IN F O R M A T IO N TO USERS This m aterial was produced from a m icrofilm copy of the original docum ent. W hile the most advanced technological means to photograph and reproduce this docum ent have been used, the qu ality is heavily dependent upon the qu ality of the original subm itted. follow ing explanation o f techniques The markings or patterns w hich may appear on this reproduction. is provided to help you understand 1. The sign or "ta rg e t" 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 in to the film along w ith adjacent pages. This m ay have necessitated cutting thru an image and duplicating adjacent pages to insure you com plete co n tin u ity. 2. When an image on the film is obliterated w ith a large round black m ark, it is an indication th at the photographer suspected th a t the copy may have moved during exposure and thus cause a blurred image. Y o u w ill find a good image o f the page in the adjacent fram e. the photographer 3 . When a m ap, drawing or chart, etc., was part o f the m aterial being photographed in "sectioning" the m aterial. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections w ith a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until com plete. a d e fin ite m ethod follow ed 4. The m a jo rity o f users indicate th at the textual content is of greatest value, however, a som ewhat higher quality reproduction could be made from "photographs" if essential to the understanding o f the dissertation. Silver prints o f "photographs" may be ordered at additional charge by w riting the Order D ep artm ent, giving the catalog num ber, title , author and specific pages you wish reproduced. 5. PLE A S E N O T E : Some pages m ay have indistinct p rin t. Filmed as received. Xerox University Microfilms 300 North Zeob Road Ann Arbor, M ichigan 48106 7 6 - 1 8 ,6 0 7 CHEEK, William Henry, 1942- MOBILE HOME PARKS IN SOUTHERN MICHIGAN: INFLUENCING THE LOCATION OF AN EVOLVING RESIDENTIAL LAND USE. FACTORS Michigan State University, Ph.D., 1976 Social Geography Xerox University Microfilms, Ann A r b o r Mi chi gan 4 8 1 0 6 MOBILE HOME PARKS IN SOUTHERN MICHIGAN: FACTORS INFLUENCING THE LOCATION OF AN EVOLVING RESIDENTIAL LAND USE By William Henry Cheek A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Geography 1976 ABSTRACT MOBILE HOME PARKS IN SOUTHERN MICHIGAN: FACTORS INFLUENCING THE LOCATION OF AN EVOLVING RESIDENTIAL LAND USE by Will lam Henry Cheek This study is concerned with the location of an evolving and often controversial residential land use. The recent accelerated rate of construction of mobile home developments in many areas of the nation has been docu­ mented . The accompanying changes and the numerous and important implications to those concerned with American housing and the growth of metropolitan areas are less well understood. The study recognizes both the relative complexity of mobile home park location and the element of change. Within this context a large number of site and area fac­ tors expected to Influence the Intrametropolitan location of parks were enumerated. These include: (1) accessibil­ ity to the urban center, employment, and major highways, (2) availability of undeveloped land, (3) characteristics of the immediate area including land use, socioeconomic prestige level, and features incompatible to residential development, and (A) institutional factors of zoning and availability of municipal services. These factors were intended to encompass both older and more recent construc­ tion. However, in view of the evolutionary process, the William Henry Cheek relative importance of the factors were expected to vary between different periods of mobile home park construc­ tion . Analysis of these factors was conducted In Genesee County, Michigan of the Flint Standard Metropolitan Sta­ tistical Area. The sixty developments containing over 8,500 spaces constructed prior to January, 1974 were used in the analysis. Thirty-six parks were constructed prior to 1958, and the remaining twenty-four opened between 1965 and 1973. Data for this research were obtained from a vari­ ety of sources. Much of the basic data were obtained from files and reports of the Michigan Department of Health which licenses all parks in the state. Low altitude aerial photography and high altitude RB-57 imagery pro­ vided adjacent and area land use data. Information per­ taining to zoning and other site characteristics