INFORMATION TO USERS The most advanced technology has been used to photo­ graph and reproduce this manuscript from the microfilm master. UMI film s the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may b e from any type of computer printer. The quality of th is reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are m issing pages, these w ill be noted. Also, if unauthorized copyright m aterial had to be removed, a note w ill indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are re­ produced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. These are also available as one exposure on a standard 35mm slide or as a 17" x 23" black and w h ite photographic print for an addi onal charge. Photographs included in the original manuscript have been reproduced xerographically in th is copy. H igher quality 6" x 9" black and w hite photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. Un iversity Microfilm s I n te r n a tio n a l A Bell & How ell In fo rm atio n C o m p a n y 3 0 0 N o rth Z e e b R o a d , A n n Arbor, Ml 4 8 1 0 6 - 1 3 4 6 US A 313/76 1 -4 7 0 0 800/521-0600 O rder N u m b er 8923839 Identification an d unique characteristics of M ichigan’s recreational tourism m arket Chung, Win-Jing, Ph.D. Michigan State University, 1989 C o p y rig h t © 1988 b y C h u n g , W in -Jin g . A ll rig h ts rese rv e d . 300 N. Zeeb Rd. Ann Arbor, MI 48106 IDENTIFICATION AND UNIQUE CHARACTERISTICS OF MICHIGAN’S RECREATIONAL TOURISM MARKET By W in-Jing Chung A DISSERTATION Subm itted to Michigan State University in partial fulfillment of the requirem ents for the degree of DOCTOR OF PHILOSOPHY D epartm ent of Park and Recreation Resources 1988 ii ABSTRACT IDENTIFICATION AND UNIQUE CHARACTERISTICS OF MICHIGAN’S RECREATIONAL TOURISM MARKET By Win-Jing Chung Expansion of a tourism market depends to a large extent on promotion. A common vehicle used in promotion is advertising designed to reach potential tourists. Since forecasting can help to predict who will travel in Michigan, the development of forecasting tools is essential for effective advertising to potential tourists. This study tests five types of variables which are assumed to be related to the tourist’s choice of Michigan as a trip destination. These variables are: (1) travel patterns (travel mileage, duration, etc.): (2) tourists’ socioeconomic background (age, income, state origin, etc.): (3) travel information from various type of media (radio, television, billboards, etc.); (4) tourists’ concerns for the trip (clear air, easy access, winter fun, etc.): and (5) tourists’ images of Michigan (nice place, friendly people, good restaurants, etc.). A forecasting tool called Linear Discriminant Analysis (LDA) is applied to investigate the effectiveness of using these variables to predict the propensity of tourists’ choosing Michigan as a trip destination. More than 8,000 records of trip information, collected by the Great Lakes Travel Monitor Study during 1985, were analyzed. Profiles of Michigan tourists were constructed. The factors significant to the choice of Michigan as a trip destination were determined. The relationships between the choice of traveling in Michigan and impacting factors were constructed. Based on these relationships, realistic targets for tourism promotion In Michigan can be established and used to predict Michigan’s tourist market. Results of the study indicate various factors are more significant In Michigan tourism than in non-Michigan tourism: married tourists, recreational vehicle owners, people of higher education, weekend tourists, and group trip participants. The chances of taking a Michigan trip were enhanced by travel information obtained from television, radio, and billboards. Through LDA, all five types of variables show evidence of their usefulness in predicting Michigan trips. In some cases, the percentage of correctness in predicting Michigan trips using a single factor such as trip mileage, duration, or season, exceeded 75%. Furthermore, using these factors together in LDA, the correct prediction rates of 83.5%, 61.4%, and 72.6% for predicting Michigan, non-Michigan, and overall trips, respectively were achieved. In conclusion, this study presents an operative approach to predicting Michigan trips. With better quality data, the effectiveness of LDA in identifying Michigan trips should be greater. Thus, using LDA and triprelated information can be considered a practical method to assist in increasing effective decisions regarding Michigan’s tourism market. C o p y rig h t b y W in-Jing C hung 1988 To My p aren ts, C hung a n d H ou My p a re n ts in law, la u a n d J i n My u n cle a n d a u n t. Dr. C hung a n d Dr. Hou My cousin, Shiao-C hung a n d Bi-Wu And m y wife, H uey-C hu vi ACKNOWLEDGEMENTS There are so many people who deserve my thanks for making this research possible. Though it is impossible to mention all of them here, the people who were especially helpful and encouraging throughout the project are listed as follows. I wish first and foremost to thank Dr. Donald F. Holecek who was always helpful, friendly, and very good at giving suggestions. This research couldn’t have been finished without his help. Many thanks are also due to Dr. Daniel E. Chappelle and Dr. Roger E. Hamlin for taking time to review this dissertation and share their extensive research knowledge. Three people deserve special thanks for advising me to keep on the right track and keep the hope alive, these people are Dr. Lewis W. Moncrief, Dr. Wayne Nierman, and Dr. Linda Nierman. They really gave me lots of good advice. Special thanks go to Mr. David Morris at the Travel Bureau, Michigan Department of Commerce for his help with obtaining the information on the history of original data used in this research. Mr. Robert Slana at Park and Recreation Resources Department was a great help in getting the original data set up on mainframe. The data downloading job could not have been accomplished without his cooperation. Gratitude is also due to the following friends: Dr. Erwin B. Sober, Sandy Bonesteel, Steven Leung, Dan Greg, and Micld Horst. Dr. Sober loaned me his computer before I could afford to buy one. The rest of them gave me their excellent editing of chapters. Sandy Bonesteel is a art student who greatly improved the sense of contents: Steven Leung is an engineering student who offered the helpful criticism; Mr. Dan Greg is a retired editor who submitted his professional opinions; and Micki Horst was an BFNEP colleague who added numerous good points. They all gave me their best service. 1 would also like to thank Expanded Food And Nutrition Education Program (EFNEP), Cooperative Extension Service program for allowing me to work as a computer consultant which helped me to pay the bills and keep this research going during the past two years. It has been a lot of fun to work with so many fine people in EFNEP such as Kendra, Barb, John, Kim, Debbie, all the student employees, and all the directors, home economists, and secretaries over each EFNEP county. I owe my parents and other family members many thanks for their love, patience, and understanding. My uncles Dr. Chung and Dr. Hou, cousin Shiao-Chung and sister in law Bi-Wu who gave me a hand when my back was against the wall with financial trouble. Especially I would like to thank my wife for her sacrifice. She worked a lot to help keep the budget balanced during the past years. She also make the purchase of my first computer possible. At last, I deeply thank God for all his blessings and for letting me know that the Ph.D. program is finally coming to an end. viii TABLE OF CONTENTS ABSTRACT iii DEDICATION vi ACKNOWLEDGEMENTS ........................................................................... vii LIST OF TABLES CHAPTER I. INTRODUCTION Objectives Assumptions Hypotheses Constraints ..................................................................... 1 ............................................................................ 8 ....................................................................... 9 ............................................................................. 10 ..................... 14 H. LITERATURE REVIEW ....................................................... 16 Tourism Definitions .................................... Michigan’s Tourism Market ................................................. The Significance of Tourism in Michigan ........................... Michigan’s Tourism Challenges .......................................... 16 19 20 21 Vigorous Competition .................................................... 21 Inadequate Market Information ................................... 22 Tourism Marketing ............................................................... 23 Emergence of The Concept of Travel Motivations ...... Factors in Attendance at Recreation Areas ............... Advertising Efficiency .................................................... Tourism Market Allocation ........................................... Linear Discriminant Analysis Applications 23 25 26 28 ....................... 30 Comparison of Linear Discriminant Analysis And Other Modeling Techniques ................................. 36 The Use of LDA, Regression Analysis. Factor Analysis, and Cluster Analysis ................... 37 Comparison between LDA and Multiple Regression .. 40 Comparison of LDA to Factor Analysis, and Cluster Analysis ............................................... 42 ix The Mechanics of Discriminant Analysis ......... EH. METHODS 44 48 Data Collection ................................................................... Data Preparation ................................................................. Research Variable Selection................. ........................................ .................... Research Design 49 50 52 55 Profiles of Regional Traveling Patterns ............................. Profile of The Tourism Market: Unique Characteristics with Descriptive Statistics ........................................ Test Variable Independence with .................................................... Chi-square Design Test of Differences between Michigan And NonMichigan Tourism Marketswith T-test .................... Developing The Prediction Functionswith LDA ........... 56 IV. ANALYSIS RESULTS AND FINDINGS ........... 57 58 58 58 64 The Regional Travel Market Profile ..................................... 65 The Michigan Recreational Tourism Market ...................... 71 Hypothesis Tests ........................................ 85 Hypothesis Hypothesis Hypothesis Hypothesis 1 2 3 4 86 93 99 103 Linear Discriminant Analysis Results: Using Individual Variable Sets in LDA to Predict Michigan Trips ..................................104 Using Trip Pattern Variables to Predict The Propensity to Travel in Michigan .......................105 Using Socioeconomic Variables to Predict The Propensity to Travel in Michigan ..................... 116 Using Travel Information Variables to Predict The Propensity to Travel in Michigan .......................119 Using Trip Attribute Variables to Predict The Propensity to Travel in Michigan .......................123 Using Michigan Image Variables to Predict The Propensity to Travel in Michigan .......................125 Using High Power Discriminators to Predict The Propensity to Travel in Michigan .......................129 x Comparing of Prediction Performance from Using Different Types of Variables to Predict The Propensity to Travel in Michigan ...................... 132 V. CONCLUSIONS ................................................................. 135 Research Problems..........................................................................136 Data Type Problem Data Quality Problem ..........................................................136 ..................................................... 136 Applications Limitations ...................................................... 138 Untested Variables Excluded Population ..........................................................138 ...................................................... 139 Implications And Issues for Future Research Conclusions ........................................... ..................140 141 Market Demand .............................................................. 142 Market Expansion ...........................................................143 Market Economic Impact .............................................. 144 Market Attributes ............. 144 Brand Loyalty............................................................................ 145 Michigan Images ............................................................145 Market Satisfaction .........................................................146 Potential Market Targets ................................................146 Useful Tourism Market Promotion Media ....................147 Market Forecast ................................................................ 147 LITERATURE CITED................................................................... APPENDICES A B. C. D. E. 150 .................................................................................... 161 Single Variable LDA Predicting Power Test Results Comparisons Using Different Research Units The SPSS/PC-f System Program The SPSS/PC+ Variables Recording Program Survey Questionnaire xi LIST OF TABLES 1. Comparison of Frequency of Use of Discriminant, Regression. Cluster, And Factor Analysis in Business Publications ............................................................... 39 2. The Yearly Distribution of Trips Data .................................... 64 3. Total And Distribution (in Percent) by Destination of Trips Generated by Each Market (Origin) State in The Great Lakes Travel Monitor Study Area ..................................................... 66 4. Total And Distribution (in Percent) by Origin of Each Vendor (Destination) .............................................................. State’s Travelers 67 5. Vendor (Destination) States’ Share of The Cold And Warm Season Travel Markets ....................... 69 6. Regional Tourism Market Comparison between Warm And Cold Season Tourism ....................................................... 70 7. Sources by State of Michigan’s Tourism Warm And Cold Season Travelers ....................................................... 73 8. Comparison of The Characteristics of Michigan and Non-Michigan Recreational Travelers Their Trips And Trip Ratings ................................................... 75 9. Comparison by Trip Purpose for Michigan And Non-Michigan Trips .......................................................... 77 10. Distribution of Michigan And Non-Michigan Trips by Month ......................................................................... 78 11. Distribution of Weekend Trips by Michigan And Non-Michigan .......................... 80 12. Group Trips between Michigan And Non-Michigan Trip Destinations ...................................................................... 80 13. Recreation Vehicle Ownership among Michigan And Non-Michigan Pleasure Travelers 81 14. Pleasure Travel Market Comparison by Gender - Michigan, Non-Michigan, And Total Region ............................ 82 15. Adult Pleasure Travel Market By Martial Status - Michigan, Non-Michigan, And TotalRegion .......................... 83 xii 16. Adult Pleasure Travel Market Comparison by Level of Education - Michigan, Non-Michigan, And Total Region ........................................................................ 84 17. Adult Pleasure Travel Market Comparison by Race Michigan, Non-Michigan, And Total Region ..................... 84 18. Adult Pleasure Travel Market Comparison by Income Michigan, Non-Michigan, And Total Region ..................... 85 19. T-tests of Differences between Michigan And Non-Michigan Recreation Trips for Selected .............................................................. Variables 87 20. Nature of The Relationship between Variables Found to Be Statistically Related (at alpha p < .05) ................................. 94 to Travel to Michigan for Recreation 21. Classification Matrix for Predicting Membership in Michigan or Non-Michigan Groups Using Trip ........................................... .........................113 Pattern Variables 22. Correlation Matrix for Trip Pattern Variables Which Entered The Michigan Traveler Analysis ............................ Discriminant 115 23. Classification Matrix for Predicting Membership in Michigan or Non-Michigan Groups Using Socioeconomic Variables ......................................................... 118 24. Correlation Matrix for Socioeconomic Variables which Entered in Michigan Traveler Discriminant Analysis........... ....................................................... 119 25. Classification Matrix for Predicting Membership in Michigan or Non-Michigan Groups Using Travel Iifformation Variables ....................................................... 122 26. Correlation Matrix for Travel Information Variables which Enter in The Michigan Traveler Discriminant Analysis „............................... 123 27. Classification Matrix for Predicting Membership in Michigan or Non-Michigan Groups Using Trip Attribute Variables ................................................................................... 125 28. Average Rating Scores Assigned to Six Destination Image Variables by Michigan And Non-Michigan .................................................................................... 126 Travelers xiii 29. Classification Matrix for Predicting Membership in Michigan or Non-Michigan Groups Using Michigan .........................................................................128 Image Variables 30. Correlation Matrix for Michigan Image Variables w hich E ntered in Michigan Traveler D iscrim inant Analysis ..............................................................................................128 31. Classification Matrix for Predicting Membership in Michigan or Non-Michigan Groups Using High Power Discriminators .........................................................................131 32. Correlation Matrix for The Selected Variables which Entered in Michigan Traveler Discriminant Analysis ................................................................. 131 33. LDA Correct Predicting Rates Comparison Using Different Types of Variables as Discriminators to Predict Michigan Trips ............................................................133 34. Percentage Improvement Over Chance Possibility in Multivariable LDA Prediction ................................................134 A-l.Trip Character Single Variable Predicting Power Test for Predicting Michigan Trips And Non-Michigan Trips A-2. Social Economic Single Variable Predictive Power Tests for Predicting Michigan Trips And Non-Michigan Trips A-3. Media Single Variable Predictive Power Tests for Predicting Michigan Trips And Non-Michigan Trips A - 4. Trip Attributes Single Variable Predictive Power Tests for Predicting Michigan Trips And Non-Michigan Trips A-5. Comparison of Michigan Single Variable Predictive Power Tests for Predicting Michigan Trips And Non-Michigan Trips B-1. Comparison of Average when Using Different Research Units: The Unit of Household Versus The Unit of Trip xiv CHAPTER I INTRODUCTION The tourism m arket, with its increasing importance, h a s become a popular subject for m any research and planning offices a t the international, national, an d local level. Though the tourism m arket is dynamic and involves m any factors, the study of factors affecting to u rist behavior can help in predicting certain tourism m arkets. This research explores the unique characteristics of Michigan’s tourism m arket and assesses the effectiveness of using existing travel information in the G reat Lakes area to predict, the propensity of tourists to choose Michigan as their destination. Though the general m eanings of tourism and to u rist are familiar to m ost people, the precise definitions are not. The World Tourism Organization defines to u rist as "a tem porary visitor staying a t least 24 hours or overnight in the country visited, whose journey is for the purpose of: (a) leisure (holiday, recreation, sport) or (b) business (family, mission, meeting, health, study, or religion)”. Tourism is defined as "a way of using leisure, and also with other activities involving travel." To narrow the scope of this research, only tourists 1 2 who traveled a t least 100 miles or sp en t a night away from hom e1, and travelled primarily for recreational purposes are considered. Tourism is more specifically defined as the com ponent of travel composed of the to u rists’ expenditures, facilities and services used, recreational activities pursued, and total experiences during trips of the types previously defined. V arious m ethods have been used by other researchers to quantify tourism . These include m easurem ents of financial volume (tourism expenditures), of movements (num bers of tourists), or of facilities used (nights of hotel or other lodging accom m odations sold). These m ake it possible to m easure, analyze, and predict various facets of tourism . Predicting tourists’ destinations is another way to assess the potential tourism m arket volume. If the factors th a t affect to u rists’ destination choices can be identified, the potential m arket can be predicted through an effective forecasting model. Many factors can influence to u rists’ trip destination choices. The m ost common factors include the traveler’s socioeconomic background, motivation, image of a destination, the location of the destination, available transportation, travel information obtained, and the traveling season. Socioeconomic factors m ay include age, sex, and income; and motivation factors may include the desires for relaxation, peace and quiet, self-satisfaction. 1 M o r e d i s c u s s i o n on the d e f i n i t i o n of t o u r i s m appe a r s in C h a p t e r II. fun, good scenery, outdoor experiences, good food, favorite recreation activities, escape from routine, family togetherness, and social statu s. All these factors are likely to differ from one individual to another. T ourists’ image of a destination is one especially im portant factor in th e selection of a trip destination. Baud-Bovy and Lawson (1976, plO) stated: The attraction o f tourist destinations arise to a large extent fro m the image ... The tourist image o f a destination is o f utm ost importance in tourism development; a choice o f destination is usually not m ade objectively but according to the image projected. Where prices are comparative, this is often the decisive factor in selection even though similar attractions and facilities m ay be available elsewhere. Travel inform ation obtained from travel agents, friends, relatives, and different m edia sources, su ch as m agazines, new spapers, television, radio, and billboards, are likely to be th e common sources of information th a t influence tourists’ images of certain destinations. This is the reason why tourism m arket prom otional strategies are designed to produce the impressive image th a t prom otes the product. In addition to price, the resources an d facilities at the destination, and the transportation to and a t th e destination are also im portant factors influencing tourists’ destination selection. Quality and availability opportunities, of attractions, transportation facilities, to the sightseeing, destination, recreational and local transportation, special events, even good re sta u ra n ts often play an im portant role in to u rists’ decisions. Knowledge of how to identify and apply these factors to predict who will choose Michigan a s their trip destination will help Michigan’s tourism industries to expand their m arkets. Also, the ability to identify the potential m arket will help tourism planners to determ ine tourism dem and. differences Accordingly, between secondly, Michigan evaluates th e th is research and first non-Michigan effectiveness of using exam ines the tourists, and existing travel information through a forecasting tool called Linear Discrim inating Analysis (LDA) to predict potential Michigan tourists. Other motivations for th is study are described below. In marketing, it is suggested th a t a careful selection of audiences can effectively increase the success of promotion. Tourism industries often u se m ass m edia as a promotional vehicle. The advertising used in promotion can be considered invitations th a t the industries send to potential tourists. Since it is unlikely th a t every audience will be interested in or able to afford the sam e type of travel, the unnecessary approach and of sending impractical. With everybody limited the invitations promotion is budgets, planners obviously need to know where and to whom to send the invitations. The value of forecasting tools th a t differentiate highpotential tourists from low-potential tourists is evident; they help to avoid sending costly invitations to low potential tourists. 