What?s in a Tip? An Exploratory Study of the Motivations Driving Consumer Tipping Behavior by Jeremy E. Whaley A dissertation submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Doctor of Philosophy Auburn, Alabama August 06, 2011 Keywords: consumer motivation, tip, service, social norm Copyright 2011 by Jeremy E. Whaley Approved by Martin A. O?Neill, Chair, Professor of Nutrition, Dietetics, and Hospitality Management Susan S. Hubbard, Professor of Nutrition, Dietetics, and Hospitality Management Alecia C. Douglas, Assistant Professor of Nutrition, Dietetics, and Hospitality Management ii Abstract According to Segrave (1998), since the late 1800?s, the study of tipping has provoked debate in a range of abstract dimensions such as economics, sociology, and psychology. To date, the studies have been largely qualitative in nature, while addressing motivating themes (service, social norm, and future service considerations) in isolation from one another. Following a thorough examination of the literature, there is a definite lack of research on the development and testing of a more holistic quantitative scale aimed at identifying the motivational Gestalt driving actual consumer tipping behavior. Therein lies the major theoretical contribution of this study, namely the development and testing of a Tipping Motivations Scale, which over three separate tests, identifies a number of drivers of consumer tipping motivation In this study exploratory and confirmatory factor analyses were conducted to test the empirical dimensions of consumer tipping motivations. The results obtained indicate a reasonable fit between the data and the proposed model across both analyses. As indicated, this was repeated on three separate occasions and the results largely remained consistent. The findings point to the key role of service in driving the consumer?s decision to tip. Other important factors included social conformity, the issue of future visitation, and server actions. It is concluded that future research is needed to explore whether these dimensions remain constant among other sample groups and across different tipped professions. . iii Acknowledgements I wish to acknowledge several people. Without their unwavering support, this project would never have been completed. My appreciation goes to Dr. Martin O?Neill, for his guidance and insistence that I finish this damn dissertation, move up from Southern poverty, and ultimately up in life; Dr. Alecia C. Douglas, for her statistical knowledge, detail, and passion for excellence in research which truly inspired me to do great work; Dr. Susan Hubbard, who, for so many years, has been an inspirational leader and mentor in my life; and Dr. Maria Witte, who always greeted me with a warm, encouraging smile, and truly gave me the confidence to believe in myself; Ms. Annette Smith, my former high school English teacher, and life-long friend, she served as my editor and tutor. My personal gratitude goes to Mrs. Lorene Guthery, who taught me compassion, empathy, and the true meaning of selfless giving; I owe my collegiate success at Auburn to her. My deepest gratitude goes to my wonderful parents, who have struggled all of their lives to make ends meet so their son could have a shot at the true American dream, liberated from the chains of oppression and poverty through the knowledge of education, and to Colby Doepel, my friend and partner. iv Table of Contents Abstract???????????????????????????................... ............ii Acknowledgments???????????????????? ...???....... ....................iii List of Tables????????????????????????????? .??? vii List of Figures? ????????????????????????...??????viii Chapter I. Introduction ???????????????????????????..?. 1 Background??????????????????????? .? ???... ....1 Problem and Statement of Significance? ?? ????.?? .???..?? ?.. ?... 4 Purpose of the Research??????????????... ??? ????.. ?... .5 Research Questions?????.?..??????..??? .?... ??? .? ??.. ?6 Definition of Terms?????????????..??? .??..??? ??? ...6 Limitations????..??????????..????? ??? ..??.?? ?..7 Summary?????????.??..?..? ..??????? ???????. ?9 Chapter II. Literature Review .????????????????????????..?10 Overview??????..????????????????? ?? ?? ...?10 Tipping: A Historical Context??????????????? ?? ? .?. ?..11 Consumer Behavior ?.??????..?????? ?? ???. ??? .??..15 Motivations for Tipping ???????..???? ?..? ???. ??? ? .?..30 Proposed Theoretical Framework ????????..?? ? ??? .??? .?.48 Research Considerations????????.?? .? ???. ? ?? .????....48 v Towards a Theoretical Framework ???.??? ???????? ?? ? .?5 0 Summary???? ????..???? ????????? .? ..?????..52 Chapter III. Methods ??????? ????????????????... ? ???.. ..53 Overview???????????????????????... ??? ?? 53 Method of Inquiry??????????????????...... ? .? .? ?? ..53 Research Questions???????????????? .?.. ??? ..?? ?..59 Plan of Research????????????????? .??? ??? ...??6 0 Qualitative Research Procedures?????????????? ...? ?? ? ?61 Focus Group Work??????????????? ????. .?? .???.61 One-on-One Interviews???????????????? ? .??.?. ??..64 The Research Instrument????????????? ??. ??? .?. ..? .?65 Pilot Work to Date???????????????? ??? ??. .??? ..65 Pilot (II) Research Findings??????????????? .?... ????..69 Final Research Instrument (TMS)??????????? ? ??????? .75 Sample?.???.?????????? .????.?? ?? ?. ..?????75 Concern for Ethics????????.???? .??? ?? ...?.. ?....???..76 Data Collection??????.??????..? ?? ?? .???????.?.78 Returned Questionnaires?? .??.?. .???... ?? ? ?? .???????...79 Summary?? ??????.???.????? ??.. ??? ???? .?.?..80 Chapter IV. Results and Findings?????.?????? ??? ?????? .?.?.?.81 Introduction.????????????????? ??? ?? ? ? ? .?.?.81 Sample Demographics?. ????????????.??? ??? ??.. ....?81 Measurement Instrument Properties?????.??????? ??. ??.?? 86 vi Validity and Reliability Tests?????????? ???????????..89 Factor Analysis?????????????.????? .? ..? ? ??.??? 89 Research Criteria????.. ? ??????? .?.??? ...? .? ? ? ?... .??.. 89 Tipping Motivations Scale (TMS)????..?.????.? .?.?. ?? ? ? .....? 91 Confirmatory Factor Analysis???????????..??.????? ? .......93 Research Questions and Testing of Central Research Hypothesis?? ...?? ............99 Reliability Checks?????. ? ??..??????.?..?? ...??????..101 Validity?. .? ?????????????????????? ...? ?..??102 Test of Non-Response and Late Response Bias? ..?????? ...??? ..??..103 Summary???.?????????????????? ...???????105 Chapter V. Conclusion??????????????.??????????.... ............106 Overview????????????????????.. ????? ??..?106 Description and Purpose of the Research?????????... ??? ???. ...107 Implications?????????????????... ??????? ?...?..109 Limitations??????. ???????????? ..? ??.. ? ..????...112 Future Research?????????? ? ??????? ...?? ? ? ?...??113 Conclusion???????????????????... ????? ?...??114 References.????????????????????????????? ???..115 Appendix A: Pilot 1-Tipping Motivations Scale?.? ? ??????????? ??? ....128 Appendix B: Pilot 2-Tipping Motivations Scale??????????? ??? ???.?130 Appendix C: Final Study- Tipping Motivations Scales...???????? ??? ? ???132 Appendix D: Tipping Motivations Scree-plot?? ??. ???????? ??? ????134 vii List of Tables Table 1: Examples of Theoretical Framework in Tourism and Tip Studies?? .?? ?? .? ...29 Table 2: Factor Analysis (Pilot 1)--Consumer Tipping Motivation Construct?? ? .................68 Table 3: Factor Analysis (Pilot 2)--Consumer Tipping Motivation Construct???.... .............74 Table 4: Demographic Profile of Respondents????????????? ? .....? ??.. ..84 Table 5: Tipping Behavioral Norms..?????????? ??.... ??.. ?? ? ?.. ??? 86 Table 6: Univariate Analysis of Tipping Motivation Scale??????????????... 88 Table 7: Rotated Component Matrix???????... ...???????? .? ??. ? .??93 Table 8: Coefficient Alpha of Scale????????????????????? ?? 102 viii List of Figures Figure 1: Consumer Decision-Making Process ?? ...???????????????? 16 Figure 2: Maslow?s Hierarchy of Needs ??????? ..?????? ??. ???.. ??20 Figure 3: Theory of Planned Behavior???????????? ..?. ???? ????... 23 Figure 4: A Proposed Model of Tipping Motivations? ?. ?????? ..????????5 1 Figure 5: Theoretical Model??? ? ? ? ?????????? .?? ...???..................94 Figure 6: Theoretical Model with Beta Weights...????? ...??? .? ? ???????95 Figure 7: Re-specified Theoretical Model with Beta Weights?? ....? ? ?..???????98 1 Chapter I Introduction Background The word tip means only one thing to most people when they think of the field of service, and that is the monetary compensation they should justly leave the person providing that service. A search for common definitions of the word results in the following designations: e.g. ?noun: gratuity, tip, pourboire, baksheesh, bakshish, bakshis, backsheesh--a relatively small amount of money given for services rendered as by a waiter; verb: tip, fee, bung--give a tip or gratuity in return for a service, beyond the compensation agreed on; Remember to tip the waiter; fee the steward? (Princeton University, ?about wordnet,? 2011). Tipping is a custom that directly affects hospitality workers? and restaurant guests? everyday lives. It is a vital way of life for many hospitality professionals around the world. ?Tipping is an important economic activity: annual tipping in the US food industry alone is estimated at $42 billion? (Azar, 2008, p. 1). Occupations such as restaurant servers, bartenders, hostesses, ma?tre d?s, concierges, housekeepers, bellmen, taxi drivers, hairstylists, exotic dancers, and night club entertainers, along with many others, derive a portion, if not all, of their monetary compensation from tips. 2 There is no question in United States restaurants whether consumers will tip, but rather when and how much they will tip. What is more interesting, however, is the reason why individuals tip. While it is hard to say with any certainty when and where the custom originated, tipping has frustrated fair wage and labor rights advocates, social scientists, economists, and reporters alike for more than a century. According to Segrave (1998), esteemed publications such as Gunton?s Magazine, Leisure Hour, and the New York Times explored the tipping culture as a social phenomena born in Europe, with hopes of dispelling the practice here in the U.S. Meanwhile, some in the growing public denounced the custom as an increasing nuisance and shunned the individuals who accepted tips; however, public discontent and outrage did little to curb Americans? appetites for receiving and ultimately giving tips (Segrave, 1998). Since the first university professor explored the subject, Crespi, 1947 (Segrave, 1998), academicians have continued to search for what are posited to be the primary drivers of consumer tipping behavior. Commonly researched motives believed to influence consumer tipping behavior include service quality (Azar, 2005; Bodvarsson and Gibson, 1999; Lynn, 2003), pressure to conform to societal norms (Azar, 2004, 2005, 2006, 2007; Boyles, Mounts, and Sowell, 2006; Conlin, Lynn, and O?Donoghue, 2003; Lynn, 2001; Shamir, 1984), and the investment in strategic and future service encounters (Azar, 2007; Azar and Yossi, 2006). On the other hand, researchers have found that while many truly believe it is the service received that drives their tipping behavior, Lynn?s (2001, 2003) work has forced a serious rethink of this attitude, as the service act has been found to minimally explain consumer tipping behavior. In addition, tipping has also been researched from an economical perspective (Azar, 2003, 2005; Bodvarsson and Gibson, 1999; Bodvarsson, Luksetich, and McDermott, 2003), a labor economics perspective (Azar, 2005), sociological phenomena (Segrave, 1998; Azar, 2004; 3 Babcock, 2007), and a psychological perspective (Cialdini and Goldstein, 2002; Boyles, Mounts, and Sowell, 2006; Azar, 2006). Such feelings as shame or disapproval are believed to be powerful mediating factors influencing societal norms. ?Consistent with the explanation of tipping, Crespi (1947) found that 34 percent of the tipping public he surveyed thought that fear of disapproval was the main reason that most people tip? (Lynn and Grassman, 1990, p. 170). Further research has examined racial differences in tipping behavior (Arnold, 2004) and tax evasion by tipped employees (Anderson and Bodvarsson, 2005). While the concept of tipping may at first glance appear simple, prior research falls short of simultaneously identifying the many realms of economic, socio-economic, and psychological dimensions which covertly influence the behavior. Tipping is a multi-dimensional, highly social behavior requiring an understanding of complex, interactive, controllable, and uncontrollable mitigating factors. Drivers such as economic exchange theory, equity theory, and diffusion of responsibility theory, coupled with strong emotions such as gratitude or guilt, influence restaurant tipping behavior. Economic exchange theory refers to the actually exchange of materials goods and service for monetary compensation, while on the other hand, equity theory refers to the balance of feelings of satisfaction/dissatisfaction with the product or service that is purchased. Diffusion of responsibility is a theory which is known to adversely affect the amount of individual responsibility which is felt within a group; as a group grows in size, individual responsibility is offset by the feeling that others will take care of the liability. When faced with the opportunity, no economists or social scientists have been able to truly pinpoint those motivations driving the consumer?s decision to tip. The phenomenon presents an interesting subject for study, as the majority of research to date has addressed related motivational concepts in isolation, mostly from a qualitative perspective. This research adds to 4 the debate in that it reports on the development and testing of a more holistic quantitative scale aimed at identifying a comprehensive motivational Gestalt driving consumer tipping behavior. Problem and Statement of Significance Currently, there is a broad body of literature which conceptualizes tipping motivation and its relationship with such fields as sociology, economics, and psychology. Researchers have proposed theoretical models which contend that service quality and societal norms are plausible consumer motivations for tipping behavior (Conlin et al, 2003). As the tipping literature has grown and developed, plentiful qualitative literature asserts what may or may not be drivers of consumer tipping behavior. A concerted effort to explain possible motives for a variety of tip settings has occurred. For example, survey research was conducted by Lynn (2006) using a national telephone survey. Results of the study found that individuals who were ?white, in their 40?s to 60?s, highly educated, wealthy, and living in metropolitan areas in the Northeast? (p. 740) were most knowledgeable about tipping. Historically, tipping evolved from a mere reward for peasant workers in the 1400?s (Segrave, 1998) to a mandated social institution impacting the lives of millions as a primary source of income. Restaurants, hotels, hair salons, barber shops, and other areas are directly affected venues which have become ultimately dependent upon it in both economic terms and in the lives of growing societies. Previous researchers in the area of tipping have attempted to explain tipping motivation and behavior with the use of theoretical models addressing such concepts as service, social norms, server actions, gender interaction and more. Despite these efforts, it appears that no one has developed a sound quantitative measure which utilizes factor 5 analysis (EFA/CFA) or structured equation modeling (SEM) to help explain tipping motivations. Because of the unique nature of tipping motivations and actual tipping behavior, this lack of a scaling instrument designed specifically for such situations is a definite weakness in the literature to date. Given this fact, the overall focus of this research will be to develop and test an evaluative measure of consumer tipping behavior, primarily focused on restaurant service provision, and then validate this scale using both exploratory and confirmatory statistical techniques. Purpose of the Research The current debate to identify the psychological underpinnings of consumer tipping motivations represents a very notable cause. This study will venture to gain a better understanding of the complex array of motivations actually driving consumer tipping behavior. To date, hospitality literature abounds with studies which advocate the sheer importance of measuring motivational intentions in relation to future behavioral intention. This study will review extant literature which will identify a number of motivational factors as being central to the consumer?s decision to tip or not. These factors include but are not limited to the service act, conforming to societal norms, server actions, and the issue of securing future service. As a matter of investigative protocol, one central research hypothesis was developed and will be presented for analytical testing. The theoretical backing for this hypothesis will be presented as well as the statistical confirmation that lends support to, or rejects, it. This project is intended to serve as the basis for further research in the area of consumer tipping behavior. 6 Research Questions Against this background the overriding research question (RQ1) for this study is presented as follows: What motivational factors drive/influence the consumer?s decision to tip? To this end the study is intent on shedding light on the actual factor structure of the motivational tipping construct and the degree of variance explained by each factor uncovered. Additional research questions are presented as follows: RQ2: To what extent will actual service delivery affect the consumer?s decision to tip or not? RQ3: To what extent will consumers continue to tip regardless of the quality of service received? Definition of Terms The following definitions are provided in order to give clarity and, above all, to support the meaning of the language used in this research: Motivation--?to give reason, incentive, enthusiasm, or interest that causes a specific action or certain behavior? (Pan, 2011, para. 1). Behavior--?the manner of conducting oneself: anything that an organism does involving action and response to stimulation: the response of an individual, group or species to its environment having requisite or adequate ability or qualities? (Corley, n.d., para. 1). Tip--?a gift or a sum of money tendered for a service performed or anticipated? (Tip, 2011, para. 1). Social Norm--?social norms are the behaviors and cues within a society or group. This sociological term has been defined as ?the rules that a group uses for appropriate and 7 inappropriate values, beliefs, attitudes and behaviors. These rules may be explicit or implicit. Failure to follow the rules can result in severe punishments, including exclusion from the group.? They have also been described as the ?customary rules of behavior that coordinates our interactions with others?? (Norm, n.d., para. 1). Service--?intangible products that are not goods (tangible products), such as accounting, banking, cleaning, consultancy, education, insurance, know how, medical treatment, transportation. Sometimes services are difficult to identify because they are closely associated with a good, such as a combination of a diagnosis with the administration of a medicine. No transfer of possession or ownership takes place when services are sold, and they cannot be stored or transported, and [are] instantly perishable, and come into existence at the time they are bought and consumed? (Service, n.d., para. 1). Limitations Every effort to minimize limitations has been made. There is no doubt, however, that the project does have limitations. The following information is intended to reveal some of those mishaps in an effort to aid in future research. While an effort was made to attain a sample group that was representative of the entire population as it relates to normative consumer tipping behavior, the high number of student respondents may or may not represent the total population. This calls into question the results of this study and how well it can be generalized to other populations. Another limitation is the inability of the researcher to collect surveys in a truly random fashion. For instance, in one pilot study (spring 2009), surveys were distributed to students in the 8 College of Human Sciences and the College of Business for convenience sampling. With hopes of attaining an older, more representative population, a second pilot study targeted tailgating fans at four Auburn University home football games during the fall of 2009. And finally, a last survey was distributed to individuals visiting Auburn University?s campus in the spring and fall of 2010-2011. It should also be stressed that several factors which were out of the researcher?s control led to a number of road blocks in data collection. One of the main issues experienced was the ease of utilizing a convenient sample such as students, and then targeting tailgating football fans. While students provided a captive, attentive audience, tailgaters, on the other hand, were difficult to survey. During a number of the home games, rainy weather played a significant part in hindering data collection. Whether or not the home team was going to win or lose also seemed to play a part in tailgaters? willingness to participate. Some tailgaters clearly did not want to be bothered by someone asking them to fill out a survey. On the other hand, some were very helpful, understanding and happy to oblige. While the level of tipping motivation for Auburn students, Auburn football fans, and visitors to Auburn University?s campus has been measured, there is no way to compare these levels with respondents from other schools. However, visiting fans from the opposing teams did complete surveys. This may lead to a statistical finding that is significant for individuals from other states or other parts of the country. Also, the approach in which the surveys were administered provided a somewhat awkward situation. As part of the survey process, respondents were approached at random by students carrying a back pack and asked if they would quickly commit to completing a survey. Because some participants may have been actively involved in the enthusiasm generated by the 9 surrounding pre-game and post-game activities, it may have been difficult for them to respond by truthfully and seriously considering their own tipping motivations. Summary In summary, this chapter has presented both an outline of and a justification for the study of consumer tipping motivations. The need and reasoning for the study have been identified along with some of the characteristics of the scale which have been used in the study. The research significance has been addressed, along with vital information about the sample groups. Key terms used in the research and limitations of the study have been given. The mere act of tipping is simple. Service is given; a tip is received. The barter of good will and trust between the server and a guest represents an economic transaction. Tipping is a multi-dimensional behavior which requires an understanding of social science, psychological science, and economics in order for one to fully appreciate the tradition. The following chapter is a wide-ranging review of the significant literature as it relates to both consumer motivation and the consumer behavior of tipping. 10 Chapter II Literature Review Overview In the past, countless researchers have conscientiously worked in order to define and ultimately understand human behavior. Comprehending the fundamental drivers of any behavior, whether good, bad, or indifferent, is the pinnacle of sociological research. Clearly, substantiated or unsubstantiated feelings are on the forefront of consumer behavior. It is a belief of the researcher that understanding consumers? motives and actions can only prove a worthwhile endeavor for operations managers, especially those in the hospitality industry. The foundation of the chapter is an in-depth review of the literature pertaining to tipping as a complex social phenomenon as well as the motivations for consumer engagement in this practice. The chapter will begin with a review of the pertinent literature on consumer behavior and, in particular, the role and influence of motivation on actual behavior. It will then proceed to offer an examination of tipping from a historical perspective and address the paramount role the custom has played in the lives of many to date. This review will draw attention to the way in which the custom has certainly bewildered and downright frustrated many cultures around the globe. A brief review of the motivational and behavioral constructs will provide a further 11 foundation and framework for this particular project. The chapter will offer a sound basis as to the sheer importance of the reasons why tipping motivations should be examined in the first place. More importantly, in-depth provisions will be offered with hopes of fully satisfying those who may need additional proof of the tipping juggernaut and its impact on the hospitality profession. In addition, this section will take a look at several tipping research studies while considering different tipping motivations believed to influence tipping behavior. Potential tipping motives which will be explored in this chapter are service, social norm, server actions, and future behavior intentions. Research from the areas of hospitality, consumer behavioral science, psychology, and economics will also be included in this review. Tipping motivations as a construct and as a topic of research will be carefully scrutinized. Prior research on tipping motivations will serve as the basis for the theoretical framework employed in this study. The covert dimensions of tipping, such as compliance with societal norms and powerful motivating emotions, will be addressed, since, undoubtedly, consumer behavior cannot be understood without an assessment of motivation. Furthermore, this chapter will explore elements of tipping motivations, with the support of prior empirical findings necessary to understand commitment to tipping behavior and the justifications of such motives. The chapter concludes with a synopsis which incorporates all afore-mentioned concepts under investigation. Tipping: A Historical Context No one knows exactly when the act of tipping began; however, when considering the origins of tipping, ?Hemenway (1993) claims that tipping was known as far back as the Roman 12 era and is probably much older? (Azar, 2007, p. 252). During Medieval times, tipping was used to reward slaves and servants. In England, brass urns and buckets within coffee houses and pubs bore the inscription ?To Insure Promptness? (Azar, 2007, p. 252). The act of tipping also took place during times of Medieval travel when ?lords met groups of beggars along their way; they tossed the beggars coins in an attempt to buy a safe passage? (Azar, 2004, p. 750). Meanwhile, the culture of tipping continued to grow, becoming a common practice among aristocrats and more privileged classes. According to Segrave (1998), who was a major contributor to nineteenth century tipping lore, ?during an age of paternalism and wardship, the master or lord of the manor might give his servant or laborer an extra few coins, from either passion or appreciation of a good deed? (p. 1). Before long, tip-giving consumed the hospitality culture in Europe where ?vails soon became expected from every guest that dined at the master?s table and slept in one of his beds? (Segrave, 1998, p. 1). In Europe, tips were given to workers in several occupations, all of which were characterized as subservient. Tipped positions included restaurant waiters, concierges, bellmen, chamber maids, and railroad porters; however, the phenomenon of tipping these individuals was regarded as a double-edged sword. On the one hand, tipping offered owners and managers a way of paying substandard or menial wages, while on the other hand, the public reluctantly provided primary compensation to servants. However, for traveling guests or restaurant patrons in Europe, tipping became an unwanted nuisance, troublesome to the point at which the custom became increasingly aggravating for those who tipped. With the onslaught of extra money, service workers became consumed with the notion of instant gratification for their work, even if the wages were miniscule and sporadic. As a result, it was not long before the culture became 13 increasingly influential as it travelled across the Atlantic, only to become ?a quintessential American trait? (Segrave, 1998, p.1). Although the founding fathers of the United States may have intended that all men were created equal in America, tipping simply defied this notion. Moreover, tipping actually condoned the thinking of those who felt that a distinct difference existed among the classes. Servants were simply servants. A tipped profession was servants? work. Once Americans were inoculated with the tipping culture, Europeans would pay the ultimate price. Traveling Americans were seen as very generous when touring throughout Europe. It was not uncommon that ?Americans were blamed for spoiling the servants to the extent that English people of modest means were not able to accept as many invitations as before for fear of being insulted by the servants in case they undertip? (Segrave, 1998, p. 3). During the nineteenth and twentieth centuries in America and Europe, tipped professions exploded in numbers due in part to their industrial revolutions. ?Industrial capitalism brought with it an increase in commercial eating and drinking establishments, hotels, and mass transportation, wherein those who received tips--maids, valets, waiters and so forth, were found in large numbers? (Segrave, 1998, p.5). Afterwards, groups began to protest the custom, and unions came to demand that workers be paid a fair day?s pay for a fair day?s work. In the wake of these demands, civil unrest continued to grow among the public sector and tipped employees. Servers revolted, causing violent riots, all in the name of tips--who should give them, and who should receive them (Segrave, 1998). Furthermore, European and American governments began to enter the tipping debate with the intent of regulating the phenomenon, while feeling that thousands, possibly millions, were eluding taxes. This did not work; servers continuously underreported their tip incomes. 