was ob­ tained from published planning and zoning reports and maps. Pertinent social and economic data were extracted from 1960 and 1970 census materials. Field investigations were carried out primarily between January and July, 1973. Numerical description, cartographic presentation, and simple statistical techniques were employed in the re­ search. Findings indicate that the dichotomy of mobile home developments based on age and the use of site and William Henry Cheek area factors Is a valid approach In the analysis of park location. The substantial differences In size, density, and amenity features are Indicative of internal change between the different periods of park construction. The relative Importance of combinations of site and area fac­ tors Influencing park location of the two periods also Indicated change. Of particular significance was the Increasing importance in factors of the area environment including adjacent and area land use, especially unmlxed adjacent land use, the socioeconomic level of the immedi­ ate area, and the absence of nearby negative features. Availability of municipal services were of less influence than had been expected. M o d e m mobile home de­ velopments, however, as Indicated by adjacent and area zoning, were more greatly Influenced by residential zoning and the absences of commercial and Industrial zoning than were parks constructed during the earlier period. The implications of this research center primarily on the evidence of change and of complexity Involved in mobile home park location. The internal changes Including larger size of parks, lower densities, and greater ameni­ ties are more easily recognized than those involving loca­ tion. Evidence indicates, however, that the modern mobile home development has evolved in a relatively short time span so that a large number of site and area factors must be in­ cluded in locational studies. The findings suggest a Will loro Henry Cheek convergence toward similar site and area factors Influenc­ ing other forms of residential development and mobile home parks. ACKNOWLEDGMENTS There are many people who helped me during this research effort. 1 wish to express my thanks to my com­ mittee which was chaired by Dr. Gary Higgs and consisted of Dr. Gary Manson, Dr. Clarence Vlnge, and Dr. Raleigh Barlowe. My colleagues In the Department of Geography and Geology at Southwest Missouri State University offered encouragement and criticism that was most appreciated. Ms. Betty Hindman did an excellent Job in typing the manuscript. Finally I express a very special thanks to my wife Maret, and my daughter Karen, for their love, patience, understanding, and encourage* sent during the work on this dissertation. ii TABLE OF CONTENTS LIST OF T A B L E S ................................. vii LIST OF F I G U R E S ..................................ix Chapter One. INTRODUCTION . . . . . ..................... Mobile Homes and the Housing Supply . . . . Page 1 1 The Increasing Importance of Mobile H o m e s ............................ Geographical Variation of Mobile H o m e s ............................ 1 1 Mobile Home Developments . . . . . . . . 3 Need for R e s e a r c h ................... 6 The Changing Role of Mobile Housing . . . . 7 Purpose of R e s e a r c h ................. 9 The Study A r e a ....................... 10 Organization of the S t u d y ........ .. 12 Two. A BACKGROUND TO THE R E S E A R C H ........... 14 I n t r o d u c t i o n ......................... 14 Mobile Home Research ............... 15 Mobile Home Developments In the United States ................. . . . . . 21 Data S o u r c e ..................... .. Distribution of Developments Relative Importance . . . . 21 21 2 3 H i Chapter Page Mobile Home Developments In Michigan . . . 33 Distribution ........................... 33 Relative Importance ..................... 33 Local S t u d i e s .......................... 38 An Approach to the I n v e s t i g a t i o n ..... 43 Three. RESEARCH D E S I G N .......................... 47 I n t r o d u c t i o n .......................... 47 Determination of Site and Area F a c t o r s .............................. 30 Literature B a s e ......................50 Modification ........................... 51 Description of Factors Used in Study . . . 53 Accessibility ........................... 53 Availability of Undeveloped Land . . . . 