«maifiafififiwfi .,. mfifi$mflu a ll :1. 3 3%; > 2‘ f5 r ‘1 2 £124: #2? I t 1-. ‘I 7.2:: .. Lilia : ‘h« :5 1...... .13. ‘4 Sit! xd. " 13$: -‘ 5;” E? . .. :«nfdflyi . .2; :32. . z 2‘ 115.131 4.. vs... x25: . Qiut 4i.‘ hi is?! on}? té 3’1, z-i“. . indufimwu; V" :4 ‘ .~ ... is» 1.43 h. I‘ h. b. I. is} . 11.42.38 (5 .5 .‘I .2; .11: :..i...\& )..\«~ . . ‘3? 6%.. iii... 9.. .‘ill‘pn‘z z) ‘I 4.. 21". 1. 1|}! t3}.61v..‘ 1.: I315 iii-Tit V , 11‘. \‘V . .25.} A. . .‘stz 32:3. «.5411; «#35431... . . «91“! a .1: Ix Ottawa‘. £12.39. In}! tiriviil 1.“.‘3583113. O... 1‘;«Iv:v».t.llll.5||. bin? :5. t .3; t nit it ‘II .. : .ivla}li.r.‘v.l1{ 1 DJ n I.M!..\. L...-.l!t. ).X|n\w\l.«.vz . 14.3.: .? iv , 51... Lu..n.fi.1 MICHIGAN STATE II I IIIIIIII LIBRARY i‘fiichigan 3 ate University I w 0311390 4515 This is to certify that the dissertation entitled Measuring The Effectiveness of Brochures in Tourism Destination Marketing presented by Zongqinq Zhou has been accepted towards fulfillment of the requirements for Ph. D . degree in Wation and Date Februarv 2. Tourism’Resources QW/ZAJM / M afin professor 1996 MS U is an Affirmative Action/Equal? Opportunity Institution 0-12771 IIIIIIII IIIIIIIIIIIIII PLACE IN RETURN BOX to remove this checkout from your rooord. TO AVOID FINES return on or botoro doto duo. DATE DUE DATE DUE DATE DUE l’ MSU lo An Affirmative ActioNEquol Opportunity Instltwon W poo-9.1 MEASURING THE EFFECTIVENESS OF BROCHURES IN TOURISM DESTINATION MARKETING BY Zongqinq Zhou A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Park, Recreation and Tourism.Resources 1996 ABSTRACT MEASURING THE EFFECTIVENESS OF BROCHURES IN TOURISM DESTINATION'MARKETING BY Zongqing Zhou Much of the travel and tourism industry relies on brochures to market its products and destinations. Research studies on improving the effectiveness, the design and the use of brochures are, thus, badly needed. A review of the existing literature reveals that effectiveness of brochures has not received enough research attention and the instruments employed in these studies do not adequately address the unique features of brochures and measure their full benefits. Accordingly, the primary objectives of this study included: 1) to define a set of criteria for adequately measuring the effectiveness of brochures, 2) to apply these criteria in a case study involving Frankenmuth, Michigan to examine how effective brochures are and to identify what factors influence their effectiveness when used in marketing tourism destinations, 3) to identify how lapsed time influences reported trip expenditures, and 4) to examine whether enclosing the previously requested brochure in a brochure study has any effect on respondents' reported impressions and effectiveness of the brochure. Findings reveal that nearly all respondents read the brochure they received. Consultation of the brochure on site was found to be correlated with reported better trip experiences. About 70% of the respondents had not made their final decisions to visit Frankenmuth at the time of requesting information. It was also found that the brochure had different influences on first time and repeat visitors. Findings suggest that respondents' recall of their trip expenditures may not be as sensitive to lapsed time as commonly assumed. The memory cue, an enclosed brochure identical to that previously requested, however, resulted in respondents' higher ratings of the impressions and effectiveness of the brochure. The findings from this study have several important implications for destination marketers and researchers. First, the almost perfect readership of the brochure suggests that inquiry programs should be taken more seriously than they generally are since inquirers as a group are without equal in their receptivity for a destination marketer's message. Second, on-site consultation of the brochure suggests the need to design brochures that not only convey promotional messages but also provide information such as maps, business hours, and other information one would expect to find in a travel guidebook. Third, since brochures have different influences on first-time and repeat visitors, it would be useful for destinations to design separate brochures to target the interests of these two market segments. Fourth, since 70% of inquirers had not decided on their destinations at the time they requested information, destination marketers are presented with an opportunity to have an impact if their marketing strategies targeting inquirers are effective. Finally, this study suggests that brochures serve a multitude of purposes in fulfilling different information needs of inquirers and therefore their impacts should be assessed in ways that go beyond simply deriving a conversion rate so that their full benefits to consumers can be truly revealed. Copyright by ZONGQING ZHOU 1996 ACKNOWLEDGEMENTS This dissertation could not have been completed without the tireless assistance of my major adviser, Professor Donald F. Holecek. His patience in reviewing various drafts of this dissertation is most appreciated. I would also like to extent my heartfelt thanks to Dr. Joseph Fridgen, Dr. Joanne Keith and Dr. Bonnie Knutson of Michigan State University for their valuable insight and helpful suggestions in the process of writing this dissertation. I am most deeply indebted to my wife Yuehui Li for her faithful support and to my dear daughter, Jenny Zhou and beloved son, William Zhou for their smiles and encouragement during the writing of this dissertation. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES CHAPTER ONE INTRODUCTION 1.1 The Importance of Brochures in Tourism Marketing 1.2 Problem Statement and Objectives CHAPTER TWO LITERATURE REVIEW NNNNN U1|l>LAJN .1 Theories of Mass Media Effects and Advertising Theories of Mass Media Effects Models of Advertising . . Measurement of Advertising Effectiveness Evaluation of Tourism Destination Marketing 2.5.1 Economic Impact Studies Conversion Studies . . Review of Brochure Studies The Uniqueness of Brochures rResearch Hypotheses NNN ubUJN .5. .5. .5. ajo CHAPTER THREE METHODS 3. wwwwwww mflmU'er-UNH Case Study Setting Definition of Effectiveness Questionnaire Development Major Variables Studied Research Design Sampling Frame Sampling Procedure Data Collection CHAPTER FOUR RESULTS 4. 4. 4. .p N Basic Findings . Brochure Use and Trip Satisfaction Comparison of 1992- 93 and 1993- 94 Reported Trip Spending . Comparison of Unaided and Aided Recall Results The Role of Prior Experience vi viii 56 56 58 59 6O 61 62 63 64 66 66 77 80 86 88 TABLE OF CONTENTS (Cont'd) 4.6 A Linear Multiple Regression Model of Effectiveness . . . . . . . . 4.7 Summary: Answers to Research Questions CHAPTER FIVE CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions . . . . 5.2 Marketing Recommendations 5.3 Research Recommendations 5.4 Limitations . . . . . . 5.4.1 Validity of the Study 5.4.2 Measurement 5.4.3 Response Bias 5.5 Future Research LIST OF REFERENCES APPENDIX A 1994 SURVEY OF FRANKENMUTH INQUIRERS APPENDIX B COVER LETTER FOR AIDED QUESTIONNAIRES vii 91 102 106 106 108 111 112 112 114 115 116 120 127 134 Table Table Table Table Table Table Table Table Table Table Table .10 .11 LIST OF TABLES Demographic and Socioeconomic Characteristics of Respondents Other Characteristics of Respondents Distribution of Dates When Trips Were Taken by 1992 and 1994 Respondents Effects and Impressions of the Brochure Chi-Square Results of the Relationship between Consultation of the Brochure on Site and Recommendation to Someone Else after Trip Correlation between Consulting Brochures on Site and Level of Satisfaction T-test for Reported Spending Means for the 1992-93 and 1993-94 Samples. Correlations between Dates of Departure for Frankenmuth and Reported Total Trip Expenditures Two-Way Analysis of Variance for Trip Expenditures. Factors: Year and Aided/Unaided Number of Nights Spent in Frankenmuth by Residence and by Sampling Year A Summary of the T-test Results for Unaided and Aided Mean Ratings of Impressions and Effects of the Brochure... viii (in Dollars) 68 73 74 76 78 79 81 82 83 86 87 Table Table Table Table .12 .13 .14 .15 LIST OF TABLES (Cont'd) T-test Results of the Influence of Brochures on One's Trip Decision by Prior Experience. Correlation between Influence of Brochures and Prior Experience Correlation Matrix for the Multiple Regression Model Regression Results for Effectiveness Model ix 89 90 97 98 Figure Figure Figure Figure Figure Figure Figure Figure LIST OF FIGURES Richardson-Haley Model of the Communication Process . . . . . . . . . . 14 The FCB Grid . . . . . . . . . . . . . . 23 Model of Consumer Involvement .. . . . . 25 The Conversion Model . . . . . . . . . . 39 Overall Research Design ... . . . . . . . . 62 Number of Responses by Year and Type of Questionnaire Sent . . . . . . 67 Among Those Not Visiting Frankenmuth Percentage of Respondents Who Were/Were Not Considering a Trip in the Near Future . .. . . . . . . . .70 Status of Decision of All Respondents to Visit Frankenmuth When Inquiry Was Made . . . . . 72 CHAPTER ONE INTRODUCTION 1.1 The Importance of Brochures in Tourism.Marketing Tourism is an important source of export earnings for most of the countries in the world. Its economic importance has been well-documented. For 1989, based on information from the World Tourism Organization, worldwide international tourism arrivals were estimated to be over 400 million, with total receipts of more than $200 billion (excluding expenditures on transportation). In 1989, worldwide expenditures for domestic and international tourism taken together were estimated at more than a trillion dollars (Edgell, 1990). Within the United States, it is the leading employer in fifteen states and is one of the top three employers in 39 of the 50 states (Burke and Gitelson, 1990). The importance of tourism in national and regional development has caused increasing competition in efforts to attract and lure tourists. According to the U.S. Travel Data Center (U.S. Travel Data Center, 1989), every state in the United States maintains an agency or division to promote inbound travel. The budget for these offices ranges from 1 2 $1.2 million in Nebraska to over $20 million in Illinois and New York. Across the states, an average amount of $2.3 million were allocated to state promotional efforts (U.S. Travel Data Center, 1989). The bulk of this money is, in turn, allocated to advertising efforts, with one of the primary objectives being the generation of requests for information packages (e.g. brochures) that have been prepared by the state. Tourism destination marketing has become a major player in this competition process. Government tourism organizations, tourism destination marketers, tourism operators and agents are constantly under pressure to seek more efficient ways and tools to evaluate as well as to market their products and services. Brochures are important tools that have been used by the tourism industry to market tourism products and services, including destinations. With the tourism marketplace becoming more and more competitive and with large amounts of marketing expenditures being spent on promoting destinations through the use of brochures (Hodgson, 1993), the need for accurate evaluation of the effectiveness of brochures has never been greater. 3 1.2 Problem Statement and Objectives Brochures are a unique medium of communication and marketing, not only in terms of their content, format, and design but also in terms of their varied means of distribution. There are many means for distributing brochures, such as mailing them to people who request them, handing them out on site or displaying them in the lobbies of hotels, in convention centers, in rest areas, in travel agency offices and in tourist information centers. Consequently, studies related to the effectiveness of brochures vary greatly in terms of focus and emphasis. In this paper, only the effectiveness of brochures requested by and mailed to potential tourists will be examined. Specifically, this paper is only concerned with the effectiveness of requested brochures when used in marketing tourism destinations. Literature on studies that are related to the general topic of effectiveness of brochures will also be reviewed to provide a broader perspective. There have been a few studies published on the effectiveness of brochures, but the reported results of these studies have been mixed. While some researchers suggest that brochures influence visitor behavior (Lime and 4 Lucas, 1977; Krumpe and Brown, 1982; Roggenbuck and Berrier, 1982; Cherem, 1982 and Fesenmaier and Vogt, 1993), other researchers indicate that brochures are either not influential in impacting travel decision-making (Baas et al., 1989) and travel behavior (Fesenmaier et al., 1993) or are not read by most people who receive them (Fazio, 1979; Cella and Keay, 1979; Middleton, 1988). There are three major issues concerning studies of the effectiveness of brochures. The first involves the question of defining the word effectiveness; that is, what are the criteria used in measuring the effectiveness of brochures. The second is concerned with the effect of lapsed time on reported data, especially in the area of trip expenditures. The third is concerned with the relationship between effectiveness of brochures and the characteristics of people who request them to include: their motivation, prior trip experience and demographic and socioeconomic factors. Change in immediate sales has been the most widely used measure of the effectiveness of brochures. Evaluation of the effectiveness of brochures is often part of a basic conversion study whose primary purpose is to estimate the so-called "conversion rate", which is the ratio between 5 people who call for information and people who actually visit the destination after requesting information. It, however, does not adequately assess the full effects of brochures. Additional complications of conversion studies as well as other brochure effectiveness studies will be further discussed in the literature review. Questions about the effect of lapsed time on reported data have been mainly focused on reporting trip expenditures. Rylander II et a1. (1993) provide an excellent review of these studies in their study of nonresponse and recall biases in a survey of traveler spending. They concluded that respondents' perceived accuracy on reported trip expenditures was related to time of completing the mailback questionnaire as well as trip complexity. However, another dimension of the effect of time on reported expenditure data was not examined in the study. This is the question of whether or not the length of lapsed time between when spending occurs and when spending data are recalled results in reporting less or more trip expenditures. Issues of memory and recall will be further addressed in the literature review. 6 Other memory and recall problems involve the effects of cues, e.g. enclosing the previously requested brochure with the questionnaire in a conversion study to assess impressions and perceived usefulness of brochures. These issues have not been addressed in the literature to date. The issue of the relationship between effectiveness of brochures and the characteristics of brochure-inquirers has not received adequate attention in most brochure studies. Fesenmaier et al. (1993) indicates that there is a need to better understand the reasons why brochures are effective or not effective. If these underlying reasons can be identified, tourism destination marketers would be able to design strategies to maximize effectiveness of brochures. The objectives of this paper were: 1) to define a set of criteria for measuring the effectiveness of brochures based on theories of mass media effects and advertising effectiveness models, 2) to apply these criteria in a case study to examine how effective brochures are and to identify what factors influence their effectiveness when used in marketing tourism destinations, 3) to identify how lapsed time (between when trips are made and when respondents are asked to report) influences reported trip expenditures, and 7 4) to examine whether enclosing the previously requested brochure with an evaluative questionnaire in a brochure study has any effect on respondents' reported impressions and effectiveness rating of the brochure. Theories of mass media effects and advertising as well as studies related to effectiveness of brochures in tourism destination marketing were reviewed to establish a theoretical basis for the hypotheses developed to guide this study. CHAPTER TWO LITERATURE REVIEW 2.1 Theories of Mass Media Effects and Advertising Tourism destination marketing comprises all communication measures designed to create awareness of, interest in and a favorable image for existing or new tourism destinations with the aim of attracting customers to them (Schmoll, 1977). It is impossible to explore the effects of brochures without examining general advertising concepts, since a brochure is only one of a set of tools (media) used in advertising and marketing. It is equally inconceivable to talk about advertising effectiveness without discussing theories of media effects. In fact, theories of media effects constitute a major part of communications theory. Further, communications theory and advertising theory have a close relationship, yet this is not commonly recognized. McDonald (1993) claimed: "It seems strange that there has been so little academic attempt to apply communication theory to advertising, ..." (p. 133). He went further to note that: 9 The strength of advertising is that its development has been driven by its practitioners over the years. I am suggesting now that this strength has become a point of weakness, and there is a need for new academic input into the subject (p. 134). In this chapter, models of mass media effects will be examined first, followed by a review of models of advertising effectiveness. A review of advertising's impact on consumers at the individual level and how direct and indirect effects are handled will also be provided. Finally, attempts will be made to relate these theories and models to tourism destination marketing and brochure research studies. 