane, worthless valuable, involvinguninvolving, not needed -needed. The Cronbachs Alpha for these ten items was .89 for the individual sport data and .85 for the team sport data. Inter item correlations ranged between r=.22 to r=.71 for individual sport respondents and r=.16 to r=.67 for team sport respondents (Table 4. 4 ). Cronbachs Alpha would not be increased by deleting any of the scale items. Table 4 4. Reliability testing for the Zaichkowsky scale on individual and team sport data Items tested Alpha ultrarunning Alpha rally car # of items Zaichkowsky scale 0.8 9 0.8 5 10 Hypothesis 1: Dimensionality of the Zaichkowsky Scale To test the hypothesis that the Zaichkowsky scale will remain unidimensional and the Laurent and Kapferer scale would remain multidimensional, confirmatory factor analysis was employed t o test goodness of fit within both the individual and team sport environments.
52 Confirmatory factor analysis explores patterns of relationships among variables. Structural equation modeling provides a confirmatory role, as the indicators are specified fo r each construct. Hypothesis 1 a stated that the Zaichkowsky scale will be unidimensional for indivi dual (endurance running) sport s. T he single factor solution was not appropriate and the hypothesis was rejected (Figure 4.1) for individual sport athletes ( 437.53, df=37, RMR=.17 GFI=.78) Hypothesis 1b stated that the Zaichkowsky scale will be unidimensional for team (rally car) sport s. The single factor solution was not appropriate and the hypothesis was rejected (Figure 4.2) for team sport athletes RMR=.18, GFI=.73) 0, .81 involvement5.94 important0, .93 d11 6.54 interesting0, .53 d2.43 1 0, 1.00 d35.83 exciting0, .85 d41 6.27 means a lot0, .60 d5.71 1 6.02 appealing0, .58 d61.14 1 5.82 fascinating0, .49 d71.11 1 6.24 valuable0, .51 d8.71 1 5.99 involving0, .74 d95.47 needed0, 1.00 d10.85 1 1.00 1.00 5.19 relevant1.05 1.08 1 1.05 Figure 4 1. Individual sport unidimensional scale model
53 0, .61 involvement5.85 important0, .86 d11 6.75 interesting0, .30 d2.35 1 0, 1.00 d36.41 exciting0, 1.03 d41 6.14 means a lot0, .55 d51.01 1 6.23 appealing0, .91 d6.93 1 6.13 fascinating0, .72 d7.81 1 5.90 valuable0, .75 d81.08 1 6.34 involving0, .82 d95.37 needed0, 1.00 d101.18 1 1.00 1.00 5.20 relevant1.08 1.18 1 .75 Figure 4 2. Team sport unidimensional scale model T he unidimensional scale was not confirmed as a one factor solution and exploratory factor analysis wa s utilized to explore the dimensions of the ten items within the Zaichkowsky scale. While the item interesting was confounded, the remainder of the results indicated that
54 affectively -processed involvement items relevant, exciting, appealing, fascinating and involving loaded onto Factor 1, while cognitive items important, means a lot, valuable and needed loaded onto Factor 2 (Table 4 5 ). Revised Hypothesis 1a stated that the Zaichkowsky scale will be two -dimensional for individual (endurance running) spor t. A retest was conducted utilizing confirmatory factor analysis to test for cognitive and affective dimensions within the unidimensional scale (Figure 4 3). Table 4 5 Varimax rotation exploratory factor analysis for individual sport athletes Factor 1 Factor 2 I mportant .25 .73 I nteresting .48 .25 R elevant .61 .39 E xciting .85 .18 M eans a lot .29 .78 A ppealing .85 .21 F ascinating .83 .26 V aluable .43 .65 I nvolving .76 .24 N eeded .13 .85 The scale performed as two factors and provided a sol ution for data related to individual sport athletes The two factor solution was accepted as modified for Hypothesis 1a. Revised Hypothesis 1b stated that the Zaichkowsky scale will be two-dimensional for team (rally car) sport. A retest was conducted utilizing confirmatory factor analysis to test for cognitive and affective dimensions within the unidimensional scale (Figure 4 4 ). The scale performed as two factors and provided a solution for data related to team spor t athletes The two factor solution was accepted as modified for Hypothesis 1 b
55 Figure 4 3. Individual sport model modified with cognitive and affective subscales 0, .62 cognitive involvement 5.94 important 0, .76 d6 1 6.54 interesting 0, .53 d1 1 0, 1.00 d7 5.83 exciting 0, .68 d2 1 6.27 means a lot 0, .42 d8 .97 1 6.0 2 appealing 0, .45 d3 1 5.82 fascinating 0, .39 d4 1 6.24 valuable 0, .44 d9 1 5.99 involving 0, .68 d5 5.47 needed 0, 1.00 d10 1 5.19 relevant 1.08 1.19 1 1.00 .87 1.21 0, .14 affective involvement 1.00 3.04 2.91 2.78 2.59 .22
56 F igure 4 4 Team sport model modified with cognitive and affective subscales .92 cognitive involvement important .74 d6 1 interesting .30 d1 1 1.00 d7 exciting .68 d2 1 means a lot .47 d8 .87 1 appealing .44 d3 1 fascinating .51 d4 1 valuable .59 d9 1 involving .70 d5 needed 1.00 d10 1 relevant 1.08 .95 1 1.00 .97 1.08 .08 affective involvement 1.00 3.00 3.52 2.75 2.39 .15
57 Retesting Reliability for the Two Factor Zaichkowsky Scale The two factor solution representing cognitive and affective subscales within the Zaichkowsky scale was tested for internal con sistency within the separate dimensions. The affective subscale tested for individual athletes presented a Cronbachs Alpha of .86, with inter item correlations ranging from f=.29 to f=.71. Alpha could be improved to .89 with the removal of interesting from the subscale. The cognitive subscale for individual athletes presented a Cronbachs Alpha of .81, with inter item correlations ranging from f=.39 to f=.57. The deletion of items within the 5 item subscale would not have improved the alpha score (Table 4. 6 ). The affective subscale tested for team athletes presented a Cronbachs Alpha of .82, with inter item correlations ranging from f=.31 to f=.67. A small increase to an alpha of .83 could be achieved by removing the interesting item from the subs cale. The cognitive subscale tested for team sport athletes presented a Cronbachs Alpha of .84 with inter item correlations ranging from f=.43 to f=.59. The deletion of items within the 5 item subscale would not have improved the alpha score. Table 4 6. Reliability testing for the Zaichkowsky scale on cognitive and affective dimensions Items tested Alpha ultrarunning Alpha rally car # of items affective 0.86 0.82 5 cognitive 0.81 0.84 5 Hypothesis 2: Dimensionality of the Laurent and Kapferer Scale Hypothesis 2 a stated that the Laurent and Kapferer scale will be multidimensional within the scales intended for individual (ultra running) athletes A goodness of fit test indicated that the five -factor solution indicated was appropriate (Figure 4 5 ) for data r elated to individual sport Hypothesis 2a was supported and accepted.
58 Hypothesis 2b stated that the Laurent and Kapferer scale will be multidimensional within the scales intended for team (rally car) athletes. A good ness of fit test indicated that the five factor solution was appropriate (Figure 4 RMR=.07, GFI=.90). Hypothesis 2b was supported and accepted. Figure 4 5. Individual sport multidimensional scale model
59 Figure 4 6 Team s port m ultidimensional s cale m odel
60 Hypothesis 3a: Relationship between the Zai ch k owsky Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2 ) Team auto racing (Team sport) Multiple regressions with the tw o Zaichkowsky scales (cognitive and affective) as independent variables and intent to participate as the dependent variable were used to test this hypothesis for ultrarunning and team auto racing. The hypothesis was not supported for ultrarunning. The mult iple regression was not significant (R = .11, F = 2.65, p = .07). The hypothesis was supported for team auto racing (R = .19, F = 5.31, p = .007) (Table 4 7 ). Between the cognitive and affective dimensions, only the cognitive had a significant standardized coefficient (.21). Table 4 7 Regression of Zaichkowsky cognitive and affective subscales on intent/plan to participate for team sport athletes Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 7 .79 1.22 6.38 .00 cognitive .53 .16 .21 3.25 .00 affective .33 .20 .11 1.62 .11 Dependent: Intent to Participate Independent: Zaichkowsky Cognitive and Affective R = .19, F = 5.31, p =.007, n = 281 Hypothesis 3b: Relationship between Zaichkowsky Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2 ) Team auto racing (Team sport) Multiple regressions with the two Zaichkowsky scales (cognitive and affective) as independent variables and behavioral participation (miles run per day for ultrarunning and hours per day invested in auto racing) as the dependent variable were used to test this hypothesis for ultrarunning and team auto racing. This hypothesis was partially supported. The multiple regression for ultrarunning miles run per day as the dependent variable and Zaichkowskys cognitive and affective dimensions as independent variables was significant (R = .13, F = 3. 31, p =.0 4 ) (Table 4 8 ). The regression for auto racing (R = .1 1 F = 1. 2 8, p = .2 8 ) was not significant.