54 Environment of the Immediate Area . . . . 55 Institutional Variables ................. 60 Type of Devel o p m e n t ....................62 Data A n a l y s i s .......................... 63 Four. MOBILE HOME DEVELOPMENT IN GENESEE COUNTY: DESCRIPTION ..................... 67 I n t r o d u c t i o n .......................... 67 Growth of Mobile Home Developments . . . . 67 The Early P e r i o d ....................67 The Intervening Y e a r s ................73 The Latter P e r i o d .................... 73 Location by Political T y p e ........... 76 i v Chapter Page Changes in P a r k s ........................... 81 S u n m a r y ...................................... 82 Five. MOBILE HOME DEVELOPMENTS IN GENESEE COUNTY: ANALYSIS OF SITE AND AREA F A C T O R S ........................................ 86 Introduction ............................. 86 Accessibility Factors ..................... 86 Distance to Urban Center ............... 86 Distance to Center of Employment . . . . 87 Distance to H i g h w a y s ..................... 90 Influence of Accessibility F a c t o r s ..................................90 Availability of Undeveloped Land ........ 92 Area Environmental F a c t o r s ................. 94 Socioeconomic Prestige ................. 94 Land Use in the Immediate A r e a ...........94 Index of Incompatible U s e s ............... 98 Distance to Elementary School .......... 102 Influence of Environmental Factors . . . 104 Institutional Factors ..................... 105 Z o n i n g ................................... 105 Type of Sewer S y s t e m .................... 107 Type of Water S y s t e m .................... 110 Influence of Institutional Factors . . . 110 S u m m a r y ..................................... 113 Six. SUMMARY AND I M P L I C A T I O N S .................... 117 I n t r o d u c t i o n ........................... . 117 v Chapter Page Summary of F a c t o r s ..........................120 Suggestions for Future Investigations . . . 129 APPENDIX A - United States D a t a ....................... 131 APPENDIX B - Michigan D a t a ............................137 APPENDIX C - Zoning Ordinances and Maps C o n s u l t e d ..............................145 LIST OF R E F E R E N C E S .................................... 146 v i LIST OF TABLES Table 1. 2. 3. 4. 5. 6. 7. 8. 9. Mobile Home Household Characteristics: 1970 Page 18 Development and Mobile Home Location Quotient and Selected Variables .......... 28 Summary of Michigan Development Char­ acteristics, 1960-1975 36 Mobile Home Development Construction in Genesee County by Year: 1950-1973 . . . . 70 Mobile Home Developments in Genesee County by Political Type: 1951-1974 . . . 7;* Size of Parks In Number of S p a c e s .......... 82 Size of Parks In Number of Developed A c r e s ........................................ 83 Density of Parks in Spaces Per A c r e ........... 84 Distance to the Urban C e n t e r .............. 88 10. Distance to Center of E m p l o y m e n t .......... 89 11. Distance to State or Federal Highway . . . . 91 12. Percent of Surrounding Area Not in Urban D e v e l o p m e n t ........................... 93 13. Socioeconomic Prestige of A r e a ................95 14. 15. Residential Land Use in Surrounding A r e a ........................................ 97 Nonresidentla1 Land Use in Surround­ ing A r e a .................................... 99 16. Land Use Adjacent Park B o u n d a r y ...............100 vii Table Page 17. Index of Incompatible U s e s .................... 101 18. 19. Distance to Nearest Elementary School In D i s t r i c t .......... 103 Residential Zoning in the Surrounding A r e a ........................................ 106 20. Zoning Adjacent Park Boundary ............... 108 21. Specific Zoning of S i t e ........................ 109 22. Type of Sewer System U s e d ...................... Ill 23. Type of Water System U s e d ......................112 Al. Mobile Home Data by S t a t e ......................131 A2. Mobile Home Development and Mobile Home Household Data by S t a t e ..................... 134 Bl. Michigan Mobile Home Data by C o u n t y ........... 