2.2 Theories of Mass Media Effects The issue of media effects has a lengthy history (see Jeffres, 1986, for a detailed discussion). Starting with the early top-to—bottom powerful or so-called hypodermic effects model popular in the 19308 and 19403, to the bottom- to-top limited effects or reinforcement effects model which came on the scene in the late 19505 and 19605 (Klapper, 1960), and to a return to the powerful effects model in the 19703 (Gerbner, 1972), most researchers have concentrated 10 their attention on the relation between mass media and audience. The hypodermic effects model suggests that the media were directly responsible for changing peoples's ideas, attitudes, and behaviors-~shooting beliefs into peoples's minds almost as a doctor inoculates patients with a hypodermic needle. This model assumes that audience members are all pretty much alike, responding to media in predictable, similar ways. The hypodermic model presents mass media as being very powerful. According to the limited effects model, the mass media merely reinforce existing attitudes and behaviors, because audiences are "stubborn" and they use a number of defense strategies such as selective attention, selective perception, and selective recall to insulate them against contradictory messages. In the 19703, the limited effects model was replaced by a direct effects model similar to that of the late 19303 and 19403. It is argued that mass media, television in particular, affects or cultivates the norms and values of a society. In general, the question of mass media effects has grown more complex rather than the reverse. Most 11 researchers today would agree that the question of whether or not mass media have effects is too simple a question. Certainly the media are not all powerful, but neither does the evidence suggest that audiences have defensive filters that wash out all media effects. Today, mass communication researchers are more interested in how people decode messages; how motivations and other information needs relate to various media effects. This latter point of view will be adopted for this paper. 2.3 Models of Advertising Theories of the effects of advertising reflect theories of mass media effects. Accordingly, most of the early models of advertising assumed that advertising operates in a straight-forward "one-way" manner; i.e. the audience is a group of passive receivers who await the advertising message to hit them. In other words, advertisers choose what they want to deliver, encode it in the form of an advertisement, and deliver it through mass media to a basically passive or receptive audience. The audience attends, partially attends, or does not attend to the message and remembers all, part, or none of it. 12 One model of this advertising process has been referred to as AIDA which stands for Attention, Interest, Desire and Action (Haley, 1985). In this model, it is assumed that the first job of advertising is to call the attention of the audience to the fact that a certain product or service is available. The message is then supposed to build interest in the product or service and to encourage potential customers to seek out more information. Once these first two jobs are accomplished, the advertising is then supposed to create an active desire to try the product or service. Then, finally, the advertising is supposed to result in an increase in sales. In recent years, there has been a substantial cultural shift away from the "passive" consumer View (McDonald, 1993). It is argued that consumers are essentially active, making choices and selecting which advertising they will attend to or not. This view recognizes the importance of selective perception, accurate target setting and the emotional elements in consumer response. In other words, the communications process is viewed as a" two-way" process rather than "one-way". 13 Representing this limited effects or "two-way" communications process is the Richardson-Haley's sophisticated model of how advertising works (Haley, 1985). Figure 2.1 is a reproduction of the Richardson-Haley model. According to this model, the first step in the planning process is to undertake a through review of the product, brand, or service to be advertised and what it means to consumers. The second step is to develop a communication strategy. This involves a definition of the product category within which the brand is competing and whether incremental volume is expected to come from primary competitors, other products, new occasions of use or larger quantities of consumption per use occasion. The third step involves the development of the advertising stimulus, e.g. the core idea, the message type, support type, format, and media considerations. At this point, the stimulus is exposed to the consumer. However, the consumer brings to the exposure situation a number of things that can have a strong effect upon his/her reactions at the moment of exposure. These include perceived and latent needs and the benefits being sought, brand salience and perceptions, interest and involvement in the product category, and the 14 W) I Maysmgy)8timulus Leaning #1161an )In'm —>Purchase “my iflflm A / I / ~>meing>0hange I I I I I Chums: Nnmmm Rmem Lms lhmmuu: A I I I I V Rigflgfm‘thms‘(—Umfimr‘F—'h““$¥<(—U3 I (Product) Figure 2.1 Richardson-Haley Model of Communications Process (Source: Russell I. Haley, Dexelgping_E££§QLiMe Communication3_5trategy, 1985. p. 27) 15 amount of risk (physical, economic, or social) that are perceived to be involved in brand choice. Also affecting responses are the information processing styles and the purchasing and use styles of the individual. Information processing styles include the ways in which people use information to make decisions, the extent to which they are seeking information and the types of information they are seeking. The final factors influencing response are the consumer's personality and style characteristics as well as demographic characteristics. It is assumed that some form of message processing takes place, which will result in changes in peoples' attitudes about and predispositions toward the advertised product. Learning is also involved. The net effect can be positive or negative; it can result in rejection or positive reinforcement. Consumers are also likely to be exposed to multiple messages from the advertised brand and to messages from competing products. If they receive no further messages, the effects of the initial exposure can be expected to diminish or "decay". Multiple messages are normally aimed at reinforcement. 16 Before buying, other influences---notably in-store influences---come into play. These include distribution, display, price, packaging, promotion, word-of—mouth, influence of sales personnel, activity of other shoppers, and the ambience of the store itself. In tourist destination marketing, since destination marketers can not provide "in-store" influences such as displays or other Shopper's activities, brochures can be used to describe the physical aspects of the advertised destination and present quotes from past satisfied visitors 2.4 Measurement of Advertising Effectiveness The discussion so far has been only concerned with how advertising is supposed to work, i.e. how advertisers communicate with consumers. The problem is even more complicated when the question is asked: "How do you measure advertising effectiveness?" Most advertising, as McDonald (1993) points out "is intended to increase, or maintain sales in some form" (p. 53). Sales, therefore, has been one of the measures that advertisers use to measure effectiveness. Some advertising is designed to change or maintain an attitude or culture to provide a context in 17 which a desired behavior is more likely to take place, e.g. corporate advertising. These differences explain why sales measures can only apply to a limited number of products and services. Moreover, this leads to another important fact which is that there is a huge variety of objectives which advertising can have and, therefore, of responses it can seek from consumers. This is, of course, not to say that some general guidelines or unifying principles of advertising can not be drawn. To quote a metaphor by McDonald (1993): "St Bernards, greyhounds and Chihuahuas are very different, but they are still all dogs." (p. 110) From the consumer's point of view, one way of categorizing responses is to measure how close they are to direct action. These responses can be classified along a scale ranging from "direct" to "indirect". Using the categories developed by McDonald (1993), direct and indirect advertising strategies are discussed below. DIRECT 1). Iakuction. This advertising aims "to get people to pull out a pen, fill in the coupon, sign a check, address an envelope 18 and mail it, or call a number." The response sought is "I'll do that or buy that now." This approach is most suitable for infrequent, low involvement products for which there are not very clear-cut branded markets and where the task is to overcome inertia or indifference. Much advertising for local and annual tourist festivals is of this type. The desired response here is "That sounds interesting; I'll find out more about it." While an immediate response is sought, it alone is not assumed to lead to a product purchase decision. This strategy would be used for infrequently purchased goods, especially if they are complex or expensive, where the decision requires thought and the comparison of alternatives: for instance, durables, holidays, cars, insurance, savings. A good example in tourism is marketing destinations. Destinations can advertise through mass media in which a phone number is provided to call for further information. This type of advertising helps to get the brand/destination onto the list of brands/destinations for the consumer to consider. 3). 4). 19 W. This type of advertising aims to get people to make the link between the brand and their needs or desires. Although an immediate effect is aimed at, there is no immediate action. This advertising is particularly aimed at trial of a new brand, and it works largely by presenting "news". For example, Marriott International has segmented its lodging product into four brands: Marriott Hotels, Resorts and Suites, the company's full-service lodging division; Courtyard, the moderate- priced lodging product; Residence Inns, the extended- stay product; and Fairfield Inn, the economy lodging offering. Marriott International might run an advertisement to "announce" this new segmentation and the availability of these services. Consumers' action might take place on the next suitable occasion, when long-term memory of the advertising may be activated by reading a brochure or by word-of—mouth recommendation. E . E . i J] . . E The desired response here is "That reminds me." This is advertising for repeat purchase, habit dominated 5). 20 products for which people develop a repertoire of brands they find acceptable. Such advertising keeps the brand at the top of the short list, and reorders the list when it has slipped. Many tourist destinations spend millions of dollars each year trying to keep their names at the top of this short list. Destinations ranging from the largest theme park, World Disney, to cities such as Traverse City and Frankenmuth in Michigan employ this form of advertising. Modif¥_attitudes. This is advertising which seeks to change perceptions and attitudes towards a brand which may take months or even years to accomplish. The intended response is "I never thought of it like that before." This advertising strategy is appropriate for overcoming some problem or when an obstacle has been identified as the main priority for a product or service, such as giving a brand a new personality to keep up with changes in fashion. This approach is especially applicable to tourism destination marketing, particularly where image-repairing is an identified need. For instance, 21 China, after the Tiananmen Square Incident of 1989, launched a huge advertising campaign to present a peaceful image of Tiananmen Square and a stable Chinese government to counter the horrible image of bloodshed as depicted by the mass media. Similarly, after terrible storm destruction, Hawaii launched an advertising campaign to reassure potential visitors that it was again a safe place to visit. INDIRECT 6). This is advertising to maintain the status quo: "I always knew I was right to buy X." Continuous advertising for package goods and much corporate advertising is of this type. It is intended to confirm the added value consumers perceive from purchasing a branded product over other brands. Tourism advertising examples discussed in item 4) above also apply here. By keeping a destination on the short list of potential destinations to visit, the destination maintains a certain level of "social status" as well as priority in consumers' minds, reassuring past visitors that their 22 choice was a right one while creating interest among potential visitors. It is not suggested here that any advertising accomplishes all of these objectives or only one of them. Most advertising works on several levels of directness to accomplish a combination of objectives. These scales, however, are useful in helping us understand what possible responses an advertiser can expect from the consumer. The above scale of response classification, however, is not the only framework that advertisers use to try to account for how consumers respond to advertising. The "involvement" theory is another popular paradigm used to explain the effect of advertising. One of the models based on this theory was developed at the advertising agency Foote, Cone & Belding (Berger, 1986; Vaughn, 1980; 1986). It has been dubbed the FCB Grid. It conceptualizes the different psychological responses which the same people will have under different circumstances to an advertising message. Figure 2.2 is a representation of the FCB Grid. Its assumption is that if the causal factors behind the responses of an individual can be determined, it will be 23 THINK FEEL Economic MW man mvowmm Inn-MD» Fed-Leanne Bummh: &mfl LOW INVOLVEMENT Maw-M ”ML“ Figure 2.2 The FCB Grid (Source: adapted from Colin McDonald, How Adlertiainmflerks, 1993, p. 117) 24 possible to aggregate individuals into response segments. One central factor in predicting response is involvement in the product category. Those planning to buy products involving risk consciously attend advertising messages for those products (Gardner, 1981). Peoples' involvement in the decision process is directly related to their involvement in the product category (Haley, 1985). In tourism marketing for instance, if a destination being marketed fits into the category (for example, fun places for children) that a consumer is planning to choose, he/she will selectively expose himself/herself to information about it. Further, if there are competing destinations being marketed, the consumer will seek further information to help make a choice. Other examples (McDonald, 1993, p. 116) of each category (See Figure 2.3 for a diagram) are: 0 Think/high involvement: choosing a camera, or a life insurance policy. Customers will seek information, compare features and prices, think about and make a carefully considered choice. A relevant example in tourism is that of a traveller deciding where to go on a long weekend or vacation trip. The mental 25 THINK FEEL 2 1 Life 0 H ' . Mums ammnmmp .EmM' 0 thQn vacuum 35mm .(hmm HMH Q Rmmc lNVOLVEMENT 0 motor oil 0 Cmnplexion Soap 3 MW“ 4 .(hm . muflfld (Ina . Visit 10011 attraction at a . Wmm travel destination - - Madeflm (The 0 Rude 0 Pins IIOW INVOLVEMENT Figure 2.3 Model of Consumer Involvement (Source: adapted and revised from Colin McDonald, WW, 1993, p. 118) 26 sequence is Learn, then Feel, then Do, in accordance with the hierarchy of effects approaches discussed previously. Think/low involvement: not major purchases, therefore not requiring effort; but they do not give personal satisfaction; we are only interested if they work. It would be common to try a product on speculation and see if it does the job required before we continue buying it. Examples might be household cleaners or pet foods. In tourism, making a visit to a local park or neighboring town would serve as a good example. The visitor does not have to seek a lot of information about the particular park or town, since it does not involve him/her in any significant social or economic way. The sequence is Do-Learn-Feel. Feel/high involvement; an expensive perfume or fashion accessory; possibly cars or houses as well, although they require thought as well as emotional attraction, and straddle the think/feel line. In tourism, examples used in the Think/high involvement category can also serve as good examples, but choosing a luxury hotel or buying a yacht may serve as more appropriate examples. 27 The sequence is Feel—Learn-Do. O Feel/low involvement: buying a Mars Bar or a pint of lager. In tourism, a visitor's on-site behavior can often fit into this category. A visitor can choose a store or a restaurant to try it out. The sequence is Do-Feel-Learn. The Relevance-Accessibility Model of Advertising Effectiveness (RAM) proposed by Baker (1993) is another effectiveness paradigm that utilizes involvement theory. Underlying the RAM model are two basic axioms: 1)the elements of the advertising message must be accessible at the time of brand choice to be effective; 2)advertising information must be relevant to be effective. The RAM assumes that advertising can be effective only when the impressions it creates in a consumer's memory are retrieved (accessed) from memory when brand evaluations are made. The RAM further assumes that accessed advertising information must explicitly or implicitly be perceived by the consumer to be relevant at the time of brand choice if it is to have a direct effect on brand choice. Explicit recognition of relevance refers to those situations when consumers 28 consciously and deliberately seek out specific information because they believe it to be a more reliable indicator of quality than other information. One important proposition in this model is the assumption that brand response involvement determines consumers' preferred level of information. In Baker's words: Simply put, involvement refers to the intensity of mental effort (Cohen, 1982, quoted by Baker). Brand response involvement (BRI) is the degree of mental effort expended by the consumer while making a brand choice. When BRI is high, consumers are motivated to think intently about their brand choice; they are motivated to explicitly search for information and compare brand alternatives. At the other extreme, when BRI is very low, consumers are not motivated to think about their brand choice; they have little desire to either seek information or to compare brand alternatives. (p.63) The BRI has two major antecedents. They are perceived product differentiation and perceived product risk (Kapferer and Laurent, 1985). Perceived product differentiation refers to the degree consumers believe there are performance differences between the brand alternatives in a product category. In tourism, the product can be a tourist product, a service or a combination of both, or even a destination. 