14 Servers? willingness to do this caused the U.S. government to fine and incarcerate waiters from the esteemed Waldorf Astoria (Segrave, 1998). Unsurprisingly, local governments quickly recognized tips as taxable income; they were reluctant, however, to consider tip income for social security benefits. In an attempt to level the playing field, the custom of tipping after services shifted towards that of tipping in advance or in the form of an added service charge or house fee. In many cases, advanced tipping proved to be inefficient, as servants expressed little need to provide further assistance. However, today Brenner (2001) claims that ?the advanced tip is the most effective method for assuring results? (Azar, 2007, p. 256). While some found it to be a waste of time, seeing that the service was to be provided first, advanced tipping acted as barter for trust, good will, and the prior recognition of a servant?s work. The advanced tip was used in more affluent venues, such as exclusive hotels and restaurants which did not pay set wages for their servers. Furthermore, some venues expected wait staff to depend solely on tips, rather than paying them a set wage (Azar, 2004). As a result, the custom was continuously met with civil unrest and outright disgust and ridicule for those who accepted tips. Perhaps because many in the country relied on slaves for service until after the Civil War (1861-65), tipping became a custom in the U.S. in the late 1800?s. According to Azar (2004) ?tips were not given in the United States until after the Civil War, possibly because the country did not have a servant class? (p. 752). Presently, the phenomenon has become such a wide- spread and common practice that it would be unimaginable for the custom to cease. In the United States particularly, almost all services industries require, if not demand, millions of workers to be compensated via tips. Given that the act of tipping is a powerful, common socially accepted occurrence, it is with little surprise that the topic has been the study of numerous social psychologists as well as economists. It is the goal of this upcoming section to explore 15 motivational theory and chronologically review much of the academic literature, as well as provide emergent theories as to why individuals are motivated to tip. Consumer Behavior Consumer behavior provides a captivating topic for researchers in the social sciences. Although a company may develop and then launch the product, consumers may or may not select it for purchase. Ultimately, the over-arching rationale for the success or failure of a product generally depends on how well a business has performed its market research. According to Palmer (2004), ?the company may have failed to understand the complex processes by which buyers make purchasing decisions? (p. 87). Moreover, entities may fall short of the mark of success if critical factors of consumer behavior are not considered. Understanding the way in which consumers behave provides a critical pre-game requirement for development of products and/or services. As defined by Solomon and Stuart (2003), ?consumer behavior is the process individuals or groups go through to select, purchase, or use goods, services, ideas, or experiences? (p. 161). The authors also posit that while many consumers may make decisions in a truly spontaneous fashion, important purchases generally follow a consumer decision-making process (Solomon and Stuart, 2003). The process is best described by the model which is provided in Figure 1. Two primary factors that should be considered in consumer behavior are the physical and psychological dimension of needs. 16 Fig. 1. Consumer Decision-Making Process Solomon and Stuart (2003), p. 161. Any behavior, in its most primitive stage, begins with some sort of a need. In terms of consumer behavior, the realization of a desire triggers the buying process (Palmer, 2004). Needs are driven by a physical or psychological state of deficiency. Palmer (2004) defines a need as a ?perceived state of deprivation, which motivates an individual to take action to eliminate that sense of deprivation? (p. 90). As the process continues to drive consumers to act, needs eventually manifest themselves in wants. Wants present the second phase in the consumer behavior process by which ?needs are shaped by culture and individual personality? (Kotler, Bowen, and Makens, 2010, p. 12). Want is the catalyst which stirs demand; demand is the ultimate goal of the buying process. In addition, demand is the crossroad where needs and wants merge with financial resources that can satisfy a desire. For sales personnel and marketers, the generation of demand is the be-all, end-all of the equation which generates financial opportunity. Problem Recognition Information Search Evaluation of Alternatives Product Choice Post-purchase Evaluation 17 However, without the recognition of a need, which then creates a want that ultimately drives demand, the consumer purchasing process would be nothing more than a physical practice. For instance, hunger represents a physical need for food and nourishment. On the other hand, a car collector may want a Lamborghini, which is more psychological and motivated by the need for self-esteem. Obviously, some needs are much more than physical. In the end, consumers experience satisfaction or dissatisfaction with the purchase decisions which they have made. In many instances, consumers are satisfied with their purchases without general regret or noticeable displeasure in the buying experience. On the other hand, some purchase decisions consumers make provide disconfirmation, or something known as the cognitive dissonance theory; ?when our thoughts and actions are inconsistent, we experience great discomfort? (Bourne and Russo, 1998, p. 438). It is important to note, however, that satisfaction or dissatisfaction with a product or service depends on the amount of involvement or perceived risk a consumer experiences during the buying process (Solomon and Stuart, 2003). Now, more than ever, consumers are inundated with countless choices with which to satisfy physical needs. Considering the physiological and psychological framework of needs and wants, further investigation into motivation is critical to understanding consumer behavior. Consumer behavior is influenced by dimensions such as internal influences, social influences, and situational influences. Perception represents an internal influence of consumer behavior. ?Perception is the process by which people select, organize, and interpret information from the outside world? (Soloman and Stuart, 2003, p. 167). Perception involves the most basic sensory evaluation of stimuli such as light, color, and sound quality, all of which use the five senses. As a result, consumers selectively retain or distort information based upon their needs and wants. The amount of continuous exposure to a particular stimulus almost always affects a consumer?s 18 affinity for a particular product or service. That person?s perception is based on reality, meaning simply that perception of a product or service may be dependent upon one?s prior experiences. In addition to perception, motivation is another internal factor to consider in consumer behavior. ?Psychologists/social psychologists generally agree that ?a motive is an internal factor that arouses, directs and integrates a person?s behavior?? (Iso-Ahola, 1982, p. 257). In hospitality and social psychological sciences, assessing and defining consumer motivation is the hallmark of academic research. Consumer sciences are studied by identifying and characterizing those motives and/or behaviors which ultimately arouse consumers. For over half a century, psychologists have been puzzled by motivation, behavior, and the drivers which contribute to individual behaviors (Maslow, 1945; Murray, 1964). They query the motives and drivers of certain behavior. They question whether these human behaviors are abstract, uninterruptable phenomena or measurable constructs. From prior research, it is evident that human needs are fundamental and basic in nature (Maslow, 1945). Driven by disequilibrium and the need for a hierarchy, self-actualization wants and desires are ultimately predictors of human behavior (Maslow, 1945). As posited by Iso-Ahola (1980), ?a motive is an internal factor that arouses, directs, and integrates a person?s behavior? (p. 230). From the consumer?s perspective, motivation is driven by basic needs and desires for products or services. The gaps between being satisfied or dissatisfied, fulfilled or unfulfilled, create motivations and intentions. Motivational influences for consumers can be innate or learned from never feeling fully satiated with a desired product or service. Moreover, feelings may be accompanied by strong, uncontrollable emotions. Other triggers can be described as the need for social approval or a simple utilitarian need. Examples include a spouse kissing a husband or wife in public in order to affirm love and the common act of refilling an automobile 19 with fuel. It is important to note that emotions such as happiness or sadness, anger, guilt or shame, coercion, disdain, concern, fear, empathy, or passion can play pivotal roles in motivation, the creation of needs and wants, and ultimately, behavior. There are two notable theories of motivations, Maslow?s hierarchy of needs and McClelland?s theory of learned needs. Maslow?s hierarchy of needs begins with basic physical needs (food, water, shelter), then the need for safety, followed by the need for belongingness, self-esteem, and ultimately, self-actualization (Maslow, 1945). McClelland?s theory focuses on an individual?s need for achievement (to get ahead), affiliation (need for friendship and association with others), followed by the need for control (power over others), and lastly, the need for individuality or uniqueness (McClelland, 1965). ?Intentions are assumed to capture the motivational factors that influence behavior; they are indications of how hard people are willing to try, of how much of an effort they are planning to exert, in order to perform the behavior? (Ajzen, 1991, p. 181). An example of Maslow?s Hierarchy is provided in Figure 2. 20 Fig. 2. Maslow?s Hierarchy of Needs Solomon and Stuart (2003), p. 169 In addition to perception and motivation, consumers are internally influenced by learning. Solomon and Stuart (2003) state that ?learning is a change in behavior caused by information or experience? (p. 169). For example, consumers search out information about a particular product or service or learn from a simple interaction with a product or service. Resultantly, psychologists have ?advanced several theories to explain the learning process? (Solomon and Stuart, 2003, p.169). According to Sheth, Mittal, and Newman (1999), ?when people talk about learning, they often are thinking of cognitive learning, or acquiring new information from written or oral communication? (p. 310). In addition, the authors posit that classical conditioning, instrumental conditioning, and modeling represent three additional modes of learning (Sheth, Mittal, and Newman, 1999). Classic condition is described by pairing two external stimuli in order to produce a conditioned behavior. For example, a patient may exhibit a conditioned response as a doctor enters the room with a needle or begin crying as pain from a wound begins to increase Physiological Safety Belongingness Ego Needs Self-Actualization 21 (Bourne and Russo, 1998). On the other hand, instrumental conditioning is a learned response as a result of positive reinforcement. For example, a consumer may tend to frequently visit the same restaurant because of great service or buy a product because it positively reinforces his or her self-esteem. The fourth mechanism of learning is modeling. Consumers learn by directly observing and imitating the behavior of others. Furthermore, consumers are internally influenced by their attitudes, personalities, age, and lifestyles (Solomon and Stuart, 2003). Situational influences such as physical environment and time cannot be discounted as factors which drive consumer behavior. The actual physical environment of any venue can influence that behavior. Consumers are generally looking for an experience, but if the venue is overly crowded, hot and noisy, disconfirmation or dissatisfaction may occur. Another way of looking at the concept is that motivation is the antecedent; behavior is the consequence (Ajzen and Fishbein, 1977). Behavior is the resultant action which is taken to balance the disequilibrium of the motivation. Many human behaviors can be predicted by inner motives such as desire, which drives individuals to meet a desired goal (Gleitman, 1986). Additionally, certain behavior has been thought to categorize certain actions in specific situations in the context of internal and external loci of controls (Rotter, 1954, 1966). Social scientists are driven to study the middle abstracts of the behavioral spectrum. On the one end, there are the cognitive and physiological processes, while on the other end lies the influence of social institutions and interactions (Ajzen, 1991). ?Concepts referring to behavioral dispositions, such as social attitude and personality trait, have played an important role in these attempts to predict and explain human behavior (see Ajzen, 1988; Campbell, 1963; Sherman & Fazio,1983)? (Ajzen, 1991, p. 179). 22 More importantly, there are applicable theories as to why individuals are driven to behave as they do. As posited by Ajzen, ?behavior is followed by presentation of a theoretical model--the theory of planned behavior--in which cognitive self-regulation plays an important part? (Ajzen, 1991, p. 180). The theory of planned behavior is an addition to Ajzen and Fishbein?s reasoned action theory (Ajzen and Fishbein, 1980; Fishbein and Ajzen, 1975). These theories explain that behavior is the product of the humanistic controllable environmental factors (resources such as money and time) and the ability of the individual to act upon them. Moreover, an aggregation of things, such as general dispositions to controlled or uncontrolled environmental stimuli, attitudes and beliefs, and personality traits, should all be considered as sequential, interdependent factors when assessing an individual?s behavior. The variables and their interrelationships are clearly seen in the following model (Figure 3). 23 Fig. 3. Theory of Planned Behavior Ajzen (1991), p. 182 Many social psychologists from the fields of consumer behavioral sciences have sought to investigate the topic of motivation (Crompton and McKay, 1997; Iso-Ahola, 1980; Iso-Ahola, 1982; Kim, Goh, and Yuan, 2010; Kim and Prideaux, 2005; Klenosky, 2002; Snepenger, King, Marshal and Muzaffer, 2006). The primary focal point of these authors? studies has been the motivational push-pull relationships in tourism-oriented behaviors. The push refers to the psychological utility or disutility individuals feel when preparing for a trip. The psychological dimensions which individuals pursue are factors such as personal seeking, personal escape, interpersonal seeking, and interpersonal escape (Snepenger et al. 2006). ?Conversely, pull factors refer to those forces that influence a person?s decisions about what specific destination is selected? (Kim and Prideaux, 2005, p. 349). Attitude Toward Behavior Subjective Norm Perceived Behavioral Control Intention Behavior 24 Another way of looking at the push/pull relationship is to consider that the push factors are more psychological, whereas the pull factors are more tangible, like natural attractions, state parks, or other specific tourist destinations. ?Considerable effort has been undertaken by motivation scholars to document and quantify general and specific motivations for tourism (Crompton, 1979; Yuan & McDonald, 1990; Pennington-Gray & Kerstetter, 2001; Sirakaya, Uysal, and Yoshioka, 2003; Pearce & Lee, 2005)? (Snepenger et al. 2006, p. 141). Furthermore, scholars in tourism have utilized such methods as ?personal interviews (Yuan & McDonald, 1990; Crompton, 1979), [and] descriptive studies using surveys, and exploratory factor analysis investigations (Card & Kestel, 1988; Dunn Ross & Iso-Ahola, 1991; Sirakaya, Uysal, and Yoshioka, 2003)? (Snepenger et al. 2006, p.141). The following information will briefly discuss the types of scales utilized in these studies in order to provide strength and examples relevant to this study. Notably here, there are several different measurement scales which social scientists have used in order to study motivation in the context of tourism and economics. The first study, by Snepenger et al. (2006), utilized a measurement scale based upon Iso-Ahola?s four dimensional model. This method characterized behaviors indentified as escapers and seekers (Snepenger et al. 2006). ?The items were measured using a 10-point response format ranging from 1=low motivational fulfillment to 10=high motivational fulfillment? (Snepenger et al., 2006, p. 142). The objective of the authors? study was to determine consumers? need to confront internal imbalances, whereby satisfying those imbalances with the tangible offerings of a specific tourism destination (Snepenger et al. 2006). Furthermore, the study sought to ?operationalize within the tourism and recreation contexts Iso-Ahola?s theory by scaling each of the four motivational dimensions? (Snepenger et al. 2006, p. 141). While utilizing a repeated measures design, two groups, four different scenarios, and a convenient 25 sample of students from a land grant university, the study found that generalizability to other populations and statistical reliability was strong. The researchers, having held demographics and personality traits constant, conducted a factor analysis on two hypothetical situations, along with two more separate factor analyses on two vacation experiences. The results of the factor analysis lent support to Iso-Ahola?s model and loaded across four latent constructs with ?reliability alpha of the four factor groupings of motivation ranging from .61-.84 for both analyses? (Snepenger, et al. 2006, p. 142). Sixteen indices were created by combining the variation of motivational factors of personal seeking, personal escape, interpersonal seeking, and interpersonal escape by four activities: sporting events, beaches, amusement parks, and natural parks. Having used factor analysis and the competing models approach, the authors found that tourist motivations were indeed multi-dimensional in nature and could not be defined by only one motivational construct. Many factors were at play in explaining tourist motivations. The methodology and statistical approaches used in their research ultimately supported their study. Factor analysis, along with structured equation modeling (SEM), is utilized in this study of tipping motivations. The second study, by Kim and Prideaux (2005), utilized a twenty-one item instrument to measure motivations for visiting Korea. While the authors based their scales on the same method of identifying push and pull factors, they also sought to identify motivational factors based on nationality (Kim and Prideaux, 2005). Divided into groups according to nationality, ?US, Australian, Japanese, Chinese (Mainland), and Chinese (Hong Kong SAR)? (Kim and Prideaux, 2005, p. 350) participants responded to a variety of Korean motivational travel items. The survey was composed of ?Likert-type scales where 1= strongly disagree, 3= neutral, and 5= strongly agree. Overall images of Korea before and after a visit were operationalized with 5-point Likert 26 scales? (Kim and Prideaux, 2005, p. 350). Additionally, the authors intended to identify different motivational factors such as ?tourist resources (activities), length of planning before traveling, information sources used, and length of stay? (Kim and Prideaux, 2005, p. 347). To analyze the proposed scale, with hopes of measuring tourist motivations, a factor analysis was used. Reliability coefficients were analyzed and domains were extracted. ?One-way analysis of variance tests (ANOVA) were subsequently conducted on motivations for visiting Korea? (Kim and Prideaux, 2005, p. 351). As a result, the authors found that five factors indentified the tourist motivations to visit Korea. The five factors were ?enjoying various tourist resources, culture and history, escaping from everyday routine, socialization, and social status? (Kim and Prideaux, 2005, p. 351). Each nationality differed on its specific motivations to visit Korea, thus adding to the debate that motivational factors were, in fact, based on cultural affiliation. Different reasons for visiting Korea were dependent upon the nationality of those who visited. Furthermore, ?other factors, such as sources of information and preferred activities, were also found to vary according to nationality? (Kim and Prideaux, 2005, p. 355). Crompton and McKay (1996) sought to identify tourism motivations for attending festivals by identifying consumer seeking factors such as ?cultural enrichment, education, novelty, and socialization? (Crompton and McKay, 1996, p. 429). In this particular study, there were seven domains researched by using a 31-item instrument in which participants were asked to rate a scale ?from 1 to 3 as being clearly representative of the motive, somewhat representative, and not representative of any motive (Lee and Crompton 1992)? (Crompton and McKay, 1996, p. 430). In this particular study, ?dimensionality and stability of the scale were evaluated by factor analysis with oblique rotation? (Crompton and McKay, 1996, p. 432). The results point to six 27 factors with Eigenvalues greater than one and Cronbach alphas which ranged from .65 to .88 (Crompton and McKay, 1996). To add further support to the authors? findings, ?mean scores of each item within each motivational domain were summated and ANOVA tests were undertaken to test for a significant difference in motives among the five event types? (Crompton and McKay, 1996, p. 434). Resultantly, the ANOVAs found significant differences among the motivational domains, which meant that the seeking motivations (cultural exploration in the companionship of recognizable faces) were tied to attending festivals rather than being tied to the tangible tourism offerings. In relation to tipping, seeking motivations are clearly seen as consumers being led to tip in order to save face or avoid being embarrassed from the act of not tipping, or stiffing (Azar, 2008). In yet another study, Klenosky?s research sought to measure tourist motivations by constructing a scale utilizing the concept of means-end theory (Klenosky, 2002, p. 385). ?Means-end theory provides a practical framework for examining the relationships between the pull attributes of destination (i. e., the means) and the higher level motivational forces important to the individual traveler? (Klenosky, 2002, p. 385). The importance of this author?s study was the notion that push/pull factors were not separate influencing dimensions but rather were interdependent constructs. For example, destinations were not only chosen for tangible offerings such as a beach, but were also chosen in order for the tourist to get a tan and feel good about himself. ?This differentiation between the tangible and functional consequences of product use and the more abstract psychological/social consequences important to the consumer is consistent with a prior means-end model advanced by Olson and Reynolds (1983)? (Klenosky, 2002, p. 392). As a result, using the means-end approach to study push/pull relationships provided another context in which consumer motivations were measured. 28 Tipping motivations have also been researched from an economic perspective. For example, Azar (2008) sought to measure tipping behavior from the influence of social/psychological motivations (push) and the investment in strategic, future service (pull) encounters. This theoretical framework provides a two-dimensional model by which tipping motivations can be measured. On the one hand, there seems to be an internal intrinsic need to tip, while on the other hand there is an extrinsic push to avoid psychological discontentment. Moreover, when these studies are compared with studies on tipping, the different authors? measurement instruments are very similar, while some things are slightly different. For example, in all of the scales used, the push/pull factors such as personal seeking/escape and interpersonal seeking/escape are defined in similar ways. However, authors Kim and Prideaux (2005) sought to define consumer motivations in a nationalistic sense, which led to a more in-depth study and, ultimately, stronger in-depth findings. Furthermore, tourism and tipping affects all cultures and nationalities. Even though these authors measured push and pull items with different approaches and somewhat different scales, there appeared to be little difference in the methodology. No matter how many dimensions seemed to make up the push/pull relationship, it appeared that only two constructs were defined. These previously mentioned studies in tourism serve as a basis and methodology for researching tipping motivations in the upcoming sections. Similarities among several of the studies which were previously mentioned are noted in tabular form (Table 1). 