5 Intuitive prediction is the forecasting tool m ost often used by people in m any different settings. Characteristics of the average cu rren t custom ers are used to identify potential custom ers. People who fit the profile are selected as targets for promotion program s. For example, the current m arket for luxury sport cars and station wagons could be described respectively as "high income, sports-oriented males" and "suburban family, sm all delivery agency, farm ers, large family and the like". Car dealers select high-income m ales who are sports-oriented, instead of low income people, as the target of luxury sports car promotions. M arried people with families, instead of single people, are the target of advertising to promote the sale of station wagons. Intuitive prediction is also used to identify criminals. Miami police compile a criminal suspect profile from p ast records of cocaine criminals, which describe the potential cocaine criminal as "black, between the ages of 18 and 40, usually driving a car with an out-ofstate licence plate" (CBS Sixty-Minutes report). Using this profile as a guide to identifying criminals, highway patrols stop whomever fits this description. Despite the success claimed by the police departm ent, the process of identification via intuitive prediction is actually unreliable in this situation because of the uncertainty of personal judgem ent. To improve the reliability of intuitive prediction, more strict statistical bases are required. Consum er psychology suggests th a t people who are Interested in a specific product are more likely to buy the product th a n those who are not. By interviewing people about their in terests in p aiticu lar products, interested people can be identified as high potential custom ers and considered the primary promotion target for each product. For instance, Michigan’s Upper P eninsula can offer more tourism experiences involving natural resources th a n can New York City. It is logical to assum e th at people who prefer traveling experiences involving n atu ra l resources will be more likely to travel to the Upper Peninsula th a n to New York City. Identifying these people and sending advertisem ent invitations to them should be more costeffective th an sending invitations to everyone. For some products, more th an one factor is required to make effective predictions. The selection process can be quite complex when the n u m b er of factors is increased. Appropriate statistical tools are required for more complex processes. In the past, a variety of forecasting techniques have been used to predict tourism , including causal m ethods, time series, qualitative m ethods, and decision analysis. Causal m ethods are used to identify relationships between the variables of interest. Exam ples are single­ equation regression models, m ulti-equation econometric models, sim ulation models, and spatial models. Time series analyses are used to analyze historical d ata patterns. These analyses include trend fitting, exponential sm oothing, and the Box-Jenkins technique. Q ualitative forecasting techniques include Judgm ental forecasting and th e Delphi method. Decision analyses include system dynam ics, m arket research, and probabilities forecasting2. However, a review of th e literature indicates th a t none of these techniques have been applied to the prediction of to u rists’ choice of trip destination. The LDA forecasting technique is considered a useful solution to the identification of Michigan’s high-potential tourists because it allows using many factors as discrim inators in predicting which people belong to the specific in terest group. According to Boyd (1981), LDA is the only forecasting technique th a t can differentiate individuals into predefined groups. In p ast decades, LDA h a s been used in m any m arketing areas to help identify and profile potential custom ers. It h a s been proved th a t LDA is a useful tool in identifying potential custom ers for certain products3. However, the literature review for this study revealed no applications of LDA for the purpose of identifying potential tourists4. To determ ine the effectiveness of LDA in tourism m arket prediction th u s requires a n actual investigation su ch as this. 2 M o r e d i scussions B.Archer, 1980. on tourism forecasting techniques can be found in 3 This includes market p r e d i c t i o n s u c h as who will be m o r e i n t e r e s t e d in s p e c i f i c b rands of cars, and what k i n d of s a l e s persons will p e r f o r m b e t t e r in s p e c i f i c kinds of businesses. M o r e r e v i e w of this topic is g i v e n in the l i t e r a t u r e r e view chapter. 4 Indeed, no l i t e r a t u r e r e view previous LDA (see Chapter II, application in tourism L i t e r a t u r e Review) . was f o u n d in the 8 In sum m ary, the need for tourism m ark et inform ation and for an effective forecasting technique to identify potential tourists motivated th is study. It attem pts to answ er th e following questions: (1) W hat are the significant m arket variables in Michigan’s recreational tourism market?; (2) W hat are th e differences between Michigan an d non-Michigan tourists?; an d (3) How effectively can existing travel information be used to identify potential tourists? Travel inform ation collected in the G reat Lakes Travel Monitor Study affords a n opportunity to seek answ ers to th ese questions. O bjectives Specifically, this study aimed to achieve the following objectives: 1. To stu d y origin-destination and other traveling p atterns of recreational to u rists in the areas covered by the G reat Lakes Travel Monitor Study. 2. To recreational generate to u rists socioeconomic in an effort inform ation to identify on Michigan’s the significant characteristics of Michigan’s recreational tourists. 3. To identify the differences betw een Michigan and non- Michigan to u rists th a t influence selection of Michigan as a tourist destination. 4. To assess the potential of several types of factors for predicting the propensity for traveling in Michigan. 9 The first two objectives resu lt in a series of descriptive statistical sum m aries from which a profile of recreational trips in Michigan emerges. The third objective is m et by testing the significance of the differences between Michigan an d non-M ichigan tourists. The la st objective is achieved through LDA applications. A s s u m p tio n s This research w as based on the following assum ptions, which serve as the basis of the research hypotheses in this study: 1. The tourism m arket in the G reat Lakes area can be categorized into two m utually exclusive sub-populations: Michigan and non-Michigan trips. 2. The travel p attern s and socioeconomic background of Michigan and non-M ichigan tourists are different. 3. Selection of Michigan as a trip destination is affected by travelers’ socioeconomic backgrounds (age, income, m arital status, etc.), images of Michigan (clear air, w inter fun, etc.), brand loyalty (visiting Michigan again), expectations for the trip, availability of travel information, accessibility of the destination (distance and available transportation), leisure time available (trip duration), holiday (date of trip), and seasonality (warm versus cold weather). 4. Prediction of tourism in Michigan’s future can be gained from an understanding of cu rren t tourism in Michigan, and a linear 10 function of certain attributes of th e population can be constructed to differentiate between Michigan to u rists and non-Michigan tourists. H y p o th e s e s Based on the above assum ptions, the following research hypotheses were proposed and tested. HYPOTHESIS 1. There is no significant differences in travel p attern s between Michigan and non-Michigan tourists. 1-a, 1-b. There is no significant difference between Michigan and non-M ichigan recreational to u rists in one way trip mileage and side trip mileage5. 1-c. There is no significant difference in trip duration6 betw een Michigan an d non-Michigan recreational tourists. 1-d, 1-e, 1-f, 1-g, 1-h, 1-i. There is no significant difference between Michigan and non-M ichigan recreational to u rists in the num ber of overnight stayed in: (1) hotel, (2) motel, (3) public ten t cam pground, (4) friend’s house, (5) relative’s house, and (6) other states. 5 Tests were c o n d u c t e d on one- w a y m a i n t r i p m i l e a g e only. m i l e a g e data were not available in the d a t a base. The r e t u r n t r i p 6 The n u m b e r of overnight stays away f r o m home in each trip was tested. 11 1-j, 1-k, 1-1, 1-m, 1-n. There is no significant difference between Michigan and non-M ichigan recreational tourists in the am ount of dollars sp en t on: (1) transportation, (2) lodging, (3) m eals, (4) entertainm ent, and (5) m iscellaneous expenses. l-o, 1-p, 1-q, 1-r, 1-s. There is no significant difference between Michigan and non-M ichigan recreational tourists in their preference scores for the following destination attributes: (1) good restau ran ts, (2) good places to stay, (3) high prestige, (4) good night life, and (5) winter fun7. 1-t, 1-u, 1-v, 1-x, 1-y, 1-aa, 1-ab, 1-ac, 1-ad, 1-ae, 1-af. Between Michigan and non-Michigan trips, there are no significant differences in travelers’ rating scores on the following images of Michigan: (1) good scenery, (2) good restaurant, (3) friendly people, (4) easy to get to, (5) reasonable prices, (6) good place to stay, (7) sum m er fun, (8) high prestige, (9) clean air, (10) good night life, and (5) winter fun8. 1-ag. There is no significant difference between Michigan and non-Michigan recreational tourists in their reported likelihood9 of 1 The terms of "good" a n d "high" me a n the q u a l i t y is h i g h e r than average. See question n o . 36 in s u r v e y quest i o n n a i r e in A p p e n d i x E. ' The terms of "good", "friendly", "easy", "high", a n d "clean" m e a n the q u ality is higher t h a n average. See question n o . 39 in s u r v e y q u e s t i o n n a i r e in A p p e n d i x E. 9 The degree of l i k e l i h o o d is m e a s u r e d by a five-p o i n t q u estion No. 28 in the q u e s t i o n n a i r e given in A p p e n d i x F ) . r a t i n g scale (see 12 revisiting the destination state. HYPOTHESIS 2. T ourists’ socioeconomic characteristics, travel patterns, available travel information, motivations for going to a state, and ownership of transportation do not positively affect to u rists' choice of Michigan as a trip destination. 2-a. Education level does not positively affect to u rists’ choice of Michigan as a trip destination. 2-b. Ownership of a recreational vehicle does n o t positively affect tourists’ choice of Michigan as a trip destination. 2-c. Marital sta tu s does not positively affect to u rists’ choice of Michigan as a trip destination. 2-d. Traveling on th e weekend does not positively affect to u rists’ choice of Michigan as a trip destination. 2-e. Taking a group trip does not positively affect to u rists’ choice of Michigan as a trip destination. 2-f. Using a personally owned vehicle to reach a destination does not positively affect to u rists’ choice of Michigan as a trip destination. 2-g. Using a personally owned vehicle at the destination site does not positively affect to u rists’ choice of Michigan as a trip destination. 13 2-h, 2-i, 2-j. Receiving travel Information from television, radio, and billboards does not positively affect to u rists’ choice of Michigan as a trip destination. 2-k. W inter fun is not a factor th a t positively affect to u rists’ choice of Michigan as a trip destination. 2-1. Michigan residency does not positively affect tourists* choice of Michigan as a trip destination. 2-m. The warm season does not positively affect tourists’ choice of Michigan as a trip destination. HYPOTHESIS 3 . Travel information and trip concerns do not negatively affect tourists* choice of Michigan as a trip destination. 3-a. Obtaining travel information does not negatively affect tourists’ choice of Michigan as a trip destination. 3-b. Obtaining travel information from a travel agent does not negatively affect tourists’ choice of Michigan as a trip destination. 3-c, 3-d, 3-e, 3-f. Good restau ran ts, clear air, good night life, and high prestige associated with a trip do not negatively affect tourists’ choice of Michigan as a trip destination. 3-g. The distance of the state of origin from Michigan does not negatively affect tourists’ choice of Michigan as a trip destination. 14 HYPOTHESIS 4. Travelers’ socioeconomic characteristics, travel behavior, available travel information, and th e m otivations for going to a state have no effect on the choice of Michigan as a trip destination. 4-a. O ccupation h a s no effect on the choice of Michigan as a trip destination. 4-b. Season h as no effect on to u rists’ choice of Michigan as a trip destination. 4-c. Travelers’ expectations of a destination have no effect on tourists’ choice of Michigan as a trip destination. 4-d. Knowing the toll-free num ber for travel inform ation has no effect on to u rists’ choice of Michigan as a trip destination. 4-e. The opportunity for outdoor activity during th e trip has no effect on to u rists’ choice of Michigan as a trip destination. 4-f. Special events connected to a trip have no effect on tourists’ choice of Michigan as a trip destination. Constraints o f The S tu d y This stu d y w as conducted with three m ajor lim itations: time, budget, an d the quality of data available. In term s of time, the primary research investigator of this study is a foreign stu d e n t from Taiwan, Republic of C hina who is permitted limited time to rem ain in the U.S. In term s of budget, currently this research is financed only by the researcher himself. The research 15 budget is limited. D ata used in this stu d y were n o t originally collected for this study, th u s analyses were lim ited to th e information available. While statistics applied in th is stu d y were appropriate for the inform ation available, some data were of poor quality and m ay have limited significance of the results. U nder these circum stances, the scope of this study reflects a compromise between manageable efforts an d the complexity necessary to dem onstrate the power of the LDA analyses in identifying factors th a t influence potential tourism in Michigan. The dissertation is organized as follows. Chapter II provides a review of the literature on tourism m arketing and LDA theories. C hapter III explains the methodology used in the study. Study findings are presented in C hapter IV. D iscussion of study resu lts, and implications for future research are discussed in C hapter V. CH A PTER H LITERATURE REVIEW lite ra tu re related to the definition of tourism , tourism m arketing, the significance of tourism in Michigan, an d LDA theories is reviewed in this chapter. Tourism definitions, tourism m arketing, factors influencing u se r attendance at recreation areas, advertising efficiency, and tourism m arket allocation, are first discussed. The theories, m echanism s, com parison and of LDA w ith applications of LDA, together w ith a other modeling techniques, are then presented. T o u ris m D e fin itio n s Tourism h as m any facets, and it is not feasible to include all of them in a simple definition. Since 1910, tourism has been described in m any ways. Some of which are more am enable th an others to operationalization for m easurem ent purposes in research. One of die earliest definitions, given by econom ist H erm ann v. Schullard in 1910, described tourism as "the sum total of operations, mainly of an economic n atu re, which directly relate to the entry, stay 16 17 and movement of foreigners inside and outside a certain country, city or region." This definition is the first to conceptualize tourism in economic terms. Since 1910, definitions of tourism have incorporated additional concepts, su ch as technology, public adm inistration, social aspects, cultural activities, attitude, and behavior aspects. Doom (1982) posited four definitions of tourism : a basic definition, a mono-disciplinary definition, a statistical definition, and a system analysis definition. In the basic definition of tourism , to u rists’ sta tu s of "stay," ’journey," and "being away from home" are the major elements. B urkart and Medlik’s (1974) definition belongs to this type. They said, 'Tourism denotes the temporary, short-term movement of people to destinations outside the places where they norm ally live and work for other th a n b u sin ess or vocational reasons, and their activities during the stay a t these destinations." This definition is representative of m ost of the basic concepts in tourism today. Mono-disciplinary motivation, pleasure, tourism and definitions tourism focus experiences. on Cohen’s people’s (1974) definition exemplifies a mono-disciplinary definition. He defined a tourist as "a voluntary, tem porary traveler, traveling in the expectation of pleasure from the novelty and change experienced on a relatively long and non-recurrent round trip." This definition is inadequate since the natu re of tourism is m uch more complex th a n th e simple linkage of people’s motivation and their experience. 18 Statistical definitions u se statistics to define tourism . They are often used by governmental and international organizations, su ch as the World Tourism Organization (WTO) and the Organization for Economic Co-operation and Development (OECD). In th is type of definition, tourism is defined as th e su m of the num ber of arriving and departing, their spending, the duration of their tours, th e purpose of stay, etc. Since this definition u se s m ainly statistical d a ta to define tourism , it is especially useful for report writing purposes. The last type of tourism definition gives the widest perspective by adopting the system analysis approach. Leiper’s (1979) definition is an example: 'The elements of th e (tourism) system are tourists, generating regions, tran sit routes, destination regions, and a to u rist industry. These five elements are arranged in spatial and functional connections." This definition includes both to u rists and trip environm ent as tourism elements, m aking it more functional th a n other definitions. Because of the inclusion of spatial concept, this definition is often adopted in tourism planning. None of these definitions can entirely describe the tourism phenom enon. However, each definition supplies certain quantitative and qualitative aspects of tourism th a t serve as the basis for research to study and m easure tourism phenom ena. 19 M ichigan’s T ourism M ark et Michigan’s tourism m arket is composed of a dom estic segm ent and an inter-state an d international segm ent. According to a Better Homes and G ardens (1977) report, 'T he Family Vacation Travel Market,"10 the G reat Lakes S tates" represent Michigan’s prim ary travel m arket, accounting for 82% of Michigan’s total tourism m arket. In 1984, the G reat Lakes States m arket still m aintained ab o u t 81% of Michigan’s total tourism m arket12 (U.S. Travel D ata Center, 1985). These figures imply th a t the Great Lakes S tates are a stable m arket for Michigan’s tourism business. Despite the fact th a t the Great Lakes S tates are the major m arket for tourism industries in the region, Michigan’s m arket share of this m arket is only 14%. Wilson (1981) suggests th a t extensive opportunities exist for expanding Michigan’s share of the m arket. He estim ates th a t as m u ch as 28% of the m arket sh are could be captured if effective tourism promotion were conducted. To expand the market, understanding and specifically identifying M ichigan’s cu rrent and potential tourism m arket is crucial. 10 This article is a t a b u l a t i o n report b a s e d o n t he 1977 National Survey conducted by the Bureau of the Census. In th i s report, r e creational trips that w e r e o v e r 100 m iles f r o m h o m e a n d i n c l u d e d at two h o u s e h o l d members were counted. Ohio, 11 In this report, the G r e a t Lakes States i n c l u d e d six Illinois, Wisconsin, Indiana, and Minnesota. 12 This was b a s e d on s e v e n Great Lakes Wisconsin, Indiana, Minnesota, Ohio, and Iowa. States: states: M i c higan, Travel only least Michigan, Illinois, 20 The Significance o f T o u ris m in M ichigan The tourism in d u stry is also called the "hospitality" in d u stry due to the fact th a t hospitality services have been used to bring in to u rist dollars, creating a local economic impact. The concept of "hospitality for sale" w as th o u g h t to be im practical by the executive levels in m ost governments u n til the economic value of th is in d u stry w as recognized. Now, a s its economic im pact becomes increasingly significant, expansion of the tourism industry h as become a topic of great importance in economic planning offices across th is country and around the world. Historically, the im portant landm ark Michigan in Tourist Michigan’s Council tourism rep resen ts development. an It w as founded in 1945 for th e purpose of promoting Michigan tourism (Wilson, 1981). Shortly after World War II, Michigan started receiving significantly increasing num b ers of tourists. Since then, the economic value of tourism h as been rapidly growing. Today, tourism in d u stry is one of the m ost im portant industries in Michigan.7 According to Wilson (1981), an increase of one percent in Michigan’s tourism m arket could produce an additional 125 an d 95 million dollars in direct and indirect expenditures, respectively. This 7. The number one M i c h i g a n i n d u s t r y is a u t o m o b i l e m a n u f a c t u r i n g . T h e cash r eceipts from Michigan's f a r m m a r k e t i n g s is 3.045 b i l l i o n in 1983. (M i c h i g a n S t atis t i c a l A b s t r a c t ) The d irect t r a v e l e x p e n d i t u r e s in M i c h i g a n is 5 . 5 4 5 b i l ­ lion. (U.S. Travel Data Center, Impact of Travel on State Economies, 1 9 8 3 ) 21 would also produce 6,100 additional Jobs and 10 million dollars in new sta te tax revenues. The economic im pact from tourism will be greater if th e state can help attract more to u rists to Michigan. Since unidentified potential custom ers always exist in m arket areas, an effective tool for identifying these people could be significant to the economy of Michigan. Michigan’s Tourism Challenges Two m ajor challenges faced by Michigan’s tourism industry are vigorous com petition and inadequate m arketing information. Vigorous C om petition It h a s been observed th at competition for tourism dollars in the G reat Lakes area has been stepped u p in the p a st decade. According to the U.S. Travel D ata Center, the Michigan Travel B ureau increased advertising budget percentage from 30.0% ($2,782,383) of the total budget in 1984-1985 fiscal year to 36.4% ($4,300,000) in the 19851986 fiscal year. The increase of the percentage of total budget allocated for tourism promotion purposes w as 6.4%. In the sam e period, a t least three neighboring G reat Lakes S tates had exceeded Michigan’s increase. The states th a t had allocated higher percentages of total budget for tourism advertising use th a n Michigan are Indiana (increased by 9.5%), M innesota (increased by 18.2%), and Wisconsin 22 (Increased by 14.8%)(U.S. Travel D ata Center, 1985). Vigorous competition also comes from d istan t sta tes such as Florida, and California, and from the C anadian province, Ontario. In the 1985-1986 fiscal year, California and Florida sp en t 62.2% and 45.9%, respectively, of their office budget for tourism promotion. In 1984, Ontario sp en t $5.7 million advertising dollars in th e U.S, more th a n twice the $2.5 million Michigan sp en t in the U.S. th a t year (Spotts, 1986). These advertising budget figures dem onstrate the growing pressu re on Michigan to increase its com petitiveness in the tourism industry. An effective marketing strategy is vital if Michigan is to improve its competitive position. Inadequate Market Information In an environm ent of intensive com petition, marketing information is crucial for the development of effective marketing strategies. Unfortunately, specific tourism m arket inform ation, such as who, where, and w hen to m arket the products, is n o t always available when needed. The inadequacy of data concerning tourism in Michigan was reflected in Holecek and Wilson’s words: ... data are indicative o f tourism’s significance in Michigan but also are one example o f the shortage o f information which exists concerning Michigan’s^ second or third most important industry. (Holecek, 1981, p i 7) ... much more specific information is needed to determine w hat m essages should be communicated and to target the m essages to the appropriate markets. (Wilson, 1981, p31) 23 Much raw data concerning to u rists’ socioeconomic background, su ch as age, income, and education, and travel inform ation su ch as trip origin, destination, purpose, and travel mode have been collected. However, more useful information, su ch as who is m ost likely to visit Michigan, how to identify them , and the factors th a t determ ine the choice of destination have rarely been obtained or estim ated through forecasting techniques. T o u ris m M a rk e tin g Gray (1981, pi) h as stated: Recreation marketing concerns itself with sending the invitation. Before w e can sen d any invitations, w e have to decide who to invite and know where their invitations are to be sent. This statem ent contains two basic concerns in tourism m arketing: "Who will be interested in the product?" and "Where can those interested be found?". These questions can be answ ered through study of tourists’ travel motivations, and other factors th a t influence purchasing of the tourist industry’s product. E m e rg e n c e o f T h e C o n c e p t o f T ra v e l M o tiv a tio n s In the p ast decade, two m ajor approaches to tourism promotion have been common: product-oriented promotion and custom er-oriented 24 promotion. The former approach tries to "sell" available attractio ns and facilities. The latter focuses on the identification of custom ers and their needs. W ahab (1976) suggested discarding the product-oriented concept because it tries to convince custom ers th a t the products are w hat they need instead of determ ining w hat products custom ers actually need. He said th a t the product-oriented approach is to con­ vince potential visitors th a t the assets and resources of a specific d es­ tination are those th a t the potential visitors desire. The approach he suggests is one which considers custom ers’ motivations, attitudes, and behavior. According to Smykay (1977), motivations come from the "relevant needs of h u m an beings." Schmoll (1977) thought th a t people’s motivation should be as im portant a factor as economic and commercial aspects in tourism plans. He said: ... psychology is concerned w ith the stu d y and analysis o f tourists’ motivations a nd behavior which, in turn, have a direct bearing on promotion plans. (Schmoll, 1977, p51) It h as been shown th at attention to a custom er’s motivations, behavior, and socio-demographic sta tu s h as a greater chance of success in advertising. Korgaonkar (1984) found th a t a careful stu dy of custom ers’ needs before advertising th e product can significantly contribute to a successful campaign. T ourists’ motivations, behaviors, and socio-demographic characteristics are also im portant 25 considerations in a tourism plan. In Schmoll’s "Model of Tourism" (1977), it is suggested th a t a tourist service plan should be guided by tourists’ attitudes, motivations, and behavior. Services th a t need custom er motivation research include transportation, accomm odations, food, sightseeing, entertainm ent activities, travel advice, travel arrangem ents, and banking and shopping facilities. A wide variety of motivations can be involved in custom er profiles. S tatus is often a m ajor consideration in buying. Smykay (1977) writes: "Status involves differentiation from the herd... It therefore implies exclusivity. Ownership of a high class automobile is a way people can visually dem onstrate their sta tu s to the crowd." Obviously, for successful tourism marketing, to u rist motivations cannot be ignored. Factors in A ttendance at Recreation Areas It is not difficult to identify some reasons why people choose a particular trip destination. Motivations can be identified by using an interview survey. If accurate data for all factors affecting the decision to visit Michigan were available, one would be able to reasonably predict Michigan’s tourism market. However, there are m any underlying factors th a t motivate choice of destination. It is probably impossible to identify all the factors involved. Clawson and Knetsch (1975, PP59-60) have identified three categories of factors which 26 directly affect tourist visits to recreational areas: (1) u se r factors, (2) area factors, and (3) a user-area interaction factor. These are defined as follows: 1. U ser factors: geographic distribution, socioeconomic characteristics (age, sex, occupation, family size and composition, education, income, and race), leisure time, knowledge of recreation opportunities, and personal tastes. 