61 T able 4 8 Regression of Zaichkowsky cognitive and affective subscales on behavioral participation for individual sport athletes Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 6.5 8 1.31 5 .0 2 .00 cognitive 07 .2 5 08 .26 79 affective 48 .2 7 12 1.78 08 Dependent: Average miles/day. Independent: Zaichkowsky Cognitive and Affective R = .13, F = 3.31, p=.04, n = 401 Hypothesis 4a: Relationship between Laurent and Kapferer Sca le and Intent to Participate in (1) Ultrarunning (Individual sport) and (2 ) Team auto racing (Team sport) Multiple regressions with the five Laurent and Kapferer sub scales as independent variables and intent to participate as the dependent variable were us ed to test this hypothesis for ultrarunning and team auto racing. This hypothesis was partially supported. The hypothesis was supported for ultrarunning. The multiple regression was significant (R = 19, F = 2.91, p =.0 1 ) (Table 4 -9 ). Among the five dimensi ons of the Laurent and Kapferer scale, only pleasure was significant with a .18 regression coefficient. Neither Importance, Risk Consequence, Risk Probability, nor Sign Value contributed significantly to intent to participate in ultrarunning. The hypothesis for team auto racing was not supported. The regression (R = .1 9 F = 1.96, p = 09) was not significant. Table 4 9. Regression of Laurent & Kapferer multidimensional subscales on intent/plan to participate for individual sport athletes Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 3.40 .5 4 6.24 .00 importance .0 3 .11 .0 1 24 81 pleasure .2 6 .10 .1 6 2. 61 .00 9 risk consequence .0 4 .06 .0 3 68 49 risk probability .1 8 12 .0 8 1. 56 .1 2 sign value .1 5 .08 .1 1 1. 92 .0 5
62 Dependent: Intent to Participate Independent: Sign Value, Risk Probability, Risk Consequence, Pleasure, Importance R = 19, F = 2.91, p=.0 1 n = 400 Hypothesis 4b: Relationship between Laurent and Kapferer Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2 ) Team auto racing (Team sport) Multiple regressions with the five Laurent and Kapferer sub scales as independent variables and behavioral participation (miles run per day for ultrarunning and hours per day invested in auto racing) as the dependent variable were used to test this hypothesis for ultrarunning and team auto racing. This hypothesis was partially supported. The hypothesis was supported for rally car racing. The multiple regression was significant (R = .24, F = 2.61, p=.03. (Table 4 10). Among the five dimensions of the Laurent and Kapferer scale, only imp or tance was significant with a 24 regression coefficient (Table 4 7 ). Neither Pleasure Risk Consequ ence, Risk Probability, nor Sign Value contributed significantly to intent to participate in ultrarunning. The hypothesis for individual running was not supported. The regression (R = .1 4 F = 1.4 8 p = .20) was not significant. Table 4 10. Regressio n of Laurent and Kapferer multidimensional subscales on behavioral participation for team sport athletes Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 1.44 1.61 0.89 0.37 importance 1.03 0. 51 0.24 2.04 0.04 pleasure 0.22 0.49 0.07 0.44 0.66 risk consequence 0.19 0.35 0.04 0.56 0.58 risk probability 0.07 0.54 0.02 0.13 0.9 sign value 0.64 0.52 0.13 1.25 0.21 Dependent: Hour/day average. Independent: Sign Value, Risk Probability, Risk Consequence, Pleasure, Importance R = .2 4 F = 2.1 p=.0 3, n = 261
63 Hypothesis 5a: Laurent and Kapferer Scale will be more predictive than the Zaichkowsky Scale for intent to participate in (1) Ultrarunning (Individual sport) and (2 ) Team auto rac ing (Team sport) Hypothesis 5a was tested using Fishers z test for independent correlation coefficients (R). The hypothesis was not supported. The one tailed Z critical score required for a directional comparison requires a 1.65 z score. Comparisons be tween the regressions of the Zaichkowsky subscales and the Laurent and Kapferer subscales, to the dependent variable of intent, had no statistical significance for either ultrarunners (z = 1.16) or rally car athletes (z = .00) (Table 4 11). Hypothesis 5b: Laurent and Kapferer Scale will be more predictive than the Zaichkowsky Scale for behavioral participation in (1) Ultrarunning (Individual sport) and (2 ) Team auto racing (Team sport) Hypothesis 5b was not supported. Comparisons between the regressions of the Zaichkowsky subscales and the Laurent and Kapferer subscales, to the dependent variable of behavior, had no statistical significance for either ultrarunners (z = 1.20) or rally car athletes (z = 1.41) (Table 4 11). Table 4 11. Fishers z test of independent Rs for Zaichkowsky and Laurent and Kapferer scales Rally Car Ultrarunning Zaichkowsky L&K Zaichkowsky L&K R n R n z R n R n z Intent 0.19 281 0.2 278 0.00 0.11 404 0.19 400 1.16 Behavior Zaichkowsky L&K Zaichkowsky L&K R n R z R N R n z 0.11 225 0.2 222 1.41 0.13 401 0.14 396 1.20 Hypothesis 6: Cognitive Experience for Team auto racing (Team sport) will be greater than Cognitive Experience for Ultrarunning (Individual Sport) This hypothesis was not supported. While t he results of the analysis of variance to test this hypothesis provided evidence of a statistically significant difference between groups (F 1,700 = 6.92, p = .01 ), the difference was not as predicted. The results were opposite to what was
64 hypothesized. Wi thin the Zaichkowsky cognitive t est, t he average cognitive experience score for individual ultrarunning was 5.84, which was significantly greater than the team auto racing average score of 5.65. The effect size, using eta squared, was .01. The Laurent and Kapferer test also determined a result opposite to what was hypothesized. Though statistically significant (F 1, 698 = 180.20, p = .00) ultrarunners held a cognitive mean of 3.63, which proved higher than the rally car average of 3.03. Hypothesis 7: Affe ctive Experience for Ultrarunning (Individual S port) will be greater than Affective Experience for Team auto racing (Team Sport). This hypothesis was partially supported. While the results of the analysis of variance to test this hypothesis provided eviden ce of a statistically significant difference between groups (F 1, 693 = 18.28, p = .00 for the unidimensional affective items and F 1, 698 = 188.55, p=.00 for the multidimensional affective items ), the difference was not as predicted. The unidimensional a ffective experience score for rally car participants was 6.35 which was significantly greater than the individual running score of 6.05. However, under multidimensional affective measures, ultrarunners held a significantly greater mean score with 4.14 ov er the rally car racing mean score of 3.34. The effect size (using eta squared) for the multidimensional affective items was strong at .21. The effect size ( using eta squared ) for the unidimensional affective items was .02. Hypothesis 8: Group Identificat ion for Team auto racing (Team Sport) will be greater than Group Identification for Ultrarunning (Individual sport). This hypothesis was not supported. While the results of the analysis of variance to test this hypothesis provided evidence of a significant difference between groups (F 1,70 5 = 155.52, p = .00), the difference was not as predicted. The results were opposite to what was hypothesized. The average group identity score for rally car team athletes was 3.14, which was significantly less than the ul trarunning individual average score of 3.87. The effect size, using eta squared, was 18.