29 In the case of a destination, this perceived product differentiation can be in the areas of image, location and service. Perceived product risk refers to the probability that a brand choice may result in undesirable consequences. The consequences may be economic, perhaps the loss of thousands of dollars that result from choosing a wrong yacht, or the involvement of a large amount of money expended for an annual family vacation. They can be physical, perhaps the loss of life when going to a warring country or a crime- stricken destination. Risks can also be social and psychological, such as the fear of being ridiculed or fooled when visiting a place with totally different cultures and languages. Taylor (1974) suggests that the consumer's decision- making process is related to the risk involved as a result of choice. Since the outcome of a choice can only be known in the future, the consumer is forced to deal with uncertainly, or risk. Risk is often perceived to be painful and uncomfortable in that it may produce anxiety, in which case it must be dealt with in some manner by the consumer. In tourism, potential tourists call for information to 30 reduce the uncertainty or risk that they perceive to exist if a wrong choice of destination is made. Gitelson and Crompton (1983) suggest that there might be three reasons underlying external information-seeking behavior in travel decisions. First, travelling usually involves investment of discretionary dollars and a considerable investment of discretionary time and is, thus, a high risk purchase action. Obtaining further information is one way to reduce the perceived risk in the decision. Second, the consumer cannot directly observe or inexpensively experience beforehand what is being purchased. The third reason for an external search of information may be due to the lack of familiarity with the destination. From this point of view, brochures can be a very helpful tool in satisfying this external search for information. 2.5 Evaluation of Tourism.Destination Marketing Previous discussions on theories of mass media effects and advertising have laid a sound foundation for examining how tourism destination marketers evaluate their advertising and promotional efforts. Tourism destination marketing has been defined in various ways. According to Mill and 31 Morrison (1992), "Marketing is a management philosophy which, in light of tourist demand, makes it possible through research, forecasting, and selection to place tourism products in the market most in line with the organization's purpose for the greatest benefits" (p. 419). They further note that there are three concepts involved in this definition. First, marketing is a way of thinking about a situation that balances the needs of the tourist with the needs of the organization or destination. Second, this definition stresses the role of tourism research in understanding marketing segmentation. Third, the concepts of positioning and product life cycle are useful to ensure the proper placement of tourism services in the market and to suggest the appropriate marketing strategies and plans resulting from that decision. Destination marketing differs, in many ways, from traditional product marketing. In the first place, destinations are sold as an "experience" rather than a product. It is "intangible" in that it can not be stored; if it is not sold on the day offered, the value of the potential experience is lost and can not be recovered. Secondly, a destination's infrastructure cost is fixed. If 32 it is not used to its full capacity, under-utilized capacity represents a waste of resources and a cost to the' destination's businesses (Mill and Morrison, 1992). A third important variable that makes a tourist destination different from industrial products is that the experience or service is an amalgam of several products and services (Mill and Morrison, 1992). The experience results from the use of various services such as transportation, lodging, food and beverage, activities and so on. The fourth difference lies in the nature of demand. Demand for destinations is highly elastic, and seasonal in nature. It is, thus, subject to fluctuations in taste and fashion as well as more objective factors such as price. In many cases, the experience sought can be satisfied by other destinations or activities (Mill and Morrison, 1992). The last, and surely not the least important difference is the risk involved in making a decision to select a destination. Choosing a destination involves more variables than choosing most other products, especially if the destination is far away from home and thus involves variables such as distance and travel time, length of stay, safety, cost, vacation planning, means of transportation, to 33 name only a few. Moreover, these costs or investments are in most cases "unrefundable" or "unrecoverable". If a customer is dissatisfied with a product he/she purchases, he/she is able, in most cases, to return it or exchange it. In the case of choosing a destination, once one makes the trip, the cost and the experiences associated with the trip are no longer recoverable or exchangeable. These special characteristics of travel destination products suggest that: 1) marketing a destination should not copy the practice used in marketing most products, and consequently, 2) measuring the effectiveness of an advertising campaign or marketing effort should be done in accordance with its unique characteristics and 3) brochures can serve as a special medium of communication in satisfying the needs of the customer for purchasing a high-involvement product such as a travel destination. As noted above, choosing a destination as compared with purchasing industrial products is generally considered a high-involvement activity. It fits somewhat in the first quarter of FCB Grid (see Figure 2.2), i.e., learn-feel-do. It implies active information searching on the part of the consumer in advance of purchase. Consumers, however, can 34 not experience the destination product before they actually take the trip. Brochures, often with colorful pictures and detailed descriptions of scenic spots available at the destination, fill the visual and tangible void that other destination marketing media can not provide at a reasonable cost. Tourism destination marketing studies that relate most closely to brochure evaluation can be generally classified into three categories: 1) economic impact studies, 2) conversion studies and 3) brochure studies. While this classification seems arbitrary because of the overlapping of these three types of studies, it is also true that each of them has a different objective and central focus. 2.5.1 Economic Impact Studies Studies of the economic impact of travel information provided at welcome centers can be found in the works of Tierney and Haas (1988) and Fesemaier and Vogt (1993). Tierney and Haas found that welcome center visitors in Colorado spent approximately $440, or 11% more than those travellers who did not visit a center ($394). They also found that information obtained at the welcome center 35 influenced respondents' length of stay. Fesemaier and Vogt (1993), in their study to evaluate the economic impact of travel information provided at Indiana welcome centers, found that users of the travel information provided at the welcome centers spent, on average, $21 per visitor group more than originally planned. There are other types of economic studies (for more coverage on this topic, see Ritchie and Goeldner (eds). 1987, W). In general, the goal of these economic impact studies is to show the benefits and costs of marketing efforts. Tourism destination marketers are interested in how much money is spent by tourists and how these expenditures affect local economies including employment. Since these studies are not directly relevant to the focus of this paper, no further review of this literature will be provided. Although the studies of economic impact of travel information provided at welcome centers noted above provide a good deal of information about the relationship between users of such information and spending behavior, it is not known whether these findings can be generalized to information inquirers who call to request travel 36 information. One of the objectives of the case study presented in this paper was to find out whether mailed brochures affect respondents' spending patterns. 2.5.2 Conversion Studies Conversion studies are probably the most popular approach to studying the effectiveness of tourism advertising (Woodside, 1990) and have been claimed to be the basis for evaluating the effectiveness of tourism advertising (Silberman and Klock, 1986). Conversion studies have often been used indirectly for examining effects of brochures on consumers' travel decisions. As Burke and Lindblom (1989) note, conversion studies are often used to evaluate response to direct marketing of tourism destinations in which marketing activities take place followed by consumers' inquiries which result in the delivery of the requested promotional material often in the form of brochures. Conversion studies involve surveying a sample of inquirers generated from advertising and asking the surveyed inquirers whether they visited the destination from which they requested information. Conversion studies are often 37 conducted several months or longer after inquiries are received via toll-free telephone calls, coupons appearing in the ads, or reader response cards. Conversion studies yield a conversion rate which is the percentage of inquirers who visit after being exposed to the direct response marketing campaign. This conversion ratio can be used to estimate effectiveness and efficiency ratios, economic impact, and return on investment (Burke and Lindblom, 1989). This conversion rate, however, is obtained under the assumption that marketing efforts are responsible for all the tourists who visit, while in fact many information inquirers have already decided to visit the destination before they are exposed to the advertising messages (Purdue, 1984; Schroeder and Kreul, 1986; and Davidson-Peterson Associates, Inc., 1990). The conversion rate thus obtained without factoring out respondents who have already decided to visit is often referred to as "gross conversion rate" and has often been criticized for inflating the conversion rate obtained (Burke and Gitelson, 1990; Woodside, 1990 and Ronkainen and Woodside, 1987). Because of this, many researchers (Burke and Lindblom, 1990; Ronkainen and Woodside, 1987; Ellerbrock, 1981 and Ballman 38 et al., 1984) have pointed out the need to use "net conversion rate" versus "gross conversion rate". The "net conversion rate" involves factoring out inquirers who report that they planned the trip before being exposed to the marketing campaign or before requesting the information package. Conversion studies appear to have their roots in the sequential or direct effects of mass media theory and the sales effect in the traditional advertising literature. Siegel and Ziff—Levine (1990) presented a "conversion model" to summarize this approach. Their model is presented in Figure 2.4. This model resembles in many ways the familiar model of advertising effects discussed previously, i.e. AIDA. (Attention, Interest, Desire and Action). Advertising is assumed to work sequentially from stage to stage. That is, consumers are exposed to advertising and are then led through a series of stages, each of which draws them closer to the final desired goal of conversion, i.e. visiting the destination in question. One important aspect of this conversion model is that the inquiry/fulfillment stage is deemed a necessary step in the selling of a destination and 39 ADVBI'ISING I \¥ TARGET AUDIENCE \ AWAREOF Anvmmsmo / UNAIDED AWARENESS 0F DESTINATION POSITIVE IMAGE 0F DESTINATION DiQUIRY/HJIFIHMENT umwmwN I l CONVERSION! BEHAVIOR Figure 2.4 The Conversion Model (Source: adapted from Siegel and Ziff—Levine, "Evaluating tourism advertising campaigns: Conversion vs. advertising tracking studies". Iournal_of_Trayel_Research, Winter. 1990. p. 52) 40 must occur prior to visitation (Siegel and Ziff—Levine, 1990). Despite the importance of the inquiry stage, which usually results in the mailing of an information package which is often in the form of brochures to inquirers, the role of brochures, a special marketing medium included in most fulfillment packages, in the consumer's decision-making process is largely ignored. Another critical element in conversion studies involves visitor spending which is used in combination with the conversion rate to assess the economic impact of marketing efforts. Expenditure data are usually collected directly from the same respondents who were questioned in the process of estimating the conversion rate. Criticisms have arisen as to how accurate these expenditure data are (Ballman et al., 1984). Clearly, a major problem associated with estimating visitor expenditures is the time factor. There are two questions involved in the time factor. The first involves how lapsed time affects the consistengy of reported data. The second involves how the length of lapsed time between the visit and when respondents are asked to recall trip expenditures affects the diregtign (inflated or deflated) of the reported expenditures. 41 Many studies can be found in the literature that attempt to address the first question. Mak et a1. (1977) reported that participants in surveys often had problems accurately remembering the amount of money which they spent. Rylander II et al. (1995) use a research design to compare early and late respondents in terms of reported confidence in reporting expenditure data. They found that early respondents reported more confidence than late respondents in their reported expenditure data. Other researchers have reported that consistency of reported expenditures decreases quickly as lapsed time increases (Muha, 1976). This has led some researchers (Rylander II et al., 1995) to recommend procedures that obtain complete responses either during or immediately upon completion of a respondent's trip. However, it is not always practical to collect expenditure data immediately upon completion of a respondent's trip. This is especially true in the case of conversion studies for a couple of reasons. First, a sufficient number of brochure-inquirers has to accrue before a reasonable sample frame can be obtained. Second, on-site expenditure data collection may not be feasible due to the cost and the fact that the tourists being interviewed may 42 not be brochure-inquirers. It is, therefore, not uncommon to find conversion studies conducted several months after respondents have made their trips. The second aspect of the time factor involves how the length of lapsed time between the visit and receipt of the questionnaire affects the direction (inflated or deflated) of reported expenditures. Mak et al. (1977) reported that survey respondents significantly underestimated their expenditures. Similar conclusions were reported by Stynes and Mahoney (1989), Frechtling (1987) and Howard et al. (1991). In other words, the longer the length of lapsed time between the visit and when the questionnaire is received, the more likely respondents will underestimate their trip expenditures. A related issue in such studies is the effect of the presence or absence of cues. Providing cues may allow respondents to recover some memories that fall outside of the reference period but are recalled with a considerable degree of detail and vividness (Sudman and Schwarz, 1989). Recognizing that cues may influence results in evaluating effectiveness of brochures, the case study presented in this paper used an aided-unaided research design. Half of the 43 sample received a questionnaire with the brochure enclosed while the other half received only the questionnaire. The former will be referred to as the aided questionnaire and the latter as the unaided questionnaire. The purpose of this design was to assess the impact of this cue on responses to the series of questions included in the research instrument and thereby provide insight into the degree that recall bias may influence results from brochure studies of this type. 2.5.3 Review of Brochure Studies A review of the literature in the area of tourism destination marketing reveals that only a few research studies have been conducted to directly assess the effectiveness of brochures in marketing tourism destinations (Etzel and Wahlers, 1985; Baas et al., 1989 and Wicks and Schuett, 1991). It is, therefore, not surprising that little is known about the overall effectiveness of brochures (Wicks and Schuett, 1991). Baas et al. (1989) undertook a study to assess the effectiveness of an informational brochure for increasing awareness, interest, and participation in charter boat trip opportunities along the 44 Oregon Coast. They concluded that the brochure did not positively or negatively affect charter trip participation and appeared to be effective only in increasing awareness of charter trip opportunities. Using involvement and uncertainty (risk) theory, Etzel and Wahlers (1985) reported that respondents requesting the brochure and having no knowledge of the attraction were much more likely to visit than would otherwise be expected. It was also found that experienced travellers and active vacationers were significantly more likely to request information. They also found that income and education are positively correlated with active information seeking. The overall conversion rate for their study was 62.4%. Other brochure studies focus on on-site behaviors of brochure users. Fesenmaier and Vogt (1993) used three major criteria to measure the effectiveness of travel information provided at Indiana welcome centers. These criteria included: extension of time spent in Indiana, selection of alternative attractions, and incremental expenditures produced by longer visits or visiting different places. They concluded that motorists were influenced to extend their stay and select alternative attractions. Lime and 45 Lucas (1977) examined the effects of a brochure for redistributing Boundary Waters Canoe Area visitors to lesser used areas and found that 33% of the study participants visited a new area after receiving the brochure. Krumpe and Brown (1982) found that 37% of visitors who received a brochure on Yellowstone National Park backcountry trail selection took a lesser used trail, as compared to 14% of the visitors who did not receive a brochure. Roggenbuck and Berrier (1982) studied the effects of a brochure designed to disperse campers in the Shining Rock Wilderness Area. Significant differences were found between individuals receiving brochures and those not receiving brochures. Cherem (1982) compared newspapers, radio announcements, and brochures for redistributing use in a crowded developed recreation area. Brochures were found to have some effect on redistributing use. Still other brochure effects studies deal with physical aspects of brochures. Gilbert and Houghton (1991) conducted a brochure study to investigate the process of selecting brochures displayed in travel agency offices. The chief objective of their study was to examine factors which may lead a consumer to choose one or more brochures in 46 preference to others. From the above discussion, it is clear that each of these studies examined part of the potential effects of brochures on travel destination choice. The overall effectiveness of brochures may be assessed by combining the strengths of these studies, namely, the immediate sale effects (e.g. Baas et al., 1989), the relationship between travel decision and information seeking behavior (e.g. Etzel and Wahlers, 1985) and the on-site behaviors (e.g. Roggenbuck and Berrier, 1982; Fesenmaier and Vogt, 1993). One of the major problems with the existing brochure effects studies is that they fail to answer the more basic questions of: a) Are sales an adequate measure of effectiveness of brochures, b) Does brochure use cause visits? and if so, to what extent does it cause visits and c) How are brochures used in the consumer's decision-making process in choosing a destination and d) Are there other possible long—term or indirect effects of brochures? A change in peoples' attitude toward a destination may not result in immediate action but may influence their decision- making in the long run. These effects are as important as any other for attracting potential visitors and establishing 47 a foundation for growth. Moreover, what makes things even more complicated is the fact that in all of these studies a similar methodological approach was used to measure effectiveness of brochures even though they were distributed via many different channels. In some studies, brochures were distributed on site; in others, travel information was provided at welcome centers and still in other cases, brochures were mailed out to respondents. As has been discussed earlier, motivations and degree of involvement can be very different from one situation to another. Because of this consideration, this research focused on one method of distributing brochures that being mailed out brochures, and a research instrument and measurement criterion were designed to specifically serve this purpose. 