INFORMATION TO USERS This was produced from a copy o f a document sent to us fo r m icrofilm ing. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality o f the material submitted. The following explanation o f techniques is provided to help you understand markings or notations which may appear on this reproduction. 1. The sign or “ target” fo r pages apparently lacking from the docum ent photographed is “ Missing Page(s)” . If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you o f complete continuity. 2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been film ed, you will find a good image of the page in the adjacent frame. If copyrighted materials were deleted you will find a target note listing the pages in the adjacent frame. 3. When a map, drawing or chart, etc., is part o f the material being photo­ graphed the photographer has followed a definite method in “ sectioning” the material. It is customary to begin film ing at the upper left hand corner of a large sheet and to continue from left to right in equal sections w ith small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete. 4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department. 5. Some pages in any document may have indistinct print. In all cases we have film ed the best available copy. University Microfilms International 300 N. ZEEB RD., ANN ARBO R, Ml 48106 8202437 G a r t n e r , W illia m C r a ig A STUDY OF OWNER CHARACTERISTICS, HOME DEVELOPMENT, AND LAND VALUE DETERMINANTS IN SELECTED AREAS OF MICHIGAN’S NORTHERN LOWER PENINSULA Ph.D. Michigan State University University Microfilms International 300 N. Zeeb Road, Ann Arbor, M l 48106 1981 PLEASE NOTE: In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this docum ent have been identified here with a check mark V 1. Glossy photographs or p ages______ 2. Colored illustrations, paper or print 3. . Photographs with dark background_____ 4. Illustrations are poor copy______ 5. Pages with black marks, not original 6. Print shows through as there is text on both sides of page_____ 7. Indistinct, broken or small print on several pages 8. Print exceeds margin requirements______ 9. Tightly bound copy with print lost in spine______ 10. Computer printout pages with indistinct print______ 11. Page(s)_____________lacking when material received, and not available from school or author. 12. Page(s)_____________seem to be missing in numbering only as text follows. 13. Two pages num bered_____________ . Text follows. 14. Curling and wrinkled pages______ 15. Other___________________________________________________________________________ copy__ University Microfilms International A STUDY OF OWNER CHARACTERISTICS, HOME DEVELOPMENT, AND LAND VALUE DETERMINANTS IN SELECTED AREAS OF MICHIGAN'S NORTHERN LOWER PENINSULA By W illiam C. Gartner A DISSERTATION Submitted to Michigan State U n iv ersity in p a r t ia l f u l f i l l m e n t o f the requirements fo r the degree of DOCTOR OF PHILOSOPHY Department o f Resource Development 1981 ABSTRACT A STUDY OF OWNER CHARACTERISTICS, HOME DEVELOPMENT, AND LAND VALUE DETERMINANTS IN SELECTED AREAS OF MICHIGAN'S NORTHERN LOWER PENINSULA by W illiam C. Gartner The purpose of th is study was to acquire and analyze data o f landed property and home owners in three representative counties (Kalkaska, Otsego, Crawford) of Michigan's northern lower peninsula. Primary data were collected via a mail survey o f a s t r a t i f i e d random sample of landowners in three selected townships o f each county. S tra ta were chosen based on homogenity fo r c e rta in natural resources ( la k e , r i v e r , no water resource). Landowners, through survey responses, were segmented in to three types: 1) permanent home owners, 2) seasonal home owners, and 3) pro­ perty owners with no home development in the area. A ttitu d e s and concerns, socio-economic c h a r a c te r is tic s , value and amount of acreage owned, in te n t to s e ll property, method and reasons fo r property acquisi­ tions and information sources o f property a v a i l a b i l i t y were obtained fo r each landowner type. Natural resources were examined to estimate what e f f e c t location r e la t iv e to c e rta in natural resources had on valua­ tio n of real property. Study results in d ica te th a t the most important source fo r learning o f a v a ila b le property was friends or r e la t iv e s . Property was generally acquired fo r investment or retirem ent home p o te n tia l although recreation al a c t i v i t i e s ranked high as a major reason fo r a c q u is itio n . The fu tu re fo r potential property sales was found to be quite high and property owners with no home development on t h e ir land are more apt to se ll than other types of property owners. Property owners thought current property tax lev e ls high but were generally s a tis f ie d with qu antity and q u a lity o f municipal services pro­ vided and f e l t property values w i l l continue to increase. Property owners would l i k e a t le a s t the present level of land use controls main­ tained , and they favor a t le a s t a l i t t l e more re s id e n tia l development. Natural resource c h a ra c te ris tic s found to be re la te d to a s ig n if ic a n tly higher value per acre of land were: location on lakes greater than 25 acres but less than 100 acres in s iz e , location on lakes greater than 500 acres in s iz e , and in one case, location close to a commercial ski area. Natural resource c h a ra c te ris tic s found to be re la te d to a s ig n if ic a n t ly lower value per acre of land were: location adjacent to public land and an increasing amount of acreage owned. ACKNOWLEDGMENTS The author wishes to thank the members of his committee; Professors Donald Holecek, M ilton S teinm ueller, Raleigh Barlowe, Glenn Johnson and e s p e c ia lly Daniel Chappelle, who also served as d is s e rta tio n adviser and e d ito r . Special thanks are extended to my s i s t e r , Beth Kish, whose patience as my ty p is t exceeded a l l normal tolerance lim i t s . My w if e , Rose Gartner, is due enormous g ratitud e fo r her economic and continued moral support during the course of my work. Very special thanks are extended to my parents, Henry and Adele Gartner; I hope th is research e f f o r t makes them as proud o f me as I am o f them. F i n a lly , g ratitu d e is extended to a l l my friends who always help make the problems one encounters seem sma11e r . ii TABLE OF CONTENTS L i s t o f Tables........................................... vi L i s t of Figures............................................................................................................. xi Chapter I II INTRODUCTION................................................................................................. 1 Problem Statement.................................................................................. 1 Objectives of the Study...................................................................... 6 1) Primary O b jective...................................................................... 6 2) S pec ific O bjectives.................................................................. 6 RESEARCH METHODS........................................................................................ 10 Defining the Study Area........................................................................ 10 Sampling S trategy.................................................................................. 15 Measurement and V ariab le I d e n t i f i c a t i o n .......................................24 SPSS Subprograms Used fo r Data A nalysis.......................................25 Variables to be Tested Using Regression.................................... 43 1) Measurement o f the Dependent V a r ia b le .............................. 44 2) Measurement of the Independent V a r ia b le s ..................... 44 Adjusting the Classical Linear Regression Model................... 49 S ignificance Level................................................................................ 52 Non-Response E rro r ................................................................................ 53 S t a t i s t i c s ................................................................................................. 53 Summary....................................................................................................... 58 iii Chapter III IV V SOCIOECONOMIC CHARACTERISTICS OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS............................................................. 59 Age of Household.............................................................................. 59 Gender o f Household H e a d ... ....................................................... 60 M a rita l Status o f Household Head............................................ 60 Family S iz e ............................... 63 Family Income.................................................................................... 63 Summary................................................................................................. 69 TYPES OF HOME DEVELOPMENT....................... ........................................ 71 Type of Home....................................................................................... 71 Seasonal Home Usage........................................................................ 75 Region of Present Residence....................................................... 77 Region of P rio r Residence........................................................... 62 Summary................................................................................................. 84 INITIAL PROPERTY PURCHASE................................................................ 85 Method o f A c q u is itio n .................................................................... 85 Reason fo r A c q u is itio n .................................................................. 86 Information Sources th a t Lead to A c q u is itio n ..................... 93 In te n t to S e ll Property................................................................ 188 Summary............................... VI 165 ATTITUDES OF NORTHERN MICHIGAN STUDYAREA PROPERTY OWNERS ON ISSUES OF CONCERN............................................................ 107 Property Tax Levels........................................................................ 167 Q u ality o f Municipal and County S ervices............................ 114 Quantity o f Municipal and Countys e rv ic e s ............................ 121 Building Regulations............................... ... ................................... 128 iv Chapter Land Use Regulations........................................................................ 133 Residential B u ilding...........................................................................139 Future Property Values...................................................................... 145 Summary................................................................................................... 151 V II INFLUENCE OF NATURAL RESOURCES ON PROPERTY LOCATION AND VALUE.................................................................................................. 153 Influence o f Water Resources....................................................... 153 Influence of Public Land.................................................................. 158 Acreage Owned per Property Owner............................................... 162 Total Value o f Property per Owner............................................. 170 Summary................................................................................................... 175 V III VALUE PER ACRE......................................................................................... 177 Value per Acre of Land Model....................................................... 177 In tern al V a lid a tio n .......................................................................... 193 External V a l i d i t y .............................................................................. 194 Summary................................................................................................... 195 IX CONCLUSIONS AND RECOMMENDATIONS..................................................... 197 Conclusions........................................................................................... 197 Recommendations..................................................................................... 201 Appendix A Questionnaire....................................................................................208 Appendix B Planning Regions..............................................................................209 Appendix C M u ltip le Regression SimpleC orrelation M a tric e s Appendix D Mean and Standard Deviations fo r the M u ltip le Regression V a ria b le s .................................................................... 212 210 L is t of References ................................................................................................. 213 v LIST OF TABLES Table 1. 2. 3. 4. 5. 6. 7. 3. 9. 10. 11. 12. 13. 14. Questionnaires Mailed and Accounted f o r , by Township, County............................................................................................................... 22 V ariab le Name, Id e n tify in g Number, Measurement Scale, and V ariab le Explanation fo r a l l Variables Extracted from the Questionnaire.............................................................................. 26 V ariab le Name, Id e n tify in g Number, and S t a t is t ic s Used to Describe the V ariab le U t i l i z i n g Subprogram FREQUENCIES and CONDESCRIPTIVE...................................................................................... 33 V ariab le Relationships Analyzed Using Subprogram CROSSTABS and D escriptive S t a t is t ic s Computed............................. 37 V ariable Relationships and S t a t is t ic s Computed with Subprogram BREAKDOWN.................................................................................. 40 Range of Values and Corresponding Strength o f Association fo r Each Variables S t a t i s t i c a l l y Tested...................................... 57 Mean Age of Northern Michigan Study Area Property Owner, by County, Township, and Type of Home Development........................ 61 Gender of Household Head fo r Northern Michigan Study Area Property Owners (Frequency and Percentage D is t r i b u t i o n ) 62 M a rita l Status o f Household Head fo r Northern Michigan Study Area Property Owners (Frequency and Percentage D is t r i b u t i o n ) ................................................................................................. 62 Family Income of Northern Michigan Study Area Property Owners (Frequency and Percentage D i s t r ib u t i o n ) .............................. 63 Family Income of Northern Michigan Study Area Property Owners, by County....................................................................................... 65 Family Income of Northern Michigan Study Area Property Owners, by Township................................................. 67 Family Income of Northern Michigan Study Area Property Owners, by Type o f HomeDevelopment................................................... 69 Type o f Home Developmentin Northern Michigan Study Area (Frequency and PercentageD is t r i b u t i o n ) ........................................... 72 vi Table 15. Type of Home Development, by County.................................................... 73 16. Type o f Home Development, by Township................................................ 74 17. Number of Days Spent a t the Seasonal Home, by Season................ 77 18. Place of Permanent Residence of Northern Michigan Study Area Property Owners with Seasonal Home Development................. 80 Region of P rio r Permanent Residence (Frequency and Percentage D is t r i b u t i o n ) ......................................................................... 83 Northern Michigan Study Area Property Owners Methods of Property A c q u is itio n .................................................................................. 87 Major Reason fo r Property Acquisition (Frequency and Percentage D is t r i b u t i o n ) .......................................................................... 87 22. Major Reason fo r Property A cq u isitio n , by County......................... 89 23. Major Reason fo r Property A cq uisition , by Township..................... 91 24. Major Reason fo r Property Acquisition, by Type of Home Development......................... 93 Sources of Information th a t Lead to Property Purchases (Frequency and Percentage D i s t r ib u t i o n ) .......................................... 94 Sources o f Information th a t Lead to Property Purchases, by - County............................................................................................................... 95 Sources o f Information th a t Lead to Property Purchases, by Township........................................................................................................... 98 Sources o f Information th a t Lead to Property Purchases, by Type of Home Development.......................................................................... 99 19. 20. 21. 25. 26. 27. 28. 29. Northern Michigan Study Area Property Owner's Inten tion s Concerning Future Property Sales (Frequency and Percentage D is t r ib u t i o n ) ................................................................................................. 100 30. Type o f Home Development, by Desire to S ell Property.................. 102 31. Property Owners Plans to Sell Property (Frequency and Percentage D is t r ib u t io n ............................................................................ 103 32. Property Owners Plans to S ell Property, by Type o f Home Development........................................................................................................104 33. A ttitu d es o f Northern Michigan Study Area Property Owners Toward Current Property Tax Levels (Frequency and Percentage D is t r ib u t i o n ) ................................................................................................... 108 v ii A ttitu d es o f Northern Michigan Study Area Property Owners Toward Current Property Tax Levels, by County........................... 109 A ttitu d es of Northern Michigan Study Area Property Owners Toward Current Property Tax Levels, by Township....................... m A ttitu d es o f Northern Michigan Study Area Property Owners Toward Current Property Tax Levels, by Type of Home Development................................................................................................... 113 A ttitu d es o f Northern Michigan Study Area Property Owners Toward Q u ality o f Provided Municipal or County Services (Frequency and Percentage D is t r i b u t i o n ) ........................................ 115 A ttitu d es o f Northern Michigan Study Area Property Owners Toward Q u ality of Municipal or County Services Provided, by County....................................................................................................... 116 A ttitud es of Northern Michigan Study Area Property Owners Toward Q u ality o f Municipal or County Services Provided, by Township......................................................................................................... 118 A ttitu d es o f Northern Michigan Study Area Property Owners Toward Q u a lity o f Municipal and County Services Provided, by Type o f Home Development................................................................. 120 A ttitu d es of Northern Michigan Study Area Property Owners Toward the Quantity o f Municipal and County Services Provided (Frequency and Percentage D i s t r ib u t i o n ) ..................... 122 A ttitu d e s o f Northern Michigan Study Area Property Owners Toward Quantity of Municipal and County Services Provided, by County....................................................................................................... 123 A ttitu d e s of Northern Michigan Study Area Property Owners Toward the Quantity of Municipal and County Services Provided, by Township.............................................................................. 125 A ttitu d e s o f Northern Michigan Study Area Property Owners Toward the Quantity o f Municipal and County Services Provided, by Type of Home Development............................................ 127 Northern Michigan Study Area Property Owners Awareness of Regulations Concerning Land Development (Frequency and Percentage D i s t r i b u t i o n ) ........................................................................ 129 Northern Michigan Study Area Property Owners Awareness of Building Regulations Concerning Land Development, by Township........................................................... .............................................. 130 Northern Michigan Study Area Property Owners Awareness of Building Regulations Concerning Land Development, by Type of Home Development.................................................................................. 132 vi i i Table 48. 49. 50. 51. 52. A ttitud es o f Northern Michigan Study Area Property Owners Concerning Present Land Use Controls (Frequency and Percentage D i s t r i b u t i o n ) ............................................................................ 134 A ttitud es o f Northern Michigan Study Area Property Owners Concerning Present Land Use Controls, by Township....................... 135 A ttitud es o f Northern Michigan Study Area Property Owners Concerning Present Land Use Controls, by Type o f Home Development....................................................................................................... 137 A ttitud es of Northern Michigan Study Area Property Owners Concerning Present Land Use Controls, by Location to Water Resources............................................................................. 139 Northern Michigan Study Area Property Owners A ttitud es Toward Future R esidential Development (Frequency and Percentage D i s t r i b u t i o n ) ............................................................................ 140 53. A ttitud es o f Northern Michigan Study Area Property Owners Toward Future R esidential Development, by County.......................... 141 54. A ttitu d es o f Northern Michigan Study Area Property Owners Toward Future R esidential Development, by Township..................... 143 A ttitud es o f Northern Michigan Study Area Property Owners Toward Future R esidential Development, by Type of Home Development....................................................................................................... 145 A ttitud es o f Northern Michigan Study Area Property Owners Toward Future Property Values (Frequency and Percentage D is t r i b u t i o n ) ................................................................................................... 146 A ttitud es o f Northern Michigan Study Area Property Owners Toward Future Property Values, by County.......................................... 147 A ttitud es o f Northern Michigan Study Area Property Owners Toward Future Property Values, by Township...................................... 149 59. Location of Property to Water Resources, by County...................... 154 60. Location of Property to Water Resources, by Township.................. 156 61. Type of Water Resource Property is Located, by Township 62. Location of P riv a te Property to Public Land (Frequency and Percentage D i s t r i b u t i o n ) ............................................................................ 160 63. Location o f P riv a te Property to Public Land, by County 161 64. Location o f P riv a te Land to Public Land, by Township.................. 65. Total Acreage Owned per Northern Michigan Study Area Property Owner, by County................................................................. 55. 56. 57. 58. ix 159 163 165 Table 66. Total Acres Owned per Northern Michigan Study Area Property Owner, by Township............................................................................................ 165 67. Total Acres Owned per Northern Michigan Study Area Property Owner, by Type o f Home Development........................................................... 167 •68. Total Acreage Owned per Northern Michigan Study Area Property Owner, by Location to Water..........................................................................168 69. Total Acres Owned per Northern Michigan Study Area Property Owner, by Type o f Water Resource on Property Location.....................169 70. Total Acres Owned per Northern Michigan Study Area Property Owner, by Location to Public Land............................................................. 169 71. Total Value of Property Owned per Northern Michigan Study Area Property Owner, by Location to Water Resource..........................172 72. Value per Acre in Northern Michigan Study Area, by Type of Home Development................................................................................................. 179 73. Information from the Three Regression Equations fo r the Dependent V ariab le "Value per Acre of Land".........................................184 Appendix C M u ltip le Regression Simple C orrelation M a trices.....................210 Appendix D Mean and Standard Deviations fo r the M u ltip le Regression V a ria b le s ................................................................................................... 212 x LIST OF FIGURES Figure 1. Problem Solving Model.............................................................................. 7 2. Location o f Surveyed Counties............................................................. 16 3. Location of Surveyed Townships in T h eir Respective Counties........................................................................................................ 18 Mean Breakdown fo r Total Value of Property Owned, by Location on Water C o n tro llin g f o r Type of Home Development................................................................................................. 174 Value per Acre o f Land by Location to Water Resource C o n trollin g fo r Type o f Home Development.................................... 180 Value per Acre f o r Property Located on Water Resource C o n trollin g fo r Type of Home Development.................................... 181 4. 5. 6. Appendix A Questionnaire............................................................................... 208 Appendix B Planning Regions......................................................................... 209 xi CHAPTER 1 INTRODUCTION Problem Statement Land is a form o f wealth. From the beginning o f recorded time wars have raged over control of land. Economic science recognized land had value, thus the development o f the concepts of land rent espoused in the w ritin g s o f Von Thunen and Ricardo. The idea of land as wealth shows up today in the consumption theories o f Ando and Modigliani (1963) and Friedman (1957). Land, however, is not owned so le ly fo r speculation or because of i t s productive q u a l i t i e s . ment (V ertree s, 1967). Land is also owned fo r enjoy­ U t i l i t y is derived from walking on one's land, hunting on i t or even viewing i t . Thus, when an in d iv id u a l's marginal value product, or u t i l i t y derived, from a parcel of land exceeds ac q u isitio n p ric e , the in divid ual w i l l purchase the land, given adequate income levels and no other more a t t r a c t iv e a lte r n a t iv e s . Conversely, when an in d iv id u a l’ s marginal value product, or u t i l i t y derived from an owned parcel o f land is less than salvage value, the land w i l l be offered fo r sa le. Marginal value product, or u t i l i t y , is d i f f e r e n t fo r each individ ual and is the reason two in divid uals with s im ila r income constraints may d i f f e r markedly in amount o f land owned. Therefore, the marginal value o f land ownership is viewed d i f f e r e n t l y by any two consumers and true land value may only p a r t i a l l y be re fle c te d in a c q u isitio n or salvage p ric e . This aspect o f consumer behavior gives ris e to the theory o f consumer surplus, (M arshall, 1891). The northern lower peninsula o f Michigan has experienced rapid growth during the te n -y ear period 1967-1977, in re s id e n tia l subdivisions. P latted subdivisions in the northern lower peninsula accounted fo r 99,862 lo ts or 37.9% o f the lo ts offered fo r sale in the state from 1967-1976 (F le tc h e r, 1979). This is a very high rate of development when r e la t iv e populations of areas in the s ta te are considered. There are no major metropolitan areas in northern Michigan and few major employers. Many of the lo ts are purchased fo r t h e i r recreational p o ten tia l and not fo r t h e i r physical productive c a p a b ilit ie s . Therefore, market exchange price may not t o t a l l y r e f l e c t i n t r i n s i c worth to an individ ual purchaser of the s i t e . Given the amount of subdivision development taking place in northern Michigan, there ex ists a p o te n tia l fo r problems to develop through un­ planned settlement p rac tic e s . Unplanned settlement practices can create problems f o r many aspects of society. The environmental, p o l i t i c a l , economic, social components of society may a l l be impacted. Economically the e ffe c ts o f past unplanned settlement practices are recently beginning to be f e l t in northern lower Michigan. Permanent residents received a b e n e fit from past settlement practices. Non-residents pay the same ra te o f property taxes as permanent residents, y e t because non-residents are only in the area fo r a portion of the ye ar, they do not demand as many services. Consequently, permanent residents, who do demand year-round services, are p a r t i a l l y subsidized by non-resident property taxpayers. This was the case fo r many years; however, with com­ p letio n o f major freeway routes and increased m o b ility of modern society in the e a rly 1970's home bu ilding began to occur. I n i t i a l l y seasonal home bu ild in g began to take place in many recrea­ tio n a l sub-divisions (F le tc h e r , 1979). The building o f seasonal homes meant th a t many county services (e .g . f i r e , ambulance service, e t c .) now 3 had to be provided to residents who were in the area on weekends or vacations. I t was not a serious problem i f the seasonal homes were located close to population centers, but many seasonal homes were scattered throughout the forested and lake areas, thereby increasing costs to counties having to provide services to ou tlying areas. Even greater problems can occur in rural areas through the rapid growth of permanent homes. During the e a rly 1970's i t was not uncommon fo r many people to move t h e i r fa m ilie s to northern Michigan and commute to work in major metropolitan areas. Commuting would take the form of v is it in g the fam ily on weekends; or f o r some in d iv id u a ls , consisted o f a four-hour round t r i p fo r work each day (conversation with Elizabeth B. Mowery, Extension S p e c ia lis t, Department of Resource Development, 1977). Counties have to provide year-lon g services to residents located in the outlying areas (e .g . snow remove!, school, bus s e rv ice , e t c . ) , causing in many cases net revenues from property taxes received to be negative (Johnson, 1973; American Society o f Planning O f f i c i a l s , 1976). Another economic problem impacts permanent residents. lots may sharply increase market values. be re fle c te d in higher assessments. Demand fo r Over time th is increase w i l l Higher property taxes w i l l r e s u lt. In some cases commodity producing lands ( e . g . , farms) may be sold because of higher property taxes increasing production costs over r e a liz e d return on investment. Environmental problems may r e s u lt from unplanned settlement practices. Stress on the natural resource base is severe when home development occurs. When homes are located on bodies o f water, seepage from septic tanks con­ t r ib u te to the eutrophication process. Erosion and s i l t a t i o n may be problems i f construction takes place on a g rad ien t. F ire danger can be increased i f development takes place in forested areas. Solid waste disposal becomes a problem as local dumps are not able to handle the waste increase or meet new regulations. W ild li f e h a b ita t destruction may be of concern e s p e c ia lly i f wetlands have to be f i l l e d . These are ju s t a few of the recognized environmental problems th a t re s u lt from unplanned settlement practices (American Society o f Planning O f f i c i a l s , 1976). In the short term, property tax receipts may be increased by allowing unregulated development, but in the long term, re s u ltin g en­ vironmental costs may o f f s e t previous tax gains. Social impacts increase because of unplanned settlement p ractices. T r a f f i c congestion, crowding a t public f a c i l i t i e s and c u ltu ra l shocks may occur. Concentrated development w i l l cause t r a f f i c congestion on many rural roads which were never intended to be major thoroughfares. Increased development w i l l cause crowding a t public f a c i l i t i e s (i.e . public access s it e s , parks) i f demand begins to exceed a v a ila b le supply. L ife s t y le and c u ltu ra l changes may occur i f new property owners come from urban and relocate in ru ra l areas. Rural tr a d itio n s may also be lo s t through resident displacement (American Society of Planning O f f i c i a l s , 1976). Opportunity fo r creation of new problems and m agnification o f old problems increase with any development. pound the problem. Unplanned settlements only com­ Problems concerning the environment, economic s tru c tu re , and social changes a l l require immediate a t te n tio n , but to be able to handle these problems adequate information must be a v a ila b le . Too often very l i t t l e is known about c it iz e n s ' development needs, pre­ ferences, or p r i o r i t i e s when developing solutions to problems (V la s in , Libby, Shelton, 1975). Even more basic, however, too often nothing is known about landowner c h a ra c te ris tic s from which assumptions can be formulated to ascertain p o te n tia l and exten t of fu tu re problems. region are property owners coming from? What Where are they locating? Which water bodies are most l i k e l y to be impacted by development? Once the answers to these questions are known, then e f f e c t iv e policy can be implemented to control some of the expected problems. At present, adequate information does not e x is t to formulate e f f e c t iv e po licy fo r northern Michigan landowners and unplanned settlement practices continue to create problems. A simple problem model schematic i l l u s t r a t e s the need fo r d e ta ile d inform ation. In Figure 1 the general problem, unplanned settlement p ra c tic e s , is l i s t e d . The problem is then disaggregated in to selected component parts ( e .g . s o c ia l, economic, e t c . ) . These components are f u r t h e r disaggregated in to t h e i r respective p arts. At th is stage, problems can be in d iv id u a lly selected and studied i f adequate informa­ tio n which re la te s to the selected problem s itu a tio n e x is ts . For example, a researcher studying the problem o f unplanned settlement practices may i n i t i a l l y e le c t to concentrate on the environmental impact components of the o v e ra ll problem. Environmental impact also has several component parts and a researcher may wish to fu rth e r define his study and concentrate on only one o f the components, possibly s o lid waste disposal. I f s u f f i c i e n t information e x is ts , a researcher can in v e s tig a te the extent of a s o lid waste disposal problem a ris in g from unplanned settlement p ra c tic e s . A lte rn a tiv e solutions can be developed and a course o f action recommended. However, i f information is lacking, then data must be c o llec ted and analyzed before a problem can be studied in depth. Collected and analyzed data allows fo r expanded knowledge and solution to some of the component problems a ris in g from the o verall problem of unplanned settlement p ra c tic e s . In summary, the problem solving model flow chart presented in Figure 1 id e n t if ie s the major problem which is then broken down in to i t s researchable component parts and, i f required, data are co llected and analyzed which feeds back into the o v erall problem component parts to allow fo r problem in vestigation and possible so lu tio n . I t was the in te n t o f th is study to develop information pertaining to the problem of unplanned settlement p rac tice s. Future research e f f o r t s should be able to proceed d i r e c t ly to the component parts of the unplanned settlement problem and develop recommended courses o f action to a ll e v ia t e present and fu tu re impacts. OBJECTIVES OF THE STUDY A. Primary Objective The primary o b je c tiv e of th is study was to c o lle c t and analyze socio-economic and selected other c h a ra c te ris tic s data o f landed property and home owners in three representative counties of Michigan's northern lower peninsula. B. S p ec ific Objectives Accomplishing the primary o b jec tive requires assimulation of numerous s p e c ific objectives in to a whole. The s p e c ific objectives o f th is study were: -explore socio-economic c h a ra c te ris tic s among property owners based on type of home development and location to c e rta in natural resources -examine use patterns among seasonal home owners ( e . g . , length of stay, seasons of high use, e t c . ) -examine factors in flu e n cin g i n i t i a l property purchase and fu ture intentions to s e ll property 7 Unplanned Settlement Practices Social Cultural Congestion Economi c Envi ronment Erosion Solid Waste Disposal Eutrophication Information C o llection And Analysis Figure I Problem Solving Model Increased Property Taxes Demand fo r A dditional Services 8 -examine acreage ownership patterns r e la t iv e to type o f home development and location to c e rta in natural resources - i d e n t i f y the magnitude of selected needs and concerns o f studied property owners -develop a d e s c rip tiv e model f o r value per acre o f land based on location r e l a t i v e to township o f property ownership and selected natural resources Once these ob jectives have been met, c e rta in questions w i l l have been answered. Questions such as who are the property owners, where do they l i v e , why did they buy property, can be answered. Policy makers w i l l be able to make use of the inform ation, as well as developers, consumers and researchers. Policy makers should be able to id e n t if y groups impacted by polciy changes. This study is intended to c l a r i f y present problems and uncover p o te n tia l problems. Policy makers should then be able to adjust accord­ in gly to meet community needs. Developers should be able to use the study to id e n t if y prospective c lie n te le . C h a ra c te ris tic s influencing demand should become apparent and developers should be able to se le c t locations fo r new developments which w i l l maximize b e n e fits . Consumers should be able to use study resu lts to id e n t ify some of the problems they may encounter i f they purchase property in the study area. Consumers should also be able to id e n t if y areas where home b u ild in g is expected to be high and adjust t h e i r purchase decisions accordingly. In addition to the above groups, there are probably additional groups which w i l l b e n e fit from th is study. A major problem in any 9 decision, whether i t be a t the governmental level or in d ivid u a l le v e l, is lack of relevant inform ation. This study is intended to f i l l in many o f the informational gaps in decisions involving a selected northern Michigan study area. i CHAPTER I I RESEARCH METHODS Defining the Study Area The study area fo r th is research centers.on three counties in Michigan's northern lower peninsula: (See Figure 1 ) . Kalkaska, Otsego, and Crawford Care was taken to se le c t counties which are s im ila r in many important respects. A ll three counties are s im ila r in s iz e , resource base, amount of tr a v e l time from major metropolitan areas, and absence o f any contiguous Great Lake. Another reason fo r choosing these three counties is th a t during the la s t f iv e years they have a l l shown ra th e r large increases in population. According to figures compiled by the Sociology Department at Michigan State U n iv e rs ity , a new migration gains fo r Otsego county have been 23.9% of the 1970 population. Crawford and Kalkaska have experienced s im ila r large gains, 22% and 46.5%, resp ectively (O'Hare, e t a l , 1976). There has been q u ite a b i t of speculation as to the reasons fo r these large population increases, the two most prominent being completion of the 1-75 expressway to the north and the conversion from seasonal residents to permanent residents through retirem ent. No matter how much care is taken in s e lec tio n of counties, there w ill s t i l l be a good deal of variance between counties. D etailed in ­ formation concerning county c h a ra c te ris tic s follows to enable readers u n fa m ilia r with the study area, an opportunity to become informed; Physical c h a ra c te ris tic s of each county are q u ite s im ila r . Kalkaska has 566 square miles o f land area whereas Crawford and Otsego have 561 and 527 square miles re s p e c tiv e ly . 