29 Tbl. 1: Examples of Theoretical Framework in Tourism and Tip Studies In terms of this research, push and pull theory seems applicable. In retrospect, Azar (2008), sought to measure tipping behavior as it related to strategic behavior to ensure future Researchers Theoretical Framework Objectives Statistical Analyses Crompton and McKay (1997) Escape-Seeking Dichotomy, Push/Pull Theory Consumer seeking: Cultural enrichment, education, novelty, and socialization Factor Analysis, ANOVA, Duncan?s Test Iso-Ahola (1982) Approach (seeking), Escape (avoidance) Seeking Intrinsic Rewards, Escaping the everyday environment Modeling Kim, Goh, and Yuan (2010) Push/Pull Theory Food tourism motivation: Knowledge & Learning, Quality of Event, and Food Variety Factor Analysis Kim and Prideaux (2005) Push/Pull Theory Cultural (American, Australian, Japanese, Chinese (Mainland), Chinese (Hong Kong SAR), preferred tourist resources (activities), length of pre-departure planning, information sources used, and length of stay. Factor Analysis, One-way ANOVA (between nationalities) Klenosky, (2002) Push/Pull Theory Means-End theory Scenic/Natural Resources, Get Sun/Tan, Look Good/Healthy {ACV: Attribute, Consequence, value} Ladder Map program Snepenger, King, Marshal and Muzaffer (2006) Push/Pull Theory Personal Seeking, Personal Escape, Interpersonal Seeking, Interpersonal Escape Factor Analysis, Competing Models, Structured Equation Modeling Azar (2008) Two-Dimensional Model Strategic Behavior, Social/Psychological Motivations Regression Analysis 30 service and social psychological motivations. Tipping to ensure future service meant that consumers had to provide tangible incentives (monetary compensation) for servers to provide service in the future. This tangible incentive should have increased as patron frequency increased; however, that was not the case. ?As part of the survey, customers were asked to rate the server on a 1?5 scale on appearance, knowledge of menu, friendliness, speed of service and attentiveness? (Azar, 2008, p. 2). When compared with previously-mentioned consumer behavioral studies in tourism, this study seems to show that tipping can be measured by two dimensions, a psychological push (adhering to a social norm), and a tangible pull (overall service encounter). Motivations for Tipping Academic research on the phenomena of tipping dates back as far as the 1940?s, when ?it led Leo Crespi, assistant professor of psychology at Princeton University, to undertake the first in-depth scholarly probe of tipping in America? (Segrave, 1998, p. 74). Having cited an editorial writing on the subject of tipping, Life magazine ?concluded that tipping was a national nuisance and as such should be eliminated? (Segrave, 1998, p. 74). Crespi?s study focused on why the custom should or should not be eliminated. For instance, one question on the survey read, ?Do you approve or disapprove of the practice of tipping by and large?? (Segrave, 1998, p. 74). Another question read, ?If service workers were given fair wages for their work, do you think that tipping should be eliminated?? (Segrave, 1998, p. 74). In the 1970?s Freeman, Walker, Boarder, and Latane (1975) explored the social psychological relationship between restaurant tipping and group size (Azar, 2006). From their 31 study, the authors found that tipping was affected by the number of guests dining with the group. The authors addressed the theory diffusion of responsibility and its influence on actual tipping behavior. This is best described as a phenomenon in which tip size decreased as the dining party increased. Furthermore, diffusion of responsibility is a theory that can best be described as the social degradation of the degree of responsibility that is experienced among participants of a large group. Even then, the act of tipping seemed to defy economic logic. Another study, conducted by Snyder (1976), found that tip size from groups may depend on consumers? perceptions that less work is required when a server waits on a group at one table as opposed to waiting on many people at different tables. In the 1980?s, authors Crusco and Wetzel (1984) studied the effects of interpersonal touch on restaurant patrons. While they were paying the bill, the server touched some guests on the shoulder or the palm, before returning the change (Crusco and Wetzel, 1984). The actual touch was found to be a contributing factor which influenced tip percentage. However, the study did not account for the gender of the guests being served (Crusco and Wetzel, 1984). Research on the effect of touch by authors Stephen and Zweigenhaft continued in 1986. In this particular study, touching females was the primary focus. During each of three treatment conditions, direct eye contact with guests was avoided, just as in Crusco?s and Wetzel?s study (Stephen & Zweigenhaft, 1986). The authors ?hypothesized that it would be more profitable for a waitress to touch the female than the male? (Stephen and Zweigenhaft, 1986, p. 142), and it was concluded that the main effect of the female touch condition was significant and contributed to a higher tip percentage. From the 1990?s to the present day, tipping has become a more thoroughly researched topic for very good reason. The sheer economic impact of the custom and the dwindling manufacturing sector of the United States has seemed to contribute to the prolific 32 works of many (Azar, 2003, 2004, 2005; Lynn, 1997; Lynn & Grassman, 1990; Lynn, Zinkhan, & Harris, 1993). In today?s restaurant sector, the act of tipping is a common occurrence. While a majority of the population does not know the exact amount to tip, many accept that it is customary and expected in certain services? venues. Over a half century ago, trade-publication magazines such as Time, Travel, and Holiday all recommended that restaurant servers should be tipped fifteen to twenty percent (Segrave, 1998), and that recommendation still holds true today. Many consumers understand that the custom of tipping provides a unique way of rewarding or punishing service providers for their efforts or lack thereof. In simplest terms, giving poor service generally results in poor tips and ultimately, no income. However, this is not necessarily true in some instances. Many servers are tipped out of pity or compassion (Azar, 2004). Empathy is a powerful motivating feeling, especially when others understand the disenchantment a server may exude when he or she is having a bad day. Respectfully, many consumers, and in particular those who work in the tipped professions, tip above and beyond the requirement due to the fact that ?they know what it is like to live off of tips so they tend to tip accordingly? (Babcock, 2007, para. 44). Babcock (2007) further explains, ?from my own experience I know that there is also a widespread belief among members of the service industry that you should take care of your own? (para. 44). Even in instances of sometimes unbearable frustrations with poor service, many consumers will tip, no matter the service experience. This is consistent with Crespi?s (1947) study, which concluded that the public tipped more out of social disapproval than reward for service (Segrave, 1998). However, Lynn posited that today, a consumer?s willingness to tip, even in the event of poor service, is largely dependent on a country?s culture (Lynn, 2000). 33 In the U. S., consumers tip a variety of individuals who perform services. Some examples of tipped professions in the United States include bartenders, busboys, casino croupiers, chamber maids, cocktail servers, and concierges. On the other hand, while many professions in the U.S. are dependent on tips, other countries, such as those in Europe, have limited tipped professions. Therefore, social customs have been studied in order to further understand the influence of different culture?s unique ways of life. For example, ?tipping may be more widespread in countries with extroverted populations than in countries with introverted populations? (Lynn, 2000, p. 396). Furthermore, tipping is one of a small number of economic customs where the barter is mandated by unofficial rules of tradition rather than openly-stated transactions (McCarty, Shrum, Conrad-Katz, Kanne, 2000). According to Lynn (2000), countries that highly regard power and status feel threatened by ambiguous situations. In addition, these countries stress a group mentality rather than individuality. Lynn?s (2000) study measured national levels of extroversion, neuroticism, and psychoticism by using the Eysenck Personality Questionnaire (EPQ). The study found significance for all of the categories that were measured. Lynn?s research concluded that national personality traits affected tipping behavior. Also, national extroversion and national desire for recognition were correlated. In addition, it was also suggested that a desire for social attention may ?affect the prevalence of tipping? (Lynn, 2000, p. 399). Tipping can calm anxieties about being served by strangers, thus leveling the power in the exchange relationship (tit for tat, quid pro quo, service for money). The study also showed that the act of tipping was positively related to psychoticism, which was correlated with masculinity driven by assertiveness, achievement, and materialism. The United States fell under the category of psychoticism which infers that the culture is driven by assertiveness and materialism. 34 The most common sense argument as to why individuals are motivated to tip is service. Tipping serves as a mechanism by which consumers can monitor the quality of service from those who provide it. Furthermore, some will argue ?the main justification for tipping is that it promotes better service, by giving the workers an incentive to do their best to satisfy the customer?s needs? (Azar, 2004, p. 761). While the debate continues to unfold on the exact nature of the relationship, it seems only logical that quality service is an intrinsic motivator of tipping behavior (Lynn, 2000). Indeed, Bodvarsson, Lukestich, and McDormott (2001) found that ?service quality significantly affects tip size and when servers expect higher tips, customers rank service quality higher? (p. 1659). Service cannot be replicated and varies from person to person. ?A price can be fixed on a hotel room, on a meal, or on a distance traveled by taxi or bus, but not on the smiles, the friendly gestures, the hospitable attitudes, etc.? (Shamir, 1984, p. 62). Consumers tip as part of an economic transaction by assessing the provided quality of service; therefore, there are many different interpretations of quality service. With multiple variations in service and the elements that complicate service delivery, two factors should be considered which may affect tipping. These dimensions are the technical and functional aspects of service delivery. The technical dimension of service refers to the more tangible and objectively measurable elements of the product or service being supplied. For example, technical measurement of services provided by a restaurant server may include the number of visits to the table (order taking, food delivery, and menu knowledge). ?Personal service, on the other hand, refers to what might be better described as the more intangible, functional, subjective, and/or relational elements of the service encounter? (O?Neill, 2000, p. 164). Interestingly, research suggests that when a company has failed in the technical aspects of service, that the functional, emotive aspects may trump the breakdown in systems (O?Neill, 35 2000). Therefore, operators need to create an environment with systems that empower staffs with tactical, operational knowledge, which ultimately gives the power to recover and make guests satisfied when they voice discontent. Service, of course, is an intangible good. It is simultaneously produced and consumed in real-time. It is characterized as perishable and labor intensive. Given this, employees must be trained in pre-established service systems and procedures in order to deliver a specific product. A pre-established service delivery system generally refers to the technical dimension of service. Guests have grown to expect a baseline of technical aspects during their service encounters. For example, the table is greeted, beverages are ordered and delivered; food is ordered and delivered; remnants, along with dishes, are taken away, and a tip is usually provided. According to Barkan and Israeli (2004), this baseline of expectations is best represented as a tit for tat relationship. Technical aspects of service delivery are generally standard. On the other hand, the emotive factor of service, which is just as critical to tipping as the technical aspects, refers to the less definable dimensions. For example, a light touch, a warm and friendly smile, or direct eye contact represents characteristics of the functional dimension of service. Support for a functional process, specifically touch in this example, was supported in Jewell?s (2008) study that included ninety-seven restaurant patrons. Additionally, age and race were significant factors that influenced tipping behavior (Jewell, 2008). Younger diners were found to tip more than middle-aged or older adults. According to the findings, there were three main reasons given for why individuals tip: service quality, to help others make a living, and the feeling of expectation (Jewell, 2008). When considering the effect of gender on tipping, men were found to tip for quality of service, whereas women were found to tip for altruist reasons (Jewell, 2008). Touch was found to be significant to some restaurant patrons, but not all. 36 Furthermore, waiters who squatted at the tableside, rather than standing over their guests, received higher tips (Jewell, 2008). Other factors which were out of a server?s control, but also influenced tips, were the weather, alcohol consumption, lighting, and atmosphere (Jewell, 2008). Moreover, when quality of service was analyzed, consumers believed that friendliness and speed were the most important factors which motivated them to tip (Jewell, 2008). Creating a positive mood, among diverse demographics, was found to be a motivating role in tipping behavior. From the study?s finding, a myriad of factors was found to influence consumers? motivations to tip. Moreover, when considering the functional dimension of the service encounter, research supports the theory that the personal interaction (functional) increases tip percentage (Israeli and Barkan, 2004). Moreover, touch has been characterized as both the most basic sensory process and the most primitive form of communication (Major, 1981). According to Videbeck (2004) a server nonchalantly touching guests when returning change may increase the tip size. Although it may seem difficult to separate the service act from the actual actions of the server, such detailed communication mimicry and eye contact have been studied as motivating factors of tipping. In addition, ?other peculiar tip-inducing behaviour includes squatting at the table, drawing a smiley face on the bill, forecasting good weather, telling a joke, and wearing a flower in your hair? (Videbeck, 2004, p. 40). Parrett (2006) found that servers from the Netherlands who mimicked their customers improved their tips. Having a server who copied guests? interactions and mannerisms (such as repeating the guests? entire orders aloud to ensure correctness) served as the actions for the mimicking in the study. The study included sixty groups of customers, thirty in the mimicked group and thirty in the non-mimicked group. The female waitresses in the study echoed mannerisms of half of their customers. The act of 37 mimicking provided a way by which individuals could build similarities with others (Parrott, 2006). Whether body, facial expressions, or attitudes were mimicked, these actions were found to create an over-all influence with guests (Parrott, 2006). To support Parrott?s findings, in the second experimental group, servers who were unaware of the experiment registered the average tip size left by guests and wrote down the full order without telling the customer the order was understood. The findings revealed that repeating orders verbatim increased tips. Support for Parrot?s (2004) type of study was previously found in Speer?s (1997) research, which posited that imitation of body movements, gestures, or attitudes can influence tipping behavior. Subtle mimicry at the table when taking a guest?s order causes perceived attentiveness by the guest. Furthermore, factors such as touch or mimicry are indeed in the server?s control; however, tipping can also be influenced by factors which are completely out of a server?s control (i.e., the behavior of others). The role that others play during the service encounter is largely out of a server?s direct control. The way in which one guest behaves during an encounter can influence the way other consumers behave during that experience. For example, the extent to which a server is liked, as well as his or her physical attractiveness, can significantly impact consumer behavior (again, these are elements out of the server?s control). More often than not, however, no matter how attentive or friendly a server may be during an encounter, an experience can be thoroughly ruined by the actions of others. As a result, a server may lose a tip or gratuity to a situation which is completely out of his or her control. In order to increase the chances of a tip, a server may create influence and likability with guests. An article by Cialdini and Goldstein (2002) claims that there are six basic principles for creating influence. ?The six basic principles that govern how one person might influence 38 another are: liking, reciprocation, consistency, scarcity, social validation, and authority? (p. 41). Upon meeting someone, the obvious initial reaction is to look for similarities or comparisons. Likability of a person generally creates partiality to that particular individual. In the case of tipping, servers who earn praise or are commended by their customers are more likely to increase their tips. The study shows that reciprocity plays a very important role in a server/guest interaction. Consumers are left with feelings of indebtedness, especially when service has performed above the expected. ?The rule of reciprocity most often takes the form of gifts and favors, and that principle also is frequently used in business negotiations? (Cialdini and Goldstein, 2002, p. 45). Another principle which may influence a person?s tipping behavior is consistency. However, the consistency presents a two-fold consideration. Quality, consistent service experiences are critical in building and maintaining trust during service-exchange relationships. On the other hand, loyal consumers who consistently visit the same restaurant or venue may be concerned with maintaining a reputable relationship among the service personnel. Frequent patronage of the same restaurant allows guests to bond with service staff. Some guests may become concerned with the personnel?s judgments or beliefs about their tipping behavior; therefore, continual interaction with servers aids in creating influence and desire impressions. In addition, the perception that a product or service may be in short supply causes the feeling of scarcity. Scarcity may cause consumers anxiety during the purchasing process; consumers may act irrationally in order to purchase the last one, thus exhibiting a get it while it?s hot mentality. Furthermore, social validation is another principle that can create influence. Servers who complement guests? ordering decisions or positively reinforce guests? dining ideas can create influence. Servers who utilize authoritative or referent menu knowledge can influence consumers? purchasing decisions while participating in the dining experience. Conclusively, all 39 of the basic principles above can be used to manipulate consumer behavior. It is important to note that while servers use these principles, they should be governed by stern ethics and honesty. While some studies have proven a positive relationship between service quality and tips, server expectations of good tips tend to encourage good quality service (Bodvarsson et al, 2003). Tipping is voluntary and rejects typical economic theory (exchange of monetary funds for products or services), in which consumers pay more than is required for services. Researchers hypothesize that while many actually leave tips with the conciseness of service quality in mind, some are more concerned with ?maintaining fairness (equity) in transactions? (Bodvarsson et. al. 2003, p. 1660). Servers who are generally friendly and polite may gain a competitive edge by the use of ingratiation or gratification. It is interesting to note that servers tend to have a preconceived notion of how a guest will tip. Often, this notion may be based on what the guest is wearing, his or her gender, or mood. While consumers are not legally bound to tip (as long as the gratuity amount is not included in the price of the bill), servers may feel that the customers have had a free ride or take advantage of the dining experience and indirectly suffer negative feelings from being stiffed. According to a study by Bodvarsson (2003), the extent to which guests free ride is dependent upon the amount and frequency of interactions between them (the consumers) and the provider (the server). His study was designed in two parts. The server would give information on the customer, such as the amount of tip, coupons redeemed, and number of courses. Customers were asked to report on frequency of visits, quality of service, and quality of food. When gender of the server was discussed, males received significantly lower scores than females. The study also suggested that servers use discretion in the amount of services which will be provided by 40 assessing certain guests? characteristics, such as dress and demeanor, as well as over-all physical appearance. Given the emotions and expectations of guests during the service encounter, there is a degree of risk involved. This risk is especially heightened if the consumer is paying a considerable amount of money or greater than expected value for a product or service. Moreover, since a product can be delivered by different individuals, variability in the way the product is perceived and/or received presents a challenge. When dealing with service delivery systems, ?there is no better place to implement specific solutions, than in the hospitality industry, where customer service is inseparable from employee performance? (Bell and Winters, 1993, p. 93). Above all, tipping provides a mechanism by which consumers are given the ultimate responsibility to monitor the quality of service. According to Videbeck (2004), ?most customers tip a percentage of the bill size rather than a fixed amount (so-called flat tippers are estimated to account for only 20% of the population)? (p. 39), while the percent tippers help monitor the quality of service provided by servers. Thus, Videbeck?s (2004) study added to the debate that tipping may be used as a form of quality management. There is a unique type of scenario which can exist between the server and consumer. Certain ?economists refer to this as the principle-agent problem? (Videbeck, 2004, p. 38). While the restaurant owner serves as the principle (authority), the servers act as the agents (informants or workers on behalf of the owner). From a continuous quality improvement perspective, since owners cannot monitor their employees? every move, part of the responsibility is given to the guests. Guests can either reward or punish their servers by leaving or not leaving a tip based on the quality of perceived services. The principle-agent scenario drives down the amount of monitoring costs for the owner. Being that restaurant owners cannot be in all places at all times, 41 monitoring costs for a restaurateur would include extra costs for management or supervision in order to monitor the quality of work which servers provide on behalf of the restaurant. In a contrary sense, however, while there is evidence to suggest a strong, positive correlation between both services provided and the likelihood to tip, there is evidence that disputes this theory, or at least proposes that it is much weaker than first thought (Lynn and Grassman, 1990; Lynn and McCall, 2000). For example, Lynn (2001) found that some may have a flawed assumption that tip amounts do not accurately judge how well a server is performing. For him, ?restaurateurs rely on tips to motivate servers to deliver good service, measure service performance and identify dissatisfied customers? (Lynn, 2001, p. 15). In psychological terms, the server?s interaction with a guest is a type of interpersonal relationship which is based on trust and equity. Therefore, ?to keep those relationships equitable, customers should give bigger tips, when they get better service? (Lynn, 2001, p. 15). However, receipt of superior service does not necessarily mean that consumers will tip above their pre-established belief or tipping norm. Quality of service received is the strongest common-sense explanation of why consumers do or do not tip. On the other hand, tipping is an economic activity, so ?equity concerns affect tipping less than do purely social actions? (Lynn, 2001, p. 15). Lynn further posited that consumers ?feel strong social pressure to tip 15 to 20 percent of the bill size? (Lynn, 2001, p. 15). To test his assumption about the differences between dining experiences and tip size, a meta-analysis was used. A ?meta-analysis is a way of statistically combing and comparing the results of different studies? (Lynn, 2001, p. 15). From the findings, tip sizes increased with quality of service; ?however, the correlation between tips and evaluations of the service or dining experience had a mean of only .11? (Lynn, 2001, p. 18). In addition, this low level of correlation between service level and tip size ultimately did not support the notion that better service quality 42 meant better tips. Moreover, ?consumers will leave 5 percent (or less) and tips of 20 percent or more at any given level of service? (Lynn, 2001, p. 18). Additionally, a weak relationship was also found for service levels in the average (or median) tip ranges. More importantly for a manager?s awareness, if servers do not perceive that better service yields better tips, then other ways of motivating servers should be applied. Some of those job motivators included better scheduling, better sections, and other rewards (Lynn, 2001). Lynn?s study suggested that managers could use tip averages to determine the servers who receive the best tips based on sales. Lynn and McCall (2000) believed that consumers are motivated to tip because of socially imposed norms. Moreover, these norms leave consumers striving for a state of equity in economic relationships. ?The idea that people tip as a reward for good service is also consistent with equity theory? (Lynn and McCall, 2000, p. 204). But in many cases, economic relationships are not equitable. Economically speaking, consumers are motivated to pay minimal costs for product and/or services. ?In order to get the most from their limited resources, consumers usually try to obtain things for the lowest possible price? (Lynn and McCall, 2000, p. 204). While consumers would like both lower cost and quality services, lower prices are usually not associated with quality services. Tipping is a way for customers to ensure a company?s success by providing a retention mechanism (compensation) for servers. Lynn and McCall?s (2000) method used in this research consisted of published and unpublished studies. Interestingly, evaluation of with-in restaurant results proved more effective than between restaurants results. From the findings, tipping greatly varied in between restaurant concepts as opposed to with-in the units. The study further utilized bill size as a dependent measure. From the findings, service-effect on tipping was small, but definitely correlational. The authors proposed that this should be especially true of infrequent diners, who have no continuous 43 interaction with the service provider. Lynn and McCall?s (2000) hypothesis of a weak service to tip interaction was further supported by Azar?s (2006) study which found that ?empirical evidence suggests that tips are hardly affected by customer service? (para. 1). Israel and Barkan (2004) gave further consideration to why consumers tip as a percentage of the bill or a flat-rate dollar amount. ?This twofold consideration frequently leads to a situation in which tips for low bills exceed the percentage norm and tips for large bills fall short of the percentage norm? (Israeli and Barkan, 2004, p. 445). This is also known as the magnitude effect. Additionally, while monitoring service, the researchers posited that technical elements increase dollar tip amounts, whereas the functional elements increase dollar amounts and percentage amounts. Accordingly, Israeli and Barkan (2004) proposed that servers could employ three strategies: technical strategy, positive functional strategy, and negative functional strategy (Israeli and Barkan, 2004). Additionally, consumers should employ three strategies when tipping: fixed percentage, fixed dollar amount, and tit-for-tat. Resultantly, in the authors? study, there was an overlap between the technical and function dimensions which tied to the magnitude effect. When serving larger groups, the functional aspects became discounted, and the experience became technical for the guests. Furthermore, the authors found correlations between dollar tip and bill size followed by increased percent tip and increased dollar amount. In terms of dollar amount, the technical dimensions of service were crucial, but percent tipping was affected by the functional dimensions. When viewed from an economic perspective, many feel that tipping is a form of pressure. More applicable in today?s hospitality settings, however, some believe tipping is driven by compliance to social norms. Consumers are motivated to tip for ?social approval, equitable relationships and future service? (Grassman and Lynn, 1990, p. 169). Azar?s study, which 44 hypothesizes that social norms influence tipping behavior, explains that ?when a norm is costly to follow and people do not derive benefits from following it other than avoiding social disapproval, the norm erodes over time? (Azar, 2004, p. 749). Unsurprisingly, the phenomenon of tipping has defied this logic. The tipping norm in the United States is widely accepted to be around fifteen percent. ?However, etiquette books report that the figure is moving toward 20 percent for excellent service in upscale restaurants, and that in buffet or smorgasbord restaurants it is only 10 percent? (Azar, 2004, p. 752). Moreover, ?many have suggested informally that tipping and the size of the tip as a percentage of the bill size is a social rule simply to be followed without much thought? (Bodvarsson et al, 1999, p. 139). In the case of tipping, individuals will conform to social norms by sacrificing money even though it may be costly to do so. While there is no definitive answer, many are thought to tip in order to bolster self esteem and prevent shameful feelings. Where tipping is required for services, consumers generally want to feel giving and fair. Some even believe that tipping generously is a way to show off in front of others. Videbeck (2004) posits ?the most widely supported theory states that people tip simply to avoid the considerable stigma that accompanies 'stiffing' (not leaving a tip)?a kind of selfish economic agent with feelings? (p. 39). It is no surprise that those who depend on tips must make tips everyday on every shift. This is due simply to the fact that wages are very low. In most states, the supplemented remuneration is $2.13 per hour. Moreover, many server positions are afforded no further compensation. In order to be successful, servers must be on-point during every interaction with a guest. 