2. Area factors: attractiveness, availability, su b stitu tes, capacity, climatic and seasonal characteristics, etc. 3. U ser-area interaction factor: travel time required, m onetary costs, comfort of travel, and the extent of stim ulation by advertising. Many other factors exist for researchers to explore, su ch as gasoline prices, accessibility, facility m anagem ent quality, and political policy. Although all of these factors can be included as the basis to project the future tourism m arket, it is not practical to do so. Therefore, only the factors th a t contribute considerable predictive power should be selected. A d v e rtis in g E ffic ie n c y According to Aaker (1982), advertising efficiency is m easured by the ratio of advertising dollars sp en t to resulting sales dollars. Mahoney and Wamell (1986) suggested th a t the m arketing strategy of "attem pting to be all things to all people" is inefficient, because 27 "strategies designed for the average custom er often result in unappealing products, prices, and prom otional m essages." An effective strategy should "match the right product or service with the right m arket or audience." The right audience is one th a t will be interested in the products and will eventually visit tourism facilities promoted by the advertisem ents. An appropriate audience can be obtained through careful analysis of survey data on the needs, motivations, and characteristics of u sers of different tourism facilities. The purpose of selecting the audience is to increase advertising efficiency. A aker (1982) found th a t u n d er the p ressu re of competition both advertising agencies and tourism in d u stries tend to advertise more th a n m ight be optimal. In m ost cases, over advertising results in inefficiency and financial resource m isallocation. In 1980, nearly 55 billion dollars were sp en t on advertising in th e U.S. In 1982, the U.S. advertising expenditure increased to over 70 billion dollars, or about 2.5% of the Gross National Product. Most of this advertising expenditure w as not optimal (Korgaonkar, 1984). To understand how advertising can be more efficient, the basic m echanism of advertising m u st be understood. Schmoll (1977, p71) described the way th at advertising works: Advertising implies indirect communication w ith selected target groups through paid m essages transmitted through suitable media-press, electronic media, by mail, etc. 28 In other words, the basic function of advertising is to reach, inform, and persuade potential tourists to p u rchase available products or services. Tourism advertising includes three m ajor elem ents: (1) m ass m edia - magazine, television, press, etc., (2) m essage - advertising program, and (3) audience - potential tourist. The first two elem ents function together like a bow and arrow, aiming a t th e target audience. Any im proper selection of these elem ents could resu lt in inefficiency. Of these three elements, target audience selection is considered m ost crucial. Kotler (1985, p35) explained how im portant the selection of the target audience is: ... the exposure value (of advertisement) depends on the readers’ characteristics and how closely they m atch those o f the consumer target groups. For a baby lotion advertisement, the exposure value might be 1,000,000 if the readers were all women and 0 if the readers w ere all men. Since the tourist industry cannot afford advertising th a t reaches all audiences, it is imperative th a t the target audience be carefully selected. Tourism Market A llocation Since the tourism m arket is highly dynamic, m ark et allocation is no easy task; it requires knowing both the m arket’s location and its characteristics. Krippendorf (1972, 122) described m ark et selection procedures as follows: 29 ...(1) Determine the market size: establish the num ber o f potential visitors o f a destination, (2) Localize the market(s): establish the geographical location and distribution o f these potential visitors. (3) Determine market characteristics: behavior and motivations o f the potential visitors, the image o f and attitudes towards the destination or service. One commonly used method for selecting target m arkets is through the u se of Areas of Dom inant Influence (ADI). ADI are geographic areas currently served by dom inant advertising media. Promotion planners can select different areas as potential tourism m arkets based on the availability of advertising media. ADI is an effective way of selecting a tourism m arket. The im portance of m arketing in each ADI can be assessed by combining inform ation on the client, the m edia available within each ADI, and each ADFs share in the total tourism m arket. Based on this assessm ent, a m arketing promotion technique can be developed. First, the county in which custom ers reside is identified. The percentage of m arket share in each ADI is then calculated. With this technique, m arket areas th a t are currently servicing clientele can be located. An example of ADI application is the development of Recreation M arketing Maps by the Recreation Resources Center a t the University of W isconsin (Madison) in 1981. ADI were used to locate w here and who custom ers were. Selection of the media for m arketing promotion, guided by the share of market, w as indicated. It w as pointed out th a t the m arket potential of ADI can be characterized by: 30 1. Area of D om inant Influence. 2. Counties included in the ADI. 3. Estim ate of ADI population. 4. Estim ated n u m b er of households in the ADI. 5. Consum er Spendable Income Per Household for the ADI. 6. Automobiles Per Household for the ADI. 7. A m ap of the ADI. The ADI approach locates potential m arket areas by drawing a profile of already existing custom ers. However, m ethods for further identifying other potential custom ers are still lacking. LDA A p p lic a tio n s The original purpose of LDA and its development are outlined in the following brief historical background. According to the theory of natu ral relationships, p lan ts and anim als can be classified into different categories. In biology, new bacteria hybrids are so sim ilar th a t sometimes th e determ ination of species of newly bred bacteria is difficult. LDA w as originally developed to help biologists to determine the species of new hybrid bacteria. This function w as then extended to profile group differences and to classify individuals into separate groups based on the nature of differences. Currently, LDA is applied in a wide variety of fields. In m ost applications, LDA research is used to determ ine characteristics 31 of interest in order to classify groups. In park and recreation, the only LDA application found in the literature reviewed is Westfall’s stu d y (1975) in which LDA w as used to help m ake adm inistrative policy decisions. This application w as intended to identify landow ners having a potential negative willingness to comply with policy decisions. Policy m akers can th e n focus on these landowners’ concerns in planning activities in those areas. From the sam pled land owners, Westfall segregated farmers who have a higher willingness to allow public access to their land for three recreational activities: hiking, hunting, and snowmobiling. The characteristics of landowners used in this application include parcel size, sex, age, the size of the parcel of land, years owned, percentage of land in crops, percentage of land as woods, prim ary ow nership objective, and residence location. This study resulted in two discrim inant functions. One LDA function resulted in a 62.5% correct predicting rate (CPR)8 com pared with a 51.0%9 CPR in a chance predicting process. Another function resulted in 58.8% CPR, w hich is considered low when com pared with a 65.3% CPR in random 8 In LDA, the p redicting rate is the p r o p o r t i o n of o b s e r v a t i o n s c o r r e c t l y classified, which is c a l c u l a t e d b y d i v i d i n g the number of observations c o r r e c t l y c l a s s i f i e d by the total n u m b e r of observations. 9 W i t h equal group size, the p r e d i c t i n g rate of a c h a n c e p r e d i c t i n g p r o c e s s is 50% for two groups. U n e q u a l g r o u p sizes affect t he c h a n c e of o b s e r v a t i o n s to be assigned in eac h group. The r a ndom p r e d i c t i n g rate thus re quires adjustment to reflect g r o u p size differences. In W e s t f a l l ' s study, the C P R was a d j usted to be 51%. 32 predicting for the m em bership of two groups of u n equal size. This study did n o t result in an outstanding CPR. According to Westfall, this w as due to the particular discrim inating variables available and the grouping m ethod used. Since no other LDA applications in the p ark and recreation field were found, applications of LDA in other fields are reviewed and sum m arized below. A sum m ary of these applications helps to clarify the purposes for which LDA was devised. In business, social science, and other areas, LDA is a useful tool for investigating the effectiveness, risk, products, m anagem ent, and custom er differences between groups. In the applications reviewed, LDA was used as a tool for the estim ation, identification or prediction in the following areas: 1. A d v ertisem en t: (1) Television commercial rating scales (Lastovicka, 1983): (2) R estaurant advertising: appeals and consum ers’ intentions (Lewis, 1981): (3) Benefit segm entation for restau ran t advertising th a t works (Lewis, 1980); (4) Advertising message and life style (Greeno and Sommers, 1977). 2. A ntrophology: (I) Fingerprint variation in P apua New Guinea for the implications of prehistory (Froehlich and Giles, 1981): (2) Q uantitative serum protein data in populations of Rwanda (Jayakar and others, 1981). 33 3. B ehavior S tudy: (1) Decide the voluntaiy union m em bership of women and men: differences in personal ch aracteristics, perceptions and attitu d es (Snyder, 1986); (2) Career goals, organizational reward system s and technical updating in engineers (Steiner, 1986); (3) Work patterns in the professional life-cycle (Raelin, 1985). 4. D ecision S cience: (1) A performance analysis of param etric and nonparam etric discrim inant approaches to b u sin e ss decision m aking (Mahmood, 1987); (2) Decision rules for increasing the rate of successfully classified respondents (Koslowsky, Locke, 1986); (3) Relationship between jo b attitu d es and the decision to retire (Schmitt and McCune, 1981); (4) Agricultural land use (Fotheringham and Reeds, 1979). 5. E m ployee S electio n , E valuation: (1) The salesm an selection process (Perreault, 1977); (2) D eterm inants of faculty ra n k (Hoffman, 1977); (3) Employee selection (Welker, 1974) (Higgins, 1970). 6. G overnm ent: (1) Classification of nations as developed and less developed by socioeconomic d ata (Dellaportas, 1983); (2) Evaluation of the success of the H ungarian economic reform an analysis using international-trade data (Murrell, 1981); (3) Response time and its significance in medical emergencies (Mayer, 1980); (4) Evaluation of a state-w ide system for identification of educationally handicapped children (Petersen and Hart, 1978); (5) City goverment structure (Dye and M acm anus, 1976). 34 7. H o tel a n d R e sta u ra n t A d m in istratio n : (1) Evaluation of the m arket position: m apping guests’ perceptions of hotel operations (Lewis, 1986). 8. M ark et S eg m en tatio n , P erform ance: (1) The assessm en t of com pany perform ance using a statistical model (Taffler, 1983); (2) M arketing of legal services (Darden and others, 1981); (3) Toward a theory of segm entation by objectives in social m arketing (Fine, 1980); (4) M arket performance of large commercial ban k s and b an k holding com panies (Simposn and Kohers, 1979); (5) M arket segm entation (Johnson, 1971)(Lease and others, 1976). 9. C o n su m ers C h a ra c te ristic s, Id en tific atio n , A nd L oyalty: (1) Reliance on life insurance agents: a demographic and psychographic analysis of consum ers (Burnett and Palmer, 1983); (2) M unicipal bond ratings (Stock and Robertson, 1981); (3) Bank credit card u se r characteristics (Martell and Fitts, 1980); (4) Perceptual m apping of consum er products and television shows (Stanton and Lowenhar, 1977); (5) Locating custom ers in a segmented m arket (Levine, 1975); (6) Industrial source loyalty (Wind, 1970); (7) The relations between consum ers’ attitudes, behavior and intentions (Perry, 1969); (8) Freight traffic of competing transportation m odes (Mildius, 1969); (9) Potential air freight u se rs (McKinnell, 1968). 10. P ro d u c ts, Service S electio n : (1) Social character an d the new automobile industry (Mccrohan an d Finkelman, 1981); (2) Sales 35 forecast u n certain ty in new product situations (More and Little, 1980); (3) Different m agazines reading between working wives and non­ working wives (Louglas, 1977); (4) Store selection by female shoppers using Age and education as predictors (Bellenger an d others, 19761977); (5) New product distribution and superm arket buyer decisions (Montgomery, 1975); (6) Effective new product decisions for super m arkets (Doyle an d Weinberg, 1973). 11. F in a n c ia l R isk Evaluation: (1) An investigation of the major influences of residential liquidity: a m ultivariate approach (Moore, 1987); (2) Logit v ersus discrim inant analysis: a specification test and application to corporate bankruptcies (Lo, 1986); (3) The demolition of downtown low-income residential buildings: a discrim inant analysis (Bell, Kelso, 1986); (4) Predicting dividend changes (Kolb, 1981); (5) Differences in risk preference between the public and private sectors (Burton, and W aldron, 1978); (6) financial failure: a re-exam ination (Moyer, 1977); (7) Financial early w arning (Altman and Loris, 1976); (8) Early w arning of changes in banks’ financial condition (Korobow, 1975); (9) B ank charge-card holders by economic, demographic, and attitudinal characteristics (Awh and W aters, 1974); (10) Rating the financial condition of b an k s as a aid of b an k supervision (Stuhr and Wicklen, 1974); (11) Small business failure using financial ratios as predictors commercial (Edmister, loan 1972) (Deakin, evaluation 1972); (Altman, (12) Implications for 1970); (13) Coorporate 36 b ankruptcy (Altman, 1968, 1970); (14) Altm an’s corporate bankruptcy model revisited: Can airline bankruptcy be predicted (Scaggs, Crawford, 1986), 12. S ta tistic s: (1) Is statistical discrim ination efficient? (Schwab, 1986) (The author adapts George Akerlof and Hayne Leland "Lemons" model to labor market); (2) Resolving certain difficulties and improving th e classification power of linear program m ing discrim inant analysis form ulations (Freed, 1986); (3) Variable selection in heteroscedastic discrim inant analysis (Fatti, Hawkins, 1986); (4) D iscrim ination with polychotomous predictor variables using orthogonal function (Butler, 1985); (5) Common principal com ponents in k groups (Flury, 1984); (6) Linear sam ples Discrim inant (Chhikera, Analysis Mckeon, with 1984); m isallocation (7) Adaptive in training classification procedures (Rukhin, 1984). C o m p a ris o n o f LDA And Other M o d e lin g T e c h n iq u e s Multivariate modeling techniques, in essence, transform raw d ata associated with a particular phenom enon into more ab stract information. The unknow n causes of a phenom enon are discovered by analyzing the relationship between the dependent and independent variables, which are utilized to describe or characterize the phenom enon. Once the relationship h as been established, the newly observed d ata can be used to predict an evolving event. For example. 37 once the degree of hum idity h a s been established, it can be used to predict the possibility of rain. These techniques involve two basic m ethods. The first m ethod is to sep arate respondents into different categories based on selected independent variables. The second m ethod is to identify interdependencies am ong a num ber of selected independent variables. An example of the first method is the identification of new bacteria hybrids into cu rren t known species. An illustration of the second method is th e categorization of new bacteria hybrids into different species w ithout giving the definition of the category in advance. Cross-Tabulation, regression analysis, LDA, and autom atic interaction detector (AID) are tools of the first m ethod. Tools of the second type include clu ster analysis, factor analysis, and conjoint analysis (Boyd, 1981). In th e following sections, three popular m ultivariate analyses, regression analysis, factor analysis, and cluster analysis are compared with LDA. The Use o f LDA. R e g re s s io n A nalysis. Factor Analysis, and Cluster Analysis According to Greenburg’s report in 1977, the frequency of use of factor, cluster and LDA in m arketing research is 23:18:14.10 From the low proportion of LDA applications in th is comparison, it may be inferred th a t LDA was still a relatively new marketing research 10 The report d i d R e g r e s s i o n Analysis. not provide information on the frequency of use of 38 technique a s recently a s 1977. A study of th e b u sin ess periodical index published by H.W. Wilson Co. from 1958 to 1987" revealed th at, am ong the four techniques, regression analysis and factor analysis have the longest application history in b u siness research. Publications employing discrim inant analysis an d cluster analysis did n o t appear until 1967 and 1971, respectively. Since its appearance, however, discrim inant analysis h a s become the m ost often applied technique other than regression analysis in th e p ast decade. The frequency of application of each technique during th is period is sum m arized by au th o r in Table 1. 11 The index covers over t h r e e h u n d r e d b u s i n e s s per i o d i c a l s . 39 Table 1. Comparison of frequency of u se of discrim inant, regression, cluster, and factor analysis in b u sin ess publications. Num ber of Publications Period (Month/Year) D iscrim inant Analysis Regression Analysis Cluster Analysis Factor Analysis 4 /8 7 -7 /8 7 8 /8 3 -7 /8 4 8 /8 2 -7 /8 3 8 /8 1 -7 /8 2 8 /8 0 -7 /8 1 8 /7 9 -7 /8 0 8 /7 8 -7 /7 9 8 /7 7 -7 /7 8 8 /7 6 -7 /7 7 8 /7 5 -7 /7 6 8 /7 4 -7 /7 5 8 /7 3 -7 /7 4 8 /7 2 -7 /7 3 8 /7 1 -7 /7 2 8 /7 0 -7 /7 1 8 /6 9 -7 /7 0 8 /6 8 -7 /6 9 8 /6 7 -7 /6 8 8 /6 6 -7 /6 7 8 /6 5 -7 /6 6 8 /6 4 -7 /6 5 8 /6 3 -7 /6 4 8 /6 2 -7 /6 3 8 /6 1 -7 /6 2 8 /6 0 -7 /6 1 8 /5 9 -7 /6 0 8 /5 8 -7 /5 9 6 12 0 21 12 26 14 20 18 12 10 5 6 3 4 3 2 1 0 0 0 0 0 0 0 0 0 6 42 39 70 50 118 112 63 79 42 25 39 44 29 31 21 19 18 18 5 11 11 6 8 2 2 6 4 8 5 9 5 8 28 2 9 2 6 4 7 12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 5 5 11 15 14 10 12 3 4 7 1 2 5 5 3 7 0 2 3 2 0 3 2 3 7 Total Percentage 175 13.32% 916 69.71% 86 6.55% 137 10.42% 40 Comparison betw een LDA and Multiple Regression In m any ways, LDA is similar to conventional linear regression analysis. Both m ethods use a linear function to predict a dependent variable. However, LDA predicts the affiliations (i.e. m em bership of a variable in a p articu lar group) a t nom inal scale, while regression analysis predicts individual values of the variable u n d er investigation on a ratio or interval scale.12 Churchill (1986, p737), Ghiselli, Campbell, and Zedeck (1981, p363), and Lansing and Morgan (1971, 300) explained th e m ajor difference between the two analyses as follows: Discriminant analysis is similar to multiple-regression analysis in that it involves the investigation o f a criterionvariable and predictor-variable relationship. Only now the criterion variable is a dichotomy or mullichatomy, whereas w ith regression analysis it is interval scaled. (Churchill, 1986, p737) Algebraically, discriminant Junction analysis is equivalent to regression analysis except that the criterion is dichotomous rather than continuous. Ghiselli (1981, p363) 12 At this point, it is appro p r i a t e to e x p l a i n t h r e e s t a t i s t i c a l terms: "nominal level," "interval level," and "ratio level." A n y n u m b e r a s s i g n e d at the nominal level can only represent a class or category. The n u m b e r so a s s i g n e d is u s e d to i d e ntify or represent the c a t e g o r y but not u s e d in calculation. We can a ssign "1" to represent an y " M i c h i g a n v a c a t i o n e r " a nd "2" to represent a n y "non- M i c h i g a n vacationer." However, it is m e a n i n g l e s s to say that "1" plus "2" e quals "3," since "1" a n d "2" are just m a r k s a n d "3" r e ceives no m e a n i n g here. The numb e r a s s i g n e d at i n terval level can be c o m p a r e d or c a l c u l a t e d b y nonpa r t i a l units. P e o p l e can be c o u n t e d by the unit of one. However, 1.5 p e o p l e is mean i n g l e s s since p e o p l e can o n l y be c o u n t e d as wholes. Numbers a s s i g n e d at the ratio level can be c o m p a r e d or c a l c u l a t e d as continuous or p a rtial units. A n example is the m e a s u r e m e n t of t he length of a p i ece of wood as 10.33 inches. 41 In regression analysis, the independent variables are regarded as fixed, while the dependent variable is regarded a s fix e d in discriminant analysis. (Lansing and Morgan, 1971, p300) The major difference between the two techniques is th e criterion th a t guides each technique. In LDA, the criterion is group m em bership. The task is to predict th e category. In regression analysis, there is no criterion and the ta sk is to predict individual values on a continuous scale. In G hisellfs words: In discriminant Junction analysis, w e are interested in a composite o f variables that has m aximum potential fo r distinguishing betw een members o f groups ... the purpose is to maximize the differences among groups or to weight tests or predictors to maximally distinguish betw een established groups. (Ghiselli, 1981, p362) Prediction from discrim inant analysis is in term s of likelihood of group mem bership and is based on the betw een-group differences explained by the composite of variables. In contrast, multiple regression is concerned w ith the composite of predictors th a t yields the best explanation of variables in the continuous, univariate criterion. The power of LDA over Regression Analysis lies in its capability of predicting m em bership (i.e. category) a t a nom inal level. For example, we m ay define people who visited Michigan before 1985 into two groups: "loyal visitors" who have revisited Michigan a t least once and "non-loyal visitors" who have not returned to Michigan. This 42 defines Michigan tourists in the term s used in relation to custom er brand loyalty. In the same way th a t it is used to establish a custom er’s characteristics profile, LDA can function to distinguish one group of tourists from another. The categorization of people according to b ran d loyalty yields noncontinuous data. T hus, the prediction of b ran d loyalty m em bership is not possible with regression functions. C o m p a ris o n o f LDA t o F a c to r A n a ly s is a n d C lu s te r A n a ly s is LDA, Factor, and Cluster Analyses all generate a grouping rule based on the d ata collected from th e objects sampled. The difference is th a t LDA uses "independent variables to characterize respondents which fall into different categories defined by the dependent variable" while factor and cluster analyses "identify interdependencies am ong a num ber of variables" (Boyd, 1981). In other words, there is a categorized dependent variable used for prediction in LDA th a t the factor and cluster analyses don’t employ. Factor analysis can also be applied to identify differences am ong individuals. However, differences, or to "when classify the concern individuals into is to identify groups, group discrim inant analysis is more appropriate" (Cooley, 1971). Hence, in order to classify tourists by their destinations (i.e. Michigan versus non- Michigan tourists) in this study, LDA is more appropriate. C luster analysis requires no prior classification of the sample, and is appropriate only when no division of the objects into categories 43 is available. Its objective is to facilitate objective formation of a natural and useful grouping rule based on sim ilarities inherent in the data. In Churchill’s words, the key difference is to get rid of the concept of "criterion variable": Factor analysis and cluster analysis are both methods o f interdependence analysis in that no variable is singled out fo r special treatment as a criterion variable. (Churchill, 1986, p737) In classifying Michigan tourists, if th e grouping information or definition is given, LDA is the appropriate analysis. For example, one may w ant to investigate the behavior of to u rists as it relates to their expenditures. If high and low spending groups are defined in advance based on existing spending information, LDA is appropriate for use in predicting which visitors fall into these two spending categories. On the other hand, if groups are simply formed on the basis of some characteristics or factors yet to be found in the visitors, then factor analysis or cluster analysis should be u sed since the variables used to set u p categories are not known beforehand. The categories or clusters so obtained usually are given a nam e based on the characteristic found within each group. In sum m ary, LDA is not designed for seeking population groupings: it simply assum es th a t su ch groupings already exist. In cases where prior classification inform ation is not available, factor or cluster analysis should be used as a grouping procedure. Since in 44 this study of the Michigan tourism market, predicting Michigan and non-Michigan tourists is the central concern, LDA is th e m ost appropriate technique for analysis. The M echanics of LDA D iscrim inant analysis, like linear regression, is a m ethod based on linear com binations of dependent and independent variables. Its m ain purpose is to distinguish the groups from one an o th er on the basis of their score profiles. This is achieved by constructing a rule which will maximize group centroid separation (i.e. the differences between groups) and minimize within-group dispersion (i.e. the differences within the groups). LDA assum es th a t new observations can be assigned to segm ents of the population on the basis of existing relationships between variables and other sam ple information. Thus, LDA can predict the m em bership of new subjects based on existing information. In automobile m arketing research, discrim inant functions are estim ated from a num ber of demographic or stratification variables. These variables include sex, age, ethnicity, social class, education, occupational statu s, and income. To estimate the buying potential of a new customer, data are analyzed using LDA and the custom er is assigned to an appropriate model buyer group. For example, if income 45 Information is available from p a st custom ers of Cadillac an d Vegas, the discrim inant function for these two brand affiliations can be estim ated from income data. To predict the b ran d th a t a new custom er will choose, the income inform ation from the new custom ers is fed into this function. First, LDA com putes "discrim inant scores." B ased on these scores, new custom ers are assigned to one of the b ran d affiliation groups. This m ethod allows an automobile com pany to conduct efficient promotion of certain brands by targeting highpotential custom ers. Similarly, differences between Michigan and non-M ichigan tou rists can be identified. Suppose th a t the tendency of traveling in Michigan depends on the composite effects of each traveler’s socioeconomic statu s. By applying LDA, the discrim inant functions can be constructed from p ast d ata to discrim inate Michigan travelers from non-Michigan travelers. Once the discrimination function h as been determined, it can be used to predict the new subject’s tendency of choosing Michigan as a trip destination. In the study of tourism , variables such as occupation, education, m arital statu s, and recreation vehicle (RV) ownership, can be used as discrim inant variables. For example, if it is found th a t RV ownership h as a positive im pact on the propensity of traveling to Michigan, th en LDA may be used to arrive a t the conclusion th a t RV owners are more likely th a n non-RV owners to visit Michigan. 