2.5.4 The uniqueness of Brochures Brochures are a unique means of marketing as well as communication. Brochures are unique in that they not only differ from the electronic media but also from other print media where they obviously belong. Being a print medium, brochures differ from the electronic media in that they 48 require more active participation on the reader's part; that being a brochure can be read again and again and kept for later reference. They differ from other print media, newspapers and magazines in particular, in that their format, layout, length, quality of paper, means of distribution can vary from case to case. They, therefore, provide a wider spectrum for creativity and ingenuity, but at the same time create greater difficulty in measuring their effectiveness and generalizing findings to a larger population. Thirdly, most brochures are requested and distributed free of charge. As a result, motivational factors such as involvement and risk become important variables in explaining the effectiveness of brochures. Finally, most brochures are mailed directly to the consumer's home, falling suspiciously into the category of "junk mail" and running the risk of being thrown away without even being opened. As a result, the first criterion in measuring a brochure's effectiveness is to ask if the brochure in question is received, opened and read. Since choosing a destination is a relatively highly involved activity as discussed earlier, for advertising to be effective, the elements of the advertising message must 49 be accessible at the time of brand choice. Brochures carry information that does not go away as electronic advertising messages do and can be kept for future reference whenever consumers feel the need to refer to them. The uniqueness of the brochure as a communication medium dictates that the study of the effects of the brochure requires establishing its own set of variables, while at the same time utilizing as much as possible the rich heritage of communication and advertising theories when appropriate. 2.6 Major Research Hypotheses The above literature review and discussion gives rise to several research hypotheses. They are based either on theories central to communications literature or.on specific research findings highlighted in the literature review. Instead of inserting these hypotheses in various places immediately before relevant discussion, they are presented together here for the sake of convenience. In addition, these hypotheses reflect the four objectives of the study which were: 50 1) To define a set of criteria for measuring the effectiveness of brochures based on theories of mass media effects and advertising effectiveness models. 2) To apply these criteria in a case study to examine how effective brochures are and to identify what factors influence their effectiveness when used in marketing tourism destinations. 3) To identify how lapsed time between when trips are made and when respondents are asked to report influences reported trip expenditures. 4) To examine whether enclosing the previously requested brochure in a conversion study has any effect on respondents' reported impressions and effectiveness ratings of the brochure. The hypotheses used to guide this study are presented below. For convenience, variables involved in each hypothesis and their coding schemes are included here prior to detailed discussion later in this dissertation. Statistical procedures used to test hypotheses are also noted under each hypothesis. Those hypotheses which address research methods issues are 30 labeled. 51 BI: The more respondents use brochures, the more likely they are to report a higher level of trip satisfaction. TESL: Chi-square and Correlation Variables: Experience*, Visit, Recom, Rating, Consultation. Variables Defined: Experience: Was your experience in Frankenmuth on this trip worse or better than expected? 1 2 3 4 5 Much worse Much better Visit: How likely are you to visit Frankenmuth again? 1 2 3 4 5 Certain not to Certain to visit visit again again Recom: After this trip, did you recommend a visit to Frankenmuth to anyone? Yes(coded as 1 ) No (coded as 2) Don't know (coded as missing) Rating: Overall, how would you rate Frankenmuth as a tourist destination? 1 2 3 4 5 6 7 Terrible Excellent Consultation: Did you consult the brochure while visiting the Frankenmuth on this trip? Yes (coded as 1) No (coded as 2) * = dependent variable E2: Brochures are more likely to have impact on those who have not visited the destination than on those with prior trip(s) in terms of influencing the trip decision. Test: T-test and Correlation Variables: Prior visit(s), influence of the brochure*. Prior visit(s): Influence of the brochure: 52 I! . 1] I E' 3' Prior to requesting information, had you ever visited Frankenmuth? Yes (coded as 1) No (coded as 2) To what extent did the brochure influence your decision to visit Frankenmuth? l 2 3 4 5 6 7 Had no influence Actually caused at all me to visit E3: The more lapsed time there is between when a trip is made and when trip expenditures are reported, the more likely respondents will underestimate trip expenditures. IeeL_l: T-teSt Variables: Spending categories*, sampling year Tesr_2: Correlation Variables: Total expenditures*, dates of visit Iesb_3: Two-way analysis of variance Variables: Total expenditure*, year(factor) and aided/unaided (factor) I! . 1] I E' 1' Spending Categories: How much money did your spending unit spend on this trip in Frankenmuth on each of the following items? (open—ended) Sampling Year: 1992 (coded as 1) and 1994 (coded as 2) aided/unaided: aided (coded as 1) and unaided (coded as 2) Dates of Visit: When did you depart on your first trip after requesting information? (open-ended) 53 Notes: 1. 920900=September 1992; the same coding scheme applies to all departure date. 2. The total time span is 24 months from September 1992 to September 1994. 3. Specific codes are as follows: (920900 thru 92123l=1) (930100 thru 930331=2) (930400 thru 930631=3) (930700 thru 930931=4) (931000 thru 931231=5) (940100 thru 940331=6) (940400 thru 94063l=7) (940700 thru 940931=8). B4: Respondents receiving aided questionnaires will more likely report higher ratings of impressions and influences of brochures than respondents receiving unaided questionnaires. (The focus of this hypothesis is on research methods.) TESL l: T-test Variables: Usefulness*, interesting*, attractive* interest*, influence*, spending*, accurate*, quality* and aided/unaided. (Note: All the scales used for the following variables employed a 1-7 rating system, with 1 representing negative or no effects and 7 representing positive or strong effects.) Usefulness: How useful was the information in the brochure? Interesting: To what extent was the brochures interesting to read? Attractive: To what extent was the brochure attractive in design? Interest: To what extent did the brochure decrease/ increase your interest? Influence: To what extent did the brochure influence your decision? 54 Spending: To what extent did the brochure cause you to spend more money? Accurate: How accurate was the information contained in the brochure? Quality: How would you rate the overall quality of brochure? Aided/Unaided: Questionnaires with/without enclosed brochures. B5. There exists a significant linear relationship between effectiveness of brochures and the independent variables including: prior experience, income, perceived usefulness of information contained in brochures, whether or not respondents spent a night in Frankenmuth and age. EFFECTIVENESS: INCOME: PRIOR EXPERIENCE: Multiple regression Effectiveness*, income, prior, experience, night spent, usefulness, and age. I! . 1] l E' 1' To what extent did the brochure influence your decision to visit Frankenmuth?" 1 2 3 4 5 6 7 Had no influence Actually caused at all me to visit What was your total annual income before taxes? (There are 12 categories of income. The lowest income category was coded as 1, and the highest income category was coded as 12) Prior to requesting information, had you ever visited Frankenmuth? Yes (coded as 1) No (coded as 2) NIGHT SPENT: USEFULNESS: Age: 55 Did you spend any nights in Frankenmuth on this trip? Yes (coded as 1) No (coded as 2) How useful was the information in the brochure? 1 2 3 4 5 6 7 Not at all Extremely Useful Useful Age of respondents. (open-ended) CHAPTER THREE METHODS 3.1 Case Study Setting In early 1993, the director of the Frankenmuth Chamber of Commerce (FCC) contacted the director of the Travel, Tourism and Recreation Resource Center (TTRRC) for assistance in evaluating a brochure that the FCC mailed to people who called toll-free for information about Frankenmuth. A team was soon formed to work out a project plan. The FCC's research objectives were: 1) to establish the conversion rate: how many information inquirers actually visit Frankenmuth, 2) to establish how much money visitors spend in Frankenmuth and 3) to collect data on the demographic characteristics of information inquirers. The research design developed for this study accommodated the FCC's objectives plus the four additional objectives noted in the previous chapters. The brochure tested in this study was designed to attract potential tourists to visit Frankenmuth, a small town in Southeastern Michigan. Frankenmuth was founded in 1845 by a group of fifteen German-Lutheran missionaries who 56 57 came to the area for the purpose of teaching Christianity to the Chippewa Indians. Today, it is a thriving community of 4,400 residents who take pride in preserving their German heritage. Frankenmuth has long claimed to be the number one visitor attraction in Michigan. It is located 86 miles north of Detroit, Michigan, approximately a two-hour drive from Detroit. Each year over three million people visit Frankenmuth. The Frankenmuth Convention and Visitors Bureau is an arm of Frankenmuth Chamber of Commerce and has been actively involved in marketing Frankenmuth as a tourist destination, using various marketing tools such as media placements, public relations, billboards as well as brochures. A toll- free 800 telephone number is provided to inquire about Frankenmuth as a tourism destination. Each inquirer is sent a brochure, regardless of the nature of the inquiry. The brochure highlights scenic places in Frankenmuth and describes various recreational activities as well as providing lodging and travel-related information. It contains colorful pictures and has an attractive overall format. 58 3.2 Definition of Effectiveness The word effectiveness has been used to denote various meanings in different types of research studies. For example, in a typical conversion study, effectiveness would mean a higher conversion rate. In an advertising campaign, effectiveness may mean the number of recalls an audience can make after viewing a particular ad, or it may mean the size of audience a particular ad has reached. For the purpose of this study, the word effectiveness is operationalized to include the following concepts: 1). Whether or not the brochure reaches those requesting it (i.e. inquirers) 2). Whether or not the brochure is read by inquirers 3). Whether or not reading the brochure affects the destination decision-making process 4). Whether or not reading the brochure influences on-site behavior of inquirers to include: length of stay, spending as well as choice of lodging 5). Whether or not the brochure is used as a guide during the visit to the destination 6). To what degree reading the brochure leads to changes in inquirers' attitudes, perceptions and expectations 59 toward the chosen destination before, during and after the visit. 3.3 Questionnaire Development A careful review of the literature was conducted to make sure the questions asked in the questionnaire were relevant to the research questions of concern and were consistent with theories of media effects and advertising related to brochure studies in tourism destination marketing. Questionnaire development also reflected the major two thrusts of this dissertation, one being concerned about methodology questions and the other with brochure effectiveness as defined above. Two seven-page questionnaires were developed for this study: an unaided recall questionnaire and an aided recall questionnaire. The brochure being evaluated was enclosed with the aided recall questionnaire to serve as a reminder of what had been previously sent to respondents; no brochure was enclosed with the unaided recall questionnaire. The two questionnaires were identical except that the brochure under study was referred to as "the enclosed brochure" in the aided recall questionnaire, but only "the brochure" in the 60 unaided recall questionnaire. A copy of the unaided recall questionnaire is presented in the Appendix A. 3.4 Major Variables Studied Variables were developed with both the FCC's objectives and this dissertation's objectives in mind. Major variables are presented below: 1) Information sources for phone number of the.FCC 2) Prior visit to Frankenmuth 3) Trip decision 4) Visited or not after requesting information 5) Length of stay in Frankenmuth 6) Trip expenditures in Frankenmuth 7) Trip satisfaction 8) Brochure use 9) Impressions of the brochure 10) Perceived usefulness of the brochure 11) Brochure effectiveness 12) Demographics of respondents 61 3.5 Research Design The research design developed for this study reflects the theoretical underpinnings of the mass media effects, advertising and tourism destination marketing literature reviewed. Figure 3.1 illustrates the research design. Through this design, research hypotheses regarding methodology that were introduced in Chapter three can be examined. Specifically, Hypotheses 3 and 4 are methodological questions. The relationship between the research design and the hypotheses is shown in Figure 3.1. As a reminder, the two hypotheses are repeated here: KB: The more lapsed time there is between when a trip is made and when trip expenditures are reported, the more likely respondents will underestimate trip expenditures. H4: Respondents receiving aided questionnaires will more likely report higher ratings of impressions and influences of brochures than respondents receiving unaided questionnaires. 62 aided 1992-93 1993-94 aided A A a a 5% BM 53 ‘:KHDQNHTB’> 8h é; EM 3 93 113 3 a 8 8 y Y unaided 1992-93 1993-94 unaided Figure 3.1. Overall Research Design Note: The 1992—93 sample frame consisted of brochure- inquirers between September 1, 1992 to March 15, 1993; and the 1993-94 sample frame from September 1, 1993 to March 15, 1994. 3.6 Sampling Frame The sampling frame consisted of the list of people who requested brochures in 1992—93 and 1993-94 periods respectively. The 1992—93 sample frame consisted of brochure-inquirers between September 1, 1992 and March 15, 63 1993; the 1993-94 sample frame consisted of brochure- inquirers between September 1, 1993 and March 15, 1994. There were two reasons for choosing this discontinuous sample frame. First, The FCC was more interested in knowing how effective their brochures were in attracting prospective late fall and winter season visitors. Second, for the purpose of this study, a compatible sample frame earlier in time was needed to conduct meaningful research comparisons. The 1992-93 sample frame consists of 3,512 inquirers and the 1993-94 sample frame consists of 6,113 inquires. The total sample frame was, thus, 9625. Both of these sample frames were provided by the Frankenmuth Chamber of Commerce. Inquiries from Mexico and overseas countries were deleted from the sampling frame, due to the fact that the questionnaire was written in English and the consideration that their reported trip expenditures might skew the reported mean spending. Canadian respondents were not excluded. 3.7 Sampling Procedure A systematic random sampling procedure was applied to both the 1992-93 and the 1993-94 lists. One-thousand-two- 64 hundred respondents were selected for the 1992-93 sample and another 1200 for 1993—94, achieving a total sample size of 2400 for the study. Of the 1200 respondents selected for the 1992-93 sample, 600 were sent the unaided questionnaire (the brochure under study was not enclosed); the other 600 were sent the aided questionnaire (the brochure under study was enclosed). The same procedure was applied to the 1993— 94 sample; six hundred received the aided and 600 the unaided questionnaire. 