65 Table 4 1 2 ANOVA comparisons for cognitive, affective and group response Mean individual Std. deviation individual Mean team Std. deviation team eta squared F df Sig. Zaichkowsky cognitive 5.84 0.92 5.65 1.02 0.01 6.92 700 .01 Laurent & Kapferer cognitive 3.63 0.4 3.03 0.78 0.21 146.4 388 .0 0 Zaichkowsky affective 6.05 0.97 6.35 0.81 0.03 19.45 672 .0 0 Laurent & Kapferer affective 4.14 0.57 3.34 0.98 0.21 15 6.25 413 .0 0 Laurent & Kapferer group ID 3.87 0.72 3.14 0.83 0.18 147.87 560 .0 0
66 CHAPTER 5 DISCUSSION Sport event organizers, sponsors, p roduct marketers and media are striving to find new and compe titive alternatives for communicating with participant sport athle tes. A cluttered marketing environment has necessitated the need for targeted m essaging that speaks to each athletes most c o mpelling rationale for participating in specific sport events and buying products and services related to those events. This study was designed to compare and contrast the involvement characteristics of two diverse sport participant groups individual and team sport athletes. Predictors of participation, cognitive and affective responses, and the power of group identity were identifi ed as key learning points in this research Two t raditional and academically tested involvement scales were utilized in this study. The Zaichkowsky involvement scale utilized semantic differen tial word pairs and was designed to measure a sum involvement score within a single factor solution. Zaichkowsky developed a 20item product involvement scale in 1985, and then made a revisio n to the scale in 1994, reducing the scale to 10 items when she addressed involvement as tool to measure the respo nse to advertis ing and the effectiveness of the advertising message. Laurent and Kapferers multidime nsional involvement measure focused on the antecedents to involvement through a 15 item scale which incorporated five subscales. The scale was ini tially devised in 1985 through qualitative research conducted with female, stay at -home respondents and their involvement response to various home and cosmetic products. The unidimensionality of the Zaichkowsky scale and the multidimensionality of the Laurent a nd Kapferer scale, were tested for both individual and team sport respondents through confirmatory factor analysis Tests were c onducted in the Statistical Package for the Social
67 Sciences ( SPSS) with A nalysis of M o ment S tructures (AMOS ) software using str uctural equation modeling Multiple regressions were utilized to test the relationship of each scale to participation scores for individual and team sport athletes The p articipation indicators of intent, and the likelihood item, were used as separate de pendent variable s. The cognitive subscale and affective subscale in the unidimensional scale and five antecedent subscales in the multidimensional scale served at independent variables Social cognition and social identity theories were used as a theoreti cal foundation to predict a higher cognitive measure in team sport athletes, a higher affective measure in individual sport athletes, and a high er measure of sign value (group identity) in team sport athletes. Cognitive, affective and group identification components of the scales were extracted to perform an analysis of variance (ANOVA) between the individual and team sport data groups. A deeper understanding of the predisposition to either cognition or emotion for participants, as well as a groups need t o s trive for social identity, will prove advantageous to communicators who are develop ing advertising and mar keting initiatives related to active sport participants This discussion will evaluate each hypothesis based on study results and offer potential e xplanations as well as implications for advertising and communications managers. Other comments will include limitations within the study and sug gestions for future research that will benefit the sport communication industry. Evaluation of the Hypothese s Hypothesis 1: Dimensionality of the Zaichkowsky Scale This hypothesis was not supported. Analysis showed the scale was two dimensional, not one. The resultant two dimensional Zaichkowsky scale represented the cognitive and affective
68 dimensions of the Za ichkowsky scale as previously discussed in the literature, e.g. Zaichkowsky, 1995 and Salmon, 1986. These results support the use of the Zaichkowsky scale in studies on involvement in individual endurance sport and motorsport teams but as a two dimension model, not as a one -dimensional model. Hypothesis 2: Dimensionality of the Laurent and Kapferer Scale This hypothesis was supported. Analysis showed the scale retained its five -factor structure and confirmed the multidimensionality of the Laurent and Kapferer Scale. This multidimensional model is appropriate for use in the study of individual endurance sport and motorsport team involvement. Hypothesis 3a: Relationship between the Zaichkowsky Scale and Intent to Participate in (1) Ultrarunning (Individua l sport) and (2) Team auto racing (Team sport) This hypothesis was partially supported. The two-dimension Z ai ch k owsky scale produced a significant relationship for intent to participate in team sports but not for individual sports (T able 5 1) While the la ck of variance in the dependent variable may partially explain the lack of a relationship for ultrarunning, the lack of variance did not adversely affect team auto racing. This suggests that the Z ai ch k owsky scale is not appropriate in measuring involvement in studies focused on intent to participat e in individual sports, e.g., ultrarunning. At the same time, only the cognitive dimension of the Z ai ch k owsky scale was related to intent to participate in team auto racing suggesting that use of the affective di mension may not be appropriate as a measure of involvement in studies focusing on intent to participate. H ypothesis 3b: Relationship between the Z ai ch k owsky Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2 ) Team auto racing (Team sport) This hypothesis was partially supported. The two -dimension Zaichkowsky scale produced a significant relationship for individual running sport, but not for team auto sport (Table 5 1) These regressions suggest that the two-dimension Zaichkow sky scale is appropriate in measuring
69 behavioral participation in individual sport, but is not appropriate for team sport. Within the individual sport regression, only the affective component was related to behavioral participation suggesting that the use of cognitive dimension may not be appropriate as a measure for involvement as related to behavioral participation. Hypothesis 4a: Relationship between Laurent and Kapferer Scale and Intent to Participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) Hy pot hesis 4 a was partially supported. The multidimensional scale held a statistically significant relationship with intent to participat e among individual ultrarunning sport athletes but no significant relationship existed with team sport athletes (Table 5 1) These regressions suggest that the Laurent and Kapferer scale is appropriate to measure for intent to participate in individual sport, but not for team sport. Only the pleasure subscale proved to be significant within the individual sport regression, suggesting that the dimensions of sign value, importance, risk consequence and risk probability may not be appropriate as a measure of involvement as related to intent to participate. Hypothesis 4b: Relationship between Laurent and Kapferer Scale and Behavioral Participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) Hypothesis 4b was partially supported. The multidimensional scale held a statistically significant relationship with behavior al participation among rally car team sport participants, but did not test as significant for ultrarunning individual sport athletes (Table 5 -1) These regressions suggest that the Laurent and Kapferer scale proves appropriate to measure behavioral partic ipation in team sport, but not for individual sport. Only the importance subscale proved significant within the team sport regression, suggesting that the dimensions of pleasure, sign value, risk consequence and risk probability may not prove appropriat e as a measure of involvement as related to behavioral participation.