46 Theoretically, inform ation on current Michigan to u rists allows LDA to predict future Michigan tourists. The rules governing application of LDA are: (1) the variable to be predicted (dependent variable) m ust be nom inally scaled, and (2) the predictors (independent variables) m u st be on a continuous, interval, or dichotom ous scale. The LDA processes, according to Nunnally (1978), are: (1) to determine w hether differences in score profiles for two or m ore groups are statistically significant; (2) to maximize the discrim ination among groups by com bining the variables in some m anner; an d (3) to establish rules for the placem ent of new individuals into one of th e groups. Given a set of independent variables, infinite linear equations can be constructed for characterizing the groups. The ta sk for LDA is to find the best possible linear combination of variables to predict the groups or categories to which the cases u n d e r investigation belong. The com binations found then serve as a rule for indicating the appropriate categorization for cases whose group s ta tu s is unknown. A linear discrim inant equation can be stated as follows: D = B0 + B|Xj + B X + ... + BpXp 2 2 Where X, is the independent variable score (ex. trip distance, family income, gender, age, m arital status, size of household, etc.); B, is the coefficient estim ated from the participant d ata (ex. Michigan or non- 47 Michigan trip data); and D is the discrim inant score calculated from these coefficients and variables a t the right side of the function. The equation is also called linear discrim inant function (LDF). A ssum ing th a t high perform ance salespeople and low performance salespeople are the two groups to consider, the b est com bination of X’s and B’s is the one th a t resu lts in the greatest difference between the two groups. In other words, the ta sk is to arrive a t the m ost sim ilar D scores for the salespeople in the highperformance group, and for the salespeople in the low performance group, while maximizing the difference between th e D’s of the two groups. G roups are distinguishable if the discrim inant scores of subjects of one group are substantially different from those of another group. Thus, the discrim inating process is realized by choosing values of B such th a t the discrim ination scores differ as m uch as possible between the groups. The distinction between the groups can th en be m easured by the ratio R. r = between-groups sum of squares within-groups sum of squares The discriminating process is most effective when R is maximum. Any other linear combination of the variables will have a smaller ratio. Once the B, values of are determined, the D score (discriminant score) of each case to be predicted is estimated. Based on the calculated D score, the case is assigned to the group whose mean group discriminant score is closest to the one just calculated. CHAPTER HI METHODS This study consists of several efforts. First, it identifies the unique characteristics of Michigan’s tourism m arket and th e factors th a t can be used to determ ine the propensity for to u rists’ choosing Michigan as a travel destination. Thus, a profile of Michigan’s tourism m arket w as produced. Secondly, on the basis of these resu lts, it evaluates the effectiveness of LDA w hen using different travel inform ation to identify Michigan’s potential recreation tourists. The study, as a whole, is designed to provide information useful in planning tourism promotion in Michigan. In th is chapter, th e procedures and m ethods used to conduct these efforts are delineated. Specifically, the scope and resu lts of the d a ta collection effort is described; variable selection and coding m ethods are explained; statistics used in the d ata analysis are presented; and problem s with the LDA model are identified and resolved. 48 49 Data Collection This stu d y centers on recreation-related trip s of over 100 miles one-way or overnight recreation-related trips taken w ithin the last 12 m onths a t the time of interview (fiscal year of 1985). The travel inform ation w as gathered from families in the A reas of Dominant Influence13 (ADIs) of the G reat Lakes region an d the Census M etropolitan A reas/C ensus Areas (CMAs/CAs) w ithin Ontario, Canada. S tates w ithin the G reat Lakes region include Illinois, Indiana, Michigan, M innesota, Ohio, Western Pennsylvania, an d Wisconsin. The trip d a ta used in this study were originally collected by a national m arket research firm, Moore & Associates Company for Ross Roy Inc. The sam ple households were random ly selected from the seven G reat Lake States and the C anadian province of Ontario. Of the 5,000 households interviewed, 4,662 interviews were successful. The 4,662 households reported taking 9,003 trip s w hich m et the criteria established for the survey. The refusal rate was 6.76%. The interviews were conducted via telephone. The questionnaire used in th is stu d y is shown in Appendix F. A screening process was used to filter out non-pleasure (e.g. business) tourists. Each respondent w as asked to supply information on h is /h e r m ost recent (last 12 m onths) trips over 100 miles (one way) or of overnight trips away from home. If a t least one of the trips w as primarily for 13 ADI is an area c u r r ently served b y dom i n a n t a d v e r t i s i n g agencies. 50 pleasure, the Interview w as continued. If not, th e interview was term inated and considered as failed (see questions 4 and 5a in Appendix E). The average length of each survey w as twenty-five to thirty m inutes. Since the screening process filtered o u t m ost ncnpleasure trips in the d ata collected, they were u n d er represented in the d ata base. This precludes use of the resu lts for conclusions regarding the non-pleasure tourism m arket. However, it does not affect the validity of th is study, since it focuses on pleasure trips and all pleasure-related trips were analyzed. D uring the d ata coding, each county w ithin the G reat Lakes Region and Province of Ontario w as assigned a code. Specific travel destinations reported by interviewees were converted into county codes. The event and activity information reported by the interviewees w as also coded according to the stan d ard categories as described in the questionnaire designed for this study (see question 13, 14, 34, and 35 in questionnaire, Appendix E). Data Preparation The m aster database contains inform ation on 9,003 trips. The d ata are coded and stored in an "OSIRIS" database on the Wayne State University Computer. The database for this stu d y w as loaded on com puter tape and sen t to Michigan State University for the CDC750 com puter to use. The database was then downloaded through a 2400 51 b au d modem to a Columbia m icrocom puter system w hich was equipped w ith a 80286 m icroprocessor which ran a t 12 MHz (Mega Hertz), two 51-MB (Mega Byte) hard disks, a 60-MB tape backup, a 8 MHz m ath coprocessor, and 6.64 MB random access memory. The SPSS/PC+ statistical m icrocom puter system program package w as used on this for d ata analyses. This com puter system provides com puting speed over ten tim es faster th a n an IBM PC and adequate storage for efficient com puting for over 9,000 pieces of trip information. It is slower, and smaller in term s of d ata processing speed and d ata storage capacity th a n the mainframe. Also, it h a s less precision. The advantage of using the microcom puter system was convenience and low cost of operation since the system w as owned by the author. From the m aster database, only recreation-related tourist inform ation w as selected for analyses. Strictly business to u rists were not included. Persons traveling for the purposes of bu sin ess and pleasure (i.e., combined pleasure and b u sin ess trips), visiting relatives or friends, shopping, outdoor recreation, sightseeing, touring, special attraction, an d others14 formed the research database. From th is database, 680 b u sin ess trips were deleted, leaving 8,323 pleasure trips, or 92.45% of the total. Both continuous and H The c a t e g o r y of c o n v e n t i o n - o n l y trips. others did not incl u d e a ny business-only trips and 52 nom inal d ata were used for the LDA modeling. Unanswered questions were coded as system -m issing data. Binary type d ata were coded as either 0 or recreational 1. Binary vehicle variables ownership, included weekend sex, trip, m arital group statu s, trip, and inform ation sources used (i.e. trip motivated by inform ation seen or heard via television, radio, newspapers, magazines, billboards, travel agents, friends, relatives, autoclubs, and toll-free phone num bers). The details of the d ata transform ation process are presented in Appendix D. Research Variable S election The selection o f the action variables is som etim es a simple reflection o f the managerial alternatives at hand or the changes under consideration. It may, however, require a high level o f creative imagination. Experimentation is then likely to be helpful and is often absolutely necessary. (Hough, 1970, p322) The relationship between influential variables and tourism behavior is potentially highly diverse. Because of this, using variables th a t influence tourism behavior to construct a model for predicting th a t behavior involves imagination as well as scientific experimentation. In constructing models for th is study, variables were selected for which information was likely to be available and applicable in th e future. Because m ost variables selected in this study, su ch as travelers’ age, income, education, trip duration, and 53 mileage are available from cen su s records, the approach can be easily duplicated in the future. O ther variables, such as travel distance, w hether traveling on a holiday, image of destination, available travel information, and travel season, could be related to th e traveling activities. The travel d ata available for this study are categorized into five types: trip p attern variables, socioeconomic variables, travel information variables, trip attrib u te variables, and Michigan image variables. There are explained below. 1. Trip p a tte rn variables: These variables are those which provide information on trip distance (one way mileage), duration (num ber of nights away from home), spending (by item), size (by num ber of persons), residency (Michigan or non-Michigan), season (warm or cold), weekend trip (travel on weekend or not), an d group trip (travel in a group or not). C ontinuous d ata variables include: trip mileage, duration, spending, an d party size. Nominal d ata variables are residency, season, group trip, and weekend trip. A ssum ing trip pattern is related to the choice of trip destination, these variables can be used to predict the propensity of tourists to choose Michigan as a destination. 2. Socioeconomic variables: These variables describe travelers’ characteristics including: sex, age, race, m arital statu s, household size, occupation, education level, and total family income. Age and household size are continuous variables and the rest of these 54 variables are categorical (nominal) variables. Assuming socioeconomic characteristics affect to u rists’ destination selection, these variable m ay be useful for predicting the propensity of choosing Michigan as a tourist destination. 3. Travel information variables: These variables provide inform ation on w hether the traveler obtained travel information from television, radio, new spapers, magazines, billboards, state toll-free telephone num bers, travel agents, friends or relatives, or state cham bers of commerce. These variables provide binary data. A "yes" answ er is coded as 1, a "no" answ er is coded as 0. Assuming th a t available travel information can affect the choice of a trip destination, these variables m ay be useful for predicting the propensity of choosing Michigan as a to u rist destination. 4. T rip a ttrib u te variables: These variables provide information on how im portant the following item s are to travelers when selecting a destination: scenery quality, re sta u ra n t services, environmental conditions (clear air, good place to stay), hospitality, accessibility, price, and pleasure (good night life, high-prestige vacations, sum m er fun, and w inter fun). These item s are m easured on a scale ranging from 1 to 5. One m eans "not at all important" and five m eans "one of the m ost im portant." Assuming these concerns affect the trip destination choice, these variables m ay be useful in predicting the propensity for choosing Michigan as a tourist destination. 55 M ichigan im ag e variables: These variables are the sam e as 5. attribute variables except th a t they pertain specifically to images of Michigan from the perspectives of tourists. The interviewees were asked how m uch they agreed or disagreed th a t Michigan is known for various specified item s. The m easurem ents ranged from 1 to 10. One m eans "strongly disagree" and ten "strongly agree." A ssum ing th a t images of Michigan affect tourists’ choice of Michigan as a trip destination, these variables are potentially useful in predicting the propensity for traveling in Michigan. Research Design In order to stu d y the unique characteristics of Michigan’s tourists and to predict their choice of Michigan as a traveling destination, the Michigan and to u rists in the database are differentiated non-M ichigan tourists. A variable called into Michigan traveling represents th ese two groups. A value of 1 was assigned to Michigan travelers, and a value of 0 was assigned to non-Michigan travelers. The analytical predictive research was designed to explore: (1) the differences between Michigan travelers and non-Michigan travelers, (2) w hich variables are related to the choice of Michigan as a tourist destination, an d (3) how effectively the various existing travel information variables can be used to predict the propensity for traveling in Michigan. The research then includes the following 56 procedures: (1) the profile an d the com parison of Michigan an d nonMichigan tourism m arkets, (2) significance tests of differences betw een the Michigan and non-M ichigan m arkets, (3) tests of the relationships between the investigated variables and traveling in Michigan, (4) the modeling of Linear Discrim ination Function (LDF), an d (5) the prediction of Michigan tourists. The statistics used in this study include descriptive statistics (frequency counts, percentages, means), crosstabulations, th e Pearson Chi-square test, pooled w ithin-group correlations, Box’s M test, the Hotelling T-test, Lambda, canonical discrim inant function coefficients, and the LDA model. The univariate statistics are used to analyze and te st the differences between Michigan and non-Michigan tourists. The m ultivariate LDA modeling technique is used to distinguish Michigan tourists from non-Michigan tourists. Along with these statistics, a m ulti-stage research design is developed to accomplish th e ta sk s of describing, testing, and modeling the travel data. The design and efforts in each stage are described as follows. Profile o f R e g io n a l T raveling P a tte rn s First the patterns of trip distribution in the study region are profiled in order to draw pictures of tourism m arket stru ctu res. A crosstabulation design is required to show the trip origin-destination, trip direction, and the trip volume th a t each state generated and received. The trip p attern and m arket stru ctu re studies allow the 57 evaluation of the competition th a t Michigan is facing in the m arket areas. Profile o f The Tourism Market: Unique C haracteristics w ith Descriptive S ta tistics In the second stage of the study, an attem pt is m ade to identify the unique characteristics of Michigan’s tourism m arkets. This involves com parison of the profiles of Michigan, non-Michigan, and regional tourism m arkets. Descriptive statistics are u sed to sum m arize tourist characteristics and trip pattern profiles. The traveler characteristic profile provides travelers’ socioeconomic characteristics such as age, gender, education, m arital sta tu s, household size, occupation, education level, total family income, and the state of trip origin. The trip p attern profile includes trip distance (one way mileage), trip duration (number of nights away from home), trip spending (by item), num ber of people in the travelling party, weekend tourists, group tourists, Michigan residency, season of travel, and trip origin and destination. Descriptive statistics used in this study are frequency count, percentage, and m ean. They are used to m easure the variables u n d er investigation and perm it com parison of the average m arket characteristics of Michigan and non-Michigan recreational tourists. 58 T est Variable Independence w ith Chi-square D esign Nominal variables such as travelers’ education level, family Income level, gender, etc., are thought to be related to the choice of traveling In Michigan. Pearson Chi-square is used to lest the hypothesis that the choice of Michigan as a trip destination is Independent of these variables. In some cases the Chi-square design permits the testing of the positive or negative impact direction of Influential variables on tourists’ choice of Michigan as a trip destination. T est o f D ifferences betw een M ichigan A nd N on-M ichigan Tourism Markets w ith H otelling T-test The study exam ines w hether the differences betw een Michigan and non-Michigan to u rists are significant. The Hotelling t-test is used to test the hypothesis th a t no differences exist between Michigan and non-Michigan to u rists in the study variables. The t-test calculates the probability th a t differences in m eans between the two groups m ay occur, and reveals th e unique characteristics of Michigan’s tourism m arket. Developing The Prediction Functions w ith LDA In the LDA modeling stage, the effectiveness of using LDA with several types of existing travel information to predict Michigan tourists is investigated. This process consists of three steps: (1) selecting cases for analysis, (2) selecting variables for the discrim inating model, and 59 (3) estim ating linear discrim inant functions (LDFs). The ta sk of this stage is to develop the functions th a t predict M ichigan to u rists15. Im balance in th e group sizes, however, is a problem in the attem pt to build an LDA model th a t effectively and correctly predicts Michigan tourists. W hen the size of the groups to be predicted are very different, it is easy to obtain a high overall correct prediction rate (CPR). The m ost likely resu lt is a very high CPR for th e larger group and a very low CPR for the sm aller group. For example, the num ber of survivors of the disease AIDS is very small (assum e the num ber is 10) compared to the num ber of AIDS p atients who have died (assum e the num ber is 10,000). Since the size difference is so great, LDA simply gives the larger group a greater weight. Thus, LDA can easily indicate a very high total CPR even though the CPR is actually very low for the survivor group. Despite th e high overall CPR obtained, the discrim inant function would be u seless in th is case due to the low effectiveness of the function in predicting AIDS survivors. In the database, the size of the Michigan trips group an d the non-M ichigan trips group are 790 and 7533 respectively. To elim inate th e problem of imbalanced group size, 790 non-Michigan to u rists were random ly selected for analysis while all 790 Michigan to u rists were included. A prior probability rate of 0.5 w as given for both groups 15 The v a r i a b l e u s e d to p r edict M i c h i g a n t o u r i s t s is M I C H I G A N TRIP. 60 w hich in su red an equal probability of choosing Michigan or not choosing Michigan as a tourist destination. This design may have som ew hat lowered the CPR for non-M ichigan to u rists and overall tourists, b u t th e chance of obtaining a useful discrim inating function for correct prediction of Michigan tourists w as increased. Single variable LDA is used to select the high predictive power variables w hich yield a t least a 75% CPR for the propensity of traveling in Michigan. These variables are th e n used to estim ate LDFs using the m ultivariate LDA process. These variables and five other types of travel inform ation variables16 are used to investigate the com ponent CPRs in LDA. A com puter program called SPSS/PC+, w hich contains LDA procedures developed by the Statistical Package for Social Sciences Inc., is used. The stren g th of association between th e variables used in LDA is also exam ined. To determ ine the optim al LDFs, the strength and n atu re of the dependency of the variables u n d er investigation m ust be assessed. Statistically, high interdependencies am ong predictors can cause m eaningless coefficients. For example, consider two highly correlated variables su ch as mileage and trip spending. The total contribution to the LDA prediction is, in fact, shared by these two 16 The five t y p e s of v a r i a b l e sets u s e d are a: t r i p p a t t e r n v ariable set, s o cio e c o n o m i c v a r i a b l e set, travel i n f ormation v a r i a b l e set, tr i p attribute v a r i a b l e set, a n d M i c h i g a n impress i o n v a riable set. M o r e details on this is p r e s e n t e d in the s e ction on research va ri a b l e s e l e c t i o n e l s e where in this chapter. 61 variables, and th u s the coefficients estim ated for these variables are m eaningless. Highly correlated variables should predictors in LDF (Norusis, not be used as 1986). Therefore, it is necessary to exam ine correlations between variables and eliminate any th a t are highly correlated as copredictors in the discrim inant function. To reduce the chance of obtaining meaningless LDFs, pooled withingroups correlation m atrices are used: (1) to assess th e contribution of individual variables to the discrim inant functions; and (2) to check the interdependencies am ong the variables used in LDFs for predicting Michigan tourists. In LDA, the LDFs are estim ated to derive a pooled w ithin-group variance value of 1. In this study, Box’s M statistic is used to te st the equality of the group covariance m atrices. This test is necessary because LDA requires th a t the covariance m atrices for the two groups (i.e. Michigan and non-Michigan trips in this study) in the analysis m u st be equal to obtain the optim um classification function. Wilk’s lam da (U statistic)17 is used to evaluate the effectiveness of LDFs during the estim ation procedure. The coefficients of LDF are 17 W i l k ' s lamda is the ratio of the wi t h i n - g r o u p s s u m of squa r e s to the t o tal s u m of squares. The values of lamda range fr o m 0 to 1. A lamda v a l u e of 1 i n d i c a t e s an extremely high v a r i a b i l i t y with i n the g r o u p observed, and a lamda of 0 indicates an e x t r e m e l y low v a r i a b i l i t y within the g r o u p observed. The lamda value indicates the total variability attributable to the d i f f e r e n c e s b e t w e e n g r o u p means. The larger the lambda v a l u e is, the larger the t e n d e n c y that the g r o u p means are equal, and the l ower t he lambda value is, the l o w e r the tendency that the g r o u p m e a n s are equal. In o t h e r words, is the p r o p o r t i o n of the total var i a n c e in the d i s c r i m i n a n t scores e x p l a i n e d b y differences among grou p s " (Norusis, 1986). "It not 62 chosen so th a t the ratio of the betw een-groups sum of squares to the w ithin-groups sum of squares is as large as possible. To te st w hether a "good" discrim inant function is obtained, Wilk’s lam da is calculated. In LDA, Wilk’s lam bda is transform ed approxim ate distribution of Chi-square, to a variable and with an the null hypothesis, which assum es the m eans of th e two trip groups are equal, can be tested. Small lam da values are associated with functions th a t have m uch variability between groups and little variability within groups. Thus, a good discrim inate function has m uch between-groups variability and little within-groups variability. The estim ated LDFs are used to predict Michigan tourists. The changes in travel information CPRs th a t resu lt from using different types of travel information as predictor are observed. The effectiveness of using selected functions in predicting Michigan tourists is determ ined by the CPRs. The resu lts are expected to aid in the selection of LDA predictors to be used in predicting Michigan’s overall tourism market. The procedure of classifying travelers is based on the discrim inant scores calculated from the LDFs and group centroids (group means). If the discrim inant score of a case is closer to the centroid of the Michigan trip group th a n it is to the centroid of the non-Michigan trip group, the traveler is classified into the Michigan traveler group. 