3.8 Data Collection A decision was made to send out the questionnaires by certified mail, based on the previous satisfactory experience of the author with this approach (Gull Lake View Golf Club Visitors Survey, 1992, unpublished). Although it costs more initially, it produces a higher response rate in a short period of time, saving time and the cost for postcards and other reminder mailings. No reminder postcards were sent. Every effort, however, was made to maximize response rates within the constraints of the project budget. Cover letters were personalized and individually signed. A coupon 65 book and a coupon worth two dollars redeemable at one of the restaurants in Frankenmuth were enclosed with the questionnaires to the potential respondents. Additionally, upon return of the questionnaire respondents were offered a chance to win one of the following gifts: 1. A Frankenmuth overnight getaway package, including a two-night stay and meals at the Bavarian Inn and Zehnder's Restaurant, a Bronner's Christmas Wonderland Gift Certificate, and passes for many of Frankenmuth's tours and activities. 2. A "Taste of Frankenmuth" gift basket. 3. A signed and numbered print of the famous Frankenmuth Holz-Brucke (covered wooden bridge). CHAPTER FOUR RESULTS 4.1 Basic Findings Questionnaires were mailed out May 15, 1994, and, by September 1, 1994, a total of 1192 completed questionnaires were received. Responses across the four subsamples was evenly distributed as can be seen in Figure 4.1. The overall response rate was 49.7 percent. The demographic characteristics of respondents are provided in Table 4.1. The majority of the respondents were female (72%). More than half of the respondents (52%) worked full time. Over 60% of the respondents have a college education. It was found that 91 percent of respondents received a brochure after requesting information from Frankenmuth Chamber of Commerce (FCC). Fifty percent of respondents said they visited Frankenmuth following requesting information from FCC, of which 99 percent stated that they read the brochure that was sent to them. This fifty percentage is what is commonly referred to as the "gross conversion rate." (The percentage of individuals who responded to advertising by requesting information and 66 67 , K gamm- ”as? K @3511) a“ l, .ll_._i agiggfiflTTTU‘fl_4v 199394 melded 25.8% I: ,, Ufifigfidedfiwl. Figure 4.1 Number of the Responses by Year and Type of Questionnaire Sent. 68 Table 4.1 Demographic and Socioeconomic Characteristics of Respondents Variable Xear IQLal IQLal 1992-93 1993-94 (Percent) Gender male 162 169 331 28% female 431 421 852 72% Education elementary 9 8 17 2 high school 197 213 410 35 undergraduate 264 274 538 46 graduate 115 91 206 18 W full time 296 315 611 52 part time 68 74 142 12 unemployed 6 8 14 1 homemaker 71 72 143 12 retired 142 93 235 20 student 4 17 21 2 other 5 10 15 1 has under 18 1 7 8 0.7 18 - 24 12 17 29 2.5 25 - 34 79 96 175 15 35 - 49 181 239 420 36 50 - 65 219 161 380 32.8 66 and over 92 63 155 13.3 Income under $25,000 22 22 44 4.2 $25,000 - $34,999 20 23 43 4.1 $35,000 - $49,999 30 31 61 5.9 $50,000 - $74,999 74 76 150 14.4 $75,000 - $104,999 150 126 276 26.5 $105,000 - $119,999 139 155 294 28.3 $120,000 and over 56 60 116 11.2 69 eventually visited the destination.) Burke and Gitelson (1990) have noted that regional and state tourism agencies regularly report 60% to 80% conversion rates. Compared with this range, the reported rate of 50% is relatively low. Ronkainen and Woodside (1987), however, compiled a list of conversion rates from several studies which showed rates ranging from 22% to 62%. Compared with this range, the conversion rate obtained for this study is relatively high. As has been pointed out from the beginning of this paper, however, gross conversion rates are controversial and, in many cases, misleading and so are net conversion rates. Neither of these ratios are of central interest in this paper and will not be discussed in further detail here. Among the other half of respondents who said they did not visit Frankenmuth after requesting information, ninety— six percent claimed that they read the brochure that was sent to them. In addition, among those respondents who did not visit Frankenmuth after requesting information, 91 percent said that they were considering a visit to Frankenmuth in the near future (Figure 4.2). 70 Considering a visit 91196 Figure 4.2 Not considering 8£B6 Among Those Not Visiting Frankenmuth, the Percentage of People Who Were/Were Not Considering A Trip in the Near Future. 71 When asked about the timing of their request for information, only 29 percent (Figure 4.3) of the respondents said that, when they requested the information, a visit to Frankenmuth was already planned. This number seems to be considerably lower than figures reported in other studies. For instance, a study of the effectiveness of the 1990 Wisconsin tourism marketing program (Davidson-Peterson Associates, Inc., 1990) found that 80% of respondents had already decided to go to Wisconsin at the time of requesting information. If the number (29%) reported here is true, it means that seventy-one percent of the respondents had either not made their final decision where to go for their trip or were not even considering a trip at all when they requested information. This is an important finding for this destination's marketers. Its implication is that there is considerable opportunity for them to effectively influence and attract this large portion of the undecided potential travel population. 72 Already decided upon 29.0% Not being considered 6.2% Being considered 64.9% Figure 4.3 Status of Decision of All Respondents to Visit Frankenmuth When Inquiry Was Made 73 More than half (51.8%) of the respondents had visited Frankenmuth on another occasion prior to requesting information in this case, and less than 25 percent(23.6%) of the respondents had requested information to help plan their prior visit(s) to Frankenmuth. More than half (64%) of the respondents reside out of Michigan and almost 80 percent (79.6%) of the respondents reported that Frankenmuth was their primary destination. These findings are presented in Table 4.2. Table 4.2 Other Characteristics of Respondents (N=1192) Variable Iear Total 1992-93 1993-94 (Percent) (Percent) Prior visit(s) 47.7 55.9 51.8 Request Information for Prior Visit(s) 27 20.7 23.6 Residence Michigan 28.7 42.7 36 Out of Michigan 71.3 57.3 64 Primary Destination 77.1 82.9 79.6 74 Seventy-five percent of the 1992-93 respondents took their trips before June 1993, highly representative of the sampling duration which ranged from September 1, 1992 to March 15, 1993. For the 1993-94 sample, the percentage was even higher. Over 95 percent of the 1993-94 sample made their trips between September 1, 1993 to June 30, 1993-94, the sample period being from September 1, 1993 to March 15, 1993—94. The findings are presented in Table 4.3. Table 4.3 Distribution of Dates When Trips Were Taken by 1992-93 and 1993-94 Respondents (N=558) Dates 1992-93 Dates 1993-94 (12 Months) (12 Months) Sep. 92-Dec. 92 37.8% Sep. 93—Dec. 93 40.7% Jan. 93-Mar. 93 10.3% Jan. 94-Mar. 94 35.3% Apr. 93-Jun. 93 26.3% Apr. 94-Jun. 94 18.9% Jul. 93-Sep. 93 25% Jul. 94—Sep. 94 4.8% 75 When asked " To what extent did the brochure decrease or increase your interest in visiting Frankenmuth?" on a scale from 1 to 7 with 1 representing "not at all" and 7 indicating "greatly increased my interest", the mean response was 5.65, which indicates that the brochure was very helpful in increasing respondents' interest in visiting Frankenmuth. However, using a similar scale, when asked "To what extent did the brochure cause you to spend more money in Frankenmuth on this trip than you would have otherwise?", the mean was only 3, which means that basically the brochure had only minimal effect on the total expenditure the respondents would have otherwise made on their trip. A minimal to moderate effect of the brochure was found when respondents were asked to what extent the brochure influenced their decision to visit Frankenmuth. The mean for this question was 4.5 on a scale from 1 to 7 with 1 being "no influence" , 4 functions as a mid—point and 7 "actually caused me to visit". Ratings on the rest of the variables in Table 4.4 were, however, relatively high. The mean for usefulness of the brochure was 5.79; the mean for question 2 (interesting to read?) was 5.61; the mean for Table 4.4 Effects and Impressions of the Brochure 76 (N=533) Variables Year Cases Mean‘ G_Meanb 1)How useful was the 1992-93 310 5.82 5.79 info in the brochure? 1993-94 235 5.75 2)To what extent was the 1992-93 310 5.59 5.61 brochure interesting to read? 1993-94 235 5.62 3)To what extent was the 1992-93 309 5.78 5.80 brochure attractive 1993-94 235 5.82 in design? 4)To what extent did the 1992-93 308 5.62 5.65 brochure decrease/increase 1993-94 235 5.68 your interest? 5)To what extent did the 1992-93 306 4.44 4.45 brochure influence your 1993-94 232 4.46 decision to visit? 6)To what extent did the 1992-93 305 3.36 3.29 brochure cause you to 1993-94 232 3.21 spend more money? 7)How accurate was the 1992-93 302 6.05 6.08 information contained 1993-94 235 6.11 in the brochure? 8)How would you rate the 1992-93 308 6.02 6.03 overall quality of 1993-94 235 6.05 the brochure? Note: a Means obtained from a scale of 1 to 7, each of them representing the opposite extremes of the scale: negative or lower rating and positive or higher rating. b G_mean stands for grand mean for the combined samples. 77 brochure design was 5.8; the mean for information accuracy was 6.08 and the mean for brochure quality was 6.03. The responses to these questions are summarized in Table 4.4. 4.2 Brochure Use and Trip Satisfaction Hypothesis 1 states that the more respondents use brochures, the more likely they are to report a higher level of trip satisfaction. When asked the question whether they consulted the brochure while visiting Frankenmuth, 78 percent of the respondents answered yes. This suggests that their information needs did not end with the beginning of the trip but continued during the period of the visit. The usefulness of the brochure was, thus, greatly increased and consequently more likely to exert its expected effects. Consulting the brochure on site was taken to represent frequent use of the brochure, since logic suggests that respondents who consulted the brochure on site are likely to use the brochure more often than respondents who did not consult the brochure. The chi-square test shows that there is a significant difference between those who consulted the brochure on site and those who did not in terms of their trip experience. A look at the Table 4.5 reveals that if 78 Table 4.5 Chi-square Results of the Relationship between Consultation of the Brochure on Site and Recommendation to Someone Else after Trip (N=533) Recommendation Consultation yes no yes 367 45 no 60 19 Phi = .258 p=.0000 Chi-square significance p=.0000 people consulted the brochure, they were more likely to report recommending a trip to Frankenmuth to someone else. Phi correlation is also significant. Pearson's correlation for other variables also reveals that consulting the brochure on site correlated significantly with reported better experience than expected. However, it does not correlate significantly with future repeat visit or higher rating of Frankenmuth as a tourist destination. The results of these tests are presented in Table 4.6. 79 Table 4.6 Correlation between Consulting Brochures on Site and the Level of Reported Satisfaction (N=533) Experience Visit Rating Consultation —.1256 .0494 -.0254 P: .000 P=. 254 P: .558 Note: Experience: Was your experience in Frankenmuth on this trip worse or better than expected? 1 2 3 4 5 Much worse Much better Visit: How likely are you to visit Frankenmuth again? 1 2 3 4 5 Certain not to Certain to visit visit again again Rating: Overall, how would you rate Frankenmuth as a tourist destination? 1 2 3 4 5 6 7 Terrible Excellent Consultation: Did you consult the brochure while visiting the Frankenmuth on this trip? Yes (coded as 1) No (coded as 2) The correlation coefficient between "experience" and "consultation" is negative, which means that respondents who said they consulted the brochure on site were more likely to report that their experience in Frankenmuth was better than expected. The relationship (r=—.13) is significant, though weak. 80 4.3 Comparison of 1992-93 and 1993-94 Reported Trip Spending So far, we have discussed the findings based on the whole sample and have not considered possible differences that may exist between the 1992-93 and 1993-94 samples. One of the mechanisms built into this research design was intended to test the hypothesis that related to memory recall. Since the same spending questions were asked of both 1992-93 and 1993-94 respondents, responses can be compared and tested for statistically significant differences. In this study, different reported spending patterns were found between the 1992-93 and 1993-94 samples. The findings are presented in Table 4.7. It shows that the 1992-93 sample of respondents consistently reported more spending than the 1993-94 sample of respondents, although only total mean spending and camping fee show significant differences when compared using the T—test (Table 4.7). 81 Table 4.7 T-test for Reported Spending Means (in Dollars) for the 1992-93 and 1993-94 Samples (N=608) X231 1992-93 1993—94 Significance Lodging 100 80 NS Camping fee 3.5 1 * Gifts, etc. 165 148 NS Grocery 15 12 NS Restaurant 110 93 NS Vehicle 16 13 NS Rental fees 0.9 0 NS Guided tours 4.5 2.6 NS Total 416 342 * Note: *=significant at p=0.05 NS=not significant at p=0.05 A correlation test was conducted to test the relationship between time of visit and amount of trip expenditures recalled. In the questionnaire, respondents were asked when they departed for Frankenmuth after requesting information. This question provided necessary information for detailed month of the year when the trip was made. Table 4.8 summarizes the correlation results. 82 Table 4.8 Correlations between Dates of Departure for Frankenmuth and Reported Total Trip Expenditures (N=506) Departure Date Total Trip Expenditure -.1101 P: .013 Note: 1. 920900=September 1992-93; the same coding scheme applies to all departure dates. 2. The total time span is 24 months from September 1992 to September 1994. 3. Specific codes are as follows: (920900 thru 921231=1) (930100 thru 930331=2) (930400 thru 930631=3) (930700 thru 930931=4) (931000 thru 931231=5) (940100 thru 940331=6) (940400 thru 940631=7) (940700 thru 940931=8). From Table 4.8, it can be seen that total spending is correlated negatively and significantly with when respondents departed. Thus, 1992-93 respondents were more likely to report more trip expenditures. To further examine whether aided and unaided questionnaires interact with the variable "year" (sampling year), two-way analysis of variance was performed. The results are shown in Table 4.9. It can be seen from Table 4.9 that under main effects, only 83 "year" is significant and there is no interaction between year and aided/unaided. In other words, respondents' recall of trip expenditures was not affected by whether or not they had the brochure in front of them at the time they were asked to recall their expenditures. Table 4.9 Two-Way Analysis of Variance for Trip Expenditures (Factors: Year and Aided/Unaided) (N=530) Sum of Mean Source of Variation Squares D. Square Sig Main Effects 974310 2 487154 .34 .097 Aided/Unaided 14679 1 14679 .07 .790 Year 963323 1 963323 .64 .032 2-Way Interactions 790589 1 790589 .82 .052 Aided/Unaided-Year 790589 1 790589 .85 .052 Residual 109491520 527 207764 Total 111174315 530 209762 84 This finding from this study is surprising in that it does not conform with the findings of previous studies in the literature. It is very tempting at this point to jump to the conclusion that it is the reverse of the previous studies, that is, the more lapsed time there is between when a trip is made and when trip expenditures are reported, the more likely respondents will overestimate trip expenditures. However, a close look at Table 4.2 reveals that there are more out-of—state visitors in the 1992—93 sample than in the 1993-94 sample. If out-of-state visitors stayed longer in Frankenmuth and, therefore, spent more than in-state visitors, then this surprising finding is only a reflection of this fact. As Table 4.10 shows, out-of—state visitors were the dominant consumers of lodging across the six categories listed with the exception of the 1993-94 category in which residents and non—residents are equally represented. Out—of-state residents were even more dominant in 1992-93 than in 1993-94. Thus, the observed expenditure difference between the two samples is correlated with length of stay difference between the two samples. 85 Table 4.10 Number of Nights Spent in Frankenmuth by Residence and by Sampling Year (N=3ll) Sampling Year Length of Stay Residence (# of Night) Michigan Out of State (Percent) (Percent) 1992-93 1 37 63 2 28.4 71.6 3 and more 5 95 1993-94 1 50 50 2 47.6 57.4 3 and more 16.7 83.3 According to the literature on spending recall, the longer the lapsed time between the time of trip and the time of recall, the more likely respondents will underestimate trip expenditures. After accounting for length of stay difference, the findings from this study suggest that reported trip expenditures are not significantly affected by lapsed time of up to 18-24 months. This is an encouraging result since it suggests that estimates of trip expenditures obtained via conversion studies may not be so sensitive to lapsed time as is commonly assumed. However, despite Woodside's claim (1987) that visitors can accurately recall 86 expenditures, the results from this study only suggest expenditure estimates obtained are consistent across a post- trip design ranging from a few months to up to 2 years. They shed no new light on the issue of the accuracy of estimates obtained many months after trips have been made. 4.4 Comparison of unaided and Aided Recall Results Recall that an aided questionnaire refers to the fact that a brochure was enclosed with the mail survey. An unaided questionnaire was one without this enclosure. Hypothesis 4 (H4) of this paper states that respondents receiving aided questionnaires will more likely report higher ratings of impressions and influences of brochures than respondents receiving unaided questionnaires. Comparisons can be made with the T-test to see if there is a difference between the mean ratings between unaided and aided respondents. The results are shown in Table 4.11. Based on the findings shown in Table 4.11, it is clear that respondents receiving aided questionnaires reported significantly higher ratings on all variables tested here except the one regarding the accuracy of the information contained in the brochure. It appears that memory cues 87 Table 4.11 A Summary of the T-test Results for Unaided and Aided Mean Ratings of Impressions and Effects of the Brochure (N=533) Variables Group Mean Significance 1)How useful was the infor- unaided 5.67 * mation in the brochure? aided 5.91 2)To what extent was the unaided 5.43 * brochure interesting to read? aided 5.81 3)To what extent was the unaided 5.54 * brochure attractive aided 6.08 in design? 