70 Table 5 1. Involvement regressions for i ntention and b ehavior Intent Behavior Ultra Rally Ultra Rally Zaichkowsky NS SIG SIG NS Cognitive 0.21 0.12 Affective 0.1 1 0.02 Laurent & Kapferer SIG NS NS SIG Importance 0.04 .24 Pleasure 0.18 .07 Risk consequence 0.02 .04 Risk probability 0.09 .02 Sign value 0.1 .13 Hypothesis 5a: Laurent and Kapferer Scale will be more predictiv e than the Zaichkowsky Scale for intent to participate in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) Hypothesis 5a was not supported. Fishers Z test for independent Rs showed no differential predictive validity between the Laurent and Kapferer scale, and the Zaichkowsky scale, regarding intention to participate. This suggests that the greater multidimensionality of the Laurent and Kapferer scale is not a benefit in predicting intention to participate. The two factor Zaich kowsky scale is equally appropriate for testing within the individual endurance sport and team motorsport communities in those instances where regressions prove to be statistically significant Hypothesis 5b: Laurent and Kapferer Scale will be more predic tive than the Zaichkowsky Scale for behavioral participation in (1) Ultrarunning (Individual sport) and (2) Team auto racing (Team sport) Hypothesis 5b was not supported. Fishers z test for independent Rs showed no differential predictive validity betwe en the Laurent and Kapferer scale, and the Zaichkowsky scale, regarding behavioral participation. This suggests that the greater multidimensionality of the Laurent and Kapferer scale is not a benefit in predicting intention to participate. The two
71 factor Zaichkowsky scale is equally appropriate for testing within the individual endurance sport and team motorsport communities in those instances where regressions prove to be statistically significant. Hypothesis 6: Cognitive Experience for Team auto racing (Team sport) will be greater than Cognitive Experience for Ultrarunning (Individual Sport) Hypothesis 6 was not supported. Ultrarunning individual sport athletes had a greater cognitive experience than did team auto racing athletes. Social cognitive the ory and social identity theory posit that cognitive activity is higher within groups, organizations and teams. This study suggests that cognitive processing among individual sport athletes is developed through the definition of event as the group or tea m. Ultrarunning athletes may take a broader view of their involvement among other competitors, the terrain and event logistics. The complexities presented may be as great, or greater, than those experienced by team sport athletes. The perspective of ind ividual athlete to event will be a fruitful area for future research. H ypothesis 7: Affective Experience for Ultrarunning (Individual sport) will be greater than Affective Experience for Team auto racing (Team Sport). Hypothesis 7 was partially supported. Analysis of variance for individual and team sport groups was significant when testing for both the affective dimension of the Zaichkowsky scale and the affective subscales of the Laurent and Kapferer scale. Ultrarunners held a higher affective mean sco re over rally car racers within the Laurent and Kapferer affective measures, but ultrarunners held a lower affective mean score than rally car racers within the Zaichkowsky affective measures. Since the Laurent and Kapferer analysis test held a strong eff ect size (representing the influence of the independent variable), the affective dimensions of that multidimensional scale prove most appropriate for testing affective processing within individual sport participant research.
72 H ypothesis 8: Group Identification for Team auto racing (Team Sport) will be greater than Group Identification for Ultrarunning (Individual sport) Hypothesis 8 was not supported. Analysis of variance between individual and team sport groups was significant, but the group identification was not as predicted. The group identity mean score was lower for team sport athletes than for individual sport athletes. Despite a strong effect size, the Laurent and Kapferer subscale of sign value may not prove to be an adequate measure for group ide ntity. Sign value items within the multidimensional scale were the only indicators available to test for group identity within the sample groups. Sign value subscale items are related to tells a lot about the participant, and about other participants i n the same sport group. The mental processing, or affective response, to these subscale items may be varied in nature, or only partially applicable to a sport setting. Implications and Limitations General The study and comparison of individual and team sport athletes raises questions about the a pplicat ion of results to advertising initiatives for sport and event communicators The unidimensional scale did not r etain a single factor solution and exploratory factor analysis led to the confirmation of cognitive and affective subscales within the scale. Factors re present cognitive and affective dimensions that can be used to understand athlete involvement beyond the sum score currently used with in the unidimensional scale. The multidimensional involvement s cale was supported with a five factor solution as intended, within both study groups F urther examination of respondent scores within subscales would determine how practitioners might approach messaging within the multidimensional construct. For example, respondents who scored exceptionally high on the importance subscale
73 would be influenced by messaging that focused on the prestige and accolades associated with a specific event. Involvement dimensions within t he multidimensional and unidimensional scale s were hypothesized to be predictors of participation for individual and team sport athletes Involvement and participation scores for both study groups were exceptionally high with little variation, which created results opposed to the predictions. Hypot heses were only partially supported in each regression test between involvement and participation Though it did not support the hypotheses, individual sport athletes demons trated higher cognition and group i dentification scores. This study would indicat e that individual sport athletes may find a team relationship with other participants at the event. Though the association among group athletes would not be as formal or defined, participant sport athlet es may identif y themselves among an alliance of in dividuals who share the same passions and sport lifestyles. Practical Implications This study would lead marketers to utilize a blended cognitive and affective marketing approach for individual sport athletes. Cognition, as operationalized by Zaichkowsky and Laurent and Kapferer, represent s important and relevance with a response based on risk approximation Messaging related to this definition of cognition would prove valid for influencing athletes within an endurance sport audience Individual sport a thletes also responded with a higher mean to Laurent and Kapferers affective measures, which incorporate sign value and pleasure. Examples of endurance sports that would respond to this form of sport communication are c ycling, running, kayak racing, moun taineering, cross country skiing, distance swimming, triathlons and adventure racing.
74 Individual spo rt athletes will connect to information presented on the magnitude of the event (importance) and possible advantages to participation, such as the prominenc e of the vent among top competitors (relevance). A description of event terrain (ri sk probability and consequence) available food and fluids on the course, would also prove to be appropriate. Individual athletes also scored a higher mean for group ident ity than did team sport athletes. A dvertising featuring the elements of sign value (the symbolic value of the association) would promote a sense of community for th is specified group. A predominant group identity among individual athletes would suggest that advertising should exemplify a positive association with the fraternity of athletes competing in the event. Advertising messages could imply that athletes who identify with participation in the event will experience increased well-being and higher sa tisfaction. Success can be identified as a positive shared experience with other competitors. The prediction that team sport athletes will use more cognition within the involvement e xperience was not supported. According to the results of this study, p ractitioners marketing to team sport athletes should consider affective messaging as indicated within the Zaichkowsky scale, while group affiliation can be minimized. P redominant dimensions within the regressions of unidimensional and multidimensional scale s as predictors of participation would give indications of how to form communication messages. Rally car team responded well to cognition and importance. Ultrarunning individual athletes had a strong respond to pleasure. A thletes that scored high on the pleasure subscale might be influenced by advertising that promoted the amenities and post -event party available at a specific sport event. Athletes that score d high on the importance subscale would respond
75 well to messaging that highlighted the elite nature of an event, or the long -standing tradition of an event as part of a na tional championship series The fascination and excitement that participants related to in motorsport would be appropriate for the higher affective scores for rally car athletes. The appeal of motor racing could be messaged through visual stimulation featuring speed and color. Affective processing with the Zaichkowsky scale also incorporates interesting. Images depicting the technology side of the sport could produc e an emoti onal attraction that would be relevant to rally car participant s Examples of motorsports that would respond to this form of sport communication are sprint cars, drag racing, powerboats, motorcycles, stock cars and open -wheel cars ( F ormula O ne and Indy Ra cing League machines). Limitations Many of the hypotheses developed for this study were not supported or were partially supported. It should be noted that sample groups represented a unique cadre of competitors with in the individual and team sport commun ities which may curtail external validity and the ability to generalize to a larger population. Ultradistance runners, defined by the nature of their sport, must be highly committed in order to participate. The rigors of training and preparation are exc eptional and may account for high involvement and participation scores and lack of variance among the sample group (Figures 5 2 and 53) Unlike a golfer who may play the game once a month yet claim to be involved, ultrarunners define active sport by ad he ring to an almost daily routine of exercise and related fitness activities. Though individual in nature, ultrarunning athletes compete in a complex environment over long distances and times. The need to cognitively process issues such as food intake, injury or dehydration could account for a cognitive response in what might otherwise be an affective setting
76 Figure 5 1. Ultrarunning individual sport involvement ( Zaichkowsky scale) Figure 5 2. Ultrarunning individual sport intent to participate Choos ing motorsport athletes for a team sport sample group also presented limitations. Though too prevalent to be dismissed as a viable choice for research, the dynamics of motorsport are distinctive Success is based on the performance of each team contrib ut or, where fractions of a second incurred during a racing pit stop can be the differe nce between winning an event and
77 falling from contention Though theory suggests that athletes within organizations must have a cognitive interaction to be successful, aff ective involvement may be heightened within the team motorsport context based on the performance of the driver. An individual component exists within the team sport context. Future Research The testing and application of modified involvement scales, desi gned specifically for participant sport athletes, could prove beneficial in future research. Scholars in fields related to sport have developed scales for sport fans, but for the most part precluded specific scale items for participant athletes. Scales u sed in this study were developed for product involvement and advertising response and have limited applicability to the participant sport experience. The interrelationship of the Zaichkowsky and Laurent and Kapferer scales also presents a future opportunit y for research. Significant correlations exist between the involvement dimensions of the scales for ultrarunning athletes (Table 5 2) The strongest positive correlation, r = .65, n = 406, p = .00 was represented by the affective and cognitive dimensions within the Zaichkowsky scale The pl easure and importance subscales, r = .54, n = 416, p = .00, within the Laurent and Kapferer scale, and Table 5 2 Correlation matrix for involvement dimensions Variable importance pleasure risk consequence risk probability sign value affective cognitive importance .54 0.03 .35 .43 .35 .47 pleasure .54 0.07 .27 .43 .42 .39 risk consequence 0.02 0.07 0.07 0.06 .152 .12 risk probability .35 .27 0.07 .24 0.07 .13 sign value .43 .43 0. 06 .24 .20 .42 affective .35 .42 .15 0.07 .20 .65 cognitive .47 .40 .12 .13 .42 .65 *. Correlation is significant at the 0.01 level (2 tailed).