63 In the following two chapters, the m ajor findings regarding the unique characteristics of the Michigan to u rist m ark et and the effectiveness of using different types of travel inform ation in the LDA model to predict Michigan tourists are presented and discussed. Suggestions are given in term s of how these findings can be useful in identifying Michigan’s potential to u rist m arket. The potential LDA applications in tourism m arketing and the implications for further study of tourism m arket allocations are given in the final chapter. CHAPTER IV ANALYSES RESULTS AND FINDINGS The following are from analyses of a d ata base of 8,323 recreational trips taken betw een 1983 and 1985. Trips taken in 1983 account for only 0.7% of th e total trips. The trips taken in 1984 and 1985 represent 56.9% an d 42.4% of the total trips, respectively (see Table 2). Table 2. The yearly distribution of Trips data. Year No. of Trips 1983 1984 1985 Total Non-response 56 731 3523 4310 13 Percentage 0.7 56.9 42.4 100.0 D atabase: The Great Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce. Five types of data analyses were performed. They are m ark et profiles for the Michigan and non-Michigan travelers, exploration of relationships between variables, significance tests on observed differences, estimation of a Linear D iscrim inant Analysis (LDA) m odels to predict Michigan to u rists and effectiveness evaluations of th e LDA model’s performance. R esults from each of the five analyses are 64 65 presented below In the order ju s t noted. T h e Regional Travel Market Profile In the following m arket profile, a trip origin state is defined as the m arket state, and a trip destination state is defined as the vendor state. It is found th a t th e tourists interviewed in th is study are from 13 m arket sta te s 18, and the destinations include 52 vendor states. The trip percentage distribution by destination and origin is shown in Table 3, an d Table 4 , respectively. The total num ber of trips generated from each m arket state is listed a t the bottom of Table 3 . Read down the colum n in this table to see how trips originating in one sta te are distributed across destinations. For example, am ong 1290 Michigan-origin trips (see bottom line in Table 3), 31.6% included M ichigan as a destination, and 6.0% were destinated for O ntario (see first and second lines in Table 3). The source of each destination’s travelers is shown in Table 4. Read across in this table. The total n u m b er of trips each vendor state received is listed in the last colum n. For example, Michigan received 790 of the total trips included in th is d a ta base (see last volume in Table 4); reading across in the first row note th a t 52.0% were trips originating in Michigan, and 2.9% were trips originating in Ontario. 18 A l l or part of t h e s e market states were i n c l u d e d in th e sample. example, N e w York, a n d P e n n s y l v a n i a were s a m p l e d at on ly the areas which the closest to Mich igan. For are 66 Table 3. Total and distribution (in percent) by destination of trips generated by each market (origin) state in the Great Lakes Travel Monitor Study area. Read down to sec how trips originating in a state arc distributed among destinations. DESTINATION ORIGIN I 1 V M ick! Ontario -gaa -rio Iowa Mlnne -sota Mis­ souri Illi­ nois Indi­ ana Ohio W is­ consin Ken tucky W.VIr glnla New PennsyYork Ivanla (b y % ) H ic llg u Ontario Iowa Minnesota Missouri Illinois Indiana Ohio W isconsin Kentucky W est Virginia New York Pennsylvania 3 1 .6 6 .0 0 .5 0 .9 0 .7 3 .8 2 .6 6 .4 1.5 1.8 0 .2 3 .3 1.5 3 .5 53.8 0 .0 0 .0 0 .3 0 .3 0 .2 1.4 0 .0 0 .0 0 .0 10.3 0 .9 2 .0 0.5 17.3 12.9 6.9 7.4 1.0 2 .0 10.9 0 .5 0 .0 2 .5 1.0 1.8 3.2 1.6 28.1 0 .7 1.8 1.6 1.4 11.8 1.2 0 .2 2.1 0 .7 0 .6 1.0 0 .6 1.0 23.1 7.4 1.3 1.0 0.6 2.2 0.3 2.9 0.6 8 .3 1.5 1.9 1.7 6.2 8.9 4.1 2.3 11.3 1.3 0.2 2.3 1.5 9 .8 0 .7 0 .7 1.5 3 .7 6 .6 15.5 7.3 3 .0 4 .5 0 .6 1.5 2 .0 6.1 3 .8 0 .5 0 .4 0 .6 2.7 3 .8 17.0 0 .8 4 .0 3 .0 4.1 5.3 6 .8 1.5 1.2 7.7 0 .6 8 .2 1.3 0 .6 3 3 .8 1.5 0 .3 0 .9 0 .6 1.4 0 .4 0 .0 0 .7 3 .6 3 .2 4 .3 9 .7 0 .7 16.6 0 .7 2 .2 2 .2 2 .5 0 .6 0 .0 0 .6 0 .6 0 .6 0 .0 13.2 0 .0 2.5 23.3 0 .6 1.9 2.5 10.6 0 .4 0 .0 1.7 3 .8 0 .0 3 .8 0.4 0.4 0.4 25.8 5.5 1.6 4.4 0 .2 0 .4 0.4 0.4 1.8 9.2 0.8 0.6 2.8 6.2 20.2 Mississippi Tennessee Maryland South Carolina Virginia North Dakota South Dakota Alaska California Hawaii Oregon Washington Arizona Colorado Idaho Montana Nevada New Mexico Utah Wyoming Kansas Nebraska Arkansas Louisiana Oklahoma Texas Alabama Delaware D. o f C. Florida Georgia North Carolina New Jersey Connecticut Maine M assachusetts New Hampshire Rhode Island Vermont 0 .4 2 .5 0 .3 1.2 0 .9 0.1 0 .0 0 .2 4.2 0 .9 0.1 0 .5 0 .7 0 .9 0 .0 0 .2 2 .0 0 .2 0.1 0 .2 0 .4 0 .3 0 .4 1.0 0 .3 3 .3 0 .8 0.1 0 .9 11.4 1.0 0 .8 1.2 0 .5 0 .5 0 .6 0 .0 0 .2 0 .0 0 .0 0 .5 0 .0 1.1 0 .2 0 .0 0 .0 0 .0 1.4 0 .8 0 .0 0 .5 0 .3 0 .6 0 .2 0 .0 1.7 0 .2 0 .0 0 .0 0 .2 0 .2 0 .0 0 .2 0 .0 0 .5 0 .2 0 .0 0 .0 17.5 0 .3 0 .5 0 .6 0 .0 0 .3 1.7 0 .2 0 .0 0 .3 1.0 0 .5 0 .0 0.5 0 .0 0 .5 1.5 0 .0 3 .0 1.0 0 .0 0 .5 2 .5 3 .5 0 .0 1.0 1.0 0 .0 0.0 0 .5 0 .5 2.0 2 .0 0 .5 0 .5 2.0 0 .0 0 .0 i.O 4.5 1.5 0 .5 0 .5 1.0 0 .5 1.0 0 .0 0 .0 0 .0 0 .0 1.2 0 .0 0 .0 0.2 1.6 3 .9 1.2 7.6 2.1 0.0 0 .7 3.2 3.0 0 .0 2.1 3.0 0 .0 0 .2 0.7 0 .5 1.2 0 .2 0 .0 0 .5 2.8 0.2 0 .0 0 .5 4.8 0.5 0 .2 0 .0 0 .0 0.2 0 .9 0 .7 0.0 0.0 1.6 3.5 0.6 0.3 1.3 0 .0 0.0 0.6 6.4 2.2 0.0 1.0 1.0 2.6 0 .0 0 .0 1.6 0 .6 0 .0 0.6 1.9 1.0 5.4 3.8 1.6 4.8 0.6 0 .0 0.3 9.6 1.3 1.0 0.3 0.0 0.6 0.0 0.3 0.3 0.3 1.4 2 .9 0 .4 1.1 0.6 0.1 0 .3 0.2 5.6 1.4 0 .2 1.1 2 .0 2.6 0 .0 0 .3 2 .8 0.4 0 .2 0 .5 0.8 0.4 0 .2 1.4 1.1 3 .5 0 .4 0 .0 0 .8 10.9 1.1 0 .7 0 .9 0 .5 0.2 1.1 0 .2 0 .0 0.1 0.1 5.2 0.1 1.5 0 .7 0 .0 0.1 0 .0 2 .4 0 .8 0 .4 0 .6 1.4 1.1 0.1 0 .3 1.3 0 .0 0 .0 0.1 0 .3 0 .6 0 .7 1.4 0 .3 2 .0 0 .8 0.1 0.7 13.8 1.5 1.5 1.0 0.1 0 .3 0 .6 0 .0 0 .0 0.1 0 .3 3 .3 0 .8 2 .8 1.6 0.1 0 .3 0.1 3.1 0 .7 0.1 0 .3 0 .8 1.2 0.1 0 .0 1.6 0 .3 0 .3 0 .3 0.1 0.1 0 .8 0.4 0.1 2 .6 0 .8 0.1 1.2 14.7 1.9 2.7 1.8 0 .4 0 .6 1.0 0.3 0 .0 0.1 0.1 1.2 0 .4 0 .0 0.1 0.1 0 .7 0.1 4 .3 0 .6 0.1 0 .7 1.6 1.8 0.1 0.1 3.1 0 .3 0 .3 0 .3 0 .6 0.1 0 .9 0 .4 0 .3 2 .4 0.1 0 .0 1.0 9 .6 0 .4 0 .3 0 .3 0 .3 0 .4 0 .7 0.1 0 .0 0 .3 0 .4 11.9 0 .7 4 .0 1.8 0 .0 0 .0 0 .0 1.8 1.1 0 .0 0 .4 0 .4 1.1 0 .0 0 .0 0 .7 0 .0 0 .4 0 .0 0 .0 0 .4 0 .4 1.4 0 .0 1.4 1.8 0 .0 2 .5 15.9 1.4 3 .2 0 .4 0 .4 0 .0 0 .0 0 .0 0 .0 0 .4 0 .0 3.1 3.1 9 .4 4 .4 0 .0 0 .0 0 .0 1.3 0.6 0 .0 0 .6 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .6 0 .0 1.9 0.6 0 .6 1.3 18.9 0.6 3 .8 1.3 0 .0 1.3 0 .0 0 .0 0.0 0 .0 0 .0 0 .4 0.4 2 .5 5.5 0 .0 0 .0 0 .0 2.1 0.0 0 .0 0 .0 1.3 1.3 0.0 0.0 1.7 0 .0 0 .0 0.0 0 .4 0 .0 0 .0 0.4 0 .0 4.2 0.8 0.0 0.4 9.7 1.3 1.7 2.5 1.7 1.7 1.7 0 .0 0.8 1.7 0.0 0.8 4.4 3.6 3.8 0 .2 0.0 0 .0 3.4 0.6 0.0 0.6 0.8 1.0 0.0 0.2 2.4 0.0 0 .0 0.6 0.0 0.2 0.0 0.6 0.0 3.0 0.0 1.0 1.4 9.4 0.8 3.8 5.2 0.8 0.4 1.4 0.0 0.2 0.4 No o f Trips by Origin 1290 66 3 202 434 312 1333 711 1290 674 277 159 236 500 Total Trips Counted 8081 Database: The Great Lakes Travel Monitor Study (1983 - 1985). Travel Bureau, Michigan Department of Commerce. Specific Analysis performed by author. 67 Table 4. Total and distribution (in percent] by origin o f each vendor (destination) state’s travelers. Read across to see from where each destination receives its travelers. DESTINATION 1 1 M ichl V -gau - -- ORIGIN ----Onta -rto Iowa Minnr -sota Mis­ souri Illi­ nois Indi­ ana Ohio W is­ consin Ken­ tucky No. of W.Vlr- New Pennsy- Trips by gtnla York lvanla D esti­ nation ( by % ) M ichigan Ontario Iowa Minnesota Missouri Illinois Indiana Ohio W isconsin Kentucky W.VIrgtnta New York Pennsylvania 5 2 .0 13.2 6.1 0 .9 0.7 13.0 11.4 15.2 3.7 11.1 2.7 12.7 7.0 2 .9 60.3 0 .0 0 .0 0 .0 0 .5 0 .3 1.7 0 .0 0 .0 0 .0 20.1 2.2 0 .5 0 .2 35.7 10.0 5.8 4 .0 0 .7 0.7 4.3 0 .5 0 .0 1.5 0.7 1.0 2.4 7.1 46.7 1.2 2.1 2.3 1.1 10.0 2 .4 0.9 2.7 1.1 0 .3 0.5 2.0 1.1 29.9 6.1 1.3 0.6 0 .4 3.4 0.9 2.7 0.7 14.2 3 .4 2 5 .5 8.8 34.4 3 1 .2 18.5 5 .7 2 9 .5 8 .2 2 .7 8.8 7 .4 8 .9 0 .8 5.1 4 .2 10.8 12.4 3 6 .9 9 .6 4.1 15.5 3 .6 3 .2 5.2 11.2 9 .3 7.1 2 .3 3 .7 10.3 18.5 4 5 .3 2.1 27.5 38.7 17.4 28.5 5.9 1.7 8.2 19.9 1.7 14.6 3 .0 0 .7 44.5 4 .8 1.8 1.8 1.5 0 .5 0 .2 0 .0 0.8 4.1 2 .4 4 .0 5.0 0 .4 22.2 1.8 1.8 2 .2 0 .5 0 .2 0 .0 0 .4 0 .4 0 .3 0 .0 3 .9 0 .0 1.9 3 3 .3 0 .3 1.1 0.8 4 .2 1.0 0.0 1.7 2.4 0 .0 1.7 0.2 0 .5 0 .9 18.0 4.6 1.0 3 .7 1.0 0 .8 0 .8 0 .5 3 .0 8 .5 0 .8 1.4 12.6 9.1 3 7 .4 7 90 5 92 98 261 241 3 78 2 98 541 5 12 207 111 339 2 70 Mississippi Tennessee Maiyland S.Carolina Virginia N.Dakota S.Dakotn Alaska California Hawaii Oregon W ashington Arizona Colorado Idaho Montana Nevada New Mexico Utah Wyoming Kansas Nebraska Arkansas Louisiana Oklahoma Texas Alabama Delaware D. o f C. Florida Georgia N.Carollna New Jersey Connecticut Maine M assachu. N.Hampshlre R lsla n d Vermont 13.2 14.0 7.0 10.4 11.9 6 .7 0 .0 21 .4 17.0 15.0 11.1 12.0 8.9 8.7 0 .0 9.1 15.8 16.7 8.3 12.0 14.3 13.3 8.1 17.8 12.5 18.9 2 0 .4 10.0 15.1 14.9 14.0 8.5 14.6 2 0 .6 18.9 11.0 0 .0 4 2 .9 0 .0 0 .0 1.3 0 .0 4 .9 1.0 0 .0 0 .0 0 .0 2 .8 6 .3 0 .0 6 .0 2 .0 3 .2 25.0 0 .0 6 .7 5.6 0 .0 0 .0 2 .9 3 .3 0 .0 1.4 0 .0 1.4 2 .0 0 .0 0 .0 11.7 2 .2 2 .5 3 .9 0 .0 5.4 15.1 8 .3 0 .0 12.5 5.3 0 .4 0.0 1.4 0 .0 6.7 8.3 0.0 1.9 2 .5 0.0 2.0 5.0 5.6 0 .0 9. 1.2 0 .0 0.0 4.0 2 .9 13.3 6 .5 1.4 3.1 1.8 0 .0 0 .0 2.7 0 .9 3.2 0 .8 1.0 5.9 2.7 2.7 0.0 0.0 0 .0 0 .0 2.2 0.0 0 .0 1.0 46.7 47.2 35.7 10.4 11.3 0.0 6.0 13.9 10.3 0.0 40.9 7.9 0.0 8.3 12.0 5.7 16.7 1.6 0.0 6.3 5.4 2.0 0.0 2.7 2.1 2.2 0.8 0.0 0.0 2.7 5.5 25.0 0.0 0.0 13.2 4.8 3.5 0.7 4.0 0.0 0.0 14.3 6.3 8.8 0.0 6 .0 3.0 6.3 0.0 0 .0 3.0 11.1 0.0 8.0 17.1 10.0 27.4 16.4 15.6 6.8 4.1 0.0 1.4 3.0 4.3 2.5 1.0 0 .0 5.4 0 .0 8.3 14.3 6.3 50.0 16.7 8.8 10.4 7.9 6 .7 11.1 14.3 2 3 .7 2 2 .5 3 3 .3 2 3 .0 2 5 .7 2 7 .0 0 .0 18.2 22.4 27 .8 2 5 .0 24 .0 28.6 16.7 19.4 24 .7 4 6 .9 2 0 .7 10.2 0 .0 15.1 14.7 16.1 7 .6 11.7 20.6 5.4 19.2 16.7 0 .0 6 .3 2 .6 16.2 1.8 9 .0 5.0 0 .0 2 .8 0 .0 5.4 7.5 33.3 8.0 9 .9 6 .3 2 5 .0 9.1 5.5 0 .0 0 .0 4 .0 5.7 13.3 8.1 13.7 6.3 8 .3 12.2 10.0 6 .8 9.9 11.8 9 .3 6 .8 2 .9 5 .4 5.5 0.0 0 .0 6 .3 10.5 21.1 21.1 27.8 22.8 6 .7 11.1 7.1 13.9 12.5 11.1 8.0 10.9 14.3 2 5 .0 0 .0 13.9 27.8 33.3 16.0 5.7 3 .3 17.7 8 .2 3.1 16.7 2 4 .5 2 0 .0 23.3 21.5 30.1 33.1 25.2 17.6 2 4 .3 19.2 33.3 0 .0 12.5 2 .6 3.5 5.3 0 .0 1.0 6 .7 13.9 7.1 9.1 5.0 11.1 10.0 10.9 9 .5 25.0 4 .5 12.7 11.1 16.7 8.0 11.4 3.3 9 .7 4.1 6 .3 7.2 2 .0 0 .0 9 .6 6 .6 3.2 1.7 1.9 5.9 8.1 6 .8 8.3 0 .0 12.5 2.6 14.5 3.5 7.6 5.0 0 .0 0.0 0 .0 1.6 3.8 0.0 2 .0 1.0 2 .4 0 .0 0 .0 1.2 0 .0 8.3 0 .0 0.0 3.3 1.6 5.5 0 .0 1.8 10.2 0.0 9.6 4.5 4 .3 7.6 1.0 2.9 0 .0 0 .0 0.0 0 .0 6 .3 0 .0 2 .2 8 .8 10.4 6 .9 0 .0 0 .0 0 .0 0 .6 1.3 0 .0 2 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 1.4 0 .0 1.4 2 .0 10.0 2 .7 3 .0 1.1 5.1 1.9 0 .0 5 .4 0 .0 0 .0 0 .0 0 .0 0 .0 0.4 1.8 4.2 12.9 0.0 0.0 0 .0 1.6 0.0 0.0 0 .0 3.0 2.4 0.0 0.0 2.4 0.0 0.0 0.0 2.9 0.0 0.0 1.4 0.0 4.5 4.1 0.0 1.4 2.3 3.2 3.4 5.8 11.8 10.8 5.5 0.0 28.6 25.0 0 .0 1.8 3 8 .6 12.5 18.8 6 .7 0 .0 0 .0 5 .4 3 .8 0 .0 6 .0 4 .0 4 .0 0 .0 4 .5 7 .3 0 .0 0 .0 12.0 0 .0 3 .3 0 .0 4 .i 0 .0 6 .8 0 .0 5 0 .0 9 .6 4 .8 4 .3 16.1 2 5 .2 11.8 5 .4 9 .6 0 .0 14.3 12.5 38 228 57 144 101 15 36 14 317 80 9 50 101 126 4 22 165 18 12 25 35 30 62 73 32 222 49 10 73 988 93 118 103 34 37 73 12 7 16 Database: The Great Lakes Travel Monitor Study (1983 - 1985). Travel Bureau, Michigan Department of Commerce. Specific Analysis performed by author. 68 As producers of trips, Illinois, Michigan, and Ohio represent the first, second, and third largest tourism m arkets respectively, accounting for nearly 50% of the total to u rists19. Michigan h as a 15.5% share of the total market. Among the to u rists originating in Michigan, 31.6% rem ain in Michigan and 68.4% visit other states. On the receiving side, Florida, Michigan and Ontario represent the first, second, and third largest tourism trip receiving states. These three vendor states share nearly 30% of total m ark et in The Great Lake region, accounting for 11.9%, 9.5%, and 8% of the m arket respectively. The seasons have been found to affect traveling p attern s and m arket shares. The shift in the traveling m arket p attern s caused by the season change is shown in Table 5 to Table ©. Table 5 shows th a t the m arket shifts between Michigan and Florida w hen the season changes. Michigan is the num ber one recreation tourism vendor state with 10.5% of m arket share during the w arm 20 season and is the num ber two vendor state with 7.4% of m arket sh are during the cold season. Florida is the num ber one cold-season tourism vendor state with 20.6% of the m arket share and is th e n u m b er three warmseason tourism vendor state with 8% of m arket share. 19 Note, however, that s ampling was c o n f i n e d to the A D I s (the Areas of D ominant Influence) a n d t he CMAs/ CAs (Census M e t r o p o l i t a n A r e a s / C e n s u s Areas) within the states in the study area. Thus the s a m p l i n g rate for some states is less than others. 20 In this study, w a r m seas on refers to the period from May and cold season includes the m o n t h s from Oc t o b e r to April. to October, 69 Table 5. Vendor (Destination) states’ share of the cold and warm season1 travel markets*. Destination State Warm Season Trip Cold Season Trips (%) Florida Michigan Ontario Ohio W isconsin Illinois New York California Indiana Pennsylvania Minnesota Missouri Tennessee Texas Kentucky Nevada South Carolina Colorado North Carolina West Virginia Iowa Virginia Arizona New Jersey Georgia Hawaii D. o f C. Louisiana M assachusetts Arkansas Maryland Alabama Washington Mississippi Oklahoma Connecticut Maine South Dakota Kansas Nebraska Montana Wyoming New Mexico Vermont North Dakota Alaska Utah New Hampshire Rhode Island Oregon Delaware Idaho 8.0 10.5 8.2 7.4 7.1 4.5 4.0 3.4 3.9 3.4 3.3 3.2 3.1 2.0 2.8 1.6 1.9 1.5 1.6 1.3 1.2 1.4 0.8 1.4 1.1 0.7 0.8 0.8 1.0 0.8 0.7 0.6 0.7 0.5 0.4 0.4 0.6 0.5 0.4 0.4 0.3 0.4 0.1 0.2 0.2 0.2 0.1 0.2 0.1 0.1 0.1 0.0 (%) (%) (3)* (1) (2) (4) (5) (6) (7) (8) (9) (10) 20.6 7.4 5.0 4.7 4.3 4.8 4.3 4.9 3.0 3.1 2.8 2.2 1.9 4.1 1.9 2.8 1.3 1.5 1.1 1.5 0.9 0.9 2.0 0.8 1.3 1.5 1.0 1.0 0.7 0.7 0.6 0.6 0.4 0.3 0.3 0.5 0.1 0.2 0.5 0.4 0.1 0.2 0.4 0.1 0.2 0.1 0.2 0.0 0.1 0.1 0.2 0.1 Total (1) (2) (3) (6) (7) (5) (7) (4) (10) (9) (8) 12.0 9 .5 7.2 6.5 6.2 4.6 4.1 3.8 3.6 3.3 3.2 2.9 2.8 2.7 2.5 2.0 1.7 1.5 1.4 1.3 1.2 1.2 1.2 1.2 1.1 1.0 0 .9 0.9 0.9 0.8 0.7 0 .6 0.6 0.5 0.4 0.4 0.4 0.4 0.4 0.4 0.3 0.3 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0 .0 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) 1. The cold and warm season were defined a s follows: May 1st through October 31st Is warm season, November 1st through April 30th is cold season. 2. The origin states are Illinois, Indiana, Michigan, Minnesota. Ohio, W. Pennsylvania, Wisconsin, and Ontario. 3. The number in parentheses shows the ranking of destination states In the column. 4. Database: The Great Lakes Travel Monitor Study (1983 - 1985). 70 Table 6. Regional tourism market comparison1 between warm and cold tourism. Origin State (Province) Recreational Trips Generated ....... ................................. Warm Season Cold Season (%) Ohio Illinois Michigan. Indiana Wisconsin Ontario Pennsylvania Minnesota Missouri Kentucky 3.4(10) New York Iowa West Virginia 17.9 15.8 15.5 8.9 7.9 7.7 6.2 4.9 3.9 3.6 (%) 16.4 16.7 15.9 7.9 8.7 8.7 5.7 5.9 3.5 2.9 (If (2) (3) (4) (5) (6) (7) (8) (9) (10) (%) (2) (1) (3) (4) (5) (6) (7) (8) (9) (10) 3.2 2.7 1.8 2.8 2.3 2.0 Total 17.4 16.1 15.6 8.6 8.2 8.0 6.0 5.3 3.8 (1) (2) (3) (4) (5) (6) (7) (8) (9) 2.9 2.4 1.9 1. In this table, trips to all destinations are compared. 2. Ranking in the column. (Database: The Great Lakes Travel Monitor Study (1983 - 1985), Travel Bureau, Michigan Department of Commerce). D uring the warm season, the perform ance difference between Michigan and Florida is not as significant as th a t during cold season. It shows th a t Michigan h as the disadvantage in competing for the travel m arket with Florida during the cold season. Across the full year, Florida captures 2.5% more of th e stu d y region’s trips th a n does Michigan. Table 6 shows th a t during the warm season, Ohio is the biggest source of trips generating 17.9% of the total tourists in the region. D uring the cold season, Illinois is the largest tourism m arket (origin) state producing 16.7% of the total to u rists in the region. Michigan 71 produces 15.5% of the warm season market. 15.9% of th e cold season m arket, and for th e full year is the third largest trip producer in the region. Ohio, Illinois, and Michigan represent the m ajor tourism m arkets (buyers) in th e G reat Lake region. Totally, these three states produce close to 50% of the entire tourism m arket generated in the study region. In Michigan’s conclusion, three Ontario, Ohio, and biggest competing neighbor W isconsin states; represent Florida California represent Michigan's two biggest d istan t tourism and rival states. Totally, over 36% of the recreation tourism m ark et is taken by the latter three com peting states (see Table 5). The M ichigan Recreational Tourism Market This section recreational tourism describes m arket. the characteristics Descriptive inform ation of Michigan’s is given to compare the Michigan and non-Michigan recreation trip m arkets. Recall th a t recreation trips were defined as trips for visiting relatives, friends, outdoor recreation, sight seeing/touring, shopping, a specific attraction, and other pleasure related trips. In th e following, tourists with Michigan destinations are defined as Michigan tourists. During the study period, Michigan received 790 recreation trips and generated 1,290 recreation trips. Thus, Michigan was a net exporter of 500 recreation trips. 72 Among 790 trips received by Michigan, ab o u t 52.0% are Michigan resident traveling parties, and 48% are out-of-state tourists (See Table 4 for details). Michigan trips by residents an d out-of-state tou rists account for 5.5% and 4% of the total m arket in the regional, respectively. Totally, Michigan’s m arket sh are is 9.5% of the total regional m arket (See Table 4). Besides the domestic market, two neighboring states, Illinois and Ohio, are Michigan’s prim ary out-of-state m arket, contributing over 25% of the total trips to Michigan. In the G reat Lake region, Michigan is one of the m ajor tourism vendor (destination) states, dominating th e warm season tourism m arket and is the second largest vendor state during cold season (see Table 5). Michigan’s warm and cold season sh ares of the study region’s tourism m arket are 10% and 7.4%, respectively (see Table 5). Table 7 shows th a t Michigan’s m ajor out-of-state tourism m arkets shift between Indiana and Illinois w hen seasons shift. Beside Michigan’s domestic tourism m arket (Michigan residents who stayed in Michigan), Indiana is Michigan’s the largest w arm -season tourism m arket and provides Michigan 18.9% of its total w arm -season tourists. During the cold season, Illinois, is the largest cold-season out-of-state tourism m arket and provides Michigan 17.6% of its total cold-season tourists. 73 In conclusion, Illinois, Ohio, and Indiana rep resen t Michigan’s three m ajor out-of-state tourism m arkets. Totally, over 34% (see Table 7) of Michigan trips are generated from th ese th ree states. Throughout the year, Illinois represents Michigan’s biggest supplier of to u rists and provides Michigan with over 14% of its total tourists. Table 7. Sources by state of Michigan’s w arm an d cold season recreation travelers. Origin State (Province) Michigan Recreational Trip ------------------ -------------------W arm Season Cold Season (%) M ichigan Illinois Ohio Indiana Wisconsin Ontario M innesota Pennsylvania New York Iowa 0.5(10) Kentucky West Virginia Missouri 52.3 13.0 11.2 18.9 5.6 2.5 1.4 1.0 0.8 0.5 0.5 0.5 0.3 (%) (l)1 (3) (4) (2) (5) (6) (7) (8) (9) (10) 51.3 (1) 17.6 (2) 11.4 (3) 7.9 (4) 6.7 (5) 4.1 (6) 4.1 (7) 1.0 (8) 0.5 (9) 0.5(10) 0.5 0.5 0.8 Total (%) 52.0 14.2 11.2 8.9 5.9 2.9 1.0 1.0 0.8 (1) (2) (3) (4) (5) (6) (7) (8) (9) 0.5 0.5 0.3 1. Ranking in th e column. (Database: The G reat Lakes Travel Monitor Study (1983 - 1985), Travel Bureau, Michigan D epartm ent of Commerce). Table 8 com pares the general characteristics between Michigan and non-M ichigan travelers. Relatively speaking, Michigan recreation travelers significantly: 1. are younger; 2. travel In sm aller parties; 3. travel sh o rter distances; 4. don’t stay a s long; 5. don’t spend as m any nights in hotels, motels friend’s or relative’s houses; 6. spend a higher percentage of overnight stay s in rented cabins, self-owned cabins, public and private te n t cam pgrounds, public RV cam pgrounds, and spa resorts; 7. expend a b it less during their trip an d a t specific destination; and 8. rate their trip an d specific destination som ew hat lower. 75 Table 8. Comparison of the characteristics of Michigan and nonMichigan recreational travelers, th eir trips and trip ratings. C haracteristics Michigan Non-Michigan D estination Trips D estination Trips Age (Year) P arty Size (Person) One-Way Mileage (Mile) Trip D uration (Day) Percentage of Night S tay in: a. b. c. d. e. Hotel Motel Rented Cabin Own Cabin Public Campground: Tent RV. Total 39.26 4.33 347.71 5.07 40.52 4.36 861.40 8.09 40.40 4.36 812.82 7.79 10.90% 9.24% 10.19% 13.27% 19.76% 10.47% 5.61% 5.16% 17.20% 10.45% 6.30% 6.22% 4.74% 4.27% 1.33% 2.95% 1.68% 3.37% 2.37% 4.98% 1.03% 2.66% 1.18% 3.03% 6.40% 27.44% 2.17% 4.03% 10.47% 33.04% 1.33% 6.19% 10.28% 32.20% 1.52% 5.57% 4.06 4.03 3.49 3.42 4.12 4.06 3.53 3.47 4.11 4.05 3.52 3.46 f. Private Campground: Tent RV. g. Friend’s House h. Relative’s House i. R esort/S pa j. O ther Rating Overall Trip Rating Destination Trip Expectation D estination Expectation (Database: The Great Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). 76 Table 9 shows th at, in descending order, the prim ary purpose of Michigan trips is: (1) visiting relatives, (2) outdoor recreation, (3) sightseeing, (4) visiting friends, (5) special attractions, (6) b u sin ess/p leasu re, and (7) shopping. The biggest m arket segment is travelers visiting relatives, which accounts for 35.32% of the total tourism m arket received by Michigan. The sm allest m arket segm ent is shoppers who accounted for only 1.14% on Michigan’s total trips. The Michigan tourism m arket includes a higher percentage of outdoor recreation tourists th an does the regional m arket indicating an outstanding dem and for Michigan’s outdoor recreation products and services. The percentage of all other types of purposes for Michigan trips are relatively lower th a n for non-M ichigan destinations, however, sightseeing stan d s out as being the m ost different. This may resu lt from a tendency for Michigan travelers on combination recreation and sightseeing trips to report outdoor recreation as their prim ary purpose more frequently th a n do non-M ichigan travelers on such dual purpose trips. Table 9 shows this comparison. ; 77 Table 9. Comparison by trip purpose for Michigan an d Non-Michigan trips. Purpose Visit relatives O utdoor recreation Sightseeing Visit friends Special attraction O ther B usiness/pleasure Shopping Total — Trip D estination — Michigan Non-Michigan Total (%) (%) (%) 35.32 28.86 13.92 9.24 5.44 3.80 2.28 1.14 100.00 35.70 18.00 19.50 10.60 7.30 3.60 3.50 1.7 100.00 35.62 19.16 18.98 10.49 7.20 3.63 3.42 1.64 100.00 (Database: The G reat Lakes Travel Monitor Study (1983 - 1985), Travel Bureau, Michigan D epartm ent of Commerce). As in m ost n o rth ern states in the country, the majority of Michigan’s tourism is concentrated in the sum m er m onths. Over half of Michigan destination trips take place between Ju ly and September. As can be seen in Table 10, A ugust is the bu siest tourism m onth in Michigan. It accounts for about 18.1% of the yearly m arket. On the other hand, Ja n u a ry tourism account for only 2.5% of the yearly m arket. Table 10 shows th a t there are different degrees of peaking in the trip percentage distribution p attern between Michigan and nonMichigan trips. The w arm season peaking for Michigan trips is more 78 pronounced th a n for non-Michigan trips21. Statistics in colum n two and three in Table 10 show the trip distribution peaking changes w hen non-M ichigan trips outside the study region are excluded. The percentage shares are higher during the warm season (May, Ju n e , July, October, and November), and are lower in the rest of m o n ths w hen the non-Michigan trips outside study region are included. Table 10. D istribution of Michigan and non-Michigan trips by m o n th 1. Month Ja n u a ry February March April May Ju n e Ju ly A ugust Septem ber October November December Total - Trip D e stin a tio n -------— Non-Michigan — Inside Region Inside Region Michigan (%) 2.5 4.1 3.5 3.2 5.7 10.9 18.0 18.1 14.7 8.1 6.7 4.7 100.0 - (%)2 3.5 3.8 6.3 7.0 6.8 10.6 15.5 15.6 11.0 8.2 6.0 5.6 100.0 (%)3 3.2 2.4 3.4 5.9 7.6 12.9 16.6 14.9 10.5 8.5 9.3 4.9 100.0 Total (%) 3.4 3.8 6.0 6.6 6.8 10.6 15.8 15.9 11.3 8.1 6.1 5.5 100.0 1. The origin states include Illinois, Indiana, Michigan, M innesota, Ohio, W estern Pennsylvania, and Wisconsin. 2 . This column show s the percentage distribution of the non-Michigan trips including outside study region, e.g. Florida. 3. This column shows the percentage distribution of the non-Michigan trips in study region only. (Database: The G reat Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). 21 Relatively, the M i c h i g a n trip p e r c e n t a g e d i s t r i b u t i o n p a t t e r n p ea ks mo re s i g n i f i c a n t l y d u r ing the w a r m season than n o n - M i c h i g a n trips (in the case of e i th er i n c l u d i n g the n o n -Mich igan trips outs ide study region or n o t ) . 