4)To what extent did the unaided 5.45 * brochure decrease/increase aided 5.86 your interest? 5)To what extent did the unaided 4.24 * brochure influence your aided 4.67 decision to visit? 6)To what extent did the unaided 3.13 * brochure cause you to aided 3.48 spend more money? 7)How accurate was the unaided 6.03 NS information contained aided 6.13 in the brochure? 8)How would you rate the unaided 5.86 * overall quality of aided 6.21 the brochure? Note: * means significant at p30.05; NS not significant. 88 (aided questionnaires) did play a role in refreshing or reinforcing memories, which led to the tendency of reporting higher ratings. 4.5 The Role of Prior Experience One important question that this research was designed to explore was that of how prior experience with Frankenmuth [i.e. visit(s)] is related to the importance of brochures in influencing the decision to choose Frankenmuth again as a destination. The issue of the influence of prior experience really touches upon two separate but related questions: (a) Was there any difference between those who had visited before and those who had not visited before? (b) What is the relation between having prior experience and the effect of the brochure. In other words, will prior experience make the brochure less important or more important in the decision-making process? The results of the T-test (Table 4.12) show that there exists a significant difference between responses by those who had visited Frankenmuth before and by those who had not in terms of the role of the brochure in the process of choosing Frankenmuth as a tourist destination. This is 89 Table 4.12 T-test Results of the Influence of Brochures on One's Trip Decision by Prior Experience (N=519) Groups Mean Significance (Influence Score) Those who had visited 4.16 * Those who had not visited 4.80 Note: *=significant at p=0.001 further supported by the correlation test shown in Table 4.13. Although the relationship is relatively weak (r=0.1656), it is positive and significant. To be specific, brochures are likely to exert more influence on the travel decision of those who have not visited Frankenmuth before than upon those who have previously visited Frankenmuth. The hypothesis that prior experience reduces decision-making risk and consequently brochures will exert less impact on brochure—inquirers is therefore supported. Caution should be exercised, though, in interpreting this correlation since it is relatively weak. 90 Table 4.13 Correlation between Influence of Brochures and Prior Experience (N=519) Brochure Influence Prior Experience .1656 P: .000 Note: Brochure Influence: Prior to requesting information, had you ever visited Frankenmuth? Yes (coded as 1) No (coded as 2) Prior Experience: To what extent did the brochure influence your decision to visit? 1 2 3 4 5 6 7 (Had no influence) (Caused me to visit) Still, this is an important finding since it has practical implications for destination marketers. For instance, it may be feasible for destination marketers to design a separate brochure that is targeted at the segment of inquirers who have visited before and who have different information needs. New and updated information can be enclosed to generate new excitement about and interest in the destination among those already familiar with it. The finding that respondents without prior visit(s) to Frankenmuth were more likely to be influenced by the 91 brochure conforms with Milman and Pizam's (1995) finding that respondents go through a process from awareness to familiarity to action. Familiarity is considered a critical stage in this process. Potential first time visitors call for information (awareness) to gain familiarity (reading the brochure received) which may lead to action (visit to Frankenmuth). In this process, reading the brochure can be a necessary step to gain familiarity which in turn increases the likelihood of making a trip to the destination. 4.6 A Linear Multiple Regression Model of Effectiveness One goal of science is prediction (Shavelson, 1988). Linear multiple regression is a statistical model used to examine the relationship between sets of independent (X) variables and a dependent (Y) variable (Neale et al., 1994). It is widely used to estimate the scale and significance of the effects of a number of independent variables on a dependent variable. It is also widely used as a forecasting tool in various disciplines. The conceptual basis of the regression model developed for this study builds on the notion that effectiveness as defined in the beginning of this paper is affected by 92 various variables derived from involvement theory and risk theory which were discussed in the literature review chapter. These variables are represented by different questions employed in this study. Specifically, the independent variables were represented by: 1) respondents' annual household income, 2) prior experience in terms of whether respondents had visited the destination before the inquiry, 3) perceived usefulness of brochures, 4) whether or not respondents spent a night in Frankenmuth and 5) age of respondents. The reason for choosing income to represent socio- economic variables is that other variables such as education and employment status tend to be highly correlated with income. In order to keep the model as parsimonious and practical as possible, it was decided to include only the income variable in the model. In their regression model for assessing travel advertising effectiveness, Messmer and Johnson (1993), using similar argument, selected income per capita to capture the socio-economic composition of their sample. The viability of using income as an independent variable in the regression model was also supported by Silberman and Klock's (1986) findings. 93 Prior experience was included in the model because it was found to be independently related to brochure effectiveness as discussed in the previous section and in previous studies (e.g. Etzel and Wahlers, 1985). The inclusion of the variable "any night spent in Frankenmuth" is based on the hypothesis that the spending one or more nights in Frankenmuth is associated with the degree of involvement and perceived risk. The reason for selecting the perceived usefulness of the brochure variable is intuitive: if the brochure was perceived to be useful, it would have more influence than if it were perceived not to be useful. The last variable, age of respondent, is included as an exploratory variable to see if age affects the perceived impact of the brochure. To ensure that the regression model accurately reflects the effectiveness of brochures, respondents who had already made the decision to visit Frankenmuth were excluded from the analysis. The general linear multiple regression model can be expressed as follows: Y1 = Bo 1' leli + Bzxzi 1' + Bpxpi '1‘ ei (l) 94 where Bo is the intercept and other B's are the slopes associated with the respective independent variables. The notation Xpi indicates the value of the pth independent variable for case i. an terms are independent random variables or the population error term, which is the difference between the actual Y and (Y') predicted by the regression model. The model assumes that there is a normal distribution of the dependent variable for every combination of the values of the independent variables in the model. One often used strategy for testing the statistical significance of multiple regression model is to use the F test. The use of the F test in regression is based on the same underlying concept of analysis of variance: dividing total variance into "accounted for" and "unaccounted for". If the F value is significant, the null hypothesis that the predictions were no better than chance can be rejected. Specifically, the mathematical formula for the regression model used in this particular study can be written as: E1 = Bo + B1111 + szzi + B3N3i + B4U4i + BsAsi + 61 (2) 95 where, again, Bo is the intercept and B's are the slopes. The symbols of this model are explained below: Ei:= EFFECTIVENESS (effectiveness of the brochure, represented here by the response to the question: "To what extent did the brochure influence your decision to visit Frankenmuth?" I1i = INCOME (a measure of total household income) P2i = PRIOR EXPERIENCE (Had the respondent visited Frankenmuth before?) N3i = NIGHT SPENT (whether or not staying overnight) U4i = USEFULNESS (perceived usefulness of the brochure) ASi = AGE (of respondents) In this equation, the variables PRIOR EXPERIENCE (Pfi) and NIGHT SPENT (Nu) are indicator variables (Norusis, 1992) or dummy variables (Williams, 1979), that is, variables coded as 0 and 1. PRIOR EXPERIENCE was coded 1 for those responses with "yes" answers and 0 for "no". The same scheme was applied to NIGHT SPENT. One of the first steps in calculating an equation with several independent variables is to calculate a correlation matrix for all variables. The purpose of producing this correlation matrix is to make sure that multicollinearity is 96 not present, that is, to see if the independent variables are really independent (are not correlated). The full correlation matrix for the multiple regression model for this study is presented in Table 4.14. An examination of Table 4.14 indicates that none of the independent variables showed more than a weak correlation with other independent variables in the model. According to Tull and Hawkins (1990), only when correlation coefficients are larger than .35, should the problem of multicollinearity be considered. Accordingly, the problem of multicollinearity can be ruled out in this case. The next step is to derive the regression coefficients. The coefficients estimated for this model as well as the related goodness of fit estimates and analysis of variance results are shown in Table 4.15. There are several noteworthy findings contained in Table 4.15. First, the sign of the regression coefficient (B) for PRIOR EXPERIENCE is negative and the coefficient is statistically significant. In this analysis, if respondents answered "yes" to the question "Have you visited Frankenmuth prior 97 Table 4.14 Correlation Matrix for the Multiple Regression Model (N=176) E: In P2: N31 U4: A51 Ei 1.000 Ili - .111 (p=.018) 921 -.212 —.026 (p=.000) (p=.312) N31 .038 .033 .005 (p=.237) (p=.269) (p=.465) U4i .334 -.097 .006 .020 (p=.000) (p=.034) (p=.456) (p=.353) A,i -.046 -.212 .007 .050 -.029 (p=.195) (p=.000) (p=.446) (p=.174) (p=.294) Note Ei== EFFECTIVENESS (effectiveness of the brochure, represented here by the response to the question: "To what extent did the brochure influence your decision to visit Frankenmuth?" I1i - INCOME (a measure of total household income) Pfi - PRIOR EXPERIENCE (Had the respondent visited Frankenmuth before?) Nu - NIGHT SPENT (whether or not staying overnight) U“ USEFULNESS (perceived usefulness of the brochure) ASi _ Age (of respondents ) 98 Table 4.15 Regression Results for Effectiveness Model Parameter Estimates Mariahle B SE B Beta T Constant 1.83 .73 2.51 Prior experience -.84 .19 -.22 -4.41 Night Spent .14 .21 .03 .66 Usefulness .57 .09 .33 6.62 Income -.09 .06 -.08 —1.74 Age .01 .01 .02 .38 Goodness of Fit Estimates Sig T .013 .000 .510 .000 .102 .707 Mariable lalne Multiple R .41 R2 .17 Adjusted R2 .15 Standard Error 1.79 Analysis of Variance Results Beam Snmf Sbnrce £122de mares Regression 5 221 Residual 347 1113 Total 352 F-Statistic 13.81 Significance of F .0000 Note: p=0.05 B=unstandardized partial regression coefficient(prc) SE B=unstandardized prc standard error Beta=standardized prc T=T~test score, and Sig T=significance level of T-test 99 to requesting information", "yes" was coded 1 and "no" was coded 0. The negative relationship between prior experience and effectiveness means that respondents with prior experience with the destination were more likely to report that the brochure had less influence on their decision to make a trip to Frankenmuth. In other words, those respondents who had not visited Frankenmuth were more likely to rely on the brochure for information to make a decision. Prior experience is, thus, a significant predictor of brochure effectiveness. Second, the predictor variable, perceived USEFULNESS, is significant and has a positive B. This suggests that the more a respondent thought that the information contained in the brochure was useful, the more likely the brochure was to influence the decision to visit Frankenmuth. Third, it is interesting to find that the variable NIGHT SPENT is not statistically significant in this model. Risk and involvement theory would have predicted that if respondents spent one or more nights in Frankenmuth, their involvement would have been high and they would have spent more money and hence would face higher risk which they would want to reduce by obtaining more information. Results here 100 suggest that perceived risk and respondent involvement were not high in the decision to visit Frankenmuth. Fourth, age and income were not significantly related to effectiveness of the brochure. Information seeking behavior appears not to be affected by how old or how rich respondents are. The goodness of fit test shows that the model explains less than 20 percent (R?=.17) of the variation in the dependent variable EFFECTIVENESS, with a standard error of 1.75. The hypothesis that no linear relationship exits between the dependent variable and the set of independent variables, i.e., R?=0, was rejected (F=13.81, Significance of F=.0000). In other words, the multiple regression model built for this study fits the data but only to a modest degree. In multiple regression, it is sometimes desirable to assign relative importance to each independent variable. For the regression model in this study, there are two possible approaches, depending on which of the following questions is asked: 1) How important are the five independent variables in the model when each one is used alone to predict effectiveness of the brochure? 101 2) How important are the five independent variables in the model when they are used to predict effectiveness of the brochure along with other independent variables in the regression equation? The first question is answered by looking at the correlation coefficients between each individual independent variable and the dependent variable. The larger the absolute value of the correlation coefficient, the stronger the linear association. Table 4.14 shows that EFFECTIVENESS correlates more highly with USEFULNESS and PRIOR EXPERIENCE (.334 and .212) than with INCOME, AGE and NIGHT SPENT (.111, .046 and .038). The answer to the second question is considerably more complicated. First, when the independent variables are intercorrelated, the unique contribution of each is difficult to assess. Second, even when multicollinearity can be ruled out, any statement about an independent variable's importance is contingent upon the other variables in the equation. Thus, it is inappropriate to interpret the partial regression coefficients B's as indicators of the relative importance of variables, the actual magnitude of the coefficients depends on the units in which the variables 102 are measured. One way to make these partial regression coefficients somewhat more comparable is to look at the beta weights, which are the coefficients of the independent variables when all variables are expressed in standardized (Z score) form (see Table 4.14). It appears from Table 4.14 that USEFULNESS and PRIOR EXPERIENCE have the largest beta weights. In summary, the multiple regression model presented above suggests that: 1) The perceived usefulness of the information contained in the brochure appears to be the best predictor of the effectiveness of the brochure, 2) Prior visit(s) to Frankenmuth also appears to be important, 3) Age, income and overnight staying were not significantly related to effectiveness of the brochure, and 4) The model was statistically significant and fits the data, although only about 17% of the variation in effectiveness was accounted for by the model. 4.7 Summary: Answers to Research Questions In this section, findings related to the hypotheses introduced at the end of chapter two will be discussed. The first hypothesis was that the more respondents use 103 brochures, the more likely they will report a higher level of satisfaction on a particular trip. The findings from this study do not present strong evidence to support this hypothesis. It was found that consulting the brochure during the visit to Frankenmuth correlated significantly, though weakly, with higher ratings of reported trip satisfaction and with respondents who were more likely to recommend Frankenmuth to someone they know after their trip. Hypothesis 2 states that brochures are more likely to have an impact on those who have not visited the destination than on those who have in terms of influencing their trip decision. The findings reveal that there exists a significant difference in responses between those who had visited and those who had not in terms of the degree to which the brochure influenced trip decision. The findings support accepting the second hypothesis. Hypothesis 3 states that the more lapsed time there is between when a trip is made and when trip expenditures are reported, the more likely respondents will underestimate trip expenditures. The findings do not support this hypothesis. Instead, the findings suggest that respondents's recall of their trip expenditures was 104 relatively insensitive to lapsed time. Hypothesis 4 states that respondents with aided questionnaires are more likely to report higher ratings of impressions and influences of brochures than respondents with unaided questionnaires. The findings reveal that aided questionnaires significantly affected respondents' rating of these variables. Aided respondents are more likely to report higher mean ratings than did unaided respondents. This hypothesis is, thus, supported. Hypothesis 5 states that there exists a significant linear relationship between effectiveness of the brochure and a set of independent variables including prior experience, income, perceived usefulness of information contained in brochures, whether or not respondents spent a night in Frankenmuth and age. According to the findings, perceived usefulness of information contained in the brochure appears to be the best predictor of the effectiveness of the brochure. Prior experience is another important variable in the regression equation in predicting the effectiveness of the brochure. Overall, the regression model explains less than 20 percent of the variation in predicting the effectiveness of 105 brochures in terms of influencing trip decisions. The model, however, is statistically significant and therefore fits the data but only to a modest degree. CHAPTER FIVE CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions The underlying premise of this paper was that the value (effectiveness) of a brochure rests on its efficiency in reaching consumers and its saliency in addressing their needs and interests. Based on the definitions of the word "effectiveness" developed for this study, the brochure under study can be said to be very effective in the following areas: I Ninety-one percent of respondents received the brochure after requesting information. I Ninety-nine percent of respondents who visited Frankenmuth after requesting information read the brochure that was sent to them. I Of the other half of respondents who did not visit Frankenmuth after requesting information, 99% reported having read the brochure. I Fifty-percent of respondents visited Frankenmuth following requesting information. 