45 Not only does compliance with social demands affect the consumer; it may also affect the service provider. The social mandate of tipping may negatively affect the effort exerted by servers. Many servers expect that they will be tipped no matter the case; therefore, poor service is often a result. Interestingly, some guests may feel embarrassed to leave a blank in the tip column on a credit card slip. It is this feeling of embarrassment that drives many consumers to tip out of guilt and pressure to socially conform. In some cases, stiffed workers may feel vengeful toward guests who do not tip. Of course the behavior of these type workers is not rational, because guests are truly required to pay for the cost of the product and nothing more (provided a service charge is not included). Interestingly, yet strange, society?s need to conform to social norms tends to erode over time. For example, a societal norm which has eroded over time is that of unmarried couples choosing to live together. The nostalgia for the old days suggests that unmarried couples should not cohabitate; however, that norm has clearly eroded over time as couples tend to live together first before getting married. Evidently as described by this example, time progresses, and the population stops conforming to preconceived societal sanctions. Furthermore, Azar?s study states that ?when a norm is costly to follow and people do not derive benefits from following it other than avoiding social disapproval, the norm erodes over time? (Azar, 2004, p. 49). Tipping, on the other hand, seems to defy this logic, as previously mentioned research posits that percent tipping amounts have increased from fifteen percent to twenty percent over the past several decades (Segrave, 1998; Azar, 2004). Therefore, it is the belief of this researcher that consumers will tip in order to conform to social norms. Consumer behavior is driven by underlying motives, wants, and needs. Action, of course, is the end result of motives reaching a climactic point. Motives may be rational or irrational, 46 intrinsic or extrinsic. When considering the act as an emotion, tipping can be explained by a number of abstract dimensions. Tipping involves personal, interdependent interaction between the provider and consumer. While the interaction is merely superficial, there is still much to be learned. As has been shown, the principle-agent theory can apply to the consumer/server relationship. In this instance, the consumer acts as principle (requiring service) and the server (delivers service) is the agent. Principle-agent theory explains the determination that consumers play a pivotal, active role in monitoring service quality. In addition, consumers? desire to feel gracious or altruistic is certainly a potential motivating principle. While some of these motives are obvious, some are much more subtle. When the custom began, tipping of a worker was believed to be out of compassion or gratitude (Azar, 2004). Presently, there is a disguised role of reciprocity found in the server/guest relationship. In addition, guilt (let down) is also driving part of the relationship. Let down refers to guilty feelings that a guest might experience from having let down a service worker when the exchange relationship remains imbalanced. For instance, service which was received, even though it may have been menial or unsatisfactory, still remains imbalanced. Some sort of service was received, but nothing was given in return. A consumer may experience guilt when the playing field is not leveled. The concept of tipping provides the ?need to reconcile the economic, social, and psychological elements of server-client interaction? (Shamir, 1984, p. 59). Evidence suggests that ?powerful non-pecuniary motives like the desire to reciprocate or the desire to avoid social disapproval also shape human behavior? (Fehr and Falk, 2002, p. 688). Tipping can influence one?s emotions, simply by providing an uplifting, generous feeling. The byproduct or effect consumers derive from tipping is termed psychological utility. However, receiving a warm, fuzzy feeling is not without costs; consumers pay for those feelings in terms of monetary 47 resources. This may lead individuals who are more patient and empathetic to have a tendency to give more generously or leave higher tips. However, it is believed that due to the feelings of disutility, many consumers tip even if they do not feel the desire. To be equitable within the service provider/ guest relationship, the customer uses a gratuity to make the transaction more equitable or balanced. In some countries, such as Germany and Britain, tipping can be offensive, and while there are no definitive purposes behind the motivation to tip, many seem to tip in order to bolster self-esteem and prevent the shameful feelings from not tipping. This was stated in Videbeck?s (2004) study. ?Consumers may tip in order to experience the positive feelings that come from showing compassion for the low income worker? (p. 39). A heavily favored theory that could reconcile the two [future service concerns] is the repeat customer scenario, where a tip is remunerated to help ensure good service or to avoid retaliation such as food tampering in the future (Videbeck, 2004). The hypothesis that consumers will tip because of concerns over future service encounters was supported in a study conducted by Grassman and Lynn (2000), where a student gave exit interviews to diners at Red Lobster restaurant. There was a customer index for satisfaction with food. In addition, there was a customer satisfaction index for service. Also, frequency of visit and the amount of the bill were questioned. The surveyor approached all paying guests except when busy with others. However, being put on the spot, customers may have inflated tip amounts, since tipping is generally a personal affair. Additionally, prior research may suggest that regular clientele will tip consistently or more generously, because their tipping practice may become a topic of conversation among other staff members (Grassman and Lynn, 2000). 48 Proposed Theoretical Framework Tipping behavior is an important activity in the life of all restaurant goers; therefore, its motivational influences should be fully understood by practitioners and academicians. The problem practitioners face is the mere fact that the act of tipping is multi-dimensional and a highly complex behavior. While some of the factors are completely out of the control of the server, other factors are understandable and calculable. It is these controllable aspects of tipping behavior which prove worthy of academic debate. In fact, the decision to tip is seldom made as the result of any one particular motive. Therefore, it would seem natural to argue that where the repeat purchase of service is sought, tipping motivation should be understood. It is necessary that service personnel have a detailed understanding of the mental and emotional underpinnings of the role of tipping motivations and clear direction on how to capitalize on those motivations. It is essential to understand how tipping motives are formed so they can be manipulated at the first moment of truth or occurrences over time. Resultantly, this should enable servers to better arm themselves with the knowledge that will ultimately increase their monetary livelihood. Research Considerations Having reviewed the literature on tipping, the obvious questions for the researcher or the reader appear to be: What are the research implications for measurement? Is it possible to gain a better understanding of the forces driving tipping behavior? It is clear from the review of the tipping literature that the need exists for the development and testing of a holistic quantitative 49 scale aimed at identifying the comprehensive myriad of tipping behavior. Tipping is a consumer behavior and is, therefore, measurable in the same way as other behaviors. In retrospect, many researchers hypothesize that tipping motivation can be explained by the service act, societal norms, and the need to secure service in the future. ? While the role of motivation and behavior are well recognized in the fields of sociology, psychology, and economics, little research has been undertaken to explain the motivational attributes of service, social norms, and future service consideration, simultaneously quantifiable, in order to find out which explains the ultimate reason why individuals tip. It is clear, however, that the different tipping motivations have a very important mediating role upon tipping behavior. To what extent any relationship might exist should be fully explored. ? Additionally, it is apparent that operations managers have much to learn from applying tipping motivational theories developed within the field of economics and hospitality to a situation by which a guest may participate in tipping behavior. This is true in all factors which are related to the behavior: service, social norms, and future service. The literature clearly suggests restaurant guests are inundated with a variety of motivational factors each visit. As a matter of fact, there are so many factors at play that all potential motivators should be explored in tandem. ? Managers must understand that tipping is not just an objective act but is clearly also subjective in nature, because it involves controllable and uncontrollable actions on the part of the service provider (service, manipulation of societal norms, future behavior considerations, peer-pressure concerns). Managers must be conscious of the fact that service personnel represent the company?s brand and, ultimately, the company?s 50 marketing message. Empowering their servers with training and development of the multi-dimensional nature of tipping can shape how consumers perceive and receive a company?s efforts. Towards a Theoretical Framework In summary, several theories stand out as potentially influencing the consumer?s motivation to tip. Social norm theory suggests that individuals will behave or comply in order to satisfy societal standards. Tipping can be described as a social norm driven practice. ?Social disapproval is a key element in the enforcement of social norms? (Fehr and Falk, 2002, p. 698). In addition, equity theory can also be applied to the tipping custom. Equity theory values the balance in exchange relationships. The outputs of a service provider are expected to match the remuneration by the patron. Evidence suggests that ?powerful non pecuniary motives like the desire to reciprocate or the desire to avoid social disapproval also shape human behavior? (Fehr and Falk, 2002, p. 688). More commonly thought to affect consumer tipping behavior is the quality of service. Usually, poor service results in poor tips; good service generally results in good tips. With these theories and/or substructures in mind, this facilitated the delineation of the construct and development of a set of 19 questions (items) to measure consumer motivations to tip. Against this background, several research hypotheses are offered. The act of tipping is a complicated motivational phenomenon best defined by a myriad of motivational influences, including the service act, the service exchange, social pressure to conform and the need to secure equitable future service. The central research hypothesis can thus be posited as follows: 51 H1: The motivational tipping construct will present itself as multi-dimensional in nature, influenced by a range of motivational themes including service, social compliance, server actions, future behavior, peer pressure and other operational processes. These motivational factors are presented in the theoretical model below (Figure 4). Below is a graphic representation of this research?s theoretical model. Tipping is multi- dimensional and very complex in nature. It is hypothesized that tipping is influenced by a myriad of motivational factors. Based upon review of the literature, those motivational factors include: service, social compliance, server actions, future service considerations, peer pressure, and operational processes. It is the intention of the researcher to show that all tipping motivations mentioned here actually play a role in motivating consumers to tip. Fig. 4: A Proposed Model of Tipping Motivations Service Social Compliance Server Actions Future Consideration Tipping Motivations Tipping Behavior Peer Pressure Operational Processes 52 Summary This chapter has reviewed the relevant literature regarding possible tipping motivations. In addition, the literature regarding the aspects of motivation and behavior has been examined in order to provide strength to the scale which was developed for the study. This review has indicated strong links between such theories as diffusion of responsibility, equity, and fairness. The links between motivation and behavior were particularly strong in respect to tipping behavior. The following chapter will provide a wide-ranging explanation of the methods used in developing the research plan, data collection and analysis. 53 Chapter III Methods Overview To put this study in the most purposeful context, a definite understanding of consumer tipping motivations is paramount in understanding consumer tipping behavior. The study of general tipping motivations, in particular, is to gain a basic consideration of the underlying importance of tipping motivations for restaurant service personnel, relative to the concept of tipping behavior, thus contributing to the level of understanding of such motives. In the framework of this research, qualitative and quantitative measures have been utilized, and empirical findings are offered to add strength and clarity to the understanding of tipping motivations and the development of a tipping motivation scale (TMS). Method of Inquiry According to Bogdan and Biklen, ?methodology is a more generic term that refers to the general logic and theoretical perspective for a research project? (Bogdan and Biklen, 2007, p. 31). Kvale and Brinkmann define methodology as ?the study of methods of a particular field? (Kvale and Brinkmann, 2009, p. 325). In simplest terms, research methodology begins with 54 general questions which are to be investigated. As a study proceeds, this process either tends to narrow the boundary of specific inquiries or creates new avenues of inquiry. This narrowing of inquiry can be visually represented by an inverted cone. Qualitative theory takes the shape of a funnel, starting with an initial, intimate inquiry. ?Theory developed this way emerges from the bottom up (rather than from the top down), from many disparate pieces of collected evidence that are interconnected? (Bogdan and Biklen, 2007, p. 6). For qualitative research, the process by which data is gathered is inductive. Qualitative researchers, ?do not search out data or evidence to prove or disprove a hypothesis they hold before entering the study; rather, the abstractions are built as the particulars that have been gathered are grouped together? (Bogdan and Biklen, 2007, p. 6). In contrast, quantitative research is deductive, with an objective to support or disprove predetermined hypotheses. Miles and Huberman (1994), when speaking about quantitative studies, state that ?like their qualitative brethren, they must be preoccupied with data reduction (computing means, standard deviations, indexes), with display (correlation tables, regression printouts), and with conclusion drawing/verification (significance levels, experimental control differences)? (p. 12). Quantitative research tends to focus on larger numbers. At the conclusion of a quantitative study, generally, information and findings are deducted from the sample population. Quantitative researchers utilize dependent and independent variables ?that either put people into distinct families built around what they say or do (Q analysis) or, alternately, cluster such actions and perceptions across informants (R analysis)? (Miles and Huberman, 1994, p. 69). Given that quantitative and qualitative studies employ different means of collecting and analyzing data, both require a theoretical basis. 55 The application of theory is the corner stone of academic research. Theory can help the researcher contextualize or serve as a frame of reference while studying and attempting to understanding human behavior; therefore, a theory-based perspective is necessary to understand tipping behavior. At its core, research utilizes some sort of theory or assumption from the beginning. ?Theory helps data cohere and enables research to go beyond an aimless, unsystematic piling up of accounts? (Bogdan and Biklen, 2007, p. 24). As Pandit reports, ?theories can?t be built with actual incidents or activities as observed or reported; that is, from raw data. The incidents, events, happenings are taken as, or analyzed as, potential indicators of phenomena, which are thereby given conceptual labels? (Pandit, 1996, p. 1). Theoretical conceptualization provides the framework for real world application. The custom of tipping represents a behavior as well as a social phenomenon. ?In other words, social phenomena, such as language, decisions, conflicts and hierarchies, exist objectively in the world and exert strong influences over human activities? (Miles and Huberman, 1994, p. 4). Therefore, the goal of phenomenological research is to gain further in-depth knowledge of the phenomena, ?although phenomenology has been called a method without techniques? (Miles and Huberman, 1994, p. 2). Bogdan and Biklen (2007) claim that ?researchers in the phenomenological mode attempt to understand the meaning of events and interactions [as applied to] ordinary people in particular situations? (p. 25). On the other hand, merely studying phenomenology seems ill-directed without some form of structure related to humanistic interaction. To aid in this structure, researchers employ the application of theories or paradigms. Bogdan and Biklen (2007) also posit that ?among quantitative researchers in education, [these methods are] sometimes restricted into a systematically stated and testable set of propositions about the empirical world? (p. 24). In contrast to the predetermined parameters of most 56 quantitative research, qualitative inquiry focuses on openness. This openness allows for researchers to study the object(s) in a naturalistic setting. Qualitative researchers believe that traditional methods of counting subjects cannot truly describe individuals in their natural settings (Bogdan and Biklen, 2007). Some supporters of quantitative research hypothesize that qualitative research is fragile and highly subjective in nature (Gherardi and Turner, 1987). These opposing viewpoints summarize one of the stark differences which researchers feel about quantitative/qualitative methodologies. It is important to note, however, that the major portion of this research is quantitative, but is preceded by qualitative inquiry; however, this study is primarily quantitative. Essentially, it is best to define the sequential methodological approaches used in this study. The definitions are taken from Robert C. Bogdan and Sari K. Biklen?s (2007) book, Qualitative Research for Education: An Introduction to Theories and Methods. Qualitative research??a research approach emphasizing descriptive data, grounded theory, participant observation, and the study of people?s understandings of all aspects of education?an approach we refer to as qualitative? (p. xiii). Quantitative research??If you want to find out what the American people think about a particular issue, survey research that relies heavily on quantitative design in picking your sample; designing and pretesting your instrument, and analyzing the data is best? (p. 43). The authors further explain that ?the emphasis on quantitative methods of research as the only evidence-based method has discouraged federal funding for this kind of work? (p. xiii). A question has been raised over the years as to the effectiveness of the two methodological approaches--qualitative versus quantitative and/or mix methods. The over-arching goal of this study is to provide further solidity and convergence of the data gathered 57 from both methods. The diversity of methods serves to make the most of the strength of each method and reduce their natural weaknesses. Through the use of statistical analyses, quantitative examination can capture the general behavior of a group. For instance, political campaigns and presidential elections can be affirmed, biologists can ascertain the number of bacteria within a colony and the likelihood that bacteria can cause disease, consumer expenditures and trends can become calculable along with the patterns of their spending habits. Even the likelihood of the way in which individuals behave in the future can be accurately predicted. It is important to note, however, that no one single statistical method can concretely define any one, specific behavior. As for qualitative research (interviews, focus groups, one-on-one interviews), it can provide what some researchers describe as deep, rich description, adding to the interpretive characteristic of explanation (Guba and Lincoln, 1989; Schwandt, 2007). Employing a qualitative approach to research can provide a rich understanding of human behavior through one-on-one interviews and interactive focus groups. Moreover, whereas quantitative research seeks to correct or normalize outliers, qualitative research can assist in the actual interpretation of a particular non-normal case. It is important, especially in social research, to remember that outliers are, in fact, human and should not be taken for granted. It can be argued that the most dynamic and important social changes have many times originated from those who stand outside the social norm, since social norms are believed to motivate behavior. Each methodology requires considerable effort on the part of the researcher to ensure the reliability of the process and the validity of the instrument to be used. For researchers who require concrete evidence by numbers, values, and calculation, qualitative research presents some skepticism, as quantitative researchers believe that qualitative research is highly subjective in nature; the researcher is the instrument. On the other hand, qualitative researchers are 58 concerned that human behaviors are theoretical abstracts that cannot be comprehensively understood by only numbers. A quantitative method seeks to investigate and ultimately report on that investigation in numerical fashion. Being that qualitative research lacks numerical values, trustworthiness and authority can be seen as one-sided. The utilization of both methods--validity and trustworthiness--is enhanced by triangulation, quantitatively through compound or confirmatory statistical analysis or qualitatively throughout the compounding of numerous sources (Schwandt, 2007). However, arguably, the beginning of all research is qualitative. A quantitative scale has to begin with a question and a clear direction that attempts to answer that inquiry. The question can be based on offered theory, or it can be unique. The steps taken to validate the research are the same: comprehensive review of the literature, interviews and/or focus groups to determine dimensions and variables, expert review, sorting techniques (Q sort, card sort), and pilot studies. All the early work is of a qualitative nature and is intrinsic to quantitative inquiry. This, however, limits qualitative research to development. In this view, each method reinforces and improves the other, regardless of the order the method is used. In this light, either method can serve as the confirmatory method of the other. Practitioners such as Newman and Shannon (2000) have posited that all research is a blend of the two methods that exist somewhere on the research spectrum. In this study, a sequence of qualitative methods, which then progress to a quantitative analysis, is used to