10 Geological features are also s im ila r . Bedrock was formed during the Paleozoic period and consists p r im a rily o f Berea shale and sandstone with some Ellsworth shale in the lower portions o f Crawford and Kalkaska Counties. In the northern tw o -th ird s o f Crawford and Kalkaska counties, Antrim shale is predominant with some intrusions o f Ellsworth shale. Otsego county bedrock is predominantly Antrim shale in the southern two-thirds area with Berea shale and Sandstone in the north. Shale in Northern Michigan is o i l r ic h , however, i t is presently d i f f i c u l t and expensive to separate the o i l from the shale. The p o te n tia l fo r fu tu re natural resource wealth remains high. C u rre n tly , there are some o i l and gas pools being tapped in Kalkaska and Otsego counties. o il a c tiv ity . Kalkaska is the center of norther lower Michigan's In terms o f statewide county t o t a l s , Kalkaska ranked fourth in the b a rrels o f o i l produced and second in cubic fe e t of natural gas. Otsego county ranked second in the state in the barrels o f o i l produced and fourth in cubic f e e t of natural gas. Crawford county having much less r e a d ily a v a ila b le o i l and gas pools ranks seventh in the s ta te in number of b a rrels o f o i l produced and f if t e e n t h in cubic fe e t o f gas. tio n . Kalkaska county also has two natural gas plants in opera­ One is owned by Amoco and the other by S h e ll. These plants co ntribute to employment opportunities w ith in the area. Importance of o i l and gas also shows up in other sectors of the economy. In 1977 construction a c t i v i t y accounted fo r 19.89% o f to ta l wages and p rop rieto rs earnings in Kalkaska county. is the highest f o r any county in the s ta te . This percentage Construction contributes 6.71% o f to ta l wages and p ro p rie to r earnings (23rd in the s ta te ) in Otsego county and 4.07% (48th in the s ta te ) in Crawford county. Location of natural gas plants in Kalkaska and subsequent construction associated 12 with them has contributed to th is higher percentage of earnings. A g ricu ltu re is o f low p r i o r i t y in a l l three counties. In terms' o f to ta l percentage of wages and p ro p rie to r earnings (1 9 7 7 ), a g ric u ltu re accounted fo r 1.13% in Kalkaska county (59th in the s t a t e ) , 1.03% in Otsego county (60th in the s t a t e ) , and 0% in Crawford county (82nd statew ide). R elative importance o f a g ric u ltu r e can be a ttr ib u te d to three factors: 1) s o i l , 2) clim a te, 3) amount o f public land ownership. The soils which were formed during the geological period consist p rim a rily o f l ig h t to moderately sandy s o ils with loamy and swamp so ils found along main watercourses. Surface gravel is present throughout much o f the area and moderate to steep slopes are common. F e r t i l i t y of the s o il is low in many areas with extensive areas o f Jack Pine present throughout. R a in fa ll is p l e n t ifu l fo r crop production averaging around 32-33 inches per ye ar, however, the growing season is short and susceptible to la te or e a rly fr o s ts . Average temperature in July is 67° and in January only 18°. Public ownership o f land reduces the amount a v a ila b le f o r a g r ic u l­ tu ra l uses. In northern Michigan there ex is ts extensive tra c ts o f land in public ownership. state ownership. Kalkaska county has 42.41% o f i t s land tie d up in State ownership in Otsego county accounts fo r 27.64% of the land and in Crawford county 66.73% is in state and federal owner­ ship. Additional land in a l l three counties is controlled by c i t i e s , townships, and school d i s t r i c t s . Given the three constraints on a g r ic u ltu r a l production i t is not surprising th a t toal wages and p ro p rie to r earnings a ttrib u te d to a g ric u ltu re are so low in the study area. This is not expected to change in the fu tu re. Tourism is o f great importance to each county's economy. One in dication o f the importance of tourism shows up in percentage of to ta l wages and p ro p rie to r earnings a t tr ib u t e d to regional trade. Kalkaska county residents receive 11.08% (36th statewide) of t h e i r to ta l wages and p ro p rie to r earnings from r e t a i l trade whereas Otsego and Crawford county residents receive 12.07% (25th statewide) and 16.5% (6th statewide) resp e c tiv e ly . One reason t o ta l wages and p ro p rie to r's earnings a ttr ib u te d to regional trade f o r Kalkaska county are low r e la t iv e to Otsego and Crawford counties is th a t i t is also fu rth e s t away from In te r s ta te highway 75. When comparing statewide rankings i t is somewhat surprising th a t the study area counties are in the top 50 p e rc e n tile . The 1980 census showed that in terms of statewide population ranking, Otsego county was 64th, Crawford county 75th, and Kalkaska county 70th. Therefore, high statewide rankings in percentage of to ta l wages and p ro p rie to r earnings indicate 1) residents of Kalkaska, Otsego and Crawford counties are big spenders a t local r e t a i l o u tle ts or 2) a great amount of sales go to out-county residents. The second reason is most l i k e l y as a l l three counties receive a substantial number o f v is it o r s . In a d d itio n , Crawford county receives a large number o f national guard troops who stay at Camp Grayling throughout the summer. Another tourism in d ic a to r is the number o f recreation al sub­ divisions developed w ith in the tr i-c o u n ty study area. Visual inspection of p la t maps and on s it e inspections showed a considerable number of recreational subdivisions. In Otsego county Chain o f Lakes, AuSable Estates P la t , Otsego Lake Plats are some of the la rg e r sub­ divisions located in Bagley township. Blue Lake township of Kalkaska county has a recreation al subdivision around most every lake. In 14 Crawford county there are many subdivisions along the Southwestern portion o f the AuSable r i v e r and around Lake Margrethe. Ski resorts also a t t r a c t to u r is ts . Crawford county has three developed ski areas and Otsego county fo ur. A to ta l of 24 ropes, 7 tows, and 8 c h a ir l i f t s are in operation among the developments. Glaciers which formed th is area l e f t a g la c ia l lobe contact l i n e , running through the area, with moderate to steep slopes and many picturesque h i l l s well suited fo r sk iin g . Other a ttra c tio n s f o r to u r is ts include the e x c e lle n t tro u t fis h in g , es p ec ially on the AuSable r iv e r system which is rated a blue ribbon stream, and hunting. The large amount of public land assures p le n t if u l access to f ie ld s and streams. Also, local areas have taken strid es to e s tab lish unique i d e n t i t i e s . Gaylord's business d i s t r i c t has taken on an a lp in e atmosphere in what appears to be a successful attempt to a t t r a c t to u r is ts . The c it y o f Kalkaska has an annual tro u t f e s t i v a l , coinciding with the opening o f tr o u t season, complete with parades and a tr o u t queen. Forestry is another important component of the tr i-c o u n ty economy. Kalkaska c it y is an important paperboard producing center. During the 1970's Otsego county p a rtic ip a te d in the expansion o f fo re s t product in d u s tries. The fu tu re f o r fo re s t industries expansion, however, is tie d to s ta te or federal p o lic ie s concerning u t i l i z a t i o n o f public land. The economic fu tu re and population growth p o ten tia l fo r the t r i county study area is paradoxical. Rising costs fo r o i l and gas insure th a t present r e fin e r ie s w i l l continue to be important components of the economy. Future expansion and growth in energy in d u s tries can 15 be expected. On the other hand, ris in g p rice o f gasoline may c u r t a il tourism tra v e l and have a dampening e f f e c t on the local econoniy. Energy costs f o r heating may also change retirem ent plans f o r many in d iv id u a ls . Living in southern climates where heating and liv in g costs are low may become more a t t r a c t iv e fo r people approaching r e t i r e ­ ment age. Speculation about fu ture growth is ris k y even when adequate information exists about present conditions. r i s k i e r when l i t t l e Speculation becomes information about the status quo is a v a ila b le . Sampling Strategy Recall th a t the study's primary o b je c tiv e was to develop an information base. Who are the northern Michigan property owners? In accomplishing th is o b je c tiv e , the groundwork was la id f o r fu tu re studies which can address such things as causal factors underlying property ownership and migration pattern s. In developing the i n f o r ­ mation base, key variables were id e n t if ie d so th a t fu tu re researchers and area planners can proceed with known s t a t i s t i c a l c h a ra c te ris tic s when developing research projects and p o lic y . study's goal was to help f i l l In essence then, the the information void in which researchers and planners are so often forced to operate. General and socioeconomic c h a ra c te ris tic s of real property and home owners in the study area were obtained via primary data c o lle c tio n - s p e c i f ic a lly a mail survey. A questionnaire was developed and sent to a sample o f property owners, having land zoned r e s id e n t ia l, in three townships o f each county. s t r a t i f i e d random sample technique. Respondents were selected using a This involved d iv id in g the popula­ tio n in to groups and a sample was then randomly drawn f o r each s t r a t a . 16 EIVUI jSCHOotciurr ( DCLTA emmct |MONT«OJti o c e o o * f A te o > u M H Z l t j f i O ' TRMV. OTSEGO J ifM u w u i *o « c o # n o « e « *« ***** LA«t O fC tO L M F c L A 0 g i i« « o 'a iA O H I * 7 AMKHAC MXCAMA (M W A V tO J KALKASKA CRAWFORD • NATIOTI **•*»*• r ■. «*"T i u m , « » 'o » M i a !««■*«»*•' CALHOUN , C A IN , | in m a n '« < * » r*AAariNA»i »**«« J IM J M U L IN A V C f !lA T .JO O tl’ H*.IN A N C H 1 I Figure 2 Location of Surveyed Counties n , “ ONNOf 17 The township selected f o r the random sample were chosen because of t h e i r resource c h a r a c te r is tic s . That i s , each township selected was determined to be s u f f i c i e n t l y homogenous f o r one resource c h a r a c te r is tic , hence, the townships are the s t ra ta . In each county one of the townships selected included p rim a rily re s id e n tia l properties around lake areas. A second township was p rim a rily a r iv e r resource based area, and the th ir d township was an area th a t has no major lake or r i v e r in the study area. township may include areas w ith in c i t y l i m i t s . The th ir d Townships were selected through visual examination o f p l a t books f o r each county. The most representative township, fo r each natural resource c h a r a c t e r is tic , was then chosen. In Kalkaska county the townships surveyed were G a rfie ld T.25N.-R.7W, selected because of the influence o f the Manistee River System; Blue Lake T.28N.-R.5W because of the influence o f many small lakes which have subdivisions surrounding them; and Orange T.26N.-R.7W because of the r e la t iv e absence of any water resource base. In Otsego county the townships surveyed were Dover T.31N.-R.2W because of the influence of Otsego Lake and many other assorted small lakes; Bagley T.30N.-R.3W because of the influence o f the AuSable River; and Chester T.29N.-R.2W because o f the influence of large tr a c ts o f state land bordering the water systems and public holdings located away from most of the water resource base. In Crawford county the townships surveyed were South Branch T.25N.-R.2VJ because o f the influence of the AuSable River System; Grayling T.26N.-R.4W because of the influence of Lake Margrethe; and Grayling T.27N.-R.2W because o f the absence of any water resource base. The location o f each township w ith in i t ' s in Figure 3. respective county is shown 18 DOVER T.3I N. R.2 W. BAGLEY T.30N. R.3W. GAYLORD CHESTER T.29N. R.2W. BLUE LAKE T.28N. R.5 W. GRAYLING T.27 N. R.2 W. □ KALKASKA ORANGE T.26N. R.7W. GRAYLING T.26N. R.4W. GARFIELD T.25N. R.7W. SOUTH BRANCH T. 25 N. R2 W. KALKASKA CRAWFORD Figure 3 Location of Surveyed Townships in Their Respective Counties 19 The next problem is how can a representative sample be drawn from each township? I f samples of equal size are drawn from the townships, then population c h a r a c te r is tic estimates obtained are more va ria b le than samples which are drawn proportional to size of the township population. Samples of equal size would allow the researcher to make more precise comparisons between townships but estimates of population ch a ra c te ris tic s in the counties would be less precise. Because more importance was placed on county c h a ra c te ris tic s rath er than comparisons between townships, a sample proportional to each township size was drawn. The mean of each county population was then computed as the mean of each township weighted by i t s size. The next question th a t has to be answered concerns the optimal size (in terms of minimizing cost with respect to sampling e rro r) of the sample in each township. One formula (Ackoff, 1962) th a t has been used to id e n t if y optimal sample size is: nh o Nh Sh N h Ch Where nhQ = Optimal sample size fo r township h Nh = Population in township h Sh = Variance in township h N = Population to ta l fo r a l l townships Ch = The cost per observation in h township h = A cost constant ( a r b i t r a r i l y derived) There is one unknown in the r ig h t hand side of the equation (Sh), which had to be determined before optimal sample size could be ascertained. 20 The only way to determine variance, without a previous sample or a census is by the best guess method. Compounding th is problem is th a t many variables (as w i l l be seen la t e r ) had to be d e a lt with and each va ria b le has i t s own unique variance. I t was decided to estimate the variance on the basis o f information from a p retest o f the questionnaire. A sample of 25 observations were selected fo r a p r e -te s t in Rapid River Township o f Kalkaska county. This township has elements o f a l l the resource c h a ra c te ris tic s which id e n t if y each s t r a t a . Answers received from Rapid River were analyzed and the variance of each v a ria b le d e ter­ mined. The highest variance obtained was used in the above formula to determine the minimum sample size fo r each township. This is s t i l l ju s t a rough approximation o f population variance because the p r e - te s t sample was small. However, because o f the lack of any previous research analogous to th is study and study area from which to draw a variance estim ate, the p r e -te s t variance approach was deemed the most appropriate. Once sample size was determined, the questionnaire was sent out to the appropriate number o f randomly selected property owners in each township. An i n i t i a l m ailing of 3,371 questionnaires were sent out in August, 1978. The f i r s t m ailing consisted o f a questionnaire and a postage paid return envelope. Postage f o r the f i r s t mailing was bulk rate re s u ltin g in a savings o f over eleven cents per piece compared to f i r s t class. The only disadvantages were th a t a l l pieces had to be sorted by zip code, which took some tim e, and undeliverable pieces were not returned to sender but instead were discarded. A follow-up mailing was conducted in September, 1978, and consisted of a questionnaire, a postage paid return envelope, and a reminder l e t t e r . This m ailing was sent out f i r s t class so th a t the number o f undeliverable questionnaires could be ascertained. A t h ir d m ailing consisting o f a post card reminder 21 was sent out in October, 1978. Postage on the post card was bulk rate which saved about seven cents per piece compared to f i r s t class. A fo u rth , and f in a l m ailin g , was sent out to non-respondents in November, 1978. I t consisted o f a questionnaire, a postage paid return envelope, and a reminder to send the questionnaire back. Postage on the fourth mailing was bulk rated and saved over eleven cents per piece compared to f i r s t class. A to ta l of 2,430 (72.1%) of the i n i t i a l 3,371 questionnaires sent out were returned. A t o ta l o f 2,006 usable responses were received, 252 undeliverables (moved, l e f t no forwarding address), and 172 responses in which the property owner had sold his property during the time mailing addresses were obtained and questionnaires sent out. A complete break­ down of questionnaires sent out to each area, and response rates is outlined in Table 1. Not a l l usable responses were complete fo r a l l questions in the survey, th e re fo re , t o ta l sample response may vary fo r each v a ria b le . In reviewing Table 1, the low return ra te due to the postcard reminder brings up serious questions as to i t s usefullness. The amount of time and money spent p r in tin g , addressing, and sorting by zipcode does not seem j u s t i f i e d in li g h t of the generally poor response r a te . Rather, i t is recommended t h a t, in fu ture surveys, of th is nature, th a t only three mailings be undertaken and the postcard reminder be elim in ated. One in te r e s tin g r e s u lt is th a t the second m ailing received almost as many responses as the f i r s t . The second mailing achieved th is response rate even though almost 25 percent less questionnaires were sent out than the f i r s t m ailin g. Therefore, marginal p ro d u c tiv ity from the second mailing is much g rea ter than th a t from the f i r s t m ailin g. three d i f f e r e n t reasons. This may be due to The second m ailing was f i r s t class and, q u ite Table 1 QUESTIONNAIRES HAILED AND ACCOUNTED FOR, BY TOWNSHIP, COUNTY A u g u s t, 1978 1 s t H a ilin g (B u lk ) ■ C raw fo rd November, 1978 4 th H a ilin g (B u lk ) S u b -T o ta l Number Number 6 6 .5 21 10.0 - 160 76.5 34 82.9 2 4 .9 - 36 87 .8 3.2 110 59.5 11 5.9 - 121 65.4 27 13.9 116 59 .8 26 13.4 - 142 73.2 16 3 .3 13 2 .7 280 57.7 32 6 .6 - 312 64 .3 4 2 .7 11 7 .3 83 55 .3 6 4 .0 - 89 59 .3 2 4 .7 95 5 .3 96 5 .4 1,059 59 .4 135 7.6 - 1,194 67 .0 33 20 .6 5 3.1 0 0 101 63.1 7 4 .4 - 108 67 .5 2 0 .6 33 2 0 .0 6 3 .6 11 6 .7 84 50.9 12 7 .3 - 96 58.2 25 .4 792 23 .5 148 4.4 208 6 .2 2,006 59 .5 252 7 .5 172 2,430 72.1 R eturned R eturned Returned Returned' 2 6 .3 49 23 .4 8 3 .8 27 12.9 139 9 21 .9 6 14.6 2 4 .9 17 41.5 185 56 30.3 44 2 3 .8 4 2.2 6 Orange 194 50 2 5 .8 31 16.0 8 4.1 B lue Lake 485 132 27 .2 119 24 .5 G a r fie ld 150 31 20 .7 37 24 .7 1,782 428 2 4 .0 440 C hester 160 63 39 .4 Dover 165 34 TOTAL 3,371 858 R eturned 209 55 G ra y lin g T27NR2W 41 G ra y lin g T26NR4W Bag le y T o ta l S old N o n -D e liv e ra b le P e rce n t R eturned o f T o ta l P erce nt R eturned o f T o ta l South Branch Kalkaska O c to b e r, 1978 3rd M a ilin g (B u lk P o s tc a rd ) P e rce n t R eturned o f T o ta l P e rce n t R eturned o f T o ta l M a ile d Otsego Septem ber, 1978 2nd H a ilin g ( 1 s t C la s s ) P e rce n t R eturned o f T o ta l P erce nt R eturned P e rce n t o f T o ta l 5.1 T o ta l Q u e s tio n n a ire s Accounted f o r % o f T o ta l M a ile d Q u e s tio n n a ire s Accounted f o r 23 possibly, people are more apt to read t h e i r mail i f i t c a rrie s a f i r s t class stamp rath er than a bulk stamped im print which to many people indicates junk m ail. Also, the second m ailing contained a reminder l e t t e r which may have prompted some people, who otherwise might not have answered, to take the time and complete the questionnaire. F in a lly , the f i r s t mailing took place in August, generally the busiest vacation month, and the second m ailing took place in September a f t e r Labor Day when people were more apt to be home. In summing the resu lts o f the survey response r a t e , i t seems th at the approach taken in questionnaire design and m ailing stra teg y provided a high return at low cost. The only change recommended f o r fu tu re research employing a mail survey is to consider e lim in a tin g the postcard reminder as i t does not seem to j u s t i f y i t s cost. The questionnaire was formulated from questions deemed p e rtin e n t to the research and includes input from advisors (W illia m Kimball and Manfred Thullen) in the area of survey design. S tructure of the question­ n aire is such th at i t could be divided in to four p a rts ; questions th a t deal exclu sively with people who own landed property with no home develop­ ment in the study area, those who own permanent homes in the area, those who own seasonal homes in the area, and questions concerning a l l landed property owners in the area, regardless o f type o f home development. This allowed fo r cross examination between d i f f e r e n t landowner factions uncovering s im ila r it ie s o f differen ces e x is tin g among property owners. When the questionnaires were returned, they were coded and analyzed using the S t a t i s t i c a l Package fo r the Social Sciences (SPSS) set of computer programs. 24 Measurement and V ariable Id e n t if ic a t io n The type o f analysis th a t can be performed on the variables is lim ite d by the level o f measurement each va ria b le lends i t s e l f to. There are four basic levels of measurement as o u tlin ed by S.S. Stevens: nominal, o r d in a l, in t e r v a l , and r a t io (Stevens, 1976). Nominal is the lowest measurement level and is b a s ic a lly a form o f la b e lin g . There are no assumptions made about the value being assigned to the data, hence the value serves as a la b e l. The values are used as symbols th a t can be e a s ily read by a computer. Ordinal level of measurement is used when a v a ria b le can be rank ordered. That i s , each category has a unique position r e l a t i v e to other categories. When a value is assigned to an ordinal level v a r ia b le , we know th a t value is higher or lower than other values. For example, a p o llu tio n level th a t is ranked high, medium, or low with values 1 f o r high, 2 fo r medium, and 3 fo r low. In te rv a l level measurement has the property th a t the lev el of measurement between categories has a d e f in it e i n t e r v a l . the units are fixed and equal. That i s , (An exception to th is is the special case o f a logarithm ic in te r v a l scale which was not encountered in th is research). However, although th is allows us to study the d iffe re n c e between things, i t does not allow us to study proportionate magnitudes. This is because an in te rv a l measurement scale does not have a true zero. R a tio -le v e l is the highest level of measurement in Steven's typology, and i t has a l l the a ttr ib u te s o f in te r v a l measurement plus a tru e zero p o in t. That is proportionate magnitudes can be studies. For example, s ix pounds is twice as heavy as three pounds. The reason a great deal o f emphasis is placed on lev el o f measure­ ment is because s t a t i s t i c s used to describe va ria b le s require s p e c ific 25 levels of measurement. S t a t is t ic s th a t require a c e rta in l e v e l, such as an in te rv a l scale, cannot be used fo r ordinal or nominal levels o f measure­ ments. However, i f a s t a t i s t i c requires a nominal level o f measurement, i t can be used with any other measurement scale. Therefore, s t a t is t ic s developed fo r a ce rtain level of measurement may be used with higher measurement scales but not with lower ones. However, a loss of s t a t i s t i c a l power results when s t a t i s t i c s designed fo r a lower level of measurement are used to describe relation ship s among variables which are measured at higher le v e ls . I t seems appropriate, at th is tim e, to introduce a l l the variables contained in the questionnaire. Table 2 l i s t s each va ria b le name, i t s appropriate id e n tify in g number, measurement sc ale , and a b r i e f descrip­ tio n of the v a ria b le . (For a complete d e s c rip tio n , the reader should r e f e r to Appendix A where the questionnaire has been reproduced). Each v a ria b le name appears as i t does on the computer program. The addition of an id e n tify in g number is fo r ease in locating any one of the variables from the questionnaire. Although the id e n tify in g number w i l l not appear on a computer p r in to u t, the reader may wish to use i t to r e fe r back to the questionnaire to determine which v a ria b le is associated with each question. The upper case l e t t e r , i . e . A, refe rs to section A on the questionnaire. The number, i . e . 5, refers to question 5 o f section A, and the small case l e t t e r , i . e . b, refers to a c e rta in section o f the s p e c ific question ( in th is example, question 5 ) . Although the questionnaire does not, in many cases, e x p l i c i t l y include subsections o f c e rta in questions, the reader by knowing the id e n tify in g number and v a ria b le name, can pick out the portion of the question re fe rre d to . For example, v a ria b le name YR1 id e n tify in g number A5b refers to section Aa question 5, and subsection b: "the f i r s t year in which some property was obtained or sold a f t e r 26 the i n i t i a l piece of property was obtained." Table 2 VARIABLE NAME, IDENTIFYING NUMBER, MEASUREMENT SCALE, AND VARIABLE EXPLANATION FOR ALL VARIABLES EXTRACTED FROM THE QUESTIONNAIRE Id e n tify in g Number Measurement Scale TOWNSHIP Aa Nominal Township where property is located COUNTY Ab Nominal County where property is located ACQUIRE A1 Nominal Method o f i n i t i a l pro­ perty ac q u is itio n REASON A2 Nominal Main reason fo r property a c q u isitio n LEARN A3 Nominal Information source leading to i n i t i a l property ac q u is itio n YROBROP A4 In te rv a l Year in which f i r s t piece of property was obtained OBTSOLD A5a In te rv a l Since f i r s t property a c q u is itio n , has any adjacent property been purchased or sold? YR1 A5b In te rv a l Year in which any addi­ tio n a l property transaction was made OBTSOLD! A5c Nominal Obtained or sold the pro­ perty in question ACRES! A5d Ratio Total acres involved in above property transaction YR2 A5e In te rv a l Year in which an ad ditional property transaction was made OBTSOLD2 A5f Nominal Obtained or sold the pro­ perty in question V ariab le V ariab le Description 27 Table 2 ( C o n t'd .) V a r ia b le Id e n tify in g Number Measurement Scale V a r ia b le D e s c rip tio n ACRES2 A5g Ratio Total acres involved in above property transaction YR3 A5h In te rv a l Year in which an additional property transaction was made 0BTS0LD3 A5i Nominal Obtained or sold the pro­ perty in question ACRES3 A5j Ratio Total acres involved in above property transaction YR4 A5k In te rv a l Year in which an additional property transaction was made 0BTS0LD4 A51 Nominal Obtained or sold the pro­ perty in question ACRES4 A5m Ratio Total acres involved in above property transaction YR5 A5n In te rv a l Year in which an additional property transaction was made 0BTS0LD5 A5o Nominal Obtained or sold the pro­ perty in question ACRES5 A5p Ratio Total acres involved in above property transaction TOTACRES A6 Ratio Total acres c u rre n tly owned or leased in the study area SELL A7a Ordinal In te n t to s e ll a l l or p a rt of property in fu tu re SELLACRE A7b Ratio Number o f acres wishing to s e ll YRSELL A7c Ordinal Number of years before d esiring to s e ll TOTVALUE A8 Ratio Total value of property owned (includes any dwelling) 28 Table 2 ( C o n t 'd .) V a r ia b le Id e n t if y in g Number Measurement Scale V a r ia b le D e s c rip tio n PROPTAX B1 Ordinal A ttitu d e towards current property tax lev els QUALSERV B2 Ordinal A ttitu d e towards q u a lit y of municipal or county services provided QUANSERV B3 Ordinal A ttitu d e towards qu an tity o f municipal or county services provided ZONING B4 Ordinal Awareness o f building regula­ tions LANDREG B5 Ordinal A ttitu d e towards present land use controls BUILDING B6 Ordinal A ttitu d e towards fu ture r e s id e n tia l building PROPVALU B7 Ordinal A ttitu d e towards fu ture o f property values SKIAREA Cla Nominal Property located close to a ski area. Yes-No SKIMILES Cl b Rati o Miles from nearest ski area PUBPROP C2 Nominal Public land adjacent to respondent's property. Yes-No 0NH20 C3a Nominal Property lo cation to water TYPEH20 C3b Nominal Type o f water property is located on SIZEH20 C3c Ordinal Size o f lake property is located on CL0SEH20 C4a Nominal Type o f water closest to property LAKEMILE C4b Ratio Miles from nearest lake RIVMILE C4c Ratio Miles from nearest r i v e r H20SYSTM D1 Nominal Type o f water system in liv in g quarters 29 Table 2 ( C o n t'd .) Measurement Scale V a r ia b le Id e n t if y in g Number SEWAGSYS D2 Nominal Type o f sewage system in liv in g quarters HOUSTYPE D3a Nominal Are liv in g quarters conven­ tio n a l housing or mobile home? MOBILMOV D3b Nominal Can mobile home be moved or is i t anchored? Yes-No ANNUVIS El Ratio Annual v i s i t s made to the seasonal home DAYSTAY E2 Ratio Average length o f stay fo r each seasonal home v i s i t . Days FALL E3a Ratio Number o f days v i s i t i n g seasonal home in f a l l WINTER E3b Ratio Number of days v i s i t i n g seasonal home in w in ter SPRING E3c Ratio Number of days v i s i t i n g seasonal home in spring SUMMER E3d Ratio Number o f days v i s i t i n g seasonal home in summer YRHOME FI In te rv a l Year in which the permanent home, in the study area, was b u ilt SEASHOME F2a Nominal P rio r usage of permanent home as a seasonal home. Yes-No YRPERM F2b In te rv a l Year conversion from sea­ sonal to permanent home took place CTYBEFOR F3a Nominal County o f residence before moving to the study area STBEFOR F3b Nominal State o f residence before moving to the study area CTYNOW Gla Nominal County o f present residence V a r ia b le D e s c rip tio n 30 Table 2 ( C o n t 'd .) Variable Id e n tify in g Number Measurement Scale V ariable Description STNOW Gib Nominal State o f present residence AGE G2a Ratio Age o f household head SEX G2b Nominal Gender of household head MARITAL G2c Nominal M a rita l status of house­ hold head UNDER5 G3a Ratio Number o f people under 5 years old residing with household head AGE5-14 G3b Ratio Number o f people between ages 5-14 residing with household head AGE15-25 G3c Ratio Number of people between ages 15-25 residing with household head AGE26-64 G3d Ratio Number of people between ages 26-64 residing with household head 0VER65 G3e Ratio Number of people over 65 years old residing with household head TOTAL G3f Ratio Total number o f people residing with household head INCOME G4 Ordinal Total fa m ily income in d o lla rs *H0ME Ordinal Type o f home development on property *VALUACRE Rati o Value per acre fo r each respondent (includes dw ellir in d o lla rs *H0ME and VALUACRE are two key variables derived from other v a ria b le s. NOTE: For a complete description of each v a r ia b le , the reader should r e fe r to Appendix A. To compute variables HOME and VALUACRE, some restru ctu rin g o f the data was required. The RECODE c a p a b ility o f SPSS was employed to tra n s ­ form variables ANNUVIS and YRHOME in to one v a r ia b le , HOME. I f any res­ ponse was recorded fo r ANNUVIS, then i t means there is a seasonal home; and i f any response is recorded fo r YRHOME, i t means there is a permanent home on the property. I f both va riables are l e f t blank, i t was recorded as no home of any type on the property. - VALUACRE is also a recoded v a ria b le derived from variables TOTVALUE and TOTACRES. The to ta l value o f property owned per respondent is divided by the to ta l acres owned per respondent. The r e s u lt is value per acre o f land expressed in 1978 d o lla rs fo r each respondent. For a d e tailed explanation of the recoding procedures used, r e fe r to the SPSS manual, Chapter 8 (N ie, e t . a l , 1975). SPSS Subprograms Used fo r Data Analysis A fte r id e n tify in g each v a ria b le contained or constructed from the questionnaire, the next step was to analyze each v a ria b le independently and then combine variables and form sets of re la tio n s h ip s . handled through the various This was subprograms a v a ila b le in SPSS. Once analyzed, through appropriate subprograms, various key variables were selected fo r presentation in the data analysis chapters. Those presented, however, are the ones containing the most useful information concerning northern Michigan study area property owners. The f i r s t step in the analysis was to describe the study area in terms of gross responses. This was accomplished through SPSS subprograms FREQUENCIES and CONDESCRIPTIVE. FREQUENCIES is appropriate f o r variables measured a t nominal or ordinal lev els and CONDESCRIPTIVE requires a t le a s t an in te rv a l level of measurement. 32, The purpose o f the FREQUENCIES program was to be able to explain in terms of simple numbers and percentages how many respondents f i t in to each category, i . e . number and percentage o f responses from each county and township, number and percentage o f permanent residents, seasonal home owners, and landed property owners with no home development. The program also allows fo r missing and non-applicable data to be analyzed and recorded. There are various s t a t is t ic s th a t can be computed fo r FREQUENCIES. Mean, standard e r r o r , median, mode, standard d e v ia tio n , variance, ku rto s is , skewness, ranges, minimum and maximum are a l l a v a i l ­ able, however, not a l l the s t a t i s t i c s were employed fo r each v a ria b le . I t would not be useful or s t a t i s t i c a l l y v a lid to compute the a rith m e tic mean fo r a nominal level and in most cases, ordinal lev el v a ria b le s . Table 3 shows the variables th a t were analyzed under subprogram FREQUENCIES and CROSSTABS and the s t a t i s t i c s used to describe the v a ria b le s . A fte r in v e s tig a tio n o f in dividual variables was fin is h e d , r e la t io n ­ ships among sets of variables was explored. SPSS subprograms employed were CROSSTABS, BREAKDOWN, NONPAR CORR, and PEARSON CORR. CROSSTABS is appropriate when both variables are of e it h e r nominal or ordinal measure­ ment. C o n trollin g fo r a th ir d v a ria b le with subprogram CROSSTABS is possible provided the th ir d va ria b le is also o f nominal or ordinal measurement. C ontrollin g fo r more than one v a ria b le is also possible with CROSSTABS but care should be taken because the numerous tables and s t a t i s t i c s th a t re s u lt can e a s ily lead to confusion. Subprogram BREAKDOWN is used when there is one nominal or ordinal level v a ria b le and one in te rv a l level v a ria b le . Also, when c o n tro llin g fo r more than one v a r ia b le , BREAKDOWN is the appropriate procedure to use as output is displayed in an easy to read and in t e r p r e t ta b le . 33 Table 3 VARIABLE NAME, IDENTIFYING NUMBER, AND STATISTICS USED TO DESCRIBE THE VARIABLE UTILIZING SUBPROGRAM FREQUENCIES OR CONDESCRIPTIVE Subprogram V ariable Id e n tify in g Number S t a t i s t i c a l Output FREQUENCIES TOWNSHIP Aa Number o f Occurences FREQUENCIES COUNTY Ab Number of Occurences FREQUENCIES ACQUIRE A1 Number of Occurences FREQUENCIES REASON A2 Number o f Occurences FREQUENCIES LEARN A3 Number of Occurences CONDESCRIPTIVE TOTACRES A6 Mean, Median,Mode, Kurtosis, Skewness FREQUENCIES SELL A7b Number of Occurences FREQUENCIES YRSELL A7c Number of Occurences CONDESCRIPTIVE TOTVALUE A8 Mean, Median, Mode, Kurtosis, Skewness FREQUENCIES PROPTAX B1 Number of Occurences FREQUENCIES QUALSERV B2 Number of Occurences FREQUENCIES QUANSERV B3 Number o f Occurences FREQUENCIES ZONING B4 Number of Occurences FREQUENCIES LANDREG B5 Number of Occurences FREQUENCIES BUILDING B6 Number of Occurences FREQUENCIES PROPVALU B7 Number of Occurences FREQUENCIES PUBPROP C2 Number of Occurences FREQUENCIES 0NH20 C3a Number of Occurences CONDESCRIPTIVE ANNUVIS El Mean, Median, Mode, K urtosis, Skewness 34 Table 3 ( C o n t 'd .) Subprogram V ariab le Id e n tify in g Number CONDESCRIPTIVE DAYSTAY E2 Mean, Median, Mode, Kurtosis, Skewness CONDESCRIPTIVE FALL E3a Mean, Median, Mode, Kurtosis, Skewness CONDESCRIPTIVE WINTER E3b Mean, Median, Mode, Kurtosis, Skewness CONDESCRIPTIVE SPRING E3c Mean, Median, Mode, K urtosis, Skewness CONDESCRIPTIVE SUMMER E3d Mean, Median, Mode, Kurtosis, Skewness FREQUENCIES CTYBEFOR F3a Number o f Occurences FREQUENCIES CTYNOW G1 a Number o f Occurences CONDESCRIPTIVE AGE G2a Mean, Median, Mode, K urtosis, Skewness FREQUENCIES SEX G2b Number of Occurences FREQUENCIES MARITAL G2c Number o f Occurences CONDESCRIPTIVE TOTAL G3f Mean, Median, Mode, Kurtosis, Skewness FREQUENCIES INCOME G4 Number of Occurences FREQUENCIES HOME - Number o f Occurences CONDESCRIPTIVE VALUACRE S t a t i s t i c a l Output Number o f Occurences 35 Subprogram NONPAR CORR is appropriate to measure the association between two variables when one or both variables are ordinal in nature. The s t a t i s t i c s computed are SpearmanR and/or Kendall rank-order Ks c o rre la tio n c o e ffic ie n ts which are non-parametric. Subprogram PEARSON CORR is used when two in te r v a l lev el variables are analyzed. Output is in the form of a Pearson R zero-order c o rre la tio n which indicates both the strength o f the lin e a r f i t to a regression lin e and the proportion of variance explained in the dependent va ria b le by the independent v a ria b le . The f i r s t program u t i l i z e d was subprogram CROSSTABS. Some of the theories formulated fo r analysis with subprogram CROSSTABS included: 1) That there is a relatio n sh ip between variables 0NH20, INCOME, PUBPROP, PROPVALU, BUILDING, LANDREG, ZONING, QUANSERV, QUALSERV, SELL, PROPTAX, LEARN, REASON, ACQUIRE, with va ria b le TOWNSHIP ( e . g . , some townships may have a s u b s ta n tia lly greater number o f property owners on lakes than others.) 2) There is a re la tio n s h ip among the aforementioned variables and v a ria b le COUNTY, ( e . g . , there is a d iffe re n c e between the income levels o f one county when compared to other counties.) 3) There is a re la tio n s h ip between variables ACQUIRE, REASON, LEARN, SELL, YRSELL, QUALSERV, QUANSERV, ZONING, LANDREG, BUILDING, PROPVALU, 0NH20, INCOME, CTYBEFOR, CTYNOW, SEX, MARITAL, TOWNSHIP, COUNTY, with v a ria b le HOME, (e .g ., a ttitu d e s toward property tax levels d i f f e r between pro­ perty owners with a permanent home and those with a seasonal home.) 36 The output from subprogram CROSSTABS consists o f tables which can be a 2 x 2 format to an N by N. u t i l i z e d in th is study are: S t a t is t ic s a v a ila b le with CROSSTABS and Chi-Square, Cramers V, Contingency C o e ffic­ ie n t and Uncertainly C o e ffic ie n t. Once again, some s t a t i s t i c s require a ce rtain level of measurement before they can be computed. s t a t i s t i c s are o f l i t t l e Also, some value in s p e c ific cases, th e re fo re , Table 4 shows CROSSTABS ta b le s , th a t were computed, along w ith appropriate s ta tis tic s . I t is quite obvious th a t subprogram CROSSTABS cannot handle a l l the v a ria b le relatio n sh ip s requiring in v e s tig a tio n in th is phase of the study. A va ria b le such as TOTVALUE, in which the value of the property is being measured against type o f home development present (v a r ia b le HOME), is o f l i t t l e use in ta b u lar form because o f the numerous values th a t TOTVALUE may have. I t would be more useful to know how TOTVALUE and HOME vary in respect to each other, and also what is the mean and standard v a ria tio n f o r TOTVALUE with each d i f f e r e n t HOME type. Subprogram BREAKDOWN is the appropriate procedure to handle th is type o f an alysis. I t allows not only f o r ta b u la r analysis of continuous or d iscrete v a ria b le s , but also analysis between two variables while c o n tro llin g f o r up to four other v a ria b le s . In a d d itio n , when the independent va ria b le is continuous, ( in te r v a l scale) and the dependent va ria b le is discrete (nominal or ordinal s c a le ), a t - t e s t can be performed on the arith m e tic means to te s t fo r s ig n if ic a n t differences between the categories of the d iscrete v a ria b le . Some of the theories and hypotheses formulated f o r analysis w ith subprogram BREAKDOWN included: 1) That there is no s ig n if ic a n t d iffe re n c e between the county means fo r variables TOTACRES, TOTVALUE, AGE, TOTAL, VALUACRE. 37 Table 4 VARIABLE RELATIONSHIPS ANALYZED USING SUBPROGRAM CROSSTABS AND DESCRIPTIVE STATISTICS COMPUTED S t a t is t ic s Chi-square Cramers V U ncertainty C o e ffic ie n t Contingency C o e ffic ie n t B iv a ria te Relationships____________ COUNTY/TOWNSHIP/HOME by ACQUIRE X X X by REASON X X X by LEARN X X X by PROPTAX X X X X (V a ria b le HOME only) X (V a ria b le HOME only) by SELL by QUALSERV X X X by QUANSERV X X X by ZONING X X X by LANDREG X X X by BUILDING X X X by PROPVALU X X X by 0NH20 X X X by PUBPROP X X X by CTYBEFOR X X X by CTYNOW X X X by SEX X X X by MARITAL X X X by INCOME X X X X (V ariab le COUNTY, HOME only) 38 Table 4 ( C o n t 'd .) S t a t is t ic s ChiSquare Cramers V Uncertainty C o e ffic ie n t Contingency C o e ffic ie n t B iv a ria te Rela­ tionships 0NH20 by PROPTAX X X X by PUBPROP X X X by CTYBEFOR X X X by CTYNOW X X X by SEX X X X by MARITAL X X X by INCOME X X X COUNTY by HOME X X X HOME by INCOME X X X ■ X 39 2) That there is no s ig n if ic a n t d iffe re n c e between the township means f o r variables TOTACRES, TOTVALUE, AGE, TOTAL, VALUACRE. 3) That there is no s ig n if ic a n t d iffe re n c e between the types of home development means fo r variables TOTACRES, TOTVALUE, AGE, TOTAL, VALUACRE. 4) There is no substantial d ifferen ce between county, township, or type of home development means f o r variables TOTACRES, TOTVALUE, AGE, TOTAL, VALUACRE, when c o n tro llin g fo r variables 0NH20, TYPEH20, SIZEH20, and PUBPROP. Table 5 shows the v a ria b le re la tio n s h ip computed with subprogram BREAKDOWN along with the d e sc rip tiv e s t a t i s t i c s employed. NONPAR CORR is the procedure required to be able to handle r e la t io n ­ ships among variables where one or more o f the variables is measured at the ordinal le v e l. ables are b i v a r ia t e . With NONPAR CORR, a l l relation ship s among the v a r i ­ There is no c o n tro llin g f o r the influence o f other variables and, th e re fo re , a l l co rrelatio ns are zero order. S ta tis tic a l output from subprogram NONPAR CORR includes the mean, standard d e v ia tio n , SpearmansD and Kendall Tau. The l i s t s o f variables used fo r b iv a r ia t e Ks c o rre la tio n analysis with NONPAR CORR were: TOTACRES, SELL, YRSELL, TOTVALUE, PROPTAX, QUALSERV, QUANSERV, ZONING, LANDREG, BUILDING, PROPVALU, SKIAREA, SKIMILES, PUBPROP, 0NH20, TYPEH20, SIZEH20, AGE, SEX, MARITAL, TOTAL, INCOME, HOME and VALUACRE. The la s t subprogram to be used in th is section is PEARSON CORR. This subprogram also produces zero order c o rre la tio n between pairs o f v a ria b le s . However, i t is a s l i g h t l y stronger c o rre la tio n procedure than NONPAR CORR as both variables in the b iv a r ia t e analysis must meet in te rv a l scale requirements. Thus, va riables with in te r v a l scale a ttrib u te s can be 40 Table 5 VARIABLE RELATIONSHIPS AND STATISTICS COMPUTED WITH SUBPROGRAM BREAKDOWN S t a t is t ic s Mean Standard Deviation X X X by TOTVALUE X X X by AGE X X X by VALUACRE X X X X X X by TOTVALUE X X X by VALUACRE X X X C o n tro llin g fo r 0NH20 X X C o n tro llin g fo r TYPEH20 X X C o n tro llin g f o r PUBPROP X X C o n tro llin g fo r 0NH20 X X C o n tro llin g f o r TYPEH20 X X C o n tro llin g f o r PUBPROP X X T-Value B iv a ria te Relationship COUNTY/TOWNSHIP/HOME/ 0NH20/PUBPR0P by TOTACRES TYPEH20/SIZEH20 by TOTACRES COUNTY/TOWNSHIP/HOME by TOTACRES by VALUACRE 41 analyzed to determine strength and sign o f the re la tio n s h ip . S t a t is t ic a l output from subprogram PEARSON CORR consist o f mean, standard d e via tio n , and Pearson r . The l i s t o f variables used fo r b iv a r ia te c o rre la tio n analysis with PEARSON CORR were: TOTACRES, TOTVALUE, SKIMILES, LAKEMILE, RIVMILE, AGE, SEX, TOTAL, VALUACRE, DAYSTAY, FALL, WINTER, SPRING, SUMMER. The preceding analysis with the various SPSS subprograms was intended to develop an information base in order th a t s p e c ific questions concerning c h a ra c te ris tic s of property owners in the study area can be answered. Some o f the questions th a t the research addressed included: 1) What reasons are most important fo r i n i t i a l property acquisition? 2) What sources o f information are most important in learning about a v a ila b le property? 3) How do median fam ily income levels d i f f e r between property owners in each o f the studied counties and townships and types of home development? 4) What is the in te n t to s e ll property among property owners in d i f f e r e n t counties, townships, and types o f home development? 5) What are the a ttitu d e s o f property owners toward land use regu lations, fu tu re r e s id e n tia l b u ild in g , and the fu ture of property values? 6) What are the a ttitu d e s toward property tax lev els and the q u a lity and q u an tity o f municipal services provided? 7) Are people aware of development regulations concerning the property they own? 8) What is the average parcel size owned by northern Michigan property owners? 42 9) What is the d is tr ib u tio n between seasonal home owners, permanent home owners, and property owners with no present housing s tru c tu re , among counties, townships, and natural resource ch aracteristics? 10) How many v i s i t s and what is the average length o f stay fo r seasonal home owners? 11) What is the current region (County group) of residence f o r property owners? 12) How do other general and socioeconomic c h a ra c te ris tic s ( i . e . age, sex) d i f f e r among land owners in d i f f e r e n t counties, townships, and with d if f e r e n t types of home development? I t is c le a r from a review o f the various s t a t is t ic s and te sts in subprograms CROSSTABS, BREAKDOWN, NONPAR CORR, and PEARSON CORR th a t c e rta in very important v a ria b le relation ship s could not be te ste d . This is due to the nature of the various computer programs. The CROSS­ TABS procedure allowed two or more variables th a t did not have many discrete categories to be compared. Subprogram BREAKDOWN allowed variables th a t were both continuous and discrete to be tested w hile at the same time c o n tro llin g fo r other v a ria b le s . However, when more than one v a ria b le was c o n tro lle d , s t a t i s t i c a l relatio n sh ip s became confused and significan ce te stin g became impossible. Subprograms NONPARR CORR and PEARSON CORR allowed fo r b iv a r ia te analysis with s t a t i s t i c a l output lim ite d to zero order p a r t ia l c o rre la tio n c o e ffic ie n ts (no other variables c o n tro lle d ). The importance o f a l l the preceeding subprograms then was: 1) To describe property owners in the study area in terms o f c e rta in va ria b le s. 43 2) To begin to look a t ce rtain relationships between va ria b le s. 3) To draw conclusions and develop hypotheses based on those v a ria b le re latio n sh ip s. This then led in to the next phase o f the research process - the development of a m u ltip le regression model. The m u ltip le regression model allows f o r relatio n sh ip s among variables to be explored while a t the same time c o n tro llin g fo r many other v a ria b le s. The m ultiple regression model serves two purposes (1) d e s c rip tio n , and (2) predic­ t io n . The d e sc rip tiv e p a rt o f the model is concerned with measuring the parameters associated with the independent variables and the c o rr e la tio n between the independent and dependent variables (more w i l l be said about th is l a t e r when the form o f the model is considered). The p red iction purpose of the model is important in th a t once the re la tio n s h ip between the independent and dependent variables has been described then any new value of an independent va ria b le can be incorporated in to the model. The r e s u lt w i l l be the p re d ic tio n , th at is the expected change in the dependent va ria b le associated with the new independent v a ria b le . The regression model is of the cross sectional v a r ie ty . That i s , data are analyzed from a sample which is assumed to be a representative cross section o f the population of in t e r e s t at a given time. In th is case, a cross section o f northern Michigan study area property owners as id e n t if ie d by questionnaire responses f o r the year 1978. The SPSS subprogram REGRESSION is used in th is section. Variables to be Tested Using Regression The dependent v a ria b le chosen fo r analysis was value per acre of land. Land value was viewed as the s in g le most important economic v a ria b le 44 in the survey. I t is r e a d ily id e n t if ie d and measurable and is the u n it o f comparison most often used in real property transactions. In a d d itio n , value per acre of land fig ures can be in f la t e d or d e flated through various indexes, and th e re fo re , can be converted over to time series analysis fo r fu ture research. The independent variables to be regressed on value per acre of land were chosen in two ways (1) to te s t c e rta in hypotheses concerning the dependent v a r ia b le , and (2 ) variables id e n t if ie d through zero order p a r t ia l c o rre la tio n which showed a high degree o f association with the dependent v a ria b le . Care should be taken in v a ria b le s e le c tio n , through zero order p a r t i a l s , so th a t multi c o l l i n e a r i t y does not become a problem. More w i l l be said about th is l a t e r . Measurement o f the Dependent Variable Y l. Value per Acre o f Land Value per acre o f land, in d o lla r s , taken from questionnaire responses and measured in 1978 d o lla r s . Survey respondents estimated to ta l worth o f t h e i r property which was then divided by amount of acres owned. The r e s u lt was value per acre of land per respondent and became the dependent v a ria b le in the regression. Measurement o f the Independent Variables X I. Type o f Home Development The v a ria b le , HOME, which contains the information concerning the type o f home development a property owner has is measured a t the ordinal le v e l. Due to the nature of regression, a l l independent va ria b le s must be measured a t le a s t on an in te rv a l scale. Therefore, HOME is converted in to new in te rv a l level 45 variables termed "dummy" or "binary" v a ria b le s. I f a property owner has no home development on the property in the study area, then a new v a ria b le , E l, receives a value of one. Likewise, i f a property owner has a seasonal home on the property, then a new v a ria b le , E2, receives a value of 1 and El receives a value o f 0. I f a property owner has a permanent home on the property, then both El and E2 receive values of zero. No E3 v a ria b le is created because serious problems of multi c o l l i n e a r i t y (s in g u la r m atrix) w i l l r e s u lt . Rather, the value of an acre of land with a per­ manent home is re fle c te d in the in te rc e p t term of the regression model (more w i l l be said about th is l a t e r ) . X2. County V ariab le COUNTY is measured a t the nominal le v e l, and th e r e fo re , must be converted to an in te rv a l level through the use of dummy v a ria b le s . The procedure is the same fo r th a t used in converting v a ria b le HOME to an in te r v a l level of measurement. The newly created variables are Cl which w i l l r e fe r to Crawford County and C2 which refe rs to Kalkaska County. Otsego County is r e fle c te d in the in te rc e p t term o f the model. X3. Township Once again a nominal level va ria b le which must be converted to an in te r v a l scale. The township dummy variables which were created are D1 South Branch, D2 Grayling T27R2W, D3 Grayling T26NR4W, D4 Orange, D5 Blue Lake, D6 G a r fie ld , D7 Bagley, and D8 Chester. The la s t township, Dover, is re fle c te d in the i n t e r ­ cept term. X4. Public Property This information concerning whether a property owner is or is not 46 adjacent to public land was derived from va riable PUBPROP. This is also a dummy v a r ia b le , J l , which receives a value of 1 i f the property in question is adjacent to public land and 0 i f not. X5. Ski Area Another dummy v a r ia b le , K l, receiving a value o f 1 i f a property owner is close to any commercially developed ski area, and a 0 i f not. X6. Water Resources The type, s iz e , and nearness o f water resources are a l l variables th a t were regressed to see i f they contribute s ig n if ic a n t ly to value per acre. The information concerning type o f water resources a property owner is located on is contained in va ria b le TYPEH20. This is a nominal lev el v a ria b le which is transformed by dummy conversion in to v a ria b le G1 which receives a value of 1 i f a property is located on a r iv e r and 0 i f not. There is no G2 v a r i­ able fo r property owners on a lake because of multi c o llin e a r it y with the next water c h a r a c te r is tic analyzed - size of lake where the property is located. I f a property owner is on a lake, then there is a c e rta in size associated with th a t lak e, and the informa­ tio n concerning lake size is located in v a ria b le SIZEH20. SIZEH20 is measured a t the ordinal level and was converted through dummy manipulation in to variables H I, H2, H3, H4. va ria b le s correspond to a lake s iz e . Each one of the new HI receives a value of 1 i f property is located on a lake or pond less than 25 acres; H2 receives a value of 1 i f property is located on a lake 25-100 acres in size ; H3 receives a value o f 1 i f property is located on a lake 101-500 acres in size and H4 receives a value of 1 i f property is located on a lake over 500 acres in s iz e . The closeness to water resources 47 was also analyzed. The variables containing th is information are LAKEMILE and RIVMILE. They are continuous variables measured a t the r a t i o sc ale, and th e re fo re , no dummy conversions are necessary fo r t h e ir inclusion in to the m u ltip le regression model. However, fo r inclusion in the model s p e c ific a tio n l i s t , LAKEMILE becomes v a ria b le X6a and RIVMILE becomes v a ria b le X6b. X7. Total Acres Owned This is a continuous lev el r a t io scale v a ria b le which was taken d ir e c t ly from the questionnaire and is lis t e d as v a ria b le TOTACRES. Measurement is in acres as id e n t if ie d from questionnaire responses. In summary then, the m u ltip le lin e a r regression model is of the form Y1 = BQ + B] X1 + B2X2 .............................. B?Xy + f which when converted in to a workable form, required because o f the use of dummy v a ria b le s , become Y1 = BQ + B1E1 + B2E2 + B3C1 + B4C2 + B5D1 + BgD2 + B7D3 + B8D4 + B9D5 + B10D6 + B11D7 + B12D8 + B13J 1 + B14K1 + B15G1 + B16H1 + B17H2 + B18H3 + B19H4 + B20X6A + B21X6B + B22X7 + ^ where: E-j = no home on the property in the study area when El=l E2 = seasonal home on the property whenE2= 1 C.