106 107 I Seventy-eight percent of respondents who visited Frankenmuth consulted the brochure during their visit. I The attributes and features of the brochure under study were rated relatively high. However, the brochure appears to have had minimal to moderate effect on changing respondents' total trip spending and a moderate effect on influencing respondents' decision to visit Frankenmuth. The brochure appears to produce other benefits as well. Respondents who consulted the brochure while visiting Frankenmuth reported a better-than-expected trip experience and were more likely to recommend a visit to Frankenmuth to others after their trip. However, use of the brochure on site did not increase respondents' likelihood for repeat visit(s) or produce higher ratings of Frankenmuth as a tourist destination. Factors affecting effectiveness of the brochure include prior visit(s) and perceived usefulness of the brochure. Respondents with prior visit(s) to Frankenmuth tended to report that their decision to visit Frankenmuth was less influenced by the brochure. Respondents who perceived that information contained in the brochure was useful were more 108 likely to state that the brochure was influential in their trip decision to Frankenmuth. Age, income and spending a night at Frankenmuth did not contribute significantly to the effectiveness of the brochure in the trip decision process. Methodological findings from this study reveal that: Aided questionnaires appear to induce higher ratings from respondents on questions that are associated with impressions and perceived usefulness of the brochure. Aided questionnaires did not affect respondents' recall of trip expenditures. Length of lapsed time between the time of visit and the time of recalling trip expenditures does not significantly affect respondents' reported trip expenditures. In other words, respondents's recall of their trip expenditures was relatively insensitive to lapsed time. Marketing Recommendations Recommendations for destination marketers which flow the findings of this study are presented below. W: The findings from this study reveal that nearly all recipients read the brochures 109 they received, and the majority who decided to visit consulted them on site during their visit. This points to a need to design a brochure that is user friendly, that is, the brochure will not only convey promotional messages but also provide information such as maps, business hours, parking areas and regulations, and other information one would expect to find in a travel guidebook. In fact, the need to produce a brochure that is user friendly has long been noticed. Hodgson (1993), in an attempt to drive this point home, quoted Ivie (1989), a senior executive from a major British tour operator, as saying: To my knowledge, virtually all research conducted in this area points to a consumer preference for one hotel per page, with all price and supplemental information on that same page. Yet all major tour operators fly in the face of this and produce separate price inserts which, by and large, are difficult to use, and are not exactly user friendly. Clearly, responding to requests for brochures is important because they are read--brochure inquirers are a most receptive audience for a destinations's marketing message. The strong tendency for inquirers to retain and bring brochures with them suggests an 110 opportunity to use them not only as tools to influence destination choice but also as devices to influence on site behavior. Maps and other destination information should be included to enhance the visitors' on-site experience and thereby the probabilities of longer stay, repeat visits and more favorable recommendations of the destination to friends and relatives. Innantixe_nffsrings: According to the findings of this study, seventy-one percent of the respondents had not made their final decision where to go for their trips. The implication of this is that there is considerable opportunity for destination marketers to influence and attract this large portion of the undecided potential travel population. Additional measures such as offering coupons, on-site prize drawing, and follow-up contacts could persuade the inquirer to make a trip to the destination. Na:ket_£ggnantatignz Through their request for information, inquirers identify themselves as high potential customers and ideal targets for aggressive marketing efforts. Study results suggest that brochures have different influences on first time and 5.3 rise 111 repeat visitors. Knowing this, a destination could design separate brochures to target the interests of these two market segments. Repeat visitors are likely to be more influenced by a brochure featuring new attractions. First time visitors may be more responsive to messages featuring a destination's more long standing and well known attractions. By investing minimal resources, destinations can simply provide targeted messages via brochures to enhance their influence on both first time and repeat visitors. Research Recommendations The two findings regarding research methodology give to the following recommendations: W W: For the time Span covered in this study (approximately 3-18 months), the findings of this study suggest that consistent trip expenditure estimates can be obtained over a fairly long span of time after trips are taken. These results do not discount the possibility of recall bias in post- trip expenditure studies but do indicate that it does 112 not escalate over the intermediate time frame explored in this study. In addition, trip expenditure recall appears not to be affected by aided questionnaires. When the purpose of a study is aimed mainly at obtaining trip expenditures, researchers do not have to be concerned about the possible impact of enclosing a brochure as a reminder of the trip. WM: Study results show that respondents with aided questionnaires tend to report higher ratings of their impressions and influences of the brochure. It is reasonable to assume that these higher ratings are more reflective of how respondents perceived the brochure when it was initially received. If a brochure is not enclosed with the questionnaire, it is unlikely that accurate measures of its impacts on respondents can be obtained. 5.4 Limitations 5.4.1 Validity of the Study The validity of a research study means the extent to 113 which the interpretation of the results of the study follow from the study itself (internal validity) and the extent to which the results may be generalized to other situations with other people (external validity). As was pointed out in the beginning of this paper, this research study dealt with only one type of brochure distribution, i.e. via mail to brochure-inquirers. These brochure-inquirers tend to be more motivated and therefore their responses to the questions may not represent situations in which brochures are acquired through other means such as those displayed in the lobbies of hotels, in convention centers, in rest areas, in travel agency offices, or in tourist information centers. Reasonable caution should, therefore, be exercised when applying the results to these different populations. Demographic, socioeconomic and motivational factors may be different from those of brochure-inquirers who call to request information. The generalizability of the results from this study is further limited by the nature of the destination and the characteristics of its customers. For example, the majority of respondents in this study did not travel long distances to reach Frankenmuth nor did they spend much time there. 114 The sample frame employed only included inquirers, who may or may not be reflective of all Frankenmuth visitors, and only those requesting information during September through March of 1992—93 and 1993—94, the period over which the brochure in question was distributed. 5.4.2 Measurement There has been considerable debate about the application of certain statistical methods to the scales from which data were obtained in this study. At the center of the debate is the issue of treating scales such as "semantic differential" as ordinal or interval. A conservative assumption is that the scale has the properties of ordinality. A more liberal assumption is that the scale cells have equal intervals, thus achieving the properties of an interval scale. Williams (1979) and Shavelson (1988) argue that the problem is not a question of whether these scales can be treated as interval and used in inferential statistics but a question of how much the researcher is willing to compromise between what he/she wants to achieve in terms of having confidence in the validity of results, and what he/she is willing to risk as error. 115 In this study, the measurement scales used for some of variables under study were scales from 1 to 7. On these scales 1 and 7 represent opposite extremes of a scale that have been labeled in each question; the values between 1 and 7 represent degrees between these extremes. In this paper, the liberal view of measurement was adopted and the scales used herein were assumed to be interval scales. It is worth while to note that most researchers in communication, psychology and education adopt this liberal view. 5.4.3 Response Bias The overall response rate obtained in this study was 49.7 percent which is above average for a mail survey. Still, it is not known if respondents may differ significantly from non-respondents, and time and budget did not permit an assessment of the possibility of nonresponse bias. A different kind of bias could also be present in this study's results, since it used incentives to boost response rate. This might induce respondents to provide more favorable responses to some questions than would have been the case were incentives not included. As in all survey 116 research in which all potential forms of bias can not be controlled for or accurately measured, reasonable care is advised in interpreting the results presented herein. Finally, the study design was developed to produce four homogeneous subgroups (92-93 aided, 92-93 unaided, 93—94 aided and 93-94 unaided); however, as analysis proceeded the desired homogeneity across subgroups was found not to have been achieved with respect to two important variables. The two 92-93 groups of respondents included more out-of—state residents who traveled longer distances and stayed longer than the 93-94 groups. These differences were especially problematic in analysis involving the relationship between lapsed time and trip expenditures. 5.5 Future Research As was discussed earlier, the methodology used in this study was designed to measure effectiveness of mailed brochures requested by inquirers, and it would need to be modified to accommodate other brochure distribution systems. For instance, readership was an essential and all important first step in measuring effectiveness for mailed brochures. However, in the case of displayed brochures, the definition 117 of effectiveness would necessarily include some measures of a brochure's ability to attract enough interest to convince passersby to select the brochure from the display, especially when presented with brochures for other destinations. Studying the process of brochure selection may reveal much information about interest in the destination and the types of people who pick up brochures and this would need to be incorporated in a study of effectiveness of a displayed brochure. Finally, two of the more interesting findings from this study merit further indepth research attention since, if they are verified in future research, they would have significant implications for destination marketers and the design of conversion studies. The first and most significant of these from a marketing perspective is that nearly 100% of those requesting information about Frankenmuth reported reading it whether or not they visited the destination. If verified in subsequent research, this would suggest that inquirers as a group are without equal in their receptivity for a destination marketer's message. Thus, inquiry response programs should be taken more seriously than they generally are and should be assigned 118 more in the way of financial support. However, in this era of junk mail, information overload and too little time, it is hard to accept the finding that 99% of respondents read the Frankenmuth brochure that they received. A host of factors could have combined to inflate respondents' reported readership such as the desire to please or not to appear uninformed. An alternative study design would probably be needed to assess actual readership. The second interesting finding from this study was that reported trip expenditures did not vary markedly with lapsed time between the trip and when expenditures were reported. It is generally assumed that recall problems increase as lapsed time increases resulting in increasing under reporting of expenditures the longer one waits to collect trip expenditure data from respondents. While the findings in this case suggest that the recall problem is less severe than is generally assumed, the sample drawn for this study, as noted earlier, lacked homogeneity across variables known to impact on visitor expenditures thereby complicating interpretation of the impact of lapsed time independent of other causal factors. Even without this limitation, results in this case only would apply to lapsed time of between 119 approximately three and eighteen months. It is probable that the recall problem is most severe before the lapsed time span covered in this study. Given the importance of this issue and the lack of definitive literature on it, a carefully designed study to assess the pattern and scope of the expenditure recall problem over varying spans of lapsed time would make a useful contribution to the literature involving both conversion and economic impact studies. LI ST OF REFERENCES LIST OF REFERENCES Baas, John M., Michael J. Manfredo, Martha E. Lee, and Deborah J. Allen. (1989). Evaluation of an informational brochure for promoting charter boat trip opportunities along the Oregon coast. Journal_of_lravel Research, 27(3): 35-37. Baker, William. (1993). The relevance accessibility model of advertising effectiveness. In Andrew A. Mitchell (ed), AdMertising_Exnosurel_Memor¥l_and_Choice. PP. 49-87. New Jersey: Lawrence Erlbaum Associates. Ballman, Gary, Jim Burke, Uel Blank and Dick Korte. (1984). Toward higher quality conversion studies: Refining the numbers game. Journal_of_Trayel_Research. 22(4): 28-33. Berger, D. (1986). Theory into practice: The FCB Grid. European_Research. 14(1): 35-46. Burke, J.F. and L.A. Lindblom. (1989). Strategies for evaluating direct response tourism marketing. Journal of_TraMel_Research. 28(2): 33-37. Burke, J.F. and Richard Gitelson. (1990). Conversion studies: Assumptions, applications, accuracy and abuse. Jonrnal_of_Trayel_Research. 28(3): 46-51. Cella, W.B. and J. Keay. (1979). Survey of bear information visitors received from park service. Annual_Bear Management_and_lncident_3epgxt, Yosemite National Park. Cherem, G. (1982). ShenangQ_Rixer_Lake1_Interpretation SI l . i E J . E E Experiment_Station, U.S. Corps of Army Engineers. Cohen, J.B. (1982). Involvement: Separating the state from its causes and effects (Working Paper #33). University of Florida. 120 121 Davidson-Peterson Associates. (1990). A study of the effectiveness of the 1990 Wisconsin tourism marketing program (unpublished). Prepared for Frankenberry, Laughlin & Constable, Inc. Edgell, David L. (1990). International_Tourism_Polic¥. NY: Van Nostrand Reinhold. Ellerbrock, Michael J. (1981). Improving coupon conversion studies. Jonrnal_of_Trayel_Research. 19(4): 37-38. Etzel, Michael J. and Russell G. Wahlers. (1985). The use of requested promotional material by pleasure travelers. Journal_of_Irayel_Research. 23(4): 2-6. Fazio, J.R. (1979). Communicating with the wilderness user. H'JiJ'E i E E . | S l' E J] . 28, University of Idaho. Fesenmaier, D.R. and C.A. Vogt. (1993). Evaluating the economic impact of travel information provided at Indiana welcome centers. Journal_of_Traxel_Research. 31(3): 33-39. Fesenmaier, D.R., C.A. Vogt and W.P. Stewart. (1993). Investigating the influence of welcome center information on travel behaviors. Journal_gf_Irayel Research, 31(3): 47-52. Frechtling, Douglas C. (1987). Assessing the impacts of travel and tourism--Measuring economic benefits. In J.R. Brent Ritchie and Charles Goeldner (eds), IranelL T0urism_and_Hanitalit¥_Research, pp. 325-31. New York: Wiley. Gardner, Burleigh B. (1981). A battle to persuade. Advertising_Age, 22(June): 52-55. Gerbner, George. (1972). Violence in television drama: Trends in symbolic functions. In George A. Comstock and Eli A. Rubinstein (eds), Television_and_Social_Behavior yQl1_l1_Media_CQntent_and_Control, pp. 128-187. Washington D.C.: US Government Printing Office. 122 Gilbert, D.C. and P. Houghton. (1991). An exploratory investigation of format, design, and use of U.K. tour operators' brochures. Jonrnal_of_Travel_Research. 30(2): 20-25. Gitelson, R.J. and J.L. Crompton. (1983). The planning horizons and sources of information used by pleasure vacationers. Jonrnal_of_Irayel_Research. 22(3): 2—7. Harley, Russell I. (1985). Deyeloning_Effectiye Communications_Strategy;;A_Benefit_Segmentation Approach. NY: Wiley. Hodgson, P. (1993). Tour operator brochure design--Research revisited. Iournal_of_Travel_Research, 32(1): 50—52. Holloway, J.C., and R.V. Plant. (1988). Marketing_for Tourism, London: Pitman. Howard, D.R., S.V. Lankford, and Mark E. Havitz. (1991). A method for authenticating pleasure travel expenditures. Journal_of_Traxel_Research. 29(4): 19-23. Ivie, K. (1989). Market research: A luxury now for tour operators? Presentation given at Market Research seminar, February. (Citation by P. Hodgson, 1993. See above reference) Jeffres, Leo w. (1986). Mass_Media_2rocesses_and_Effects. Illinois: Waveland Press. Kapferer, J.N., and G. Laurent. (1985). Consumer involvement profiles. Journal_of_Exnerimental_Es¥cholog¥: Learningl_Memorx_and_CoSnition. 