79 These distribution p attern s In Table 10 indicate: (1) Michigan’s tourism industry perform ances is marginally better th a n non-Michigan areas in the w arm season, while its m arket share is lower except for February th a n its non-M ichigan competition for the re st of the year; (2) a strong dem and for the warm season in Michigan since the trip percentage shares in Michigan are higher during w arm season in Michigan th a n non-M ichigan areas: (3) the tourism m arket competition from th e outside-study-region states have affected the trip distribution in different seasons; (4) the outside-study-region trips, Florida for example, h a s some obvious w inter season attractions such as winter resorts w hich resu lt in lower peaking in th e trip distribution pattern when these trip s are included (see colum n two and three in Table 10). The following tables present the percentage distribution comparisons between various types (i.e. weekend trip, group trip, etc.) of Michigan trips an d non-M ichigan trips. The significance tests of the differences between Michigan trips and non-M ichigan trips will be presented later. The percentage distribution com parisons between various types (i.e. weekend trip, group trip, etc.) of Michigan trips and non-Michigan trips are presented in Table 11 to Table 19. W eekend travelers account for over four-fifths of Michigan’s m arket. Com pared with the 80 non-M ichigan tourism m arket, Michigan h a s a higher percentage of weekend tourists, indicating a strong w eekend pleasure travel dem and in Michigan (see Table 11). Table 11. D istribution of Weekend trips by Michigan and nonMichigan trip destinations. Weekend Trip — Trip D estination — Michigan Non-Michigan (%) 81.2 18.8 100.0 Yes No Total (%) 76.2 23.8 100.0 Total (%) 76.9 23.1 100.0 (Database: The G reat Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). Table 12 shows th a t group to u rists account for only 2.9% of Michigan’s total pleasure travel m arket. This percentage is lower th an th a t in the non-M ichigan m arket indicating th a t group tourism is less popular in Michigan. Table 12. D istribution of group trips between Michigan and nonMichigan trip destinations. Group Trip Yes No Total — Trip D estination — Michigan Trip Non-Michigan (%) 2.9 97.1 100.0 (%) 4.8 95.2 100,0 Total (%) 4.6 95.4 100.0 (Database: The G reat Lakes Travel M onitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). 81 Compared with other areas in the study region, Michigan a ttra c ts a greater percentage of recreation vehicle owners (19.7% v ersu s 14.9%) indicating th a t Michigan is more popular to recreational vehicle owners th a n resource other areas based attractions probably because or other of superior differences in n atu ral charges and regulation (see Table 13). Table 13. Recreation vehicle ownership among Michigan an d nonMichigan pleasure travelers. Own R.V. Yes No Total — Trip Destination — Michigan Trip Non-Michigan Total (%) (%) (%) 19.7 80.3 100.0 14.9 85.1 100.0 15.4 84.6 100.0 (Database: The Great Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). The telephone survey results show th a t females account for over 60% of all respondents (see Table 14). Females account for a higher percent (65.2%) of Michigan trip respondents th an for non-M ichigan trips (63.2%). However, there is a bias in these results. Since women are more likely to be home th a n men, the chance for women to answ er the phone is larger. Also these results can only be applied to respondents’ gender distribution b u t not to travelers per se. If knowledge regarding travelers’s gender distribution in the study region 82 is desired, fu rth er adjustm ent is required. Table 14. Pleasure travel m arket com parison by gender - Michigan, non-Michigan, an d total region. Respondent Gender Michigan Trip D estination — Non-Michigan (%) (%) Male Female Total 34.8 65.2 100.0 36.8 63.2 100.0 Total (%) 36.6 63.4 100.0 (Database: The G reat Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). The resu lts from the telephone survey show th a t the percentage of m arried respondents interviewed are larger for Michigan trips than for non-M ichigan trips. In Michigan, 73.1% of adult pleasure trip participants are reported married while only 68.1% of non-Michigan pleasure trip participants are married (see Table 15). Though the results indicate th a t family tourism in Michigan is stronger th an in non-Michigan areas, it is possible th a t the telephone survey design and time of interviews may have favored reaching married non­ employed spouses. T hus, there is a bias in the d ata base toward married non-employed women. 83 Table 15. Adult pleasure travel m arket by m artial s ta tu s - Michigan, non-M ichigan, and total region. — Trip D estination — Michigan Non-Michigan M arital S tatu s Married U nm arried Total Total (%) (%) (%) 73.1 26.9 100.0 68.1 31.8 100.0 68.6 31.3 100.0 (Database: The G reat Lakes Travel Monitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). It was found th a t over 95% of respondents who reported Michigan pleasure trips have a t least a high school education com pared with only 91.8% for the respondents who reported nonMichigan trips (see Table 16). About 56% of Michigan traveling respondents have a t least some college experience com pared with 54.4% of non-Michigan traveling respondents. Though the telephone survey results show th a t Michigan’s high-education m arket segm ent is larger th a n th a t of the non-M ichigan destinations, there is again likely to be a bias toward female respondents because women are more likely to be home to answ er the phone th a n men. These resu lts m ay not accurately reflect the education level distributions of actual travelers. 84 Table 16. Adult pleasure travel m ark et com parison by level of education - Michigan, non-Michigan, an d total region. — Trip D estination — Michigan Non-Michigan Education Level (%) 5.0 33.6 5.4 25.2 23.4 7.4 0.0 100.0 Less th a n high school High school graduate Trade, technical Some college College graduate Post degree Refuse Total Total (%) 8.2 31.7 5.8 22.6 23.0 8.7 0.1 100.0 (%) 7.9 30.9 5.8 22.8 23.0 8.5 0.1 100.0 (Database: The G reat Lakes Travel M onitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). The race distribution of Michigan pleasure travelers is not m uch different th a n th a t of non-Michigan travelers (see Table 17). Michigan’s tourism m arket is composed of 93.4% white, 5.9% black and 0.5% of Hispanic people. Table 17. Adult pleasure travel m arket Michigan, non-Michigan, and total region. Race White Black Hispanic O ther Total com parison — Trip D estination — Non-Michigan Michigan (%) 93.4 5.9 0.5 0.1 100.0 by race - Total (%) (%) 92.7 6.4 0.5 0.4 92.7 6.4 0.5 0.4 100.0 100.0 (Database: The Great Lakes Travel M onitor Study (1983 - 1985), Travel B ureau, Michigan D epartm ent of Commerce). 85 Middle income ($20,000 to $40,000) travelers compose 53.5% of Michigan tourism m arket representing the sta te ’s m ajor m arket. This portion com pares favorably with both the 48.1% sh are in the nonMichigan tourism m arket, and the 48.6% in regional m arket (see Table 18). Michigan, however, attracts relatively fewer high income travelers (69,000+) th a n does the rest of the stu d y region. Table 18. Adult pleasure travel m arket com parison by income Michigan, non-Michigan, and total region. — Trip D estination — Michigan Non-Michigan Income Less th en 10,000 to 20,000 to 30,000 to 40,000 to 50,000 to 60,000 to 70,000 or Total 10,000 19,999 29,999 39,999 49,999 59,999 69,999 more Total (%) (%) (%) 6.0 15.2 30.0 23.5 12.2 7.6 1.1 4.6 100.0 6.1 15.3 25.8 22.3 13.9 7.9 2.9 5.7 100.0 6.1 15.4 26.2 22.4 13.7 7.9 2.7 5.6 100.0 (Database: The G reat Lakes Travel Monitor S tudy (1983 - 1985), Travel Bureau, Michigan D epartm ent of Commerce). H ypothesis T ests There are four general hypotheses tested in th is study. These general hypotheses contain various specific hypotheses which are tested using the t-test and Chi-square statistic. The results of these 86 te sts are tabulated in Tables 19 and 2 0 which appear later in this Chapter. The trip p attern differences found between Michigan and non-Michigan trips are also discussed including the direction of the im pacts on propensity to travel in Michigan. H ypothesis 1 There are no significant differences in trip between Michigan and non-M ichigan tourists. p attern s H ypothesis 1-a. There is no significant difference in one way trip mileage between Michigan and non-M ichigan pleasure trips. This null hypothesis is rejected a t a p < .05 (see Table 19). This resu lt indicates th a t travel distance of Michigan pleasure trip s are significantly different from those of non-Michigan trips. The average travel distance of Michigan trips is 347 mile, and the average travel distance of non-M ichigan trips is 861 miles. Michigan pleasure trips are significantly shorter th a n non-M ichigan trips. 87 Table 19. T-tests of differences between Michigan and non-Michigan recreation trips for selected variables. Hypothesis Number Variable Tested 1-a. One way m iles traveled 1-b. Total side trip miles 1-c. Trip length in day* 1-d. Mean nights stayed In hotels 1-e. Mean nights stayed In motels 1-f. Mean nights stayed in public tent campgrounds 1-g. Mean nights stayed In friend houses 1-h. Mean nights stayed in relative houses 1-1. Mean nights stayed in other states 1-J. Amount spent on transportation^) 1-k. Amount spent on lodging($) 1-1. Amount spent on meals($) 1-m. Amount spent on entertal nment($) 1-n. Amount spent on mlscellaneous($) 1-o. Good restaurant* 1-p. Good place to stay* 1 -a. High prestige* 1-r. Good night life* l-s . Winter fun3 1-t. Good Scenaiy3 1-u. Good Restaurant3 1-v. Friendly People3 1-x. Easy to Get to3 1-y. Reasonable Price3 1-aa. Good Place to Stay3 1-ab. Summer Fun3 1-ac. High Prestige3 1-ad. Clean Air3 1-ae. Good Night Life3 1-af. Winter Fun3 1-ag. Likely to visit destination state again3 Michigan Trips Non-Michigan Trips Test Results ...... M e a n .................... Significance 347.71 861.40 85.56 153.37 • 5.07 8.09 • 0.46 1.33 * 0.39 0.71 • 0.20 0.09 • 0.27 0.71 • 0.12 2.24 • 1.98 1.83 • 563.15 643.60 • 556.34 623.88 * 568.05 620.84 * 560.01 616.51 • 552.39 4.15 4.33 2.51 2.97 3.07 8.58 7.43 7.89 8.41 7.01 7.79 8.53 6.01 7.59 6.68 8.42 609.32 4.29 4.40 2.63 3.17 2.82 7.42 6.41 6.76 7.49 6.22 6.88 7.20 5.25 6.46 6.02 7.51 * • • • * * • • • • • » * • • * • 4.53 4.45 • 1. *, t-tests significant at 0 .0 5 level of significance. 2. Number of days derived by adding one to total nights spent away from home. 3. Ratings were based on a five point scale with 5 very likely. 4. The ratings were based on a five point scale with one being not at all important to the trip and five being very important to the trip. 5. The ratings were based on a ten point scale with one strongly disagree that Michigan is know for the item stated and ten being strongly agree that Michigan is know for the item stated. (Database: The Great Lakes Travel Monitor Study (1983 1985), Travel Bureau, Michigan Department of Commerce). 88 H ypothesis 1-b. There is no significant difference in total side trip mileage between Michigan and non-M ichigan pleasure trips. This null hypothesis is rejected a t a p < .05 (see Table 19). This result indicates th a t side trip traveling distance of Michigan trips are significantly different from those of non-Michigan trips. The average side trip distance of Michigan trips is 85.56 miles, an d the average side trip distance of non-M ichigan trips is 153.37 miles. Michigan trips’s average side trip distance is significantly sh o rter th a n the nonMichigan trip s's. H ypothesis 1-c. There is no significant difference in trip d u ratio n 22 between Michigan and non-M ichigan tourists. This null hypothesis is rejected at a p < .05 indicating th a t there is a significant difference in duration between Michigan and non-Michigan trips. The average trip duration is 5.07 days and 8.09 days for Michigan and non-M ichigan trips, respectively. The average length of Michigan trips is significantly shorter th a n the average nonMichigan trips. 22 Trip dur ation is d e r i v e d nights spent away from home. by adding one to the number of r e p orte d 89 Hypotheses 1-d, 1-e, 1-f, 1-g, 1-h, 1-1. There is no significant difference in the n u m b er of overnight stays in: (1) hotel, (2) motel, (3) public te n t campground, (4) friend’s house, (5) relative’s house, an d (6) other accom m odations while on Michigan and nonMichigan trips. All these null hypotheses are rejected a t a p < .05 (see Table 19). The rejections indicate th a t there are significant differences in the num ber of nights stayed in these accomm odations while on Michigan and non-Michigan trips. For Michigan trips, the average num ber of overnight sp en t in the following accom m odations: hotel, motel, public tent cam pground, friend’s house, relative’s house, and other are 0.46, 0.39, 0.20, 0.27, 0.12, and 1.98 nights, respectively. For non-Michigan trips, the average overnight sp en t in these accomm odations are 1.33, 0.71, 0.09, 0.71, 2.24, and 1.83 nights, respectively. Michigan trips involve significantly more nights in public tent cam pgrounds th a n non-M ichigan trips, while non-M ichigan trips involve more nights in hotels, motel, relative’s houses, and friend’s houses. H ypotheses !•). 1-k. 1-1. 1-m. 1-n. There is no significant difference in the am ount of money spent on: (1) transportation, (2) lodging, (3) m eals, (4) entertainm ent, and (5) m iscellaneous, by Michigan and non-Michigan travelers while pleasure trips. All of these null hypotheses are rejected at a p < .05 indicating 90 th a t there are significant differences in these expenditures between Michigan and non-M ichigan pleasure trips (see Table 19). For Michigan trips, th e average spent on tran sp o rta tio n , lodging, meals, entertainm ent, an d miscellaneous per trip is $563.15, $556.34, $568.05, $560.01, and $552.39, respectively. For non-M ichigan trips, the average spending on transportation, lodging, m eals, entertainm ent, and m iscellaneous per trip is $643.6, $623.88, $620.84, $616.51, and $609.32 respectively. The spending during Michigan traveling is significantly less th a n the spending during non-M ichigan traveling. Because of th e sh o rter duration of Michigan trips, however, average total spending p er day on Michigan trips averaged $552.26 while nonMichigan trip spending totaled only $384.94 on average per day. H ypotheses 1-0. 1-p. l-o . 1-r. 1-s. Between Michigan and non-Michigan trips, there are no significant differences in travelers’ rating scores on the degree of im portance to their trips for the following elements: (1) good restaurant, (2) good place to stay, (.3) high prestige, (4) good night life, and (5) w inter fun. The im portance of these concerns to the trip are rated from one to five. A score of one m eans not a t all im portant, while a score of five m eans very im portant. All these null hypotheses are rejected a t a p < .05 (see Table 19). The rejections indicate th a t the rating scores for these elements are significantly different between Michigan and non-Michigan 91 to u rists. For Michigan trips, the average ratings for the elements: "good restaurant", "good place to stay", "high prestige trip", "good nig h t life", and "winter fun" were 4.15, 4.33, 2.51, 2.97, 3.07 respectively. For non-M ichigan trips, the average scores ratings on th ese sam e five elem ents were 4.29, 4.40, 2.63, 3.17, and 2.82 respectively. With the exception of "winter fun", Michigan scores slightly below its competition in this region on the four other elem ents of travel experiences examined in this study. It should be noted, however, th a t only about 10% of the sample involved Michigan trips. Hence these data are dom inated by respondents’ im ages ra th e r th an by actual experiences w ith Michigan as a travel destination. H y p o th e s e s 1-t. 1-u. 1-v. 1-x. 1-v. 1-aa. 1 -ab . 1 -ac. 1 -ad , 1a e . 1-af. Between Michigan and non-Michigan trips, there are no significant differences in travelers’ rating scores on the following Michigan image : ( ) good scenery, (2) good restaurant, (3) friendly people, (4) easy to get to, (5) reasonable price, ( ) good place to stay, (7) su m m er fun, ( ) high prestige, (9) clean air, (10) good night life, and (5) winter fun. 1 6 8 These hypotheses were conducted to determ ine im pressions based on w hat the respondents had seen or read. These item s were rated from one to ten. A score of one m eans strongly disagree and a score of ten m eans strongly agree. The question (see Q39 in questionnaire) "Do you 1, strongly disagree, 10, strongly agree, or would you choose some num ber in between th a t Michigan is known 92 for good weather?". All these null hypotheses are rejected a t a p < .05. The rejections indicate th a t the rating scores for these elem ents are significantly higher for Michigan trips th a n non-M ichigan trips. The average ratings for Michigan and Non-Michigan trips on these items are listed in Table 19. The higher rating scores may help explain why Michigan travelers chose Michigan as trips destination. Hyp o th esis 1-ag. There is no significant difference in th e likelihood of revisiting a destination state between Michigan and nonMichigan pleasure travelers. The probability of revisiting the destination state w as rated by respondents using a five point scale where one equals n o t a t all likely and five equals very likely. This null hypothesis is rejected a t a p < .05 indicating a significant difference existing in the rating scores on the propensity of revisiting the destination state between Michigan and non-Michigan pleasure travelers. For Michigan travelers, the average rating was 4.53. For non-Michigan travelers, the average rating w as 4.45. T hus the Michigan travelers reported a slightly, b u t statistically significant, greater propensity to retu rn to Michigan for a future visit th a n was reported by the non-Michigan travelers for non-M ichigan destinations. The significantly higher scores indicating Michigan travelers have 93 greater b ran d loyalty th a n non-Michigan travelers. However, other behavioral explanations are possible, the expected lower costs for the vacation, for instance. H ypothesis 2. Travelers’ socioeconomic characteristics, type of transportation used, sources of travel information used, an d trip attributes sought do n o t positively affect to u rists’ choice of Michigan as a trip destination. H ypothesis 2-a. E ducation level difference does n o t positively respondents’ choice of Michigan as a trip destination. 2 3 affect This null hypothesis is rejected at a p < .05 level indicating th a t education level difference h as a significant, positive influence on people’s choice of Michigan as a trip destination. People with higher th an high school education are more likely to travel in Michigan than to non-Michigan destinations in the study region. H ypothesis 2-b. Recreation vehicle ownership does not positively respondents’ choice of Michigan as a trip destination. affect This null hypothesis is rejected a t a p < .05 indicating th a t RV 23 The comparison was c o n d u c t e d on the g r o u p s of ’’wi t h h igher education" v ersus "less t h a n high school education". high school or 94 Table 20. Nature of the relationship between variables found to be statistically related (at a p < .05) to travel to Michigan for recreation. (If +. preference for Michigan as a travel destination increases as the variable tested increases.) Hypotheses Number 2 -a. -b. -c. -d. -e. 2 -f. 2 2 2 2 2 2 2 2 2 2 2 2 2 -g-h. -i. -j. -k. -. -m. -n. -o. 1 Variable Tested Chi-Square Significance Education Ownership of RV. Married vs unmarried Weekend trip Type Of transportation to destination Type of transportation at destination TV seen-heard Radio seen-heard Billboard seen-heard Michigan toll free 800 # seen-heard Winter fun Michigan resident Warm season Activity on trip Importance of special event Travel information seen-heard Travel agent information seen-heard Importance of good restaurant Importance of clear air Importance of good night life High prestige vacation State origin by distance Group trip Occupation Destination expectation 1 1 1 1 1 3-a. 3-b. 3-c. 1 3-d. 3-e. 1 1 3-f. 3g. 3-h. 4-a. 4-b. 1 Nature of Impact 0 . 0 2 0 . 0 2 +2 + + + 0 . 0 0 + 0 . 0 0 0 . 0 0 0 . 0 0 0 . 0 0 0.03 0 . 0 0 + + + + 0 . 0 0 0 . 0 0 0 . 0 0 + + + + + 0 . 0 0 + 0 . 0 0 0 . 0 1 0 . 0 0 0 . 0 1 _3 - 0 . 0 1 0 . 0 1 0 . 0 1 0 . 0 0 0 . 0 0 - 0 . 0 2 - 0 . 0 0 0 . 0 2 . - NA NA 4 1. Scores for degree of importance ranged from one to five. A score of one means not at all important; a score of five means very important. . + = positive effect. 3. - = negative effect. 4. NA, not applicable. (Database: The Great Lakes Travel Monitor Study (1983 -1985), Travel Bureau, Michigan Department of Commerce).ownership has a significant, positive influence on people’s choice of Michigan as a trip destination (see Table 20). 2 95 Hyp o th esis 2-c. M artial sta tu s does not positively affect respondents’ choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t m arital sta tu s h a s a significant, positive influence on th e choice of Michigan as a trip destination. Significantly more m arried people choose Michigan th a n choose other sta tes in the study region. However, this resu lt was probably confounded by u rb an -ru ral differences. H ypothesis 2-d. Traveling on the weekend does n o t positively affect travelers’ choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 level indicating th a t weekend travel h as a significant positive influence on the choice of Michigan as a trip destination (see Table 20). Weekend travelers are more likely to choose Michigan th an non-M ichigan destinations. H ypothesis 2-©. Using a personally owned vehicle (POV) rath er th a n a rented vehicle to travel to a pleasure trip destination does not positively affect tourists’ choice of Michigan as a trip destination. This null hypothesis was rejected a t a p < .05 indicating th a t driving a POV to a destination h as a significant positive influence on 96 the choice of Michigan as a trip destination (see Table 20). Significantly m ore of Michigan’s pleasure travelers arrive by POV. Rented vehicles are m ore common a t non-M ichigan destinations. H ypothesis 2-f. Use of a POV a t the destination does n o t positively affect to u rists’ choice of Michigan as a trip destination. This null hypothesis is rejected at a p < .05 indicating th a t use of a POV at the destination h a s a significant, positive influence on the choice of Michigan as a trip destination (see Table 20). Significantly more travelers who u se their own cars a t th eir destination choose traveling in Michigan th a n people who ren t a car a t their destination. Non-Michigan trips to, for example, Florida are m ore likely to involve a fly-drive (a rental car) com bination of transportation. H ypothesis 2-g. 2-fa. 2 -i. Receiving travel inform ation from (1) television, (2) radio, and (3) billboard have no positive affect on the choice of Michigan as a trip destination. These null hypotheses are rejected a t a p < .05 indicating th at travel information from television, radios, and billboards have a significant, positive influence on the choice of Michigan as a trip destination (see Table 20). In this study, significantly more travelers who received travel information from television, radio, and billboards chose Michigan as a trip destination th a n those who do did receive 97 travel information from these media. Hyp o th esis 2-1. Knowledge of th e Michigan toll free num ber for travel information h a s no positive effect on tourists’ choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t knowing the Michigan toll free n um ber for travel inform ation h a s a significant, positive influence on people’s choice of Michigan as a trip destination. A significantly higher percentage of respondents who know Michigan toll free n um ber chose Michigan as a trip destination th a n those who did not know the Michigan toll free num ber. H ypothesis 2 -k . The degree of im portance assigned to "winter fun" by travelers does not positively affect th eir choice of Michigan a s a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t the degree of im portance assigned to w inter fun has a significant, positive influence on people’s choice of Michigan as a trip destination (see Table 20). Significantly more people who are highly concerned with w inter fun choose Michigan as a trip destination th a n th o se not concerned with this trip attribute. 98 H ypothesis 2-1. Michigan residency does not affect to u rists’ choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t living in Michigan h a s a significant, positive influence on the choice of Michigan as a trip destination (see Table 20). In this study, more Michigan respondents traveled in Michigan th a n non-Michigan respondents. H ypothesis 2-m. The season during which the trip w as taken does not affect to u rists’ choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t season (warm versus cold) h as a significant, positive influence on the choice of Michigan as a trip destination (see Table 20). Significantly more w arm season travelers choose Michigan as their trip destination th a n cold season travelers. H ypothesis 2-n. The degree of im portance assigned to outdoor activity during the trip h a s no positive effect on the choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th at the degree of im portance assigned to outdoor activity h a s a significant, positive effect on peoples’ choice of Michigan as a trip destination. 