| = Crawford County when C l=1 C2 = Otsego County when C2=l D.j = South Branch T25NR2W Township whenDl=l D2 = Grayling T27R2W Township whenD2=l = Grayling T26NR4W Township when D3=l 48 = Orange Township T26NR7W when D4=l Dg = Blue Lake Township T28NR5W when D5=l Dg = G a rfie ld Township T25NR7W when D6=l Dy = Bagley Township T30NR3W whdn D7=l Dg = Chester Township T29NR2W when D8=l = Property adjacent to public land when J 1=1 K-J = Property located close to a commercial ski area when Kl=l G-j = Property located on a r iv e r when G1=1 H, = Property located on a lake which is less than 25 acres when Hl=l H2 = Property located on a lake which is between 25 and 100 acres when H2=l Ho = Property located on a lake which is over 100 acres but less than 500 acres when H3=l H. = Property located on a lake which is over 500 acres when ^ H4=l * X6^= Distance property is from a lak e, miles X6g= Distance property is from a r i v e r , miles Xy = Total number o f acres in the property f- = E rror or residual not explained by independent v a r iib le s in the regression Bq........................ B2 2 are parameters to be f i t t e d . This model can then be broken down in to three separate regressions which are: Y1 = B0 + B1E1 + B2E2 + B3C1 + B4C2 + B13J 1 + B14K1 + B15G1 + B16H1 + B17H2 + B18H3 + B19H4 + B20X6A + B21X6B + B22X7 + ^ This regression is county s p e c ific in th a t i t uses the independent county variables as proxy v a ria b le s . Any unique county c h a ra c te ris tic s should be re fle c te d in the c o e ff ic ie n t value f o r each county in the regression. A second regression was: Y1 = B0 + B1E1 + B2E2 + B5D1+ B6D2 + B7D3 + B8D4 + B9D5 + B10D6 + B11D7 + B12D8 + B13J1 + B14K1 + B15G1 + B16H1 + B17H2 + B18H3 + B19H4 + B20X6A + B21X6B + B22X7 + ^ This regression is township s p e c ific as each township's independent vari able is used as a proxy f o r any unique c h a ra c te ris tic s in each township. The th ir d regression was: Y! = B0 + B-j E] + B2E2 + B13J 1 + B14K] + B156 1 + B ^ B17H2 + B18H3 + B19H4 + B20X6A + B21X6B + B22X7 + + ^ This is the s im p lifie d regression and is n e ith e r township or county s p e c ific , but ra th e r represents study area property owners without respect to area o f residence. The best regression of the three depends on what use is to be made from i t . For example, local planners may want to use the township s p e c ific regression whereas regional planners may opt f o r the county or the s im p lifie d regression. Each of the three regressions were analyzed and are presented in the data analysis section Adjusting the C lassical Linear Regression Model The basic form of the model is the same f o r a l l three regressions. This form o f the model was chosen because i t is the cla ssical lin e a r regression model and there was no evidence to suggest th a t lo g arith m ic , polynomial, e tc . model s p e c ific a tio n s would be more appropriate. Tests 50 performed which supported the decision to se le c t the c la ss ical lin e a r form included p a r t ia l c o rre la tio n an aly sis, F -te s t f o r c u r v i l i n e a r i t y , examinations o f residuals and run o f signs. ( B q ................ B 2 2 ) f ° rrn The parameters to be f i t t e d the basis fo r the d escrip tive nature of the model. The parameter Bq represents value per acre of land in d o lla rs given ce rtain property c h a ra c te r is tic s . For example, in the measurement of independent v a ria b le s , reference was made as to how there was no need fo r the creation of a dummy v a ria b le representing property owners in Otsego County. Instead, the value per acre fo r property owners in Otsego County is r e fle c te d in the Bq parameter. For property owners in Crawford County, the value per acre is an addition or subtraction to the value re fle c te d in B q . When a l l the parameters are f i t t e d , through the regression, the d e scrip tive nature of the model is complete. The parameters then describe how each independent va ria b le is re la te d to the dependent v a ria b le . Once the model parameters were ascertained, v a ria b le sig n ifican ce te s tin g and confidence in te rv a ls were e a s ily obtained. Even though great pains are taken to elim in ate multi c o l l i n e a r i t y problems from the regressions, they s t i l l a r is e . model was modified to deal with th is problem. The basic lin e a r The model was checked as the analysis proceeded, and i t became necessary to develop a model with in te r a c tio n terms. Any p a ir of variables which showed a c o rr e la tio n greater than .20 in the simple c o rre la tio n matrix (Appendix C) were transformed in to an in t e r a c t iv e term. These new combinations were: C1G1 = Cl x G1 C2H2 = C2 x H2 C2H3 = C2 x H3 D3G1 = D3 x G1 51 D5H2 = D5 x H2 D5H3 = D5 x H3 D7J1 = D7 x J1 E2H3 = E2 x H3 In building the model, the option th a t was used was stepwise regression. The preferred option is usually a l l possible regressions, however, there are too many variables in the model and the output would be confusing a t best. Stepwise regression allows the variables to enter according to t h e i r r e la t iv e importance. The v a ria b le which accounts fo r the most variance around the mean enters f i r s t , and so on u n til a l l the variables are in to the regression or do not meet the sig nifican ce level requirements chosen f o r inclusion in to the regression. The advantages o f stepwise regression, are th a t i t allows the researcher to see the model developing and see the importance o f some variables change as others enter in to the model. A disadvantage is th a t one va ria b le which may explain a good deal of variance may not meet the s ig n ifican ce level requirements to enter in to the model because i t s influence is being suppressed by a v a ria b le already in the model. However, a combination of stepwise and h ie ra rc h ia l regression can be u t i l i z e d i f a suppressed va ria b le is sus­ pected. Some of the questions th a t th is p art of the study s p e c if ic a lly addressed included: 1) What variables tested are most useful in explaining value per acre o f land? 2) What regression also resu lts in 3) Is there enough model explains the most variance and the lowest residual mean square error? evidence to support the theory th a t cer­ ta in natural resource c h a ra c te ris tic s play an important 52 role in real property valuation? The hypotheses used in regression analysis are usually re fe rre d to as null hypotheses. That i s , instead o f asking what value a population parameter is l i k e l y to have, the null hypothesis says the value is zero and the a lte r n a tiv e hypothesis then states the value is d i f f e r e n t than zero. The nu ll hypothesis is te s te d , and i f i t cannot be disproved, is accepted. 1) The null hypotheses tested in th is study were: There is no lin e a r re la tio n s h ip between the dependent v a ria b le , Value per Acre of land, and the independent va ria b le s , Type of Home Development ( E - ^ ) , County ( C - ^ ) , Township (D^ D g ) , Water Resources (G-j, H4 ^ 6AX6B^’ Public Land (J -j), Ski Area (K-j), and Total Acres Owned (X^). 2) Each independent va ria b le lis t e d above has no s ig n if ic a n t e f f e c t on value per acre of land once the e ffe c ts o f the other independent variables are adjusted fo r . 3) The relatio n sh ip between value per acre o f land and any p a r t ic u la r independent v a ria b le is n o n -lin ea r and the e f f e c t of two or more independent va riables are not a d d itiv e . S ignificance Level The level o f significan ce chosen f o r th is study is .05. This level was chosen f o r two reasons, (1) i t was the predominant sig n ifican ce level encountered in the l i t e r a t u r e review f o r research dealing with property purchase and home development; (2) i t provides a basis fo r external v a lid a tio n . I f resu lts from th is research are to be compared to resu lts from any other study, then i t is important th a t s ig n ifican ce levels be the same. 53 Non-Response E rror In any survey there w i l l always be a c e rta in percentage of people who w i l l not respond. This group of non-respondents may represent a d if f e r e n t population e n t i r e l y than the group which did respond. To determine i f non-respondents do indeed make up a separate population, i t is necessary to obtain some responses from the group of non-respondents. This was achieved through a telephone survey of a randomly selected portion of the non-respondents. A telephone survey is more intensive than the mail survey and did receive additional responses. Also, to help achieve a measureable response from the previous non-respondents, the i n i t i a l questionnaire was reduced to f iv e key questions. These questions deal with type of home development, t o ta l acreage owned, t o ta l value of the property, information leading to i n i t i a l purchase and in te n t to s e l l . Responses were analyzed and compared with the survey re s u lts . No s ig ­ n if ic a n t differences were encountered between values from i n i t i a l respondents and the random sample of non-respondents, fo r the f iv e key variables tested. S t a t is t ic s There were many d e sc rip tiv e and i n f e r e n t ia l s t a t i s t i c s used in th is study. computes. Each subprogram u t i l i z e d has a set of s t a t i s t i c s th a t i t Many o f the s t a t i s t i c s computed are understood by almost a l l researchers (e .g . mean and median), however, there may be some s t a t i s t i c s used in th is research u n fa m ilia r to even an ardent researcher. Therefore, the follow ing l i s t of s t a t i s t i c a l d e fin itio n s has been compiled f o r each d e scrip tive s t a t i s t i c used in th is repo rt. Mean or average is the sum of in d ivid u a l case values divided by the number of cases. the mean. In te rv a l scale measurement is required to compute 54 Median is the value o f the case lyin g on the 50th p e rc e n tile . One-half of the cases have values higher than the median and on e-h alf are lower. Ordinal level measurement is required to compute the median. Mode is the value o f the v a ria b le th a t occurs most often. Any level o f measurement is adequate to compute the mode. Skewness is a measure in d ic a tin g the degree to which a sample d is tr ib u tio n o f cases approximates a normal curve. A p o s itiv e value indicates a clu ste rin g o f cases to the l e f t of the arith m e tic mean while a negative value in dicates c lu s te rin g to the r ig h t of the a r i t h ­ metic mean. In te rv a l lev el measurement is required to compute a skewness value. Kurtosis is also a measure in d ic a tin g the degree to which a sample of cases approximate a normal curve. I f the kurtosis value is p o s itiv e , i t indicates th a t the curve is more peaked than a normal d i s t r ib u t io n . A negative value indicates the curve is f l a t t e r than a normal d i s t r ib u ­ tio n . A kurtosis value of 0 indicates a normal d i s t r ib u t io n . In te rv a l level measurement is required to compute a kurtosis value. Variance is a measure o f data dispersion about the arith m e tic mean. The smaller the variance, the more homogeneity in the data. In te r v a l level measurement is required to compute the variance. Standard Deviation is the mathematical square root of the variance. I t is another measure o f dispersion about the a rith m e tic mean. Standard deviation has more i n t u i t i v e meaning than the variance because i t is based on the same units as the o r ig in a l v a ria b le . For example, i f the variance in a sample of to ta l acres owned per property owners is 100, then what is being referred to is 100 squared acres. The standard d e v ia tio n , however, is 10 acres per property owner and is re a d ily comprehendable. 55 In te rv a l level measurement is required to compute the standard d e via tio n . Chi-Square is a te s t fo r s t a t i s t i c a l s ig n ific a n c e . I t s main function is helping in determining whether a systematic re la tio n s h ip ex ists between two v a ria b le s. The chi-square s t a t i s t i c reported in th is te x t is followed by degrees of freedom and whether the s t a t i s t i c is s ig n if ic a n t a t the .05 le v e l. For example, i f a chi-square value o f 97.62 is obtained with 8 degrees of freedom, reference to a chi-square values ta b le w i l l in dicate th a t there is a s ig n if ic a n t re la tio n s h ip . Care should be taken, however, in in te r p r e tin g chi-square s t a t is t ic s as to the r e la t iv e strength o f the re la tio n s h ip . A large chi-square value does not necessarily mean a v a r i ­ able relatio n sh ip is strong. Rather chi-square should be used only to , i n f e r th a t a re la tio n s h ip does or does not e x is t . re la tio n s h ip is a matter fo r other s t a t i s t i c s . The strength o f th a t Chi-square is the le a s t powerful of s t a t i s t i c s used to determine s ig n ific a n c e . Only nominal level measurement is required fo r one or both va ria b le s to compute the chi-square s t a t i s t i c . Cramers V is a measure o f the strength of re la tio n s h ip between v a ria b le s . Values fo r Cramers V range from 0 to +1. indicates th a t the association is strong. dicates the re la tio n s h ip is weak. A large value A low value (close to 0) in ­ Only nominal level measurement is required fo r one or both variables to compute Cramers V. Contingency C o e ffic ie n t is another s t a t i s t i c which measures the degree of association between two v a ria b le s . I t has a minimum value o f 0 in d ic a tin g the absence of any re la tio n s h ip between the v a ria b le s , but i t s maximum value is dependent on the size o f the crosstabs ta b le . Therefore, the Contingency C o e ffic ie n t should only be used with tables having the same number of rows as columns. Only nominal level measurement is required fo r one or both variables to compute the Contingency Coe f f i c ie n t. 56 Uncertainty C o e ffic ie n t (asymmetric) is a s t a t i s t i c which has d ire c t meaning. The computed value is a c tu a lly the proportion of uncertainty in the dependent va ria b le reduced by knowledge of the independent v a ria b le . For example, two va ria b le s HOME (type of home development) the dependent v a ria b le , and TYPEH20 (w a ter, r i v e r , no water resource) the independent v a ria b le . When s t a t i s t i c a l l y analyzed, y ie ld an Uncertainty C o e ffic ie n t (asymmetric) value o f .4732 in d ic a tin g th a t 47% of the uncertainty in knowing what type o f home development is located on the land is eliminated when type of water resource th a t land is located on is known. An Uncer­ t a in t y C o e ffic ie n t value o f 1 indicates a l l uncertainty is removed and each individual type of home development is associated with one s p e c ific type of water resource. Only nominal level measurement is required fo r one or both variables to compute the Uncertainty C o e ffic ie n t (asymmetric). T-Test is a s t a t i s t i c computed to measure whether or not s ig n if ic a n t differences e x is t between two groups. Depending on the degrees of freedom, variance in the sample and level o f sig nifican ce chosen, a t - t e s t can be performed and the t - s t a t i s t i c computed to see whether or not two groups means are s ig n if ic a n t ly d i f f e r e n t . In te rv a l level measurement is required fo r both variables to compute the t - s t a t i s t i c and u t i l i z e a t - t e s t . SpearmansD (Spearman Rank C o rrelation C o e ffic ie n t) is a measure of Ks c o rre la tio n between two v a ria b le s . C orrelation c o e ffic ie n ts vary between +1 and - 1 . A value of +1 or -1 indicates complete association between the two v a ria b le s . However, a + indicates the va ria b le moves in the same d ire c tio n and a - indicates they move in opposite d ire c tio n s . SpearmansD is a nonparametric s t a t i s t i c which means i t is derived from Ks variables which do not necessarily have a normal or known d is tr ib u tio n . In a d d itio n , only ordinal lev el measurement is required fo r computation of SpearmansD . Ks 57 Kendall Tau is also a non-parametric s t a t i s t i c and is quite s im ila r to SpearmansD . The major d iffe re n c e between the two in th a t Kendall's Ks tau is somewhat more meaningful when data are ranked as opposed to con­ tinuous. Pearsons r is a parametric c o rre la tio n c o e ff ic ie n t ranging between +1 and -1 . I t has the same i n t u i t i v e meaning as SpearmansD and Kendall Ks tau, but i t has s lig h t ly more power because of the assumption of in te rv a l level measurement requirements. Certain s t a t is t ic s u t i l i z e d in the study ( i . e . Cramers V, Contingency C o e ffic ie n t, SpearmansD » Kendall Tau, Pearsons r) in d ic a te the strength Ks of relatio n sh ip between two v a ria b le s . The absolute value of these s t a t is t ic s range from 0 to 1 and as value increases the strength of va ria b le relationships becomes stronger. However, there is no ru le as to which range o f values indicates a r e la tio n s h ip is weak, moderate or strong. Strength of re la tio n s h ip is a r b i t r a r i l y determined by each researcher. In th is study the range of values and corresponding strength of association fo r each affec ted v a ria b le is reported in Table 6. Table 6 RANGE OF VALUES AND CORRESPONDING STRENGTH OF ASSOCIATION FOR EACH VARIABLE STATISTICALLY TESTED Cramers V Weak Moderate Strong Contingency C o e ffic ie n t Spearmansps Kendall Tau Pearsons r 0 -.1 0 - .1 5 0 -.1 0 -.1 0 -.1 5 .1 1 -.2 .1 6 -.2 5 .1 1 -.2 .1 1 -. 2 .1 6 -.2 5 .21 .26 .21 .21 .26 58 Summary In th is section the tools and model used in th is study have been i d e n t if ie d . The basic tools used were a questionnaire and the subprograms contained in SPSS. The model used was cross sectional m u ltip le regression w ith dummy v a ria b le s . A s t r a t i f i e d random sampling technique was employed to c o lle c t data. Three townships in each o f three counties (Kalkaska, Crawford, Otsego) were surveyed. SPSS subprograms used in i n i t i a l data analysis included FREQUENCIES, CONDESCRIPTIVE, CROSSTAB, BREAKDOWN, NONPAR CORR, and PEARSON CORR. The SPSS subprogram REGRESSION was u t i l i z e d in developing the cross sectional m u ltip le regression model. Variables tested in the regression model included location c h a ra c te ris tic s ( p o l i t i c a l , natural resource), property c h a ra c te ris tic s ( s iz e , home development) and combination v a r i ­ ables ( i n t e r - a c t i v e term considered to be highly co rrela ted with each o th e r ). A ll tests fo r s t a t i s t i c a l s ig n ifican ce were performed at the .05 p r o b a b ility le v e l. CHAPTER I I I SOCIOECONOMIC CHARACTERISTICS OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS What are the c h a ra c te ris tic s o f northern Michigan study area property owners? Before the a t titu d e s , concerns, and other general c h a ra c te ris tic s o f northern Michigan study area property owners can be explored, i t is important th a t socioeconomic c h a ra c te ris tic s be known. T h e re a fte r, when describing a ttitu d e s and concerns o f northern Michigan study area property owners, i t w i l l be c le a r what group of people is being referenced. follow ing analysis w i l l be divided in to three segments. includes property owners in s p e c ific counties. property owners in s p e c ific townships. The The f i r s t group The second group includes The th ir d group includes property owners with d if f e r e n t types of home development. The reason fo r these groupings is so th a t any unique c h a ra c te ris tic s which may r e la t e to a s p e c ific county, township, or type o f home development may quickly become apparent. Age of Houshold In terms o f age, there is l i t t l e v a ria tio n in the sample. A 95% Confidence In te rv a l about the mean has a minimum of 52.119 and a maximum of 53.394, with a mode o f 50.00 and a median value of 53.25. In a d d itio n , a Kurtosis value o f -.561 indicates the curve is only s l i g h t l y f l a t t e r than normal and a skewness value of -.9 5 4 means there is only a s lig h t grouping o f values to the r ig h t o f the mean. Due to the fa c t of such a small confidence in te rv a l f o r mean age le v e ls , a problem in s t a t i s t i c a l analysis occurs. Each county's mean when compared to the overall mean is s t a t i s t i c a l l y s ig n if ic a n t . d iffe re n c e may be s t a t i s t i c a l l y s ig n if ic a n t but is not considered 59 This 60 o p e ra tio n a lly s ig n ific a n t (unless we are considering l i f e insurance). Therefore, separate analysis by county, township, and type o f home development by median age of property owners is not considered impor­ ta n t. A complete breakdown o f age by area and type o f home development is shown in Table 7. Gender o f Household Head Gender of head of household is another important c h a r a c te r is tic of property owners th a t has to be considered. Few decisions to purchase or s e ll property are made without substantial input from the household head. In the study area, heads of household who own property are pre­ dominantly males. In f a c t , Table 8 shows th a t 88.7% o f property owning households have male heads. This is not a t a l l s u rp ris in g . males are considered the head fo r 91.3% o f the households. S t a t i s t i c a l A bstract, 1978, pg. 77). In Michigan, (Michigan Therefore, there seems to be l i t t l e d i f f e r e n t i a t i o n between the northern Michigan study area property owners survey and resu lts fo r the e n tir e State of Michigan. M a rital Status of Household Head M a rita l status is another important c h a r a c t e r is t ic , e s p e c ia lly fo r developers. I t is generally considered th a t l i f e s t y l e s between married and ngn-married in divid uals d i f f e r considerably. A developer designing a subdivision fo r unmarried individuals may not s e ll many parcels o f land i f a l l o f his prospective c lie n t e le are married couples. As shown in Table 9, the overwhelming m ajority of study area property owners are married (83.7%). widowers. The second la rg e s t percentage (9.3%) are widows or The lowest recorded percentage (3.1%) are sin gle-never married. Once again, the analysis changes very l i t t l e between d i f f e r e n t counties or townships and type o f home development. 61 Table 7 MEAN AGE OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS, BY COUNTY, TOWNSHIP, AND TYPE OF HOME DEVELOPMENT Age Size of Sample Mean Standard Deviation Crawford 266 53.85 13.06 Kalkaska 455 54.78 12.73 Otsego 1,187 51.78 13.36 Total 1,908 52.78 13.22 128 53.78 12,32 Grayling T27NR2W 31 55.84 13.74 Grayling T26NR4W 106 53.42 13.82 Orange T26NR7W 112 51.63 13.56 Blue Lake T28NR5W 266 55.53 12.49 76 56.50 11.45 1,002 51.77 13.45 Chester T29NR2W 93 52.92 12.62 Dover T31NR2W 78 50.68 13.02 1,892 52.78 13.22 No Home 620 51.45 12.6 Seasonal Home 686 54.15 11.16 Permanent Home 586 52.62 15.72 1,892 52.78 13.22 County Township South Branch T25NR2W G a rfie ld T25NR7W Bagley T30NR3W Total Home Total NOTE: Sample size may be d i f f e r e n t f o r county, township, and type of home variables due to missing responses on some survey questions. 62 Table 8 GENDER OF HOUSEHOLD HEAD FOR NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS (Frequency and Percentage D is trib u tio n ) Sex Male Female Total Frequency Frequency Percentage 1,678 88.7 213 11.3 1,891 100.0 Table 9 MARITAL STATUS OF HOUSEHOLD HEAD FOR NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS (Frequency and Percentage D is tr ib u tio n ) Frequency Frequency Percentage 1,603 83.7 Single (Never Married) 60 3.1 Di vorced 74 3.9 178 9.3 1,915 100.0 M a rital Married Widow or Widower Total Family Size Total fam ily size fo r northern Michigan study area property owners was also considered. The mean fam ily s iz e is 3.24 w ith a median value o f 2.795 and a mode o f 2 .0 . was from 3.161 to 3.320. The 95% Confidence In te r v a l about the mean This t o t a l fa m ily size mean is q u ite s im ila r 63 to th a t reported f o r the State of Michigan in the 1970 Census which was 3.27 (Michigan S t a t i s t i c a l Abstract 1978, p. 79) When in d ivid u a l counties, townships, and type of home developments were considered, each category's mean d iffe r e d very l i t t l e o v erall mean. from the Some categories were found to be s t a t i s t i c a l l y s ig n if ic a n t (by only a fr a c tio n ) but none were judged o p eratio n a lly s ig n if ic a n t . Therefore, each category's mean can be considered the same as the o verall mean. Family Income Income, or lack of i t , is a c r i t i c a l fa c to r in decisions concerning major purchases such as real property. Inadequate income levels can prevent people from owning land and an abundant income can lead to an increase in more expensive purchases such as second or seasonal homes. (Nelson, 1973). Table 10 shows t h a t, in the study area 60.5% o f the property owners have fam ily income levels over $15,000. Somewhat sur­ p ris in g is the f a c t th at 32.4% o f a l l study area property owners have incomes exceeding $25,000. Table 10 FAMILY INCOME OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage - $ 5,999.00 172 9.4 $ 6,0 00 .00 - $ 9,999.00 263 14.4 $10,000.00 - $14,999.00 288 15.7 $15,000.00 - $25,000.00 504 27 .6 Over $25,000.00 602 32.9 1,829 100.0 Income $0.0 Total 64 When in divid ual counties are considered, as presented in Table 11, i t is found th a t property owners in Kalkaska county have generally less fam ily income than t h e i r counterparts in Crawford or Otsego counties. Only 53.3% of the property owners in Kalkaska county have fam ily incomes over $15,000/year whereas in Crawford 60.3% and in Otsego 63.2% of the property owners exceed $15,000 in median fam ily income. The differen ce is q u ite pronounced between Otsego and Kalkaska when the over $25,000/ year category is considered. In Otsego county 36.0% o f the property owners have fam ily incomes exceeding $25,000/year, but only 26.6% of the property owners in Kalkaska county exceed $25,000.00. In the lowest fam ily income category (0 - $ 5 ,9 9 9 /y e a r ), Kalkaska also has the largest percentage of property owners (12.3%) when compared to Crawford and Otsego (10.5% and 8.0%, re s p e c tiv e ly ). S t a t i s t i c s , presented at the bottom of Table 11, in d ic a te a r e la t io n ­ ship ex ists between property owners of ce rtain counties and fam ily income le v e ls . A Chi-square value o f 21.97381, s ig n if ic a n t a t the .05 pro­ b a b i l i t y le v e l, implies property owners o f c e rta in counties d i f f e r in terms of median fam ily income. Families with higher median incomes are more l i k e l y to own property located in Crawford county or Otsego counties and less l i k e l y to own land in Kalkaska county. However, other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between high median fam ily incomes and property owners o f c e rta in counties, although s ig n if ic a n t , is weak. The re la tio n s h ip is not strong enough to accurately p re d ic t which county a property owner would have his land l o ­ cated, given a ce rtain median fam ily income le v e l. When in dividual townships are considered as presented in Table 12, each township generally approximates the percentage d is tr ib u tio n Table 11 FAMILY INCOME OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS, BY COUNTY Count Row Percentage Row Cumulative Percentage Income 0 $5,999.99 $ 6,0 00 .00$9,999.99 $10,000.00$14,999.99 $15,000.00$25,000.00 Over $25,000.00 Crawford 27 10.5 10.5 40 15.6 26.1 35 13.6 39.7 79 30.7 70.4 76 29.6 100.0 257 14.1 Kalkaska 54 12.3 12.3 69 15.7 28.0 82 18.7 46.7 117 26.65 73.35 117 26.65 100.0 439 24.0 Otsego 91 8.0 8.0 154 13.6 21.6 171 15.1 86.7 308 27.2 63.9 408 36.0 99.9 1,132 61.9 172 9 .4 263 14.4 288 15.7 504 27.5 601 32.9 1,828 100.0 Row Total % of Total County Column Total % of Total Chi-square = 21.97381with 8 degrees of freedom S ig n ific a n t a t .05 p ro b a b ility level Cramers V = .07751 Uncertainty C o e ffic ie n t (asymmetirc) = .00654 with County dependent = .00395 with Income dependent *Due to rounding, to ta l percentages may not add up to 100. 66 c h a r a c te r is tic of i t s respective county. Otsego county townships fo r example, a l l had large percentages o f property owners with fam ily incomes over $15,000.0 0 /y e a r, led by Bagley township which had 64.3% of i t s pro­ perty owners making over $1 5,0 00.00/year. In Kalkaska county, a l l three townships generally had a much lower percentage o f property owners making over $15,000.0 0 /y e a r. Orange township was the lowest with only 45.5% of i t s property owners making over $15,0 00.0 0/y ear. Crawford county townships were the exception when i t came to township trends supporting county trends. South Branch township had 67.8% of i t s property owners making over $15,000.0 0 /y e a r, but Grayling T27NR2W had only 41.9% of i t s property owners in the same category. S t a t i s t i c s , presented a t the bottom of Table 12, in d ic a te a r e la t io n ­ ship exists between property owners o f ce rtain townships and fam ily income le v e ls . A Chi-square value of 71.64135, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l, implies property owners o f c e rta in townships d i f f e r in terms of median fam ily incomes. Other q u a lify in g s t a t i s t i c s , (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between higher median fam ily incomes and property owners o f c e rta in townships is weak. The re la tio n s h ip is not strong enough to accurately p re d ic t which township a property owner would have his land located, given a c e rta in median fam ily income le v e l. When types o f home development are considered, very sharp differences are noted. As shown in Table 13, property owners with no home in the study area and seasonal home owners both show a much la rg e r percentage in the higher fam ily income categories than property owners with permanent homes. Property owners with seasonal homes have the highest percentage (42.7%) in the "Over $25,000.00" category. By comparison, property owners with permanent homes have only 14.3% in the "Over $25,000.00" category. Property 67 Table 12 FAMILY INCOME OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS, BY TOWNSHIP Count Row Percentage Row Cumulative Percentage Income 0$5,999 $6,000$9,999 $10,000$14,999 $15,000$25,000 Over $25,000 South Branch T25NR2W 15 12.4 12.4 12 9.9 22.3 12 9.9 32.2 45 37.2 69.4 37 30.6 100.0 121 6.6 Gray!i ng T27NR2W 4 12.9 12.9 4 12.9 25.8 10 32.3 58.1 8 25.8 83.9 5 16.1 100.0 31 1.7 Grayling T26NR4W 8 7.5 7.5 24 22.6 30.1 13 12.3 42.4 26 24.5 66.9 35 33.0 99.9 106 5.8 Orange T26NR7W 14 12.7 12.7 21 19.1 31.8 25 22.7 54.5 23 20.9 75.4 27 24.5 99.9 no Blue Lake T28R5W 30 11.8 11.8 32 12.6 24.4 44 17.3 41.7 68 26.8 68.5 80 31.5 100.0 254 13.9 G a rfie ld T25NR7W 10 13.7 13.7 15 20.5 34.2 13 17.8 52.0 25 34.2 86.2 10 13.7 99.9 73 4.0 Bagley T30NR3W 69 7.2 7.2 135 14.0 21.2 140 14.5 35.7 269 28.0 64.7 349 36.3 100.0 962 52.6 Chester T29R2W 8 8.6 8.6 9 18.3 18.3 18 19.3 37.7 21 22.6 60.3 37 39.8 100.0 93 5.1 Dover T31NR2W 14 17.7 17.7 11 13.9 31.6 13 16.4 48.0 19 24.1 72.1 22 27.8 99.9 79 4.3 172 9 .4 263 14.4 288 15.7 504 27.5 602 32.9 1,829 100.0 Row Total % of Total Township | Column Total % of Total 6.0 Chi-square = 71.64135 with 32 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .09896 Uncertainty C o e ffic ie n t (asymmetric) = .02282 with Township dependent = .01285 with Income dependent *Due to rounding,percentage to ta ls may not equal 100 owners with permanent homes also have a higher percentage (17.2%) in the lowest income category (0 -$ 5 ,9 9 9 .9 9 ), than e it h e r property owners with no home in the study area (6.4% ), or seasonal homes (52%). This sharp percentage d iffe re n c e in median fam ily income levels between permanent home owners and a l l other property owners has been noted in previous research. Permanent home owners generally have lower incomes than fa m ilie s in the downstate area. Many permanent home owners are c o lle c tin g unemployment insurance or social se c u rity be n efits (G a lin , 1976; Michigan Public Opinion Survey, 1977; Marans and Wellman, 1978). This tends to keep median fam ily incomes fo r permanent home owners on the low side. Median fam ily income f o r the s ta te o f Michigan in 1976 was $15,758.00 (Michigan S t a t i s t i c a l A b s tra ct, 1979). Results from th is survey in d ic a te a large proportion o f seasonal home owners and property owners w ith no home development on t h e i r land s u b s ta n tia lly exceed the statewide median fam ily income le v e l. However, permanent home owners in the study area generally have median fa m ily incomes below the statewide le v e l. S t a t i s t i c s , presented a t the bottom of Table 13, in d ic a te a r e l a ­ tionship exists between property owners with d is s im ila r types o f home development and median fa m ily income le v e ls . A Chi-square value of 237.79638, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners with d is s im ila r types o f home development d i f f e r in terms of median fam ily incomes. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners with d is s im ila r types o f home development and median fam ily incomes is strong. Seasonal home owners and property owners with no home in the study area are more l i k e l y to have higher median fam ily incomes than permanent home owners. 69 Table 13 FAMILY INCOME OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS,BY TYPE OF HOME DEVELOPMENT Count Row Percentage Row Cumulative Percentage Income | 0$5,999 $6,000$9,999 $10,000$14,999 $15,000$25,000 Over $25,000 Row Total % o f Total Home No Home 38 6.4 6 .4 60 10.1 16.5 74 12.4 28.9 185 31.1 59.9 238 40.0 100.0 595 32.7 Seasonal Home 34 5.2 5.2 61 9 .4 14.6 84 12.9 27.5 193 29.7 57.2 277 42.7 99.9 649 35.7 Permanent Home 99 17.2 17.2 140 24.4 41.6 129 22.5 64.1 124 21.6 85.7 82 14.3 100.0 574 31.6 171 9 .4 261 14.4 287 15.8 502 27.6 597 32.8 1,818 100.0 Column Total I o f Total Chi-square = 237.79638 w ith 8 degrees of freedom S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .25574 Uncertainty C o e ffic ie n t (assymetric) = .06064 with Home dependent = .04750 with Income dependent *Due to rounding,totals may not add up to 100.0 When considering location of water resource to fam ily income le v e ls , little re la tio n s h ip was found. E m p iric a lly , there are s l i g h t l y more property owners who are located on a water resource in the highest income category. S t a t i s t i c a l l y , there was only a very weak re la tio n s h ip . Therefore, location on water resources is not viewed as a function of fam ily income le v e ls . Summary This chapter explored the socioeconomic c h a ra c te ris tic s of northern Michigan study area property owners, including county o f property 70 ownership, township of property ownership, and type o f home development. The analysis shows th a t: 1) The median age level f o r property owners is 52.78 years. Property owners with seasonal homes were s l i g h t l y older than those with permanent homes or no home development in the study area. 2) The head o f household f o r study area property owners is overwhelmingly male. 3) Property owners, in the m a jo rity o f cases, are married. The m arital status of the next larg e st group is widow or widower. Very few property owners are single or divorced. 4) The mean fam ily size fo r study area property owners is 3.24. 5) Family income is d i r e c t ly re la te d to type of home develop­ ment. Property owners w ith no home in the study area or seasonal homes have many more respondents in the higher income categories than property owners with permanent homes. In ad d itio n , c e rta in counties and townships have property owners showing higher fam ily income when compared to other counties or townships. CHAPTER IV TYPES OF HOME DEVELOPMENT Few people who li v e in Michigan have not, a t one tim e, v is it e d i t s northern lower peninsula. For many, the desire to become property owners was too strong to r e s is t . The area has much to o f f e r the vacationer as w ell as the permanent resid en t. Forests, streams, lakes, w i l d l i f e , small town l i v i n g , rural settings are a l l advantages offered property owners in northern Michigan. The a t tr a c tio n o f the area has led to the construction o f many seasonal homes as well as a p r o f it a b le market fo r undeveloped land. I t is the objective o f th is chapter to describe property owners by type of home development on t h e ir land. In a d d itio n , current place o f permanent residence w i l l be investigated fo r property owners with seasonal homes and no home development in the study area. Also, p r io r place of residence f o r permanent home owners w i l l be examined. Type o f Home Many property owners in the study area have some type o f liv in g quarters on t h e i r land. These liv in g quarters may be o f the seasonal type ( i . e . cabin) or a permanent fam ily residence. In the three study counties, taken as a whole, the o verall frequency'percentages are divided almost evenly between property owners with no home development (33.6% ), seasonal homes (36.2%) and permanent homes (30.2%). This r e s u lt , shown in Table 14, is quite in te re s tin g because i t shows th a t m u n ic ip a litie s now c o lle c t property taxes from a l l property owners but only have to provide year round services to less than one t h ir d . I t is not surprising to learn th a t the la rg e s t segment of property owners in the northern Michigan study area are seasonal home owners. 71 72 Table 14 TYPE OF HOME DEVELOPMENT IN NORTHERN MICHIGAN STUDY AREA (Frequency and Percentage D is trib u tio n ) Home Frequency Frequency Percentage No Home 668 33.6 Seasonal Home 721 36.2 Permanent Home 600 30.2 1,989 100.0 Total P rio r research has indicated th a t the highest concentration o f seasonal homes occurs in the Great Lakes region with Michigan ranking f i r s t in the nation in terms o f numbers o f seasonal homes (American Society o f Planning O f f i c i a l s , 1976). When individ ual counties are considered, as presented in Table 15, Otsego County has the larg e st percentage o f property owners with no home development in the study area (40.8%). Otsego also has the smallest number of property owners with seasonal homes (28.9%). The percentage of seasonal home owners in Crawford and Kalkaska is much higher at 40.6% and 52.4% resp e c tiv e ly . In terms of permanent home owners, there is not a l o t of d iffe re n c e in the three counties, although Kalkaska has the lowest percentage, 27.3%. S t a t is t ic s presented a t the bottom of Table 15, in d ica te a r e la t io n ­ ship ex ists between property owners o f c e rta in counties and type o f home development. A Chi-square value of 111.36622, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners between counties d i f f e r in terms of type o f home development. Other q u a lify in g s t a t i s t i c s , (Cramers V, Contingency C o e ffic ie n t, Uncertainty C o e ffic ie n t) in d ic a te the re la tio n s h ip 73 Table 15 TYPE OF HOME DEVELOPMENT, BY COUNTY Count Row Percentage Row Cumulative Percentage Home No Home Seasonal Home Permanent Home Row Total % of Total County Crawford 68 24.5 24.5 113 40.6 65.1 97 34.9 100.0 278 14.0 Kalkaska 97 20.3 20.3 250 52.4 72.7 130 27.3 100.0 477 24.0 Otsego 503 40.8 40.8 357 28.9 69.7 373 30.3 100.0 1,233 62.0 Column Total 668 33.6 720 36.2 600 30.2 1,988 100.0 % o f Total Chi-square = 111.36622 with 4 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .13603 Contingency C o e ffic ie n t = .22934 Uncertainty C o e ffic ie n t (asymmetric) = .02461 with Home dependent = .03042 with County dependent between property owners o f c e rta in counties and types o f home develop­ ment is moderate. S t a t is t ic s support the percentage d iffe re n c e s , d is ­ played in Table 15, th erefo re knowledge of type of home development w i l l help p re d ic t county of property lo c a tio n . When in d ivid u a l townships are considered, as presented in Table 16, the resu lts d i f f e r widely with very l i t t l e discernable p a tte rn . A ll three townships surveyed in Kalkaska county have a low percentage o f property owners with no home development and a l l three surveyed town­ ships in Otsego county have a high percentage of property owners with no home development. However, when seasonal homes are considered, townships seem independent and are not re la te d to t h e i r respective county. In Chester Township o f Otsego County, 47.5% of the property 74 Table 16 TYPE OF HOME DEVELOPMENT, BY TOWNSHIP Count Row Percentage Row Cumulative Percentage Home | No Home Seasonal Home South Branch 44 31.9 31.9 69 50.0 81.9 25 18.1 100.0 138 6.9 Grayling T27NR2W 11 32.4 32.4 10 29.4 61.8 13 38.2 100.0 34 1.7 Grayling T26NR4W 13 12.1 12.1 35 32.7 44.8 59 55.1 99.9 107 5.4 Orange 24 20.9 20.9 49 42.6 63.5 42 36.5 100.0 115 5.8 Blue Lake 60 21.7 21.7 158 57.0 78.7 59 21.3 100.0 277 13.9 G a rfie ld 12 14.5 14.5 43 51.8 66.3 28 33.7 100.0 83 4.2 419 39.9 39.9 294 28.0 67.9 337 32.1 100.0 1,050 52.8 Chester 48 47.5 47.5 48 47.5 95.0 5 5.0 100.0 101 5.1 Dover 37 44.0 44.0 15 17.9 61.9 32 38.1 100.0 84 4.2 668 33.6 721 36.2 600 30.2 1,989 100.0 Permanent Home Row Total % of Total Township Bagley Column Total % o f Total Chi-square = 211.14607 with 16 degrees of freedom S ig n ific a n t at .05 p r o b a b ility level Cramers V = .18731 Uncertainty C o e ffic ie n t (asymmetric) = .04963 with Home dependent = .03510 with Township dependent 75 owners own seasonal homes whereas in Dover township, another Otsego county township, only 17.9% of the property owners own seasonal homes. Also in Chester township, only 5.0% of the property owners have permanent homes compared to 38.1% fo r Dover township. Other townships with high percentages o f seasonal home owners include Blue Lake 57.0% and G a rfie ld 51.8% (Kalkaska county), and South Branch 50.0% (Crawford county). In terms of permanent home owners, both South Branch and Blue Lake have low lev els (18.1% and 21.3%), resp ective ly. S t a t is t ic s presented a t the bottom of Table 16, in d ic a te a r e la t io n ­ ship exists between property owners of c e rta in townships and type o f home development. A Chi-square value of 211.14607, s ig n if ic a n t at the .05 p r o b a b ility le v e l, implies property owners between townships d i f f e r in terms o f type o f home development. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in dicate the' re la tio n s h ip between property owners o f c e rta in townships and type of home development is moderate. S t a t is t ic s support percentage differences displayed in Table 16, th e re fo re , knowledge of type of home development w i l l help p red ict county of property lo cation. Seasonal Home Usage As shown previously, seasonal home owners make up almost o n e-third of northern Michigan study area property owners. The economic impact seasonal home owners have on northern Michigan communities is d ir e c t ly related to the value of t h e i r property and the amount and length of v i s i t s to t h e i r seasonal home. The market value of the seasonal home is re fle c te d in the assessed value of property owned, and th e re fo re , property taxes which in turn have a d ir e c t economic impact on local schools and governments. A second d ir e c t monetary impact resu lts from 76 purchases made in the area where the seasonal home is located. In d ir e c t impacts also re s u lt as purchases make t h e i r way through the economy. Input-ouput models are designed to obtain estimates o f in d ir e c t impacts (Is a r d and Langford, 1971). As seasonal home owners spend more time at t h e i r second home, they are more l i k e l y to spend larg e r amounts o f money in th at area. Food and gasoline are ju s t two commodities th a t require frequent renewal. Therefore, i t is theorized th a t the longer a seasonal home owner stays at his second home, the greater the d ir e c t and in d ir e c t monetary impacts there are. The average length o f stay per seasonal home owner per v i s i t in the stucty area is 10.7 days. However, th is fig u re is somewhat misleading as some seasonal home owners may stay up to six months, others ju s t a day. A Skewness value of 5.283 and a Kurtosis value of 29.84 indicates th a t the curve is peaked and generally to the l e f t o f the mean in d ic a tin g the average stay is usually less than 10.7 days. Indeed, th is is the case as a median value o f 3.39 and a mode o f 3.0 indicates th a t the usual t r i p to the seasonal home consists of a weekend v i s i t . The average number o f v i s i t s a seasonal home owner makes a year is 13.89 with summer being the time of heaviest use. An average of 27.27 days are spent at the seasonal home in the summer compared to 10.58 days in the f a l l , 8 .5 8 in spring, and only 7.13 days in the w in te r. As shown in Table 17, the greatest economic impact occurs in the summer when seasonal home owners are more apt to use t h e ir second home and stay a longer length of time. Results from th is study contrast with some previous research. Marans and Wellman reported in a 1976 study of seasonal homeowners in northern Michigan an average length of stay during the summer of 60 days 77 (Marans & Wellman, 1976). However, another study o f seasonal residence in northern Michigan reported an average length of stay o f approximately 22 days (V ertree s, 1967). Table 17 NUMBER OF DAYS SPENT AT THE SEASONAL HOME, BY SEASON Mean Mode Median 10.58 10.0 7.238 4.428 16.321 Winter 7.13 0 3.55 2.578 9.752 Spring 8.58 0 5.95 3.493 18.247 Summer 27.27 30.0 19.915 1.323 .768 Annually 53.56 F a ll — Skewness Kurtosis — — Region o f Present Residence Michigan consists o f 83 counties which are s p l i t in to 14 d if f e r e n t economic and planning regions. These regions plan not only f o r t h e ir permanent residents, but also fo r seasonal v i s i t o r s . Seasonal v is i t o r s , es p e c ia lly those owning property, have the p o te n tia l fo r becoming per­ manent residents. Therefore, local planners should know where t h e ir seasonal v is it o r s c a ll home. tourism, highway planning, In a d d itio n , s ta te agencies dealing with and recreation also b e n e fit when the o rig in o f seasonal v is ito r s is known. P rio r research (Marans and Wellman, 1978; American Society of Planning O f f i c i a l s , 1970) has found th a t large percentages o f seasonal home owners have t h e i r primary residence in metropolitan areas. For Michigan th is tran slates to the southeastern Michigan area, s p e c if ic a lly Wayne, Oakland, and Macomb counties. Therefore, fo r th is study, Region 15 78 was created which consists only o f Wayne, Oakland, and Macomb counties. Also, a Region 16 was created consisting o f only the study area counties Otsego, Crawford, and Kalkaska. Regions 1, 9, and 10 t o t a l s , th e re fo re , r e f l e c t the absence of removed counties. Regions 17, 18 and 19 were also created consisting o f the state o f Ohio, F lo rid a , and other s t a t e s , re s p e c tiv e ly . The 14 s ta te economic and planning regions plus the newly created regions are l is t e d in Appendix B. As expected, the m a jo rity of seasonal home owners, 52.0%, have t h e i r permanent residence in the metro­ p o lita n D e tr o it area, Region 15. A large percentage o f the property owners with no home development in the study area also o rig in a te in Region 15, 55.6%. Region 1 which consists of the other counties surround­ ing Wayne, Oakland, and Macomb counties accounts fo r 6.0% of the property owners with no home in the study area. Therefore, the impact on the study areas seasonal economy is d i r e c t ly tie d to the economy o f south­ eastern Michigan. The high aggregate level o f property ownership by residents of southeastern Michigan is not unusual when comparing regional populations w ith in the s ta te . Over 45% o f Michigan's population li v e in three counties (Wayne, Oakland, Macomb, Region 15 fo r th is study). An ad ditional 6.4% li v e in other counties (Region 1 ), which surround Wayne, Oakland, and Macomb counties (Michigan S t a t i s t i c a l A bstract, 1979). It s till seems as though more study area property owners li v e in southeastern Michigan than comparison by regional populations would in d ic a te , however, th is can be fu r th e r explained by examining r e la t iv e fam ily median income le v e ls . Higher incomes are generally found w ithin the southeastern Michigan area than any other area w ith in the state (Michigan S t a t i s t i c a l A b s tra ct, 1979). Therefore, more disposable income would make property ownership e a s ie r fo r southeastern Michigan residents. 79 One in te r e s tin g s t a t i s t i c is th a t 7.7% of the permanent residents in the study area (Region 1 6 ), also own at le a s t one additional piece of property in the area w ith no home development on t h e i r land. In a d d itio n , 3.5% o f the permanent residents in the study area also own a seasonal home in the area. Ohio residents (Region 17 ), also have a share in study area property as they own 1.8% o f the property with no home development and 3.5% o f the seasonal homes in the study area. Ohio is followed clo se ly by F lo rid a residents (Region 1 8 ), who own 1.4% of the property with no home development and 1.9% o f the seasonal homes. Residents in other states (Region 19) account f o r 9.4% o f the property owners with no home development and 3.1% o f the seasonal homes in the study area. A complete breakdown on ownership by region can be found in Table 18. S t a t i s t i c s , presented a t the bottom o f Table 18, in dicate a r e la t io n ­ ship e x is ts between property owners with d is s im ila r types of home develop­ ment and region o f permanent residence. A Chi-square value of 67.69007 s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners with d is s im ila r types o f home development d i f f e r in terms of region of permanent residence. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e f f ic ie n t ) , in d ic a te the re la tio n s h ip between property owners with d is s im ila r types o f home development and region o f permanent residence is strong. These s t a t i s t i c s support percentage differences displayed in Table 18, th e re fo re , knowledge o f property owners region o f permanent residence w i l l g r e a tly help p re d ic t type of home development on the property owners land in the study area. Table 18 PLACE OF PERMANENT RESIDENCE OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS WITH SEASONAL HOME DEVELOPMENT OR NO HOME DEVELOPMENT Count Row Percentage Row Cumulative Percentage CTYNOW Reg. 1 Reg. 2 Reg. 3 Reg. 4 Reg. 5 Reg. 6 Reg. 7 Reg. 8 Reg. 9 Reg. 10 No Home 37 6.0 6.0 10 1.6 7.6 6 1.0 8 .6 0 0 8.6 13 2.1 10.7 13 2.1 12.8 32 5.2 18.0 15 2 .4 20.4 5 .8 21.2 11 1.8 23.0 Seasonal Home 45 6.6 6 .6 11 1.6 8.2 23 3.4 11.6 3 .4 12.0 31 4.5 16.5 46 6.7 23.2 48 7.0 30.2 21 3.1 33.3 0 0 33.3 9 1.3 34.6 Column Total 82 6.3 21 1.6 29 2.2 3 .2 44 3.4 59 4.5 80 6.1 36 2.8 5 .4 20 1.5 581,300 273,000 6 .4 3.0 6 .4 9 .4 473,000 5.2 14.6 227,000 2.5 17.1 577,000 6.4 23.5 398,000 4.4 27.9 740,000 8.2 36.1 699,000 7.7 43.8 87,400 1.0 44.8 174,800 1.9 46.7 Home % of Total Statewide Population (1975) Table 18 - Continued Count Row Percentage Row Cumulative Percentage Row Total Reg. 11 Reg. 12 Reg. 13 Reg. 14 Reg. 15 Reg. 16 Reg. 17 Reg. 18 Reg. 19 % of Total Home No Home 0 0 23.0 0 0 23.0 4 .6 23.6 3 .5 24.1 345 55.6 79.7 48 7.7 87.4 11 1.8 89.2 9 1.4 90.6 58 9.4 100.0 620 47.6 Seasonal Home 2 .3 34.9 2 .3 35.2 0 0 35.2 5 .7 35.9 355 52.0 87.9 24 3.5 91.4 24 3.5 94.9 13 1.9 96.8 21 3.1 99.9 683 52.4 % of Total 2 .1 2 .1 4 .3 8 .6 700 53.7 72 5.5 35 2.7 22 1.7 79 6.1 1,303 100.0 Statewide Population (1975) 54,000 .6 47.3 176,000 1.9 49.2 93,000 1.0 50.2 318,000 4,158,700 3.5 45.9 53.7 99.6 29,800 .3 99.9 N/A N/A N/A 9,060,000 Column Total Source (Michigan S t a t is t ic a l Abstract) Chi-square = 67.69067 with 13 degrees of Freedom Cramers V = S ig n ific a n t at .05 p ro b a b ility le v e l. .22784 Uncertainty C o e ffic ie n t (asymmetric) = .04228 with Home dependent = .01660 with Region dependent *Due to rounding, percentage to ta ls may not add up to 100. 100.0 82 Region o f P rio r Residence The preceding analysis placed strong emphasis on the p o te n tia l fo r fu tu re settlement in the study area by seasonal v is it o r s from the metro­ p o lita n D e tro it area. Further analysis shows th a t a great deal of th a t p o te n tia l is being r e a liz e d . Table 19 shows th a t almost on e-third (28.5%) of the permanent residents in the study area used to li v e in the metropolitan D e tro it area. An ad ditional 5.5% o r ig in a ll y liv ed in Region 1, implying the influence o f southeastern Michigan already plays an important ro le in decision making in the study area. influence may show up by a demand fo r more public service. This P rio r re ­ search indicates rural resid en ts, who relocated from a metropolitan area, are l i k e l y to demand an increasing level o f public services. This is due in most p a rt to the lev el o f service provided residents o f metropolitan areas and the desire to maintain th a t level even when relocating to rural areas (American Society of Planning O f f i c i a l , 1976). The next chapter w i l l explore th is trend in greater d e t a i l . Somewhat surprising is th a t only 36.5% o f the study areas permanent residents used to liv e in the study area p r io r to building a permanent home there. Therefore, almost tw o -th ird s o f the present permanent residents have migrated in to the area. One reason fo r th is in migration can be tie d to the median age level o f the study area residents. The high median age level shown in Table 2 indicates study areas permanent r e s i­ dents are near retirem ent age and thus, may not be dependent on the local economy fo r a job. At the present time jobs are not abundant in northern Michigan. Scattered manufacturing and some in d u s tr ia l development provide l i t t l e in terms of an economic base. Severe competition ex is ts fo r the few jobs a v a ila b le and many would-be permanent residents have to fin d employment 83 Table 19 REGION OF PRIOR PERMANENT RESIDENCE (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage Cumulative Frequency Percentage Region 1 30 5.5 5.5 Region 2 7 1.3 6 .8 Region 3 14 2.6 9 .4 Region 4 3 .5 9.9 Region 5 16 2.9 12.8 Region 6 21 3.9 16.7 Region 7 30 5.5 22.2 Region 8 9 1.7 23.9 Region 9 10 1 .8 25.7 Region 10 16 2.9 28.6 Region 11 2 .4 29.0 Region 12 0 0 29.0 Region 13 2 .4 29.4 Region 14 3 .5 29.9 Region 15 155 28.5 58.4 Region 16 198 36.5 94.9 Region 17 7 1.3 96.2 Region 18 0 0 96.2 Region 19 20 3.7 99.9 543 100.0 CTYBEFOR Total *Due to rounding, percentage to ta ls may not add up to 100.0 84 downstate in the more in d u s tr ia l counties (G a lin , 1976). This trend is not viewed as abating in the near fu tu re . Summary This chapter explored the types o f home development owned by northern Michigan study area property owners. In a d d itio n , usage of seasonal residence, p r io r lo cation o f residence fo r permanent home owners, and present place of residence fo r seasonal home owners and property owners with no home in the study area was also examined. The analysis indicated th at: 1) There are s l i g h t l y more seasonal homes in the study area than permanent homes. Kalkaska county has the highest percentage of seasonal home owners followed by Crawford and Otsego. 2) The average length o f stay, per v i s i t , f o r seasonal home owners is 10.7 days with summer being the high use season. F a ll was the second most popular season o f use fo r seasonal home owners. 3) Total use per year averaged 53.56 days. A large percentage of seasonal home owners and property owners with no home development in the study area, 52.0% and 55.6%, re s p e c tiv e ly , have t h e i r permanent residence in the southeastern Michigan, m etropolitan D e tr o it area. 4) Only s l i g h t l y over o n e -th ird o f the permanent residents in the study area liv e d in the area p r io r to locating t h e ir permanent residence th ere. Almost o n e-third o f the study area's permanent residents previously liv e d in the south­ eastern Michigan, m etropolitan D e tr o it area. CHAPTER V INITIAL PROPERTY PURCHASE Acquiring the r ig h t to use property is usually not an easy under­ taking. In the case of acquisitio n in fee simple, t i t l e searches are conducted, c r e d it references s c ru tin iz e d , mineral rig h ts decided upon and so on. Leasing is no simple matter e i t h e r , as anyone who has even read through a complete rental contract knows. Obtaining the r ig h t to use property is time consuming and c o s tly , th e re fo re , the rewards of ownership must o ffs e t costs incurred. In th is section property owners are analyzed as to why and how they s e ttle d on a p a r t ic u l a r piece of land along with future intentions on s e llin g t h e i r property. Method o f Acquisition Generally i t is assumed th a t almost a l l property is acquired through o u trig h t purchase, however, there are some exceptions. The most recognized has to do with in h erita n ce , many parcels o f land are handed down through generations. is through leasing. Another method of land a c q u isitio n Normally a lease is not thought of as a c q u isitio n of land, but in northern Michigan there is a unique type o f lease arrangement th a t has many aspects o f ac q u is itio n in fee simple. Con­ sumers Power Company owns many acres, p rim a r ily along r i v e r s , which i t leases out on a long-term basis (some leases may run fo r 99 y e a rs ). Holders o f the leases can b u ild on the land and have rig h ts s im ila r to other property owners, including payment of property taxes on the assessed value o f the property. Table 20 shows th a t in the study area, 91.1% o f the property owners acquire t h e ir land through o u trig h t purchase. 85 Inheritance 86 accounts fo r 6.3% o f property a c q u isitio n and leasing only 1.2%. Even though only 1.4% of property ac q u is itio n is acquired through other means, some of the other methods o f a c q u is itio n given are quite c o lo rfu l such as, "won i t in a poker game" or "crap shoot". When in ­ dividual counties and townships are considered, no substantial changes occur from th at displayed in Table 20. Reason fo r Acquisition As previously mentioned, ac q u isitio n o f a piece of property requires not only a monetary investment but usually a large time investment. Therefore, there should be good reasons f o r land a c q u is itio n . An in d ivid u a l can probably think of f if t e e n d i f f e r e n t reasons fo r land acquisitions w ith in a m atter of minutes. L ite r a tu r e review enabled our survey to concentrate on the major reasons fo r property ac q u isitio n in northern Michigan. In a previous study, with Kalkaska county as the sample area, i t was found th a t "Hunting and Fishing" was the major reason f o r property a c q u isitio n by 33.8% of the absentee landowners (who owned over 10 ac re s ). This was followed clo sely by 30.1% who purchased the land as a retirem ent s i t e (V e rtre e s , 1967). Other research id e n t if ie d major reasons fo r property ac q u is itio n as a means to get out of the c i t y and escape urban problems (G a lin , 1976). Investment was also considered an important reason fo r property ac q u is itio n (American Society o f Planning O f f i c i a l s , 1976). Based on the resu lts from previous research, the categories used in th is study to d e lin e ate major reasons fo r property a c q u is itio n , were formulated. As presented in Table 21, the la rg e s t percentage of property in the study area is acquired fo r an investment or retirem ent home (45.1%). Id e a lly investment and retirem ent homes should be two categories, but 87 Table 20 NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS METHODS OF PROPERTY ACQUISITION Frequency Frequency Percentage Frequency Cumulative Percentage Purchased 1,818 91.1 91.1 In h e rite d 125 6.3 97.3 Leased 24 1.2 98.5 Other 28 1.4 100.0 Total 1,995 100.0 Acqui re Table 21 MAJOR REASON FOR PROPERTY ACQUISITION (Frequency and Percentage D is tr ib u tio n ) Frequency Frequency Percentage Frequency Cumulative Percentage Reason Investment or Retirement Home 866 45.1 45.1 Get Out of City 283 14.8 59.9 Hunt and Fish 405 21.1 81.0 97 5.1 86.1 Other 267 13.9 100.0 Total 1,918 100.0 In h erite d 88 due to an e r r o r in the questionnaire, they became one. There was an attempt made to separate investment and retirem ent homes in to two ca te­ gories. Some respondents would cross out investment and leave retirem ent home, and v ic e-v ersa, as t h e i r major reason fo r buying property. When separate categories were set up, by counting the number of respondents who a lte re d the category "Investment or Retirement Home", retirem ent homes accounted f o r 4.1% of property purchases and investment was the main reason fo r 2.7%. I f these proportions were the same fo r a l l property owners selecting category "Investment and Retirement Home", then th is category could be broken down in to the following two categories; "Investment" 17.9% and "Retirement Home" 27.1%. The recreation al pursuit of hunting and fishin g was the major reason fo r property purchase by 21.1% of the study areas property owners. Also, the category "Other" contained many responses f o r skiing as the major reason fo r property a c q u is itio n . The recreational p o ten tial in the study area then is probably a strong a ttr a c tio n fo r property a c q u isitio n s . In a d d itio n , 14.8% o f the study area's property owners f e l t the need to get out of a c i t y and in to a more rural atmosphere was the major reason f o r i n i t i a l property a c q u is itio n . When in d ivid u a l counties are considered, many substantial differences are noted. As shown in Table 22, 40.8% o f the property owners in Crawford county checked investment or retirem ent home as t h e i r major reason fo r property ac q u is itio n whereas in Otsego county 47.1% indicated investment or retirem ent home as the major reason. Also, in Kalkaska and Crawford counties the percentage of property owners who indicated hunting and fis h in g as t h e i r major reason (30.1% and 27.7%, r e s p e c tiv e ly ), was much higher than in Otsego county (16.2%). Correspondingly in Otsego, 89 Table 22 MAJOR REASON FOR PROPERTY ACQUISITION, BY COUNTY Reason Investment or Retirement Home Get Out of City Hunt and Fish In h e rite d Other Crawford 109 40.8 40.8 39 14.6 55.4 74 27.7 83.1 13 4.9 88.0 32 12.0 100.0 267 13.9 Kalkaska 196 42.8 42.8 61 13.3 56.1 138 30.1 86.2 20 4.4 90.6 43 9.4 100.0 458 23.9 Otsego 561 47.1 47.1 182 15.3 62.4 193 16.2 78.6 64 5.4 84.0 192 16.1 100.1 1,192 62.2 Column Total % o f Total 866 45.2 282 14.7 405 21.1 97 5.1 267 13.9 1,917 100.0 Count Row Percentage Row Cumulative Percentage Row Total % o f Total County | Chi-square = 74.79379 with 8 degrees o f freedom S ig n ific a n t a t .05 level Cramers V = .11401 Uncertainty C o e ffic ie n t (asymmetric) = .02111 with County dependent = .01185 with Reason dependent *Due to rounding, percentage to ta ls may not add up to 100. more property owners indicated some other reason was responsible fo r property a c q u isitio n (16.1%) than in e it h e r Kalkaska or Crawford counties (9.4% and 12.0%, re s p e c tiv e ly ). This may in d ica te th a t other recreation pursuits played a la rg e r role in property a c q u is itio n f o r Otsego county than fo r Crawford or Kalkaska counties. There are more ski areas operating in Otsego county than in Crawford or Kalkaska combined. S t a t i s t i c s , presented a t the bottom of Table 22, in d ic a te a r e la t io n ­ ship exists between property owners o f c e rta in counties and s p e c ific reasons f o r property a c q u is itio n . A Chi-square value of 74.79379, s ig n if ic a n t a t 90 the .05 p r o b a b ility le v e l, implies property owners o f c e rta in counties d i f f e r in terms o f major reasons fo r property a c q u is itio n . Other q u a l i f y ­ ing s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in dicate the r e la t io n ­ ship between property owners o f c e rta in counties and major reasons fo r property ac q u is itio n is moderate. These s t a t i s t i c s support the percentage d iffe re n c e displayed in Table 22, th e re fo re , knowledge of major reasons fo r property ac q u isitio n w i l l help p re d ic t county of property lo c a tio n . When in divid ual townships are considered, empirical evidence, reported in ta b le 23, shows some ra th e r in te re s tin g re s u lts . Six out of the nine townships s u b s ta n tia lly exceed the o v erall percentage mean fo r property owners whose major reason f o r property acq u isitio n was to have a place to hunt and f is h . In Chester township 42.5% of the property owners major reason fo r property a c q u isitio n was hunt or f is h . s l i g h t l y smaller a t 39.4%. In South Branch, the number was Even in Orange township, which ranked sixth in percentage f o r the category, 28.6% of the property owners indicated th a t was t h e ir major reason f o r property a c q u is itio n . The influence of Bagley township, with i t s large number o f respondents and only 13.5% of i t s property owners checking the "Hunt or Fish" category, tended to d e fla te the o v erall percentage d is tr ib u tio n when compared to the other townships percentage d is tr ib u tio n s . Therefore, recreation pursuits as a major reason f o r i n i t i a l property ac q u is itio n may be thought of as a potent force fo r development in many areas o f northern Michigan. S t a t i s t i c s , presented a t the bottom o f Table 23, in d ic a te a r e l a ­ tionship' ex is ts between property owners of c e rta in townships and the major reason fo r property a c q u is itio n . A Chi-square value of 184.25446, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners of 91 Table 23 MAJOR REASON FOR PROPERTY ACQUISITION, BY TOWNSHIP Reason Investment or Retirement Home Get Out o f C ity Hunt and Fish South Branch T25NR2W 44 33.3 33.3 17 12.9 46.2 52 39.4 85.6 7 5.3 90.9 12 9.1 100.0 132 6.9 Grayling T27NR2W 9 26.5 26.5 4 11.8 38.2 13 38.2 76.4 3 8.8 85.2 5 14.7 100.0 34 1.8 Gray!i ng T26NR4W 56 54.9 54.9 19 18.6 73.5 9 8 .8 82.3 3 2.9 85.2 15 14.7 99.9 102 5. j Orange T26NR7W 48 42.9 42.9 - 12 10.7 53.6 32 28.6 82.2 6 5 .4 87.6 14 12.5 100.0 112 6 .8 Count Row Percentage Row Cumulative Percentage Row Total In h erite d Other % o f Total Township | - Blue Lake T28NR5W 113 42.5 42.5 39 14.7 57.2 81 30.4 87.6 11 4.1 91.7 22 8.3 100.0 266 13.9 G arfield T25NR7W 34 43.6 43.6 10 12.8 56.4 24 30.8 87.2 3 3 .8 91.0 7 9.0 100.0 78 4.1 Bagley T30NR3W 503 49.4 49.4 161 15.8 65.2 137 13.5 78.7 49 4 .8 83.5 168 16.5 100.0 1,018 53.1 Chester T29NR2W 33 35.1 35.1 7 7.4 42.5 40 42.5 85.0 8 8.5 93.5 6 6.4 99.9 94 4.9 Dover T31NR2W 26 31.7 31.7 14 17.1 48.8 17 20.7 69.5 7 8.5 78.0 18 21.9 99.9 82 4.3 886 45.1 283 14.7 405 21.1 97 5.1 267 13.9 1,918 100.0 Column Total % o f Total Chi-square = 184.25446 w ith 32 degrees of freedom S ig n ific a n t a t .05 p ro b a b ility le v e l. Cramers V = .12653 Uncertainty C o e ffic ie n t (asymmetric) = .03157 with Township dependent = .3097 w ith Reason dependent *Due to rounding, percentage to ta ls may not equal 100. 92 c e rta in townships d i f f e r in terms of major reason fo r property ac q u isi­ tio n . Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e f f ic ie n t ) , in d ic a te the re la tio n s h ip between property owners of c e rta in townships and the major reason fo r property ac q u isitio n is moderate. S t a t is t ic s support the percentage d iffe re n c e , displayed in Table 23, th e re fo re , knowledge o f major reason fo r property ac q u is itio n w i l l help p re d ic t townships o f property lo c a tio n . When type of home development is considered, some very discernable trends are noted. As shown in Table 24, more land without homes were purchased as an investment or retirem ent home s it e (56.3%) than e it h e r property w ith permanent homes (44.6%) or seasonal homes (35.3%). Property owners with seasonal homes purchase the land f o r hunting and fis h in g (38.4%) much more than property owners with no home (17.2%) or permanent homes (5.7% ). Property owners with permanent homes include a much higher percentage who purchase the property to get out of the c it y (25.6% ), than property owners w ith no home in the study area (6.3%) or seasonal homes (13.3%). S t a t i s t i c s , presented a t the bottom o f Table 24, in d ica te a r e l a t i o n ­ ship ex is ts between property owners with d is s im ila r types o f home devel­ opment and major reasons f o r property a c q u is itio n . A Chi-square value o f 462.36731, s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners with d is s im ila r types o f home development d i f f e r in terms o f major reasons fo r property a c q u is itio n . Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the re la tio n s h ip between property owners with d is s im ila r types o f home development and major reasons fo r property ac q u isitio n is strong. 93 Table 24 MAJOR REASON FOR PROPERTY ACQUISITION BY TYPE OF HOME DEVELOPMENT Reason Count Row Percentage Row Cumulative Percentage Investment or Reti rement Home Get Out o f C ity Hunt and Fish In h erite d Other No Home 364 56.3 56.3 41 6.3 62.6 111 17.2 79.8 48 7.4 87.2 83 12.8 100.0 647 34.0 ' Seasonal Home 239 35.3 35.3 90 13.3 48.6 260 38.4 87.0 32 4 .7 91.7 56 8.3 100.0 677 35.6 Permanent Home 258 44.6 44.6 148 25.6 70.2 33 5.7 75.9 15 2.6 78.5 125 21.7 100.1 579 30.4 Column Total % o f Total 861 45.2 279 14.7 404 21.2 95 5.0 264 13.9 1,903 100.0 Row Total % of Total Home | Chi-square = 462.36731 w ith 8 degrees o f Freedom, S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .28347 Uncertainty C o e ffic ie n t (asymmetric) = .10567 with Home dependent = .07342 with Reason dependent *Due to rounding, percentage to ta ls may not equal 100. Information Sources th a t Lead to Acquisition The question o f how property owners f i r s t learn o f an a v aila b le piece o f land is important. Obviously i f the p e rfect information assump­ tio n o f neoclassical economics were v a lid , every prospective buyer would know o f every a v a ila b le piece of property. However, perfect information does not e x is t thereby necessitating a market structure which allows buyers to contact re a lto rs or read advertisements in newspapers/magazines to fin d out what is a v a ila b le . However, only 33.6% 94 o f property owners in the study area f i r s t learned of t h e i r property through these sources (see Table 2 5 ). More important sources of i n f o r ­ mation are r e la tiv e s and friends who may know o f a v a ila b le land parcels. The r e la t iv e and frie n d information connection accounts fo r 45.9% of property acquired in the study area. I t is important to c le a r up some confusion concerning the category 'Other' since i t accounts fo r 20.6% o f i n i t i a l information sources. Question A3 (see Appendix A) l i s t s 'Other' and has a space fo r explanation. I t should be noticed th a t there is no category 'Friends' in question A3. A ll responses fo r 'Friends' came from respondents who checked 'Other' and wrote in fr ie n d . Many respondents checked 'Other' but did not id e n t if y what th a t meant to them. Therefore, the to ta l percentage of property owners who f i r s t learned about t h e i r property from friends and r e la tiv e s l i k e l y is higher than what is a c tu a lly recorded. in a conservative bias towards the category 'F rie n d s '. This resu lts Another response i d e n t if ie d often in the 'Other' category was " ju s t drivin g through the area." However, these responses were not numerous enough to j u s t i f y creation of a new category. Table 25 SOURCES OF INFORMATION THAT LEAD TO PROPERTY PURCHASES (Frequency and’ Percentage D is trib u tio n ) Absolute Frequency Frequency Percentage Cumulative Frequency Percentage Learn Newspapers and Magazines 174 8.9 8.9 Real Estate Salespersons 484 24.7 33.6 Relatives 535 27.3 60.9 Friends 364 18.6 79.5 Other 404 20.6 100.1 Total 1,961 100.1 95 When in d iv id u a l counties are considered, in te re s tin g trends begin to develop. As shown in Table 26, in Crawford county the r e la t iv e and frie n d connection s t i l l accounts fo r a high proportion of f i r s t informa­ tio n concerning a v a ila b le property (44.6%) and d if f e r s l i t t l e o verall percentage d is t r ib u t io n . from the The magazine/newspaper/real estate salesperson, connection increased to 38.4% in Crawford county, mostly a t the expense o f the category, 'O th e r'. Table 26 SOURCES OF INFORMATION THAT LEAD TO PROPERTY PURCHASES, BY COUNTY Learn Count Row Percentage Row Cumulative Percentage Newspaper Real Estate or Magazine Salesperson Relatives Friends Other % of Total County Crawford 24 8.7 8.7 82 29.7 38.4 77 27.9 66.3 46 16.7 83.0 47 17.0 100.0 276 14.1 Kalkaska 36 7.6 7.6 54 11.4 19.0 157 33.2 52.2 115 24.3 76.5 111 23.5 100.0 473 24.1 Otsego 113 9 .3 9 .3 348 28.7 38.0 301 24.8 62.8 203 16.8 79.6 246 20.3 99.9 1,211 61.8 Column Total 173 8 .8 484 24.7 535 27.3 364 18.6 404 20.6 1,960 100.0 Chi-square = 69.28935 w ith 8 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .13292 Uncertainty C o e ffic ie n t (asymmetric) = .02123 with County dependent = .01253 with Learn dependent *Due to rounding, percentage to ta ls may not equal 100. 96 In Kalkaska county, the r e la t iv e and frie n d connection accounts fo r 57.5% o f f i r s t inform ation, s u b s ta n tia lly higher than the overall percen­ tage d is t r ib u t io n . The newspapers/magazines, and real estate sales­ person connection declined to 19% fo r f i r s t information sources. Obviously, the r e l a t i v e and frie n d connection is much more important in learning of a v a ila b le property in Kalkaska county than t r a d it io n a l market information sources. S t a t is t ic s presented a t the bottom of Table 26, in dicate a r e la t io n ­ ship ex is ts between property owners of c e rta in counties and information sources o f property a v a i l a b i l i t y . A Chi-square value of 69.28935, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners of c e rta in counties d i f f e r in terms o f information sources of property a v a ila b ility . Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners of c e rta in counties and information sources of property a v a i l a b i l i t y is weak. S t a t i s t i c s support the percentage d iffe re n c e s , displayed in Table 26, however, due to the weak nature o f the s t a t i s t i c a l r e la t io n ­ ship, knowledge of how a property owner learned o f property a v a i l a b i l i t y w i l l only s l i g h t l y help p re d ic t the county where his property is located. When in d ivid u a l townships are considered, the trends th a t surfaced in in d iv id u a l counties above (more re lia n c e on friends and r e la tiv e s f o r inform ation) is fu r th e r supported. I t seems th a t a l l Kalkaska townships studied show a large number o f property owners who f i r s t learned o f t h e i r property through friends or r e la tiv e s and a s u b s ta n tia lly sm aller number who learned o f t h e i r property through tr a d itio n a l market sources. Obviously the frien ds and r e la t iv e s information connection is much more important in Kalkaska county than e it h e r Crawford or Otsego counties. 97 The only other substantial deviation from the o verall percentage d is tr ib u tio n was noted in Grayling T27NR2W township (Crawford county) where 55.8% o f property owners f i r s t learned o f t h e ir property through newspapers/magazines/real estate salespeople compared to only 29.4% who use the r e la t iv e and frie n d connection. S t a t i s t i c s , presented a t the bottom of Table 27, in d ica te a r e l a ­ tionship ex ists between property owners o f c e rta in townships and in f o r ­ mation sources o f property a v a i l a b i l i t y . A Chi-square value of 109.37203, s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners of c e rta in townships d i f f e r in terms of information sources o f property a v a ila b ility . Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners o f c e rta in townships and information sources of property a v a i l a b i l i t y is moderate. S t a t is t ic s support the percentage differences displayed in Table 27, th e re fo re , knowledge of information sources o f property a v a i l a b i l i t y w i l l help p re d ic t township o f property lo catio n . When type of home development is considered, in Table 28, with respect to information sources leading to property a c q u is itio n , one sharp differen ce is noted. Property owners with a seasonal home r e lie d more heavily on r e la tiv e s and friends to fin d out about t h e ir property (52.6%), than e it h e r property owners with no home in the study area (44.1%) or permanent homes, (39.8%). Property owners with no home in the study area and permanent homes both r e lie d more heavily on the tr a d itio n a l market sources (37.5% and 37.3%, resp ective ly) to f i r s t learn of t h e i r property than did property owners with seasonal homes (26.3%). Results in d ic a te , though, th a t the r e la t iv e and frie n d connection is s t i l l the most important source fo r learning o f a v a ila b le property no matter what type of home development. 98 Table 27 SOURCES OF INFORMATION THAT LEAD TO PROPERTY PURCHASES BY TOWNSHIP Learn | Count Row Percentage Newspaper or Row Cumulative Magazine Salesperson Percentage Relatives Friends Other Row Total % o f Total Township South Branch T25NR2W 11 8.1 8.1 38 28.1 36.2 45 33.3 69.5 24 17.8 87.3 17 12.6 99.9 135 6.9 Grayling T27NR2W 8 23.5 23.5 11 32.3 55.8 6 17.6 73.4 4 11.8 85.2 5 14.7 99.9 34 1.7 Grayling T26NR4W 6 5.5 5.5 33 30.5 36.0 26 24.1 60.1 18 16.7 76.8 25 23.1 99.9 108 5.5 Orange T26NR7W 9 7.9 7.9 16 14.0 21.9 48 42.1 64.0 19 16.7 80.7 22 19.3 100.0 114 5 .8 Blue Lake T28NR5W 23 8.3 8.3 29 10.5 18.8 81 29.2 48.0 72 26.0 74.0 72 26.0 100.0 111 G a rfie ld T25NR7W 4 4.9 4.9 9 11.1 16.0 28 34.6 50.6 24 29.6 80.2 16 19.7 99.9 81 4.1 Bagley T30NR3W 97 9.4 9 .4 305 29.5 38.9 245 23.7 62.6 172 16.6 79.2 216 20.9 100.1 1,035 52.8 Chester T29NR2W 12 12.6 12.6 23 24.2 36.8 29 30.5 67.3 16 16.8 84.1 15 15.8 99.9 95 4 .8 Dover T31NR2W 4 4.9 4.9 20 24.4 29.3 27 32.9 62.2 15 18.3 80.5 16 19.5 100.0 82 4.2 174 8.9 484 24.7 535 27.3 364 18.6 404 20.6 1,961 100.0 Column Total % of Total 14.1 Chi-square = 109.37203 with 32 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .11808 Uncertainty C o e ffic ie n t (asymmetric) = .01800 with Township dependent = .01860 with Learn dependent *Due to rounding, percentage to ta ls may not equal 100. 99 Table 28 SOURCES OF INFORMATION THAT LEAD TO PROPERTY PURCHASES BY TYPE OF HOME DEVELOPMENT Learn I Count Row Percentage Row Cumulative Percentage Newspaper or Real Estate Magazine Salesperson Row Total Relatives Friends Other % of Total Home | No Home 76 11.7 11.7 167 • 25.8 37.5 190 29.4 66.9 95 14.7 81.6 119 18.4 100.0 647 33.2 Seasonal Home 57 8.0 8.0 130 18.3 26.3 212 29.9 56.2 161 22.7 78.9 150 21.1 100.0 710 36.5 Permanent Home 40 6.8 6 .8 180 30.5 37.3 130 22.0 59.3 105 17.8 77.1 136 23.0 100.1 591 30.3 173 8.9 477 24.5 532 27.3 361 18.5 402 20.6 1,948 100.0 Column Total Chi-square = 59.29720 w ith 8 degrees of freedom S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .10032 U ncertainty C o e ffic ie n t (asymmetric) = .01355 with Home dependent = .00984 with Learn dependent *Due to rounding, percentage to ta ls may not equal 100. S t a t i s t i c s , presented a t the bottom o f Table 28, indicates a r e la t io n ­ ship ex ists between property owners with d is s im ila r types of home develop­ ment and information sources o f property a v a i l a b i l i t y . A Chi-square value o f 59.29720, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners w ith d is s im ila r home developments d i f f e r in terms o f information source o f property a v a i l a b i l i t y . Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the relatio n sh ip between property owners with d is s im ila r types of home development and information sources of property a v a i l a b i l i t y is weak. S t a t is t ic s support the percentage d iffe re n c e s , 100 displayed in Table 28. However, because of the weak nature of the re la tio n s h ip , knowledge o f information sources of property a v a i l a b i l i t y w i l l only s l i g h t l y help p re d ic t type of home development on land in the study area. In te n t to Sell Property Many property owners buy property with the in te n t to s e ll i t in the near fu tu re . This is the nature of an investment. Also property goes up fo r sale when people get d is s a tis fie d with i t or a need arises to liq u id a te assets to pay other expenses. I t comes as no surprise th a t 21.9% of northern Michigan study area property owners intend to s e ll a l l or p a rt of t h e i r property in the near fu tu re (Table 2 9 ). What is somewhat surprising is th a t 24.8% o f northern Michigan study area property owners are not sure whether they intend to s e l l . This makes the p o ten tia l fo r fu tu re sales q u ite high. Table 29 NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER'S INTENTIONS CONCERNING FUTURE PROPERTY SALES (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage Cumulative Frequency Percentage Intend to Sell 419 21.9 21.9 Not Sure 473 24.8 46.7 Do Not Intend to S ell 1,017 53.3 100.0 Total 1,909 100.0 S ell 101 When in d ivid ual counties and townships are considered, there is l i t t l e noticeable d ifferen ce between each area and the overall percentage d is tr ib u tio n presented in Table 29. Also, s t a t i s t i c s in d ica te th a t any re la tio n s h ip th a t exists between area of property location and desire to s e ll is very weak. Therefore, each area (county or township) can be considered independent and it s respective percentage d is tr ib u tio n clo se ly approximates the overall percentage d i s tr ib u tio n . When type of home development is considered with desire to s e ll in Table 30, i t is found th a t property owners with no type of home develop­ ment on the land are more apt to s e ll than e it h e r seasonal or permanent home owners. Almost o n e-th ird (30.9%) o f the property owners with no home on t h e i r land in the study area intend to s e ll t h e i r property with an ad ditional 33.8% not sure whether they wish to s e l l . Only 16.5% of the seasonal home owners and 19.5% o f the permanent home owners wish to s e ll t h e i r property. Indecision is lower on the p a rt of the seasonal and permanent home owners as only 20.5% o f property owners with a seasonal home and 20.0% o f the property owners with a permanent home are not sure whether they intend to s e l l . The resu lts outlined here are s im ila r to resu lts found in a study conducted in Emmet and Cheyboygan counties o f northern Michigan, (Marans and Wellman, 1978). In th a t study, 20% o f the permanent home owners intended to s e ll t h e i r property and 10% of the seasonal home owners i n ­ tended to s e l l . Results are s im ila r enough to conclude th a t there is no d iffe re n c e between homeowners in Crawford, Kalkaska, and Otsego counties compared to homeowners in Emmet and Cheyboygan counties concerning desire to s e ll t h e i r property. S t a t i s t i c s , presented at the bottom of Table 30, in dicate a r e la t io n ­ ship ex ists between property owners with d is s im ila r types o f home development 102 Table 30 TYPE OF HOME DEVELOPMENT BY DESIRE TO SELL PROPERTY Sell Count Row Percentage Row Cumulative Percentage Intend To S ell Property Do Not Intend to S ell Property Not Sure Row Total % of Total Home No Home 198 30.9 30.9 226 35.3 66.2 217 33.8 100.0 641 33.7 Seasonal Home 115 16.5 16 .5 439 63.0 79.5 143 20.5 100.0 697 36.6 Permanent Home 110 19 .5 19 .5 341 60.5 80.0 113 20.0 100.0 564 29.6 Column Total % of Total 423 22.2 1,006 52.9 473 24.9 1,902 100.0 Chi-square = 128.65296 with 4 degrees o f freedom S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .18439 Contingency C o e ffic ie n t .25764 Uncertainty C o e ffic ie n t (asymmetric) = .03144 w ith Home dependent = .03394 w ith S ell dependent SpearmansD Ks Kendall Tau = -.1 845 = -.1741 and in te n t to s e ll property. A Chi-square value o f 128.65296, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners with d is s im ila r types of home development d i f f e r in terms o f in te n t to s e ll property. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t, Contingency C o e ffic ie n t) in d ic a te the re la tio n s h ip between property owners with d is s im ila r types o f home development and in te n t to s e ll is moderate. S t a t i s t i c s , displayed in Table 30, in d ica te a moderate re la tio n s h ip exists between in te n t to s e ll property and type o f home development on the land 103 in the study area. Therefore, knowledge of in te n t to s e ll w i l l help p red ict type o f home development on the land in the study area. Additional s t a t i s t i c s , SpearmansRs value o f -.1845 and a Kendall Tau value o f -.1 7 4 1 , in d ica te th a t as type o f home development proceeds from no home to permanent home to seasonal home, the percentage o f property owners in ­ tending to s e ll declines. As previously ascertained, many property owners intend to s e ll t h e i r property in the near fu tu re but how soon is the near future? As presented in Table 31, the near fu tu re fo r 50.7% o f those property owners who intend to s e ll is w ith in a year. An additional 33.0% intend to s e ll w ith in a 1-5 year period. Table 31 PROPERTY OWNERS PLANS TO SELL PROPERTY (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage Cumulative Frequency Percentage Within a Year 186 50.7 50.7 1-5 Years 121 33.0 83.7 6-10 Years 42 11.4 95.1 Over 10 Years 18 4.9 100.0 367 100.0 YRSELL Total When in d ivid u a l counties and townships are considered, l i t t l e change is noted from the o v erall d is t r ib u t io n , however, when type o f home development is considered in Table 32, s li g h t ly more property owners with no home development in the study area (54.2%) intend to s e ll t h e i r property w ith in a year than property owners with seasonal 104 Table 32 PROPERTY OWNERS PLANS TO SELL PROPERTY BY TYPE OF HOME DEVELOPMENT YRSELL Count Row Percentage Row Cumulative Percentage Within a Year 1-5 Years 6-10 Years Over 10 Years No Home 90 54.2 54.2 52 31.3 85.5 14 8.4 93.9 10 6.0 99.9 166 45.9 Seasonal Home 42 44.7 44.7 33 35.1 79.8 15 16.0 95.8 4 4.2 100.0 94 26.0 Permanent Home 50 49.0 49.0 36 35.3 84.3 13 11.8 96.1 4 3.9 100.0 102 28.2 182 50.3 121 33.4 41 11.3 18 5.0 362 100.0 Row Total % o f Total Home Column Total % of Total Chi-square = 8.33061 with 6 degrees of freedom Not s ig n if ic a n t at .05 p r o b a b ility le v e l. Cramers V = .08699 Uncertainty C o e ffic ie n t (asymmetric) = .01213 with Home dependent = .01230 with YRSELL dependent SpearmansD = -.0459 s Kendall Tau = -.0 4 0 8 *Due to rounding, percentage to ta ls may not equal 100. homes (44.7%) or property owners with permanent homes (49.0%). Property owners with seasonal or permanent homes e x h ib it a desire to hold onto t h e i r property a few more years before s e llin g . S t a t i s t i c s , presented a t the bottom o f Table 32, in dicate a re la tio n s h ip does not e x is t between property owners with d is s im ila r types o f home development and length o f time before property is offered 105 fo r sale. A Chi-sauare value of 8.33061, not s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners with d is s im ila r types of home development do not d i f f e r in terms o f length of time before pro­ perty is offered fo r sale. Therefore, knowledge of length of time before a property owner wishes to s e ll his property w i l l not help p re d ic t type of home development located on the land in the study area. Summary This chapter explored factors th a t influenced the i n i t i a l property purchases of northern Michigan study area property owners. desire to s e ll presently owned property and the In a d d itio n , time frame r e la t iv e to desire to s e ll were explored. 