most fully investigate the projected hypotheses. The study posits that using mixed methods approach (focus groups, one-on-one, interviews, item generations, survey development and data, participant comments, field surveillance) will provide the most comprehensive look at tipping behavior. Quite simply, this approach will attempt to determine 59 what the intrinsic and extrinsic, push or pull motives are which drive consumers? tipping behaviors. Specifically, this study takes the realistic view of academic research. Research Questions While the importance of tipping research has been highlighted in the literature and this study as stated earlier, quantitative examination has been found to be lacking. To fill this void, a new scale, the Tipping Motivations Scale (TMS), has been devised by the researcher to uncover and identify those dimensions most directly associated with consumer tipping behavior. The TMS scale was developed to answer the following primary research questions: What motivational factors drive the consumer?s decision to tip (RQ1)? Additionally, the TMS has been designed to address the following secondary research question: To what extent will actual service delivery affect the consumer?s decision to tip or not (RQ2)? To what extent will consumers continue to tip regardless of the quality of service received (RQ3)? Since the primary purpose of exploring tipping motivations is to provide practitioners (management and servers) with knowledge, skills, and abilities to ultimately attain a greater degree of satisfaction, self-fulfillment, and quality of life, the need for an instrument that measures the resultant behavior of the activity is obvious. The Tipping Motivations Scale (TMS) seeks to identify the various motivations driving consumer tipping behavior and the extent to which they explain this behavior. The TMS consists of multiple domains previously identified within the literature as follows: 60 ? Service?with the dimensions of technical (order taking, order delivering, tangible aspects) and functional (friendliness, personal service, eye contact, intangible aspects) elements. ? Social Compliance?which includes feelings of shame, guilt, regret, equitable relationships, and altruism. ? Server Actions?which includes the aspects of touch. ? Future Behavior?which includes the dimensions of gratitude, gratuity, and frequency of visit, along with fellowship, social respect and esteem. ? Peer pressure?which includes the dimensions of diffusion of responsibility for obligation to others. ? Operational processes?which include the dimensions of restaurant space (physical environment or offerings). Plan of Research The research plan included identification and solicitation of a sample group in order to collect the qualitative and quantitative data necessary to develop the research scale and answer the research questions proposed above. The starting point for the research began with a literature review on motivation, motivational scaling, and tipping behavior. The goal here was to identify potential drivers of actual consumer tipping behavior. As indicated, this review pointed to the existence of multiple factors that seemed to drive consumer tipping behavior. These factors included service, social compliance, server actions, future behavior, peer pressure, and operational processes. Subsequently a series of individual items was generated that was to form 61 the basis of the initial TMS. These items were then validated through a variety of qualitative and quantitative procedures. Qualitative Research Procedures The following sections will provide a detailed examination of the processes leading to the development of the central research instrument?the Tipping Motivation Scale. Focus Group Work The first focus group was conducted with a class of approximately twenty-five students. This class was chosen based on its level of education (mainly because they were juniors and seniors in an upper division course in hospitality education who were studying an abstract curriculum concerned with the implications of quality management). Being they were students, this class of participants was also chosen because their lives have been overwhelmingly involved in restaurant patronage and, ultimately, tipping behavior. Furthermore, the fact that a large percentage of the population had unique experiences of working in the field while obtaining a degree in hospitality served as an important factor, since many of these students were currently employed in the hospitality industry. The ratio of males to females in the class was considered. From Auburn University?s demographic composition, the population (60% female and 40% male) was considered a close split between male and female students. During the initial meeting with the first focus group, class attendees were asked to sort a variety of statements based on similar comparisons. Next, statements that were grouped together 62 were sorted into like constructs. This organizing of statements or potential variables is commonly referred to as the card sort method or card sorting technique, which tested for face validity. ?The card sort method is proven to be a viable tool to ascertain the individual subject?s perceptions regarding competencies? (Jahrami, Marnoch, and Gray, 2009, p. 176). The students were divided into groups of three, and each group was given a list of statements. They were then asked to group questions which they felt to represent the same category. Next, the focus group was asked to identify a name for categories, along with the questions in that specific grouping. For example, variables which represented service were grouped in the category of service. Questions that seemed to involve obligation, guilt, or regret were grouped into the category of social norm, and thus went the process of categorizing. This exercise led to the identification of thirty-two items which were deemed to fit the area of study. Following this process, twenty-eight of the original sixty items were dismissed as having very little or nothing at all to do with the study at hand. As a result, thirty-two items comprised the main body of the Tipping Motivations Scale. Another focus group was conducted with doctoral students as a part of a Qualitative Statistics course. A review of the work of the second focus group provided an unusual amount of information from the survey which seemed to correlate and/or group into the same categories the first focus group had found. This was done in order to make sure that the information gathered from the first focus group of students was represented and supported in the six initial factors developed by the research. Those factors were service, social compliance, server actions, future service concerns, peer pressure, and operational processes. While the card sort technique was not applied in this setting, the general framework and goals were the same in this focus group as with the class of students. The main goal of this group was to assess the survey after the changes 63 based on the recommendations from the class room focus group had occurred, thus re-confirming that any issues of importance pertaining to consumer tipping behavior were covered in the survey. This information, in tandem with previous research, created a thirty-two item questionnaire (Pilot-1) intended to measure the multiple factors identified by the two focus groups. Students and instructors were used as a medium of survey distribution and collection. The survey instrument and interview protocols were given close participant scrutiny. Since this research was conducted under the support of Auburn University, IRB (Internal Review Board), approval was sought and obtained. After the two focus groups? work was completed, industry experts were chosen to review the research instrument. The career professionals were given the thirty-two questions and note cards with the six previously identified factors on them and asked to place each question under the corresponding headings of service, social compliance, server actions, peer pressure, future service, and operational processes. This was a crucial step in confirming that participants not only understood the questions, but were able to identify which factor each question was addressing. Thus, clarity of the questions was addressed in the minds of respondents as well as an evaluation of the six factors. Lastly, the participants (both focus groups and industry experts) were asked to proof the remaining items and/or survey by checking for grammatical correctness, wording, design and ease of interpretation; any reservations concerning the scales, or any additional concerns were addressed. Changes based on these responses included the re-wording of some questions, redesigning the layout of part of the survey, and altering the size of the font used on the entire questionnaire. These changes were made in an effort to ensure as little confusion as possible on the part of the respondents in order to attain the best possible outcome. 64 In addition to the two focus groups, individual or one-on-one interviews were held with experts in the hospitality fields in order to further validate the initial research instrument. One-on-one Interviews Careered professionals within the hospitality industry were interviewed individually. In order to include all possible tipping motivations, one-on-one interview participants were selected based upon length of employment in the restaurant and hotel industry, gender, ethnicity, and over-arching operational experience. During the time that the one-on-one interviews were being conducted, each participant was tape recorded so that the responses of the subjects could later be checked for accuracy. On a positive note, there were no other topics discussed which impeded the research process. The researcher was aware of the nature of the relationship possibly affecting the interviews and conscientiously avoided straying from the current topic of tipping. It is important to stress that the one-on-one interviews were conducted with individuals who had career experience in the restaurant and hotel industries. The combined years of experience in the hospitality sector totaled more than three-quarters of a century. One interviewee was a former general manager of Darden Restaurants, a male Caucasian with a twenty-five year cumulative experience with company-owned Red Lobster and Longhorn Steakhouse brands. Another interview took place with a Caucasian female restaurateur who had worked as a cook, a server, and a bartender, and who now has been an owner/operator for ten years. Her company is currently incorporated and has three units. The third interview was with a female Caucasian career server of more than ten years experience, a recent graduate with a Hotel and Restaurant Management degree, who is currently a Public Health Inspector for the Alabama 65 Department of State Public Health. The final interviewee, an African-American female, was chosen based on the fact that she was a career server with more than ten years experience; she currently works as an assistant general manager of a corporate hotel chain. It was the researcher?s belief that gender and race had to be taken into account, especially before the instrument?s piloting, in order that appropriate selection would provide an inclusive representation of tipping motivations based on gender and culture differences. The Research Instrument As a result of the preceding qualitative work, the investigator was able to finalize an initial draft research instrument (Pilot-1) for the purposes of preliminary testing. The initial instrument comprised a 32-item Likert type scale anchored at 1 (Strongly Disagree) through to 5 (Strongly Agree). Additionally, the research instrument sought respondent demographics and an additional measure of respondent tipping norms. Pilot Work to Date In an attempt to better understand consumer tipping motivations and to test the initial Tipping Motivations Scale (Pilot-1), the pilot studies were conducted at a prominent land grant university in the south eastern United States. Pilot study one was conducted in the spring semester 2009. The instrument (Pilot-1) was distributed across campus over a one-week period during the spring semester 2009. One-thousand surveys were distributed and 831 responses were received and analyzed, representing a response rate of approximately 83%. While the 66 questionnaire predominantly sought to measure students? motivations for tipping, demographic data were also collected. Turning first to a description of the research sample, just over 63% of respondents were female, 83% Caucasian, and just over 13% were directly employed in the hospitality industry. Of those who worked in the hospitality industry, over 51% of respondents had worked for one year or less in their current hospitality position, and some 10% indicated that they themselves held a tipped job. Fifty-eight percent of respondents classified themselves as either freshmen (15.6%) or sophomore (42.4%) on their respective programs. When asked if they tipped, 98% of respondents said yes, with a mere 0.5% (4) respondents indicating flatly that they did not tip. Some 41.5 % of respondents reported their tipping norm as being between 11-15%, with 27.5% indicating their norm as 16-20% and 4.1% of respondents indicating their norm as being above 20%. This figure increased greatly to 34.8% of respondents who indicated that they tipped above 20% for excellent service While the overriding goal of the study was to shed light on those motivations driving the consumers? decision to tip, it also proved useful to test the psychometric performance of the scale employed. Results show that the instrument performed well with coefficient ?=0.79. This reliability score exceeds the usual recommendation of ?=0.70 for establishing internal consistency of a scale for exploratory research. Construct validity was also addressed in terms of both convergence and the research instrument?s ability to discriminate between the underlying dimensionality of the consumer tipping motivation construct. Convergence was investigated by calculating the mean score for each of the thirty scale items and correlating (Pearson?s product moment correlation) this with the mean score from an overall single item variable addressing the 67 correlation between tipping behavior and the service received. This variable was presented by asking: please tell us how strongly you (1) Disagree through to (5) Agree that ?My tipping behavior is directly related to the service received.? A correlation of 0.281 was found which, while low, was nonetheless significant at the 1% level (p<.001). The next stage of the analysis was to conduct an exploratory factor analysis of respondent motivations for tipping. This test was designed to give structure to the motivational domains explaining consumer tipping behavior. An exploratory factor analysis using the principal component extraction technique was performed on the various motivational variables. The analysis made use of the VARIMAX factor rotation procedure in SPSS version 17. A component matrix was initially generated to ensure that the analyzed variables had reasonable correlations (greater than or equal to 0.4) with other variables. Unrotated and rotated component matrices were inspected, and variables that did not correlate or correlated weakly with others were excluded (De Vaus, 1996). All but four variables (variables 3, 4, 8 and 12) correlated well, which led to their removal from the subsequent analysis. The result of the corresponding KMO of ?sampling adequacy? was 0.821, and Bartlett?s test for sphericity was 5819.690, which is considered a high Chi-Square, yet significant at the level of 1 percent (sig.=0.001). The results of these tests rendered the data factorable, and consequently, the factor analysis was generated. 68 Table 2 ? Factor Analysis (Pilot 1) ? Consumer Tipping Motivation Construct Variable Component Matrix Pilot 1 Service Social Compliance Server Actions Future Behavior Peer Pressure Operational Process V11 .758 V10 .711 V19 .662 V26 .589 V21 .583 V25 .518 V22 .815 V24 .815 V16 .678 V14 .627 V7 .734 V6 .672 V23 .560 V18 .520 V5 .501 V31 .821 V32 .807 V9 .772 V13 .730 V30 .550 V29 .724 V27 .578 V28 .566 Eigenvalue 4.032 2.610 1.795 1.499 1.469 1.069 % Variation 17.529 11.349 7.803 6.517 6.388 4.648 Coefficient ? .73 .73 .65 .68 .51 .52 69 This initial analysis identified factor loadings (item to total correlations) along nine dimensions, accounting for approximately 56% of the explained variance. Further analysis of the rotated component matrix identified one variable (Variable 15) as having cross-loaded on two separate factors. The offending item was removed and the analysis rerun. A total of three additional analyses were run due to subsequent cross-loading, leaving a total of 23 items, loading cleanly across six factors and explaining approximately 54% of the variance. The corresponding reliability co-efficient for this revised scale was ?=0.71 and the related KMO of ?sampling adequacy? was 0.775; Bartlett?s test for sphericity was 3956.839, a lower Chi-Square, again significant at the level of 1 percent (sig.=.001). Table 2 reveals moderate to good alpha co-efficients for each of the six extracted factors, ranging from ?=.51 for what has been termed ?Social Peer Pressure? through to ?=.73 for what has been termed ?Service Received? and ?Social Compliance.? Pilot (II) Research Findings The results from pilot study one present reasonable support for the (Pilot-1) research instrument; nonetheless, the decision was made to revisit the instrument with a view to potentially improving its performance. Two weaknesses were observed during the initial pilot. First and foremost was the exclusion of a number of important variables that were overlooked in the development of the initial instrument. Quite a number of participants from the first pilot identified a range of variables that, at least in their minds, were important when it came to actual tipping behavior. These were volunteered at the end of each survey administration and on the face of it, there seemed to be a lot of repetition with respect to the comments offered. These 70 included such items as: a server?s gender, the timeliness of the service received, food quality, expense, payment method, seating assignment, and surprisingly?the weather. As a consequence, the researcher immediately began looking for theoretical support from the literature for the inclusion of these variables in a revised Tipping Motivation Scale (Pilot-2). While theoretical support was found for the inclusion of a number of these variables (namely: gender, seating, food quality and timeliness), support could not be found for the others. The strength of feeling among the respondent population was such, though, that the investigator chose to include all seven items in a revised TMS scale (Pilot-2), and to conduct another pilot with this revised instrument. A second weakness that was observed from the first pilot was the clear lack of any worthy outcome and/or dependent variable(s). The investigator determined that it was not only necessary to identify those variables driving the decision to tip or not, but also their influence in predicting and explaining sense upon actual consumer tipping behavior; in other words, form(s) of tipping behavior consumers engage in when good service is received. A review of the extant literature pointed to six main behavioral norms in this instance, namely (1) the amount of the tip during the actual visit, (2) same server requests, (3) word-of-mouth recommendation, (4) increased check amount/expenditure, (5) increased tip amount on future visits, and (6) actual future visitation. The investigator sought expert commentary from a variety of academic and industry sources in order to validate the inclusion of these variables in any subsequent pilot. This included two university professors and three food and beverage specialists from the local hospitality industry. Upon final review the revised Pilot-2 instrument included a total of 30 variables (23 from the original study and 7 new items), again anchored on a Likert type scale ranging from 1 (Strongly Disagree) through to 5 (Strongly Agree). This revised survey also 71 contained the same demographic variables as TMS-1. Additionally, the six outcome variables referred to above were included on a similar 5-point Likert scale anchored at (1) Strongly Disagree) through to 5 (Strongly Agree). The instrument (TMS/Pilot-2) was distributed across campus over a one-month period during the fall semester of 2009. A total of 400 surveys were distributed and 245 responses were received and analyzed, representing a response rate of just over 61%. Unlike pilot study one, which sought out student responses only, the second pilot sought the opinion of students, faculty and staff, and was administered on a face-to-face interception basis across campus. This helps explain the lower response rate, as many individuals declined the opportunity to participate in the research following the initial approach. Turning first to a description of the research sample, just over 52% of respondents were female, 87% Caucasian, almost 25% classified themselves as students and just over 24% were directly employed in the hospitality industry. Of those who worked in the hospitality industry, almost 12% of respondents had worked for one year or less; 25% had worked in their current hospitality position for three years or more, and just over 14% indicated that they themselves held a tipped job. Turning to the psychometric performance of the scale employed, the results show that the revised instrument performed well with coefficient ?=0.85. Once again, this reliability score well exceeds the usual recommendation of ?=0.70 for establishing internal consistency of a scale for exploratory research. The reliability score for the 6-item outcome scale measure was similarly good at ? =0.81. Construct validity was also addressed in terms of both convergence and the research instrument?s ability to discriminate between the underlying dimensionality of the consumer tipping motivation construct. Convergence was investigated by calculating the 72 overall mean score for each of the 30 scale items (m=3.34) and correlating (Pearson?s product moment correlation) this with the overall mean score for the 6 outcome variables (m=3.31). A correlation of 0.396 was found and was significant at the 1% level (p<0.001). The next stage of the analysis was to conduct a further exploratory factor analysis of respondent motivations for tipping. This test was again designed to give structure to the motivational domains explaining consumer tipping behavior; in particular, it should determine whether the repeating items from the first pilot (Pilot-1) performed well in the second pilot study and whether the newly added items played any role in explaining the variance from this respondent group. An exploratory factor analysis using the principal component extraction technique was performed on the various motivational variables. The analysis again made use of the VARIMAX factor rotation procedure in SPSS version 17. The result of the corresponding KMO of ?sampling adequacy? was 0.787 and Bartlett?s test for sphericity was 2265.228, significant at the level of 1 percent (sig.=0.001). The results of these tests rendered the data factorable and, consequently, the factor analysis was generated. This analysis identified factor loadings (item to total correlations) along nine dimensions accounting for approximately 62% of the explained variance. Further analysis of the rotated component matrix identified five variables as having cross-loaded?strangely, these variables included four (v.8 ? Food Quality, v.18 ? Method of Payment, v. 26 ? The Weather and v. 29 ? Restaurant Cleanliness) out of the six additions to the instrument following the first pilot. The offending items were removed and the analysis rerun. A total of three additional analyses were run due to subsequent cross-loading, leaving a total of 19 items loading cleanly across six factors and explaining approximately 62% of the variance. The corresponding reliability co-efficient for this revised 19-items explaining consumer tipping motivations was ?=0.78 and 73 the related KMO of ?sampling adequacy? was 0.766; and Bartlett?s test for sphericity was 1299.710, a lower Chi-Square again significant at the level of 1 percent (sig.=0.001). Reliability co-efficients for each of the extracted factors (shown in Table 3) ranged from ? =0.50 (Other Operational Variables) to ? =0.80 (Service). Other extracted factors included ?Social Compliance? (? =0.71), ?Server Action? (? =0.68), ?Future Service? (? =0.73) and ?Logistics? (? =0.55). Of particular note here was the fact that 17 of the original 30 items (TMS-1) continued to play a role in explaining consumer tipping motivation as part of this second pilot study. Indeed the degree of variance explained increased considerably from 54% across six factors in pilot study one to almost 63% in pilot study two. This presents strong support for their inclusion in the finalized research instrument?the Tipping Motivation Scale. 74 Tbl. 3 ? Factor Analysis (Pilot 2) ? Consumer Tipping Motivation Construct Variable Component Matrix Pilot 2 Service Social Compliance Server Actions Future Behavior Operational Process Other V4 .812 V11 .781 V10 .774 V3 .730 V22 .771 V24 .741 V16 .709 V14 .597 V6 .758 V7 .756 V15 .591 V12 .556 V30 .833 V23 .813 V17 .796 V9 .634 V20 .622 V1 .827 V5 .687 Eigenvalue 4.127 2.665 1.592 1.311 1.103 1.043 % Variation 21.959 14.026 8.377 6.902 5.807 5.492 Coefficient ? .80 .71 .68 .73 .50 .50 75 Final Research Instrument ? TMS The final research instrument again comprised a mix of demographic and scaled variables. Respondents were again quizzed on their ethnicity, gender, age, employment status, years of employment, whether they held a tipped job, annual household income, and tipping norm. Additionally, the instrument contained a 6-item tipping behavior intention scale anchored at 1 (Strongly Disagree) through to 5 (Strongly Agree), and a 19-item Tipping Motivation Scale, again anchored at 1 (Strongly Disagree) through to 5 (Strongly Agree). The instrument also contained a section soliciting comments, thus offering participants the opportunity to express what is important to them regarding tipping. The comments obtained provided a rich source of qualitative information which served to shed light on the strength of connection with the behavior and the a priori assumptions, while illuminating the emergent themes. Sample The sample in this survey was composed of randomly selected individuals who visited Auburn University?s campus in the fall of 2010. Auburn University is a land-grant university in Alabama, located in the South-Eastern United States. With a student population of approximately 26,000 individuals, Auburn presented an ideal population from which to draw a sample to study the tipping motivations experience. While some will argue that students represent a convenient sample and that those kinds of samples may not be generalizable to other populations, students at such a university come from very diverse backgrounds, have large amounts of disposable income, and participate in restaurant patronage probably more than any other population sample. 