99 Significantly more respondents who feel outdoor activity is im portant to their trip choose Michigan as a trip destination. H ypothesis 2-o. The degree of im portance assigned to special events available during the trip h a s no effect on the choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t the degree of im portance assigned to available special events h as a significant positive effect on tourists’ choice of Michigan as a trip destination. Significantly more respondents with a high degree of concern for special events choose Michigan as a trip destination. H ypothesis 3. Travel information provided and attributes of th e trip and the destination do n o t negatively affect tourists’ choice of Michigan as a trip destination. H yp o th e sis 3-a. O btaining travel inform ation about destination sta te does not negatively affect tourists’ choice of Michigan as a trip destination. This null hypothesis is rejected a t the a p < .05 indicating th at receiving travel information about destination state h as a significant, negative influence on taking a Michigan trip (see Table 20). Significantly fewer people who receive destination travel information 100 choose Michigan as a trip destination th a n people who do not receive travel information. This may be because people taking out of region trips (e.g. Florida) are more likely to seek information th a n those traveling w ithin the region. However, the statistics used in this study establish only correlation; to explain causality requires more exploration. H ypothesis 3-b. O btaining travel information from negatively affect tourists’ choice destination. a travel agent does not of Michigan as a trip The null hypothesis is rejected a t a p < .05 indicating th a t the travel inform ation from a travel agent h a s a significant, negative influence on people’s choice of Michigan a s a trip destination (see Table 20). Significantly fewer respondents who received travel inform ation from a travel agent chose Michigan as a trip destination th a n those who did not receive travel inform ation from travel agent. Hyp oth esis 3-c. The degree of importance assigned by respondents to good re sta u ra n ts does not negatively affect their choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t the degree of im portance of a good re sta u ra n t to the trip h a s a significant, negative influence on the choice of Michigan as a trip 101 destination (see Table 20). Significantly fewer people who consider good restau ran ts as a very im portant trip attrib u te chose Michigan as a trip destination th a n people who were less concerned about availability of good restau ran ts. H ypothesis 3-d. The degree of im portance of clear air to travelers does not negatively affect their choice of Michigan as a trip destination. This null hypothesis is rejected at a p < .05 indicating th a t the degree of im portance of clear air to the trip h a s a significant, negative influence on th e choice of Michigan as a trip destination (see Table 20). Significantly fewer people who consider clear air as veiy im portant chose Michigan as a trip destination th a n people who considered clear air as not th a t im portant to their trip. H ypothesis 3-e. The degree of im portance assigned to good night life by travelers does n o t negatively affect their choice of Michigan as a trip destination. This null hypothesis is rejected at a p < .05 indicating th a t the degree of im portance assigned to good night life h a s a significant, negative influence on the choice of Michigan as a trip destination (see Table 20). Significantly fewer people who consider good night life as very im portant chose Michigan as a trip destination th a n people who did not consider good night life as im portant. However, the rural- 102 u rb a n factor may affect the difference. To explain causality requires more exploration need to be done. H ypothesis 3-f. The degree of im portance assigned to a high prestige vacation by travelers does not negatively affect their choice of Michigan as a trip destination. This null hypothesis is rejected a t a p < .05 indicating th a t the degree of importance assigned to a high prestige vacation h as a significant, negative influence on the choice of Michigan as a trip destination (see Table 20). Significantly fewer people who consider having a high prestige vacation as very im portant chose Michigan as a trip destination. Hypothesis. 3-g. The distance of a state from Michigan does not negatively affect tourists’ choice of Michigan as a trip destination. This null hypothesis is rejected at a p < .05 indicating th a t location h as a significant negative affect on tourists’ choice of travel to Michigan (see Table 20). Significantly fewer people from d istan t states chose Michigan as a trip destination th a n people who are from closer states. 103 H ypothesis 3-h. Taking a group trip does not negatively affect travelers’ choice of Michigan as a trip destination. The null hypothesis is rejected a t a p < .05 indicating th a t taking a group trip h as a significant, negative influence on the choice of Michigan as a trip destination (see Table 20). Fewer group trips involve Michigan as a trip destination th a n other destinations considered in this study. H ypothesis 4. Travelers’ socioeconomic characteristics, expectations, and activities sought have no effect on to u rists’ choice of Michigan as a trip destination. H ypothesis 4-a. Differences in destination expectations have no to u rists’ choice of Michigan as a trip destination. A five-point scale was used to m easure effect on respondents’ expectations of their destinations. One m eans th a t the destination is a lot worse th a n expected and five m eans th a t the destination is a lot better th a n expected. This null hypothesis is rejected at a p < .05 indicating th a t travelers’ destination expectations have a significant influence on respondents’ choice of Michigan as a trip destination. H ypothesis 4-b. O ccupation h as no effect on to u rists’ choice of Michigan as a 104 trip destination. This null hypothesis is rejected at a p < .05 indicating th at occupation differences have a significant effect on to u rists’ choice of Michigan as a trip destination. Respondents in different occupation categories exhibit different frequencies for choosing Michigan as a trip destination. None of the following socioeconomic characteristics were found to be statistically significant in the choice of Michigan as a travel destination: sex, race, an d family income. LDA Results: Using Individual Variable S ets in LDA to Predict Michigan Trios Results from th e linear discrim inant analysis (LDA) performed are reported in this section. Both the results of LDA modeling and the comparison of the predictive effectiveness of the modeling are covered. In order to explore the effectiveness of using different types of variables as discrim inators in LDA to predict the propensity of traveling in Michigan, the variables used in this study are grouped into six categories including: socioeconomic variables, attribute variables, (5) ( ) 1 trip pattern variables, (3) travel information variables, Michigan image variables, and ( ) 2 (4) trip ( ) high 6 prediction power variables. The results of each analysis are discussed below. 105 U sing Trip Pattern Variables to Predict The P ropensity of Traveling in Michigan In this section, trip pattern variables are used to estim ate the linear discrim inant function (LDF) for predicting the propensity of traveling in Michigan. A stepwise variable selection procedure was used to select significant independent variables as discrim inators. The variables used in this analysis include one way travel mileage, trip duration, num ber of nights spent in a hotel, num ber of nights sp ent in other states, side trip mileage, weekend trip, group trip, Michigan residency, and trip season. As the discussions in Chapter II highlighted, LDF is the best, m axim al linear com bination of variables for predicting trie group m em bership of the cases under investigation. A LDF is composed of a dependent variable and a varying num ber of independent variables. The objective of LDA is to select good, useful, significant discrim inators (independent variables) which can contribute significant discrim inating power to the model and resu lt in the b est prediction of th e group m em bership (dependent variable). To select an optimal set of discrim inating variables, the criteria used in this analysis include: (1) Wilks’ Lambda, (2) m inim um 106 tolerance value, (3) F * . ^ , and (4) F ^ , * ^ 24. The basic rule is to select a variable w ith minimum Wilks’ Lambda w hich also m eets the other three criteria: m inim um tolerance value (MTV), m inim um F to enter (MFE), and m axim um F to remove (MFR). In this analysis, the value for MFE and MFR is set a t 1.00, an d MTV is se t a t 0.001. During the selection procedure, each en tiy an d removal of a variable is called a step, and th e maximum num ber of steps allowed in this analysis is 16. The m eanings and the procedure for using these criteria are explained below. Wilks* L a m b d a : Wilks’ lambda is a m ultivariate statistic which is used to m easure the differences between homogeneity (cohesiveness) within groups. groups and the In other words, Wilks’ lam bda is a m easurem ent of the degree to w hich cases cluster near their group centroid. For the two-group analysis, Wilks’ lam bda is calculated as the ratio of the total sum of squares. It is the within-groups su m ofsquares to the proportion of the total variance in the discrim ination scores not explained by differences betw een groups. The maximum value of lambda is one, and the m inim um value is zero. A variable with a Wilks’ There are several o ther lam bda value criteria for of zero denotes a high selecting t he s i g nificant v ariables in LDA including: (1) Rao's V, (2) M a h a l a n o b i s ' D i s t a n c e C loset Groups, (3) B e t w e e n - G r o u p s F, (4) M i n i m i z i n g R e s i d u a l V a r i ance, M i n i m u m Conditions for S e l e c t i o n (Norusis, 1986) (Klecka, 1980). betw e e n and (5) 107 discrim ination power (i.e. the variable can effectively discrim inate the cases into groups with great variability between groups and little variability within groups). A variable with a Wilks’ lam bda value of one denotes veiy poor discrim ination power since th e groups it discrim inates have identical group centroids and no group differences exist. A variable with small Wilks’ lam bda is desired in a LDA model, since it can be used as an effective discriminator. T o le ra n c e T e s t : By m easuring the degree of linear association between the independent variables to be entered and the other independent variables in the model, tolerance level is u sed to decide if a variable is a useful discriminator. It is also used to monitor potential com putational inaccuracy. The calculation of tolerance is 1R2,, where R2, is the squared multiple correlation coefficient between ith independent variable and all other variables already entered. According to Klecka (1980), a variable with a tolerance of less th an 0 . 0 0 1 indicates th a t the variable is almost a linear com bination of the rest of variables in the model. Since this variable contributes very little extra discriminating power to the model, including this variable is not necessary. Another reason for excluding su c h a variable is because the com putations involved tend to result in rounding error if the tolerance becomes sm aller th an required to have a tolerance greater th an 0.001. T hus, 0 . 0 0 1 a variable is in order to be entered 108 a s a discrim inator in the model. F V alues: Both F.to0 THNG I DID WHEN KID’/ ’MEET NEW PEOPLE’/ ’LIVE IN LUXURY’/ ’PLCS AWAY FROM HOME’/ ’OVERNIGHTVISITORSLAST3MT/ ’HOWMANYTIMESOVERNIGHT/ ’NUMBEROFPEOPLE1’/ ’NUMBEROFPEOPLE2’/ ’NUMBEROFPEOPLE3’/ ’NUMBEROFPEOPLE4’/ ’NUMBEROFDAYS1’/ ’NUMBEROFDAYS2’/ ’NUMBEROFDAYS3’/ ’NUMBEROFDAYS4’/ ’STATE FROM1’/ ’STATE FROM2’/ ’STATE FROM3’/ ’STATE FROM4’/ ’SEX’/ ’AGE’/ 174 V1271 V1272 V1273 V1274 V1275 V1276 V1277 V1278 V1279 V1280 V1281 V1282 V I194 V I195 V I196 V I197 V I198 V I199 V1200 V1201 ’MARSTAT/ ’PEOPLEINHOUSEHOLD’/ ’OCCUPATION’/ ’OWNCAMPER-KV’/ ’EDUCATION’/ TiOUSEHOLDRACE’/ TOTALFAMILYINCOME’/ ’ZIPCODE’/ ’DATE OF INTERVW7 ’LENGTHOFINTERVIEW’/ ’ADICODE’/ ’STATECODE’/ ’RATINGROTATIONV ’FRWD-BCKWRD ROT*/ ’FIRST STATE’/ ’SECOND STATE’/ ’GOODSCENERYMI7 ’PEACEANDQUIETMI’/ ’FAMILYFUNMI7 ’GOODRESTAURANTSMI’/ V 1202 ’FRIENDLYPEOPLEMiy V1203 ’EASYTOGETTOMI’/ V ’ExcrrEM EN nvny 1 2 0 4 V1205 V1206 V1207 V1208 V1209 V1210 V V1212 V1213 V1214 V1215 V1216 V1217 V1218 V1219 V1220 V1221 V V1223 V1224 V1225 V1226 V1227 V1228 V1229 V1230 V1231 V1232 V1233 V1234 V1235 1 1 2 2 1 2 1 2 ’REASONABLEPRICESMI’/ ’ GOODPLACESTOSTAYMI’/ ’LOTSOFTHINGSTODOMiy ’SUMMERFUNMT / ’HIGHPRESTIGEVALMr/ ’CLEANAIRMI’/ ’GOODNIGHTUFEMiy ’WINTERFUNMT/ ’GOODSCENERYSTATE1’/ ’PEACEANDQUIETSTATE1’/ ’FAMILYFUNSTATE1’/ ’GOODRESTAURANTSSTATE1’/ ’FRIENDLYPEOPLESTATE1’/ ’EASYTOGETTOSTATE1’/ ’EXCFTEMENTSTATE1’/ ’REASQNABLEPRICESSTATE1’/ ’GOODPLACESTOSTATESTATE1’/ ’LOTSOFTHINGSTODOSTATEiy ’SUMMERFUNSTATE1’/ ’HIGHPREST1GEVALSTATE1 / ’CLEANAIRSTATE1’/ ’GOODNIGHTLIFESTATE1’/ ’WINTERFUNSTATE1’/ ’GOODSCENERYSTATTE2'/ ’PEACEANDQUIETSTATE2’/ ’FAMILYFUNSTATE2’/ ’GOODRESTAURANTSTATE2’/ ’FRIENDLYPEOPLESTATE2’/ ’EASYTOGETTOSTATE2’/ ’EXCITEMENTSTATE2y ’REASONABLEPRICESSTATE2’/ 175 V1236 V1237 V1238 V1239 V1240 V1241 ’GOODPLACESTOSTAYSTATE2’/ LOTSOFTHINGSTODOSTATE2’/ ’SUMMERFUNSTATE2’/ ’HIGHPRESTIGEVACSTATE2’/ ’CLEANAIRSTATE27 ’GOODNIGHTUFESTATE27 V I242 WINTERFUNSTATE2’/ V 1002 TRIP ID7 VALUE LABELS V 1003 to V 1004 9 9 8 T>K* 997 ’HV 0 ’NONE’/ V1005 1 YES’ 2 ’NO’ 9 ’DK’/ V1007 900 ’OUT OF STATE REGION’ 99 ’NOT ON CODESHEET ’9 ’DK’ 101 ’MACOMB’ 102 ’OAKLAND’ 103 ’WAYNE’ 201 ’BAY 202 ’CLINTON’ 203 ’EATON’ 2 0 4 ’GENESSEE’ 205 ’GRATOIT 206 ’HILLSDALE’ 207 ’HURON’ 208 ’INGHAM’ 209 ’ISABELLA’ 210 ’JACKSON’ 211 ’LAPEER’ 212 ’LENAWEE’ 213 ’LIVINGSTON’ 214 ’MIDLAND’ 215 ’MONROE’ 217 ’SAGINAW 218’SANILLAC’ 219’SHIAWASSEE ’ 220 ’ST.CLAIR’ 221 TUSCOLA’ 222 WASHTENAW 301 ’ALLEGAN’ 302 ’BARRY 303 ’BERRIEN’ 304’BRANCH’ 305 ’CALHOUN’ 306 ’CASS’ 309 ’IONIA’ 310 ’KALAMAZOO’ 311 ’KENT 312 ’MECOSTA’ 313 ’MONTCALM’ 314 ’MUSKEGON’ 31 5 ’NEWAYGO’ 316 ’OCEANA’ 317 ’OTTAWA’ 318 ’ST.JOSEPH’ 319 VAN BUREN’ 401 ’ALCONA’ 402 ’ALPENA’ 403 ’ARENAC’ 404 ’CHEBOYGAN’ 405 ’CLARE’ 406 ’CRAWFORD’ 407 ’GLADWIN’ 408 TOSCO’ 410 ’MONTMORENCY 411 ’OGEMAW412 ’OSCODA’413 ’OTSEGO’ 414 ’PRESQUE ISLE’415 ’ROSCOMMON’ 501 ’ANTRIM’ 502 ’BENZIE’ 503 ’CHARLEVQIX’504 ’EMMET 505 ’GRAND TRAVERSE’ 506 ’KALKASKA’ 507 TAKE’ 508 ’LEELANAU’ 509 510 ’MASON’ 511 ’MISSAUKEE’ 512 ’OSCEOLA’ 601 ’ALGER’ 602 ’CHIPPEWA ’ 603’LUCE’ 604 ’MACKINAC’ 701 ’ALGER’702 ’DELTA’703 ’DICKINSON’ 704 ’MARQUETTE ’705 ’MENOMINEE’ 706 ’SCHOOLCRAFT 801 ’BARAGA’ 802 ’GOGEBIC’ 803 ’HOUGHTON’ 804 ’IRON’ 805 ’KEWEENAW806 / V1281 395 ’ALEXANDRIA’ 627 ’ALPENA’ 135 ’BUFFALO’ 257 ’CHARLESTON’ 051 ’CHICAGO’ ’093 ’CINCINATn’035 ’CLEVELAND’121’COLUMBUS’ 177 ’DAVENPORT 095 ’DAYTON’057 ’DETROIT 381 ’DULUTH’147 ’ERIE’ ’207 ’EVANSVILLE’ 393 ’FARGO’ 063 ’FUNT-SAGINAW 091 ’FT.W AYNE ’059 ’GRAND RAPIDS’ 315 ’GREEN BAY 083 ’INDIANAPOLIS’ 117 ’LACROSSE’ 085 ’LAFAYETTE’061 ’LANSING’ 101 ’LIMA’ 209 ’LOUISVILLE’ 113 ’MADISON’ 449 ’MANKATO’ 317 ’MARQUETTE’ 111 ’MILWAUKEE’ 107 ’MINN.ST.PAUL’ 187 ’PADUCAH’ 175 ’PEORIA’ 029’PITTSBURGH’ 227 ’QUINCY-HANNIBAL’165 ’ROCHESTER’ 11 ’ROCKFORD’ 389 ’SIOUX FALLS’ 053 ’SOUTH BEND’ 077 ’SPRINGFIELD’ 075 ’ST.LOUIS 087 TERRE HAUTE’ 055 TOLEDO’ 451 TRAVERSE CITY 115 WAUSAU’103 WHEELING’ 031 YOUNGSTOWN’ 125 ’ZANESVILLE’ 73 5’BELLEVDLLE’ 750 ’BRANTFORD’770 ’CHATHAM’ 760 ’FERGUS’ 755 ’GUELPH’ 710 ’HAMILTON’ 720 ’KITCHENER’ 775 ’LEMINGTON’ 745 LINDSAY 725 ’LONDON’ 795 ’MIDLAND’790 ’ORILLIA’ 700 ’OSHAWA’ 785’OWEN SOUND’740 ’PETERBOROUGH’ 780 ’SARNIA ’ 715 ’ST.CATH.-NLAGARA’ 765 ’STRATFORD’ 705 TORONTO’ 730 WINDSOR’ 173 ’CEDAR RAPIDS’/ V1006 1 ’DESTINATION’ 2 ’MAIN DESTINATION’ 3 ’PLACE FARTHEST/ V I134 TO V I139. V I103. V1008. V1265 TO V1268, V1282 601 ALABAMA 102 ’ALASKA’ 203 ’ARIZONA’105 CALIFORNIA’ 206 ’COLORADO 907 ’CONNECTICUT 708 ’DELAWARE’ 709 ’D. OF C.’ 710 ’FLORIDA’ 711 ’GEORGIA’ 112 HAWAII’ 213 ’IDAHO’ 514 ’ILLINOIS’ 515 ’INDIANA 316 IOWA 317 ’KANSAS’ 618 ’KENTUCKY 419 ’LOUISIANA’ 920 ’MAINE’ 721 ’MARYLAND’ 922 ’MASSACHUSETTS’ 523 ’MICHIGAN’ 324 ’MINNESOTA’ 625 ’MISSIPPI’ 326 ’MISSOURI’ 227 ’MONTANA’ 328 ’NEBRASKA’ 229 ’NEVADA 930 ’NEW HAMPSHIRE’ 831 8 176 •NEW JERSEY* 232’NEW MEXICO’ 833 ’NEW YORK’ 734 ’NORTH CAROLINA’ 335 ’NORTH DAKOTA’ 536 ’OHIO’ 437 ’OKLAHOMA’ 138 ’OREGON' 83 •PENNSYLVANIA’ 940 ’RHODE ISLAND’ 741 ’SOUTH CAROLINA’ 342 ’SOUTH DAKOTA’ 643 TENNESSEE’ 444 TEXAS’ 245 ’UTAH’ 946 VERMONT 747 VIRGINIA’ 148 WASHINGTON’ 749 WEST VIRGINIA’ 550WISCONSIN* 251 WYOMING’ 052 ’ONTARIO’ 053 INTERNATIONAL’ 600 ’SOUTH REGION’ 200 ’MOUNT.SW, WEST 300 ’CENTRAL’ 800 ’EAST 100 ’PACIFIC’ 404 ’ARKANSAS’/ V1G06.V100899 ’NOT ON CODESHEET 9 ’DK’/ V1009 1 ’JANUARY’ 2 ’FEBRUARY’ 3 ’MARCH’ 4 ’APRIL’ 5 ’MAY* ’JUNE’ 7 ’JULY* ’AUGUST 9 ’SEPTEMBER’ 10 ’OCTOBER’ 11 ’NOVEMBER’ 12 ’DECEMBER’/ V1010 3 ’1983’ 4 ’1984’ 5 *1985’/ V1012 1 VISIT RELATIVES’ 2 VISIT FRIENDS’ 3 ’SHOPPING’ 4 ’OUTDOOR RCREAHON’ 5 ’SIGHTSEEING,TOURING’ ’A SPECIFIC ATTRACTION’ 7 ’OTHER’/ V1011 ’CONVENTION ONLY’ 2 ’BUSINESS ONLY’ 3 ’BUS.OR CONV.& PLEASURE’ 4 VISIT RELATIVES’ 5 VISIT FRIENDS’ ’SHOPPING’ 7 ’OUTDO OR RECREATION’ ’SIGHTSEEING, TOURING’ 9 ’A SPECIFIC ATTRACTION’ 10 ’ OTHER’/ V1015, V1016, V 1022, V10131 ’YES’ 2 ’NO’ 9 ’DK’/ V1Q14 98 ’DK’ 97 ’HV 0 ’NONE’/ V1G18 TO V1Q218 ’DK’ 7 ’HV 0 ’NONE’/ V1023 TO V10268 ’DK’ 7 ’HV 0 ’NONE’/ V1028 TO V10351 ’BICYCLING’ 2 ’POWER BOATING ’3 ’CANOEING’ 5 ’RAFTING’ 6 ’CAMPING’ 7 ’CHARTER BOAT FISHING’ ’STREAM RIVER FISHING’ 9 ’LAKE FISHING’ 10 ’ICE FISING’ 11 ’GOLF 4 ’SAILING’ 12 ’ TENNIS’ 13 ’HUNTING’ 14 ’HIKING BACKPACKING’ 15 ’HORSEBACK RIDING’ 16 ’DOWNHILL SKIING’ 17 ’CROSS COUNTRY SKIING’ 18 ’SNOWMOBILING’ 19 ’ SWIMMING’ 20 WATERSIOING’ 21 ’SUNBATHING’ 22 ’MISCELLANEOUS SPORTS’ 23 ’OTHER’/ V1036 TO V10401 ’NATURAL ATTRACTIONS’ 2 ’LANDMARKS ’3 ’HISTORICAL SITES’ 4 ’MAN MADE ATTRACTIONS ’5 ’MUSEUMS’ 6 ’CULTURAL EVENTS’ 7 ’ PROFESSIONAL SPECTATOR SPORTS’ ’FAIRS EXHIBITS’ 9 ’NIGHT CLUBS SHOWS RESTAURANTS’ 10 ’INDUSTRY TOURS’ 11 ’FESTIVALS ’12 ’MOVIES’ 13 ’ OTHER’/V1049 TO V I060 98 ’DK’ 97 ’HV 0 ’NONE’/ V1061 TO VI064 9 ’DK’/ V1065 9 ’DK’ 1 YES’ 2 ’NO’/ V1066 TO VI0938 ’DK’ 7 ’H V / V1094.V109599 ’DK’ 1 ’RENTED CAR’ 2 ’OWNED CAR’ 3 ’RENTED KV 4 ’OWNED KV 5 ’PLANE’ TRAIN’ 7 ’BUS’ BOAT 9 ’OTHER’/ V1096 998 TDK 997 ’HV 0 ’NONE’/ V1098 0 NONE’ 98 ’DK’ 97 ’HV/ V1099 0 ’NONE’ 98 ’DK’ 97 ’HV/ V I100 98 TDK’ 97 ’HV/ V1101 1 ’SELF ONLY 2 ’OPP.SEX COMPANION’ 3 ’CHILD UNDER 18’ 4 ’CHILD 18 OR OLDER’ 5 ’RELATIVE UNDER 18’ ’RELATIVE 180R OLDER’ 7 ’FRIEND UNDER 18’ ’FRIEND 18 OR OLDER’ 9 BUS.ASSOC.OR OTHER’/ V I102 I YES’ 2 ’NO 9 ’DK’/ V I103 999 ’DK’/ V I104 99 ’DK’/ V1105.V11069 ’DK’ 1 WORST 5 ’BEST/ VI 107.V1108 9 ’DK’ 1 ’A LOT WORSE’ 2 ’SOMEWHAT WORSE’ 3 ’ABOUT THE SAME’ 4 ’SOMEWHAT BETTER’ 5 ’A LOT BETTER’/ V I109 1 NOT AT ALL LIKELY 2 NOT TOO LIKELY 3 ’NEURTRAL’ 4 ’SOMEWHAT LIKELY 5 VERY LIKELY/ V I110 1 NOT AT ALL WELL’ 2 NOT TOO WELL’ 3 ’SOMEWHAT WELL’ 4 VERY WELL’ 9 ’DK’/ V l l l l TO V I115998 ’DK’ 997 ’HV 0 NONE’/ V1116 9998 ’DK’ 9997 ’HV 0 ’NONE’/ V I117 1 YES’ 2 ’NO’ 9 ’DK’/ V1118 TO V1129.V1130 TO V I1321 NOT SEEN 2 BEFORE’ 3 ’DURING’ 4 ’AFTER’ 5 ’BEFORE AND DURING’ ’DURING AND AFTER’ 7 ’BEFORE AND AFTER’ BEFORE.DURING.AFTER’ 9 ’DK’/ V I133 1 YES’ 2 ’NO’ 9 ’DK’/ V I134 TO V I139999 ’DK’/ V I140 TO V I1451 ’LITERATURE NOT READ’ 2 ’BEFORE’ 3 ’DURING’ 4 ’AFTER’ 5 ’BEFORE AND DURING’ ’DURING AND AFTER’ 7 BEFORE AND AFTER’ BEFORE.DURING.AFTER’ 9 ’DK’/ V I146 TO V I148998 ’DK’ 997 ’HV 0 NONE’/ V I149 TO V I1551 ’BICYCLING’ 2 ’POWERBOATTNG’ 3 6 8 6 1 6 8 8 8 6 8 6 8 6 8 6 8 177 ’CANOEING’ 4 ’SAILING' 5 ’RAFTING’ ’CAMPING’ 7 ’CHARTER BOAT FISHING’ ’STREAM.RIVER FISHING’ 9 ’ LAKE FISHING ’ 10 ’ICE FISHING’ 11 ’GOLF 12 TENNIS’ 13 ’HUNTING’ 14 HIKING,BACKPACKING ’15 ’HORSEBACK RIDING’ 16 ’DOWNHILL SKIING ’17 ’CROSS COUNTRY SKIING’ 18 ’SNOWMOBIUNG’ 19 ’SWIMMING’ 20 WATER SKIING’ 21 ’SUNBATHING ’22 TvOSCELLANEOUS SPORTS’ 23 ’OTHER’ 99 ’DK’/ V I156 TO V I1601 ’NATURAL ATTRACTIONS’ 2 -LANDMARKS’ 3 ’HISTORICAL SITES’ 4 ’MAN-MADE ATTRACTIONS’ 5 ’MUSEUMS ’ ’CULTURAL EVENTS’ 7 ’PROFESSIONAL SPEC. SPORTS ’8 'FAIRS.EXHIBITS ’9 ’NIGHT CLUBS.SHOWS.RESTAURANTS’ 10 ’INDUSTRY TOURS ’11 ’FESTIVALS’ 12 ’MOVIES’ 13 ’OTHER’ 99 ’DK’/ V1161.V11941 ’GOOD SCENERY 2 ’PEACE AND QUIET ’3 ’FAMILY FUN’ 4 ’GOOD RSTRNT5 ’FRIENDLY PEOPLE ’EASY TO GET TO ’ 7 ’EXCITEMENT ’REASONABLE PRICES’ 9 ’GOOD PLACES TO STAY TO TOTS OF THINGS TO DO ’ 11 ’SUMMER FUN’ 12 ’HIGH PRESTIGE VACATIONS’ 13 ’CLEAN AIR’ 14 ’GOOD NIGHT LIFE’ 15 WINTER FUN’/ V I162 TO V I1761 ’NOT AT ALL IMPORTANT 9 ’DK 5 ’ONE OF THE MOST IMPORTANT/ V I 177 TO V I1840 ’NONE’ 7 ’HV ’DK’ 9 ’LIVES THERE OR MISSING DATA’/ VI 185.V1196.V11971 ’OHIO’ 2 WISCONSIN’ 3 ’ILLINIOIS ’4 ’MICHIGAN 5 ’ PENNSYLVANIA’ ’ONTARIO’ 7 ’INDIANA ’MINNESOTA’/ V I186 TO V I1939 ’DK’ 1 WORST 5 ’BEST/ V1243 1 ’MAKE LASTING MEMORIES’ 2 ’KEEP ME BUSY 3 ’LET ME ESCAPE’ 4 ’ RETURN TO NATURE’ 5 ’PLACES I VE NEVER BEEN’ ’BRING NEW EXPERIENCES’ 7 ’CHANCE TO REST ’DO KID THINGS ’9 ’NEW PEOPLE’ 10 ’LIVE IN LUXURY 11 ’PLACES AWAY ’FROM HOME’/ V1244 TO V12549 ’DK 1 ’STRONGLY DISAGREE’ 5 ’STRONGLY AGREE’/ V10131 YES’ 2 ’NO’ 9 ’DK’/ V1014.V1049 TO VI0600 ’NONE’ 98 ’DK’ 97 ’HV/ V1061 TO V10649 ’DK’/ V I110 1 "NOT AT ALL WELL’ 2 ’NOT TOO WELL' 3 ’SOMEWHAT WELL’ 4 VERY WELL’ 9 ’DK’/ VI111 998 TDK* 997 ’HV/ V1112 TO V I115998 ’DK’ 997 ’HV 0 ’NONE’/ V I116 9998 ’DK’ 9997 ’HV 0 ’NONE’/ V1255 1 YES’ 2 ’NO’ 9 ’DK’/ V1256 99 ’DK’ 98 H V / V1257 TO V126498 ’DK’ 97 VARIES’ 96 ’HV/ V1265 TO VI268999 ’DK’/ V1269 1 ’MALE 2 ’FEMALE’ 9 ’REFUSED’/ V1270 9 ’REFUSED’ 99 ’DK’/ V1271 1 ’MARRIED’ 2 UNMARRIED’ 9 ’REFUSED’/ V1272 HV 9 ’REFUSED’/ V I273 99 ’DK 1 ’OWNERPROPRIETER’ 2 *MANAGERADMINISTRATOR’ 3 ’PROFESSIONAL’ 4 TECHNICAL’ 5 ’SALES’ 6 ’CLERICAL, OFFICE’ 7 ’SKILLED,SEMI- SKILLED .TRADE’ 8 ’UNSKILLED LABOR’ 9 ’FARMER’ 10 ’POLICE.POSTAL.FIRE’ 11 ’ARMED SERVICES’ 12 HOMEMAKER’ 13 ’STUDENT 14 ’RETIRED’ 15 ’OTHE R’ 16 ’REFUSED’/ VI274 1 YES ’ 2 ’NO’ 9 ’REFUSED’/ V1275 9 DK 1 LESS THAN HIGH SCHOOL’ 2 ’HIGH SCHOOL GRADUATE’ 3 TRADE.TECHNICAL’ 4 ’SOME COLLEGE’ 5 ’COLLEGE GRAD’ TOST DEGREE’ 7 ’REFUSED’/ V1276 1 WHITE’ 2 ’BLACK 3 HISPANIC’ 4 ’OTHER’ 5 ’REFUSED’ 9 ’DK’/ V1277 1 LESS THAN 10,000’ 2 ’10,000 TO 19,999’ 3 ’20,000 TO 29,999’ 4 ’30,000 TO 39,999’ 5 ’40,000 TO 49,999’ ’50,000 TO 59,999’ 7 ’60,000 TO 69,999’ ’70,000 OR MORE’ 9 ’REFUSED’/ V1278999999 D K / VI28098 ’HV 9 D K / V I195 1 ’UP ’ 2 ’DOWN ’/ V I198 TO V I242 1 ’STRONGLY DISAGREE ’ 10 ’STRONGLY AGREE ’ 99 ’DK ’ 98 ’REFUSED ’. FREQUENCIES GENERAL=ALL. FINISH. 6 8 6 * 6 8 8 6 8 6 8 8 6 6 8 178 APPENDIX D SPSS/PC+ Variables Recoding Program 179 SPSS/PC+ Variables Recording Program get fll ’e:alltrip.sys’. set pri=on ptr=on echo=on. rec vl011(3 thr 10=1)(1 thr 2=0). (recode trip purpose l=recreation trip, O=nonrecreation trip com ssfw=vl009. rec ssfw(3 thr 5 =1)(6 thr =2) (9 thr 11=3) (12 1 2=4). val lab ssfw 1 ’Spring’ 2 ’Summer’ 3 ’Fall’ 4 ’Winter’. (create season variable four seasoncom wc=vl009. 8 rec wc (1 2 3 4 11 12=0) (5 7 9 10=1). val lab wc 1 ’warm season 1 to 4 11 12’ 0 ’cold season 5 to 10’. (create season variable warm season versus cold season 6 8 if (vl282 eq vl008) inout= , if (vl282 ne vl008) inout= . (create new var inout: = in-state travel, l=out-of-state travel 0 1 0 if (vl008 eq 523) minmit= 1. if (vl008 ne 523) minmit=0. val lab minmit 1 ’michigan trip’ ’nonmichigan trip’. (create new variable minmit: l=Michigan trip. O=non-Michigan trip 0 if (vl282 eq 523) minmir=l. if (vl282 ne 523) minmir=0. val lab minmir 1 ’michigan resident’ ’nonmichigan resident’. [create new variable minmir: l=Michigan resident, O=non-Michigan (resident 0 rec vl269 (2=0). [record male=l female = 0 rec vl271 (2=0). [record married= unmarried 1 = 0 rec vl270 (99=sysmis). [record age dont know into system missing rec vl274 (2=0). [record non-rv owner = 0 rec vl275 (7=sysmis). [record education variable refused=sys missing rec vl276 (5 9=sysmis). (record race dk re into sysmis rec vl277 (9=sysmis). (last line record family income refuse into sysmis. rec vl272 9=sysmis). (last line record household size dk re into sysmis ( 8 180 {last line record weekend trip=l nonweekend trip rec vl016 ( = ). [last line record group trip=l norigroup trip 2 = 0 0 = 0 rec v l 117 (2=0). [last line record travel infomation heard or not rec v l l l (1 4=0) (2 3 5 7 8=1). [last line record TV seen or heard 8 6 rec v l l l 9 (1 4=0) (2 3 7 8=1). [last line record radio seen heard rec v l 120 (1 4=0) (2 3 5 7 8=1). [last line record nwstrav seen-heard 6 rec v l 121 (1 4=0) (2 3 5 7 8=1) (9=sys). [last line record Nwspoth seen heard 6 rec v l 122 (1 4=0) (2 3 5 7 8=1) (9=sys). [last line record magazine seen heard rec v l 123 (1 4=0) (2 3 5 7 8=1). [last line record billboard seen-heard 6 rec v l 124 (1 4=0) (2 3 5 7 8=1). [last line record travel agent seen heard 6 rec v l 125 (1 4=0) (2 3 5 7 8=1). [last line record friend relative seen heard 6 rec v l 126 (1 4=0) (2 3 5 7 8=1) (9=sys). [last line record autoclub seen heard rec v l 127 (1 4=0) (2 3 5 7 8=1). [last line record 800 number seen heard 6 rec v l 128 (1 4=0) (2 3 5 7 8=1). [last line record chmcomm seen heard 6 rec v l 129 (1 4=0) (2 3 5 8=1). [last line record at dest seen heard rec 6 v l l 3 0 (1 4=0) (2 3 5 7 8=1). [last line record other seen heard 6 sel If (vlOll eq ). sav out=’e:\allrec.sysVcom. pro If (minmit eq out=’e:\mirect.sys com. 1 7 1 ). sav 181 APPENDIX E S u r v e y Q u estio n n a ire 182 ( ) ( > ( ) c h a n n e l? A t r a v e l a g en cy o r t o u r i s t b u r ea u ? k IF YES TO AX? Q1 ITEMS, TERMINATE. OTHERWISE, CONTINUE. 2. Have you o r anyone i n y o u r h o u s e h o ld t a k e n a n y k in d o f a t r i p s i n c e ( DATE 6 MONTHS ACO) ? T h is i n c lu d e s t r i p s f o r p l e a s u r e , b u s i n e s s , o r b u s in e s s m ixed v l t b p le a s u r e . 2. V ere any o f t h e s e o v e r n ig h t o r o v e r 1 0 0 m i l e s o n e way? 4. G e n e r a lly s p e a k in g who i n y o u r h o u s e h o ld d e c i d e s w h y, w here and when you t r a v e l f o r p le a s u r e ? (1 ) (1 ) (1 ) Yes - CONTINUE (2 ) Yes - CONTINUE (2 ) (2 ) S e l f , i n c lu d in g sh a r e d d e c i s i o n ­ making - CONTINUE No - TERMINATE Ho - TERMINATE O th e r - "Nay I p le a s e sp eak t o h im /h e r ? " ( I F HOT BONE, ARRANGE CALLBACK) xJ IF CET DIFFERENT PERSON ON PHONE, REPEAT INTRODUCTU ION 4. 3a. Bow many t r i p s o f o v e r 1 0 0 m i l e s o n e way o r o v e r n i g h t h ave you ta k en s i n c e (DATE 12 MONTHS ACO) ...(R E A D . WRITE I N .) ------------a. P r im a r ily f o r p le a s u r e b. P r im a r ily f o r b u s i n e s s IF HONE, ASK Q4b AND TERMINATE. ' OF BUSINESS TRIPS. I P 4 OR MORE, ASK Q Sa. RECORD NUMBER OTHERWISE CO TO Q6 . (6 -8 ) (9 -1 1 ) Were any o f you r l a s t 4 t r i p s p r i m a r i l y f o r p le a s u r e ? (1 ) Y es - CONTINUE <2) No - CET 3 MOST RECENT BUSINESS AND ONE MOST RECDfT PLEASURE. -1 - ( 12) 183 {t r ip a c t iv it y ! c m p l e t e m s s e c t i c h , Q M tis e , f o r m t o p ou r h o s t r e c e n t r u A i u n t r i p s n PAST 1 2 HOWTOS. A R U LIST W f QO 1 0 TRIP SPECIFICS, f i l l . . o».| XT LAST T U T , GO TO I M S SPSCIFICS, S. Nhere d id y o u Qo o n y o u r ( e z s t ) c o s t m n l t r i p ? (RECORD DESTIMATICB AMD DESTINATION STATS BELCH.) (XT 2 HOST R E O *! T U T S BUSINESS, AT FOURTH T R IP, "Where d id y o u f o o o y o u r B o a t r e c a n t p l a a m r e t r ip } " ) XT SEVERAL DOMESTXC/ONIRftlO BESTISATIONiS XB OWE T U P ; t r i p t h a t y o u c o n s i d e r e d t o h e y o u r B a in d e s t i n a t i o n ? t h a r o a n y p la c e (1 ) b. BELOW AKD 8M P TO Q7 Y ea - "What W hich p l a c e w as f a r t h e s t f r o a you r b o a s? Xn w h ic h a t a t a d id you (2 ) .) 12) a t . t . .p a n t b d s c t l s a (0 6 c) TRIP ?• (44; (4 3 ] (4 2 ) ... ..................