78 importance and cognition dimensions between the two scales r = .47, n = 411, p = .00 als o held a medium to large effect size. Qualitative research with participant athletes could provide a fruitful area of exploration and would serve as a foundation for scale development. Trends identified within qualitative interviews might indicate anteced ents specific to sport involvement, which would be useful in creating items within multidimensional subscales. Other existing scale items could also be incorporated and tested. For example, the integration of job burnout scale items (Maslach & Jackson, 1981) could be incorporated and tested in a refined sport involvement scale. A study of other associations or relationships between involvement and behavioral indicators may be practical for future applications An example would be the study of involvement and its correlation to related sponsors associated with events The d ependent variable of sponsor awareness might offer a greater variation in response and present a higher opportunity for statistical significance with involvement measures. Regressions testing the relationship between involvement and media usage would also prove to be advantageous to event promoters and sponsors. The complexities of team sp orts would allow for profiling related to specific team roles Data extrapolated for drivers in a motorsports sample group could be compare d to th e involvement scores of crew chiefs an d other support personnel. Similarly, scores related to team captains in soccer or pitchers in baseball could also be extracted and studied for differences in invol vement indicators over other members of the team. The cognitive and affective scores of specific team individuals may also vary over the role of each participant and the responsibility of the participant within the sport or team framework
79 An opportunity exists to expand the study of individual and team sports to other study groups, where the potential for external validity will allow for gene ralization to other sport communities. I ndividual sports to be considered might be billiards, golf or recreationa l fishing where psychological and intellectual skills may be as predominant as physical ability. Team sports to be considered can include soccer and recreational softball. Another viable area of study for team sport r esearch would be sample respondent g roups from senior sports leagues. A study of involvement and the influence of age would prove enlightening.
80 APPENDIX A INDIVIDUAL SPORT INV OLVEMENT SURVEY The I nner M ind of U ltrarunning A thletes Little is known about the behavioral aspects related to u ltrarunning. The sport focuses on the physical side of athletic training, where injuries and dietary supplements dominate journalism and research. But unique to ultra endurance sport is the psychological component, allowing the mind to drive individuals forward in events that can take anywhere from 12 to 30 hours to compete; beyond that barrier are 5, 10 and even 30 day ultradistance races, where sleep deprivation and fatigue wreak havoc with the human body. What is the mental and emotional composition of this stout group of adventurers, and what motivates them to compete? The primary purpose of this study is to develop a measurement tool to benchmark the level of involvement in ultrarunning athletes, and then determine which antecedents are most prevalent in forming behavioral intentions towards attending ultrarunning events. The sport is attracting a broad range of adventuresome athletes, and an enhanced understanding of the behaviors associated with ultrarunning will assist event organizers, personal coaches, and the media in communicating with this elite niche of competitors. Informed Consent Protocol Title: Antecedents to behavior for ultrarunning athletes." Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to examine the behaviors of ultrarunning athletes, and determine the key indicators of involvement in ultra racing events. What you will be asked to do in the study: We have several questions we wish to answer regarding ultrarunning. We would like you to respond as completely as possible. Time required: Approximately 10 minutes Risks and Benefits: There are no risks or benefits associated with this study. Compensation: There is no compensa tion associated with this study.
81 Confidentiality: Your identity will remain anonymous. No names or other identifying information obtained will be associated with data or IP addresses, or emails will be saved. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from the study at anytime without consequence. Whom to contact if you have questions about the study: Tom Mueller, PhD student, Department of Advertising, 2039D Weimer Hall Gainesville, FL 32611, 3522731640. Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 326112250; phone 3523920433. Please acknowledge the informed consent above by checking the appropriate box: ___ I agree. ___ I do not agree. I will not participate in this study. How many years have you been ultrarunning? _________ What is the average number of miles you run each week? _________ What is the average number of days you run each week? _________ I have competed in a 50K (or longer) ultrarunning event in the past year. ___ Yes ___ No How many ultrarunning events did you enter and participate in throughout 2007? _________ In the past year, approximately how much have you spent (travel, entry fees, equipment, etc.) participating in ultrarunning events? _________ Please check the appropriate boxes regarding the ultrarunning events you have partic ipated in during the past year.
82 ___ 50K ___ 50 mile ___ 100K ___ 100 mile ___ 12hour ___ 24hour ___ Multi -day How many miles did you travel (one way) to your last ultrarunning event? _________ In the next year, how likely is it that you will participa te in an ultrarunning event? ___ Very likely ___ Somewhat likely ___ Neutral ___ Somewhat unlikely ___ Very unlikely It is my intention to participate in an ultrarunning event in the next year. ___ Strongly disagree ___ Somewhat disagree ___ Neutral ___ S omewhat agree ___ Strongly agree In the next year I plan to participate in an ultrarunning event. ___ Strongly disagree ___ Somewhat disagree ___ Neutral ___ Somewhat agree ___ Strongly agree What is your age (in years)? __________ Gender ___ Male ___ Female What is your annual gross income? _________ What state do you reside in? __________ What is your highest level of education? ___ High School ___ College Degree ___ Graduate Degree ___ Doctorate
83 What is your marital status? ___Single ___Married ___ Divorced ___ Widowed ___ Separated ___ Cohabitating Optimum nutrition is an essential component for successful ultrarunning event participation. ___ Strongly disagree ___ Somewhat disagree ___ Neutral ___ Somewhat agree ___ Strongly agree How upset would you be if you were unable to participate in ultrarunning events? ___ Not upset ___Somewhat upset ___Moderately upset ___Noticeably upset ___Extremely upset How often do you read magazines devoted to ultrarunning? ___Never ___Seldom ___Sometimes ___R egularly ___Constantly How often do you browse web sites devoted to ultrarunning? ___Never ___Seldom ___Sometimes ___Regularly ___Constantly How do you perceive your skill level in ultrarunning? ___Very low ___Beginner ___Intermediate ___Accomplished runner ___Very high Please consider the following statements, and choose the response that best depicts your opinion regarding the sport of ultrarunning.
84 When choosing to compete in ultrarunning events from among other activities I always feel confident tha t I will make the right choice. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Ultrarunning is important to me. ___ Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Ultrarunni ng never leaves me indifferent. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Competing in ultrarunning events gives a glimpse of the type of person I am. ___Strongly disagree ___Somewhat disagree ___Neutral ___S omewhat agree ___Strongly agree Ultrarunning interests me a lot. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree When I compete in an ultrarunning event it is like giving a gift to myself. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree I really enjoy competing in ultrarunning events. ___Strongly disagree ___Somewhat disagree
85 ___Neutral ___Somewhat agree ___Strongly agree Whenever I compete in ultrarunning events I am conf ident that it is the right activity choice. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Being an ultrarunner tells a lot about me. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Competing in ultrarunning events is pleasurable. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree I get annoyed if I participate in an ultrarunning event, and it proved to be the wrong event choice. ___St rongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree When I mistakenly prioritize an ultrarunning event from among other activities, it really matters to me. __Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat a gree ___Strongly agree I can tell a lot about a person by whether or not they are ultrarunners. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree
86 If, after I have competed in an ultrarunning event, my choice proved t o be poor, I would be upset. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree It is complicated to choose ultrarunning events over other activities. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat a gree ___Strongly agree Please relate your experience with participation in ultrarunning events to the pairs of words below. Please check the number in each pair that best describes your opinion. Participating in ultrarunning events is: 1. important 2 3 4 5 6 7. unimportant 1. boring 2 3 4 5 6 7. interesting 1. relevant 2 3 4 5 6 7. irrelevant 1. exciting 2 3 4 5 6 7. unexciting 1. means nothing 2 3 4 5 6 7. means a lot to me 1. ap pealing 2 3 4 5 6 7. unappealing 1. fascinating 2 3 4 5 6 7. mundane 1. worthless 2 3 4 5 6 7. valuable 1. involving 2 3 4 5 6 7. uninvolving 1. not needed 2 3 4 5 6 7. needed Thank you for your participation. Click the "submit" button below and you are finished!