76 This particular sample gave this researcher access to a sufficient data set, by virtue of the number of participants; however, the geographic diversity of the membership may not allow for both meaningful statistical evaluation and generalizability to all populations. In this case, it is the belief of the researcher, because there is a smaller minority population in Auburn in comparison to their Caucasian counterparts, that minority demographics may not be generalizable; however, the research may support earlier evidence of such findings. One thousand surveys were distributed via the researcher, student helpers, and other faculty, resulting in 784 usable surveys. Concern for Ethics The researcher followed proper procedures and research protocol while conducting this research on tipping motivations. The graduate student and supervising faculty were all CITI (Collaborative Institutional Training Initiative) certified. The Internal Review Board (IRB) of Auburn University reviewed the study prior to administration, examining the study considerations, survey instrument, supporting literature, and potential for injury. For each round of quantitative data collection, IRB approval for addition and amendment of the research was obtained. All ethical considerations having been met, the review resulted in approval by IRB to conduct the research. The initial instrument was administered directly to students and was collected by the principle investigator and his Committee Chair, with the help of other support faculty and staff members. The researcher analyzed data by using Microsoft Excel and PASW 18 statistical software. No identifying information, such as the participants? names, was retrieved during the data collection process. The protection of each participant?s identity was made an obligation. In order to conscientiously consider minorities and tipping behavior, securing the 77 participants? identities and the presentation of the data was crucial to ensure anonymity of each case. Data was obtained and presented in such a way to ensure anonymity of the participants. All of the surveys were collected in random fashion and numbered in sequence after data collection ended. All research methods share certain ethical challenges. The research process creates a natural imbalance between the aims of research to make generalizations for the good of others and the privacy rights of participants. The researcher must follow proper protocol with regards to seeking Internal Review Board (IRB) approval while obtaining an informed consent, and giving privacy assurance. The investigator must also fulfill the promise to instill no harm or injustice to participants. Through an unspoken, but implied relationship, the researcher is required to ultimately and accurately represent the population under question in an unbiased and fair approach. One notable challenge that is inherent in quantitative research is that of impartiality in creating and operationalizing the research instrument. On the other hand, qualitative research involves the researcher in a very intimate way; the researcher becomes the instrument. Thus, in a quantitative instrument design, the investigator must guard against coercing statements or prejudiced questions which may be unconsciously biased in order to achieve a predestined final result; in qualitative inquiry, the researcher must protect against compromising his/her objectivity through the direct contact with subjects. Interaction between the investigator and participant may create power imbalances and further complicate the research process. As posited by Eisner (1998), the researcher must be fully aware of his/her own tendency towards subjectivity and always strive to be as consistent and objective as possible. 78 This study benefited from the priceless insight and direction provided by experienced social science researchers. The committee managing this research includes four faculty members with extensive quantitative research experience, three of whom are senior faculty at the university. Multiple sessions were held with these researchers individually, during which the general outline of the study was formed and specific contribution was obtained. As for the sequential qualitative portion of the study, interviews were recorded and transcribed verbatim. By utilizing the process of member-checking, recorded interviews were constantly slowed or reviewed multiple times in order to solidify the researcher?s understanding of what the interviewees were saying. Clarification of the subject matter was sought at all times during the interview process. The researcher constantly read aloud and summarized what he believed he was hearing and obtained corroboration from the participant that the quote and/or point was correct. By taking and utilizing the skills received in graduate qualitative and quantitative statistics courses, the researcher successfully realized the inductive process. Frequent differentiation throughout the interactive process as confirmation of triangulated material began to mold the study. Respondents were assured that their responses were clear and well understood and that no meaning was compromised. Data Collection The survey was administered by hand to visitors on Auburn University?s campus in the fall semester of 2010 and spring semester of 2011. The instrument, including an informational letter that had been previously approved by the Internal Review Board of Auburn University, was administered throughout the university. The surveys were administered by a trained graduate 79 research assistant, senior students from a Food and Beverage Operations course (as part of an extra credit assignment), and by helpful faculty and staff. Potential participants were asked to complete a paper-based survey which would require approximately ten to fourteen minutes of their time. Upon acquiescence, the respondents were given the survey with the information letter attached to it. Survey administration was limited to two semesters. Paper surveys were returned to the researcher and the findings were input into PASW 18. Since the surveys were returned immediately, the need for envelopes, stamps, and mailings was eliminated. It is important to note that certain guidelines were followed during survey distribution. Individuals were questioned if they had participated in any of the other pilot studies. If so, then the survey was taken away and those individuals did not complete another survey. In addition, it is the belief of the researcher that all surveys were completed in earnest with each individual?s utmost integrity and truthfulness in responding to the instrument. Returned questionnaires The sample was collected from persons who visited Auburn?s campus. The questionnaire was designed to collect demographic and other information in order to enable analysis of the motivation derived from engagement in tipping. The survey was administered over a two- semester period encompassing August 2010 to April 2011, when the IRB provision expired. Seven hundred and eighty-four individuals responded to the solicitation of one thousand surveys. Acceptably completed returns (approx. 95%), offered an early assumption that almost everyone is clearly interested in his/her willingness to tip. 80 Summary This chapter provided a full description of the research undertaken and the measures and methods used to obtain the data. The sample group, data collection, and research tools used were described as well. Support for the design and approaches used in this study is based on literature regarding methodology, design, and analysis (Clark, Riley, Wilkie, and Wood, 2005; Creswell, 2009; Maxwell and Delany, 2004). The next chapter will present the data analysis, research results, and disposition of the hypotheses offered. 81 Chapter IV Results and Findings Introduction This chapter will present the results and findings obtained from the TMS questionnaire. The chapter will be divided into four sections. Section one will present a short description of the collected responses in the form of a detailed breakdown of the respondent demographic information and self-identification results, along with the scale?s behavioral measurement outcomes. Section two will present the measurement instrument properties. Section three will address the validity and reliability of the instrument, along with the technique used to determine non-response bias. Section four will report on the findings pertaining to the various research hypotheses. Lastly, the chapter summary will recap the overall results. Sample Demographics Table 4 revealed a female dominant environment with approximately 55% of all respondents classifying themselves as female; additionally, approximately 87% of the respondents self-described as Caucasian. The age range indicates a predominantly youthful respondent data set, with approximately 75% falling in the 19-24 year old age grouping. 82 Approximately 43% of the sample classified themselves as employed, with just over 10% indicating that they were unemployed and approximately 45% of respondents classifying themselves as students. Some 76% of the sample indicated that they did not work in the hospitality field. A further 82% of respondents indicated that they did not work in any tipped profession. Almost 99% of respondents indicated that they tipped when dining out, and just over 29% indicated that their own employment influenced their tipping behavior. Exactly 45% of the sample reported that they earned less than $15,000 per year, with approximately 23% earning more than $70,000 per year. The survey also went on to address tipping norms among the respondent group. This has long been a passionately debated topic within the literature as it relates to tipping behavior for normal, good and excellent service (Azar, 2005; Azar and Yossi, 2006; Bodvarsson and Gibson, 1999; Grassman and Lynn, 1990; Lynn and McCall, 2000). Looking first at tipping norms, some 23% of respondents indicated that their tipping norm was in the range of 1-10% of the check; 66% of respondents tipped in the range of 11-20% as a norm, and some 6.5% tipped in excess of 20% for normal service. In regards to tipping behavior for good service, just over 12% of respondents indicated that they would tip in the range of 1-10%. A little over 71% of respondents indicated that they would tip in the 11-20% range for good service, and some 14% of respondents indicated that they would tip in excess of 20% for good service. Respondent behaviors changed remarkably when it came to rewarding what they considered to be excellent service. Approximately 7% of respondents indicated that they would tip in the range of 1-10%, while 40% of respondents indicated that they would tip in the range of 11-20% and barely more than 40% of respondents indicated that they would tip above 20%. These figures certainly do help to shed a little light on the differences between what servers and 83 patrons consider adequate reward for all three levels of service. In addition, this is a clear indication that studying consumer motivations to tip is an important undertaking. In addition, close to 30% of the sample claimed that their particular, current occupation caused them to tip more. This finding is supported by research indicating that some in the hospitality industry tip more because they empathize with the server (Babcock, 2007). Moreover, the level of debate in gratuity amounts (15% or 20%) squared nicely with the literature reviewed regarding the current tipping norms (Azar, 2004; Videbeck, 2004) in the mind of restaurant goers. 84 Gender f % Male 355 45.3 Female 425 54.2 Missing 4 0.5 Age 1-24 585 74.6 25-39 96 12.2 40-59 79 10.1 60 + 18 2.3 Missing 6 0.8 Employment Status Employed 337 43.0 Unemployed 81 10.3 Student 354 45.2 Retired 8 1.0 Missing 4 0.5 Tipping Norm 1-5% 43 5.5 6-10% 139 17.7 11-15% 274 34.9 16-20% 244 31.1 20% + 51 6.5 Missing 33 4.2 Tbl 4. ? Demographic Profile of Respondents In the next section of the scale, the respondents were asked to self-identify their reasons for engaging in tipping in the future based on six options in Table 5. The table shows that when service is perceived as good by the consumer, over 70% of the respondents either agreed or strongly agreed that they would tip above their own pre-established tip norm. On the other hand, according to this sample of respondents, servers who provide good services are not likely to be requested in the future or recommended by customers in the future. This finding was interesting. From the researcher?s own personal perspective, this finding does have some merit. When consumers experience good service in an establishment, it is a common misconception that the 85 same quality or level of service will be provided by other servers in the same restaurant in the future. However, some consumers do not realize the importance of the server until after the fact. They (consumers) return to the restaurant for a second time, do not request the same server, and a poor experience or dissatisfaction occurs. It is in this moment that customers realize that all services are not standard, regardless of the server and, ultimately, the circumstances surrounding the visit. Moreover, consumers then will recognize that they should have asked for the same server as before. Even more interesting is that when patrons felt that service received was good, they would spend more money at the same restaurant in another visit. Over half of the respondents (approximately 55%) said that they would be willing to spend more money in the future. In addition, when service has been perceived as good, more than 35% of the sample agreed or strongly agreed that they would tip more in a future visit. This percentage does hold true to a common restaurant server belief that one should not anger a customer today, because he/she could be someone else?s return customer tomorrow. Furthermore, servers who work in teams (working in and around the same sections) can truly benefit from making sure that all guests are happy when they leave, because they (customers) will come back in the future and could potentially be someone else?s patrons. The results reported in Table 5 pair nicely with the tipping literature suggesting that service, even though it is believed to minimally explain tipping behavior (Lynn, 2001; Lynn, 2003), provides additional support for the theory. This is congruent with the survey results in that approximately 75% of the respondents stated that they would be more likely to revisit the establishment in the future due to the service being good. Good service is indeed a driver of future behavior. 86 Tbl. 5: Tipping Behavioral Norms Measurement Instrument Properties The measurement instrument used in this study was composed of the scale outlined in the earlier chapter, namely, the Tipping Motivation Scale. The scale was utilized to identify and measure those variables that best drive and/or explain actual tipping motivation. It was revised on two occasions as a result of initial pilot work; in its final form, it was thought to be representative of the multiple domains that had previously been identified during these same pilot studies as explaining the tipping motivation construct. Below is the description of the scale along with the descriptive statistics for each scale item. Strongly Disagree N (#) % Disagree N (#) % Neither N (#) % Agree N (#) % Strongly Agree N (#) % Tip Above Norm 29 3.7% 42 5.4% 158 20.2% 291 37.1% 261 33.3% Same Server 241 30.7% 168 21.4% 176 22.4% 116 14.8% 78 9.9% Recommend Server 235 30% 155 19.8% 188 24.0% 120 15.3% 81 10.3% Spend More 182 23.2% 146 18.6% 241 30.7% 139 17.7% 70 8.9% Tip More 136 17.3% 108 13.8% 243 31.0% 191 24.4% 102 13.0% Revisit More 30 3.8% 22 2.8% 136 17.3% 282 36.0% 308 39.3% 87 The final TMS is divided into multiple hypothesized domains and encompasses a total of 19 variables. The domains are as follows: 1. Service: composed of dimensions such as the technical (order taking, order delivering, tangible aspects) and functional (friendliness, personal service, eye contact, intangible aspects). 2. Social norm: consisting of dimensions of shame, guilt, regret, equitable relationships, and altruism. 3. Server actions: with the dimensions of touch, body language, and eye contact. 4. Future behavior: which includes the dimensions of gratitude, gratuity, and frequency of visit along with fellowship, rapport building, and self-esteem. 5. Peer pressure: consisting of dimensions of diffusion of responsibility or obligation to others. 6. Other operational processes: which examines the dimensions of restaurant space (physical environment or offerings). Table 6 illustrates the full complement of variables with each of the individual items divided by domain with their accompanying descriptive statistics. 88 Dimension / Variable Mean SD Skew Service V2 Timeliness of the service 4.21 .821 -1.123 V3 Poor service received 4.37 .801 -1.413 V6 Poor service influence 4.46 .823 -1.846 V7 Server?s attitude influences my tip 4.42 .771 -1.546 Social Norm V12 I feel obligated to tip when service is bad 2.99 1.267 -0.172 V15 I feel regret if I do not leave a tip 3.55 1.265 -0.608 Server Actions V4 Server?s body language influences my tip 3.62 1.004 -0.579 V5 Direct eye contact influences my tip 3.51 .968 -0.456 V8 Server?s menu knowledge affects my tip 3.48 1.039 -0.384 V11 Server?s ability to sell menu 3.14 1.137 -0.225 Future Behavior V16 Consider future service when tipping 2.97 1.160 -0.055 V19 Consider future visitation when tipping 3.07 1.164 -0.138 Peer Pressure V9 I feel obligated to tip when dining with family and friends 2.93 1.145 -0.074 V10 I tip more when others in my party do not tip 3.68 1.108 -0.573 Other Operational Influences V1 Gender 2.23 1.272 0.593 V13 The general restaurant expense 3.40 1.064 -0.520 V14 Where I am seated 2.37 1.131 0.444 Tbl. 6: Univariate Analysis of Tipping Motivation Scale When examined from a descending mean value perspective, ?Service? truly stands out in most respondents? minds when it comes to explaining their actual tipping motivations and/or behavior; V6, ?poor service influence? (m=4.46) appears to be most significant in explaining actual motivation, followed by V7, ?server?s attitude? (m=4.42) and V3, ?actual service received? (m=4.37). Of least significance seem to be those other operational variables that pertain to seating assignment V14, (m=2.37) and V1, gender (m=2.23). 89 Validity and Reliability Tests Factor Analysis Attention now turns to psychometric performance of the research instrument and, in particular, the factors that influence the consumer?s actual tipping motivation and/or behavior. Application of the factor analysis technique in this research enables the researcher to search for and reveal coherent subscales specific to tipping. The sample (n=784) was divided randomly into two data sets. The first set was examined using exploratory factor analysis (EFA). The specific EFA analysis method employed in this study is the principle components analysis (PCA). The PCA method was chosen specifically because it is psychometrically sound and, according to Stevens (2002), avoids factor indeterminacy. The further advantage to using PCA is that it provides an empirical summary of the data set (Tabachnick & Fidel, 2007). The resultant output was examined, and a theoretical model based on the empirical data from the EFA results was designed. The second data set was subjected to a confirmatory factor analysis (CFA), using the AMOS 18.0 statistical software package to seek confirmation of the theoretical model. Research Criteria For variables to be factorable, certain assumptions and circumstances regarding the data set should be met; the following describes the standards used in this study: 90 1. General assumptions?the data must be shown to have inter-item correlation; variable pairs must be normally distributed, each case independent of the others and exhibiting linearity to the relationships between variables (Mertler and Vannatta, 2005). The results of the analyses showed that the distribution was normal; the factors were sufficiently correlated to hang together, yet sufficiently differentiated to be measuring different aspects. 2. Sample size?for meaningful factor analysis, Tabachnick and Fidel (2007) recommend that the sample have at least 300 cases and a minimum of a five to one ratio of subjects to variables. The n of 784 makes the sampling adequacy robust to this assumption. 3. Variance?according to Tabachnick and Fidel (2001), to be robust the factor solution should account for at least 50% of the variance. 4. Quantitative scales?the variables must be measured continuously. 5. Factorability of the correlation matrix?Bartlett?s test of sphericity should obtain an alpha of .05 or smaller and the Kaiser-Meyer-Olkin (KMO) measuring sample adequacy should obtain a minimum value of .6 or higher. Steven?s (1992) benchmark of .4 as the minimum standard for considerable values is used. 6. Factor selection?it is usually recommended that the study employs the eigenvalue is greater than one criterion, but often, as described by Patil, Singh, Mishra, and Donavan (2008), this leads to too many factors. For this study, only factors that have an eigenvalue greater than one will be considered. In addition, the application of Catell?s (1966) scree test is used only with eigenvalues greater than one. The combination of these two criteria better limits the factor selection to the most expressive subscale factors. 91 7. Rotated component matrix?the concept of rotated factors is intended to best present the solutions in a pattern of loadings for ease of identification. The choice resides between the use of orthogonal (most commonly Varimax rotation) and oblique (most commonly Direct Oblimin) factor solutions. In practice, both approaches usually result in similar solutions (Tabachnick and Fidel, 2007). For this study Varimax rotation has been selected for its ease of interpretation and reporting. Tipping Motivations Scale (TMS) The 19 items of the final TMS were tested to identify the factors that best explained respondent motivation as it applied to tipping behavior. Prior to performing PCA, the ability of the data to meet the assumptions of normality of distribution, independence, linearity, and sample size was tested and accepted. The PCA initially revealed the presence of six factors with eigenvalues above 1.0, explaining 63.31% of the variance. Upon further analysis, three variables (V1, V8 & V14) were found to cross-load across multiple factors and, in keeping with the perspective of Patil et al (2008) that reliance on eigenvalues alone can defeat parsimony, the pertaining scree plot was examined; it was determined that a five factor solution better represented the underlying structure, using fewer variables and dimensions. The Kaiser-Meyer- Olkin (KMO) measure of sampling adequacy result of .759 exceeded the recommended value of .6, and Bartlett?s test for sphericity result of 2195.18 p. < 0.001 supported the initial factorability of the correlation matrix. PCA was re-run on the newly obtained five-factor model with 16 variables. The data remained favorable for factor analysis with Bartlett?s Sphericity test score 1806.02 p. <0.001, 92 and the KMO result .749 again remained consistent above the .6 benchmark. The five-factor analysis explained a robust 62.13% of the total variance; it used fewer variables, thus realizing the purpose of factor analysis to reduce the data to the fewest factors possible to explain the underlying structure. The modified TMS with 19 items was further tested and, to aid in the interpretation of these five components, Varimax rotation was performed. The rotated solution revealed an optimal loading result offering the simplicity of structure called for seminally by Thurstone (1947). The loadings were clear, each with considerable values, all of them loading on only one component. The five components are identified as Service (F1), Conformity (F2), Other (F3), Future (F4), and Server (F5). Cronbach?s Alpha for the revised scale is ?=0.745, indicating strong reliability for the factors. The validity of the instrument is supported by the factor loadings and clarity of the underlying factor structure; the reliability scores for each of the factors uncovered a range from a mediocre ?=0.534 for ?Other? operational processes to ?=0.821 for ?Service.? 93 VARIABLES F1 F2 F3 F4 F5 Service Conformity Other Future Server V6 V2 V7 V3 V17 V15 V10 V12 V9 V11 V18 V13 V16 V19 V4 V5 Eigenvalue % Variation Coefficient ? .799 .796 .786 .769 .801 .779 .647 .597 .732 .648 .593 .496 .877 .853 .819 .812 3.432 2.857 1.424 1.165 1.064 21.448 17.853 8.898 7.283 6.653 .821 .704 .534 .762 .751 Tbl. 7: Rotated Component Matrix Confirmatory Factor Analysis The variables and dimensions identified in the EFA provided the empirical data from which a theoretical model could be designed and tested for confirmation with the second half of the sample reserved for this test. The theoretical model is presented in Figure 5. 94 Fig. 5: Theoretical Model 95 Fig. 6: Model with Beta Weights Illustrates the theoretical model with the computed standardized beta estimates 96 The confirmatory factor analysis was conducted using the second half of the data set. The study examined the 16 Tipping Motivation indicator variables and five factors revealed in the EFA. The theoretical model was assessed through the AMOS 18 software package using the maximum likelihood factor analysis (Arbuckle, 2007). The model was evaluated by five fit measures: (a) the chi square, (b) the normal fit index (NFI), (c) the comparative fit index (CFI), (d) the root mean square error of approximation (RMSEA) and Hoelter?s (1983) critical N (CN)?labeled as Hoelter?s .05 and .01 indexes. The results generally supported the proposed model. While the chi-square had a value of 386.36, (df 94, n=399) p = <.001, pointing to the fact that the model did not fit the data, other fit measures were good. The NFI and CFI are measures of relative fit comparing the theoretical model with the null model. The optimum value of .95 for these indexes, though very close, was not obtained; the NFI value was .87 and the CFI value was .90, indicating moderate goodness of fit across both categories. The RMSEA measures the discrepancy between the sample and population coefficients, with a value <.08 indicative of a well-fitting model. The RMSEA was .063. Since the closer to zero the RMSEA, the better the model fit, the obtained .063 score indicates a reasonable fit to the model (Meyers, Gamst, and Guarino, 2006). The final goodness- of-fit statistic examined was Hoelter?s critical N, which differs substantially from those previously discussed in that it focuses directly on the adequacy of sample size, rather than on model fit. Specifically, its purpose is to estimate a sample size that would be sufficient to yield an adequate model fit for a ?2 test (Hu and Bentler, 1995). Hoelter proposed that a value in excess of 200 is indicative of a model that adequately represents the sample data. The results confirm that the size of our random sample (N=399) was satisfactory according to Hoelter?s benchmark that the CN (239) should exceed 200 (Byrne, 2001). 97 Further examination of the model identified two offending items on the F4 (Other) Dimension. Variables V9 and V13 had less than satisfactory inter-item correlations (0.45 and 0.34 respectively). This seems to indicate that there may have been some confusion as to the nature of both questions in the final TMS survey. Both items were subsequently removed from the model and the analysis was re-run. The respecified model was again evaluated by five fit measures and the results pointed to a stronger fit between the model and the data set. While the chi-square had a value of 284.337, (df 67, n=399) p = <.001, again pointing to the fact that the model did not fit the data, other fit measures improved considerably. The NFI and CFI are measures of relative fit comparing the theoretical model with the null model. The NFI value was .90 and the CFI value was .92, indicating moderate goodness of fit across both categories. The RMSEA was .064 which again indicates a reasonable fit to the model (Meyers, Gamst, and Guarino, 2006). The final goodness-of-fit statistic examined was Hoelter?s critical N. The results confirm that the size of the random sample (N=399) was satisfactory according to Hoelter?s benchmark that the CN (240) should exceed 200 (Byrne, 2001). 