1) Generally, property in the study area is acquired through o u trig h t purchase although 6.2% of present property owners in h e rite d t h e i r property. 2) The major reason fo r property acq u isitio n is fo r investment or a retirem ent home, however, recreation al a c t i v i t i e s place high as reasons f o r a c q u is itio n . 3) Seasonal home owners place g rea ter emphasis on recreational opportunities f o r property a c q u isitio n than do permanent residents or property owners with no home development in the study area. 4) Friends and r e la t iv e s are the most important sources of information leading to property purchases. T ra d itio n a l market information sources (newspaper/magazine ads, and real estate salespersons) are more important fo r learning o f a v a ila b le property fo r permanent homes than fo r seasonal 106 homes or property owners with no home in the study area. However, r e la tiv e s and friends are s t i l l the most important source o f information about a v a ila b le property in the study area. 5) Over o n e - f if t h o f northern Michigan study area property owners intend to s e ll t h e ir property in the fu tu re. Property owners with no home development on t h e ir land are more apt to s e ll t h e i r property than seasonal or permanent home owners. 6) Of the o n e - f if t h who desire to s e ll t h e i r property, h a lf desire to s e ll w ith in one year and an additional one-third w ith in 1-5 years. CHAPTER IV ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS ON ISSUES OF CONCERN The ac q u is itio n p ric e of property in northern Michigan is d ir e c t ly affec ted by county or municipal decisions. The level of taxes, building r e s t r ic t io n s , amount of services provided, e tc . a l l play important roles in ra is in g or lowering not only the p rice but the u t i l i t y fo r an individual property to the owner. Perceived u t i l i t y fo r a t r a c t of land must exceed the a c q u is itio n price or an individ ual would not purchase the land given adequate income and no other more desirable purchase options. On the other hand, when u t i l i t y diminishes to such an extent th a t is is below salvage p r ic e , then an in d ivid u a l w i l l s e l l . d i r e c t ly on u t i l i t y . Property tax levels impact Property tax lev els also d i r e c t ly impact the level o f municipal services which can be provided. Perceived q u a lity o f muni­ cipal services have an in d ir e c t influence on u t i l i t y . This section w i l l explore many p e rtin e n t issues th a t can a f f e c t an in d ivid ual property's u t i l i t y to the owner. Property Tax Levels Recent property tax re v o lts have indicated an ever increasing resistance to r is in g taxes. Even though b a llo t proposals to reduce property taxes were defeated in Michigan in the 1976 general e le c tio n , the f a c t th a t enough signatures were s o lic it e d to place the proposals on the b a l l o t in dicates th a t there is a fe e lin g of resentment toward r is in g property taxes. This is q u ite evident fo r northern Michigan study area property owners because, as shown in Table 33, 65.4% thought property taxes were too high. In a d d itio n , 33.7% thought property taxes were about r i g h t , and only .9% thought property taxes too low. 107 P rio r 108 Table 33 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD CURRENT PROPERTY TAX LEVELS (Frequency and Percentage D is trib u tio n ) Frequency Percentage Cumulative Frequency Percentage 1,267 65.4 65.4 652 33.7 99.1 18 .9 100.0 1,937 100.0 Absolute Frequency PR0PTAX High About Right Low Total research (Marans, Wellman, 1978) reported th a t tw o-thirds of northern Michigan property owners f e l t property taxes too high r e la t iv e to ser­ vices they supported. Results from th is survey closely p a r a lle l those from Marans and Wellman. However, e a r l i e r research studies indicated generally only 15-20% of northern Michigan property owners f e l t property taxes too high. (V e rtre e s , 1967; McEwan, 1970). Obviously, resentment toward property tax lev e ls has g re a tly increased in the la s t few years. Rising resentment probably p a r a lle ls ris in g property tax levels during the time span between studies. When individ ual counties are considered in Table 34, more property owners in Crawford and Kalkaska counties (75.1% and 81.1%, resp ective ly) f e l t property taxes were too high than in Otsego county where only 56.9% o f the property owners f e l t property taxes were too high. Also, 41.6% of the property owners in Otsego thought property taxes were gust r ig h t . The property taxes in each o f the counties, although somewhat d i f f e r e n t due to d i f f e r e n t school d i s t r i c t s , were comparable. per $1,000 o f valuation averaged about $40.00. The assessed value 109 Table 34 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD CURRENT PROPERTY TAX LEVELS, BY COUNTY PROPTAX Count Row Percentage Row Cumulative Percentage Row Total % o f Total High About Right Low 208 75.1 75.1 69 24.9 100.0 0 0 100.0 381 81.1 81.1 88 18.7 99.8 1 .2 100.0 677 56.9 56.9 495 41.6 98.6 17 1.4 99.9 1,189 61.4 1,266 65.4 652 33.7 18 1,937 100.0 COUNTY Crawford Kalkaska Otsego Column Total % o f Total .9 111 14.3 470 24.3 Chi-square = 103.42485 with 4 degrees of Freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .16339 Contingency C o e ffic ie n t .22514 Uncertainty C o e ffic ie n t (asymmetric) = .03105 with County dependent = .04177 with PROPTAX dependent *Due to rounding, percentage to ta ls may not equal 100. One explanation fo r the high percentages of property owners in Kalkaska county who f e l t property taxes were too high may be because of Michigan's tax e q u iliz a tio n program. B a s ic a lly , th is is a program whereby the s ta te assesses property values, in addition to the local assessor, so th a t they are commensurate among counties throughout the s ta te . The s ta te had ju s t finished i t s tax e q u iliz a tio n work in Kalkaska county a t the time th is study's questionnaires were being sent out. One e f f e c t of property tax e q u iliz a tio n was to s u b s ta n tia lly increase the number of delinquent taxpayers. (County personnel in the tax assessor's 110 o f f ic e reported th a t delinquent property taxes were up 300% in the span o f one year accounting f o r over 15% of a l l property owned). A second e f f e c t o f the tax e q u iliz a tio n program is th a t the sudden increase in taxes would tend to make property owners fe e l th a t property taxes were too high. Even without the e f f e c t of the tax e q u iliz a tio n program, there was a great deal of resentment to current property tax levels in the northern Michigan study area. Crawford and Otsego counties both had substantial percentages o f property owners who f e l t current property tax levels high. S t a t i s t i c s , presented a t the bottom of Table 34, in d ic a te a r e la t io n ­ ship ex is ts between property owners of ce rtain counties and a ttitu d e s toward property tax le v e ls . A Chi-square value of 103.42485, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners of c e rta in counties d i f f e r in terms of a ttitu d e s toward property tax le v e ls . Other q u a lify in g s t a t i s t i c s (Cramers V, U ncertainty, C o e ffic ie n t, Contingency C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners of c e rta in counties and a ttitu d e s toward property tax levels is moderate. S t a t is t ic s support the percentage d iffe re n c e s , displayed in Table 34, therefo re knowledge of a ttitu d e s toward property tax levels w i l l help p red ict county of pro­ perty lo c a tio n . When in d ivid ual townships are considered, in Table 35, an in te re s tin g development occurs. A ll townships but one, Bagley, have a higher percen­ tage of property owners who fe el property taxes are higher than the overall percentage d i s tr ib u tio n . I t should be remembered th at because Bagley had such a high number o f respondents th a t i t can, and in th is case d id , a f f e c t the o v e ra ll re s u lts . The townships ranged from a high of 87.6% o f property owners in G a rfie ld township who f e l t property taxes were m Table 35 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD CURRENT PROPERTY TAX LEVELS, BY TOWNSHIP Count Row Percentage Row Cumulative Percentage PROPTAX Row Total High About Right Low 102 74.4 74.4 35 25.5 100.0 0 0 99.9 137 7.1 Grayling T27NR2W 24 70.6 70.6 10 29.4 100.0 0 0 100.0 34 , 1.7 Grayling T26NR4W 83 77.6 77.6 24 22.4 100.0 0 0 100.0 107 5.5 Orange T26NR7W 92 80.7 80.7 21 18.4 99.1 1 .9 100.0 114 5.9 217 79.5 79.5 56 20.5 100.0 0 0 100.0 273 14.1 G a rfie ld T25NR7W 71 87.6 87.6 10 12.3 100.0 0 0 99.9 81 4.2 Bagley T30NR3W 547 54.2 54.2 447 44.3 98.5 15 1.5 100.0 1,009 52.1 Chester T29NR2W 67 67.0 67.0 31 31.0 98.0 2 2 .0 100.0 100 5.2 82 4.2 % o f Total Township I South Branch T25NR2W Blue Lake T28NR5W Dover T31NR2W . 64 78.0 78.0 18 22.0 100.0 0 0 100.0 Column Total % o f Total 1,267 65.4 652 33.7 18 .9 1,937 100.0 Chi-square=132.5531 with 16 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .18498 Uncertainty C o e ffic ie n t (asymmetric) = .02284 with Township dependent = .05382 with PROPTAX dependent *Due to rounding, percentage t o t a ls may not equal 100. 112 high to a low o f 54.2% in Bagley township. The three townships o f Kalkaska county a l l recorded large numbers o f property owners who f e l t property taxes were too high. In f a c t , in terms o f ranking, from highest to lowest, Kalkaska county townships were the top three. S t a t i s t i c s , presented a t the bottom o f Table 35, in d ica te a r e la t io n ­ ship exists between property owners of c e rta in townships and t h e ir a ttitu d e s toward property tax le v e ls . A Chi-square value of 132.5531, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners of ce rtain townships d i f f e r in terms of a ttitu d e s toward property tax lev els . Other q u a lify in g s t a t i s t i c s , (Cramers V, Uncertainty C o e ffic ie n t) indicate the relatio n sh ip between property owners o f c e rta in townships and a ttitu d e s toward property tax lev e ls is moderate. Therefore, knowledge of a ttitu d e toward property tax lev els w i l l help p red ict townships o f property loca­ tio n . When type of home development is considered with a ttitu d e s toward property tax le v e ls , a c e rta in trend is noted. As type o f home develop­ ment progresses from no home in the study area to seasonal home to permanent home, the percentage of property owners who feel property taxes are high increases. As shown in Table 36, a to ta l of 73.1% of the property owners with permanent homes view property taxes as too high compared to 69.9% f o r seasonal home owners and only 53.1% fo r property owners with no home development in the study area. S t a t i s t i c s , displayed a t the bottom o f Table 36, in d ica te a r e la t io n ­ ship exists between property owners w ith d is s im ila r types of home development and a ttitu d e s toward property tax le v e ls . A Chi-square value of 76.75619, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies 113 Table 36 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD CURRENT PROPERTY TAX LEVELS, BY TYPE OF HOME DEVELOPMENT Count Row Percentage Row Cumulative Percentage PROPTAX High About Right Low Row Total No Home 332 53.1 53.1 278 44.5 97.6 15 2 .4 100.0 625 32.5 Seasonal Home 492 69.9 69.9 209 29.7 99.6 3 .4 100.0 704 36.7 432 73.1 73.1 159 26.9 100.0 0 0 100.0 1,256 65.4 646 33.6 18 % o f Total Home | Permanent Home Column Total % o f Total 591 30.8 1,920 100.0 .9 Chi-square = 76.75619 with 4 degrees o f freedom S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .14138 Contingency C o e ffic ie n t = .19606 Uncertainty C o e ffic ie n t (asymmetric) = .01852 with Home dependent = .02950 with PROPTAX dependent SpearmansRs = .1745 Kendall Tau = .1638 *Due to rounding, percentage t o ta ls may not equal 100. property owners with d is s im ila r types of home development d i f f e r in terms of a ttitu d e s toward property tax le v e ls . Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the r e la t io n ­ ship between property owners with d is s im ila r types o f home development and a ttitu d e s towards property tax levels is moderate. This re la tio n s h ip is fu r th e r supported by a Spearmansps value of .1745 and a Kendall Tau value of .1638. In other words, as type o f home development progresses from no home in the study area to a seasonal home to a permanent home, 114 the number o f property owners who fe el property taxes are high increases. Q u ality of Municipal and County Services Related to property tax levels is the q u a lit y of municipal or county services provided. I f property tax levels are perceived as high but q u a lity of services is viewed as very good, then there may well be a balancing e f f e c t on t o ta l u t i l i t y . On the other hand, i f q u a lity of services is perceived as poor and property taxes high, then there may very well be q u ite a lo t o f d is s a tis f ie d property owners who wish to s e ll. O v e ra ll, as shown in Table 37, there does not seem to be too much unhappiness with q u a lit y o f services. Although only 4.5% of the property owners thought the q u a lity of services very good compared to 10.3% who f e l t q u a lity poor; in general, property owners seemed s a t is ­ f ie d . Between the two a t t it u d e extremes, 24.3% o f the property owners thought the q u a lity o f services provided was good and an additional 30.9% thought q u a lity of services average compared to only 7.2% who f e l t q u a lity was below average. Of great in te r e s t is the 22.7% of property owners who are not sure o f the q u a lity of services provided. This may r e f l e c t many non-resident property owners who occasionally v i s i t the area and are not aware, or do not wish to take advantage o f , the ser­ vices provided. Previous research in d ica ted , in general, property owners were content with the q u a lit y o f local public services (Marans and Wellman, 1978). Results fo r th is study support previous research re s u lts . When in d ivid ual counties are considered, in Table 38, some devia­ tions from the o v e ra ll percentage d is trib u tio n s are noted. Crawford county approximates the o verall percentage d is tr ib u tio n best with only a s lig h t increase in the percentage o f property owners (30.4%) viewing 115 Table 37 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD QUALITY OF PROVIDED MUNICIPAL OR COUNTY SERVICES (Frequency and Percentage D is trib u tio n ) PROPTAX Very Good Frequency Cumulative Frequency Percentage Frequency Percentage 86 4.5 4.5 Good 464 24.3 28.8 Average 589 30.9 59.7 Below Average 138 7.2 66.9 Poor 197 10.3 77.2 Not Sure 434 22.7 99.9 1,908 100.0 Total *Due to rounding, percentage to ta ls may not equal 100. q u a lity of services as being e it h e r very good or good. There is also a small increase in the percentage (20.8%) who view q u a lity as e it h e r poor or below average. In Kalkaska county, there was generally greater unhappiness toward q u a lity of services provided. Only 21.9% of property owners responded th at q u a lit y o f services was e it h e r good or very good w hile 25.9% responded th a t i t was below average or poor. In Otsego county, property owners were generally s a tis f ie d with q u a lit y of services as 31.2% thought q u a lity was good or very good while only 13.4% thought q u a lit y was below average or poor. In a d d itio n , 25.1% o f property owners in Otsego county were not sure of the q u a lity o f services provided. S t a t i s t i c s , presented at the bottom o f Table 38, in d ica te a re la tio n s h ip exists between property owners o f c e rta in counties and t h e i r a ttitu d e s toward q u a lity of municipal or county services provided. A 116 Table 38 ATTITUDES OF NORTHER MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD QUALITY OF MUNICIPAL OR COUNTY SERVICES PROVIDED, BY COUNTY County Row Percentage Row Cumulative Percentage QUALSERV Very Good Good Average Below Average Row Total Poor Sure % o f Total County Crawford 12 4 .4 4 .4 71 26.0 30.4 89 32.6 63.0 20 7.3 70.3 37 13.5 83.8 44 16.1 99.9 273 14.3 Kalkaska 16 3.5 3.5 85 18.4 21.9 145 31.4 53.3 51 11.0 64.3 69 14.9 79.2 96 20.8 100.0 462 24.2 Otsego 58 4.9 4.9 308 26.3 31.2 355 30.3 61.5 67 5.7 67.2 90 7.7 74.9 294 25.1 100.0 1,172 61.5 Column Total 86 4 .5 464 24.3 589 30.9 138 7.2 196 10.3 434 22.7 1,907 100.0 Chi-square = 61.65390 w ith 10 degrees of freedom Signi f ic a n t a t .05 p r o b a b ility le v e l. Cramers V = .10378 U ncertainty C o e ffic ie n t (asymmetric) = .01616 with County dependent = .00930 with QUALSERV dependent *Due to rounding, percentage to ta ls may not equal 100. Chi-square value o f 61.6539, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners of c e rta in counties d i f f e r in terms of a ttitu d e s toward q u a lit y of municipal or county services provided. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in dicate the re la tio n s h ip between property owners of c e rta in counties and a ttitu d e s toward q u a lity of municipal or county services is weak. S t a t is t ic s support the percent­ age d iffe re n c e s , displayed in Table 38, however, due to the weak nature o f the r e la tio n s h ip , knowledge o f a t t itu d e toward q u a lity of municipal or county services provided w i l l help l i t t l e to p red ict county of property lo c a ti on. 117 When considering in divid ual townships, in Table 39, i t is noted th a t Bagley township, once again, exerts a strong influence on the overall percentage d is tr ib u tio n . Property owners in Bagley township recorded the lowest percentage (12.5%) of those who thought q u a lit y of service provided was below average or poor. This compares to Blue Lake township where 30.5% o f the property owners f e l t q u a lit y o f services provided was below average or poor. Bagley township also recorded the second highest percentage (31.9%) fo r property owners who f e l t q u a lity of services provided was good or very good. South Branch township had the highest percentage (33.6%) o f property owners who f e l t the q u a lity of services provided was good or very good. G enerally, the presence o f Bagley township, and to a lesser e x te n t, South Branch township, tended to i n f l a t e the overall percentage d is tr ib u tio n in favor of the very good and good categories and d e fla t e the overall percentage d is tr ib u tio n in the below average and poor categories. S t a t i s t i c s , presented a t the bottom o f Table 39, in d ica te a r e la ­ tionship exists between property owners o f c e rta in townships and a ttitu d e s toward q u a lity of municipal or county services provided. South Branch, Bagley and Orange township property owners, in general, gave a higher q u a lity ra tin g toward municipal or county services pro­ vided than property owners in other townships. A Chi-square value o f 110.5774, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l, implies property owners of ce rtain townships d i f f e r in terms o f a ttitu d e s toward q u a lity o f municipal or county services provided. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e f f ic ie n t ) , in d ica te the re la tio n s h ip between property owners of c e rta in townships and a ttitu d e s toward q u a lity o f municipal or county services provided is weak. the percentage d ifferen ces , displayed in Table 39. S t a t is t ic s support However, due to 118 Table 39 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD QUALITY OF MUNICIPAL OR COUNTY SERVICES PROVIDED, BY TOWNSHIP Count Row Percentage Row Cumulative Percentage QUALSERV Very Good Good Average Below Average Poor Not Sure South Branch T25NR2W 8 6 .0 6.0 37 27 .6 33 .6 41 30.6 64.2 5 3.7 67.9 15 11.2 79.1 28 20.9 100.0 134 7.0 Grayling T27NR2W 3 8 .8 8 .8 4 11 .8 20 .6 14 41.2 61.8 1 2.9 64.7 6 17.6 82.3 6 17.6 99.9 34 1.8 Grayli ng T26NR4W 1 .9 .9 30 28 .3 29 .2 34 32.1 61.3 14 13.2 74.5 17 16.0 90.5 10 9 .4 99.9 106 5.6 Orange T26NR7W 6 5 .3 5.3 21 18.6 23 .9 44 38.9 62.8 5 4.4 67.2 13 11.5 78.7 24 21.2 99.9 113 5.9 Blue Lake T28NR5W 6 2.2 2.2 51 19 .0 21 .2 75 28.0 49.2 39 14.5 63.7 43 16.0 79.7 54 20.1 99.8 268 14.0 G a rfie ld T25NR7W 4 5.1 5.1 12 15 .2 20 .3 26 32.9 53.2 7 8.9 62.1 12 15.2 77.3 18 22.8 100.0 79 4.1 Bagley T30NR3W 50 5.0 5.0 269 26 .9 31.9 308 30.9 62.8 55 5.5 68.3 70 7.0 75.3 246 24.6 99.9 998 52.3 Chester T29NR2W 2 2.1 2.1 24 24 .7 26 .8 24 24.7 51.5 4 4.1 55.6 11 11.3 66.9 32 33.0 99.9 97 5.1 Dover T31NR2W 6 7.6 7.6 16 20.2 27 .8 23 29.1 56.9 8 10.1 67.0 10 12.7 79.7 16 20.2 99.9 79 4.1 86 4 .5 464 24.3 589 30.9 138 7.2 197 10.3 434 22.7 1,908 100.0 Row Total % of Total Township I Column Total % o f Total Chi-square = 110.5774 w ith 40 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .10766 U ncertainty C o e ffic ie n t (asymmetric) = .01786 with Township dependent = .01795 with QUALSERV dependent *Due to rounding, percentage to t a ls may not equal 100. 119 the weak nature o f the re la tio n s h ip knowledge of a t t itu d e toward q u a lity o f municipal or county service provided w i l l help l i t t l e to p red ict townships o f property lo c a tio n . When type of home development is considered in Table 40, resu lts are q u ite in te r e s tin g . There is a d e f in it e d ifferen ce in the number of property owners s a t is f ie d with q u a lit y o f services provided when each type o f home development is compared. Permanent homeowners were generally s a t is f ie d with the q u a lit y of services provided as evidenced by the 36.4% who f e l t q u a lit y of services was good or very good. S u rp ris in g ly , permanent home owners also recorded the highest percentage (19.6% ), who thought q u a lity of services was below average or poor. At the same time, then, permanent home owners had a substantial number s a tis f ie d with q u a lit y o f services provided and a substantial number d is s a t is f ie d . The reason fo r th is in te r e s tin g re s u lt probably has to do with u n certainty. Only 4.8% o f the permanent home owners were not sure how they f e l t about q u a lit y o f services provided compared to 45.4% of the property owners with no home in the study area. Obviously then, people who li v e in an area year round are much more aware o f the services provided, and t h e i r q u a lit y , than are property owners with no home development on the land. Property owners with no home development in the study area recorded the lowest percentage (23.2%) who f e l t q u a lity o f services provided was good or very good, and also the lowest percentage (9.4%) who thought q u a lity was below average or poor. Although property taxes were viewed as high by a l l property owners, regardless o f type of home development, q u a lit y o f services were generally viewed as good in d ic a tin g th a t overall u t i l i t y may be balanced. Ignorance o f a community's stru ctu re and p o lic ie s is not a new phenomenon fo r property owners with undeveloped land or seasonal homes. 120 Table 40 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS ' TOWARD QUALITY OF MUNICIPAL AND COUNTY SERVICES PROVIDED, BY TYPE OF HOME DEVELOPMENT RowPercentage Row Cumulative Percentage lH|ALSERV I Very Good Good Average Below Average Poor Not Sure Row Total % of Total Home No Home 23 3.7 3.7 120 19.5 23.2 135 21.9 45.1 24 3.9 49.0 34 5.5 54.5 280 45.4 99.9 616 32.5 Seasonal Home 30 4.3 4.3 160 23.1 27.4 221 31.9 59.3 59 8.5 67.8 98 14.1 81.9 125 18.0 99.9 693 36.6 Permanent Home 33 5.6 5.6 180 30.8 36.4 228 39.0 75.4 54 9 .2 84.6 61 10.4 95.0 28 4.8 99.8 584 30.8 Column Total 86 4.5 460 24.3 584 30.8 137 7.2 193 10.2 433 22.9 1,893 100.0 % o f Total Chi-square = 321 .30754 with 10 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .23692 Uncertainty C o e ffic ie n t (asymmetric) = .07718 with Home dependent = .05441 w ith QUALSERV dependent *Due to rounding, percentage to ta ls may not equal 100.0. Previous research revealed no opinion responses or non response were high among property owners when asked t h e i r opinion about local community services. There seems to e x is t a high degree of uncertainty toward county functions among property owners with undeveloped land and seasonal home owners in northern Michigan (V ertrees, 1967; McEwan, 1970). Uncertainty and confusion about community services were viewed as such a problem, a recommendation to communicate with absentee landowners through the mail was proposed (McEwan, 1970). As f a r as th is researcher knows, the proposed reconriendation was never implemented and confusion and uncertainty levels remain high among property owners with no home develop ment on t h e ir land and seasonal home owners in the study area. S t a t i s t i c s , presented a t the bottom of Table 40, in d ic a te a re la tio n s h ip exists between property owners with d is s im ila r type of home development and a ttitu d e s toward q u a lity o f municipal or county services provided. A Chi-square value of 321.30754, s ig n if ic a n t at the .05 p r o b a b ility le v e l , implies property owners with d is s im ila r types o f home development d if f e r s in terms of a ttitu d e s toward q u a lit y of muni­ cip al or county services provided. Other q u a lify in g s t a t i s t i c s (Cramers Uncertainty C o e ffic ie n t) in dicate the rela tio n s h ip between property owners w ith d is s im ila r types of home development and a ttitu d e s toward q u a lity of municipal or county services provided is strong. S t a t is t ic s support the percentage differences displayed in Table 40, th e re fo re , knowledge of a ttitu d e s toward q u a lity of county services provided w i l l g re a tly help p red ict property owners type of home development. Quantity of Municipal and County Services Related very closely with q u a lity of municipal and county services is the qu antity of services provided. Sometimes q u an tity and q u a lit y can be confused, th e re fo re , the questionnaire was designed to t r y and a l l e v i a t e th a t problem. O v e ra ll, property owners seem s a t is f ie d with the current lev el of services provided. t e r i s t i c is noted. However, one important charac­ As shown in Table 41, a t h ir d (33.5%) o f the property owners were not sure how they f e l t about the current level o f services provided. This indecision may be caused by the lack of information about what services are a v a ila b le . This problem may not be as great in urban areas where the presence of p o lic e , h e alth , f i r e , bus services, e tc . is well recognized. However, in rural areas, even a few property 122 Table 41 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD THE QUANTITY OF MUNICIPAL AND COUNTY SERVICES PROVIDED (Frequency and Percentage D is trib u tio n ) Absolute Frequency Frequency Percentage Cumulative Frequency Percentage Could Use a Lot More 118 6.2 6.2 Could Use Some More 302 16.0 22.2 About Right 649 34.3 56.5 Don't Need so Many 136 7.2 63.7 53 2 .8 66.5 634 33.5 100.0 1,892 100.0 QUANSERV Would Like to see a Lot Less Not Sure Total owners with permanent residency may not be aware o f some services th a t are provided. Whatever the reason fo r the lack of inform ation, i t s p ills over in to the category 'Not Sure' and flags a p o te n tia l problem. Of those property owners who had a fe e lin g towards the q u an tity o f services provided, 22.2% f e l t th at a t le a s t some more should be provided and 10% f e l t less services would be desirable. The remaining 34.3% of property owners responding f e l t the level of services provided was about r ig h t (Table 41). When in divid ual counties are considered in Table 42, resu lts d i f f e r very l i t t l e from the o v erall percentages d i s tr ib u tio n . The only d iffe re n c e among the counties is th a t fewer property owners in Otsego county favor more services to be provided than e it h e r Crawford or Kalkaska. Only 19.8% o f property owners in Otsego would l i k e to see some more services provided compared to 25.7% fo r Crawford and 26% fo r Kalkaska. 123 Table 42 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD QUANTITY OF MUNICIPAL AND COUNTY SERVICES PROVIDED, BY COUNTY Count Row Percentage Row Cumulative Percentage QUANSERV Could Use a Lot More Could Use Some More About Right Don't Need So Many Like to See a Lot Less Not Sure Crawford 26 9.7 9.7 43 16.0 25.7 98 36.6 62.3 21 7.8 70.1 10 3.7 73.8 70 26.1 99.9 268 14.2 Kalkaska 41 8 .8 8 .8 80 17.2 26.0 146 31.5 57.5 27 5 .8 63.3 12 2 .6 65.9 158 34.0 99.9 464 24.5 Otsego 51 4.4 4 .4 179 15.4 19.8 405 34.9 54.7 88 7.6 62.3 30 2 .6 64.9 406 35.0 99.9 1,159 61.3 118 6.2 302 16.0 649 34.3 136 7.2 52 2 .8 634 33.5 1,891 100.0 Row Total % of Total County Column Total % of Total Chi-square = 61.38107 with 10 degrees o f freedom S ig n ific a n t at .05 p r o b a b ility level Cramers V = .10399 Uncertainty C o e ffic ie n t (asymmetric) = .00959 with County dependent = .07484 with QUANSERV dependent *Due to rounding, percentage to ta ls may not equal 100. S t a t i s t i c s , presented a t the bottom o f Table 42 in d ic a te a r e la t io n ­ ship ex is ts between property owners of c e rta in counties and a ttitu d e s toward qu an tity o f municipal and county services provided. Crawford county property owners, in general, desired more county or municipal services than did Otsego or Kalkaska property owners. A Chi-square value of 61.38107, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners o f c e rta in counties d i f f e r in terms of a ttitu d e s toward q u an tity o f municipal or county services provided. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) indicates the re la tio n s h ip 124 between property owners o f ce rtain counties and a ttitu d e s toward quantity o f municipal or county services provided is weak. S t a t is t ic s support the percentage d iffe re n c e s , displayed in Table 42, however, due to the weak nature of the r e la tio n s h ip knowledge o f a t t itu d e toward qu antity o f municipal or county services provided w i l l help l i t t l e to pred ict county o f property lo c a tio n . When in d ivid u a l townships are considered in Table 43, many differences are noted. In Grayling T27NR2W township only 9.4% of the property owners would l i k e to see a t le a s t some more services provided and 21.8% would l i k e to see less. This same fe e lin g is prevalent in Dover township where only 14.1% o f the property owners wanted more services and 20.5% wanted less. Uncertainty levels were also high in a l l townships except f o r Grayling T26NR4W where only 13.6% of the property owners were not sure how they f e l t about the qu an tity o f municipal and county services provided. In contrast to Grayling T26NR4W township, the next lowest level of uncertainty was recorded in G a rfie ld township where 31.6% o f the property owners were not sure how they f e l t about the quantity o f municipal and county services provided. S t a t is t ic s presented a t the bottom of Table 43 in d ica te a r e la t io n ­ ship ex is ts between property owners of c e rta in townships and a ttitu d e s toward qu an tity of municipal or county services provided. Grayling T26NR4W township property owners, in p a r t ic u l a r , would lik e to see more municipal and county services provided. A Chi-square value of 90.33334, s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners o f c e rta in townships d i f f e r in terms o f a t titu d e toward qu an tity o f municipal or county service provided. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in dicate the re la tio n s h ip between property owners of c e rta in townships and a ttitu d e s toward qu an tity of 125 Table 43 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD THE QUANTITY OF MUNICIPAL AND COUNTY SERVICES PROVIDED, BY TOWNSHIP Count Row Percentage Row Cumulative Percentage QUANSERV Could Use a Lot More Like to Could Don11 Need see a Use Some About Right So Many Lot Less More Not Sure Row Total % o f Total Township | South Branch T25NR2W 13 9.7 9.7 15 11.2 20.9 46 34.3 55.2 11 8.2 63.4 4 3.0 66 .4 45 33.6 100.0 134 7.1 Gray!ing T27NR2W 0 0 0 3 9.4 9.4 11 34.4 43.8 2 6.2 50.0 5 15.6 65.6 11 34.4 100.0 32 1.7 Grayling T26NR4W 13 12.6 12.6 25 24.3 36.9 41 39.8 76.7 8 7.8 84.5 2 1.9 86.4 14 13.6 100.0 103 5 .4 OrangeT26NR7W 5 4.4 4 .4 19 16.7 21.1 42 36.8 57.9 6 5.3 63.2 1 .9 64.1 41 36.0 100.1 114 6.0 Blue Lake T28NR5W 26 9 .6 9.6 49 18.0 27.6 79 29.0 56.6 16 5.9 62.5 9 3.3 65.8 93 34.2 100.0 272 14.4 G a rfie ld T25NR7W 10 13.2 13.2 10 13.2 26.4 25 32.9 59.3 5 6.6 65.9 2 2 .6 68.5 24 31.6 100.1 76 4.0 Bagley T30NR3W 41 4.2 4.2 159 16.1 20.3 349 35.4 55.7 70 7.1 62.8 23 2.3 65.1 344 34.9 100.0 986 52.1 Chester T29NR2W 7 7.2 7.2 14 14.4 21 .6 30 30.9 52.5 6 6.2 58.7 3 3.1 61.8 37 38.1 99.9 97 5.1 Dover T31NR2W 3 3 .8 3.8 8 10.3 14.1 26 33.3 47.4 12 15.4 62.8 4 5.1 67.9 25 32.0 99.9 78 4.1 118 6 .2 302 16,0 649 34.3 136 7.2 53 2 .8 634 33.5 1,892 100.0 Column Total % o f Total Chi-square = 90.33334 with 40 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .09772 Uncertainty C o e ffic ie n t (asymmetric) = .01338 with Township dependent = .01689 with QUANSERV dependent *Due to rounding, percentage to t a ls may not equal 100. 126 municipal or county services provided is weak. S t a t is t ic s support the percentage differences displayed in Table 43. However, because o f the weak r e la tio n s h ip , knowledge o f a t titu d e toward qu an tity of municipal or county services provided w i l l help l i t t l e to pred ict township o f property lo cation. When type of home development is considered in Table 44, resu lts in d ic a te the longer a property owner is in the study area the more services w i l l be requested. Permanent home owners e x h ib it the highest percentage o f property owners desiring a t le a s t a few more services to be provided (30.0%) and property owners with no home in the study area have the lowest percentage desiring a t le a s t a few more services (12.0%). In terms o f u n certa in ty , property owners with no home in the study area had the highest percentage (55.7% ), and permanent home owners had the lowest percent (13.9%) who were not sure about the qu an tity of municipal and county services provided. Obviously, permanent home owners would stand to b e n e fit most by increased services and property owners with no home in the study area would probably b e n e fit le a s t. In a d d itio n , property owners with no home probably are le a s t aware of the current level of services provided. S t a t i s t i c s , presented at the bottom of Table 44, in d ica te a r e l a t i o n ­ ship exists between property owners with d is s im ila r types o f home development and a t t it u d e toward q u an tity of municipal or county services provided. A Chi-square value o f 258.43889, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners with d is s im ila r types of home development d i f f e r in terms o f a ttitu d e s toward qu an tity of municipal or county services provided. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the re la tio n s h ip between property owners with d is s im ila r types o f home development and a ttitu d e s toward 127 Table 44 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD THE QUANTITY OF MUNICIPAL AND COUNTY SERVICES PROVIDED, BY TYPE OF HOME DEVELOPMENT Count Row Percentage Row Cumulative Percentage QUANSERV Could Use a Lot More Could Use Some More About Right Don11 Need So Many Like to See a Lot Less Not Sure Row Total % o f Tota Home No Home 15 2 .5 2 .5 58 9 .5 12.0 160 26.3 38.3 24 3.9 42.2 13 2.1 44.3 339 55.7 100.0 609 32.4 Seasonal Home 58 8 .4 8 .4 113 16.3 24.7 249 36.0 60.7 45 6 .4 67.2 16 2.3 69.5 211 30.5 100.0 692 36.9 Permanent Home 45 7.8 7.8 128 22.2 30.0 234 40.5 70.5 66 11.4 81.9 24 4.2 86.1 80 13.9 100.0 577 30.7 118 6.3 299 15.9 643 34.2 135 7.2 53 2 .8 630 33.5 1,878 100.0 Column Total % of Total Chi-square = 258 .43889 with 10 degrees o f freedom S ig n ific a n t a t .05 p ro b a b ility level Cramers V = .21338 Uncertainty C o e ffic ie n t (asymmetric) = .06268 w ith Home dependent = .04763 with QUANSERV dependent quantity of municipal or county services provided is strong. The r e la t io n ­ ships between c e rta in types o f home development and property owners attitu d e s toward the qu an tity of municipal and county services provided, as discussed above, is strongly supported by s t a t i s t i c a l te s ts . S t a t is t ic s support the percentage d iffe re n c e s , displayed in Table 44, th e re fo re , knowledge o f a t t itu d e toward qu an tity o f municipal or county services provided w i l l g re a tly help p re d ic t type o f home development located on the land in the study area. Because resu lts f o r q u a n tity o f services provided w ity type of home development were so s im ila r to resu lts fo r q u a lit y o f services 128 provided with type o f home development, a t e s t fo r a c o rre la tio n was performed between qu antity of services provided and q u a lit y of services provided. A Spearmansps value of .3749 and a Kendall Tau value of .2873 in dicate a strong p o s itiv e c o rre la tio n . This indicates th a t property owners who thought the q u a lit y of services provided was good also were generally the same ones who desired more services to be provided. Also, the same property owners who were uncertain about the q u a lit y o f services provided were uncertain about the qu antity provided. Building Regulations Regulations are of concern to almost every property owner. Zoning regu lations, in p a r t ic u l a r , d ir e c t ly a f f e c t the value of property. It is usually assumed th at someone purchasing a piece of property finds out what he can or cannot do in terms of development. assumption is in v a lid . However, th is Not everyone enters in to a land transaction with complete knowledge o f zoning and other development regu lations. Even i f the regulations are known a t the time of purchase, they may change without the property owner being aware. In the northern Michigan study area, over one th ir d (35.4%) of the property owners e it h e r are not aware o f regulations concerning development or are not sure (Table 45 ). When in d ivid ual counties are considered, empirical and s t a t i s t i c a l evidence indicates th a t there is l i t t l e i f any deviation from the o v erall percentage d is tr ib u tio n reported in Table 45. The percentage of property owners in each county, who are aware/not aware of building reg u la tio n s , is approximately the same. When in d ivid ual townships are considered in Table 46, one very in te re s tin g trend is noted. In terms o f the percentage d is trib u tio n s of property owners who are aware of development reg u latio n s , three of the top four are townships considered lake resource 129 Table 45 NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS AWARENESS OF REGULATIONS CONCERNING LAND DEVELOPMENT (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage Cumulative Frequency Percentage 1,273 64.5 64.5 Not Sure o f B u ilding , Zoning, Development Regulations 290 14.7 79.2 Not Aware o f B u ilding , Zoning, Development Regulations 409 20.7 99.9 1,972 99.9 Zoning I Aware o f B uilding , Zoning, Development Regulations Total *Due to rounding, percentage to ta l does not equal 100. based f o r the s t r a t i f i e d random sample. Blue Lake (70.5% ), and Bagley (65.4%). Grayling T26NR4W (77.1% ), A ll these townships are above the o verall percentage fig u re (64.5%) f o r property owners who are aware o f bu ilding reg u lations. Grayling T27NR2W township, with 75.0% o f property owners aware of bu ilding regulations was the only other township to exceed the o v erall percentage fig u r e . This connection between lake resource based townships and awareness o f building regulations may be only coincidental because a c o rrela tio n c o e ff ic ie n t (Kendall Tau = .0600) does not support th is theory. S t a t i s t i c s , presented a t the bottom of Table 46, in d ic a te a re la tio n s h ip ex is ts between property owners of c e rta in townships and awareness of regulations concerning land development. A Chi-square value o f 41.07012, s ig n if ic a n t a t the .05 p ro b a b ility le v e l, implies 130 Table 46 NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS AWARENESS OF BUILDING REGULATIONS CONCERNING LAND DEVELOPMENT, BY TOWNSHIP Aware of B uilding , Zoning, Development Regulations Not Sure of B uilding, Zoning, Development Regulations Not Aware of B uilding , Zoning, Development Regulations South Branch T25NR2W 84 60.4 60.4 19 13.7 74.1 36 25.9 100.0 139 7.0 Grayling T27NR2W 24 75.0 75.0 1 3.1 78.1 7 21.9 100.0 32 1.6 Grayling T26NR4W 84 77.1 77.1 9 8 .3 85.4 16 14.7 100.1 109 5.5 Orange T26NR7W 67 58.3 58.3 19 16.5 74.8 29 25.2 100.0 115 5.8 196 70.5 70.5 40 14.4 84.9 42 15.1 100.0 278 14.1 40 50.0 50.0 16 20.0 70.0 24 30.0 100.0 80 4.1 Bagley T30NR3W 678 65.4 65.4 158 15.2 80.6 200 19.3 99.9 1,036 52.5 Chester T29NR2W 53 53.0 53.0 15 15.0 68.0 32 32.0 100.0 100 5.1 Dover T31NR2W 47 56.6 56.6 13 15.7 72.3 23 27.7 100.0 83 4.2 1,273 64.5 290 14.7 409 20.7 1,972 100.0 Count Row Percentage Row Cumulative Percentage Row Total % o f Total Township | Blue Lake T28NR5W G a rfie ld T25NR7W Column Total % o f Total Chi-square = 41.07012 w ith 16 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .10205 Uncertainty C o e ffic ie n t (asymmetric) = .00664 with Township dependent = .01199 with Zoning dependent *Due to rounding, percentage t o t a ls may not equal 100. 131 t property owners of c e rta in townships d i f f e r in terms o f awareness of regulations concerning land development. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners of c e rta in townships and awareness o f regulations concerning land development is weak. d iffe re n c e s , displayed in Table 46. S t a t is t ic s support the percentage However, because o f the weak r e l a ­ tio n s h ip , knowledge of level of awareness about regulations concerning land development w i l l help l i t t l e to p re d ic t township o f property lo c a ti on. When type o f home development is considered in Table 47, a strong re la tio n s h ip is found to e x is t . home development (no home — As expected with increasing lev el of seasonal — permanent) more property owners become aware of regulations concerning development. A t o t a l of 79.8% o f permanent home owners are aware o f land development regulations compared to only 47.5% o f the property owners with no home in the study area. However, the percentage o f property owners lacking complete information concerning land development regulations is d is tu rb in g ly high f o r a l l types o f home development. For property owners with no home in the study area, a to ta l of 52.5% are not aware or are not sure o f land development regulations. This fig u re f a l l s to 33.0% f o r seasonal home owners and drops to 20.2% fo r permanent home owners. R e s tric tiv e land development regulations may not be as crucial fo r permanent or seasonal home owners, as they already have a stru ctu re on the land. However, f o r property owners with no home, r e s t r i c t i v e land development regulations may severely d e fla te property value. Entering in to a property transaction without knowledge o f the regulations concerning fu tu re s it e development is risky at best. 132 Table 47 NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS AWARENESS OF BUILDING REGULATIONS CONCERNING LAND DEVELOPMENT, BY TYPE OF HOME DEVELOPMENT Zoning Count Row Percentage Row Cumulative Percentage Aware o f B ldg ., Zoning Development Regulations Not Sure of B ldg ., Zoning Development Regulations Not Aware of B ld g ., Zoning, Development Regulations Row Total % o f Total Home | No Home 308 47.5 47.5 122 18.8 66.3 219 33.7 100.0 649 33.2 Seasonal Home 478 67.0 67.0 109 15.3 82.3 126 17.7 100.0 713 36.4 Permanent Home 474 79.8 79.8 58 9.8 89.6 62 10.4 100.0 594 30.4 1,260 64.4 289 14.8 407 20.8 1,956 100.0 Column Total o f Total % Chi-square = 157.41261 with 4 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .19978 Contingency C o e ffic ie n t = .27189 Uncertainty C o e ffic ie n t (asymmetric) = .03541 with Home Development dependent = .04509 with Zoning dependent SpearmansRs = .2821 Kendall Tau = .2660 S t a t is t ic s presented a t the bottom o f Table 47, in d ic a te a r e l a t i o n ­ ship ex ists between property owners with d is s im ila r types o f home develop­ ment and awareness of bu ilding regulations concerning land development. A Chi-square value of 157.41261, s ig n if ic a n t at the .05 p r o b a b ility level, implies property owners with d is s im ila r types o f home development d i f f e r in terms o f awareness of building regulations concerning land development. Other q u a lify in g s t a t i s t i c s (Cramers V, Contingency 133 C o e ffic ie n t, Uncertainty C o e ffic ie n t) indicate the re la tio n s h ip between property owners with d is s im ila r types o f home development and awareness o f building regulations concerning land development is strong. This strong re la tio n s h ip is f u r t h e r supported by a Spearmansps value of .2821 and a Kendall Tau value of .2660 in d ic a tin g as level of home development progresses from none to a seasonal home to a permanent home, knowledge of building regulations increases. S t a t is t ic s support the percentage differences displayed in Table 47, th e re fo re , knowledge of a property owners awareness about building regulations concerning land development g re a tly help to pred ict type of home development the property owner has on the land in the study area. One in te re s tin g re la tio n s h ip occurs between awareness of land development regulations and propensity to s e ll as evidenced by a Kendall Tau value o f .1067 and a Spearmans^ value of .1172. This indicates th a t property owners who intend to s e ll t h e ir land are some­ what more aware of regulations concerning development than are property owners who have no in te n t to s e l l . Land Use Regulations In the past few years, the Michigan le g is la tu r e has t r ie d numerous times to pass land use le g is la t io n . fa ilu re . Usually each attempt has met with Some land use le g is la t io n opponents have taken the stance th a t there are already too many controls r e s t r ic t in g individual freedom. It was s u rp ris in g , th e re fo re , th a t when northern Michigan study area property owners were asked how they f e l t about land use controls th a t only 16.8% said they would lik e to see fewer controls (See Table 48 ). hand, 24.2% would li k e to see s t r i c t e r controls controls are adequate. On the other and 33.6% fe el present 134 Table 48 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS CONCERNING PRESENT LAND USE CONTROLS (Frequency and Percentage D is trib u tio n ) Absolute Frequency Frequency Percentage Cumulative Frequency Percentage Favor S t r i c t e r Land Use Controls 467 24.2 24.2 Present Land Use Controls are Adequate 648 33.6 57.8 Favor Lessening of Present Land Use Controls 201 10.4 68.2 Against a l l Land Use Controls 123 6.4 74.6 Not Sure 489 25.4 100.0 1,928 100.0 LANDREG Total Once again, the portion o f property owners lacking s u f f ic ie n t information to make a judgment was quite high as 25.4% were not sure how they f e l t about land use controls. When considering in dividual counties, em pirical and s t a t i s t i c a l evidence indicates th a t there is no noticeable deviation from the o verall percentage d is tr ib u tio n s . Therefore, no evidence ex is ts to es ta b lis h a re la tio n s h ip between individual counties and a ttitu d e s toward land use controls. When in dividual townships are considered in Table 49, some deviation from the o v erall percentage d is tr ib u tio n are noted. In Dover township, only 12.3% o f the property owners favored s t r i c t e r land use controls and 40.7% were in favor of a few less controls. This contrasted sharply with Blue Lake township, where 29.8% o f the property owners desired more land use controls and 12.3% favored a t le a s t a few less. Chester and Grayling 135 Table 49 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS CONCERNING PRESENT LAND USE CONTROLS, BY TOWNSHIP LANDREG Count Row Percentage Row Cumulative Percentage Favor S tric te r Land Use Controls Favor Present Lessening of Land Use Present Controls Land Use are Adequate Controls Against A ll Land Use Controls Not Sure Row Total % of Total Township South Branch T27NR2W 35 25.5 25.5 38 27.7 53.2 13 9 .5 62.7 5 3.6 66.3 46 33.6 99.9 137 7.1 Gray!ing T27NR2W 6 17.6 17.6 8 23.5 41.1 6 17.6 58.7 6 17.6 76.3 8 23.5 99.8 34 1.8 Grayling T26NR4W 29 27.6 27.6 39 37.1 64.7 13 12.4 77.1 7 6.7 83.8 17 16.2 100.0 105 5.4 Orange T26NR7W 20 17.5 17.5 32 28.1 45.6 13 11.4 57.0 13 11.4 68.4 36 31.6 100.0 114 5.9 Blue Lake T28NR5W 82 29.8 29.8 97 35.3 65.1 19 6.9 72.0 15 5.4 77.4 62 22.5 99.9 275 14.3 G a rfie ld T25NR7W 16 20.2 20.2 20 25.3 45.5 5 6.3 51.8 8 10.1 61.9 30 38.0 99.9 79 4.1 Bagley T30NR3W 240 23.9 23.9 367 36.5 60.4 106 10.5 70.9 49 4.9 75.8 243 24.2 100.0 1,005 52.1 Chester T29NR2W 29 29.6 29.6 27 27.5 57.1 8 8.2 65.3 5 5.1 70.4 29 29.6 100.0 98 5.1 Dover T31NR2W 10 12.3 12.3 20 24.7 37.0 18 22.2 59.2 15 18.5 77.7 18 22.2 99.9 81 4.2 467 24.2 648 33.6 201 10.4 123 6 .4 489 25.4 1,928 100.0 . Column Total % of Total Chi-square = 96.48141 with 32 degrees o f freedom S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .11185 U ncertainty C o e ffic ie n t (asymmetric) = .01380 with Township dependent = .01520 with LANDREG dependent *Due to rounding, percentage t o t a ls may not equal 100. 136 T26NR4W townships also recorded high percentages o f property owners favoring more land use controls (29.6% and 27.6%, re s p e c tiv e ly ). S t a t is t ic s presented a t the bottom o f Table 49, in d ic a te a r e la t io n ­ ship exists between property owners o f c e rta in townships and a ttitu d e s concerning present land use regulations. In p a r t ic u l a r , Grayling T26NR4W, Blue Lake and Chester township property owners, in general, favored s t r i c t e r land use controls than property owners in other townships. A Chi-square value o f 96.48141, s ig n if ic a n t at the .05 p r o b a b ility le v e l, implies property owners o f ce rtain townships d i f f e r in t h e i r a ttitu d e s concerning present land use regulations. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the re la tio n s h ip between property owners o f c e rta in townships and t h e i r awareness o f present land use regulations is moderate. S t a t i s t i c s support the percentage differences displayed in Table 49. I t appears, th e re fo re , th a t knowledge of level o f awareness about present land use regulations w i l l help p re d ic t township o f property lo catio n . When type of home development is considered, the resu lts are q u ite in te r e s tin g . As shown in Table 50, property owners with a seasonal home e x h ib it a tendency to favor s t r i c t e r land use controls (28.9% ), as com­ pared to property owners with permanent homes or no homes in the study area (21.6% each). More permanent home owners favored lessening of present land use controls (24.5%) than e it h e r seasonal home owners (13.7%) or property owners with no home in the study area (13.4%). As before, property owners w ith no home in the study area showed the highest uncertainty levels as 33.0% were not sure how they f e l t about present land use controls. Property owners with seasonal and permanent homes were more c e rta in as to how they f e l t about present land use co n tro ls, but t h e ir uncertainty lev els were s t i l l high a t 22.6% and 20.0%, re s p e c tiv e ly . 137 Table 50 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS CONCERNING PRESENT LAND USE CONTROLS, BY TYPE OF HOME DEVELOPMENT LANDREG Count Row Percentage Row Cumulative Percentage Favor S tric te r Land Use Controls Present Land Use Controls are Adequate Favor Lessening o f Present Land Use Controls Against A ll Land Use Controls Not Sure Row Total % 0 o f Total Home No Home 137 21.6 21.6 202 31.9 53.5 52 8.2 61.7 33 5.2 66.9 209 33.0 99.9 633 33.1 Seasonal Home 202 28.9 28.9 244 34.9 63.8 54 7.7 71.5 42 6.0 77.5 158 22.6 100.1 700 36.6 Permanent Home 125 21.6 21.6 196 33.8 55.4 94 16.2 71.6 48 8.2 79.9 116 20.0 99.9 579 30.3 Column Total 464 24.3 642 33.6 200 10.5 123 6 .4 483 25.3 1,912 100.0 °l o f Total Chi-square = 65.12994 with 8 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .13051 Uncertainty C o e ffic ie n t (asymmetric) = .01481 with Home dependent = .01104 with LANDREG dependent *Due to rounding, percentage to ta ls may not equal 100. S t a t i s t i c s , presented a t the bottom o f Table 50, in d ic a te a r e l a t i o n ­ ship ex ists between property owners with d is s im ila r types o f home develop­ ment and t h e i r a ttitu d e s concerning present land use reg u latio n s . A Chi-square value o f 65.12994, s ig n if ic a n t at the .05 p r o b a b ility le v e l , implies th a t property owners with d is s im ila r types of home development d i f f e r in terms of t h e i r a ttitu d e s concerning present land use regu la­ tio n s . Other q u a lify in g s t a t is t ic s (Cramers V, Uncertainty C o e ffic ie n t) 138 in d ica te the re la tio n s h ip between property owners with d is s im ila r types o f home development and a ttitu d e s concerning present land use regulations is moderate. S t a t is t ic s support the percentage differences displayed in Table 50. Therefore, knowledge of a ttitu d e s concerning present land use regulation w i l l help p red ict type of home development on the land in the study area. When considering location to water resource with a ttitu d e s toward present land use co n tro ls, a d e f in it e re la tio n s h ip is found to e x is t as shown in Table 51. Property owners located on a body of water favored s t r i c t e r land use controls than those not on a body o f water. In a d d itio n , only 14.3% o f the property owners located on a body of water favored lessening o f present land use controls compared to 18.4% fo r property owners not located on a body o f water. Uncertainty was also lower among property owners located on water as opposed to those not on water. S t a t i s t i c s , presented at the bottom of Table 51, in d ica te a re la tio n s h ip exists between property owners located on a water resource and a ttitu d e s concerning present land use regulations. In general, s t a t i s t i c s in d ica te property owners located on a water resource favor s t r i c t e r land use controls. A Chi-square value of 32.21594, s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners located on water d i f f e r in terms of a ttitu d e s concerning present land use regulations than property owners not located on water. Other q u a lify in g s t a t i s t i c s , Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners located on water and t h e i r a t titu d e concerning present land use regulations is moderate. The moderate re la tio n s h ip is fu rth e r supported by a Spearmansps value o f .1147 and a Kendal Tau value of .1072 in d ic a tin g th a t property owners located on a water resource are more l i k e l y to favor s t r i c t e r land use regulations then property owners 139 Table 51 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS CONCERNING PRESENT LAND USE CONTROLS, BY LOCATION TO WATER RESOURCES LANDREG Count Row Percentage Row Cumulative Percentage Favor S tric te r Land Use Controls Present Land Use Controls are Adequate Favor Lessening of Present Land Use Controls Against A ll Land Use Controls Not Sure Row Total % of Total 0NH20 Property Located on Water 239 29.9 29.9 277 34.6 64.5 67 8.4 72.9 47 5.9 78.8 170 21.2 100.0 800 42.1 Property Not Located on Water 223 20.3 20.3 365 33.2 53.5 130 11.8 65.3 73 6 .6 71.9 309 28.1 100.0 1,100 57.9 Column Total 462 24.3 642 33.8 197 10.4 120 6.3 479 25.2 1,900 100.0 % of Total Chi-square = 32.21594 w ith 3 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .13015 Uncertainty C o e ffic ie n t (asymmetric) = .01245 with 0NH20 dependent = .00575 with LANDREG dependent SpearmansRs = .1147 Kendall Tau = .1072 not located on a water resource. Residential Building Residential building in northern lower Michigan is increasing a t a very fa s t r a te . As mentioned in the in tro d u ctio n , the period between 1970-1975 shows extremely large increases o f permanent residents in northern Michigan counties. Many new residents to ru ral area experience what has become to be known as the " la s t in syndrome." That i s , each prospective resident th a t desires to become a member o f a c e rta in community wants to 140 be allowed to s e t t l e there but then does not want anyone else to move in . On the other hand, some property owners view r e s id e n tia l development as economic growth providing a basic work force and supporting expanding local businesses. In the northern Michigan study area, 56.0% of the property owners would li k e to see at le a s t a l i t t l e more re s id e n tia l development with 10.7% favoring a l o t more r e s id e n tia l development (See Table 5 2 ). Uncer­ t a in t y was also high as 21.8% o f the property owners were not sure how they viewed fu ture re s id e n tia l development. Previous research (Marans and Wellman, 1978) reported th a t 40% of northern Michigan residents favored a r e s t r ic te d or no growth policy and only 1 in 15 favored ex­ tensive growth. Results from th is survey indicates growth seems more desirable by more property owners than reported in previous research. Table 52 NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS ATTITUDES TOWARD FUTURE RESIDENTIAL DEVELOPMENT (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage Cumulative Frequency Percentage Would Like a Lot of Development 208 10.7 10.7 Would Like a L i t t l e Development 876 45.3 56.0 Oppose Future Development 430 22.2 78.2 Not Sure 421 21 .8 100.0 1,935 100.0 Bui 1ding Total When in d ivid u a l counties are considered in Table 53, some deviation from the o v erall percentage d is tr ib u tio n is noted. Kalkaska property 141 Table 53 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD FUTURE RESIDENTIAL DEVELOPMENT, BY COUNTY BUILDING Would Like Count Row Percentage a Lot More Row Cumulative Residential Percentage Development Would Like a L ittle Oppose More Any Future Residential Residential Development Development Not Sure Row Total % of Total County Crawford 34 12.4 12.4 124 45.4 57.8 66 24.2 82.0 49 17.9 99.9 273 14.1 Kalkaska 24 5 .2 5.2 198 42.6 47.8 124 26.7 74.5 119 25.6 100.1 465 24.0 Otsego 150 12.5 12.5 553 46.2 58.7 240 20.1 78.8 253 21.1 99.9 1,196 61.8 Column Total % o f Total 208 10.7 875 45.2 430 22.2 421 21 .8 1,934 100.0 Chi-square = 31.11519 with 6 degrees of freedom S ig n ific a n t at .05 p r o b a b ility level Cramers V = .08967 Uncertainty C o e ffic ie n t (asymmetric) = .00957 with County dependent = .00693 with Building dependent *Due to rounding, percentage to ta ls may not equal 100. owners are s l i g h t l y more opposed to fu tu re re s id e n tia l development (26.7%) than e it h e r Crawford or Otsego (24.2% and 20.1%, resp ective ly) county property owners. Also, in Kalkaska county only 5.2% of the property owners desired a l o t more development compared to 12.5% in Otsego and 12.4% in Crawford county. S t a t i s t i c s , presented a t the bottom of Table 53, in d ic a te a re la tio n s h ip e x is ts between property owners o f ce rtain counties and a ttitu d e s towards fu tu re r e s id e n tia l development. A Chi-square value 142 o f 31.11519, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners o f c e rta in counties d i f f e r in terms o f a ttitu d e s toward fu ture re s id e n tia l development. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in dicate the re la tio n s h ip between property owners of ce rtain counties and a ttitu d e s toward fu tu re re s id e n tia l development is weak. displayed in Table 53. S t a t is t ic s support the percentage d iffe re n c e s , However, due to the weak nature o f the r e la t io n ­ ship, knowledge o f a t t it u d e toward fu tu re r e s id e n tia l development w i l l help l i t t l e to p red ict county of property lo catio n . When individ ual townships are considered in Table 54, there is a l o t more opposition to re s id e n tia l development than previously noted. One th ir d o f the property owners in both Chester and Dover townships (33.3% in each), opposed any fu ture r e s id e n tia l development. G a rfie ld and South Branch property owners followed close behind in t h e i r opposi­ tio n to fu tu re re s id e n tia l development (32.5% and 30.9%, r e s p e c tiv e ly ). Bagley, Grayling T27NR2W and Grayling T26NR4W township property owners were, in general, more favorable to a t le a s t a l i t t l e more re s id e n tia l development than a l l other townships. S t a t i s t i c s , presented a t the bottom of Table 54, in d ic a te a re la tio n s h ip exists between property owners o f c e rta in townships and a t t it u d e toward fu ture re s id e n tia l development. In p a r t i c u l a r , s t a t i s ­ tic s in dicate more property owners in Grayling T26NR4W township wanted more re s id e n tia l development than property owners in other townships. A Chi-square value o f 91.24037, s ig n if ic a n t a t the .05 p r o b a b ility le v e l , implies property owners of c e rta in townships d i f f e r in terms of a ttitu d e s toward fu tu re r e s id e n tia l development. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the r e la t io n ­ ship between property owners of c e rta in townships and a ttitu d e s toward 143 Table 54 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD FUTURE RESIDENTIAL DEVELOPMENT, BY TOWNSHIP BUILDING Would Like a Lot More Residential Development Would Like a L ittle More Residential Development Oppose any Future Residential Development Not Sure South Branch T25NR2W 13 9.6 9 .6 56 41.2 50.8 42 30.9 81.7 25 18.4 100.1 136 7.0 Grayling G27NR2W 7 21.2 21.3 12 36.4 57.6 5 15.1 72.7 9 27.3 100.0 33 1.7 Grayling T26NR4W 14 13.3 13.3 57 54.3 67.6 19 18.1 85.7 15 14.3 100.0 105 5.4 Orange T26NR7W 8 7.0 7.0 42 36.8 43.8 30 26.3 70.1 34 29.8 99.9 114 5.9 12 4.5 4.5 130 48.3 52.8 68 25.3 78.1 59 21.9 100.0 269 13.9 3 3.7 3.7 25 31.2 34.9 26 32.5 67.4 26 32.5 99.9 80 4.1 Bagley T30NR3W 142 13.9 13.9 483 47.3 61.2 181 17.7 78.9 215 21.1 100.0 1,021 52.8 Chester T29NR2W 1 1.0 1.0 41 42.7 43.7 32 33.3 77.0 22 22.9 99.9 96 5.0 Dover T31NR2W 8 9.9 9.9 30 37.0 46.9 27 33.3 80.2 16 19.7 99.9 81 4.2 208 10.7 876 45.3 430 22.2 421 2 1 .8 1,935 100.0 Count Row Percentage Row Cumulative Percentage Row Total % o f Total Township 1 Blue Lake T28NR5W G a rfie ld T25NR7W Column Total % o f Total Chi-square = 91.24037 with 24 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .12537 Uncertainty C o e ffic ie n t (asymmetric) = .01575 with Township dependent = .02001 with Building dependent *Due to rounding, percentage t o t a ls may not equal 100. 144 fu ture re s id e n tia l development is moderate. S t a t is t ic s support the percentage d iffe re n c e s , displayed in Table 54, th e re fo re , knowledge of a t titu d e toward fu tu re r e s id e n tia l development w i l l help p re d ic t township o f property lo c a tio n . When type o f home development is considered in Table 55, the resu lts are quite in te r e s tin g . Opposition to fu ture r e s id e n tia l development centers p rim a rily w ith seasonal home owners. Permanent home owners also opposed fu tu re r e s id e n tia l development but not q u ite so strongly as property owners with seasonal homes. As expected, more property owners with no home in the study area desired (16.6%) a l o t more re s id e n tia l development than e it h e r property owners with permanent homes (11.5%) or property owners with seasonal homes (5.0% ). Previous research (Marans and Wellman, 1978) also reported seasonal home owners as having more resistance to fu tu re growth and development than permanent home owners. S t a t i s t i c s , presented a t the bottom o f Table 55, in d ic a te a re la tio n s h ip ex ists between property owners with d is s im ila r types o f home development and a ttitu d e s toward fu tu re r e s id e n tia l develop­ ment. A Chi-square value o f 125.3369, s ig n if ic a n t a t the .05 pro­ b a b i l i t y le v e l , implies property owners with d is s im ila r types o f home development d i f f e r in terms o f a ttitu d e s toward fu tu re re s id e n tia l development. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty Co­ e f f i c i e n t ) in d ic a te the re la tio n s h ip between property owners with d is s im ila r types of home development and a ttitu d e s toward fu tu re r e s i­ dential development is moderate. This moderate re la tio n s h ip is fu rth e r supported by a SpearmansRs value o f -.1222 and a Kendall Tau value of -.1 1 0 4 in d ic a tin g th a t as level of home development proceeds from no home to permanent home, and then to seasonal home, th a t property owners 145 T able 55 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD FUTURE RESIDENTIAL DEVELOPMENT, BY TYPE OF HOME DEVELOPMENT BUILDING | Count Row Percentage Row Cumulative Percentage Would Like Would Like A Oppose A Lot More Any Future L i t t l e More Residential Residential Residential Development Development Development Not Sure Row Total % o f Total Home j No Home 106 16.6 16.6 293 45.8 62.4 71 11.1 73.5 170 26.6 100.1 640 33.3 Seasonal Home 35 5.0 5.0 299 43.0 48.0 230 33.0 81.0 132 19.0 100.0 696 36.3 Permanent Home 67 11.5 11.5 275 47.2 58.7 126 21.6 80.3 115 19.7 100.0 583 30.4 208 10.8 867 45.2 427 10.8 417 21.7 1,919 100.0 Column Total % o f Total Chi-square = 125.3369 with 6 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility level Cramers V = .13051 Uncertainty C o e ffic ie n t (asymmetric) = .03123 with Home dependent = .02794 w ith Building dependent Spearmans^ = -.1222 Kendall Tau = -.1 1 0 4 *Due to rounding, percentage to ta ls may not equal 100. are more l i k e l y to oppose fu tu re r e s id e n tia l development. Future Property Values The ris in g property values experienced in the la s t few yeare are seen as continuing a t le a s t in to the middle 1980's. Expansion of values is not only a phenomenon of metropolitan areas as evidenced by high value per acre assessments given by northern Michigan study area property owners (explained in greater d e ta il in the next s e c tio n ). I t comes as no surprise 146 to learn th a t 80.8% o f the property owners in the northern Michigan study area expect property values to increase a t le a s t moderately. Only 5.9% fe el property values w i l l stay the same and even less (2.4%) fe el property values w i l l decline (See Table 5 6 ). Table 56 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD FUTURE PROPERTY VALUES (Frequency and Percentage D is trib u tio n ) Frequency Frequency Percentage Cumulative Frequency Percentage 364 18.7 18.7 1,211 62.1 80.8 115 5.9 86.7 Decrease Moderately 34 1.7 88.4 Decrease Radically 14 .7 89.1 212 10.9 100.0 1,950 100.0 Property Value PROPVALUE Increase Radically Increase Moderately Stay Same Not Sure Total When in d iv id u a l counties are considered in Table 57, there is only one noticeable d iffe re n c e . Crawford and Kalkaska counties have s l i g h t l y sm aller percentages (77.2% and 75.2%, respectively) of property owners who fe el property values w i l l increase at le a s t moderately than in Otsego county (83.7%). S t a t i s t i c s , presented a t the bottom o f Table 57, in d ic a te a r e la t io n ­ ship ex ists between property owners of c e rta in counties and a ttitu d e s toward fu tu re property values. A Chi-square value of 29.95991, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners o f c e rta in counties Table 57 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD FUTURE PROPERTY VALUES, BY COUNTY PROPVALU | Count Row Percentage Row Cumulative Percentage Property Values W ill Increase Radically Property Values W ill Increase Moderately Property Values W ill Stay The Same Property Values W ill Decrease Moderately Property Values W ill Decrease Radically 46 16.6 16.6 168 60.6 77.2 21 7.6 84.8 5 1.8 86.6 2 100 21.3 21.3 253 53.9 75.2 33 7.0 82.2 15 3.2 85.4 218 18.1 18.1 790 65.6 83.7 61 5.1 88.8 14 1.2 90.0 364 18.7 1,211 62.1 115 5.9 34 1.7 Not Sure Row Total % of Total County Crawford Kalkaska Otsego Column Total % o f Total .7 87.3 3 .6 86.0 9 .7 90.7 14 .7 35 12.6 99.9 14.2 65 13.9 99.9 469 24.0 112 9.3 100.0 1,204 61.7 212 10.9 1,950 100.0 Chi-square = 29.95991 with 10 degrees o f freedom S ig n ific a n t at .05 p ro b a b ility level Cramers V = .07155 Uncertainty C o e ffic ie n t (asymmetric) = .00815 with County dependent = .00669 with PROPVALU dependent *Due to rounding, percentage to ta ls may not equal 100. 111 148 * d i f f e r in terms of a ttitu d e s toward fu ture property values. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners of c e rta in counties and a ttitu d e s toward fu ture property values is weak. S t a t is t ic s support the percentage differences displayed in Table 57. However, because of the weak re la tio n s h ip , knowledge o f a ttitu d e toward fu tu re property values w i l l help l i t t l e to pred ict county o f property lo c a tio n . When in d ivid ual townships are considered in Table 58, two substantial deviations from the overall percentage d is t r ib u t io n are noted. G a rfie ld and Grayling T27NR2W townships have many fewer pro­ perty owners who fe el property values w i l l increase a t le a s t moderately than the other townships (59.2% and 54.5%, re s p e c tiv e ly ). This decrease is not taken up by a corresponding increase in the number o f property owners who fe el property values w i l l d e clin e, instead the percentage of property owners who are not sure o f fu tu re property values is higher. In G a rfie ld township 24.7% of property owners are not sure of the fu tu re of property values and in Grayling T27NR2W township 21.2% are not sure. There is obviously a good deal o f uncertainty operating in both G a rfie ld and Grayling T27NR2W townships concerning fu ture property values. S t a t i s t i c s , presented at the bottom o f Table 58, in d ic a te a re la tio n s h ip exists between property owners of c e rta in townships and a ttitu d e s toward fu ture property values. A Chi-square value o f 77.7444, s ig n if ic a n t a t the .05 p r o b a b ility le v e l, implies property owners o f c e rta in townships d i f f e r in terms of a ttitu d e s toward fu tu re property values. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ica te the re la tio n s h ip between property owners o f c e rta in townships Table 58 ATTITUDES OF NORTHERN MICHIGAN STUDY AREA PROPERTY OWNERS TOWARD FUTURE PROPERTY VALUES, BY TOWNSHIP PROPVALU Count Row Percentage Row Cumulative Percentage Property Values W ill Stay The Same Property Values W ill Decrease Moderately Property Values W ill Decrease Radically Property Values W ill Increase Radically Property Values W ill Increase Moderately South Branch T25NR2W 22 16.1 16.1 79 57.7 73.8 11 8.0 81.8 2 1.5 83.3 .7 84.0 22 16.1 100.0 137 7.0 Grayling T27NR2W 4 12.1 12.1 18 54.5 66.6 3 9.1 75.7 0 0 75.7 1 3.0 78.7 7 21.2 99.9 33 1.7 Grayli ng T26NR4W 20 18.5 18.5 72 66.7 85.2 7 6.5 91.7 3 2 .8 94.5 0 0 94.5 6 5.5 100.0 108 5.5 Orange T26NR7W 23 20.2 20.2 61 53.5 73.7 8 7.0 80.7 6 5.3 86.0 1 15 13.2 100.1 114 5.8 Blue Lake T28NR5W 60 22.1 22.1 159 58.5 80.6 15 5.5 86.1 6 2.2 88.3 .7 89.0 30 11.0 100.0 272 13.9 G arfie ld T25NR7W 16 19.7 19.7 32 39.5 59.2 10 12.3 71.5 3 3.7 75.2 0 0 75.2 20 24.7 100.1 81 4.1 Bagley T30NR3W 189 18.4 18.4 671 65.5 83.9 55 5.3 89.2 13 1.3 90.5 8 90 8 .8 100.1 1,025 52.5 Not Sure Row Total % of Total Township . 1 .9 86.9 2 .8 91.3 Table 58 - C o n t'd . Count Row Percentage Row Cumulative Percentage PROPVALU Property Values W ill Increase Radically Property Values W ill Increase Moderately Property Values W ill Stay The Same Property Values W ill Decrease Moderately Property Values W ill Decrease Radically Not Sure Row Total % of Total Township Chester T29NR2W 15 15.3 15.3 69 70.4 85.7 5 5.1 90.8 0 0 90.8 1 1.0 91.8 8 8.2 100.0 98 5.0 Dover T31NR2W 15 18.1 18.1 51 61.4 79.5 2 2.4 81.9 1 1.2 83.1 0 0 83.1 14 16.9 100.0 83 4.2 364 18.7 1,212 62.1 115 5.9 34 1.7 14 212 10.9 1,951 100.0 Column Total % o f Total .7 Chi-square = 77.7444 with 40 degrees of freedom S ig n ific a n t at .05 p ro b a b ility Level. Cramers V = .08927 Uncertainty C o e ffic ie n t (asymmetric) = .01164 with Township dependent = .01670 with PROPVALU dependent *Due to rounding, percentage to ta ls may not equal 100. 151 and a ttitu d e s toward fu tu re property values is weak. the percentage differen ce displayed in Table 58. S t a t is t ic s support However, bacause of the weak r e la tio n s h ip , knowledge o f a t titu d e toward fu ture property values helps l i t t l e to p red ict township o f property location. When type of home development is considered, almost no d iffe re n c e is noted e it h e r e m p iric a lly or s t a t i s t i c a l l y . A ttitud es toward fu tu re property values does not seem to be related to type of home development. Summary f This chapter examined a ttitu d e s on issues o f concern to northern Michigan study area property owners. A ttitud es on property tax lev e ls to awareness of land use regulations were s o lic it e d . The analysis indicated th a t: 1) About tw o-thirds of the property owners f e l t property tax lev els were too high. In Kalkaska county over f o u r - f i f t h s o f the property owners f e l t property tax lev e ls were high. 2) The q u a lit y o f municipal services provided was, in general, perceived as average or good. Property owners with no home development in the study area were generally uncertain about the q u a lity o f services provided. 3) One-third o f the property owners f e l t the quantity of services provided was about r ig h t with an additional one-quarter desiring a t le a s t a few more services. Over h a lf o f the property owners with no home in the study area were uncertain how they f e l t about the quantity of services Drovided. 152 4) O n e -fifth o f northern Michigan study area property owners are not aware of b u ild in g , zoning, or other land use regu­ la tio n s concerning t h e i r property. An additional 15% are not sure of b u ild in g , zoning, or other land use regulations a ffe c tin g t h e i r land. Property owners with permanent and seasonal homes were much more aware of land use regulations than property owners with no home development in the study area. 5) One-third o f the property owners f e l t th a t present land use controls are adequate, however, an ad ditional one-fourth favor s t r i c t e r land use co ntrols. Seasonal home owners are the ones most in favor of s t r i c t e r land use controls, and property owners with no home development in the study area are generally not sure how they fe el about land use controls. 6) O n e -fifth o f northern Michigan study area property owners oppose any fu tu re re s id e n tia l development. Property owners in Kalkaska county are more opposed to fu ture re s id e n tia l development than those in Crawford or Otsego county. One- th ir d of the seasonal home owners are against fu tu re residen­ t i a l development as opposed to permanent homeowners, who were more l i k e l y to favor more r e s id e n tia l development. 7) The overwhelming m ajo rity of property owners envision property values increasing a t le a s t moderately with almost o n e - f if t h expecting property values to increase r a d ic a lly w ith in the near fu tu re . CHAPTER V II INFLUENCE OF NATURAL RESOURCES ON PROPERTY LOCATION AND VALUE Influence of Water Resources Water resources play an important ro le f o r many people in determining where to vacation and many times where to l i v e . State tourism agencies emphasize advantages o f vacationing in the "Water Winter Wonderland" or "The Land o f 10,000 Lakes". No mention is made of the mosquito or black f l y population near the 10,000 lakes, ra th e r they are portrayed as great places to v i s i t and enjoy. Many states also boast o f w ild and b e a u tifu l riv e rs a v a ila b le fo r r a f t i n g , fis h in g , swimming and many other recrea tio n al pursuits. Not only are water resources viewed as a great place to v i s i t , they are also advertised as a wonderful place to own property and l i v e . Numerous developments occur around man-made lakes ' where prospective buyers are informed of a l l the wonderful advantages o f lake liv in g or of the appreciation value o f water property. Water resources are viewed by many as a great a ttr a c tio n influencing property location decisions (Nelson, 1973; Tombaugh, 1967). Opportunities f o r obtaining land on water resources are re a d ily a v a ila b le in the study area. In f a c t , the type of water resources prevalent in a township was a determining fa c to r when se lecting townships fo r the s t r a t i f i e d random sample (See Chapter I I ) . The abundance of water resources in northern Michigan is brought out by the f a c t th a t 41.8% of the property owned in the study area is located on some water resource. When in d ivid ual counties are considered in Table 59, Kalkaska has the highest percentage of property owners with land on some type of water resource (56.8%). Crawford and Otsego counties have much lower pencentages of property owners with land on some type o f water resource (35.8% and 37.2%, re s p e c tiv e ly ). 153 154 S t a t i s t i c s , presented a t the bottom o f Table 59, in dicate a rela tio n s h ip ex is ts between property owners o f c e rta in counties and property location to water resources. A Chi-square value of 58.61954, s ig n ific a n t a t the .05 p r o b a b ility le v e l, implies property owners of c e rtain counties d i f f e r in terms o f property location to water resources. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) indicate the re la tio n s h ip between property owners o f c e rta in counties and property location to water resources is moderate. centage differen ces displayed in Table 55. S t a t is t ic s support the per­ Therefore, knowledge of property lo cation to water resources w i l l help p red ict county of pro­ perty lo catio n . Table 59 LOCATION OF PROPERTY TO WATER RESOURCES, BY COUNTY 0NH20 Count Row Percentage Row Cumulative Percentage Property on a Water Resource Property Not on a Water Resource Row Total % of Total Crawford 100 35.8 35.8 179 64.2 100.0 279 14.0 Kalkaska 273 56.8 56.8 208 43.2 100.0 481 24.2 Otsego 456 37.2 37.2 769 62.8 100.0 1,225 61.7 Column Total % of Total 829 41.8 1,156 58.2 1,985 100.0 County Chi-sqaure = 58.61954 w ith 2 degrees o f freedom S ig n ific a n t at .05 p r o b a b ility le v e l. Cramers V = .17180 Uncertainty C o e ffic ie n t (asymmetric) = .01593 with County dependent = .02149 with 0NH20 dependent 155 When in dividual townships are considered in Table 60, the resu lts d i f f e r somewhat from what was expected. Remembering th a t the s t r a t i f i e d random sample was based on selecting fo r each county a lake township, a r iv e r township, and a township with no major waterway (See Chapter I I ) survey results showed th a t the s t r a t i f i c a t i o n procedure was not e n t ir e ly successful. The three townships selected fo r Crawford county show the best conformity to the s t r a t i f i c a t i o n procedure. The percentage of property owners whose land is located on a body o f water is highest in Grayling T26NR2W township where 55.0% of the property owners have land on some type o f water. Grayling T26NR4W was selected as a "lake" town­ ship and of a l l the property owners whose land is located on a body of water only 46.7% had land on a lake whereas 50% had land on a r i v e r . South Branch township, selected as a "riv e r" township, had 24.6% o f i t s property owners located on a body o f water and o f th a t number, 82.9% were on a r i v e r . Therefore, the s t r a t i f i c a t i o n procedure worked much b e tte r f o r South Branch township. Grayling T27NR2W, selected as a town­ ship with no major water resource, had only 21.2% of i t s property owners located on a body of water. In Kalkaska county, Blue Lake township, selected as a "lake" township, has 78.2% o f i t s property owners located on a body o f water and 100% of those are located on a lake. Obviously, the s t r a t i f i c a t i o n procedure which chose Blue Lake as a lake township gave nearly pe rfect resu lts in th is case. G a rfie ld township, selected as a "riv e r" township has only 12.0% of i t s property owners located on a body of water but 81.8% o f those property owners are located on a r iv e r . In Orange town­ ship, the s t r a t i f i c a t i o n procedure was not e f f e c t iv e as 37.1% of the property owners are located on a body o f water and 90.7% o f those located 156 Table 60 LOCATION OF PROPERTY TO WATER RESOURCES, BY TOWNSHIP 0NH20 Count Row Percentage Row Cumulative Percentage Property Not Located on Water Property Located on Water Row Total % of Total Township | 104 75.4 75.4 34 24.6 100.0 138 6.9 Gray!i ng T27NR2W 26 78.8 78.8 7 21.2 100.0 33 1.7 Grayling T26NR4W 49 44.9 44.9 60 55.0 99.9 109 5.0 Orange T26NR7W 73 62.9 62.9 43 37.1 100.0 116 5 .8 Blue Lake T28NR5W 61 21 .8 21 .8 219 78.2 100.0 280 14.1 G arfie ld T25NR7W 73 87.9 87.9 10 12.0 99.9 83 4.2 Bagley T30NR3W 665 63.7 63.7 379 36.3 100.0 1,047 52.6 Chester T29NR2W 41 41.0 41.0 59 59.0 100.0 100 5.0 Dover T31NR2W 64 77.1 77.1 19 22.9 100.0 83 4.2 1,156 58.2 830 41.8 1,986 100.0 South Branch T25NR2W Column Total % of Total Chi-square = 251.54324 with 8 degrees of freedom S ig n ific a n t a t .05 p ro b a b ility le v e l. Cramers V = .35589 Uncertainty C o e ffic ie n t (asymmetric) = .04085 with Township dependent = .09671 with0NH20 dependent *Due to rounding, percentage t o t a ls may not equal 100. 157 on a lake. These are high levels f o r a township selected fo r i t s lack of water resources. Otsego county has probably the most devastating resu lts of a l l counties when i t comes to selecting water s t r a t a fo r random samples. Bagley, selected as a lake county has 36.3% o f i t s property owners located on a body o f water and 95.5% of those are located on a lake. The problem is Chester township which was considered lacking any major water resources. A to ta l o f 59.0% o f the property owners in Chester township had property located on some type o f water and 72.9% o f those were located on a lake. Obviously, these are very high percentages fo r a township considered lacking a major water resource. Dover township has 22.9% of i t s property owners with land located on water and 55.5% of these are located on a r i v e r with an a d d itio n a l 11.1% located on both a r iv e r and a lake. I t is obvious from the preceeding discussion th a t s t r a t i f i c a t i o n through the use o f p la t maps is hazardous. In the case o f Blue Lake township, the s t r a t i f i c a t i o n procedure worked q u ite w e ll. of Chester township, the procedure was w oefu lly inadequate. In the case The problem was in te r p r e ta tio n o f p la t maps to accurately estimate in d ivid u a l lo ts . Large tra c ts o f land are e a s ily distin guishable but numerous small tr a c ts , usually around lakes or r iv e r s , are represented by small dots. Absence of d e ta il l o t representation precludes determination o f the amount or size o f lots in some areas by visual inspection o f p l a t maps. In fu tu re research, where natural resource s t r a t i f i c a t i o n is desired, p la t maps should only be used as a prelim inary mechanism fo r s e lec tio n with f in a l determination o f areas to be surveyed made a f t e r visual inspection of the area. In a d d itio n , information obtained from county personnel would be h e lp f u l. 158 S t a t i s t i c s , presented at the bottom of Table 60, in d ica te a r e la t io n ­ ship ex ists between property owners o f c e rta in townships and property loca­ tio n r e la t iv e to water resources. erences displayed in Table 60. S t a t is t ic s support the percentage d i f f ­ Therefore, knowledge of property location to water resource w i l l g re a tly help p re d ic t township o f property lo c a tio n . There is also a re la tio n s h ip between the type of water a property owner is located on and individ ual townships. As expected, c e rta in town­ ships have more property located on lakes ( i . e . Blue Lake), or on riv e rs ( i . e . South Branch) than other townships (See Table 60). However, due to the fa c t th a t townships were s t r a t i f i e d and then selected because of t h e i r water resource type, the re la tio n s h ip should be stronger to j u s t i f y the procedure used to choose the s tra ta fo r the random sample. S t a t i s t i c s , presented at the bottom of Table 61, in d ica te a r e la t io n ­ ship exists between property owners of c e rta in townships and property location on c e rta in types o f water resource. A Chi-square value of 395.45014, s ig n if ic a n t a t the .05 p r o b a b ility l e v e l , implies property owners of c e rta in townships d i f f e r in terms of property location on c e rta in types o f water resources. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the re la tio n s h ip between property owners of c e rta in townships and property lo cation on c e rta in types o f water resources is very strong. Therefore, a ce rtain type o f water resource w i l l knowledge o f property location on g re a tly help p re d ic t township of property lo c a tio n . Influence of Public Land In northern Michigan there e x is t ownership. large tra c ts o f land in public S tate and national forests make up the bulk of p u b lic ly owned land, but c i t i e s , counties, townships, and school d i s t r i c t s own 159 Table 61 TYPE OF WATER RESOURCE PROPERTY IS LOCATED, BY TOWNSHIP TYPEH20 | Count Row Percentage Row Cumulative Percentage Row Total Lake or Pond River Both % o f Total South Branch T25NR2W 6 17.1 17.1 29 82.9 100.0 0 0 100.0 35 4.2 Grayling T27NR2W 4 57.1 57.1 3 42.9 100.0 0 0 100.0 7 Grayling T26NR4W 28 46.7 46.7 30 50.0 96.7 2 3.3 100.0 60 7.2 Orange T26NR7W 39 90.7 90.7 4 9 .3 100.0 0 0 . 100.0 43 5.2 219 100.0 100.0 0 0 100.0 0 0 100.0 219 26.3 1 9.1 9.1 9 81.8 90.9 1 9.1 100.0 11 1.3 Bagley T30NR3W 362 95.5 95.5 16 4 .2 99.7 1 .3 100.0 379 45.6 Chester T29NR2W 43 72.9 72.9 15 25.4 98.3 1 1.7 100.1 59 7.1 Dover T31NR3W 6 33.3 33.3 10 55.5 88 .8 2 11.1 99.9 18 2.2 708 85.2 116 14.0 7 831 100.0 Township Blue Lake T28NR5W G a rfie ld T25NR7W Column Total % o f Total .8 .8 Chi-square = 395.45014 v/ith 16 degrees of freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .48779 Uncertainty C o e ffic ie n t (asymmetric) = .12181 w ith Township dependent = .41923 w ith TYPEH20 dependent *Due to rounding, percentage t o t a ls may not equal 100. 1 60 acreage scattered throughout the area. Also, some areas may have many acres in public ownership and some may have few,'such as in the case of Grayling T27NR2W or Chester townships, which are almost completely p u b lic ly owned, and Bagley township which has few areas in public ownership. As presented in Table 62, 18.2% of the property owners in the study area have public property touching t h e ir property on a t le a s t one side. When in d ivid u a l counties are considered in Table 63, the differences become s u b s ta n tia l. A much la rg e r percentage (39.6%) of property owners in Crawford county have property adjacent to public land than in e ith e r Kalkaska or Otsego counties (21.2% and 12.0%, r e s p e c tiv e ly ). A c tu a lly , th is d iffe re n c e is not surprising because 66.73% of the to ta l land in Crawford county, 42.41% in Kalkaska county, and 27.64% in Otsego county is in public ownership. (Michigan S t a t i s t i c a l A bstract, 1978, p. 724-727). Table 62 LOCATION OF PRIVATE PROPERTY TO PUBLIC LAND (Frequency and Percentage D is trib u tio n ) PUBPROP P riv a te Property Adjacent to Public Land Frequency Frequency Percentage 355 18.2 P riv ate Property Not Adjacent to Public Land 1,598 81.8 Total 1,953 100.0 S t a t i s t i c s , presented a t the bottom o f Table 63, in d ica te a re la tio n s h ip ex ists between property owners o f c e rta in counties and location o f p riv a te property to adjacent public property. This s t a t i s t i c a l 161 Table 63 LOCATION OF PRIVATE PROPERTY TO PUBLIC LAND, BY COUNTY PUBPROP Count Row Percentage Row Cumulative Percentage P riv ate Land Adjacent To Public Land P rivate Land Not Adjacent To Public Land Row Total % o f Total County Crawford 111 39.6 39.6 169 60.4 100.0 280 14.3 Kalkaska 101 21.2 21.2 376 78.8 100.0 477 24.4 Otsego 143 12.0 12.0 1,052 88.0 100.0 1,195 61.2 Column Total 355 18.2 1,597 81.8 1,952 100.0 Chi-square = 120.74314 with 2 degrees o f freedom S ig n ific a n t a t .05 p ro b a b ility level Cramers V = .24865 Uncertainty C o e ffic ie n t = .02983 with County dependent = .05807 with PUBPROP dependent relatio n sh ip is not unusual rath er i t was expected based on a previous chapters discussion o f each counties public land acreage. A Chi-square value of 120.7434, s ig n if ic a n t at the .05 p r o b a b ility l e v e l , implies property owners o f c e rta in counties d i f f e r in terms o f location of p riv a te property to adjacent public property. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in dicate the re la tio n s h ip between property owners o f c e rta in counties and location o f p riv a te property to adjacent public property is strong. ences displayed in Table 63. S t a t is t ic s support percentage d i f f e r ­ Therefore, knowledge o f p r iv a te property location to public property w i l l g re a tly help p re d ic t county of property lo catio n . 