8 7 APPENDIX B TEAM SPORT INVOLVEME NT SURVEY The I nner M ind of Team M embers in R ally C ar R acing Little is known about what makes rally car teams tick on a psychological lev el. The sport focuses on the performance of teams at regional and national rally car events. But unique to rally car competition is the relationships among team members over multiple days and across varying terrain and distances. The primary purpose of t his study is to measure the level of involvement among rally car team members and to determine what causes rally car enthusiasts to participate in rally car events. Rally car racing is attracting a broad range of individuals. A more complete understandin g of behaviors associated with rally car competition will assist event organizers, Rally America and the media in communicating and promoting the sport. Informed Consent Protocol Title: Antecedents to behavior for rally car team athletes." Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to examine the behaviors of rally car team athletes, and determine the key indicators of involvement in rally car r acing events. What you will be asked to do in the study: We have several questions we wish to answer regarding rally car racing. We would like you to respond as completely as possible. Time required: Approximately 10 minutes Risks and Benefits: There ar e no risks or benefits associated with this study. Compensation: There is no compensation associated with this study. Confidentiality: Your identity will remain anonymous. No names or other identifying information obtained will be associated with data or IP addresses, or emails will be saved. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty for not participating.
88 Right to withdraw from the study: You have the right to withdraw from the study at anytim e without consequence. Whom to contact if you have questions about the study: Tom Mueller, PhD candidate, Department of Advertising, 2039D Weimer Hall Gainesville, FL 32611, 3522731640. Whom to contact about your rights as a research participant in th e study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 326112250; phone 3523920433. Please acknowledge the informed consent above by checking the appropriate box: ___I agree. ___I do not agree. I will not participate in this study. How many years have you been rally car racing? _________ What is the average number of hours you commit to rally cars each week? _________ What is the average number of days you commit to rally cars each week? _________ I have been on a team that compet ed in at least one (1) rally car event in the past year. ___Yes ___No Which of the following best describes how you participate in rally car racing? Choose one response only. ___driver ___navigator ___mechanic ___fabricator ___marketing ___sponsorship ___media relations ___team coordinator ___event volunteer ___sanctioning body official Other: _________ How many rally car events did you participate in throughout 2008? _________
89 In the past year, approximately how much have you spent (travel, entry fees, equipment, etc.) participating in rally car events ? _________ Please check the appropriate boxes regarding the rally car events you have participated in during the past year. ___Regional rally events ___National rally events ___Rallycross ___TSD rallys __Rally Sprints ___Superspecial style events (i.e. X Games) ___Hill Climbs How many miles did you travel (one way) to your last rally car event? _________ In the next year, how likely is it that you will participate in a rally car event? ___Very likely __Somewhat likely ___Neutral ___Somewhat unlikely ___Very unlikely It is my intention to participate in a rally car event in the next year. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree In the next year I plan t o participate in a rally car event. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree What is your age (in years)? _________ Gender ___Male ___ Female What is your annual gross income? _________
90 What state do you reside in? _________ What is your highest level of education? ___High School ___College Degree ___Graduate Degree ___Doctorate What is your marital status? ___Single ___Married ___Divorced ___Widowed ___Separated ___Cohabitating Optimum nutrition and physical training is an essential component for successful rally car event participation. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree How upset would you be if you were unable to participate in rally car events ? ___Not upset ___Somewhat upset ___Moderately upset ___Noticeably upset ___Extremely upset How often do you read magazines devoted to rally cars? ___Never ___Seldom ___Sometimes ___Regularly ___Constantly How often do you browse web sites devoted to ral ly cars? ___Never ___Seldom ___Sometimes ___Regularly ___Constantly
91 How do you perceive your skill level in rally car team participation? ___Very low ___Beginner ___Intermediate ___Accomplished ___Very high Please consider the following statements, and c hoose the response that best depicts your opinion regarding the sport of rally car racing. When choosing to compete in rally car events from among other activities I always feel confident that I will make the right choice. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Rally car racing is important to me. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Rally car racing never leaves me indifferent. ___Strongly disagree ___S omewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Competing in rally car racing events gives a glimpse of the type of person I am. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Rally car racing int erests me a lot. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree
92 When I compete in a rally car racing event it is like giving a gift to myself. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree I really enjoy competing in rally car racing events. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Whenever I compete in rally car racing events I am confident that it is the right activity ch oice. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Being part of a rally car racing team tells a lot about me. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree Competing in rally car racing events is pleasurable. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree I get annoyed if I participate in a rally car racing event, and it proved to be the wrong event choice. ___Strongly disagre e ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree When I mistakenly prioritize a rally car racing event from among other activities, it really matters to me. ___Strongly disagree
93 ___Somewhat disagree ___Neutral ___Somewhat agree ___Str ongly agree I can tell a lot about a person by whether or not they are part of a rally car racing team. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree If, after I have competed in a rally car racing event, my cho ice proved to be poor, I would be upset. ___Strongly disagree ___Somewhat disagree ___Neutral ___Somewhat agree ___Strongly agree It is complicated to choose rally car racing events over other activities. ___Strongly disagree ___Somewhat disagree ___Neutr al ___Somewhat agree ___Strongly agree Please relate your experience with participation in rally car racing events to the pairs of words below. Please check the number in each pair that best describes your opinion. Participating in rally car racing event s is: 1. important 2 3 4 5 6 7. unimportant 1. boring 2 3 4 5 6 7. interesting 1. relevant 2 3 4 5 6 7. irrelevant 1. exciting 2 3 4 5 6 7. unexciting 1. means nothing 2 3 4 5 6 7. me ans a lot to me 1. appealing 2 3 4 5 6 7. unappealing 1. fascinating 2 3 4 5 6 7. mundane
94 1. worthless 2 3 4 5 6 7. valuable 1. involving 2 3 4 5 6 7. uninvolving 1. not needed 2 3 4 5 6 7. needed Thank you for your participation. Click the "submit" button below and you are finished!
95 APPENDIX C HAVITZ AND DIMANCHE S ADAPTATION OF LAURENT AND KAPFERER S INVOLVEMENT PROFILE (IP) SCALE (AS APPLIED TO ULTRA RUNNING PARTICIPATION)* __________________________________________________ Component Item __________________________________________________ Importance 1: Ultrarunning is important to me. Importance 2: Ultrarunning never leaves me indifferent. Importance 3: Ultrarunning inte rests me a lot. Pleasure 1: When I compete in an ultrarunning event it is like giving a gift to myself. Pleasure 2: I really enjoy competing in ultrarunning events. Pleasure 3: Co mpeting in ultrarunning events is pleasurable. Risk consequence 1: I g et annoyed if I participate in an ultrarunning event, and it proved to be the wrong event choice. Risk consequence 2: When I mistakenly prioritize an ultrarunning event from among other activities, it really matters to me. Risk consequence 3: If, after I have competed in an ultrarunning event, my choice proved to be poor, I would be upset. Risk probability 1: It is complicated to choose ultrarunning events over other activities. Risk probability 2: Whenever I compete in ultrarunning events I am c onfident that it is the right activity choice. Risk probability 3: When choosing to compete in ultrarunning events from among other activities I always feel confident that I will make the right choice. Sign Value 1: I can tell a lot about a person by whether or not they are ultrarunners. Sign value 2: Being an ultrarunner tells a lot about me. Sign value 3: Competing in ultrarunning events gives a glimpse of the type of person I am. items were randomly ordered throughout the second and final porti on of the online survey. From Kerstetter and Kovich (1997).