98 Fig. 7: Re-specified Theoretical Model with Beta Weights 99 Research Questions and Testing of Central Research Hypothesis As previously indicated, the overriding research question (RQ1) for this study was presented as: What motivational factors drive/influence the consumer?s decision to tip? Put simply, the study was intended to shed light on the actual factor structure of the motivational tipping construct. To this end, the central research hypothesis was posited as follows: H1 - The motivational tipping construct will present itself as multi-dimensional in nature, influenced by a range of motivational themes including service, social compliance, server actions, future behavior, peer pressure and other operational processes. The results of the preceding EFA and CFA analyses not only confirm this hypothesis, but help to shed light, at least with respect to this sample group, on the actual factor structure, which in this case comprised five unique dimensions explaining just over 63% of the variance. These results also support the results from earlier pilot studies undertaken by the researcher in development of his scale, and certainly point to a high degree of repetition when it comes to such dimensions as service, social pressure to conform, server action and future service-related issues. These motivational factors are presented in the theoretical model in figure 6. Attention next turned to addressing the additional research questions (RQ2 and RQ3). The second research question (RQ2) pertained to the extent to which actual service delivery influences the consumer?s decision of whether or not to tip. Regardless of social pressure to conform and the need to manage risk aversion, the overriding tipping motivation for the vast majority of consumers remains the actual service received during the normal service exchange. The second research question was thus presented as follows: 100 RQ2 ?To what extent will the actual service delivery be the most important motivational influence for consumers in their decision to tip or not? The preceding EFA and CFA results clearly point to the influential role of actual service delivery as a motivator for actual tipping behavior. The revealed ?Service? dimension accounts for the greater proportion of explained variance (21.44%). This portion testifies to the importance of the service act for most consumers when it comes to their motivation to tip or not. The final research question (RQ3) speaks to the issue of social conformity and the extent to which consumers will continue to tip regardless of the quality of service received. There is a paucity of evidence (Azar, 2004; Azar, 2005; Azar, 2007, Bodvarsson, Luksetich, and McDermott, 2003; Conlin, Lynn, and O?Donoghue, 2003; Lynn and Grassman, 1990) that regardless of the service received, most consumers still feel obliged to tip due to social pressure to conform. Thus, the third research hypothesis was presented as follows: RQ3 ? To what extent will consumers continue to tip regardless of whether they receive good or bad service? In order to test this research question, a simple, yet crude descriptive statistic was first run on scale variable 12, which simply read ?I feel obligated to tip even when service is bad.? The overall mean for this variable was m=2.99, which indicated a stronger degree of agreement than disagreement on the scale employed. An additional analysis of related frequencies lent further tentative support to this issue, with approximately 40% of respondents indicating that they either ?agreed? or ?strongly agreed? that they would tip regardless of bad service. A further 21% were undecided, with just over 35% of respondents indicating that they either ?Disagreed? or ?Strongly Disagreed? that they would tip in the event of bad service. 101 Reliability Checks Quantitatively, Huck (2004) defines reliability as the consistency of a measure. Put simply, the ability of solutions to reach valid reliability scores indicates that the measure itself, not error or chance, explains the result. Reliable research instruments allow future research to consistently measure the same factors, and different results can be reliably attributed to differences in the sample and not to the instrument itself. Internal consistency was calculated for the scale and then for each of the components identified within the scale from the PCA. While there remains disagreement within the literature as to the standard for reliability scores (with some calling for .50 and above and others calling for the more stringent .70 or higher), there is general agreement that the higher the number (closer to one), the greater the internal reliability of the instrument, thus increasing the likelihood that error or chance did not produce the result. Table 8 illustrates the results of the scale and each dimension created from the factor analysis outlined above. The table clearly indicates a robust coefficient for the TMS both in the total scores and within the identified dimensions of each scale. Both the initial (19 item) and modified (16 item) scales demonstrate good reliability scores in excess of .70 (Cronbach, 1951). Similarly, all but one of the revealed dimensions (Other / ?=.534) demonstrated good reliability. 102 Tipping Motivation Scale Coefficient ? Tipping Behavior Scale .793 Initial Scale (19 items) .747 Modified Scale (16 items) .745 Factor 1 ? Service .821 Factor 2 ? Conformity .704 Factor 3 ? Other .534 Factor 4 ? Future .762 Factor 5 ? Server .751 Tbl. 8: Coefficient Alpha of Scale Validity Quantitatively, validity measures accuracy (Huck, 2004). Put in another way, validity is the underlying soundness of the instrument, signaling sufficiency that the instrument does indeed measure what it is purported to measure. Validity for this study has been determined using content validity and construct validity, with the attendant sub headings of convergent and discriminant validity. Content, or face validity as it is more often termed, was assessed through expert input during the qualitative work leading up to the development of the instrument (focus groups, card sort and expert interviews). As previously documented in the methods section of the dissertation, the feedback obtained led to changes and improvements in specificity of the final TMS items, clarity of the questions, and the questions relevant to the activity of tipping. Additionally and as a result of two consecutive pilot studies, it is felt that the items contained in the final TMS were adequately assessed in terms of time necessary for completion, their clarity, ease of understanding and interpretability. The feedback obtained resulted in further changes 103 designed to enhance the final TMS instrument. The instrument was found to have content validity as determined by the expert review and field test population. Construct validity was addressed in terms of both convergence and the research instrument?s ability to discriminate between the underlying dimensionality of the consumer tipping motivation construct. Convergence was investigated by calculating the mean score for each of the 19 scale items and correlating (Pearson?s product moment correlation) this with the mean score from the six item scale attesting to actual consumer tipping behavior as an outcome of receiving good service. Each of these scale items was presented as--when I tip for good service I tend to: (1) tip above the norm, (2) ask for the same server on future visits, (3) recommend the server to friends and family, (4) spend more time on future visits, (5) tip more money on future visits and (6) visit a restaurant more often. Each item was anchored at (1) Strongly disagree through to (5) Strongly agree. A correlation of 0.396 was found which, while low, was nonetheless significant at the 1% level (p<0.001). Tests of Non-Response and Late Response Bias A further issue that needs to be addressed relates to the issue of non-response and late-response bias, the existence of which limits the ability of the researcher to generalize findings from a respondent sample to a population of interest. According to Churchill (1996) this represents a failure to obtain information ?late or not at all? from some elements of the population selected and designated for the sample. The previous description of sample results makes it clear that the researcher achieved a very high response rate of some 78% (784 returns) out of a total of 1000 distributed questionnaires. While no opportunity presents itself to track 104 issues related to non-response, many of the non-responders indicated to the researcher that time was very much an issue and thus they did not wish to be inconvenienced. While the researcher can hypothesize on the causes of non-response, he can offer much more concrete evidence on the non-existence of any late-response bias. The researcher was able to track those early responders who completed the questionnaire in the fall semester 2010 (cases 1-361) and those who completed the questionnaire in the spring semester of 2011 (cases 362-784). One widely supported method to assess non-response bias is to compare characteristics of early respondents with late respondents. If differences are found between these groupings, the indications are that non-respondents are likely to be different as well. Conversely, if there are no significant differences found between early and late respondents, then support is provided that the survey results are more likely to be generalizable to the population under consideration. The underlying rationale is that early respondents are more likely to be motivated and exhibit higher enthusiasm than late respondents or non-respondents. This is because early respondents tend to have stronger feelings in the area under examination. To determine if non-response bias was a problem, the sample was split between fall 2010 and spring 2011 respondents who participated in the survey. This created a comparison base of 361 respondents classified as early respondents and 423 classified as late respondents. Statistical analysis comparing the group means and the total scores from the TMS and TMB was computed using the independent samples t-test, with no statistically significant differences found between the two groups. This finding offers strong support that non-respondents too would not be different and increases the confidence level of the generalizability of the results obtained from this study. 105 Summary Chapter four presented the statistical results compiled from the research instrument in relation to each of the research questions and related hypotheses. Additionally, results pertaining to reliability, validity and response bias check tests were reported. These results included a comprehensive overview of the respondents? demographic characteristics, such as age, income, education, employment information, and important information pertaining to their normal tipping practice. Analysis of the measurement properties of the instrument was conducted as well, using the exploratory factor analysis technique of Principle Component Analysis, supported by obtaining the eigenvalues of the data and analysis of the scree plots. The modified measurement scale that most accurately revealed the relevant subscales and variables for the participant?s tipping motivation was determined and presented. The theoretical model was subjected to Confirmatory Factor Analysis, and the corresponding goodness-of-fit tests were found to be good. The following chapter will review the findings, discuss the implications of the study, and identify future research potential. 106 Chapter V Conclusion Overview As previously mentioned in the methodology section, the research associated with the study involved both qualitative and quantitative research. Qualitative research consisted of focus groups and one-on-one interviews, the results of which were used to establish a basic understanding of what was important to consumers in their motivations to engage in tipping behavior. Quantitative research consisted of two-pilot studies with sample groups composed of students and tailgating football fans at Auburn University. Confirmatory study was composed of visitors to Auburn?s main campus. This researcher intended the TMS to measure and assess tipping motivations and future behavioral intention when service was characterized as good. In order to measure this custom, survey administration was conducted over the course of the fall semester of 2010 and the spring semester of 201l. This was done to ensure that an ample amount of completed surveys would be collected and in an effort to measure not only tipping motivations, but tipping behavior as well. This chapter is divided into four sections. First, a brief description of the study is given. Secondly, the research questions, along with results, will be synthesized. Third, a review of the significance and also the contributions of the study is presented, along with derived implications. 107 Next, the researcher will present future research opportunities designed to improve and advance this research. Finally, a brief summarization of both the chapters and the study as a whole will conclude the chapter. Description and Purpose of the Research As outlined in Chapter I, this study has been undertaken primarily to determine whether or not multiple factors such as service, social compliance, server actions, future service, peer pressure, and operational processes could be simultaneously addressed in terms of a newly devised quantitative scale. The secondary aim was to identify those factors, through the use of qualitative and quantitative methods, to reveal which, if any, were necessary to add to the scale to support the tradition of tipping as multi-dimensional in nature. Utilization of two pilot studies and exploratory factor analysis of half the data set accomplished this goal. These steps were taken in order to provide the empirical grounds from which to design the theoretical model. The model was then subjected to confirmatory factor analysis using the second half of the data set to measure the fit to the model, achieving moderate goodness of fit. However, the instrument may need additional information added to the scale; moreover, other dimensions could plausibly be taken away in order to improve the model and results. A certain disadvantage of this study may be the lack of further qualitative data. Tipping is a behavior; therefore, it is necessary to physically observe the participants in the environment in which they engage in the activity. The use of a more mix-methods approach, triangulation of the data, along with the convergence of the dimensions and added quotes obtained from the survey, interviews, and focus group members, may solidify the results of this research. 108 Certainly, the interest and, ultimately, participation of the respondents lie at the very core as to why examining the consumer?s motivation to tip is important. Overall, the findings which were obtained in this study give clear indication that the instrument under development is on track; through the addition of the revealed dimensions, it is probable that the model will be confirmed in future research. The development of a quantitative instrument to be applied to tipping has been the specific primary purpose of this research. Adding this methodological element, that simultaneously utilizes multiple factors (potential motivating dimensions) in order to explain the custom, represents an original contribution to and a further development of the tipping literature. The development of an instrument that expands the understanding of tipping serves both an operational (micro level) and cultural function (macro level). As noted in the first chapter, participation in tipping is entrenched in the lives of everyday Americans. This research adds a tangible benefit to an entire society?s way of life. Seemingly, the benefit of understanding tipping motivations as they relate to the overall impact of restaurant workers can increase employee and occupational satisfaction, along with emotional health and, ultimately, financial success. The ability to control the dining experience to the extent to which those participating are satisfied and, therefore, leave monetary compensation to the server who ultimately provides the experience is, without a doubt, worthy of the effort needed to learn more about tipping behavior. This is especially true in such a dynamic and rapidly changing economic environment. The chance to progress the knowledge base in an area that speaks so deeply to both the individual and the societal level supports the significance of the study in both specificity and overall. The sample population is adequate (N=782); however, the cultural implications of this research may not be fully realized, as minorities are not well represented, thus presenting a low 109 response rate. On the other hand, the respondents represent a wide spectrum of participation engaging in tipping behavior. Resultantly, the research utilized an ample number of participants to satisfy the statistical output; and the shared experience of tipping motivations certainly support the claim of generalizability to other populations, particularly by qualitative and quantitative methodologists. Considering the similarity of the findings among tipping researchers, it is a reasonable proposal that the instrument, once finalized, could provide theoretical framework for future research. Implications The implications that can be taken away from this research are profound. From an academic standpoint, it is the belief of the researcher that this is the first quantitative scale aimed at identifying tipping motivations to be developed and tested utilizing statistical rigor. The results quite evidently call for training and development of efficient operations managers and service staff. Understanding the motives that drive consumers to tip can be useful in building training programs for servers and restaurant managers. Academicians and occupational counselors can benefit from a deepened understanding of tipping motivations in terms of counseling potential students/workers as to what seems to be the appropriate fit of employment. Based on a working knowledge of the motivational factors driving consumers to tip, these educators can certainly advise their students with a two-fold consideration. First, based on certain motivational dimensions, educators can suggest specific restaurant venues that may best benefit a particular individual, based upon economic or monetary needs and personal fulfillment. For example, an upscale restaurant, or in particular, a restaurant having a clientele base willing to 110 pay higher prices, will potentially lead to higher monetary reward. Secondly, adding a familiarity of tipping motivations to a career server?s work knowledge can offer the intrinsic and extrinsic benefits that can enable him to persevere in what might otherwise be an unfulfilling occupation. Moreover, this adds to the benefit of both the employer and employee. In terms of the employer, a tried and true principle is that when employees are happy, the guests are happy; happy guests are satisfied guests who will return and, hopefully, increase brand loyalty and repeat business. As previously mentioned, knowledge of tipping motivations will provide coping mechanisms that enable workers to understand that certain behaviors (within their control) can increase the chances or likelihood of a tip. The broad scope of the restaurant industry can all benefit from the empirical understanding offered in this research. Furthermore, and even more interesting, is that this research has broad implications for not only those in the hospitality industry (travel, leisure, and food and beverage), but potentially all service industries which benefit from the custom of tipping. The literature is clear that tipping is a very important topic and is highly relevant to the lives of many restaurant workers. The survey participants themselves have directly indicated that they tip based on service. In addition, they indicated a strong preference towards making plans both to revisit and consider future restaurant visitation based upon the service they receive. An inference can be made that this research can truly be used to give strong support to tipping literature. In addition, restaurateurs who display commitment and knowledge to a broad-market base can gain competitive advantage due to the commitment and show of expertise. This seems especially true of restaurateurs who can provide the knowledge which, ultimately, satisfies and further aids their employees in economic and over-all job satisfaction. 111 The development of a scale addressing tipping motivations has numerous implications transcending multiple disciplines, especially for those who are both the applied and theoretical aspects of this research inquiry. In essence, business scholars and human resource managers can determine better employee fit and long-term success through application of tipping motivations construct and application of this research. For example, positively increasing an employee?s productivity and working knowledge may lead to reduced overhead (costs) that are associated with hiring, retaining, and developing staff members. Put simply, while this does not provide a panacea to all of the possible considerations of those in tipped professions, awareness of tipping motivations that can be derived from the scale may play an important part in the consumer?s willingness to tip. What is evident and consistent in this research is that the category of service explained the most variance in all three studies--17.52 % (pilot 1), 21.96 % (pilot 2), and 21.49 % (confirmatory study). More importantly, this finding is supported by a prolific amount of research that confirms service as a common tipping motivator (Azar, 2003; Azar, 2004; Azar, 2005, Babcock, 2007; Bodvarsson and Gibson, 1999; Bodvarsson, Luksetich, and McDermott, 2003; Grassman and Lynn, 1990; Kwortnik, Lynn, and Ross, 2005). Also, in all three studies, the category of social norms and/or compliance explained the second largest contribution of the variance found in each of the studies--11.35 % (pilot 1), 15.35 % (pilot 2), 17.85 % (confirmatory study). This explanation is consistent with the literature on tipping and the role of social norms (Azar, 2004; Azar, 2005; Azar, 2007; Conlin, Lynn, O?Donoghue, 2003; Lynn and Grassman, 1990). Understanding and acting on the motivational drivers can play an important role and make a significant contribution to the well-being of the restaurant server and, ultimately, provide him/her with further arsenal in their respective fields of restaurant operations. 112 Limitations Again, while effort was taken to minimize the amount of limitations this research may contain, caution is advised in generalizing this study to other populations. The following is offered to highlight potential short comings in an effort to improve the theoretical research design for a future researcher who may be inclined to build upon this study. The limitation lies with the sample groups. A major portion in all three of the studies sample populations were students from Auburn University. Therefore, tipping motivations may, in fact, be different for students from other regions. The survey was administered by several different people. This may add to the issue of whether or not the instruments were true depictions of each individual respondent. In addition, there may have been an issue with clarity or understanding of the survey, along with the effective administration and retrieval of completed responses. Considering the exorbitant amount of surveys which are given by companies today, potential participants may have thought twice before taking or failed to complete the survey in its entirety (i.e., survey fatigue). The survey was administered by means of paper and pencil. This method may lend itself to error in transposing the data from paper copy to electronic form (in terms of data set entry). Therefore, this did potentially exclude respondents who may have wanted to complete the survey and otherwise did not get the chance. On a final note, the overriding objective of this study was to develop a scale which measured tipping motivations. Since this was the uniqueness of this particular study, it is difficult to compare this research as empirical literature which is lacking in this field. 113 It is also noteworthy that the researcher himself is a career server and hospitality professional. His own belief is that service is the number one driver of consumer motivations to tip. In addition, given the fact that he is a career server, he has experienced firsthand that other restaurant servers do, indeed, take care of their own. He has a firm belief that this research is important in the lives of other restaurant servers. Future research The first step for future research should be a qualitative one. Qualitative analysis such as further one-on-one interviews with industry experts and additional focus groups, to aid in the refinement of the scale, is warranted. Secondly, in order to support this additional information, actually observing consumers in the act of tipping might help solidify some of the short comings in this research. In retrospect, the model tested here obtained a result which was approaching goodness of fit to moderate goodness of fit. Once the scale has been modified with additional factors or lessened dimensions, a modified or improved theoretical model could be examined. Resultantly, the newly devised instrument can be tested for potential improved goodness based upon an improvised theoretical model. If, in fact, confirmation of the newly devised model was obtained, the next step would be to repeat the study by applying it to other motivations and behaviors. The improvement of a well-established quantitative scale that could be applied across restaurant operations could be a significant contribution to the hospitality industry as a whole. Understanding the underlying motivations of tipping within the restaurant industry will give an array of benefits to restaurant hostesses, servers, bartenders, managers, vocational trainers and many others. 114 Conclusion This study has presented supportive analysis of possible tipping motivations. The benefits obtained from this research are obvious. More importantly, in terms of a restaurant server?s economic way of life, the development of a scale that would aid in possibly furthering his/her economic position can only prove to have a positive impact. This study developed an in- depth understanding of consumer tipping motivations and behavior and gave a further direction to the development further motivation scales. 115 References Ajzen, I. (1988). Attitudes, personality, and behavior. Chicago: Dorsey Press: Prentice-Hall. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes. 50, 179-211. Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84, 888-918. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Engle-wood-Cliffs, NY: Prentice Hall. Anderson, J. E., & Bodvarsson, O.B. (2005). Tax evasion on gratuities. Public Finance Review, 33, 466-487. Arbuckle, J. (2007). AMOS 16.0 User's Guide Arnold, S. (2004). Racial differences in tipping; Evidence from the field. Cornell Hospitality Quarterly, 45(1), 23-29. Azar, O. H. (2003). The implications of tipping for economics and management. International Journal of Social Economics 30(10), 1084-1094. Azar, O. H. (2004). The history of tipping--from sixteenth-century England to the United States in the 1910?s. Journal of Socio-Economics 33(6), 745-764. Azar, O. H. (2004). What sustains social norms and how they evolve? The case of tipping. Journal of Economic Behavior & Organization 54(1), 49-64. 