( 3 9 - 4 (4 «t-3: H o s t R e cen t 1__________1 What a o n t h d i d y o u s t a r t y o u r t r i p t o (D CST.)P JAH"1, 1 - 1 2 RECORD HOtrm AMD YEAR. I P HOWTO OUT OP 1 2 HOWTO PERIOD, DO WOT RECORD TRIP. YEAR: I T ALL TRIPS OUT OP 12 HOWTO PERIOD, TERMINATE. (5 4 - 3 5 : (5 6 -3 7 ; (5 6 -3 9 ) (4 0 - 6 1 ) (6 2 ) (63] (6 4 ) (6 5 ) ENTER LAST OXCXT IP INTERNATIONAL DESTINATION (EXCEPT ONTARIO) , CO TO QS FOR NEXT TRIP. IP ALL TRIPS INTERNATIONAL, TERMINATE. RECORD WimnCR GP INTERNATIONAL TRIPS. 184 8. V h at w as th e M i n p Q f f o i i o f t h i s i l i p ? p (ENTER ONE COOE BEtOW r o t PURPOSE) IF IU51NESS OB COHVEHTION V a t I t p la n n ed f o r b u s i n e s s ( a c o n v e n tio n ) o n l y , o r f o r b u s i n e s s ( a c o n v e n tio n ) n ix e d v ttb p liiiu t t l ( 1 ) C o n v en tio n o n ly - CO TO Q6 .FOR NEXT TB1P. ( 2 ) B u s in e s s o o ly - CO TO Q6 FOR NEXT T U P . ( 1 ) B u s in e s s o r c o n v e n tio n e i s e d w ith p le e r u n O ld a n o th e r p e r s o n g o w ith you f o r p le a s u r e on t h i s t r i p o r n o t? (1 ) (2 ) Yea - CODE AS BUSINESS'/PLEASURE AND SKIP TO Q9. No - CONTINUE O ld you e x te n d th e c r i p a t l e a s t erne a d d i t i o n a l day f o r p le a s u r e o r n o t? (1 ) (2 ) Yea - CODE AS BUSINESS/PLEASURE AND SKIP TO Q9 Ho - COOE AS BUSINESS 0 5 CONVENTION AND GO TO Q6 FOR NEXT TRIP. IP VACATION W hich o f t h e s e b e s t d e s c r ib e s v h a t you d id ? (A ) V i s i t r e l a t i v e s ( 5 ) V i s i t f r ie n d s ( 6 ) S h op p in g (READ 3 - 9 , ROTATE. (? ) (6) (9 ) (1 0 ) HARKROTATION START.) O utdoor r e c r e a t i o n S ig h t s e e i n g / t o u r i n g A s p e c if ic a tt r a c t io n O th er (WRITE IN) ( 6 6 ). H ost r e c e n t 1 2 (6 7 -6 8 ) (6 9 -7 0 ) « } ENTER ONE CODE FOR PURPOSE (7 1 - 7 ? ) (7 3 -7 6 ) I F CONVENTION ROT MENTIONED IN Q9: D id you a t t e n d a c o n v e n tio n on t h i s t r i p ? (1 ) Y es (2 ) Ho ENTER ONE CODE 10. ENTER NUMBER OF H1CHTS ( 1 - 4 ) D U P(5)2 10a .* (7 5 ) (7 6 ) (6 -7 ) (B -9 ) ( 10- 11) (1 2 -1 3 ) (1 4 ) < liL <16) (1 7 ) (I B ) (1 9 ). ( 20) ( 21) .. (7 8 ) (77.) How a en y n i g h t s , i f a o y , d id y o i s t a y away f r o a h ose? (I F ZERO BIGHTS, SKIP TO Q l l ) IF 1 , 2 OR 3 HICHTS IB QiO: ( a s t h i s a w eekend t r i p o r n o t? (1 ) Yea (2 ) Ho ENTER ONE CODE 11. Was t h i s t r i p p a r t o f a grou p to u r? (1 ) Y es (2 ) Bo ENTER ONE CODE_ -3 185 Most R ecent 1 REENTER DESTINATION 12. 2__________5_________ 4 . Bow a n y p e o p le v e n t w i t h y ou ? (I F SELF ORLY* CO TO Q6 FOR NEXT TRIP OR SKIP TO Q l l . I F MORE TUAN SELF CONTINUE.) ENTER TOTAL PEOPLE__ (2 2 - 2 3 ) I 2. r - | i r h o u n u n b. F r ie n d s < 2 8 -2 9 ) 1 1 (READ) 1 SELF_ F am ily (2 6 -2 7 ) s e l f omly | Not I n c lu d in g y o u r s e l f , how many w e r e . . . T a. (2 4 -2 5 ) ENTER NUMBER OF PEOPLE c. R e la t iv e s d. B u s in e s s a s s o c i a t e s / o t h e r (DO HOT READ) FAMILY (3 0 ) (3 4 ) (3B ) (4 2 ) (3 1 ) (3 5 ) (3 9 ) (4 3 ) FRIENDS RELATIVES__ (3 2 ) (3 6 ) (4 0 ) (4 4 ) (3 3 ) (3 7 ) (4 1 ) (4 5 ) OTHER • e a yn AL MUST :q u al 012 * a • 2 2 b . V cre th e r e any c h i l d r e n u n d e r 1BT (1 ) Y es - CONTINUE (? ) No - CO TO Q6 FOR NEXT TRIP OR SKIP TO Q13. ENTER ONE CODE__ (4 6 ) 1 2 c . Bov many o f th e c h i l d r e n w e r e . . . (1 ) L ese th an 4 y e a r s o l d ... , , (3 ) B -12 (4 ) 1 3 -1 7 (4 9 ) L .T . 4 ENTER NUMBER CHILDREN . , — (5 0 ) (5 4 ) (5 6 ) (6 2 ) (5 1 ) (5 5 ) (5 9 ) (6 3 ) (5 2 ) (5 6 ) (6 0 ) (6 4 ) (5 3 ) (5 7 ) (6 1 ) (6 5 ) S " 12— 1 3 -1 7 _ RETURN TO Q6 FOR NEXT T R IP. H r ip (4 8 ) (4 7 ) (READ) I F LAST TRIP, CONTINUE. sp e c ific s; COMPLETE THIS SECTZON, Q13 TO Q 30a, FOR TVO MOST RECEHT PLEASURE TP.IPS. H o st R ecen t 1 2 (6 6 - 6 7 ) ( 8 -9 ) (6 6 -6 9 ) ( 10- 1 1 ) (7 0 - 7 1 ) (1 2 -1 3 ) (7 2 - 7 3 ) (1 4 -1 5 ) REENTER DESTINATION Nov we h a v e a fe w q u e s t i o n s a b o u t y o u r t r i p t o (D E S T .). 13. On t h i s t r i p v h a t o u td o o r r e c r e a t i o n a c t i v i t i e s , i f a n y . d id you p a r t i c i p a t e In i n (DEST. STATE)7 A n y th in g e l s e ? (DO NOT READ. ACCEPT UP TO 7 ACT IVIT IES.) ( 1) ( 2) (3 ) (4 ) (5 ) (6) (7 ) ( 8) (9 ) ( 10) (ID ( 12) B ic y c l i n g Power b o a t in g C anoeing S a il i n g R a ftin g Camping C h a rter b o a t f i s h i n g S tr e a a /r lv e r f is h in g Lake f i s h i n g Ice fis h in g C o if T ea n le (1 3 ) (1 4 ) (1 5 ) (1 6 ) (1 7 ) (1 8 ) (1 9 ) ( 20) ( 21) ( 22) (2 3 ) U u a tia g H ik in g /b a c k p a c k in g H o rseb a ck r i d i n g D o w n h ill s k ii n g C roea c o u n tr y s k ii n g S n O M sob llln g S v la a l n g W ater s k ii n g S u n b a th in g M i s c e l la n e o u s s p o r t s O th e r (WRITE 1M) TRIP I . ENTER CODES OP ACTIVITIES (1 -4 ) DUP (7 4 - 7 5 ) (1 6 -1 7 ) (7 6 - 7 7 ) (1 8 -1 9 ) (7 8 - 7 9 ) ( 20- 2 1 ) ( 6 -7 ) (2 2 -2 3 ) 186 KOSt tlC C D t 1 REENTER DESTINATION 14, Vhat e v e n t s o r o l g h t o d id y ou v i s i t o r v h a t o t h e r e n t e r t a i n m e n t , I f a n y , d id you p a r t i c i p a t e l a ? A n y th in g a l e e ? (DO NOT READ. ACCEPT OP TO S . ) (1 ) (2 ) (j) (4 ) (5 ) (6) (? ) (B ) (9 ) N ig h t c l u b s , s h o w s , r e s t a u r a n t s N a tu r a l a t t r a c t i o n # (1 0 ) I n d u s tr y t o u r s Landmarks (1 1 ) F e s t iv a ls H i s t o r i c a l n ig h t s ENTER (1 2 ) M ovies M an-aade a t t r a c t i o n s LETTERS (1 3 ) O th er (WRITE IN) H useuas OF C u lt u r a l e v a o t a ACTIVITIES P r o f e a a i o n a l/ e p e c t a t o r a p o r ta TRIP l . _ F a ir s , e x h ib its .TRIP 2 . (2 4 -2 5 ) (34 - 3 5 ) (2 6 - 2 7 ) TlbO?) (2 8 - 2 9 ) (3 8 -3 9 ) (3 0 - 3 1 ) (4 0 - 4 1 ) n izin (4 2 -4 3 ) XF STAYED AWAY FROM DOME IN Q 10, ASK Q15 AND CONTINUE. OTHERWISE, SKIP TO Q17. 15. What k in d o f l o d g i n g d id you u s e i n (DEST. STATE)? How o a n y n ig h t s ? (PROBE UNTIL HAVE NUMBER OF NICHTS FOR EACH TYPE OF LODCINC USED.) ENTER NUMBER OF NICHTS FOR EACH ( 1) H o te l (2) H o te l 1 I-----ASK Q15a 1 (3 ) R en ted c a b i n , c o t t a g e , v a c a t io n b o u se (6) Owned c a b i n , c o t t a g e , v a c a t io n (5 ) P u b lic c a a p g r o u n d s -t e n t ( 6) P u b l i c ca a p g ro u n d s-R .V . (7 ) (8) (9 ) ( 10) ( 11) ( 12) (6 4 - 4 5 ) (7 2 -7 3 ) (4 6 -4 7 ) (7 4 -7 5 ) ___ (4 8 -4 9 ) (7 6 - 7 7 ) (5 0 -5 1 ) ttouse (5 2 - 5 3 ) (7 8 -7 9 * 1 qup ( 6 - 7 ) 5 ). (5 4 -5 5 ) ( 8 -9 ) ( 5 6-57) ( 10- 11) ( 5 8 -5 * ( 12- 11) (6 0 -6 1 ) (1 4 - 1 5 ) P r i v a t e c a a p g r o u n d s -t e n t P r i v a t e ca a p g ro u n d s-R .V . ___ F r i e n d ' s h o u se R e l a t i v e ' s h o u se „ (6 2 -6 3 ) (1 6 - 1 7 ) (6 4 - 6 5 ) (1 8 - 1 9 ) (6 6 - 6 7 ) ( 20- 21) (6 8 -7 1 ) (2 2 -2 5 ) R e s o r t , s p a , dude ran ch O th er (WRITE IH) — TRIP 1 . TRIP 2 . 1 5 a . D id you a r e / a t a . . . ji d tc g (ASK FOR UP TO 2 HOTELS AND 2 HOTELS.) (1 ) A c h a in h o t e l ( a o t e l ) ? (2 ) An In d e p e n d e n t h o t e l ( a o t e l ) ? ENTER ONE CODE FOR EACH HOTEL/ HOTEL 7 2 i ) 7 2 8 ) ( 30) T32) W 16. )T S ) JM D id you s t a y o v e r n i g h t i n any o t h e r s t a t e s ( o r p r o v i n c e s ) w h ile you w e re t r a v e l i n g t o o r f r o a ( DEST. STATE)? (1 ) Tea - CONTINUE (2 ) Ho - SKIP TO Q17 ENTER ONE CODE JQSL 1 6 a . V h lch s t a t e s and how su n y n ig h t s ? SEE CODING NEXT PACE. -5 - 187 H ost R e ca n t REENTER DESTINATION 16a. W hich •caeca and how oan y m ig h ts ! STATE 1 XF MICHIGAN MENTIONED, ASK Q16b AND CONTINUE OTHERWISE, SKIP TO Q17. 7 ENTER NUMBER 3 OF RIGHTS, UNDER MICHIGAN 4 0 1 OTHER. 5_ (36) 144) (5 2 ) (6 0 ) (3 7 ) [4 5 ) (5 3 ) (6 1 ) 08) M) (5 4 ) (6 2 ) 0 9 ) (4 7 ) ( 5 5 ) (6 3 ) (4 0 ) [4 8 ) (5 6 ) (6 4 ) 6_ (4 1 ) 4 9 ) (5 7 ) (6 5 ) (4 2 ) 5 0 ) (5 8 ) (6 6 ) (4 3 ) (5 9 ) 7 <1-4)MT What ty p e o f lo d g in g d id you u aa l o M ic h ig a n ? How many n ig h t s ? (PROBE UNTIL HAVE NUMBER OF NICHTS WITH EACH TYPE OF LODCINC USED.) (1 ) H o te l ( 2) M otel (5 )5 (6 -7 ) ASK Q16c (3 ) ENTER NUMBER OF NICHTS FOR EACH \~ ( B) ( 9) (7 0 ) ( 10) R ented c a b i n , c o t t a g e , v a c a t io n houae (4 ) O w ed c a b i n , c o t t a g e , v a c a t io n h ouae _ (5 ) P u b lic ca a p g r o u n d s-te n t ( 6) P u b lic c a a p g r o u n d a -R .V . (7 ) P r iv a te ca a p g r o u n d s-te n t (8) P r i v a t e c a a p g r o u n d s -R .V . (9 ) F r i e n d ' s h o u se (10) (6 9 ) (7 1 ) ( 11) (7 2 ) ( 12 ) (7 3 ) (1 3 ) (7 4 ) (1 4 ) (7 5 ) (1 5 ) (7 6 ) ( 16) (7 7 ) (1 7 ) R e la t iv e ' s b ou se (11 ) R e s o r t , s p a , d ud e ra n ch [12) O th er (VRITE IN ) (7 8 ) (7 9 ) • • TrflAL MUSI • • 1C CHICAN 5 * • II Q 1 6 .. TRIP I .. TRIP 2 . 16c. 17. D id you I t a y a t a . . . (1 8 ) DUAL*• (1 9 ) AL » * tIO -lfl * (A 5 t ron UP TO 7 HOTELS AND 2 HOTELS) (1 ) A c h a in h o t e l (m o te l)? (2 ) An in d e p e n d e n t h o t e l (m o te l)T ENTER ONE CODE 1 . FOR EACH HOTEL/ MOTEL 2 . (2 2 ) (2 4 ) (2 6 ) (2 8 1211 m H cv many m i l e s o n e way d id you t r a v e l when y o u v e n t t o ( DEST. ) ? ENTER NUMBER OF MILES (.3 0 -1 3 ) (3 4 -3 7 ) -6 - 188 M ott R ecen t 1 2 REEHTEt DESTINATION IS . i Vhat h in d o f t r a n s p o r t a t i o n d id you u s e g e t t i n g t o (DE5T. STATE)? (1 ) (2 ) (3 ) (4 ) (5 ) (6) (7 ) (B ) (9 ) ENTER one CODE B anted c a r Owned c a r B anted RV Owned RV P la n e T ra in Bus B oat O th er (VE1TE IN) ( 3 8 - 3 9 ) (4 0 - 4 1 ) TRIP 1 . TRIP 2 . 19. V hat k in d o f t r a n s p o r t a t i o n d id you u s e I n (DEST . STATE)? ( 1) (2 ) (3 ) (4 ) (5 ) ( 6) (7 ) (8) (9 ) ENTER ONE CODE B anted c a r Owned ca r R anted RV Owned RV P la n e T ra in Bus Boat O cher (WRITE IN ) (4 2 - 4 3 ) (4 4 - 4 5 ) TRIP 1 . TRIP 2 . 20. Bow many B i l e s , I f a n y , d id y o u t r a v e l I n (DEST. STATE) f o r to u r in g o r a id e t r i p s ? ENTER NUMBER OF NILES__ 21. ( 4 6 -4 0 ) (4 9 - 5 1 ) <5Z> C 53) (5 4 - 5 5 ) ( 56-57) ( 58) (5 9 ) Have you o v e r b een t o (D E S T .) b e f o r e t U) T ea - CONTINUE ( 2 ) Bo - SKIP TO Q22 ENTER ONE CODE_ 2 1 a . B ot in c l u d i n g t h i s t r i p , how oa n y t l s e s In t h e p a s t 3 y e a r s ? ENTER NUMBER OF TIMES__ IF BOTH DESTINATION STATES THE SAME, ASK FIRST TIME THROUCH ONLY. 2 2 . - IF LIVES(D) IH DESTINATION STATE, DO NOT ASK. SKIP TO Q23. Have you a v e r b een t o (DEST. STATE) b e fo r e ? (1 ) T ea - CONTINUE ( 2 ) Bo - SKIP TO Q23 ^ SKIP*TO^Q23*r " ENTER ONE CODE__ 2 2 a . Not I n c lu d in g t b l o t r i p , h o v a a n y t l a e s i n t h e p a s t 3 y e a r s ? ENTER NUMBER OF TIMES ( 62-63) ( 4 0 -6 1 ) a a nu T BE EQU X TO OR • a CR ATER THA 02 1 a . Mow j u s t a few q u e s t i o n s a b o u t p la n n in g f o r y o u r t r i p . 23. How many w eek s b e f o r e y o u w e n t t o (D EST.) d id you c h o o se (DEST.) a s you r f i n a l d e s t i n a t i o n ? ENTER NUMBER OF WEEKS < 6 4 -6 9 ( 6 6 - 6 ?) 189 Moat R e cen t _ 2_ REENTER DESTINATION 24. Who b e s i d e s y o u r s e l f ch o a a (D E S T .)? (CET ACE VHERE NECESSARY: " l a t h a t u n d er 16 o r 16 and o v a r t " ) ( 1) ( 2) O) ENTER (4 ) ONE (5 ) CODE ( 6 ) (7 ) (8) (? ) 25. S a i f o n ly S p o u s e / o p p o s i t e s e x co a p a n io n C h ild u n d er 16 C h ild 18 and o ld e r R e l a t i v e u nd er 18 R e l a t i v e 16 and o ld e r F r ie n d u n d er 18 F r ie n d 16 and o l d e r b u s in e s s a s s o c ia t e /o t h e r (6 8 -6 9 ) (7 0 -7 1 ) When y ou w ere c o n s i d e r i n g y o u r t r i p t o ( DEST. ) , d id you h ave a se c o n d c h o ic e ? (1 ) Yea - CONTINUE (2 ) No - SKIP TO Q26 ENTER ONE CODE 2 5 a . I d v h a t a t a t a waa y o u r aacond c h o ic e ? <73) jm (7 4 -7 6 ) (7 7 -7 9 ) (URITE I N .) (1-4)D U P (5 )6 2 5 b . V h at l a th e bo i t Im p o rta n t r e a e o n yog c h o s e ( DEST. ) o v e r you r a ec o o d c h o i c e ? (VRITE IN ) TRIP 1 . ________________________________________________ DO MOT < fc-7 ) ( 8 -9 ) ( 10) (1 1 ) ( 12) (1 3 ) On a a c a l c f r o n 1 t o 5 v l t h 1 b e in g th e v o r a t t r i p you e v e r had and 5 b e in g t h e b e a t y ou e v e r h a d , a l l I n a l l , b o v would you r a t e y o u r o v e r a l l t r ip ? You c a n c h o o s e 1 w h ich l a w o r s t , 5 w h ich l a b e a t o r any n uab er l a b e t v e e o . ENTER ONE CODE (1 ) (2) O) (4 ) (5 ) V orat B eat 2 6 a . U s in g t h e aa a e e c a l e , bow w ou ld you r a t e (D EST .) (o r " your tour**)? ENTER ONE COOE (1 ) (2) (3 ) (4 ) (5 ) Vorat R e st -8 - 190 H ost R ecen t I REENTER DESTINATION 27. A l l l a a i l , was you r t r i p W t t i r , a b o u t t h e y o u e x p e c te d ? rnn ONE CODE M c’r“*'Ci(i) (3 ) k tta r < ^ 5) m u , o r t n r s i th an A l o t w o r s t th a n e x p e c te d S o a cw h a t w o r s e th a o e x p e c te d A b ou t Ch« M M S o e e w h a t b a t t e r th a n e x p e c te d A l o t b e t t e r th a n e x p e c te d ( 13) (1 6 ) (1 7 ) (IB ) (1 9 ) ( 20) ( 21 ) (2 2 - 2 4 ) (4 1 -4 3 ) (2 5 - 2 7 ) (4 4 -4 6 ) 2 7 a . A l l I n a l l , was (BEST. ) ( o r “y o u r t o u r " ) b e t t e r , a b o u t th e e a s e , o r w o r se th an you e x p e c te d ? t EKTER ONE CODE 26. O) * ^ (2 ) (3 ) . - —( 4 ) ^ (5 ) How l i k e l y o r u n l i k e l y a r e you t o v i s i t A l o t w o r s e th a n e x p e c te d Soekevhnt w o r s e th a n e x p e c te d A b out t h e sa n e S o a e w h e t b e t t e r th a n e x p e c te d A l o t b e t t e r th a n e x p e c te d ( DEST. STATE) a g a in ? ( 1) (2> (3 ) (4 ) (5 ) ENTER ONE CODE 29. How w e l l d o you r c e e a b e r how ou ch y ou s p e n t o n t h i s t r i p ? EKTER ONE CODE 2 9 a .“ IF Q29 IS ( 1 ) OR ( 2 ) : “H e l l , l e t ’ s t r y i t How s u c h d id you spend on . . . ( 1) ( 2) (3 ) (4 ) (READ) Hot a t a l l w e ll Hot t o o w e ll Somewhat w e ll V ery w e ll a n y w a y ." (READ) a. ACCEPT 0HL? TOTAL IF DOESN’ T DOW DETAIL (READ) H ot a t a l l l i k e l y H ot to o l i k e l y N eu tra l Somewhat l i k e l y V ery l i k e l y T r a n s p o r ta tio n ENTER b DOLLARS SPENT c FOR EACH d (2 8 - 3 0 ) (3 1 - 3 3 ) e O th e r B l s c e l l a n e o u s Ite m s (5 0 -5 2 ) _ (5 3 -5 5 ) TOTAL (3 7 - 4 0 ) (5 6 -5 9 ) -9 - 191 ASK rot EACH DESTINATION STATE, riKST TIME TOROUGH ONLY. 30. ir BOTH T tlP S IN THE SAME STATE, ASK D id you ( « « , b e a r , o r re a d a n y th in g a b o u t t r a v e l t o ( DEST. STATE) b e f o r e , d u r in g o r a f t e r you r t r i p t o (PEST.I T (RECORD ON C8 ID BELOW.) (1) Tea - CONTINUE (2 ) No RETURN TO TRIP SPECIFICS, Q 13, FOR SECOND TRIP. COMPLETED, SKIP TO Q 31. 3 0 a . Where d id you s e e o r b e a r I t ? (ACCEPT AS RANT AS OFFERED. IF SECOND TRIP RECORD ON GRID BELOW.) j— ASK FOR EACH SOURCE: Was I t b e f o r e , d u r in g , o r a f t e r you r t r ip ? Q 30. SEEN/HEAXD7 Q 30a. WHERE? DESTINATION STATE 2 ( 1 ) T ea ( 2 ) Bo (6 1 ) (CHECK) DESTITUTION STATE J. ( 1) Yea (21 Ho (6 0 ) (PUT "X" IN APPROPRIATE BOXES.) HOT SEEN/ HEARD BEFORE DUR1NC AFTER NOT SEEN/ HEARD BEFORE DURING AFTER a . TV (6 2 ) ( 9) b . RADIO (6 3 ) (10) (6 4 ) (1 1 ) c . N ew spaper: Was t h a t th e tr a v e l s e c tio n o r s o e e o th e r s e c t i o n ? (CHECK) ( 1) t r a v e ( 2) tth er (6 5 ) ( 1) t r a v e ( 2 ) < th e r (12) 6 6 -6 ) d . M agazine Which one? (WRITE IN MAGAZINE NAME.) 1 3-1 ) (1 5 ) ( 16-1 ) (6 8 ) 69-7C ) (1 8 ) (7 1 ) (7 2 ) (1 9 ) f . T r a v e l a g en t (7 3 ) (2 0 ) r. e. B illb o a r d s (7 4 ) (2 1 ) h . AAA o r o th e r a u to c lu b (7 5 ) (2 2 ) 1 . S t a t e 's to u r is t c e n t e r , 600 D u m b e r (RECORD HERE AND ON C ID ON hEXT PACE. (7 6 ) (2 3 ) J . C haabcr o f C oas e r c e o r o th e r s t a t e so u r c e (7 7 ) (2 4 ) k . At d e s t i n a t i o n (7 8 ) (2 5 ) F r ie n d /r e la t iv e 1 . O th er (WRITE IN OTHER.) ( - 4 ) Dl 5(5)1 ( 2( -2 7 ) (6 - 7 ) (2 8 ) (8 ) RETURN TO TRIP SPECIFICS. Q13 FOR SECOND TRIP. IF SECOND TRIP CUMPLETED, CONTINUE. -1 0 - 192 31. O ld you u s * u y a u sb ara? •00 (1 ) Y es - COKHHUE 3 1 * . U h tch o n e s ? 32. —| ( o t h e r ) i u t « i ' o r p r o v in c e * * t o u r i s t c e n t e r s o r t o l l ( r e * (I) Ho - SKIP TO Q33 (29) (RECORD OH CHID BELOW.) fok each s t a t e hakked on o l id s e w a. D id you g e t l i t e r a t u r e f r o a ( STATE)T (RECORD ON CRZD BELOW.) b« O ld you r e c e i v e I t b e f o r e , d u r in g o r a f t e r you t r a v e l e d ? (RECORD ON GRID BELOW.) Q31a. STATES u s e d t o u r l a t c t r / t o l l f r e e n u ab er (WRITE IN STATE) Q 32a. A b . R e c e iv e d l i t e r a t u r e ? When? (fU T HI M IN APPROPRIATE BOXES) L it e r a t u r e Mot r e c e iv e d B e fo r e L i t e r a t u r e R e c e iv e d D u rin g A fte r 0 0 -3 2 ) (4 8 ) (1 3 -1 5 ) (4 9 ) (3 6 -3 8 ) (5 0 ) 0 9 -4 1 ) (5 1 ) (4 2 -4 4 ) (5 2 ) (4 5 - 4 7 ) (5 3 ) |TRAVEL PREFERENCE^ N ov v c h ave J u s t a fe w q u e s t i o n s ab out v h a t you l i k e t o d o when you t r a v e l f o r p l e a s u r e . 33. Whet I s th e f a r t h e s t d i s t a n c e you a r e v l l l i n g t o t T a v e l o n e way b y c a r f o r a . . . (READ. WRITE IN NUMBER OP MILES.) a. b. c. 34. One d ay t r i p _______ Weekend t r i p _ _ _ _ _ _ One week t r i p (5 * -5 6 ) (5 7 -5 9 ) (6 0 -6 2 ) What o u td o o r a c t i v i t i e s d o you l i k e t o d o when you t r a v e l ? ACCEPT UP TO 7 . ) ( 1) ( 2) (3 ) (4 ) (5 ) (6) (7 ) (0 ) (9 ) ( 10) ( 11) ( 12) B ic y c l i n g Power B o a tin g C an oein g S a ilin g R a f tin g C aap lng C h a r te r b o a t f i s h i n g S t r e a a /r iv e r f is h in g Lake f i s h i n g I c e f is h in g C o if T e n n is (1 3 ) 04) (1 5 ) (1 6 ) (1 7 ) (1 6 ) (1 9 ) ( 20) (21) (22) (2 3 ) (DO MOT READ. H u n tin g H lk in g /b e c k p a c k ln g H orseb ack r i d i n g D o w n h ill s k i i n g C r o s s c o u n tr y s k i i n g S n o v u o b llln g S w ln s lo g W ater s k i i n g S u n b a th ln g /b e a c h M i s c e lla n e o u s s p o r t s O th er (WRITE IN) CHECK. (6 3 -6 4 ) (6 5 -6 6 ) (6 7 -6 8 ) (6 9 -7 0 ) (7 1 -7 2 ) (7 3 -7 4 ) (7 5 -7 6 ) - 11- 193 (1-4)DU1 )). What e i g h t s , i v n t i o r a U T t a l n a t n t d o y o u l i k e t o u r t l e i M t i l o t CHECK. ACCEPT UP TO 5 . ) . «> (2) (J ) (4 ) (5 ) <*> (7 ) (8) B .t u r .1 . n r o c t l o o u L io d a ir k . l l . t o r l c .1 s ig h t . K u -* » d « a t t r o c t i o a s M u ltu al C u ltu r a l c v c a t s r r o ia .B lo a a l/.p .c t a t o r sp o r tu f a i r ., e x h ib it. M ig h t d u b s , sh o w s, r e s t a o r a a t s I n d u s t r y to u r s fe s tiv a ls M ovies O th e r (WRITE IN) (» ) ( 10) ( 11) ( 12) (1 3 ) o>» (DO MOT ICAD. ( *-7 ) ( 8 -9 ) ( 10- 1 1 ) (1 2 -1 3 ) (1 4 -1 5 ) lATTRIBUTES/lHAGEAYl 36. Mow I ’o g o in g t o read b o m w o r d s t o y o u t h a t d e s c r i b e s t a t e s . I ' d l i k e you t o t e l l m how im p ortan t each o n e l a t o y o u v h s o you t r a v e l t o a s t a t e . The s c a l e w e ' l l b e u s in g I s s 1 t o 3 s c a l e w ith 1 b e l o g " n o t a t a l l I m p o r ta n t” end 5 b e in g "one o f t h e B oat Im p ortan t" . L e t ' s t r y o n e : go o d w e a t h e r . How im p o r ta n t t o you l a "good w e a th er" when you t r e v e l t o a s t a t e : 1 , w h ic h w e a n s c o t a t a l l I w p o r ta n t, 2 , 3 , 4 o r 5 , w h ich M a n s o n e o f th e w oat I w p o r ta n t? (READ. CHECK. ROTATE. HARK ROTATION START. REPEAT SCALE AS NECESSARY.) (1 6 -1 7 ) Out o f R ot a t a l l th e D o st 1st p o r t e n t ls o o r ta o t 2 3 4 1 5 a. b. c. d. c. f. f. h. 1. J. k. 1. m. n. 0. 37. Good s c e n e r y P eace and q u ie t F a a lly fu n Good r e s ta u r a n ts F r ie n d ly p eo p le E asy t o g e t t o E x c lte s e n c R eason ab le p r i c e s Good p la c e s t o s t a y L o ts o f t h in g s t o do S u a ser fun H igh p r e s t i g e v a c a t i o n s C lean a i r Good n ig h t l i f e W inter fu n ( 1) (1) ( 1) ( 1) ( 1) ( 1) ( 1) ( 1) ( 1) a ) (i) < i) a ) <» <» <2 > a> (2) (2) ( 2) ( 2) (2) (2) (2) ( 2) ( 2) ( 2) ( 2) ( 2) (2) (3 ) (3 ) (3 ) (3 ) (3 ) (3 ) (3 ) O) (3 ) (3 ) (3 ) (3 ) <3) (3 ) (3 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) <4) (4 ) (4 ) <4> (3 ) (i) (5 ) (3 ) (3 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (1 6 ) (1 9 ) ( 20) (2 0 ( 22) (2 3 ) (2 4 ) (2 5 ) (2 6 ) (2 7 ) (2 8 ) (2 9 ) (3 0 ) (3 1 ) (3 2 ) T od ay, w a 'rc c o n c e n t r a t in g o n s t a t e s I n t h e C r e a t L akes R egion o f th e c o u n tr y . How wany t l w i have you b e e n t o . . . i n t h e l a s t 3 y e a r s ? (READ. WRITE IN BUKRER OF T1HES. INDICATE IP UVES(D) THERE.) a . O hio (3 3 ) e. P e n n s y lv a n ia b. W isco n sin ( 34) f. O n ta r io e. I llin o is ( 35) g. I n d ia n s d. M ichigan (3 6 ) ____________ ( 37) (J g ) (3 9 ) b* M in n e so ta (4 0 ) We know th a t you way n o t b a v e b e e o t o e v e r y s t a t e b u t b aaed j u s t 00 your is r p r e s e lo n a , on what y o u 'v e s e e n o r h e a r d o r r e a d , o v e r a l l , bow would you r a t e th e s e s t a t e s a s good o r bad p la c e s t o t a k e a v a c a t io n ? 1 s c a n s a s t a t e t o aaong th e worse p la c e s t o ta k e s v a c a t io n end 5 w e a n s a s t a t e l a s a o n g t b e b e s t p la c e s t o ta k e a v a c a t io n and you can c h o o se any m taber i n b e tw e e n . (READ. CHECK. ROTATE. MARK ROTATION START. REPEAT SCALE AS NECESSARY.) (4 1 ) B est J W orst a. b. c. d. e. f. I* h. Ohio W iscon sin I llin o is M ichigan P e n n s y lv a n ia O n ta rio lo d la o a M in nesota IC O TO 1 ~ r ~r (1) (1) (1) ( i) (2 ) (2 ) (2 ) (2) (2 ) (2 ) (2 ) (2) (3) (3) (i) I A T I H C S H E E T (3 ) (3 ) (3 ) (3 ) (3 ) (3 ) POR (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (4 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (5 ) (3 ) Q39. (4 2 ) (4 3 ) (4 4 ) (4 3 ) (4 6 ) (4 7 ) (4 6 ) (4 9 ) 194 Q )9 b t l « | S h e e t 3>. Sew w s 'd l i k e y o u t o r e t s a fe w o f t h e s e s t a t e s on s e v e r a l I t e m s . A g a in , w e'd l i k e t o koow y o u r I m p r e s s io n s b a s e d l u s t e o what y o u 'v e s e e n o r h e s r d o r r e a d , l i t want t o know how o u ch you a g r e e o r d is a g r e e th a t a s t a t e l a known f o r s o m e th in g , t h e s c a l e w e ' l l h e v s t s g t h i s tim e l o t 1 I s s t r o n g l y d i s a g r e e an d 1 0 I s s t r o n g l y a g r e e , anJ you c a o c h o o s e s a y number I n b e tw e e n . For e x a m p le , " good w e a t h e r ." Do y o u 1 , stvonjr.ty d i s a g r e e , 1 0 , s t r o n g l y a g r e e , o r w ou ld you c h o o s e som e num ber i n b etw een t h a t C a l i ­ f o r n i a l a known f o r good w e a th e r ? l o t ' s s t a r t w i t h M ic h ig a n . l a M ich igan known f o r . . . ? READ. KHTEI WMBER OF RATING. REFEAT SCALE AS NECESSARY. ROTATE. MARI ROTATION START. HALF ROTATE FORWARD, HALF ROTATE BACKWARD. INDICATE FORUARD OR BACKWARD ROTATION WITH ARROW. (7 1 -7 2 ) (7 3 )______ (7*>_____ AFTER MICHIGAN RATINGS, READ RACK MICHICAN RATING FOR EACH ITEM IN TURN AND GIT 2 OTHER STATES' RATINCS FOR THAT ITEM. "Now you r a t e d M ic h ig a n a Ite m I I And s t a t e IT on I t c e 1 . (7 5 )_____ How, t h e n , w o u ld you r o t e e t a t e 7 eo (1 - 4 ) DUP (S-b)i2 M ic h ig a n W is c o n s in M in n esota a. Cood s c e n e r y ( 7 -* ) (3 7 -1 8 ) (6 7 -6 8 ) b. P ea c e and q u i t e < 9 -1 0 ) < 3 9 -4 o ) (6 9 - 7 0 ) c. F a m ily fu n d. <11- 1 2 ) < 4 1 -4 7 ) (7 1 - 7 7 ) (1 3 -1 4 ) (4 3 -4 4 ) (7 3 - 7 4 ) e. F r ie n d ly p e o p le (1 5 -1 6 ) (4 5 -4 6 ) (7 5 - 7 6 ) f . E asy t o n e t t o (1 7 -1 B ) (4 7 -6 0 ) (7 7 - 7 0 ) It* E x c ite m e n t (1 9 -2 0 ) (4 9 -5 0 ) (7 9 - 0 0 ) h. R e a s o n a b le p r i c e s ( 21- 2 2 ) (5 1 -5 2 ) ( 7 -0 ) 1. Good p l a c e s t o s t a y (2 3 -2 4 ) (5 3 -5 4 ) ( 9 -1 0 ) 1- l o t s o f th in g s t o do (2 5 -2 6 ) (5 5 -5 6 ) (1 1 - 1 3 ) (1 3 - 1 4 ) k. Summer fu n (2 7 - 2 B ) (5 7 -5 0 ) 1. H igh p r e s t i g e v*r> '? io n s (2 9 -3 0 ) (5 9 -6 0 ) (1 5 - 1 6 ) m. C le a n a i r (3 1 -3 2 ) (6 1 -6 2 ) (1 7 - 1 0 ) 0. Cood n i g h t l i f e (1 3 -3 4 ) (6 3 -6 4 ) (1 9 - 2 0 ) o. W in ter fu n (3 5 -3 6 ) (6 5 -6 6 ) (2 1 - 7 2 ) 195 40. T he f o l l o w i n g t e w e t a t e a c o t a a r e co n ce rn ed v i t h y o u r o p in i o n s a b o u t v h a t u k < a a good v a c a t io n . I ' d I l k a t o know how a u ch you a g r e e o r d i s a g r e e w it h e a c h i t a t m a t . Th« a c a l a t h l a t l a a l a X, s t r o n g l y d i s a g r e e , 5 , s t r o n g l y a g r e e o r you c a n c b o o ta an y au ab ar l a b ttw e a n . L e t ' s t r y ) (I S ) ( 1 J> KOTEL/HOTEL | 2 . __ 0 2 ) (1 4 ) (1 6 ) (1 8 ) Bow w e l l d o you r e s e a b e r how s u c h t h i s ( t i p c o s t ? (1 ) ENTER (2 ) Oftfe (3 ) CODE (A ) d I T r a n s p o r t a t io n (2 6 -2 8 ) < 4 5 -4 7 ) (6 4 -4 5 ) (2 9 -3 1 ) (4 6 -5 0 ) (6 7 -6 9 ) (3 2 -3 4 ) (5 1 - 3 3 ) (7 0 -7 2 ) (3 5 -3 7 ) (5 4 - 5 6 ) (7 3 -7 3 ) (3 6 -4 0 ) (5 7 -3 9 ) (7 6 -7 8 ) (4 1 -4 4 ) (6 0 -6 3 ) ( 7 -1 0 ) b. C. d. e. O th e r a l s c e l l a n e o u s I te m s < TOTAL c o t o Q41 r o e u r n b u s in e s s m r . IF BO MORE BUSINESS TB1FS, CONTINUE. -1 4 - (1 -4 ) DUP - 6 )1 1 197 fpDIXRAfHlCS) f i n a l l y , your an sw ers w i l l ha m a y o u r a tlf a a f you r l a s t l y . a e n h e l p f u l t o u a I f y ou a a i v i r a fa w q u e s t i o n s a b o u t [ I f MICHIGAN RESIDENT, ASK Q 45-- 15- 198 33. S o y o u own a c o o p e r o r IV ? (1 ) 34. L ass th a n h ig h s c h o o l H ig h s c h o o l g r a d u a te T r a d e /te c h c lc a l Soae c o lle g e W h ite H ack (HEAD. CHECK.) ( 5 ) C o lle g e g r a d (4 ) l o s t d e g r e e ( 7 ) R e fu se d ( 0 0 HOT HEAD) (SO) (CHECK. I F MIXED, CET RACE OF HALE AND FEMALE BEADS OF HOUSEHOLD.) ( 3 ) H isp a n ic ( 4 ) O th er (3 ) R e fu se d (DO HOT READ) (S I) What w a s y o u r t o t a l f a a l l y ln c o a e i s 1983 f r o a a l l s o u r c e s b e f o r e t a x e s ? (READ. CHECK.) L e a s th a o 9 1 0 ,0 0 0 •1 2 0 ,0 0 0 -• 9 3 0 ,0 0 0 -• ( 1) (2) (3 ) (4 ) 37. (4 9 ) I s t h i s a w h i t e , b lo c k o r H isp a n ic h o u seh old ? (1 ) (2 ) 34. ( 2 ) Bo ifh o t w oo t h s h ig h e s t g ra d e you com p leted l a s c h o o l o r t r a i n i n g ? (1 ) (2 ) (3 ) (4 ) 35. {CHECK.) T oo RIO ,000 9 1 9 ,9 9 9 9 2 9 ,9 9 9 9 3 9 ,9 9 9 (5 ) (6) (7 ) (8) (9 ) 9 4 0 ,0 0 0 9 5 0 ,0 0 0 9 6 0 ,0 0 0 9 7 0 .0 0 0 R e fu se d - 9 4 9 ,9 9 9 - 8 5 9 ,9 9 9 - 9 6 9 ,9 9 9 or so re (DO HOT READ) Would you p l e a s e t e l l a e y o u r I I P (o r p o s t a l ) c o d e ? ______________ (5 3 - 5 8 ) TUAHK YOU VERT SC O FOR YOUR COOPERATION lo te r v le w c r D a t e _________________ (5 9 -6 1 ). L e o g th o f I n t e r v ie w (6 2 -6 3 > ADI_________________ (6 4 -6 6 ) STATE (6 7 -6 9 ) (1 ) -C rca t L a k e s (2 ) D e s t i n a t io n (3 ) I n q u ir y ( 70) -1 6 -