162 When in d ivid ual townships are considered in Table 64, a good deal o f v a ria tio n is noted. Grayling T27NR2W township has the largest percentage o f property owners (61.8%) whose land is adjacent to public land. Chester township has the second larg e st percentage (43.0%) followed closely by South Branch and Grayling T26NR4W (38.0% and 34.5%, resp ective ly) townships. By co n tras t, only 2.4% of the property owners in Dover township have land adjacent to public land. Orange (8.6%) and Bagley (9.7%) townships follow closely behind Dover. S t a t i s t i c s , presented a t the bottom of Table 64, indicate a re la tio n s h ip ex ists between property owners of c e rta in townships and location o f p riv a te property to public property. A Chi-square value o f 120.74314, s ig n if ic a n t at the .05 p r o b a b ility le v e l , implies property owners of c e rta in townships d i f f e r in terms o f location o f p riva te property to public property. Other q u a lify in g s t a t i s t i c s (Cramers V, Uncertainty C o e ffic ie n t) in d ic a te the re la tio n s h ip between property owners o f c e rta in townships and location of p riv a te property to public property is very strong. displayed in Table 64. S t a t is t ic s support percentage differen ces , Therefore, knowledge of location of p riv a te property to public property w i l l g re a tly help p red ict township of property lo c a tio n . Acreage Owned Per Property Owner The mean to ta l acreage owned, per property owner, in the study area was reported a t 17.516 acres. fig u r e to be highly misleading. However, fu rth e r analysis indicates th is A skewness value o f 22.84 indicates th a t many o f the cases analyzed are clustered to the l e f t (below) of the mean with most of the extreme cases to the r ig h t (above). In a d d itio n , a kurtosis value of 694.723 indicates th a t the curve defined 163 Table 64 LOCATION OF PRIVATE LAND TO PUBLIC LAND, BY TOWNSHIP PUBPROP | Count Row Percentage Row Cumulative Percentage P riv a te Property Adjacent to Public Land P riv a te Property Not Adjacent to Public Land South Branch T25NR2W 52 38.0 38.0 85 62.0 100.0 137 7.0 Grayling T27NR2W 21 61.8 61.8 13 38.2 100.0 34 1.7 Grayli ng T26NR4W 38 34.5 34.5 72 65.4 99.9 Orange T26NR7W 10 8 .6 8 .6 106 91.4 100.0 116 5.9 Blue Lake T28NR5W 75 27.0 27.0 203 73.0 100.0 278 14.2 G a rfie ld T25NR7W 16 19.7 19.7 65 80.2 99.9 81 4.1 Bagley T30NR3W 98 9 .7 9 .7 915 90.3 100.0 1,013 51.9 Chester T29NR2W 43 43.0 43.0 57 57.0 100.0 100 5.1 Dover T31NR2W 2 2 .4 2 .4 82 97.6 100.0 84 4.3 355 18.2 1,598 81.8 1,953 100.0 Row Total of Total %o Township I Column Total % of Total no 5.6 Chi-square = 255.78325 with 8 degrees o f freedom S ig n ific a n t a t .05 p r o b a b ility le v e l. Cramers V = .34001 Uncertainty C o e ffic ie n t (asymmetric) = .03317 w ith Township dependent = .11363 with PUBPROP dependent *Due to rounding, percentage t o t a ls may not equal 100. 164 by the d is tr ib u tio n of cases is peaked. Therefore, the acreage owned by the m ajo rity o f property owners is generally smaller than the mean of 17.516. Two other s t a t i s t i c s (mode and median) used to describe variables also in dicate fo r the m a jo rity of property owners the amount of land owned is less than the mean. The median value is 1.004 meaning at le a s t 50% o f the property owners own one acre or less. The mode value o f .50 in d ic a tin g the most frequent acreage size owned is only one h a lf acre. One other s t a t i s t i c is q u ite important and th at is the 95% Confidence In te rv a l about the mean. For study area property owners, the 95% confidence in te rv a l about the mean ranges from 13.042 acres to 21.99 acres. When in d ivid ual counties are considered in Table 65, there is l i t t l e deviation from the overall sample mean. county own an average o f 20.113 acres. Property owners in Crawford In Kalkaska county, the average acreage owned is 13.884 and in Otsego county, the average is 18.360. A ll three counties, th e re fo re , f a l l w ith in the 95% confidence in te rv a l fo r the overall mean acreage owned. However, there does seem to be q u ite a d iffe re n c e between the mean fo r Crawford county and the mean fo r Kalkaska county. s ig n if ic a n t . This d iffe re n c e was tested and was found to be Therefore, mean acreage owned in Crawford county is s ig ­ n i f i c a n t l y higher than in Kalkaska county. When in d ivid u a l townships are considered in Table 66, a great deal of v a ria tio n is noted. Six of the nine township means fo r acreage owned do not f a l l w ith in the 95% confidence in te rv a l lim its f o r the o v erall sample mean. In Dover township, the average acre size owned is 119.477 which d i f f e r s markedly from Bagley township where the average property size is only 9.1056 acres. Blue Lake township also has a low 165 4 Table 65 TOTAL ACREAGE OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER, BY COUNTY Size of Sample Mean Standard Deviation Crawford 267 20.113 69.395 Kalkaska 456 13.884 40.657 Otsego 1,141 18.360 118.571 Total 1,864 17.516 98.075 County t value t value t value fo r Mean of f o r Mean of fo r Mean o f Crawford vs. Mean o f Kalkaska = 2 . 2 7 * Crawford vs. Mean of Otsego = 1.57 Kalkaska vs. Mean of Otsego =-1.29 95% C .I 13.042 acres to 21.99 acres i n d i c a t e s s ig n if ic a n t d iffe re n c e a t .05 p r o b a b ility le v e l. Table 66 TOTAL ACRES OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER, BY TOWNSHIP Township Size of Sample Mean Standard Deviation 22.2122 * 38.8756 South Branch T25NR2W 134 Grayling T27NR2W 31 20.6006 36.2776 Grayling T26NR4W 103 17.0439 100.9411 Orange T26NR7W 109 23.9640 * 66.2253 Blue Lake T28NR5W 267 8.8439 * 29.2995 G arfield T25NR7W 79 Bagley T30NR3W 968 9.1056 * 55.7617 Chester T30NR3W 93 27.8074 * 53.0860 Dover T31NR2W 80 119.4770 * 387.3304 Total 1,864 17.0557 17.516 20.0698 98.4831 in d i c a t e s township mean f a l l s out of the 95% Confidence In te rv a l fo r the o verall sample mean. 166 average property size a t only 8.8439 acres. When type of home development is considered, there is l i t t l e devia­ tio n in to ta l acreage owned per property owner. Mean to ta l acreage owned fo r seasonal homeowners is 15.627 acres (See Table 67). This is almost id e n tic a l fo r the mean t o ta l acres owned per property owner with no home development (15.391 acres) in the study area. Only the permanent home property owners have a s l i g h t l y higher mean fo r t o ta l acres owned (22.437%). Therefore, empirical evidence indicates th a t to ta l acres owned per property owner is not affec ted much by type of home development or the land. Also, s t a t i s t i c a l evidence indicates no s ig n if ic a n t differences between the means and very l i t t l e c o rre la tio n between to ta l acres owned and type of home development. In comparing the resu lts o f th is study, f o r to ta l acreage owned, to one conducted in Emmet and Cheyboygan counties (Marans and Wellman, 1978, p. 175), the results are q u ite s im ila r . The Marans and Wellman study found th a t permanent home owners owned more to ta l acres than did seasonal home owners (28 acres vs. 13 acres). The mean fo r permanent home owners is s l i g h t l y higher in Emmet and Cheyboygan counties than in Kalkaska, Crawford, and Otsego counties, but th is d ifferen ce is not s t a t i s t i c a l l y s ig n if ic a n t . Also, the mean to ta l acres owned f o r seasonal home owners in Emmet and Cheyboygan counties is only s l i g h t l y smaller than the mean in Crawford, Kalkaska, and Otsego counties. as s t a t i s t i c a l l y s ig n if ic a n t. The d iffe re n c e is not viewed Therefore, s t a t i s t i c a l l y , seasonal and permanent home owners are s im ila r in both study areas in terms of mean acreage owned. When considering location of water resource to amount of acres owned, there is no s t a t i s t i c a l l y s ig n if ic a n t d ifferen ce between the means. The 167 Table 67 TOTAL ACRES OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER,BY TYPE OF HOME DEVELOPMENT Standard Deviation Size o f Sample Mean No Home 601 15.391 49.586 Seasonal Home 680 15.627 72.041 Permanent Home 567 22.43-7 151.827 17.64 98.898 HOME Total 1,848 t Value fo r mean of no home vs. mean o f seasonal home t Value fo r mean of no home vs. mean o f permanent home t Value fo r mean o f seasonal home vs. mean of permanent home = = - .07 = -1 .0 5 - .98 SpearmansRs = .0125 Kendall Tau = .0098 mean acreage owned per property owner whose land is on a body o f water is 19.756 acres. The mean acreage owned per property owner not located on a body of water is 16.032. E m p iric a lly , there is a s lig h t differen ce f o r to ta l acres owned between locating on a body of water as opposed to locating o f f a body of water, but s t a t i s t i c a l l y , there is no d iffe re n c e . An F Value o f only .6348, not s ig n if ic a n t a t the .05 p r o b a b ility le v e l, supports the conclusion o f no s t a t i s t i c a l d iffe re n c e between the means. Total acreage owned per property owner has been shown to have little re la tio n to whether or not property is on the water. However, there may be a re la tio n s h ip between to ta l acres owned and type o f water on which the property is located. This seems to be the case in the study area as the mean to ta l acres owned f o r property owners with property located on a r iv e r is 78.4952 acres compared to only 10.5236 f o r property 168 Table 68 TOTAL ACREAGE OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER,BY LOCATION TO WATER Total Acreage Size of Sample Mean Standard Deviation 764 19.7557 136.9784 Property Not Located on Water 1,083 16.0324 58.7417 Total 1,847 17.5725 98.8996 0NH20 | Property Located on Water F Value fo r d ifferen ce between the mean = .6348 Spearmans^ = .1215 Kendall Tau = .1029 owners located on a lake. Obviously, the relatio n sh ip between location on water and to ta l acreage owned was being masked by the average between lake and r i v e r acreage. A t - t e s t between the means fo r to ta l acreage owned f o r lake property owners versus r i v e r property owners in d ica te th a t there is a s ig n if ic a n t difference between the means. Therefore, empirical and s t a t i s t i c a l evidence indicates th a t to ta l acres owned per property owner is a function of the type o f water on which the property is located. When location of p riv a te property to public property is considered in Table 70, another strong relatio n sh ip is uncovered. The mean acreage owned by property owners whose property is adjacent to public land is 30.03 acres. Mean acreage owned fo r property owners whose land is not adjacent to public land is only 14.80. In a d d itio n , an F -te s t between the two means y ie ld s a value of 6.4468 in d ic a tin g a s ig n if ic a n t d iffe re n c e between property owners with land adjacent to public land and those owning 169 Table 69 TOTAL ACRES OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER, BY TYPE OF WATER RESOURCE OF PROPERTY LOCATION Size of Sample Mean Standard Deviation Lake or Pond 653 10.5286 42.3489 River 105 78.4952 351.8302 7 70.8857 93.4830 765 20.4096 TYPEH20 River and Lake or Pond Total 137.913 t - t e s t fo r mean of Lake or Pond vs. mean of River = - 1 .9 8 * SpearmansRs = .1402 Kendall Tau = .1161 ♦Indicates s ig n if ic a n t ly d i f f e r e n t at .05 p r o b a b ility le v e l. Table 70 TOTAL ACRES OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER, BY LOCATION TO PUBLIC LAND PUBPROP Private Property Adjacent to Public Property Size of Sample Mean Standard Deviation 336 30.026 76.487 P riv ate Property Not Adjacent to Public Property 1,487 14.798 103.736 Total 1,823 17.605 99.434 F Value fo r mean of p riv a te property adjacent to public property vs. p riv a te property not adjacent to public property = 6.4468* Spearmans^ = .2258 Kendall Tau = .1914 *In d ic a te s s ig n if ic a n t d iffe re n c e a t .05 p r o b a b ility le v e l. 170 land not adjacent to public land. One possible explanation fo r th is s ig n if ic a n t d iffe re n c e in means may be because of subdividing. Descrip­ t iv e s t a t i s t i c s indicated th a t the size of acreage owned by the m ajority o f property owners was much sm aller. When subdividing large parcels of land, contact with public land may be reduced. example il l u s t r a t e s th is point. A simple arith m e tic A single parcel of 19,200 acres is completely surrounded by public land. A fte r subdividing in to t h i r t y - two equal 640 acre parcels; only 50% (s ixte en) are now adjacent to public land and 50% (s ixte en ) are not. Therefore, in th is example, subdividing accomplishes two things: (1) reduces average size per property owner, and (2) increases the percentage of land not adjacent to public land. Therefore, mean acreage owned can be seen to be a function of p riv a te land being adjacent to public land. Total Value of Property Per Owner Valuation of land and development is often very su b je ctiv e. Many methods e x is t fo r "objective" v a lu a tio n , such as appraised values fo r ta xatio n purposes. However, short o f a c tu a lly putting property up fo r s a le , an accurate current market value is d i f f i c u l t to obtain. Property owners, in th is study were asked to use t h e i r own judgement to estimate what t h e ir property was c u rre n tly worth. In the study area, the average property owner reported owning 17.516 acres of land valued a t $32,361.51 or $1,847.54 per acre. Although there are obvious weaknesses to th is approach to deriving value estimates ( i . e . subjective measurement e r r o r s ) , th is was the only p ra c tic a l approach to use, in th is instance, because of research budget lim ita tio n s . A c tu a lly , fo r the m ajority o f property owners, the value of t h e i r property is s l i g h t l y less than the mean o f $32,361.51. This is evidenced 171 by a median value o f $22,001.67 which indicates th a t a t le a s t 50% of the property owners have property valued a t approximately $22,000.00 or less. The most common value (mode) is also smaller than the mean a t $30,000.00. A kurtosis value of 118.868 indicates th a t the standard curve fo r property values is s l i g h t l y peaked and th is occurs to the l e f t (below) of the mean (o r less than $ 3 2,3 61.5 1). supported by a skewness value o f 8.8 6 . This is fu rth e r Also, fo r northern Michigan study area property owners, the 95% confidence in te rv a l about the mean fo r property valuation ranges from $29,639.36 to $35,083.66. There are some very severe problems when analyzing to ta l value per property owner th a t are not inherent when analyzing to ta l acreage owned per property owner. One major problem is q u ite obvious, the t o ta l value o f property owned is d i r e c t l y influenced by the type of home development on the land. Another problem is to ta l value which is a p o sitive function o f t o ta l acres owned as evidenced by a Pearson^ value of .6826. The more acres owned, the higher t o ta l property value. S t i l l another problem arises depending on where the property is located. There may be forces present w ith in a c e rta in county or township th a t a f f e c t property valuation (i.e . location to c i t i e s ) . Therefore, l i t t l e analysis w i l l be presented fo r to ta l value in th is se ctio n, instead most o f the analysis w i l l be presented in the next chapter on value per acre of land. There are a few c h a ra c te ris tic s concerning to ta l value o f property owned th a t were enlightening and w i l l be addressed in th is section. One in te r e s tin g r e s u lt of the survey is th a t 20.86% of the property owners do not know the value of t h e i r property. This may be a conserva­ tiv e fig u r e as there wasn't a category to check i f value was unknown. Only property owners who s p e c i f ic a lly indicated they did not know the 172 value were counted. Blank responses were f i l l e d out as missing and not as "value unknown". Therefore, uncertainty over current market value . of property is high among northern Michigan study area property owners. There is some evidence to suggest th a t to ta l value of property owned is re la te d to water resources. The to ta l value mean f o r property owners whose land is on a body o f water is $40,884.21 and fo r those property owners not on a body o f w ater, the mean to ta l value fo r t h e ir property is $25,817.07. N either of these means f a l l w ith in the 95% confidence in te rv a l fo r the o v erall to ta l value mean. Also, an F -te s t value of 29.3184 fo r analysis between the means indicates th a t they are sig ­ n i f i c a n t l y d i f f e r e n t from each other (See Table 71). Table 71 TOTAL VALUE OF PROPERTY OWNED PER NORTHERN MICHIGAN STUDY AREA PROPERTY OWNER,BY LOCATION TO WATER RESOURCE Size of Sample 0NH20 Mean Standard Deviation Property Located on Water 586 $40,884,212 * + 67,386.75 Property Not Located on Water 807 $25,817,077 * + $32,155,441 36,149.204 Total 1,393 52,157.418 *In d icates mean value e it h e r exceeds or f a l l s short of the 95% Confidence In te rv a l +Indicates s ig n if ic a n t ly d i f f e r e n t from other means. F = 29.3184 SpearmanSRs = .2380 Kendall Tau = .970 173 Further analysis indicates th a t even when s t a t i s t i c a l l y c o n tro llin g fo r the influence of type o f home development, there is s t i l l a high co rrela tio n between to ta l value o f the property and location on a body of water. In every case, no home, seasonal home or permanent home, the property on the body o f water has a higher value than property not on a body o f water (See Figure 4 ) . Therefore, location on a water resource is e m p iric a lly and s t a t i s t i c a l l y p o s itiv e ly re la te d to the to ta l value of the property. Having public property touch on a t le a s t one side o f an in d iv id u a l's private property is usually considered a favorable c h a r a c te r is tic . The fa c t th a t no one may b u ild on at le a s t one side may increase the sub­ je c tiv e evaluation of land owned and property value. In the study area, property owners who have land adjacent to public land have a mean to ta l value fo r t h e i r property o f $35,906.56. For property owners whose land is not adjacent to public land, the mean value is $31,840.99. An F -te s t indicates th a t there is no s ig n if ic a n t d iffe re n c e between the two means and SpearmansRs value o f -.0 3 2 8 and a Kendall Tau value of -.0271 indicates there to be only a very s l i g h t negative c o rre la tio n between location to public property and t o ta l value o f the property. P riv ate property located next to public property w i l l generally have a lower value than p riv a te property adjacent to p riv a te property. This negative c o rre la tio n w i l l be fu rth e r explored in the next chapter. The empirical d iffe re n c e showing a higher mean f o r p riv a te property located next to public property may be explained by the f a c t th a t s t a t i s t i c a l l y t o ta l value o f acreage owned is strongly related to t o ta l acreage owned (Pearson R = .6826). 174 Total Value f o r E n tire Population Mean Standard Deviation $32,155,441 $52,157,418 N(1393) Property Located on Water Mean $40,884,212 Standard Deviation 67,386.75 N (586) Home Development Property Not Located on Water Mean $25,817,077 Standard Deviation 36,149.204 N (807) Home Development No Home Mean Standard Deviation N 12,716.016 25,079.924 (128) No Home Mean Standard Deviation N 9,920.12 13,082.094 (292) Seasonal Home Mean Standard Deviation N 40,767.188 50,958.075 (320) Seasonal Home Mean Standard Deviation N 24,857.517 46,078.109 (207) Permanent Home Mean Standard Deviation N 67,282.594 106,383.714 (138) Permanent Home Mean Standard Deviation N 41,533.117 36,811.692 (308) Figure 4 Mean Breakdown f o r Total Value o f Property Owned, by Location on Water C o n tro llin g f o r Type of Home Development 175 Therefore, because property owners with land adjacent to public land own more acres than property owners not adjacent to public land, t h e ir to ta l value of property owned is also higher. Holding acreage owned constant, t o ta l value would be less f o r property owners adjacent to public land than fo r property owners not adjacent to public land. Summary This chapter examined the influence o f natural resources, acreage owned, and to ta l value of property owned. The natural resource base in each county and township was id e n t if ie d and v a ria b le relatio n sh ip s were explored to see i f location to c e rta in natural resources has any e f f e c t on t o ta l acreage owned or t o t a l value of property. The analysis shows th a t: 1) At le a s t one-third of the property in the study area is located on some water resource. Kalkaska county has over h a lf of i t s property owners on a water resource. 2) Almost o n e - f if t h o f land owned in the study area is ad­ ja c e n t to public land. In Crawfor' county, almost 40% of property owners are adjacent to public land. 3) The average amount of acreage owned per property owner in the study area is s l i g h t l y over 17 acres. However, q u a lif y ­ ing s t a t i s t i c s in d ic a te th a t the most common parcel size owned is only h a lf of an acre. 4) Property owners located on a r i v e r have s i g n if ic a n t ly higher acreage owned than property.owners on a lake. 5) Property owners located next to public land have s i g n if ic a n t ly higher acreage owned than property owners not adjacent to pub­ l i c land. 176 6) O n e -fifth o f the present property owners in the study area have no idea what the value of t h e i r property is . 7) Property located on water has a s ig n if ic a n t ly higher average value than property not located on water. This trend holds even when accounting f o r type of home develop­ ment. CHAPTER V I I I VALUE PER ACRE Value Per Acre of Land Model Value per acre of land is probably the most important v a ria b le in th is study. A ll land is not homogenous. Location of land r e l a t i v e to markets, f e r t i l i t y , natural resources, type of s o i l , extent o f ground cover, e tc . a l l play important roles in determining the value of land. Owing to the fa c t th a t not a l l parcels of land are the same s iz e , a basic u n it of measurement is needed fo r comparison purposes. This basic u n it is value per acre of land which r e fle c ts most o f the external in ­ fluences o f land price. Value per acre of land, in th is study, is a subjective measure of property value estimated by each surveyed pro­ perty owner fo r t h e i r individ ual property. Total predicted property exchange value was divided by actual acreage owned with the r e s u lt being value per acre o f land. In th is section, value per acre o f land is analyzed as a dependent v a ria b le with some important independent v a ria b le s. Zero-order p a r t ia l c o rre la tio n c o e ff ic ie n t are computed to assess the strength o f b iv a r ia te re la tio n s h ip s . The analysis then proceeds to m u ltip le regression in which selected variables are analyzed to see how much variance in value per acre o f land they account fo r when c o n tro llin g fo r other independent va ria b le s . (See Chapter I I fo r a d e ta ile d explanation o f th is procedure). In other words, an in d ivid u a l w i l l be able to make some general statements concerning land value when c e rta in property c h a ra c te ris tic s are known. Probably the greatest variance in value per acre o f land is re la te d to the type o f home development on the property. 177 What is r e a l l y being 178 measured is both the value o f a dwelling and the land. Therefore, the only accurate measure fo r value per acre o f land resu lts when no type o f home development is located on the land. As expected when considering value per acre o f land by type of home development, as presented in Table 72, property with permanent homes show an extremely high value o f $41,322.51. Properties with seasonal homes also show a high value per acre o f land fig u re ($ 3 1 ,8 2 7 .0 9 ). What is very s u rp ris in g , though, is th a t even when there is no home on the land, the value per acre o f land is s t i l l q u ite high a t $7,596.18. The most plausible reason fo r th is high value was touched upon e a r l i e r - subdividing. When large parcels are broken down in to smaller parcels, the combined s e llin g price o f a l l the small parcels is usually greater than the i n i t i a l price fo r the large p a rc el. Remembering th a t the mode and median value fo r t o ta l acres owned per study area property owner was much smaller than the mean, value per acre o f land was tested against t o ta l acres. The r e s u lt was a negative c o rre la tio n w ith a Pearson's R value o f -.0 9 1 6 . When s t a t i s t i c a l l y c o n tro llin g fo r type of home development, the c o rre la ­ tio n was -.1 0 7 2 . This is a moderate re la tio n s h ip in d ic a tin g th a t as to ta l acres owned per property owner increase, the value per acre o f land decreases. When location to water resource is considered, a strong re la tio n s h ip is found to e x is t . The value per acre o f land fo r property located on water was $35,244.26 and $22,216.01 fo r property not located on water. An F - te s t between the means indicates th a t property located on water has a s i g n if ic a n t ly higher value per acre of land than fo r property not located on water. Even when s t a t i s t i c a l l y c o n tr o llin g fo r the e f f e c t o f type o f home development, value per acre of land fo r water property is s ig n if ic a n t ly higher than fo r property not on water (See Figure 5). 179 Table 72 VALUE PER ACRE IN NORTHERN MICHIGAN STUDY AREA BY TYPE OF HOME DEVELOPMENT Size of Sample Mean Standard Deviation Home No Home 404 $ 7,596.18 10,373.034 Seasonal Home 508 $31,827.03 39,314.453 Permanent Home 436 $41,322.51 48,727.07 1,348 $27,636.21 39,603.3271 Total SpearmansRs = .4926 Kendall Tau = .3329 When type of water resource associated with a property is considered, evidence indicates th a t lake and pond property has a much higher value per acre o f land than r i v e r property as presented in Figure 6. Property located on a lake or pond has a mean value per acre o f land of $38,236.66 compared to only $18,115.24 fo r property located on a r i v e r . Even when s t a t i s t i c a l l y c o n tro llin g fo r the e ffe c ts o f type o f home development, lake or pond property s t i l l maintains a s u b s ta n tia lly higher per acre of land value than r iv e r property. In f a c t , one surprising r e s u lt is th a t land located on a r i v e r has a lower per acre of land value than land not located on any water. subdividing. This is once again probably due to the e f f e c t of I t was reported e a r l i e r in Table 69 th a t average acreage owned per property onwer located on a r i v e r was 78.4952 compared to 16.0324 (Table 68) fo r property not on any water resource. Having pre­ viously ascertained th a t value per acre o f land is negatively r e la te d , although weakly, to to ta l acres owned i t is easy to see how property located on a r i v e r can have a lower per acre o f land value than property not located on any water resource. 180 Mean Standard Deviation Size o f Sample Property Not Located on Water Property Located on Water Mean Standard Deviation Size of Sample Mean Standard Deviation Size o f Sample $35,244.26 44,043.98 563 Mean Standard Deviation Size of Sample $ 9,391.86 11,901.90 122 Mean Standard Deviation Size o f Sample $40,761.62 40,860.65 306 $18,368.53 32,614.50 201 Permanent Home Permanent Home Mean Standard Deviation Size of Sample $ 6,835.55 9,585.78 280 Seasonal Home Seasonal Home Mean Standard Deviation Size of Sample $22,216.01 35,045.14 776 No Home No Home Mean Standard Deviation Size o f Sample $27,597.15 39,611.66 1339 Mean Standard Deviation Size of Sample $46,266.42 48,372.76 306 Figure 5 Value per Acre o f Land by Location to Water Resource C ontrolling f o r Type of Home Development $38,921.10 43,719.23 295 181 Mean Standard Deviation Size of Sample $35,283.89 44,043.98 563 Lake or Pond Mean Standard Deviation Size of Sample Ri ver $38,236.66 46,233.64 477 Mean $18,115.24 Standard Deviation 22,597.83 Size o f Sample 86 No Home Mean Standard Deviation Size o f Sample No Home $10,013.05 12,063.455 102 Seasonal Home Mean Standard Deviation Size of Sample Seasonal Home $43,209.72 41,955.07 269 Mean $21,448.40 Standard Deviation 25,346.27 Size o f Sample 36 Permanent Home Mean Standard Deviation Size of Sample Mean $ 8,064.30 Standard Deviation 13,990.833 Size o f Sample 21 Permanent Home $52,774.91 63,300.26 106 Mean $22,249.25 Standard Deviation 22,359.76 Size o f Sample 29 Figure 6 Value per Acre f o r Property Located on Water Resource C on trolling f o r Type of Home Development 182 Value per acre of land is s ig n if ic a n t ly affected’ by location o f p r i ­ vate property being adjacent to public land. The value per acre of land fo r those property owners whose land is adjacent to public land is $13,992.14. The mean value per acre of land not adjacent to public land is $30,912.76. Both mean values occur outside of the 95% Confidence In te rv a l boundaries fo r the o v erall mean f o r value per acre of land. In a d d itio n , an F -te s t indicates th a t the means fo r property owners whose land is adjacent to public land and the mean fo r property owners whose land is not adjacent to public land are s ig n if ic a n t ly d i f f e r e n t . The resu lts are somewhat surprising when the influence of other variables are not considered. I n i t i a l l y , i t seems because mean acreage size is much la rg e r fo r p riv a te property owners with land adjacent to public property value per acre of land and to ta l acreage owned would be strongly re la te d in the negative d ir e c tio n . However, resu lts from the section on i n i t i a l property purchase, acreage and valuation show a weak negative re la tio n s h ip e x is ts . One explanation is possibly the ov e ra ll re la tio n s h ip between value per acre of land and to ta l acreage owned is being suppressed by another v a ria b le . The next section w i l l explore th is p o s s i b ilit y but a t th is p o in t, empirical evidence indicates having public land adjacent to p riv a te land resu lts in a lower value per acre o f land. Land value per acre was also found to be highly correlated w ith size o f lake or pond on which the land is located. A SpearmansRs value of .4137 and a Kendall Tau value o f .3226 in d ic a te the la rg e r the size of lake or pond, the greater the value per acre o f land. The next step in the analysis is to control fo r e ffe c ts o f other variables and uncover suppressed relation ship s and tr y to elim in ate spurious re la tio n s h ip s . This is accomplished by using a m u ltip le 183 regression technique which controls fo r the e f f e c t of a l l independent variables entered in to the regression model. As mentioned in Chapter I I , three m u ltip le regressions were performed on survey data. The simple c o rre la tio n m atrix r e la t in g to variables in the equations are lis te d in Appendix C and the means and standard deviations in Appendix D. One o f the null hypotheses o f th is study was th a t natural resource c h a ra c te ris tic s do not ex ert a s ig n if ic a n t influence on value per acre o f land. In te r p r e ta tio n o f regression resu lts in d ica te th is p a r t ic u la r nu ll hypothesis must be rejected but only under ce rtain conditions. That i s , some natural resource c h a ra c te ris tic s exert a s ig n if ic a n t influence on value per acre o f land and others have no s ig n if ic a n t e f f e c t . As expected, in a l l three regression equations the independent v a ria b le exerting the most influence on value per acre of land was type o f home development. In a d d itio n , to ta l acres owned was found to con­ t r ib u t e s ig n if ic a n t ly in the negative d ire c tio n fo r value per acre of land. There were three natural resource c h a ra c te ris tic s found to be s ig n if ic a n t in a l l three equations. They were H4 (s ize of lake greater than 500 acres) which contributed s ig n if ic a n t ly in a p o s itiv e d ir e c tio n , H2 (s iz e of lake 25-100 acres) which contributed s ig n if ic a n t ly in a p o s itiv e d ir e c tio n , and J1 (adjacent to public property) which c o n tr i­ buted s ig n if ic a n t ly in a negative d ire c tio n . When considering only the s im p lifie d regression equation, which was n e ith e r county or township s p e c ific , the independent variable K1 (close to ski area) was found to contribute s ig n if ic a n t ly in the p o s itiv e d ire c ­ t io n . A complete l i s t of a l l independent v a ria b le s , t h e i r c o e f f ic ie n t , t value, and 95% confidence in te rv a ls fo r each regression equation, along 2 with each regression's R and residual mean square values can be found in Table 73. Observations, from Table 73 reveal the regression equation Table 73 INFORMATION FROM THE THREE REGRESSION EQUATIONS FOR THE DEPENDENT VARIABLE "VALUE PER ACRE OF LAND" S im p lified Regression C o e ffic ie n t Value Constant 38991.398 t-Value 16.7335 * County S pec ific Regression 95% Confidence In te rv a l C o e ffic ie n t Value 34420.32 46190.47 t-Value 18.1917 * 43562.48 Cl - - - 95% Confidence In te rv a l 41209.45 Township S pecific Regression C o e ffic ie n t Val ue 26281.1 t-Value 5.892 * 51171.49 -15402.47 - 4.7156 * -21810.00 95% Confidence In te rv a l 17530.83 35031.36 - - - — — — - 8994.96 C2 “ "■ “ -15889.676 - 5.2734 * -21800.69 - 9978.66 D1 “ - — — - — 3867.97 .77298 - 6627.36 14363.30 D2 - - - D3 “ — “ - - - — “ + 11363.37 + + 1.8264 - 841.95 23568.68 D4 _ _ _ _ _ _ 347.76 .061694 -10710.50 11406.04 Table 73 - Continued S im p lified Regression C o e ffic ie n t Value D5 t-Value - - County S pecific Regression 95% Confidence In te rv a l - C o e ffic ie n t Value t-Value “ - 95% Confidence In te rv a l - Township Regression C o e ffic ien t Value 17146.43 t-Value 3.215 * 95% Confidence In terval 6684.12 27608.73 D6 - - — - - — - 6445.89 -1.0480 -18511.69 5619.917 D7 “ - — — — “ 22777.05 5.4138 * 14523.56 31030.53 D8 - - — — — — 9123.59 1.5608 - 2343.51 20590.70 El -31772.67 -13.1014 * -36530.11 -33463.39 -13.8680 * -15690.74 - 6.2398 * -20623.76 -14357.71 - 5.7916 * -10757.72 61 - 3434.70 - .8361 -11493.57 4624.16 -33594.64 -14.0169 * -19220.92 -14043.93 - 5.6162 * - .6586 -15034.36 7477.17 -18949.48 - 9138.38 - 9494.50 - 3778.60 -38296.40 -28892.89 -28729.73 -27015.22 E2 -38197.05 6891.41 1.4584 - 2378.15 16160.98 Table 73 - Continued S im p lifie d Regression C o e ffic ie n t Value HI - 954.36 County S pecific Regression t-Value 95% Confidence In te rv a l C o e ffic ie n t Value - .28529 - 7516.67 - 1648.19 t-Value - .4968 95% Confidence In te rv a l C o e ffic ie n t Value - 8156.22 - 3521.84 8330.43 2.4798 * 1740.39 9774.49 2.1312 * 777.13 14920.46 H3 3332.29 .53549 - 8875.27 t-Value -1.0744 4859.82 5607.94 H2 Township Regression + + 11776.13 2.7145 * 33070.74 8.6689 * 25587.03 1552.74 .2461 -11035.62 -4.3294 * -16036.05 30398.27 22974.48 8.0327 * 4733.42 2.3586 * 796.52 - 7705.25 -2.9883 * -12763.44 95.12 1.1609 - 65.61 255.86 7.1504 * 1324.98 .6439 - 2711.49 - 9101.45 -2.8964 * .4717 - 120.95 197.52 -15265.72 - 2937.18 603.68 .2938 - 3426.77 4634.13 5361.46 38.29 19654.84 34517.07 - 2647.02 8670.33 LKMILE X6a 27085.95 37822.07 - 6035.19 K1 -10824.65 13930.13 40554.44 J1 3265.85 20287.41 15539.86 H4 - 9952.06 2908.38 18771.85 + 95% Confidence In te rv a l + + + Table 73 - Continued S im p lifie d Regression C o e ffic ie n t Value RIVMILE X6b - 110.37 t-Value -1.2548 County S pecific Regression 95% Confidence In te rv a l - 282.91 C o e ffic ie n t Value - 62.98 t-Value - .7239 95% Confidence In te rv a l - 62.18 TOTACRES X7 C1G1 - 31.699 “ -3.2174 * - 51.03 - 12.37 C o e ffic ie n t Value + 233.64 t-Value 95% Confidence In te rv a l + + 107.68 - 33.32 7708.66 “ Township Regression -3.4106 * .9311 - 52.487 - 14.155 - 8532.34 - 24.26 -2.4935 * - 43.3410 5.1732 “ — **■ — “ 23949.66 C2H2 - 1629.13 .2465 -11336.78 14595.05 C2H3 “ — — 15315.96 2.3462 * — 2509.58 28122.34 D3G1 — “ - — - - -15711.76 -1.5825 -35188.43 3764.92 “ D5H2 -13119.64 -1.8860 -26765.86 ' 526.58 Table 73 - Continued S im p lifie d Regression C o e ffic ie n t Value D5H3 - D7J1 - t-Value County S pecific Regression 95% Confidence In te rv a l — C o e ffic ie n t Value - - “ - t-Value 95% Confidence In te rv a l - — “ Township Regression C o e ffic ien t Value t-Value 95% Confidence In te rv a l + + .5156 - 7819.68 + 2787.79 13395.26 E2G3 23535.53 2.9947 * 8115.25 20868.43 3.2356 * Residual Mean Square .21389 1242120711.57672 22645.17 7661.47 2.9648 * 37628.86 33520.61 38937.82 R2 8216.25 .23993 .26480 1204562948.09946 1168622125.78786 * Variable is s ig n ific a n t at .05 p ro b a b ility le v e l. + V ariable does not meet tolerance level requirement and does not enter in to the regression. 189 2 with the highest R value and lowest residual mean square value is the 2 one which is township s p e c ific . However, the gain in R is small and is due mostly to the addition o f more independent variables in the regression equation. Therefore, to choose one of the three regression equations over another as the most appropriate fo r use in value per acre assess­ ments in northern Michigan, without additional inform ation, is hazardous. For example, i f an area in Roscommon county is chosen fo r an aly sis, i t would be wise to s t a r t with the s im p lifie d regression equation. I f there is p r io r knowledge which indicates th at Roscommon county is quite s im ila r to any of the three counties in th is study, then i t may be appropriate to use the county s p e c ific regression equation. Additional analysis may show th a t the area chosen in Roscommon county nnrrors a township used in th is study, i t would then seem advisable to use the township s p e c ific model. Care must be exercised in which model is chosen fo r fu rth e r analysis in areas other than the study area used in th is p r o je c t. Care must also be exercised in using any of the regressions over time without additional work to estimate change th a t may have occurred. A su rprising r e s u lt o f a l l three regression equations was va ria b le J1 (adjacent to public property) which contributed s ig n if ic a n t ly to value per acre of land in a negative d ire c tio n . I n i t i a l l y i t was thought v a ria b le J1 would contribute s ig n if ic a n t ly in a p o s itiv e d ire c tio n . was based on the open spaces theory. This assumption Having undeveloped land next to developed land generally gives a fe e lin g of owning more land than in a c t u a lit y . However, l i t e r a t u r e review revealed open spaces a c tu a lly may impart a negative influence on value per acre of land due to tre s s ­ passing problems. A m a jo rity (54%) o f landowners in northern Michigan post and fence t h e i r land against tresspassers with many more (11%) in ­ tending to post and fence in the near fu tu re (McEwan, 1970). 190 Another depressant fo r value per acre of land fo r p riv a te land adjacent to public land may be the fa c t th a t fo r most landowners seclusion is not a desirable t r a i t , ra th e r, close contact with neighbors was more desirable (McEwan, 1970). Examination o f Table 70 reveals property owners with land adjacent to public land own s ig n if ic a n t ly more land than property owners not adjacent to public land. The high mean (30.026) acres owned by property owners adjacent to public land increases the chances fo r seclusion. Because seclusion is not a desirable t r a i t value per acre of land fo r p riv a te property next to public land w i l l be de­ pressed. Also, th is study revealed th a t value per acre o f land decreases as amount of land owned increases. A ll these factors w i l l in te r a c t to reduce value per acre of land fo r p riv a te property adjacent to public land. Further observation o f Table 73 reveals many of the independent variables to be s t a t i s t i c a l l y s ig n if ic a n t and enter in to the regression, however, they only account fo r around one-quarter o f the variance in the dependent v a ria b le . In a d d itio n , the large confidence in te r v a ls shown fo r each independent v a ria b le lead to a poor p re d ic tiv e c a p a b ility f o r the model. Obviously, there are many more unknown variables in the study area which contribute s ig n if ic a n t ly to value per acre of land than those id e n t if ie d here. This is the greatest problem in attempting to choose any o f the three regressions fo r use in a s p e c ific area. The amount of variance controlled fo r by the independent v a ria b le is not large enough to j u s t i f y p ra c tic a l use. The three regression equations can now be summarized in to the follow ing form. The county s p e c ific model becomes: 191 Y1 = 46190.47-33463.39 ( E l) * -14357.71 (E 2 )* - 15402.47 ( C l ) * - 15889.68 (C 2)* -7705.25 ( J l ) * + 1324.98 (K l) - 3778.60 (G l) 1648.19 (H I) + 9774.49 (H2) + 0 .0 (H3) + 30398.27 (H 4)* + 38.29 (Y6a) = 62.98 (Y6b) - 33.32 (Y 7 )* + 7708.66 (1 G l) + 1629.13 (C2H2) + 15315.96 (C2H3)* + 20868.43 (E2G3)* + f *In d ic a te s va ria b le s ig n if ic a n t a t .05 p r o b a b ility level 2 R = .23993 The township s p e c ific model becomes: Y1 = 26281.10 - 33594.64 ( E l ) * - 14043.93 (E 2 )* + 3867,97 (D l) + 0.0 (D2) + 11363.37 (D3) + 347.76 (D4) + 17146.43 (D 5)* - 6445.89 (D6) + 22777.05 (D7)* + 923.59 (D8) - 9101.45 ( J l ) * + 603.68 (K l) + 6891.41 (G l) - 3521.84 (H I) + 11776.13 (H2) + 1552.74 (H3) + 27085.95 (H 4)* + 0 .0 (Y6a) + 0.0 (Y6b) - 24.26 (Y 7 )* - 15771.76 (D3G1) - 13119.64 (D5H2) + 0.0 (D5H3) + 2787.79 (D7J1) + 22645.17 (E2G3)* + f 2 ♦Indicates v a ria b le s ig n if ic a n t a t .05 p r o b a b ility lev el R =.26480 and the s im p lifie d model becomes: Y1 = 38991.398 - 31772.67 ( E l ) * - 15690.74 (E 2 )* - 11035.62 ( J l ) * + 4733.42 ( K l ) * - 3434.70 (G l) - 954.36 (H I) + 8330.43 (H 2)* + 3332.29 (H3) + 33070.74 (H 4)* + 95.12 (Y6a) - 110.37 (Y6b) 31.699 ( Y7 ) * + 23525.53 (E2G3)* + (♦Indicates v a ria b le s ig n if ic a n t a t .05 p r o b a b ility le v e l R^=.21389 192 where: 1 when there is no home development on the property El — E2 = 1 when there is a seasonal home on the property Cl = 1 when the property is located in Crawford county C2 = 1 when the property is located in Otsego county D1 = 1 when the property is D2 = 1 when the property is located in Grayling T27NR2W township D3 = 1 when the property is located in Grayling T26NR4W township D4 = 1 when the property is located in Orange T26NR7W township D5 = 1 when the property is located in Blue Lake T28NR5W township D6 = 1 when the property is located in G a rfie ld T25NR7W township D7 = 1 when the property is D8 = 1 when the property is located in Chester T29NR2W township Jl = 1 when the property is located adjacent to public land Kl = 1 when the property is located close to a commercial ski area Gl = 1 when the property is located on a r iv e r HI = 1 when the property is located on a lake less than 25 acres located in South Branch T25NR2W township located in Bagley T30NR3W township in size H2 = 1 when the property is located on a lake g rea ter than 25 acres but less than 100 acres in size H3 = 1 when the property is located on a lake g rea ter than 100 acres but less than 50C acres in size H4 = 1 when the property is located on a lake g rea ter than 500 acres in size Y6a = actual miles property is located away from a lake Y6b = actual miles property is located away from a r iv e r Y7 = actual size o f property in acres C1G1 = 1 when r i ver the property is located in Crawford county and C2G2 = 1 when the property is located in Otsego county and on a lake g rea ter than 25 acres but less than 100 acres in size ona 193 C2H3 = 1 when the property is located in Otsego county and on a lake greater than 100 acres but less than 500 acres in size D3G1 = 1 when the property is located in Grayling T26NR4W township and on a r iv e r D5H2 = 1 when the property is located in Blue Lake T28NR5W township and on a lake g rea ter than 25 acres but less than 100 acres in size D5H3 = 1 when the property is located in Blue Lake T28NR5W township and on a lake greater than 100 acres but less than 500 acres in size D7JI = 1 when the property is located in Bagley T30NR3W township and is adjacent to public land E2H3 = 1 when the property has a seasonal home on i t and is located on a lake g rea ter than 100 acres bue less than 500 acres in size In te rn a l V a lid a tio n One hypothesis is th a t there is no, lin e a r re la tio n s h ip between the dependent and the set o f independent v a ria b le s. To te s t th is hypothesis, i t is necessary to see i f any of the assumptions of the c la ss ical lin e a r regression model were v io la te d . As mentioned before, great care was taken to ensure th a t m u lt ic o llin e a r it y d id not become a problem. I n i t i a l model s p e c ific a tio n was such th a t m ulti c o l l i n e a r i t y was severely co n tro lled . To f u r t h e r guard against any multi c o l l i n e a r i t y problems, combination variables were created when zero order p a r t ia l c o rrela tio n s showed any appreciable re la tio n s h ip amongst a set o f independent v a ria b le s . Creation of dummy va riables was necessary to ensure th a t the assumption of in te r v a l level measurement in a l l variables was met. The level of analysis i n ­ herent in regression requires in te r v a l measurement. Homoscedasticity was also evaluated through visual inspection of a p lo t o f residuals and various s t a t i s t i c s output by SPSS subprogram REGRESSION. Visual inspection shows th a t (1) there was no pattern which 194 indicated the need to introduce more multi p i i c i t i v e terms or polynomial terms in to the equation, (2) no visual pattern existed'among the p lo t of residuals in d ic a tin g any problems with homoscedasticity. output also indicated no homoscedasticity problems. S t a t is t ic s An expected run o f signs was calculated to be 648 and actual run o f signs was 629. This indicates a normal d is tr ib u tio n of the e rro r term around values of x and th a t the expected value of the e r r o r term is equal to zero. There­ f o r e , a l l the assumptions o f a general lin e a r model have been met and the null hypothesis which states there is no lin e a r re la tio n s h ip between the dependent v a ria b le and a set o f independent variables must be rejected as must the null hypothesis which states th a t any re la tio n s h ip between the dependent v a ria b le and any independent v a ria b le is not lin e a r and the e f f e c t between two or more independent variables is non-additive. The r e la tio n s h ip between the dependent and a set o f independent v a r i ­ ables is li n e a r , and the cla ss ical lin e a r regression model, modified with a few combination v a ria b le s , appeared to be the appropriate model f o r the study. External V a lid it y The te s t o f any model is how well does i t work. In th is study the p re d ic tiv e power of the m u ltip le regression model was checked by randomly se le c tin g cases, which went in to building the model, and te s tin g them. Normally a model is tested with data which did not go in to building the model but th a t e n ta ils another survey. To save time and e f f o r t , the model is tested to see i f i t predicts well with biased data and i f so, then i t would be necessary to conduct a survey to see how well the model works with unbiased data. I t is biased data because the model is being tested under the most advantageous manner, as i t is being tested against information from which i t was derived. 