137 B2. Michigan Mobile Home Development Data by C o u n t y ..............................140 v i i i LIST OF FIGURES Percent of annual starts by type of housing: 1956-1974 ................... Mobile homes as a percent of year-round housing, 1970 ......................... Michigan showing study area and SMSA's Location and size of mobile home developments in the Genesee County study area . . . . ................... Number of mobile home development spaces by state, 1970 ................. Development quotient by state, 1970 . . . Percent heads of mobile home house­ holds age 65 and over by state, 1970 Percent of mobile homes in SMSA's by state, 1970 ...................... . . Median income of owner occupied mobile homes by state, 1969 . Percent change In mobile homes 1960-1970 by s t a t e ..................... Number of mobile home development spaces in Michigan by county, 1975 .......... Development quotient in Michigan by county, 1970 ......................... Density ranges by type of residential development ........................... An example of a crosstabulation table Illustrating kinds of Information Included from CROSSTABS routine . . . . ix Page 2 4 11 13 22 25 27 29 31 32 34 37 42 65 Figure 15. Location of mobile home developments, cities, villages, and townships In Genesee County, 1973 Page 68 16. 17. 18. 19. Location of mobile home developments constructed prior to 1957 .................. 72 Location of mobile home developments constructed 1965-1973 ...................... 75 Location of pre-1957 mobile home de­ velopments by s e c t o r ....................... 77 Location of mobile home developments constructed after 1965 by s e c t o r ........... 78 x CHAPTER ONE INTRODUCTION Mobile Homes and the Housing Supply The Increasing Importance of1 Mobile Homes The single family dwelling has long been the domi­ nant form of housing in the United States. As recently as 1958, approximately 80 percent of new dwellings added annually were single family detached homes (U.S. Depart­ ment of Commerce, 1959) (Figure 1). In recent years, how­ ever, this percentage has decreased. Apartments have accounted for much of the decline. Comprising slightly over 10 percent constructed in 1956, two-or-more family dwellings made up over 35 percent of new housing starts between 1969-19 73 (U.S. Department of Commerce, 19 74). A more recently emerged housing alternative is the mobile home. Since 1956, mobile home production increased proportionally from 10 percent to almost 22 percent of new housing additions by 1974. Between 1969-1974, the propor­ tion has been in the 20-22 percent range (U.S. Department of Commerce, 1974). Geographical Variation of Mobile Homes Aggregate national data conceal very important 1 2 ao 70 00 — so uj O 40 — 0 20 0 ■ I I I I I I I I I I I I I I I I I I 19 56 58 60 62 64 66 66 70 72 74 Y E A R Figure 1. Percent of annual starts by type of housing: 1956-1974 (Source: Computed from data in Construction Review. U.S. Department of Coomerce). SOURCE C O N S T R U C T IO N R E V I E W 3 information about mobile housing. Figures derived from the 1970 Census of Housing show that mobile homes ac­ counted for 3.1 percent of year-round housing in 1970 (U.S. Bureau of the Census, 1972b: 242-247). Based on census regions, however, this proportion varied from 1.45 percent in the Middle Atlantic States to 6.43 per­ cent in the Mountain States. Within the 48 contiguous states, the extremes in proportions of mobile homes to year-round housing were from 0.59 percent in Massachu­ setts to Nevada's 11.95 percent (Figure 2).*" Thus, mobile homes are an increasingly important part of the nation's housing, but their importance varies consider­ ably from region to region. Mobile Home Developments At first glance, 3.1 percent of total year-round housing for the nation as a whole, or even Nevada's approximately 12 percent, seems almost insignificant. However, it should be remembered that only during the past ten years has annual production reached even 200,000 mobile homes. Presently, an estimated one-half of all mobile homes are on individual lots particularly in small towns and rural areas (Swaback, 1971: 327; Cloos and Bir- gells, 1972: 9). Within metropolitan areas, however, and ^The computer mapping routine, CPLETH, was written by Dr. Waldo Tobler of the Department of Geography, Univer­ sity of Michigan. The program is on permanent file at the Computer Center, Southwest Missouri State University. « * + 44+4+4444--- + + + RRHMHHWHHHHHHHfl«HHHftHR«t» + ♦ ♦♦♦** + +++44*-------------- . * u r r m r u h ^ r m r h RHRRRRRk HHh R * + + + + + .,*♦.»,, +------------ t4^MHHHRURURMHMRRNMRRHRBHM+ * . + + + + t + 4+* >4 ------ - t + HRHHRHHHR4SO«4 ^ K lL A H ilO O f t l A N S I U S m , i | i i F 4 ( M * - l # u S < c F Q O » H C I O M T 9 JO 4A-0IMA* I I 101(00 m**'-. Figure 3. Michigan showing study area and SMSA's. 12 than 8500 spaces at the end of 1973 (Figure 4). 