96 LIST OF REFERENCES Ajzen, I. & Driver, B. (1992). Application of the theory of planned behavior to leisure choice. Journal of Leisure Research, 24 (3), 207224. Andrews, J., Durvasula, S. & Akhter, S. (2001). A framework for conceptualizing and measuring the involvement construct in advertising research. Journal of Advertising, 19 (4), 27 40. Armstrong, K. (2007). Self, situations, and sport consumption: An exploratory study of symbolic interactionism. Journal of Sport Behavior, 30 (2), 111129. Ashforth, B. &, Mael, F. (1989). Social identity theory and the organization. The Academy of Management Review, 14(1), 20 39. Babbie, E. (2007). The practice of social research (11th ed.) Belmont, CA: Thomson Wasworth. Barendse, M. (1983). Individualism, technology, and sport: the speedway nexus. Journal of Sport and Social Issues, 7(1), 15 23. Bloch, P. (1981). An exploration into the scaling of consumers involvement with a produc t class. Advances in Consumer Research, 8, 61 65. Bloch, P., & Richins, M. (1983). A theoretical model for the study of product importance perceptions. Journal of Marketing, 47(2), 69 81. Brief rally background History Part 1. From World Rally Champi onship background. Retrieved on October 6, 2008 from: http://www.rallycars.com/Cars/Cars_Background1.html Cacioppo, J.T. & Petty, R.E. (1979). Effects of message repetition and positi on on cognitive response, recall and persuasion Journal of Personality and Social Psychology, 37, 97 109. Cacioppo, J.T. & Petty, R.E. (1984). The elaboration likelihood model of persuasion. Proceedings of the Association for Consumer Research, 11, 6 68 672. Celsi, R. & Olson, J. (1988). The role of involvement in attention and comprehension processes. The Journal of Consumer Research, 15 (2), 210224. Capella, M. (2001). Measuring sports fans involvement: The fan behavior questionnaire. Southern Business Review, 27(2), 30 36. Cortina, J. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98 104. Couper, M. (2001). Web surveys: A review of issues and approaches. Public Opi nion Quarterly, 64 (4), 464494.
97 Dimanche, F., Havitz, M. & Howard, D. (1991). Testing the Involvement Profile (IP) scale in the context of selected recreational and touristic activities. Journal of Leisure Research, 23(1), 51 66. Eccles, D. & Tenen baum G. (2004). Why an e xpert t eam is m ore t han a t eam of e xperts: A s ocial -c ognitive c onceptualization of t eam c oordination and c ommunication in s port Journal of Sport and Exercise Psychology, 26, 542 560. Eagley, A. & Chaiken, S. (1993). The psychology of attitudes. Orlando, F L : Harcourt, Brace Jovanovich. Fisher, R., & Wakefield, K. (1998). Factors leading to group identification: A field study of winners and losers. Psychology & Marketing, 15(1), 23 40. Fryer, J ( 2008). Equity firm takes over Pe tty Enterprises USA Today Retrieved June 17, 2008 from: http://www.usatoday.com/sports/motor/2008 06 112269494866_x.htm Fuller, C., Myerscough, F. (2001). Stakeholde r perceptions of risk in motorsport. Journal of Safety Research, 32 345 358. Funk, D., Ridinger, L. & Moorman, A. (2004). Exploring origins of involvement: Understanding the relationship between consumer motives and involvement with professional sport teams. Leisure Sciences 26(1), 35 61. Gibson, H. J. (1998). Active sport tourism: who participates? Leisure Studies, 17(2), 155. Global sports market to be worth $141 billion by 2012. Sport Business. Retrieved June 19, 2008, from www.sportbusiness.com/news/167017/global -sports -market to -be -worth 141-billion by 2012. Gwinner, K. & Swanson, S. (2003). A model of f an identification: A ntecedents and sponsorship outcomes. Journal of Services Marketing, 17 (3), 275294. Hair, J., Anderson, R., Tatham, R. & Black, W. (1998). Multivariate data analysis. New Jersey: Prentice Hall. Havitz, M. & D i manche, F. (1990) Propositions for testing the involvement construct in recreational and tourism contexts. Leisure Sciences, 12 19 195. Havitz, M. & Samdahl, D. (1996). Unpublished manuscript. University of Waterloo, Waterloo, Ontario. Hill, J. S. & Vincent, J. (2 006). Globalisation and sports branding: T he case of Manchester United. International Journal of Sports Marketing and Sponsorship, 5 213230.
98 Houston, M. & Rothschild, M. (1977). A paradigm for research on consumer involvement. Unpublished paper, gr aduate school of business, University of Wisconsin, Madison. Houston, M. & Rothschild, M. (1978). Conceptual and methodological perspectives in involvement. In S. Jain (Ed.), Research Frontiers in Marketing: Dialogues and Directions Chicago: American Marketing Association 184187. IEG Sponsorship: Sponsorship Spending (2007) International Events Group. Retrieved December 4, 2007 from: http://www.sponsorship.com/Resource s/SponsorshipSpending.aspx Jamrosy, U., Backman, S. & Backman, K. (1996). Involvement and opinion leadership in tourism. Annals of Tourism Research, 23(4), 908924. Jensen, T., Carlson, L., & Tripp, C. (1989). The dimensionality of involvement. Ad vances in Consumer Research, 16, 680689. Jones, P. (2005). Ultrarunners and chance encounters with absolute unitary being. Anthropology of Consciousness, 15(2), 39 50. Josiam, B., Smeaton, G., & Clements, C. (1999). Involvement: Travel motivation a nd destination selection. Journal of Vacation Marketing, 5(2), 167175. Kapferer, J., & Laurent, G. (1993). Further evidence on the consumer involvement profile: Five antecedents of involvement. Psychology & Marketing, 10(4), 347355. Karnazes, D (2006). Ultramarathon Man: Confessions of an All Night Runner. New York: The Penguin Group. Kerstetter, D. & Kovich, G. (1997). An involvement profile of Division I womens basketball spectators. Journal of Sport Management, 11, 234249. Krugman, H.E. (1 965). The impact of television advertising: Learning without involvement. Public Opinion Quarterly 29, 349356. Krugman, H. E. (1966). The measurement of advertising involvement. Public Opinion Quarterly, 30, 583596. Krugman, H.E. (1971). Brain wa ve measures of media involvement. Journal of A dvertising Research 11 (1) 3 9. Kyle, G., Graefe, A., Manning, R., & Bacon J. (2003). An examination of the relationship between leisure activity involvement and place attachment among hikers along the Appalachian trail. Journal of Leisure Research, 35(3), 249273.