10: 649-662. Klapper, Joseph. (1960). The_Effects_of_Mass_Communication. NY: Free Press. Krumpe, E.E. and P.J. Brown. (1982). Using information to disperse wilderness hikers. Journal_of_£orestry, 79: 92-94. 123 Lime, D.W. and R.C. Lucas. (1977). Good information improves the wilderness experience. Naturalist. 28: 18-20. Mak, J., James Moncur, and David Yonamine. (1977). How or how not to measure visitor expenditures. Journal_of Travel_Research. 16(3): 1-4. McDonald. Colin. (1993). How_AdMertising_florks;;A_Reyiew_of CnrranL_Ihinking. The Advertising Association in association with NTC Publications LTD. Messmer, D.J. and R.R. Johnson. (1993). Inquiry conversion and travel advertising effectiveness. Journal_of_1rayel Research. (Spring): 14-21. Middleton. V.T.C. (1988). Marketing_in_Traxel_and_Tourism. Oxford: Heinemann. Mill, R.C., and A.M. Morrison. (1992). Ihfi_IQnIifim_S¥fiLQm;; An_Interucthy_Iext, (2nd Ed.). New Jersey: Prentice Hall. Millman, A. and Abraham Pizam. (1995). The role of awareness and familiarity with a destination: The Central Florida case. Journal_of_Trayel_Research. 33(3): 21-27. Muha, S.L. (1976). Evaluating travel advertising: A survey of existing studies. Paper presented at the Fifth Annual Educational Seminar for State Travel Officials. Lincoln, Nebraska, 4 October. Neale, M.C., Lindon J. Eaves, Kenneth S. Kendler, Andrew C. Heath and Ronald C. Kessler. (1994). Multiple regression with data collected from relatives: Testing assumptions of the model. Multivariate_3ehayigral Research. 29(1): 33-61. NorUSiS: Mariga J. (1992)..SESSLPCi_Base.$xstem_Userls_Guide lyersion_5101. SPss/PC+. 124 Purdue, Richard. (1984). Segmenting state travel information inquirers by timing of the destination decision and previous experience. Journal_of_Trayel_Researeh. 23(3): 6-11. Stynes, Daniel J. and E. Mahoney. (1989). Measurement and analysis of recreational and travel spending. Ahsrrasrs l2a2_Sympos1um_on_Leisure_Researeh. Alexandria. VA: National Recreation and Park Association. Ritchie, Brent J.R. and C.R. Goeldner (eds). (1987). Irayelr Managers_and_Researehers. NY: Wiley. Roggenbuck, J. W. and D. L. Berrier. (1982). A comparison of the effectiveness of two communication strategies in dispersing wilderness campers. J9nrnal_gf__Lsisnre Resaarsh, 14: 77- 89. Ronkainen, I.A. and A.G. Woodside. (1987). Advertising conversion studies. In J.R. Brent Ritchie and Charles R. Goeldner. (eds). Traxel1_Tourism1_and_Hosnitalit¥ Research, pp. 481—487. New York: Wiley. Rylander II, Roy G., Dennis E. Propst, and Terri R. McMurtry. (1995). Nonresponse and recall biases in a survey of traveler spending. Jaurnal_gf_Trayel Rsssarsh, 33(4): 39-45. Schmoll, G.A. (1977). Ignrimm_2rgmgrign. London: Tourism International Press. Schroeder, J.J and L.M. Kreul. (1986). The effectiveness of travel information in converting the "inquirer" into a "traveler" to Indiana (Unpublished). Purdue University. Prepared for Indiana Department of Commerce. Shavelson. R.J. (1988). Statistieal_Reason1ng_Eor_the Behaxioral_Seienees. (2nd ed). MA: Allyn and Bacon. 125 Siegel, William and William Ziff-Levine. (1990). Evaluating tourism advertising campaigns: Conversion vs. advertising tracking studies. Janrnal_gf_Iraysl Rassarsh, (Winter): 51-54. Silberman, Jonathan, and Mark Klock. (1986). An alternative to conversion studies for measuring the impact of travel ads. Jenrnal_of_Iravel_Researeh. 24(4): 12-16. Sudman, Seymour and Norbert Schwarz. (1989). Contributions of cognitive psychology to advertising research. Journal_of_Adyertisin9_Researeh. 29(3): 43-53. Taylor, J.W. (1974). The role of risk in consumer behavior. Journal_of_Market1ng. 38(April): 54-60. Tierney, P., and Glenn Haas. (1988). Colorado welcome centers: Their users and influence on length of stay and expenditures. Ft. Collins: Department of Recreation Resources and Landscape Architecture. Tull, 0.3. and Del I. Hawkins. (1990). W Measurement_and_Method. NY: Macmillan. U.S. Travel Data Center. (1989). Suryex_of_State_Trayel Offices_1288;12a&. Washington, DC: USTDC. Vaughn, D. (1980). How advertising works: a planning model. Journal_of_Adyertisin9_Researeh. 20(5): 27-33. Vaughn, D. (1986). How advertising works: A planning model revisited. Jenrnal_of_Adyertis1ng_Research. 26(1):57- 66. Wicks, Bruce E., and Michael A. Schuett. (1991). Examining the role of tourism promotion through the use of brochures. Igurism_Managemanr, 12(4): 301-12. WilliamS. F. (1979). Reasoning_w1th_Statisties. NY: Holt. Rinehart and Winston. 126 Woodside, A.G. (1990). Measuring advertising effectiveness in destination marketing strategies. MW]. Research. (Fall): 3-8. APPENDICES APPENDIX A 1993-94 SURVEY OF FRANKENMUTH INQUIRERS 1 9 93 - 94 SURVEY OF FRANKENMUTH INQUIRERS Thank you very much for participating in our study! Please answer the following questions sequentially, unless directed to skip over a block of one or more questions by the statement "GO TO QUESTION. . . ." Please follow these "GO TO" directions very carefully. 1. How did you obtain the phone number or address of Frankenmuth Chamber of Commerce/Visitors Bureau? Newspaper article ] Friend/relative/co-worker [ Newspaper ads Frankenmuth area business [ Magazine ads At a travel show [ [ [ Michigan Travel Bureau [ [ Magazine article [ [ [ ] ] Telephone directory ] Directory assistance ] HHHO—‘f—‘H HHL—DA—IHI—J L—ll—ll—Jh—lL—l Travel agent Brochure Radio ads Regional tourist association Television ads Other; please specify: 2. Prior to requesting information on / / , had you ever visited Frankenmuth? [ ] Yes [ ] No .4 GO TO QUESTION 3 2a. Approximately when did the most recent of these visits occur? (Please provide dates for up to three visits.) Month Year Month Year Month Year 2b. Did you request information from Frankenmuth Chamber of Commerce/Visitors Bureau to help plan any of these visits? [ ] Yes [ ] No [ ] Don't remember GO TO QUESTION 3 3. Was the information you requested from Frankenmuth Chamber of Commerce/Visitors Bureau on / / for YOUR USE or for USE BY SOMEONE ELSE? [ ] For my use [ ] For use by someone else —+ GO TO QUESTION 28 4. When you requested information, was a visit to Frankenmuth. [ ] Already decided upon [ ] Being considered [ ] Not being considered, but desired information anyway 1127 128 Did you visit Frankenmuth AFTER REQUESTING information on / / from Frankenmuth Chamber of Commerce/Visitors Bureau? [ ] Yes-4 GO TO QUESTION 6 ON NEXT PAGE []No l 51 Are you considering a visit to Frankenmuth? [ ] Yes [ ]No ... GO TO QUESTION 5c 1 5b. When will this visit most likely occur? [ ] Within the next 3 months [ ] 4-6 months in the future [ ] 7-9 months in the future [ ] 10 or more months in the Sc. Did you receive any information from Frankenmuth Chamber of CommerceNisitors Bureau in response to your request? [ ] Yes [ ]No _. GO TO QUESTION 28 5d Was the information you received the information you requested? [ ]Yes [ ]No [ ]Partially 5e. Approximately when did you receive the information? / / Month Day Year 5f. Did you read the brochures that were sent to you? []Yw []N04 59. What did you do with them? Q 'onSShand [ ] Kept them for future reference 5i scales from 1 to 7 [ ] Gave them to someone are shown On these [ ] Discarded them scales 1 and7repre— [ 10th“ sent extremes that have Go To QUESTION 2 8 been labeled in the case of each question; the values between 1 and 7 represent degrees between these extremes. Please answer each question by circling ONE number in the case of each scale. 511 To what extent did the brochures decrease or increase your interest in visiting Frankenmuth? l 2 3 4 5 6 7 Greatly decreased my interest Greatly increased my interest 5i. How would you rate the overall quality of the brochures? l 2 3 4 5 6 7 Terrible Excellent GO TO QUESTION 28 2129 6. Approximately when did you depart on your first trip involving a visit to Frankenmuth after requesting information from Frankenmuth Chamber of Commerce/Visitors Bureau on / / ? / / Month Day Year NOTE: IN THE FOLLOWING QUESTIONS, "THIS TRIP" REFERS TO THE TRIP THAT BEGAN ON THIS DATE 7. Prior to this trip, did you obtain any information about Frankenmuth from any sources other than Frankenmuth Chamber of Commerce/Visitors Bureau? [ ] Yes [ ] No GO TO QUESTION 8 7a. What other source(s)? [ ] Newspaper article [ ] Travel show [ ] Newspaper advertisement [ ] Travel agent [ ] Magazine article [ ] Michigan Travel Bureau [ ] Magazine advertisement [ ] Regional tourist association [ ] Radio advertisement [ ] Telephone directory [ ] Television advertisement [ ] Brochure [ ] Friend/relative/co-worker [ ] Other; please [ ] Frankenmuth area business Specify: 8. Was Frankenmuth the PRIMARY DESTINATION of this trip? [ ] Yes [ ] NO'* 8a. What was the primary destination? 9. What was the PRIMARY PURPOSE of this trip? [ ] Recreation/pleasure [ Business or convention/meeting [ ] Combined business and recreation/pleasure [ ] Other; please specify: 10. Did you spend any NIGHTS AWAY FROM HOME on this trip? [ ] Yes [ ] No —+ GO TO QUESTION 11 l 10a. Did you spend any nights IN FRANKENMUTH on this trip? [ ] Yes [ ] No 1 10b. In what cities did you spend the night? GO TO QUESTION 11 10c. How many nights did you spend IN FRANKENMUTH? 10d. Where were these nights IN FRANKENMUTH spent? (Check all that apply) [ ] Hotel or motel [ ] Friend's/relative's home [ ] Bed & Breakfast [ ] Second home you own [ ] Campground [ ] Other; please specify: 130 11. When you visited Frankenmuth on this trip, were you a member of an organized tour group, such as a motor coach tour group? [ ] Yes [ ] No _. 11a. Did anyone accompany you on this trip? [ ] Yes [ ] No GO TO QUESTION 12 11b. How many persons (not including yourself)accompanied you? 11c. Were these persons: ] Friends ] Relatives ] Friends and relatives ] Business associates [ [ [ [ We'd also like to know approximately how much (if any) money your "spending unit" spent in Frankenmuth on this trip. A "spending unit" is an individual who purchases things for himself or herself, or a group that purchases things as a group. An individual who pays his or her own expenses is one spending unit whether he or she is traveling alone or with others. Two married couples traveling together are two spending units if they purchase things separately, although they may split some expenses. 12. While you were in Frankenmuth on this trip, did you or any other members of your SPENDING UNIT spend any money? [ ] Yes [ ] No —0 GO TO QUESTION 13 12a. How many persons, including yourself, were in your SPENDING UNIT on this trip? 12b. Approximately how much money did your SPENDING UNIT spend on this trip IN FRANKENMUTH on each of the following items? PLEASE DO NOT LEAVE ANY SPACES BLANK; WRITE "0" TO INDICATE NO EXPENDITURES ON A GIVEN ITEM. INCLUDE ONLY EXPENDITURES MADE IN FRANKENMUTH. EXCLUDE FEES PAID TO MOTORCOACH OPERATORS. Lodging .................................................. $ Camping fees ............................................. $ Gifts, crafts, souvenirs, clothing, and/or specialty food items ............................................. $ Grocery and convenience store food and beverages ......... $ Restaurant and bar meals and drinks ...................... $ Gasoline, oil, repairs, and other vehicle-related items..$ Rental fees (for golf carts, cross-country skis, etc.)...$ Guided tours ............................................. $ All other items .......................................... $ TOTAL .................................................... $ GO TO QUESTION 13 131. 13. Was your experience in Frankenmuth on this trip. [1 [l [l [1 [1 Much worse Somewhat About Somewhat Much bet- than ex— worse than what you better than ter than pected expected expected expected expected 14. How likely are you to visit Frankenmuth again? [1 [l [l [l [1 Certain Unlikely Uncertain whether Likely to Certain to to not to visit I will visit visit again visit again visit again again again 15. After this trip, did you recommend a visit to Frankenmuth to anyone? [ ] Yes [ ] No [ ] Don't remember 16. Overall, how would you rate Frankenmuth as a tourist destination on a scale from 1 to 7, where 1 is "terrible" and 7 is "excellent"? Please circle one number on the scale below. 1 2 3 4 5 6 7 Terrible Excellent Now we'd like to learn a little more about your request for information from Frankenmuth Chamber of Commerce/Visitors Bureau on / / 17. Did you receive any information in response to your request? [ ] Yes [ ] No GO TO QUESTION 28 17a. Was the information you received the information you requested? [ ] Yes [ ] No [ ] Partially 17b. Approximately when did you receive the information? / / Month Day Year 17c. Did you receive information from Frankenmuth Chamber of Commerce/ Visitors Bureau BEFORE this trip or AFTER this trip? [ ] Before this trip GO TO QUESTION 18 [ ] After this trip GO TO QUESTION 28 18. Did you read the brochures that were sent to you? [ 1 Yes i ] N°-* 18a. What did you do with them? [ ] Kept them for future [ ] Gave them to reference someone [ ] Discarded them [ ] Other 19. Did you consult the brochures while visiting Frankenmuth on this trip? [ ] Yes [ ] No [ ] Don't remember 132 In Questions 20 through 27, scales from 1 to 7 are shown. On these scales 1 and 7 represent extremes that have been labeled in the case of each question; the values between 1 and 7 represent degrees between these extremes. Please answer each question by circling ONE number on each scale. 20. How useful was the information in the brochures? 1 2 3 4 5 6 7 Not at all Extremely useful useful 21. To what extent were the brochures interesting to read? 1 2 3 4 5 6 7 Very un- VerY interesting interesting 22. To what extent were the brochures attractive in design? 1 2 3 4 5 6 7 Very un- Very attractive attractive 23. To what extent did the brochures decrease or increase your interest in visiting Frankenmuth? 1 2 3 4 5 6 7 Greatly Greatly decreased increased my interest my interest 24. To what extent did the brochures influence your decision to visit Frankenmuth? 1 2 3 4 5 6 7 Had no Actually influence caused me at all to visit 25. To what extent did the brochures cause you to spend more money in Frankenmuth on this trip than you would have otherwise? 1 2 3 4 5 6 7 Not at A great all deal 26. Based on your experience on this trip, how accurate was the information in the brochures? 1 2 3 4 5 6 7 Completely Completely inaccurate accurate 27. How would you rate the overall quality of the brochures? 1 2 3 4 5 6 7 Terrible Excellent 133 Please answer the remaining questions so that we will be able to develop a profile of the types of people who request information from Frankenmuth Chamber of Commerce/Visitors Bureau. Your responses will, of course, remain strictly confidential; they will simply be combined with those of other respondents to compute percentages and averages. 28. What is the ZIP CODE of your permanent residence? 29. What is your gender? [ ] Male [ ] Female 30. What is your present employment situation? [ ] Working full time [ ] Homemaker [ ] Other; please [ ] Working part time [ ] Retired specify: [ ] Temporarily unemployed [ ] Student 31. Please circle the highest year of formal schooling you have completed. 12345678 1234 1234 1234567+ Grade School High School Undergraduate Graduate college college education education 32. In what year were you born? 33. How many people reside in your household (including yourself)? 34. How many EMPLOYED persons age 18 or older reside in your household? 35. Do any children under age 18 reside in your household? [ ] Yes [ ] NO —o GO TO QUESTION 36 l 35a. What is the AGE of the oldest child living at home? 35b. What is the AGE of the youngest child living at home? GO TO QUESTION 36 36. What was your total 1993 household income before taxes? [ ] Under $15,000 [ ] $35,000 to $49,999 [ ] $120,000 to $134,999 [ ] $15,000 to $19,999 [ ] $50,000 to $74,999 [ 1 $135,000 to $149,999 [ ] $20,000 to $24,999 [ ] $75,000 to $104,999 [ 1 $150,000 to $299,999 [ ] $25,000 to $34,999 [ ] $105,000 to $119,999 [ ] $300,000 or more THANKS FOR YOUR HELP! Please return your questionnaire in the postage-paid envelope provided. If you misplaced this envelope, please return the questionnaire to: 1993-94 Survey of Frankenmuth Inquirers Travel, Tourism, and Recreation Resource Center 172 Natural Resources Building Michigan State University East Lansing, MI 48824-1222 APPENDIX B COVER LETTER FOR AIDED QUESTIONNAIRE 134 Date (Name and address inserted by computer) Dear (name inserted by computer): We are conducting a survey of people who have requested information about Frankenmuth, Michigan from the Frankenmuth Chamber of Commerce/Visitors Bureau. Your name was randomly selected from a list of such inquirers. We are very interested in determining if you received the information you requested, and if so, whether you found it to be useful. Would you please take a few minutes to complete the enclosed questionnaire? Your candid views will be greatly appreciated. While your participation in this study is, of course, voluntary, it is essential that we obtain cooperation from everyone we contact so that our relatively small sample will be representative of all Frankenmuth inquirers. Please complete the questionnaire even if you never received any information from the Frankenmuth Chamber of Commerce/Visitors Bureau and/or did not visit Frankenmuth after requesting information. To express our sincere appreciation for your help, we've enclosed "Dining Dollars" and discount coupons from 135 many Frankenmuth businesses. These coupons may be used on your next visit to Frankenmuth. Please accept them as tokens of our gratitude for your time and effort. Additionally, upon return of your questionnaire you will earn a chance to win one of the following gifts: 1. A Frankenmuth overnight getaway package, including a two-night stay and meals at the Bavarian Inn and Zehnder's Restaurant, a Bronner's CHRISTmas Wonderland Gift Certificate, and passes for many of Frankenmuth's tours and activities. 2. A "Taste of Frankenmuth" gift basket. 3. A signed and numbered print of the famous Frankenmuth Holz-Brucke (covered bridge). Please follow the instructions in the questionnaire very carefully since they direct you to only those questions that pertain to your own experience. Your responses will remain strictly confidential and will be used only for the purpose of evaluating the inquiry servicing and promotional efforts of the Frankenmuth Chamber of Commerce/Visitors Bureau. Many of the questions in the questionnaire ask you about the brochure that were sent to you by the Frankenmuth 136 Chamber of Commerce/Visitors Bureau. To help you recall the publication, we've enclosed an additional copy of it. You're welcome to keep it. After you have completed your questionnaire, please return it to us in the enclosed postage-paid envelope. Again, your assistance will be greatly appreciated. Sincerely, Donald F. Holecek Professor and Director Enclosures (IlMilli(Illlllllllllll(llllllllHlllllllH(I