3) Variety in size, ages, and amenities of the sixty developments. Organization of the Study Chapter Two is devoted to a review of literature pertinent to the problem. Also included will be a descrip­ tion of the regional variation of mobile home developments and of the approach used in this research. The determina­ tion of site and area factors is presented in Chapter Three. Chapter Four is an overview of mobile home devel­ opments in Genesee County (Flint), Michigan. The analysis of site and area factors influencing parks in the study area is found in Chapter Five. Chapter Six presents the major implications and conclusions of this research. 13 Figure 4* Location and size of mobile home developments In the Genesee County atudy area (Source! Michigan De­ partment of Health and field data). CHAPTER TWO A BACKGROUND TO THE RESEARCH Introduc tion Even with the rapid increase in the use of mobile housing in the United States, literature concerning many of its facets is limited. It is true that articles con­ cerning mobile housing in the popular press are more numerous than in the past. These tend to be reports on the rapid growth of their manufacture and sale or expres­ sions of concern about the effects mobile homes and parks have on American housing or the local community. Except as sometimes interesting reading, these reports are of little direct utility to either public decision makers or researchers. Of concern in this section will be that portion of the literature consisting of research reports. These too are limited but perhaps this should be expected. Mobile homes comprise only slightly over 3 percent of the nation's housing supply. Further, mobile home developments are of very recent origin as a part of residential land use. It could be that mobile housing has been thought too inconse­ quential to date to be of major concern. As stated In a recent publication: 14 15 It may well be In the future that the mobile home industry will play a much larger role in providing housing for Americans, especially for lower-income people, and land conversions to this type of residential occupancy will thus grow In importance (Clawson, 1971: 91). Mobile Home Research A common characteristic of much mobile home re­ search has been an all inclusive approach. The tendency has been to cover numerous topics; a typical format in­ cludes a brief history of the industry, a section de­ scribing household characteristics, perhaps one on manu­ facturing, and a portion devoted to mobile home parks. No doubt there has been a need for this approach in the past. The time has come, however, to concentrate on certain identifiable problems of mobile housing. The focus of this research is location of mobile home developments. Location has not been an important consideration in most previous studies of mobile hous­ ing. This is partly due to few geographers conducting research on the subject (See Abu-Ayyash, 1972: 28-30). With few exceptions whenever location was considered, it has been treated superficially, it should also be noted that few researchers have separated mobile home location in general from mobile home development location. While parks are obviously related to the distribution of mobile homes, an estimated 50 percent are on individual lots and there are likely to be different factors involved. How­ ever, certain information from this literature bears on 16 the question of development location and provides a start­ ing point for this study. The desire to know more about the characteristics of mobile home households has been one of the most common themes and dates back to the 1940's (Cowgill, 1941). At first, the concern was the adverse effects of "trailer life" (Fuller and Meyers, 1941). Later, findings indi­ cated a rather wide range of characteristics and more heterogeneous mobile home households than originally thought (Lemert, 1951; Edwards, 1964; French and Hadden, 1965: 136-138; Drury, 1972: 53-71; Newcomb, 1971: 27- 32) . One of the problems with drawing definitive con­ clusions from these studies was the sample on which they were based (Newcomb, 1972: 28-30). For example, a survey taken by the Census Bureau for the Department of