116 Azar, O.H. (2005). Incentives and service quality in the restaurant industry: Tipping-service puzzle. Retrieved online from http://mpra.ub.uni-muenchen.de/4457/ Azar, O.H. (2005). The social norm of tipping: Does it improve social welfare? Journal of Economics 85(2), 141-173. Azar, O.H. (2006). Who do we tip and why? An empirical investigation. Applied Economics 37(16), 1871-1879. Azar, O.H. (2007). Do people tip strategically, to improve future service? Theory and evidence. Canadian Journal of Economics 40(2), 515-527. Azar, O.H. (2007). Why pay extra? Tipping and the importance of social norms and feelings in economic theory. Journal of Socio-Economics 36(2), 250-265. Azar, O. H. (2008). Do people tip because of psychological or strategic motivations? An empirical analysis of restaurant tipping. Applied Economics 42(23), 3039-3034. Azar, O.H., & Yossi, T. (2006). Tipping as a strategic investment in service quality: An optimal-control analysis of repeated interactions in the service industry. Retrieved online from http://mpra.ub.uni-muenchen.de/4452/ Babcock, M. (2007). The effects of personal background on tipping behavior. Retrieved online from http://www.deedspublishing.com/objects/shortStories/ Mark%20-%20Tipping%20Paper%20-%20short%20story.pdf. Barkan, B. & Israeli, A.A. (2004). Developing a framework for rewards in combined production/service businesses. International Journal of Service Industry Management 15(5), 444-459. Barron, R.B.V. (2005). The Parrot Effect. How to increase tip size. Cornell Hotel and Restaurant Administration Quarterly 46(1), 79-84. 117 ?Behavior.? (n.d.). In Merriam-Webster?s online dictionary. Retrieved January 28, 2011, http://www.merriam-webster.com/dictionary/behavior Bell, R. A, & Winters, L.C. (1993). Using marketing tools to improve employee relations. Cornell Hotel and Restaurant Administration Quarterly 34(6), 38-42. Bodvarsson, O.B., & Gibson, W.A. (1999). An economic approach to tips and service quality: Results of a survey. The Social Science Journal 36(1), 137-147. Bodvarsson, O.B., Luksetich, W.A., & McDermott, S. (2003). Why do diners tip: Rule-of-thumb or valuation of service? Applied Economics 35, 1659-1665. Bogdan, R. C., & Biklen, S. K. (2007). Qualitative research for education: an introduction to theories and methods: Fifth Edition: Pearson. Bourne, L.E., & Russo, N.F. (1998). Psychology: Behavior in context. New York, NY: W.W. Norton & Company, Inc. Boyles, W.J., Mounts, W.S., & Sowell, C. (2006). Restaurant tipping: Free riding, social acceptance, and gender differences. Journal of Applied Social Psychology 34 (12), 2616-2625. Brenner, M. L. (2001). Tipping for Success: Secrets for How to Get In and Get Great Service. Sherman Oaks, CA: Brenmark House. Brigid, C. (2001). Tipping in New Zealand?s restaurants. International Journal of Hospitality Management, 20(2), 219-225. Byrne, B.M. (2001). Structural equation modeling with AMOS: basic concepts, applications, and programming (multivariate applications series). Mahwah, NJ: Lawrence Erlbaum Associates. Campbell, D. T. (1963). Social attitudes and other acquired behavioral dispositions. In S. 118 Koch (Ed.), Psychology: A study of a science. (6, pp. 94-172). New York: Mc-Graw Hill. Card, J.A., & Kestel, C. (1988). Motivational factors and demographic characteristics of travelers to and from Germany. Society and Leisure 11, 49-58. Cattell, R. (1966). The scree test for the number of factors. Multivariate and Behavioural Research. 1 (2), 245-276. Cialdini, R.B., & Goldstein, N.J. (2002). The science and practice of persuasion. Cornell Hotel and Restaurant Administration Quarterly 43(2), 40-50. Clark, M., Riley, M., Wilkie, E., & Wood, R. (2005). Research and writing dissertations in hospitality and tourism. London, UK: Thomson Learning. Conlin, M., Lynn, M., & O?Donoghue, T. (2003). The norm of restaurant tipping. Journal of Economic Behavior and Organization 52 297-321. Corley, J. (n.d.). The behavior of business knowledge, behavior and the company?s bottom line. Retrieved January 30, 201l http://www.joanncorley.com/uploads/The_Behavior_of_Biz_-_c.pdf Creswell, J. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (Third ed.). Thousand Oaks, CA: Sage Publications. Crompton, J.L.(1979). ?Motivations for Pleasure Vacation.? Annals of Tourism Research 6 408-424. Crompton, J.L. & McKay, S. L. (1997). Motives of visitors attending festival events. Annals of Tourism Research 24(2), 425-439. Cronbach, L. (1951). Coefficient and alpha and the internal structure of tests. Psychometrika, 16(3), 297-334. 119 Crusco, A.H. & Wetzel, C.G. (1984). The Midas touch: The effects of interpersonal touch on restaurant tipping. Personality and Social Psychology Bulletin 10, 512-517. De Vaus, D.A. (1996). Survey and social research. 4th Edition UCL Press Limited. London, England. Dirks D., & Rice S.K. (2004). Dining while black: Tipping as social artifact. Cornell Hotel and Restaurant Administration Quarterly 45(1), 30-47. Dunn Ross, E.L., & Iso-Ahola, S.E. (1991). Sightseeing tourists? motivation and satisfaction. Annals of Tourism Research 18, 226-237. Eisner, E. (1998). The enlightened eye. Upper Saddle River, NJ: Prentice Hall. Fehr, E., & Falk, A. (2002). Psychological foundations of incentives. European Economic Review 46(4-5), 687-724. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley. Freeman, S., Walker, M.R., Borden, R., & Latane?, B. (1975). Diffusion of responsibility and restaurant tipping: cheaper by the bunch. Personality and Social Psychology Bulletin 1, 584-587. Gleitman, H. (1986). Psychology (2nd Ed.). New York: Norton. Guba, E.G., & Lincoln, Y.S. (1982). Epistemological evaluation: Improving the usefulness of evaluation results through responsiveness and naturalistic approaches. San Francisco, CA: Jossey-Bass. Gunton, G. (1896). The social economist. Gunton?s Magazine of American economics and political science, 10, 504 pgs. Hair, J.F., Anderson, R.E., Tatham, R.L., and Black, W. (1998). Multivariate data analysis. 5th 120 Ed. Upper Saddle River, NJ: Prentice Hall. Holloway, W., & Jefferson, T. (1997). Eliciting narrative through the in-depth interview. Qualitative Inquiry, 3(1), 53-64. Hsieh, A.N., & Wu, D.H. (2007). The relationship between timing of tipping and service effort. The Service Industries Journal 27(1), 1-14. Hu, L.T., & Bentler, P.M. (1995). Evaluating model fit. In R.H. Hoyle (Ed.). Structural equation modeling: Concepts, issues, and applications. (pp.76-99). Thousand Oaks, CA: Sage. Huck, S. (2004). Reading statistics and research. Boston, MA: Pearson Education Inc. Iso-Ahola, S.E. (1980). The psychology of leisure and recreation. Dubuque IA: Wm. C. Brown. Iso-Ahola, S.E. (1982). Toward a social psychological theory of tourism motivation: a rejoinder. Annals of Tourism Research 9, 256-262. Jahrami, H., Marnoch, G., Gray, A.M. (2009). Use of card sort methodology in the testing of a clinical leadership competencies model. Health Services Management Resources. 22 (4),176-183. Jewell, C. N. (2008). Factors influencing tipping behavior in a restaurant. Psi Chi Journal of Undergraduate Research 13(1), 38-48. Kim, J.Y., Natter, M., & Spann, M. (2009). Pay what you want: A new participative pricing mechanism. Journal of Marketing 73, 44-58. Kim, S.S. & Lee, C.K. (2001). Push and pull relationships. Annals of Tourism Research 29 (1), 257-260. Kim, S.S. & Prideaux, B. (2005). Marketing Implications arising from a comparative study of international pleasure tourists? motivations, and other travel-related characteristics of victors in Korea. Tourism Management 26, 347-357. 121 Kim, Y.H. Goh, B.K., & Yuan, J. (2010). Development of a multi-dimensional scale for measuring food tourists motivations. Journal of Quality Assurance in Hospitality & Tourism 11(1), 56-71. Klenosky, D. B. (2002). The ?pull? of tourism destinations: a means-end investigation. Journal of Travel Research 40, 385-395. Koku, P.S. (2005). Is there a difference in tipping in restaurant versus non-restaurant service encounters, and do ethnicity and gender matter? The Journal of Services Marketing 19 (6/7), 445-452. Kotler, P., Bowen, J.T., & Makens, J.C. (2010). Marketing for hospitality and tourism (5th Edition). Upper Saddle River, NJ: Prentice Hall. Kvale, S. & Brinkmann, S. (2009). Interviews, learning the craft of qualitative research interviewing. (2nd Edition) Sage Publications. Kwortnik, R.J., Lynn, W.M., & Ross, W.T. (2005). Does a voluntary tipping policy really improve service? An examination of consumer agency in the cruise industry. Retrieved from ssrn:http://ssrn/abstract=779505. Laffie, L.S. (2000). Got tips? Better report them. Journal of Accountancy 75-78. Lin, T.C. (2007). Economic behavior of restaurant tipping. Economic Bulletin 4(2), 1-10. Lincoln, Y.S. & Guba, E.G. (1985). Naturalistic Inquiry. Beverly Hills, CA: Sage. Lynn, M. (1997). Tipping customs and status seeking: a cross-country study. International Journal of Hospitality Management 16(2), 221-224. Lynn, M. (2000). National personality and tipping customs. Personality and Individual Differences 28(2), 395-404. Lynn, M. (2001). Restaurant tipping and service quality: A tenuous relationship. Cornell Hotel 122 and Restaurant Administration Quarterly 42(14), 14-20. Lynn, M. (2003). Tip levels and service: An update, extension, and reconciliation. Cornell Hotel and Restaurant Administration Quarterly 44(5/6), 139-148. Lynn, M. (2006). Geodemographic differences in knowledge about the restaurant tipping norm. Journal of Applied Social Psychology 36(3), 740-750. Lynn, M. (2006). Tipping in restaurants and around the globe: An interdisciplinary review. Handbook of Contemporary Behavioral Economics: Foundations and Developments, Morris Altman, ed., M.E. Sharpe. Lynn, M. (2009). Determinants and consequences of female attractiveness and sexiness: realistic tests with restaurant waitresses. Arch Sex Behavior 38, 737-745. Lynn, M., & Grassman, A. (1990). Restaurant tipping: An examination of three rational explanations. Journal of Economic Psychology 11, 169-181. Lynn, M., & McCall, M. (2000). Gratitude and gratuity: a meta-analysis of research on the service-tipping relationship. Journal of Socio-Economics 29, 203-214. Lynn, M., & Mynier, K. (2006). Effect of server posture on restaurant tipping. Journal of Applied Social Psychology 23(8), 678-685. Lynn, M., Zinkhan, G.M., & Harris, J. (1993). Consumer tipping: A cross-country study. Journal of Consumer Research 20, 478-488. Major, B. (1981). Gender patterns in touching behavior. In C. Mayo & M.N. Henley (Eds.), Gender and Nonverbal Behavior (pp. 15-37). New York: Springer-Verlag. Maslow, A. H. (1945). A theory of human motivation. Psychological Review 50(4), 370-396. Mattila, A.S. (2006). The power of explanations in mitigating the ill-effects of service failures. Journal of Services Marketing 20(7), 422-428. 123 Maxwell, S., Delany, H. (2004). Designing experiments and analyzing data: A model comparison perspective (2nd ed.). Mahwah, NJ: Lawrence Erlbaum Associates. McCall, M., & Belmont, H.J. (1996). Credit card insignia and restaurant tipping: Evidence for an associative link. Journal of Applied Psychology 81(5), 609-613. McCarty, J.A., Shrum, L.J., Conrad-Katz, T.E., Kanne, Z. (1990). Tipping as a consumer behavior: a qualitative investigation. In: Goldberg, M. (Ed.), Advances in Consumer Research, Association for Consumer Research 17, 723?728. McClelland, D.C. (1965). Toward a theory of motive acquisition. American Psychologist 20(5), 321-333. McMullan, R. (2005). A multiple-item scale for measuring customer loyalty development. The Journal of Services Marketing 19 (6/7), 470-481. Mertler, C. A. & Vannatta, R. A. (2005). Advanced and multivariate statistical methods. Practical application and interpretation (3rd Ed). Glendale, CA: Pyrczak Publicaiton. Meyers, L., Gamst, G., & Guarino, A. (2006). Applied multivariate research: Design and interpretation. Thousand Oaks, CA.: Sage Publications. Miles, M., & Hubberman, M. (1994). Qualitative Data Analysis: An Expanded Sourcebook (2nd Edition) US: Sage Publications. Murray, E.J. (1964). Motivation and Emotion. Englewood Cliffs, NJ: Prentice Hall. Needleman, R.L. (1937). Tipping as a factor in wages. Monthly Labor Review 45, 1302-1322. Newman, I., & Shannon, D. (2000). A conceptualization of mixed methods: A need for an inductive/deductive approach to conducting research. Journal of Research in Education, 10(1), 81-84. Norm. (n.d.) In Wikipedia online. Retrieved January 31, 2011 124 http://en.wikipedia.org/wiki/Norm_(sociology). Ogbonna, E., & Harris, L.C. (2002). Institutionalization of tipping as a source of managerial control. British Journal of Industrial Relations 40(4), 725-752. O?Neill, M.A.(2000). The measurement of service quality in the hospitality industry, In J. Kandampully., B. Sparks and C. Moks (Eds.) Service Quality and Customer Satisfaction in the Hospitality Industry (159-186). New York, John Wiley. Palmer, A. (2004). Introduction to Marketing: Theory and Practice. Oxford University; New York. Pan, W. (2011). Definition of Motivaion. Retrieved online January 19, 2011 from http://ezinearticles.com/?Definition-of-Motivation&id=1567108 Pandit, N.R. (1996). The creation theory: a recent application of the grounded theory method. The Qualitative Report 2(4), Article retrieved November 28, 2009 from http://www.nova.edu/ssss/QR/QR2-4/pandit.html Parret, M. (2006). An analysis of the determinants of tipping behavior: A laboratory experiment and evidence from restaurant tipping. Southern Economic Journal, (73, 2), 489-514. Pearce, P.L. & Lee, U. (2005). Developing the travel career approach to tourist motivation. Journal of Travel Research 43(3), 226-237. Pennington-Gray, L.A., & Kerstetter, D.L. (2001). ?What do university educated women want from their travel experiences? Journal of Travel Research 40, 49-56. Princeton University. (2000). About wordnet; retrieved January 31, 2011. http://www.wordnet.princeton.edu Rotter, J. B. (1954). Social Learning and Clinical Psychology. Prentice- 125 Hall: Englewood Cliffs, NJ. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. (monographs). Psychological Monographs, 80(1, Whole No. 699). Saunders, S.G. & Lynn, M. (2010). Why tip? An empirical test of motivations for tipping car guards. Journal of Economic Psychology 31, 106-113. Schwandt, T. (2007). Dictionary of qualitative inquiry (3rd ed.). Thousand Oaks, CA.: Sage Publications. Segrave, K. (1998). Tipping: An American Social History of Gratuities, McFarland & Company, Jefferson, NC. Service. (n.d.) The Business Dictionary.com Retrieved January 31, 2011 from http://www.businessdictionary.com/definition/services.html. Shaw, F. G. (1950). Clinical psychology and behavior theory. Journal of Abnormal and Social Psychology 45(2), 388-391. Sherman, S. J., & Fazio, R. H. (1983). Parallels between attitudes and traits as predictors of behavior. Journal of Personality 51, 308-345. Sheth, J., Mittal, B., & Newman, B. (1999). Customer behavior: Consumer behavior & beyond. (Fort Worth, TX):Dryden Press. Snepenger, D., King, J., Marshall, & Muzaffer, U. (2006). Modeling Iso-Ahola?s motivation theory in the tourism context. Journal of Travel Research 45, 140-149. Snyder, M. L. (1976). The inverse relationship between restaurant party size and tip percentage: Diffusion of responsibility or equity? Personality and Social Psychology Bulletin 2, 308. Soloman, M.R. & Stuart, E.W. (2003). Marketing: Real people, real choices (3rd Edition). Upper 126 Saddle River, NJ: Prentice Hall. Speer, T, L. (1997). The give and take of tipping. American Demographics 19(2), 50-54. Stephen, R. & Zweigenhaft, R.L. (1986). The effect on tipping of a waitress touching male and female customers. The Journal of Social Psychology 126(1), 141-142. Stevens, J., (2002). Applied multivariate statistics for the social sciences (4th Ed.). Hillsdale, N.J.: Lawrence Erlbaum Publishing. Tabachnick, B., & Fidel, l. (2001). Using multivariate statistics (4th Ed.).Needham Heights, MA: Allyn & Bacon. Tabachnick, B., & Fidel, l. (2007). Using multivariate statistics (5th Ed.).Boston, MA: Pearson Education. Tanggaard, L. (2009). The research interview as a dialogical context for the production of social life and personal narratives. Qualitative Inquiry, 15(9), 1498-1515. Thurstone, L. (1947). Multiple factor analysis. Chicago, Il., Chicago University Press. Tip. (n.d.). In Merriam Webster?s Dictionary Online. Retrieved January 30, 2011 http://www.merriam-webster.com/dictionary/tip?show=0&t=1296421873. Uysal, M., Gahan, L., & Martin, B. (1993). An examination of event motivations: a case study. Festival Management and Events Tourism 1, 15-29. Videbeck, S. (2004-05). The economics and etiquette of tipping. Policy, 20(4), 38-41. Watts, M. (2007). They have tied me to a stake. Qualitative Inquiry, 13(2), 204-217. Wicker, A. W. (1969). Attitudes versus actions: The relationship of verbal and overt behavioral responses to attitude objects. Journal of Social Issues 25, 41-78. Witz, K. (2006). The participant as ally and essentialist portraiture. Qualitative Inquiry, 12(2), 246-268. 127 Yuan, S., & McDonald, C. (1990). ?Motivational Determinants of International Pleasure Time.? Journal of Travel Research 29, 42-44. 128 Appendix A: Pilot 1-Tipping Motivations Scale What?s in a Tip? We are interested in your tipping behavior and more importantly why you tip the way you do. We?d be grateful therefore if you would take a few minutes and complete the following short survey on tipping motivation. Simply check the appropriate answer in all cases. Thank you! Are you ? Male O Female O What is your ethnicity? Caucasian O African American O Hispanic O Asian O Other O Are you a ? Freshman O Sophomore O Junior O Senior O What is your major area of study? ____________________________ Are you currently employed in the hospitality industry? Yes O No O If you answered yes, in what field are you employed? Food production O Restaurant service O Bar service O Hosting O Banquets O Housekeeping O Concierge O Front desk O Reservations O Maintenance O Other O ____________________ Not applicable O Are you ? < 21 years old O 21 + O How many years of employment experience do you have in the hospitality industry? < 1 year O 1 ? 2 years O 2 ? 3 years O 3 ? 4 years O 3 ? 5 years O 5 years + O When dining out, do you tip? Yes O No O Do you work ? Part time O Full time O Do you hold a tipped job? Yes O No O What is your tipping norm 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O for good service? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O for excellent service? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O Please tell us how strongly you disagree (1) or Agree (5) with each of the following statements (please circle): Leaving a good tip should ensure great service in the future 1 2 3 4 5 I always consider future visitation when tipping 1 2 3 4 5 My tipping behavior is directly related to the service received 1 2 3 4 5 129 How strongly you agree or disagree with the following? SD ? Strongly Disagree / D ? Disagree / N- Neither / A ? Agree / SA ? Strongly Agree SD D N A SA Please circle As a rule, I tip a percentage of the total bill 1 2 3 4 5 Employee appearance/hygiene influences my tipping behavior 1 2 3 4 5 The promptness of a server?s greeting influences my tipping behavior 1 2 3 4 5 I tend to tip above my norm when service is excellent 1 2 3 4 5 On occasion, I tip to impress 1 2 3 4 5 A server?s body language influences my tipping 1 2 3 4 5 Direct eye contact with a server influences my tipping behavior 1 2 3 4 5 I feel generous when I leave a tip 1 2 3 4 5 Sometimes I feel pressured to tip 1 2 3 4 5 Unsatisfactory service influences my tipping behavior 1 2 3 4 5 A server?s attitude influences my tipping behavior 1 2 3 4 5 A server?s confidence at table side affects my tipping behavior 1 2 3 4 5 I feel more obligated to tip when dining with friends and/or family 1 2 3 4 5 I leave a larger tip when others I have dined with do not tip 1 2 3 4 5 A server squatting at table side affects the way I tip 1 2 3 4 5 I feel obligated to tip even when service is bad 1 2 3 4 5 I leave a larger tip on special occasions (birthdays etc) 1 2 3 4 5 A server?s menu knowledge influences the way I tip 1 2 3 4 5 A server?s ability to be caring and patient affects the way I tip 1 2 3 4 5 I tend to tip a dollar amount when the total check is large 1 2 3 4 5 If a server is caring and empathetic this influences the amount I tip 1 2 3 4 5 I feel regret if I do not leave a tip 1 2 3 4 5 A server?s stance at tableside influences the way I tip 1 2 3 4 5 I feel embarrassed when others in my party do not tip 1 2 3 4 5 The number of times a server visits my table influences my tipping 1 2 3 4 5 I tip below my norm when service is bad 1 2 3 4 5 A company?s service recovery efforts influences my tipping behavior 1 2 3 4 5 A server making physical contacts affects my tipping behavior 1 2 3 4 5 Overall restaurant cleanliness affects my tipping behavior 1 2 3 4 5 My tipping behavior changes in times of economic hardship 1 2 3 4 5 130 Appendix B: Pilot 2-Tipping Motivations Scale What?s in a Tip? We are interested in your tipping behavior and more importantly why you tip the way you do. We?d be grateful therefore if you would take a few minutes and complete the following short survey on tipping motivation. Simply check (?) the appropriate answer in all cases. Thank you! Are you ? Male O Female O What is your ethnicity? Caucasian O African American O Hispanic O Asian O Other O How old are you? 19 ? 24 O 25 ? 39 O 40 ? 59 O 60 + O Where do you live? City/State? __________________________________ Are you currently ? Employed O Un-employed O Student O Retired O Are you employed in the hospitality industry? Yes O No O How many years of employment experience do you have in the hospitality industry? < 1 year O 1 ? 2 years O 2 ? 3 years O 3 ? 4 years O 3 ? 5 years O 5 years + O N/A O Do you hold a tipped job? Yes O No O When dining out, do you tip? Yes O No O Does your job cause you to tip more ? Yes O No O What is your average annual household income ? < $15,000 O $16 ? $39,000 O $40 ? $69,000 O $70 - $99,000 O $100 - $129,000 O $130,000 + O What is your tipping norm? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O for good service 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O for excellent service? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O Please tell us how strongly you disagree (1) or Agree (5) with each of the following statements (please circle) When I tip for good service I tend to: Tip above my norm 1 2 3 4 5 Ask for the same server on future visits 1 2 3 4 5 Recommend the server to friends and family 1 2 3 4 5 Spend more money on future visits 1 2 3 4 5 Tip more money on future visits 1 2 3 4 5 Visit a restaurant more often 1 2 3 4 5 131 How strongly you agree or disagree with the following? SD ? Strongly Disagree / D ? Disagree / N- Neither / A ? Agree / SA ? Strongly Agree SD D N A SA Please circle 1. A servers gender (male/female) influences my tipping behavior 1 2 3 4 5 2. A servers appearance influences my service behavior 1 2 3 4 5 3. Timeliness of service influences my tipping behavior 1 2 3 4 5 4. The service received influences my tipping behavior 1 2 3 4 5 5. On occasion, I tip to impress 1 2 3 4 5 6. A server?s body language influences my tipping behavior 1 2 3 4 5 7. Direct eye contact with a server influences my tipping behavior 1 2 3 4 5 8. The quality of a restaurants food influences my tipping behavior 1 2 3 4 5 9. Sometimes I feel pressured to tip 1 2 3 4 5 10. Poor service influences my tipping behavior 1 2 3 4 5 11. A server?s attitude influences my tipping behavior 1 2 3 4 5 12. A server?s menu knowledge affects my tipping behavior 1 2 3 4 5 13. I feel more obligated to tip when dining with friends and/or family 1 2 3 4 5 14. I leave a larger tip when others I have dined with do not tip 1 2 3 4 5 15. A server?s ability to sell the menu influences my tipping behavior 1 2 3 4 5 16. I feel obligated to tip even when service is bad 1 2 3 4 5 17. The general expense of a restaurant influences my tipping behavior 1 2 3 4 5 18. I tend to tip more when I can pay by credit card 1 2 3 4 5 19. A server?s ability to be caring and patient affects the way I tip 1 2 3 4 5 20. Where I am seated tends to influence my tipping behavior 1 2 3 4 5 21. If a server is caring and empathetic this influences the amount I tip 1 2 3 4 5 22. I feel regret if I do not leave a tip 1 2 3 4 5 23. I always consider future service when I tip 1 2 3 4 5 24. I feel embarrassed when others in my party do not tip 1 2 3 4 5 25. The number of times a server visits my table influences my tipping 1 2 3 4 5 26. The weather influences my tipping behavior 1 2 3 4 5 27. How my service complaints are handled influences my tipping behavior 1 2 3 4 5 28. A server making physical contact with me affects my tipping behavior 1 2 3 4 5 29. Overall restaurant cleanliness affects my tipping behavior 1 2 3 4 5 30. I always consider future visitation when I tip 1 2 3 4 5 132 Appendix C: Final Study-Tipping Motivations Scale What?s in a Tip? We are interested in your tipping behavior and more importantly why you tip the way you do. We?d be grateful therefore if you would take a few minutes and complete the following short survey on tipping motivation. Simply check (?) the appropriate answer in all cases. Thank you! Are you ? Male O Female O What is your ethnicity? Caucasian O African American O Hispanic O Asian O Other O Are you currently ? Employed O Un-employed O Student O Retired O Where do you live? City/State? _______________________________ ___ Are you employed in the hospitality industry? Yes O No O How old are you? 19 ? 24 O 25 ? 39 O 40 ? 59 O 60 + O What is your average annual household income ? < $15,000 O $16 ? $39,000 O $40 ? $69,000 O $70 - $99,000 O $100 - $129,000 O $130,000 + O How many years of employment experience do you have in the hospitality industry? < 1 year O 1 ? 2 years O 2 ? 3 years O 3 ? 4 years O 3 ? 5 years O 5 years + O N/A O When dining out, do you tip? Yes O No O Does your job cause you to tip more? Yes O No O Do you hold a tipped job? Yes O No O What is your tipping norm? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O for good service? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O for excellent service? 1 ? 5% O 6 ? 10% O 11 ? 15% O 16 ? 20% O 20% + O Please tell us how strongly you disagree (1) or Agree (5) with each of the following statements (please circle) When I tip for good service I tend to: Tip above my norm 1 2 3 4 5 Ask for the same server on future visits 1 2 3 4 5 Recommend the server to friends and family 1 2 3 4 5 Spend more money on future visits 1 2 3 4 5 Tip more money on future visits 1 2 3 4 5 Visit a restaurant more often 1 2 3 4 5 133 How strongly you agree or disagree with the following? SD ? Strongly Disagree / D ? Disagree / N- Neither / A ? Agree / SA ? Strongly Agree SD D N A SA Please circle 31. A servers gender (male/female) influences my tipping behavior 1 2 3 4 5 32. Timeliness of service influences my tipping behavior 1 2 3 4 5 33. The service received influences my tipping behavior 1 2 3 4 5 34. A server?s body language influences my tipping behavior 1 2 3 4 5 35. Direct eye contact with a server influences my tipping behavior 1 2 3 4 5 36. Poor service influences my tipping behavior 1 2 3 4 5 37. A server?s attitude influences my tipping behavior 1 2 3 4 5 38. A server?s menu knowledge affects my tipping behavior 1 2 3 4 5 39. I feel more obligated to tip when dining with friends and/or family 1 2 3 4 5 40. I leave a larger tip when others I have dined with do not tip 1 2 3 4 5 41. A server?s ability to sell the menu influences my tipping behavior 1 2 3 4 5 42. I feel obligated to tip even when service is bad 1 2 3 4 5 43. The general expense of a restaurant influences my tipping behavior 1 2 3 4 5 44. Where I am seated tends to influence my tipping behavior 1 2 3 4 5 45. I feel regret if I do not leave a tip 1 2 3 4 5 46. I always consider future service when I tip 1 2 3 4 5 47. I feel embarrassed when others in my party do not tip 1 2 3 4 5 48. Overall restaurant cleanliness affects my tipping behavior 1 2 3 4 5 49. I always consider future visitation when I tip 1 2 3 4 5 Additional Comments: 134 Appendix D: Tipping Motivations Scree Plot