195 As expected, the model does not p re d ic t w e ll. The amount of variance explained by any of the regression equations is too small and the c o n fi­ dence in te rv a ls f o r the independent va riables are too large to allow fo r repeated good p red ictio n s . Future work should be directed toward i d e n t i f i ­ cation of other variables which should en ter in to a regression equation of th is type, and, i f successful, the p re d ic tiv e power o f the model should improve. Summary In th is chapter relation ship s between value per acre o f land and selected independent variables have been explored including type of home development, p o l i t i c a l boundary o f lo c a tio n , and location next to , or close to , selected natural resources. The analysis shows th a t: 1) Location on large lakes (g re a te r than 500 acres) and medium lakes (25-100 acres) is s ig n if ic a n t ly and p o s it iv e ly corre­ lated with value per acre o f land. 2) Location to public land contributes s i g n if ic a n t ly in the negative d ire c tio n fo r value per acre o f land. 3) There is no evidence to suggest th a t location on lakes (100-500 acres or less than 25 a c re s ), location on r iv e r s , distance from lakes or riv e r s has any e f f e c t on value per acre o f land. 4) There is some evidence th a t suggests closeness to ski areas may contribute in the p o s itiv e d ire c tio n to value per acre of land. The independent v a ria b le Kl (close to a ski a re a ), was s ig n if ic a n t in only one of the regression equations. 196 5) There are many other variables which may influence land value which were not included in th is study. CHAPTER IX CONCLUSIONS AND RECOMMENDATIONS Conclusion The major objective o f th is study was to develop an information base concerning socioeconomic and general c h a ra c te ris tic s of landed property and home owners in three representative counties o f Michigan's northern lower peninsula. The information base has been established; however, i t is by no means complete. With what has been ascertained, though, i t is possible to formulate some general conclusions and recommendations that impact on fu tu re p o licy . The major conclusions from th is study are out­ lined below. 1) The typical property owner in the northern Michigan study area was male, married and approximately f i f t y = t h r e e years old. Overall fam ily income levels o f property owners in the study area were higher among seasonal home owners and property owners with no home in the area than f o r permanent home owners. In a d d itio n , property owners in Kalkaska and Crawford counties generally had lower fam ily income levels than property owners in Otsego county. 2) There were s l i g h t l y more property owners with seasonal homes in the study area than e it h e r property owners with permanent homes or no type of home development. The Kalkaska townships had more seasonal home owners than those in Crawford or Otsego counties. A large percentage o f property owners with a seasonal home or no home development on t h e i r property in the study area liv e d in the southeastern Michigan region. In a d d itio n , almost o n e -th ird o f the study areas' permanent 197 198 residents used to li v e in the southeastern Michigan area p r io r to re lo c a tin g in the study area. 3) The major reason fo r property a c q u isitio n in the study area was fo r investment or a retirem ent home, although the influence of recreation ranks high as a major reason fo r property ac­ q u is it io n . The influence o f recreation was probably under­ estimated as a major reason fo r property a c q u is itio n . 4) Most information about a v a ila b le property came from friends and r e la t iv e s . (i.e . The t r a d itio n a l market information sources real estate salespersons, newspapers/magazine ads) accounted fo r less than one-fourth o f f i r s t information sources leading to property acquisitio n in the northern Michigan study area. 5) Over o n e - f if t h o f the current property owners in the study area intended to s e ll t h e i r property in the fu tu re . Eighty percent who desired to s e ll vis u a liz e d s e llin g w ith in a f iv e year period. More property owners with no home develop­ ment on t h e i r land in the study area desired to s e ll than e it h e r property owners with a permanent home or seasonal home. This is not surprising as a la rg e r percentage of property owners with no home development on t h e i r land in the study area purchased t h e i r property f o r i t s investment p o te n tia l than e it h e r property owners with permanent homes or seasonal homes. 6) There is a d e f in it e fe e lin g th a t levels of property taxes in the study area were too high. Kalkaska townships had a higher percentage of property owners who f e l t property 199 tax levels too high than the Crawford or Otsego townships. A higher percentage o f property owners with a permanent home in the study area had f e l t property tax levels too high than did property owners with a seasonal home or no home develop­ ment in the study area. 7) Twenty-five percent of the property owners in the study area desired a t le a s t a few more county or municipal services to be provided. Permanent home owners in the study area generally desired more services to be provided than property owners with a seasonal home or no home in the area. Kalkaska and Crawford counties had higher percentages o f property owners desiring more services than property owners in Otsego county. The q u a lit y o f the county or municipal services provided was gener­ a l l y viewed as average or good. 8) Only o n e - f if t h of the northern Michigan study area property owners opposed any fu tu re r e s id e n tia l development. Kalkaska county property owners were more opposed to fu tu re r e s id e n tia l growth than property owners in Otsego or Crawford county. Seasonal home owners were more opposed to r e s id e n tia l growth than property owners with a permanent home or no home develop­ ment in the study area. There is also a higher proportion o f seasonal home owners who favored s t r i c t e r land use controls ra th e r than e it h e r property owners with a permanent home or no home development in the study area. 9) Uncertainty over present land development regulations was high among a l l northern Michigan study area property owners. However, property owners w ith seasonal homes or no home development 200 in the study area were much less informed of land development r e s t r ic tio n s than property owners with a permanent home in the study area. Uncertainty was also a major problem fo r seasonal home owners and es p ec ially property owners with no home development in the study area when t h e i r a ttitu d e s were s o lic it e d on the qu an tity and q u a lity of services provided, present land development regulations and fu ture re s id e n tia l growth. This indicated th a t property owners not liv in g in the study area were generally unaware of what services were a v a ila b le , what regulations and r e s tr ic tio n s there were fo r developing t h e i r land, and trends concerning re s id e n tia l growth in the study area. 10) The average length o f stay in the study area per seasonal home owner was approximately 11 days, however, the most common length o f stay was a weekend and summer was the time o f heaviest use. 11) The mean acreage owned per northern Michigan study area pro­ perty owner was approximately 17 acres. However, th is fig ure is somewhat misleading as the most common size parcel owned was determined to be only on e-half acre. 12) One-third o f the property owned in the study area was on some type of water resource. Location of property on large lakes (g re a te r than 500 acres) and medium size lakes (25-100 acres) was determined to ra is e value per acre o f land. There was no evidence to suggest th a t lo cation on lakes 100-500 acres or less than 25 acres in size or location on riv e rs had any e f f e c t on value per acre o f land. 201 13) Almost o n e - f if t h o f land owned in the study area was adjacent to public land. In a d d itio n , property owners located next to public land owned s ig n if ic a n t ly larg e r acreage than property owners not so located. Analysis showed th a t location next to public land s ig n if ic a n t ly lowered land value per acre even when c o n tro llin g fo r to ta l acres qwned per property owner. Recommendations 1) Counties in the northern Michigan study area should seriously consider via b le a lte rn a tiv e s to present methods of administ­ ering local and municipal services. There are q u ite a few trends which surfaced in the survey resu lts leading to th is suggestion. I t is quite evident th a t the influence of southeastern Michigan w i l l continue to be f e l t in the study area. More services w i l l be demanded as residents o f southeastern Michigan, who are in general used to services provided by a metropolitan area, con­ tinue to influence p o lic y . Growth in outlying areas w i l l also stress present levels o f public service. A severe problem arises when more services are demanded but there is an unwillingness to pay or even an i n a b i l i t y to pay fo r these services. This is what is happening in the northern Michigan study area a t the present time. The analysis indicated th a t property owners with permanent homes in the study area desired more services, were more upset about current property tax levels and had lower fam ily income than property owners with a seasonal home or no home development in the area. Meeting the increased demand fo r services without ra is in g property taxes, requires a county to consider a lte r n a tiv e s . 202 One fe a s ib le a lt e r n a t iv e may be forming regional cooperatives to take advantage o f economies o f scale. Regional health services are one example of cooperative arrangements already working in northern Michigan to hold down costs and improve q u a lit y o f services provided. Another a lt e r n a t iv e th a t may be worthwhile is to contract with p riv a te companies fo r services now provided at the public le v e l. A th ird a lt e r n a t iv e would p r i o r i t i z e service presently provided p u b lic ly and s h i f t tax revenues to increase q u a lity of high p r i o r i t y service and leave low p r i o r i t y services fo r in divid ua ls to handle on t h e i r own. There are many other a lte r n a tiv e s which professional managers recog­ nize and may be b e tte r suited fo r the region. Results of th is study in dicate a p o te n tia l supply and demand problem fo r municipal and county services is present. Therefore, a lt e r n a t iv e arrangements to handle the fu tu re dispo sitio n o f public services should be considered. 2) Local and regional tourism agencies may wish to consider a lte r n a tiv e s aimed at a t tr a c tin g seasonal residents to the area in the o f f season. There are two reasons fo r th is suggestion. I t is obvious th a t the more money seasonal residents spend in an area, the greater the economic impact on th a t area. I t is also obvious from survey resu lts th a t seasonal home owners have the highest lev els o f fam ily income of any group o f property owners. Consequently, seasonal home owners w i l l have more disposable income. Seasonal home owners already have a substantial c a p ita l investment in the study area and any a c t i v i t y th a t a t tr a c ts them to the area during the o f f season may economically b e n e fit the community. However, i t is also possible th a t expected costs of services required to a t t r a c t seasonal home owners to the area during the o f f season may exceed expected community b e n e fits . 203 The survey also indicated th a t recreational considerations were a major reason fo r i n i t i a l property a c q u isitio n . P o lic ie s th a t would lengthen the t o u r is t season (establishment of an extensive snowmobile t r a i l ) or e x p lo it any comparative advantage th a t e x is t , or could be created, should be considered as methods to create ad ditional t o u r is t a c tiv ity . The spring Kalkaska tr o u t fe s tiv a l and unique Alpine atmos­ phere o f Gaylord are j u s t two examples of methods u t i l i z e d to economically strengthen the host community. I t is e n t ir e ly possible th at a f t e r careful co n su ltation , i t may be determined th a t economic and social costs o f a t tr a c tin g more to u ris ts w i l l o ffs e t expected b e n e fits . In this case the co rrect decision may be to maintain the status quo. However, study resu lts in d ica te a p o ten tial fo r economic b e n e fit and counties would be best advised to consider a l l p o s s i b i l i t i e s . 3) I n i t i a t e a program to reduce uncertainty and confusion over issues th a t could a f fe c t the valuation of property. Rules and regulations are in constant flu x when i t comes to land use. Local, re g io n a l, s ta te , and federal p o lic ie s can a f f e c t one's use and subsequent valuation of real property. The major r e s p o n s ib ility f o r c o n tro llin g location and q u a lit y of land develop­ ment rests with the local government (American Society of Planning O f f i c i a l s , 1976). The major re s p o n s ib ility fo r informing property owners of any changes th a t could a f fe c t perceived q u a lity , as re fle c te d in expected market exchange values, of th e ir land should also re s t with local government. I t is c le a r from survey resu lts th a t many property owners are unaware of current land development regulations in the study area. In a d d itio n , many property owners are unaware of 204 the lev el and type o f public services provided. Uncertainty and confusion is very high among property owners with seasonal homes or no type of home development. This is due in large p art to the present p rac tice o f announcing zoning and development changes through local newspapers. Most property owners not permanently located in the study area do not receive the area's local newspaper. I t should be the r e s p o n s ib ility o f the local governing agency to i n i t i a t e a program to make property owners not permanently located in the area aware o f any proposed or enacted changes concerning land development. One approach th a t could be considered is to attach a new sletter, along with the property tax b ill, id e n tify in g key proposed or enacted changes fo r the community. Costs would be nominal and uncertainty could be s u b s ta n tia lly reduced. 4) Formulate e f f e c t iv e p o licy to control unplanned s e t t l e ­ ment practices e s p e c ia lly around environmentally se n sitive areas. I t is c le a r from survey resu lts th a t p o te n tia l fo r fu tu re residen­ t i a l growth is high in the northern Michigan study area. Present property owners expect growth to continue and many desire i t . At the present time adequate regulations to control unplanned settlement have been missing in Michigan (Nelson, 1973). A major problem develops when unplanned settlement takes place in environmentally se nsitive areas. Survey resu lts in d ica te increased demand f o r lo ts on c e rta in size lakes e x is t , as evidenced by higher value per acre o f land values f o r lo ts located on those lakes. The survey did not s p e c i f ic a lly address reasons behind increased demand. However, p r io r research indicated th a t developers' preference may be more important than con­ sumers' preferences f o r location decisions concerning new development 205 (K a is e r, 1968). The producer does not merely r e f l e c t consumer pre­ ferences, but considers other inputs such as s it e c h a r a c te ris tic s , acting as constraints and parameters, and decision agent c h a ra c te ris tic s - those which a f f e c t the p r o f i t and production functions.' Consumer preferences play a p a rt in the f i n a l location decision, but only in an in d ir e c t and p a r t ia l manner. When dealing with re s id e n tia l develop­ ment, i t may be more appropriate to concentrate on c o n tro llin g developers ra th e r than consumers. One recent trend in Michigan has been towards the planned develop­ ment type of recreation communities as opposed to the individual lo t type (Nelson, 1973; F le tc h e r, 1978). I f th is trend continues, s o c ia l, environmental, and economic impacts can be reduced through th is c lu s te r ­ ing. The major burden of land use control in the state rests with local government which generally does not have resources to do an adequate job (Nelson, 1973; American Society o f Planning O f f i c i a l s , 1976). The trend o f controlled growth in subdivisions can be very b e n e fic ia l in slowing down unplanned settlement practices. However, e f f o r t must be expended by local governments to prevent damage from unplanned s e t t l e ­ ment p rac tice s. Not a l l property owners w i l l follow the trend and not a l l recreation al subdivisions w i l l be compatable with the area, y e t local governments w i l l s t i l l be charged with protecting an areas' resources. Therefore, policy should be considered to provide resources local govern­ ments need to control unplanned settlement practices. One other trend surfacing in recent years is creation o f Undivided In te re s ts (UDI) (Dickinson, and Hansen, 1975). This trend toward UDI's has grown out o f problems experiences with t r a d itio n a l 206 recreational developments. Members in a UDI own shares o f land w ithin the club but any in d iv id u a lly owned permanent structures are concentrated in an area where environmental harm is minimized. Land th a t is not developed is used f o r recreation al a c t i v i t y (h ik in g , hunting, e t c . ) o f which only members o f the club may p a r t ic ip a te . The advantages of the UDI are less environmental damage and la rg e r property tax returns fo r the county in the long run. I n i t i a l l y , property taxes are less because most o f the land is not assessed f o r home s it e use in the UDI's. However, in the long run, because no permanent structures w i l l be scattered over the area, net property taxes w i l l be la rg e r than fo r tr a d itio n a l development, because county services w i l l not have to be provided to remote c lie n t s . These are ju s t a few approaches to control unplanned settlement practices. Fu rther'consideration should be given to analyzing a l t e r ­ natives and deciding on a course o f action . Survey resu lts in d ica te unplanned settlement practices w i l l continue to be a problem in northern Michigan, th e re fo re , e f f e c t iv e p o licy is needed to control problems expected to a ris e . 5. Continue development o f land value models. The primary purpose o f the value per acre of land model in th is study was to determine what e f f e c t , i f any, ce rtain natural resources had on the estimated market value o f land. Results showed th a t c e rta in natural resource c h a ra c te ris tic s did contribute s i g n if ic a n t ly to land values. However, the amount of v a ria tio n explained by the model was low in d ica tin g there are l i k e l y more independent variables which con­ t r ib u te s ig n if ic a n t ly to land va lu a tio n . Addition o f more location variables ( i . e . a c c e s s i b i l i t y ) , inclusion o f a wider array of s it e 207 a ttrib u te s ( i . e . shape, s o i l s ) , services provided ( i . e . u t i l i t i e s , roads), use po ten tia l ( i . e . re c re a tio n , investment), ownership patterns ( i . e . zoning, bu ilding codes) w i l l probably lead to a b e tte r prediction model f o r value per acre of land (Real Estate Research Corporation, 1974). Quite possibly aggregated value per acre o f land regression models ( i . e . county or regional) would be more accurate fo r pred iction purposes than the in d ivid ual property owner's model examined in th is study. Research should proceed in th is area as a good chance ex ists to develop useful p re d ic tiv e aggregated models. APPENDICES APPENDIX A Please See the Folder in the Back Cover o f th is Manuscript fo r a Copy o f the Questionnaire 208 APPENDIX B Planning Regions REGION 1 REGION 2 REGION 3 REGION 4 REGION 5 REGION 6 Livingston H ills d a le Barry Berrien Genesee Clinton Monroe Jackson Branch Cass Lapeer Eaton S t. C la ir Lenawee Calhoun Van Buren Shiawasee Ingham Washtenaw Kalamazoo S t. Joseph REGION 7 REGION 8 REGION 9 REGION 10 REGION,11 REGION 12 Arenac Allegan Alcona Antrim Chippewa Alger Bay Ionia Alpena Benzie Luce Delta Clare Kent Cheboygan Charlevoix Mackinac Dickinson Gladwin Lake Montmorency Emmet Marquette G ra tio t Mason Oscoda Grand Traverse Menominee Huron Mecosta Presue Isle Leelanaw Schoolcraft Iosco Montcalm Manistee Is a b e lla Newaygo Missaukee Midland Osceola Wexford Ogemaw Roscommon Saginaw Sanilac Tuscola REGION 13 REGION 14 REGION 15 REGION 16 REGION 17 REGION 18 Baraga Oceana Wayne Otsego Ohio Flo rid a Gogebic Muskegon Oakland Crawford Houghton Ottawa Macomb Kalkaska REGION 19 Iron Keweenaw Other States Ontonagon 209 APPENDIX C M u ltip le Regression Simple C o rre la tio n M atrices CORRELATION COEFFICIENTS. A VALUE OF 99.00000 IS PRINTED IF A COEFFICIENT CANNOT BE COMPUTED. Cl C2 D1 D2 D3 D4 D5 D6 D7 D8 El E2 G1 HI H2 H3 H4 J1 C1G1 C2H2 C2H3 D3G1 D5H2 D5H3 D7J1 E2H3 TOTACRES K1 -.10588 -.06600 -.12071 -.06403 .01006 -.07057 .02684 -.10889 .21492 -.08894 -.33010 .08016 -.06408 -.07480 .03764 .11520 .23717 -.15944 -.02444 -.01654 .10646 -.02824 -.01635 .11047 -.02714 .13213 -.09167 .07349 -.22818 .68679 .32449 .59533 -.10052 -.16423 -.08361 -.42785 -.09385 -.07539 .03275 .27873 -.08823 -.13567 -.08401 .01125 .22858 .44865 -.09531 -.09188 .31464 -.08936 -.08361 -.09237 -.06816 .01082 -.04468 -.15389 -.07404 -.13584 .44052 .71974 .36640 -.58732 -.12986 -.15754 .18832 -.27089 .00865 .21907 .28952 -.13677 .04858 -.10237 .41170 .40265 -.07180 .39162 .36640 -.12781 .24301 -/02094 -.24486 -.03609 -.06621 -.06779 -.11076 -.05639 -.28858 -.06330 -.00686 .07663 .16714 -.06242 -.08494 -.06791 -.07759 .14019 .27414 -.06428 -.06197 -.13500 -.06027 -.05639 -.06230 -.05992 .01330 -.11539 -.03186 -.03262 -.05329 -.02713 -.13884 -.03045 -.00134 -.01661 .01450 -.00342 -.04482 -.03637 -.00778 .15131 .04193 -.03193 -.02981 -.01684 -.02900 -.02713 -.02997 -.02883 .00409 -.06901 -.05984 -.09777 -.04977 -.25473 -.05587 -.10818 -.02202 .23018 -.06380 -.28177 -.03245 .10750 .10565 .35478 .05674 -.05470 .52852 -.05320 -.04977 -.05499 -.02132 -.00114 .10132 VALUACRE Cl C2 D1 D2 D3 210 Appendix C - Continued M u ltip le Regression Simple C o rre la tio n M atrices LMILE RMILE G1 HI H2 H3 H4 J1 C1G1 C2H2 C2H3 D3G1 D5H2 D5H3 D7J1 E2H3 TOTACRES K1 LMILE RMILE D7J1 E2H3 TOTACRES K1 LMILE RMILE .00252 .05197 -.04581 -.06116 -.03418 .01343 -.07198 -.04156 VALUACRE Cl C2 D1 D2 D3 .01945 -.00402 .15854 .16279 .14417 .04506 .01680 .14256 .14508 -.06300 .12792 .12526 -.05057 .29049 -.01447 -.04014 -.11276 -.09924 -.00562 -.08570 -.07080 -.05677 .12833 .71021 -.06021 -.05834 .49808 -.05645 -.05282 -.03907 -.05613 .15436 -.03153 -.07624 -.05805 -.10879 -.08988 -.09226 .03993 -.04000 -.07643 -.0 736 8 -.02805 -.07168 -.06705 .03629 -.07125 -.00188 -.02409 -.08154 -.01317 -.09220 -.09464 -.02899 -.06087 .70254 -.07558 -.04269 .65867 -.06878 -.06055 -.07309 -.03463 -.01009 -.09147 -.17589 -.07819 -.00577 -.05029 -.06478 .81974 -.03527 -.06073 .74594 -.02658 .79273 -.02001 -.01410 -.07557 -.06269 -.05690 -.04176 -.06849 -.06410 -.02897 -.06284 -.05833 -.02984 -.06198 -.02484 .06395 -.07757 -.16435 E2 G1 HI H2 H3 H4 -.04683 .60014 -.02476 -.00218 -.05637 -.04677 -.03865 .00323 -.04249 -.03647 -.03559 -.03209 -.02368 -.05930 -.04970 D5H3 D7J1 L2H3 .01509 .00584 .01410 TOTACRES .00646 .01950 -.07715 -.08967 K1 -.05925 -.04903 .82070 LMILE APPENDIX D Mean and Standard D e v ia tio n s f o r the M u ltip le Regression V a ria b le s V ariab le Mean Standard Deviation VALUACRE 27721.8301 39559.7260 Cl .1416 .3487 C2 .2400 .4272 D1 ' .0698 .2248 D2 .0171 .1296 D3 .0552 .2285 D4 .0577 .2333 D5 .1406 .3477 D6 .0407 .1976 D7 .5261 .4994 D8 .0507 .2194 El .3283 .4697 E2 .3619 .4807 61 .0617 .2408 HI .0959 .2945 H2 .1004 .3006 H3 .0708 .2565 H4 .0743 .2623 J1 .1782 .3828 C1G1 .0321 .1764 C2H2 .0522 .2225 C2H3 .0487 .2153 D3G1 .0161 .1258 D5H2 .0462 .2099 D5H3 .0407 .1976 D7J1 .0492 .2163 E2H3 .0457 .2088 17.5069 98.4575 .6345 .4817 LMILE 5.6376 20.5957 RMILE 4.3273 19.0551 TOTACRES K1 LIST OF REFERENCES LIST OF REFERENCES A ckoff, Russell L. I n c . , 1962. S c i e n t i f i c Method. New York: John Wiley and Sons, American Society o f Planning O f f i c i a l s . Subidividing Rural America: Impacts of Recreational Lot and Second Home Development. Prepared f o r the Council on Environmental Q u a lity , and the Department of Housing and Urban Development. Washington D.C.: Government P rin tin g O ffic e , 1976. Anderson, W illiam Dyer. "Ways to Regulate the Aesthetics of our Rural Environment." Yearbook o f A g r ic u ltu r e , 1970. Washington D.C.: Government P rin tin g O f f ic e , 1970. Ando, A lb e rt and M o d ig lia n i, Franco. "The L if e Cycle Hypothesis of Saving: Aggregate Im plications and Tests." American Economic Review. (March 1963): 55-84. B a ile y , Warren R. "Rural and Urban: The InterFaces." Yearbook of A g ric u ltu re , 1970. Washington D.C.: Government P rin tin g O f f ic e , 1970. Bari owe, Raleigh. Land Resource Economics: The Economics o f Real Property. 2nd ed. Englewood C l i f f s , New Jersey: P r e n t ic e - H a ll, I n c . , 1972. Barlowe, Raleigh and A l t e r , Theodore R. Use-Value Assessment of Farm and Open Space Lands. Development and Public A f f a ir s B u lle tin 308. East Lansing: Michigan State U n iv e r s ity , A g ric u ltu ra l Experiment S ta tio n , 1976. Beale, C. L. "People on the Land Rural Population." Rural U.S.A. Persistence and Change, ed. T. R. Ford, 1978. pp 19-34, 229-232. Beale, C. L. The Revival o f Population Growth in Non-Metropolitan America. United States Department o f A g ric u ltu re Report No. 605. Washington D.C.: Government P rin tin g O f f ic e , 1975. B ird , R. and Kampe, R. 25 Years o f Housing Progress in Rural America. United States Department of A g ric u ltu re Report No. 373. Washington D.C.: Government P rin tin g O f f ic e , 1977. Boxley, R. F. Land Ownership Issues in Rural America. United States Department o f A g ricu ltu re Report No. 655. Washington D.C.: Government P rin tin g O ffic e , 1971. 213 214 Brown, C. J. D. "Michigan Streams - Their Length D is trib u tio n and Drainage Areas." Michigan Conservation X I I I Vol. 5, 1944, 9. Buse, Reuben C. and Bromley, Daniel W. Applied Economics Resource A llo catio n in Rural America. Ames: Iowa State U niversity Press, 1975. C a r lin e r , Geoffrey. 50 (May 1978): "Determinants of Home Ownership." 109-119. C l i f f , Andrew D .; Frey, A llan ; and Haggett, Peter. New York: John Wiley and Sons, I n c ., 1977. Land Economics Location Methods Coughenour, C. M. and Busch, L. "A lternative Futures fo r Rural America: The Cloudy Crystal B a ll," Rural U.S.A. Persistence and Change, ed. T. R. Ford, 1978. pp. 211-228, 242-245. Crapo, Douglas. "Recreation A c tiv ity Choice and Weather: The S i g n i f i ­ cance of Various Weather Preceptions." Ph.D. d is s e rta tio n , Michigan State U n iv e rs ity , 1970. Dickinson, Thomas E. and Hansen, David E. "Undivided In te re s ts : Im plications of a New Approach to Recreational Land Development," Land Economics 51 L I , 2 (May 1975) 124-132. Erdos, Paul L. Professional Mail Surveys. Company, 1970. New York: McGraw H i l l Book F letch er, James. "A Systematic Approach to the Analysis o f Land Sales Regulatory Programs: A Case Study of the Michigan Land Sales Act of 1972." Ph.D. d is s e rta tio n , Michigan State U n iv e rs ity , 1978. Ford, T. R . , ed. "Contemporary Rural America: Persistence and Change." Rural U.S.A. Persistence and Change. 1978. pp 3-16, 229. Fremon, Suzanne and Wilson, Morrow., ed. H. W. Wilson, 1976. Rural America. New York: Friedman, M ilton . A Theory of the Consumption Function. National Bureau of Economic Research. Princeton: Princeton U n iversity Press, 1957. G alin, J. Summaries of Interviews Concerning the Impact of Rapid Popula­ tio n Growth Upon Selected Areas in Northern Michigan. Ann Arbor: The U n iversity of Michigan, I n s t it u t e of Social Research, 1976. Gambel, Hay B. et a l . Environmental Q uality Effects Associated with Seasonal Home Communities. A g ricu ltu ral Experiment Station B u lle tin 801. U n iversity Park: The Pennsylvania State U n iv e rs ity , 1975. Glahe, Fred R. Macroeconomic Theory and P o lic y . Brace Tovanavich, I n c . , 1973. New York: Harcourt 215 Gregory, S. S t a t i s t i c a l Methods and the Geographer. York: Zongman, 1978. 4th ed. New Humphreys, C. R. and Green, R. F. Michigan Lakes and Ponds. Lake Inventory B u lle tin s 18 and 21. East Lansing: Michigan State U n iv e rs ity , 1962. Is a rd , Walter and Langford, Thomas W. Regional Input-Output Study: R ecollectio ns, R eflections and Diverse Notes on the Philadelphia Experience. Cambridge: The M .I.T . Press, 1971. K aiser, Edward J. "Location Decision Factors in a Producer Model o f Residential Development," Land Economics 44 (August 1968) 351-362. Marans, Robert W., e t a l . W aterfront Living: A Report on Permanent and Seasonal Residents in Northern Michigan. Ann Arbor: The U niversity o f Michigan, I n s t i t u t e of Social Research, 1976. Marans, Robert W. and Wellman, John D. The Q u ality o f Non-metropolitan Living: Evaluations, Behaviors, and Expectations of Northern Michigan Residents. Ann Arbor: The U n iv ersity of Michigan, I n s t i t u t e o f Social Research, 1978. M arshall, A lfr e d . P rin c ip le s of Economics. York: The MacMillan Company, 1891. Vol. 1 . , 2nd ed. New McEwan, Douglas. "A Survey of Non-resident Landowners in Clare and Gladwin Counties." M. S. theses, Michigan State U n iv e rs ity , 1970. Michigan Public Opinion Survey. Prepared by the Department o f Resource Development, Cooperative Extension Service and the A g ric u ltu ra l Experiment S ta tio n . East Lansing: Michigan State U n iv e rs ity , 1977. Morley, Frederick. Michigan and i t s Resources. and Company, 1881. Lansing: W. S. George Nelson, Burton D. Second Home Development in Michigan. Prepared fo r Northern Michigan T o u rist and Development Association, 1973. N ie , Norman H ., e t a l . S t a t i s t i c a l Package f o r the Social Sciences (SPSS) . New York: McGraw-Hill Book Company, 1975. O'Hare, W illiam ; Beegle, A llan J . ; and Leonard, Jim. Recent Changes in Population Growth and D is trib u tio n in Michigan. Forthcoming, A g ric u ltu ra l Experiment S tation Report. Payne, Brian R.; Gannon, Richard C.; Ir la n d , Lloyd C. The Second-Home Recreation Market in the Northeast. Prepared fo r the Bureau of Outdoor Recreation, U.S. Department o f the I n t e r i o r by the Forest S ervice, U.S. Department o f A g ric u ltu re . Washington D.C.: Government P rin tin g O f f ic e , 1974. Peterson, Roger G. Exercises in S t a t i s t i c a l In fe re n c e . O.S.U. Book Store, I n c . , 1972. C o rv a llis : 216 Ragatz, Richard L. Vacation Homes: An Analysis of the Market fo r Seasonal - Recreational Housing. Ithaca: Cornell U n iv e rs ity , 1969. Real Estate Research Corporation. The Costs of Sprawl: Environmental and Economic Costs o f A lte rn a tiv e Residential Development Patterns a t the Urban Frin ge. Prepared fo r the Council on Environmental Q u a lity , the O ffic e o f Policy Development and Research in the Department o f Housing and Urban Development, and the O ffic e o f Planning and Management in the Environmental Protection Agency. Washington D.C.: U.S. Government P rin tin g O ffic e , 1974. Santer, Richard A. Michigan: Heart of the Great Lakes. Kendall/Hunt Publishing Company, 1977. Dubuque: Schertz, L. P. Rural America and Environmental Q u ality: Issues and P otential Impacts S itu a tio n Studies. National A g ric u ltu ra l Outlook Conference Proceedings, 1976, 88-99. Stevens, S. S. "On the Theory of Scales and Measurement.11 Science 103, 1946, 677-680. Thomas, Edwin. "Maps o f Residuals from Regression." S patial Analysis: A Reader in S t a t i s t i c a l Geography, ed. Brian Perry and Duane Marble, Englewood C l i f f s , N .J .: Prentice H a ll, 1968. Tombaugh, Larry W. "Factors A ffe c tin g Home Location." Journal Leisure Research. Vol. 2. No. 1 (Winter 1970) 54. U.S. of Department o f A g ric u ltu re , Forest Service. Questionnaires fo r Research: An Annotated Bibliography on Design, Construction and Use. USDA Forest Service Research Paper PNW-140. P ortland, Oregon, 1972. U.S. Department o f Commerce. Second Homes in the United States Current Housing Reports. Series H-121, No. 16. Washington D.C.: Govern­ ment P rin tin g O f f ic e , 1969. U.S. Department o f Commerce. Weather Bureau. Climate o f Michigan by Stations No. 20. Washington D.C.: Government P rin tin g O f f ic e , 1963. U.S. Department of I n t e r i o r , Bureau of Outdoor Recreation. Northern New England Vacation Home Study, Washington D.C.: Government P rin tin g O f f ic e , 1966. U.S. Rural Development Service. Rural Development: Sixth Annual Report of the President to the Congress on Government Service to Rural America Programs. Washington D.C.: Government P rin tin g O f f ic e , 1976. Veatch, J. 0. S oils and Land of Michigan. State College Press, 1953. East Lansing: The Michigan 217 V ertrees , Robert. "A Survey o f Non-Resident Landowners o f Ten or More Acres in Antrim and Kalkaska Counties, Michigan." M.S. theses, Michigan State U n iv e rs ity , 1967. Verway, David I . , ed. Michigan S t a t i s t i c a l A bstract. Michigan State U n iv e rs ity , 14th e d itio n , 1979. East Lansing: V la s in , R. D .; Libby, L. W.; and Shelton, R. L. "Economic and Social Information fo r Rural America: P r i o r i t i e s fo r Immediate Improve­ ment of Information Systems." American Journal o f A g ric u ltu ra l Economics. 59 ( 5 ) , (December 1975), 900-909. W a llis , Kenneth F. Introductory Econometrics. Publishing Company, 1972. Chicago: Aldine Ward, David J. "Serving Man's Needs: Our Rural Environment." Yearbook o f A g ric u ltu re , 1970. Washington D.C.: Government P rin tin g O f f ic e , 1970. White, Terrence H. "The R elative Importance of Education and Income as Predictors in Outdoor Recreation P a r tic ip a t io n ." Journal of Leisure Research. Vol. 7, No. 3. (1 9 7 5 ), 191. W ilkening, E. A. and Klessig, L. "The Rural Environment: Q u ality and C o n flic ts in Land Use." Rural U.S.A. Persistence and Change, ed. T. R. Ford (1978) pp 19-34, 229-232. Wolfe, R. I . "Communication." No. 1 (W inter, 1970), 54. Journal of Leisure Research. Vol. 2, We need your help! I'm sure vou re probably aware of many of the problems associated with an increasing level of development (i.e. prop­ erty tax increases because more local services are required). The Department of Resource Development at Michigan btate University is conducting a study of property owners in nine townships in Michigan. When you answer the questions that follow it will help us get an idea of what causes property ownership and home development to occur. This will also benefit you because the results of our study will be made available to you through your county agent. This information can be used to get an idea of what to expect in your area, in terms of property ownership, and home development patterns, for the future. The questionnaire should take no more than ten minutes to fill out. Your privacy is assured as we have no way of knowing the name of the person who completes the questionnaire. Please put the completed questionnaire in the . enclosed self addressed stamped envelope and slip it into tfie mailbox at your earliest con­ venience. Thankyou. Sincerely, Bill Gartner Research Assistant \ "1 mm ms A. Our first section deals with property you own or lease in Baglay Township T30W-R3W in Otsego County. .sold . obtained “ bought ____ . Other (please explain) leased __ 4. Are you aware of the zoi codes or percolation te ment on your land? .year yes 6. W hat is the current total acreage size of the prop­ erty you own or lease? . acres , 2. What was your main reason for obtaining (buying or leasing) your property? (Please chock only the most important reason.) . year .sold .obtained 1, How did you acquire your first piece of property? inherited , acres 5. How do you feel about tions? _ J favor stricter lam 7. Do you intend to sell any of your property in the future; if yes how many acres and within how many years? . yes _____ present land use c i favor lessoning ; trols . number of acres investment for a retirement homo location . I'm against ui: lam . within a year to get out of the city . no .o th e r_____ ____ 6, How do you feel about ft: . not sure I oppose any futui _ ___ local newspaper and magazine ad _ _ 4. In what year did you obtain (or obtain right to use) your first piece of property? B. In th is section w e would lik e to g e t y o u r a t­ titu d e s on various issues th a t are o f concern to property owners in Bagley Township T30M-R3W in O tsego County. ; . it will increase ran) will increase mode; ! will decrease modi . high Since you first obtained your land have you ob­ tained or sold any adjacent land? . about right . low 2. How do you feel about the quality of municipal or county services? . no very good if yes, please give the amount of acreage obtain­ ed or sold and the year the transaction took place. . acres . year . sold »obtained . acres „ year _ sold _ obtained . acres . year ____ good below average 3. . sold _ obtained . soid _obtained 7. How do you fee! about values? will stay the same i year yes _____ I would like to see I'm not sure 1. How do you feel about property tax levels? 5. I would like to see _____ dollars -.re a l estate salesperson __ other ..._______ .not sure .over 10 years 8. Please indicate what you think you could get for your property if you put it all up for sale (include any dwelling). 3. Where did you first learn of your property? relatives ....... 1-0 years 6-10 years __ _ to have a place to hunt/fish inherited the property ,_ __™_average poor How do you feel about the quantity of municipal or county services? could use a lot more ... could use some more ___ about right . we don't need so many , would like to see quite a tot less . acre*; . year not sure , not sure _ _ _ will decrease radics not sure i C. This se ctio n cisals w itf p ro p e rty in ' ;y ToO tsego C o unty and itf natural resources. j 1. Is your property located ef or publicly developed ski j miles is it from your propel yes i . no 2. Does any state or fed era one side, your property? .yes year 4. Are you aware o f the zoning regulations, building codes or percolation tests concerning develop­ ment on your land? .year .yes •no .not sure 5, How do you feel about present land use regula­ tions? .1 favor stricter land use controls jnurnber of acres pi ft . I favor lessening of present land use con­ trols b. If yes, what type body of water? 6. How do you feel about future residential building? I oppose any future development I would like to see a little development , I would like to see a lot more development c. If the answer to b was "la k e o r p o n d ," then what size is the lake or pond your property is located on? .less than 25 acres _ 2 5 -1 0 0 acres .100-500 acres , over 500 acres 4. a. If the answer to 3a was " n o " then what type of body of water is your property closest to? lake 7. How do you feel about the future of property values? ____ it will increase radically will stay the same will decrease moderately fit __ tow |y of municipal or ___ _ average pr not sure ntity of municipal will decrease radically -lake. . miles , river following questions. not sure C. This section deals with the location of your property in Bagley Township T30N-R3W in Otsego County and its closeness to certain 1. What kind of water system do you use in your liv­ ing quarters (check the one most commonly us­ ed)? _____ well w ith electric or gas pump natural resources, 1. Is your property located close to any commercial or publicly developed ski area? If yes, how many miles is it from your property? , yes .miles , no ot less . miles D. M a n y people w h o o w n p ro p e rty , in Bagley Township T30N-R3W in Otsego County also have som e ty p e o f liv in g q u a rters (either seasonal or permanent) on the land. If you happen to have some ty p e o f living quarters on the land w e would like you to answer the _____ will increase moderately |ax levels? both . river to. How may miles by road is your property from that body of water? I'm not sure Jto get your at[ are of concern jley Township .both not sure .over 10 years Sou could get for fto r sale (include no . present land use controls are adequate . I'm against all land use controls _ ___1-5 years yes __ lake or p o n d _____river ;ize of the prop- property in the snd within how 3. a. is your property located on any body of water? 2. Does any state or federal land touch, on at least one side, your property? yes hand pump municipal water system 2. What kind o f sewage system does your living quarters have (check the one most commonly used)? individual septic tank hook up to a municipal sewage system outhouse of dry well _ _ other ......................... ................. F. We now would like responses from those people who have located their permanent home in Bagiev Township T30M R3W in b. What is the sex of tf male Otsego County. 3. a. Are your living quarters of the conventional housing typo (fremo, cementblock, etc.) os the mobiic home typo? 1. Approximately in what year was your home built? —: c. What is the mas it. household? ___ _ married year conventional housing single (never ______ _ _ __ not sure mobile home b. If you checked mol ie home" in a, is the mobile home abl“ to no moved easily or is it anchored in placer _____ divorced 2. a. Before it became your permanent hoi no was your home used on a seasonal basis by you or someone else? _ .widowed 3. Please indicate the nur propriate age brackets j you. I I yes _ can be moved easily no anchored in place under 5 not sure b. If yes, in what year did you or someone else j '.......j 5-14' .....' ....... make the seasonal home a permanent home? j ■15-25 __ _______ _ E. in this section w e would like responses from those people who have located a seasonal home in Bagley Township T30N-R3VV in Otsego County. 1. About how many annua! visits do you, your friends, or relatives make to your seasonal home? ______;_________ _ ' | year ' 26-64 __ :...... ......... ..... ■ .______________ not sure 3. Before moving into your permanent home in what county and state did you reside? ___ ________________ days 3. Piease indicate about how many days do you visit your seasonal home during each season of the year? _________ number of days in Fall _ number of days in Winter ................. . number of days in Spring number of days in Summer . 1 ' ' ' 4.■ What is the approxtnjI before taxes, for your hi it 1 ■' county _ 0 -$ 5 ,9 9 9 ■' . . j ...........................state _ _ _ _ VlSitS 2. In general what is the average length of stay for each visit? - 65 and over ________ I G. Finally, we have a few background ques­ tions we wo uld like to ask. It is important to keep In mind that these questions are asked for statistical purposes only and your privacy will be insured. There is no way of knowing the identity of tire person filling out the questionnaire. 1. What county and state do you presently reside ■' fn? ;■ ________ c o u n t y _____________ _ state 2, a. What is the age of the head of household? years $6,000-$S,999 I * ' I 14,9986 ■""11111111,1$10,000-$ ' ' ' 1 .' . . ■ ' ■ 1 $15,000-525,000| I - i over $25,000 v • a! ii ' 1 ' •' i 1 Please review the q have answered a ll the -el! rters of the conventional cementblonk, etc.) or the F. We now would like responses from those people w ho have located their permanent home in Bagley Township T30N-R3W in Otsego County. 1. Approximately in what year was your home built? ... ■ b. What fsthe sex of the head of household? _ _ _ _ _ female mala c. What is the marital status of the head of household?.. married year . lousing single (never married) ____________ _ not sure sbite h o m e " in a, is the i be moved easily or is it divorced 2. a. Before it became your permanent home was your home used on a seasonal basis by you or someone else? yes .widowed 3. Please indicate the number of people in the ap­ propriate age brackets who currently reside with you. easily no ace not sure b. If yes, in what year did you or someone else make the seasonal nome a permanent home? under 5 ____________________ no.ofpeople 5-14 no.ofpeople :__________ 15-25_______________________ no.ofpeople ____________ year uk! like responses from ave located a seasonal >wnship T30N-R3W in lual visits do you, your e to your seasonal home? 26-64'_________________________no. of people ____________ not sure 65 and over ___________________ no. of people 3. Before moving into your permanent home what county and state did you reside? in ___________ _ _ _ _ _ _______ __ _ county 0-$5,999 __________________ state $6,000-$9,999 its >w many days do you visit -ring each season of the G. Finally, we have a fe w background ques­ tions we would like to ask. It is important to keep in mind that these questions are asked for statistical purposes only and your privacy will be insured. There is no way of knowing the identity of the person filling out the questionnaire. ays in Fall 1. What county and state do you presently reside in? iverage length of stay for re ays in Winter ays in Spring ays in Summer 4. What is the approximate total family income before taxes, for your household in 1977? ___ c o u n t y ______________ _ state $10,000-$14,999 $15.000-$25,000 — over $25.000 Please review the questionnaire to see i f you have answered a ll the questions th at apply. Thank you. 2. a. What is the age of the head of household? ____________ years M ic h ig a n S ta te U n iv ers ity P rin tin g