99 Kyle, G., Kerstetter, D. & Guadagnolo, F. (2002). Market segmentation using participant involvement profiles. Journal of Park and Recreation Administration, 20 (1) 1 21. Lastovicka, J. & Gard ner, M. (1979). Components of involvement. In J.C. Maloney & D.M Gardner (Eds.) Attitude Research Plays for High Stakes (pp. 5373). Chicago: Ameri can Marketing Association Laurent, G. & Kapferer, J. (1985). Measuring consumer involvement profiles. Journal of Marketing Research, 22, 4153. Lipsky, R. (1979). The athleticization of politics: The political implications of sport. Journal of Sport and Social Issues, 3(2), 6188. Marsh, H., Balla, J. & McDonald, R. (1988). Goodness of -fit indexes in confirmatory factor analysis: The effect of sample size. Psychological Bulletin, 103(3), 391410. Maslach, C., & Jackson, S. (1981). The measurement of experienced burnout. Journal of Occupational Behavior 2 (2), 99 113. McDaniel, S. (1999). An investigation of match up effects in sport sponsorship advertising: The implications of consumer advertising schemas. Psychology and Marketing, 16(2), 163184. McGehee, N., Yoon, Y. & Cardenas, D. (2003). Involvement and travel for recreational runners in North Carolina. Journal of Sport Management 17(3), 305 324. McIntyre, N. (1989). The personal meaning of participation: Enduring i nvolvement. Journal of Leisure Research, 21, 167179. Medinger, J. (2008, March ). Ultrarunning 2007 year in review. UltraRunning, 27(9), 27. Meng, X., Rosenthal, R., & Rubin, D. (1992). Comparing correlated correlation coefficients. Psychological Bulletin, 111(1), 172175. Mitchell, A. (1979) Predicting c hoice with p reference f unctions e stimated from p erceptions o f brands within a product category. In Allan Shocker (Ed.) Analytic Approaches to Product and Marketing Planning (pp. 368387) Cambridge, MA: Marketing Science In stitute. National Center for Chronic Disease Prevention and Health Promotion. (2008). Pat terns and trends in physical activity. Retrieved October, 2. 2008 from h ttp://www.cdc.gov/nccdphp/sgr/chapcon.htm OCass, A. (2000). An assessment of consumers product, purchase decision, adver tising and consumption involvement in fashion clothing. Journal of Economic Psychology, 21, 545576.
100 Park W., & Mit tal, B. ( 1985) A theory of involvement in consumer behavior: P roblems and issues. Research in Consumer Behavior, 1, 201 231. Pham, T. (19 92). Effects of involvement, arousal, and pleasure on the recognition of sponsorship stimuli. Advances in Consumer Research, 19, 85 93. Ray, M. L. (1973). Marketing communication and the hierarchy-of -effects. In P. Clark (Ed.) New Models for Communicat ion Research (pp. 147176). Beverly Hills, CA: Sage. Roberts, J., & Kundrat, D. (1978). Variation in expressive balances and competence for sport car rally teams. Journal of Contemporary Ethnography, 7, 231251. Rothschild, M. (1979). Advertising str ategies for high and low involvement situations I n J.C. Maloney & B. Silverman (Eds.) Attitude Research Plays for High Stakes (pp. 7493) Chicago: American Marketing Association. Rothschild, M. (1984). Perspectives on involvement: Current problems and future direction. Advances in Consumer Research, 1 1 (1), 216 217. Running USA state of the sport series. (2008, December). Florida Running & Triathlon, p. 62 63. Salmon, C.T. (1986). Perspectives on involvement in consumer and communication researc h. In B. Dervin & M.J. Voigt (Eds.) Progress in Communications Sciences (pp. 243 268). Norwood, NJ: Ablex. Schmitt, N. (1996). Uses and abuses of coefficient alpha. Psychological Assessment, (8) 4, 350353. Schumacker, R. & Lomax, R. (1996). A beginner s guide to structural equation modeling. Mahwah, New Jersey: Lawrence Erlbaum Associates Shank, M. & Beasley, F. (1998). Fan or fanatic: Refining a measure of sport involvement. Journal of Sport Behavior, 2 1 (4), 435443. Sheehan, B. & Hoy, G. (1999). Using e -mail t o s urvey i nternet u sers i n t he United States: m ethodology a nd a ssessment Journal of Computer -Mediated Communication, 4(3). Retrieved September 10, 2008, from: http://jc mc.indiana.edu/vol4/issue3/sheehan.html Sherif, M. & Cantril, H. (1945). The psychology of attitudes, part 1. Psychological Review, 52 (6), 295319. Sherif, M. & Cantril, H. (1947). The psychology of ego involvement. New York: John Wiley & Sons
101 Sherif, M. & Hovland, C. (1961). Social judgment: Assimilation and contrast effects in communication and attitude change Santa Barbara, CA: Greenwood Publishing Group. Slama, M. & Tashchian, A. (1985). Selected socioeconomic and demographic charact eristics associated with purchasing involvement. Journal of Marketing, 49 (4), 7282. Sloan, L. (1989). The motives of sports fans. In J.H. Goldstein (Ed.) Sports Games and Play: Social and Psychological Viewpoints (pp. 175240). Hillsdale, NJ: Lawrenc e Erlbaum Associates Sport ing Goods Manufacturers Association (2009). Sports industry sales drop in 08; participation buffers major declines Retrieved July 9, 2009 from http://www.runblogrun.com/2009/03/sgma_survey_supports_sports_is.html Toor, R. (2008). Personal record: A love affair with running. Lincoln: U niversity of Nebraska Press. Trail, G., Robinson, M., Dick, A. & Gilletine, A. (2003). Motives and points of attachment: Fans versus spectators in intercollegiate athletics. Sport Marketing Quarterly, 12 (4), 217227. Traylor, M. (1981). Product involvement and brand commitment. Journal of Advertising Research, 21 (6) 51 56. Traylor, M. & Joseph, W. (1984). Measuring consumer involvement in products: Developing a g eneral s cale. Psychology and Marketing, 1 (2), 6577. Turner, J. C. (1982) Towards a cognitive redefinition of the social group. In H. Tajfel (Ed.) Social Identity and Inter -group R el ations (pp. 15 40). Cambridge, England: Cambridge University Press. Wang, A. (2006). Advertising engagement: A driver of message involvement on message effects. Journal of Advertising Research, 46 (4), 355368. Wann, D. (1995). Preliminary validation of the sport fan motivation scale. Journal of Sport and Social Sciences, 19 (4), 377396. Wann, S., Schrader, M., & Wilson, A. (1999). Sport fan motivation, questionnaire validation, comparisons by sport, and relationship to athletic motivation. Journal of Sport Behavior. 22(1), 114139. Ward, C. D. (1965). Ego involvement and the absolute judgment of attitude statements. Journal of Personality and Social Psychology, 2, 202208. Wiley, C., Shaw, S., & Havitz, M. (2000). Mens and womens involvement in sports: An examination of the gendered aspects of leisure involvement. Leisure Sciences, 22, 1931.
102 Wilson, C. (1999). Private email sent to Babbie in The practice of social research (11th ed.) Belmont, CA: Thomson Wa d sworth. Woodman, T., & Hard y, L. (2001). A case study of organizational stress in elite sport. Journal of Applied Sport Psychology 13, 207238. Zaichkowsky, J. (1985). Measuring the involvement construct. Journal of Consumer Research, 12, 341352. Zaichkowsky, J. (1994). The p ersonal involvement inventory: Reduction, r evision, and application to advertising. Journal of Advertising, 23(4), 59 70.
103 BIOGRAPHICAL SKETCH Thomas S. Mueller was born in Milwaukee, Wisconsin. He was a graduate of Wausau West High School in Wausau, W isconsin. He began his undergraduate education at the University of Wisconsin, Marathon Campus and complete d his B.A. degree in journalism at the University of Wisconsin Eau Claire. Tom was highly interested in motorcycles and began his career as an ass ociate editor at Cycle News East, a weekly tabloid newspaper in Atlanta, Georgia. He went on to form a publicity firm and contracted special events marketing with the Wrangler Brand and its p romotions in NASCAR, Supercross motorcycle racing and professional rodeo. Tom later became marketing director for Mercury Marines high performance division, then became executive director for the American Motorcyclist Associations professional racing su bsidiary. Tom returned to his entrepreneurial roots and created Sport Management, Inc., the agency he would own and operate for the next eight years. During that time Tom returned to academics and earned a MBA from Otterbein College in Westerville, Ohio. He would move on to become an account supervisor for Wasserman Media Group, where he managed the Nationwide Insurance mobile interactive tour and the Rally America motorsports account where Tom integrated auto racing into the 2006 ESPN Summer X Games. To m began his Ph.D. program in the College of Journalism and Communications at the University of Florida in August, 2006. He was awarded the 2009 Outstanding Student Teaching Award and participated in the mass communications study abroad program in 2007, 2008 and 2009. After completi ng his Ph.D. program in 2009, Tom relocated to Boone, North Carolina, where he is an assistant professor in the Department of Communication at Appalachian State University.