Housing and Urban Development in 1965-66 was not a survey of all mobile home households (U.S. Department of Housing and Urban Development, 1968: 66-148). Rather, it was a sample of purchasers of new mobile homes. It found the "typical" mobile home household (meaning modal group or median) to be younger and smaller than the national average. The head of household typically had a high school education and was employed in a blue collar occupation. Addition­ ally, the family's income was generally below the national median. This report, however, indicated a wide range of 17 characteristics for purchasers of new mobile homes- In­ terestingly, the spatial variation of these characteris­ tics was not reported even in tabular form. The most comprehensive report of the social and economic characteristics of mobile home residents to date was a special publication of the 1970 Census of Housing (U.S. Bureau of the Census, 1973b). The sample was drawn from all mobile home households and thus provided a much broader base from which to draw conclusions than did pre­ vious surveys. Certain of the findings are important to the present investigation. Several items are particularly striking (Table 1). The vast majority of mobile homes were owner occupied (84.3 percent compared to 62.9 percent of all housing) and an even higher percentage were occupied by whites (97.7 percent compared to 89.2 percent nationally). Most mobile homes were classified as rural (60.9 percent) and were located outside Standard Metropolitan Statistical Areas (33 percent). The rural, non-metropolitan location was reported in studies based on the more limited information contained in the 1960 Census of Housing. For example, it was found that mobile homes, whether in a park or on an individual lot, were inversely related to population densities within 3 National data used for comparative purposes in this section were derived from the 1970 Census of Popula­ tion (U.S. Bureau of the Census, 1972e) and Census of Housing (U.S. Bureau of the Census, 1972f). 18 Table 1 Mobile Home Household Characteristics: 1970 Gene ra1 Owner occupied units In SMSA's Rural Urban Percent with Black Head Total mobile home units of Household 84.507* 45.01 60.92 39.08 2.35 2,073,994 Family Income Below $4,000 $4,000-$9,999 $10,000-24,999 over $25,000 Median family income Owner occupied Renter occupied Age, Head of Household Under 2 5 25-34 35-64 65 and over Occupation. Head of Household Professional, technical & kindred Managers and Administrative Clerical and Sales Workers Craftsmen & kindred Other Blue Collar Farm Workers Service, including private household Occupation not listed Source: U.S. Bureau of the Census, 1973^ 18.7 77. 52.71 27.69 0.83 $7,800 5,800 18.117* 24.33 40.35 17.21 6.307* 4.99 8.42 17.97 22.56 2.71 6.17 30.88 19 SMSA's (French and Hadden, 1963: 133-134). They were most important in metropolitan counties not containing the cen­ tral city and were T,in fact, a kind of suburbia" (French and Hadden, 1965: 133). The HUD survey mentioned previ­ ously concluded that new mobile homes typically were located in parks and: . . . contribute significantly to housing sup­ ply outside of the central city and particularly in smaller communities outside of Standard Metro­ politan Statistical Areas (U.S. Department of Housing and Urban Development, 1968: 67-68). The special report on mobile homes (U.S. Bureau of the Census, 1973b) substantiates previous surveys in other respects as well. For example, the median income of owner occupied mobile homes was almost $1,800 below that of the national norm. Further, while all Income categories were represented, almost three quarters (71.5 percent) had an income under $10,000 in 1969. Nationally, 52.7 percent of all families were in that category. Compared to national norms, two age groups were more heavily represented in mobile home households. There were 17.2 percent of household heads in the age 65 and over category in mobile homes. The largest single age group occupying mobile homes were under age 35 with 41.4 percent in this category. While all occupation categories were represented, "craftsmen and kindred workers" and "other blue collar workers" comprised two-fifths (40.5 percent) of the occu­ pations of heads of households. This supports the findings of previous reports . 2 0 The Census Report tabulated data from a sample of all mobile homes with no distinction between units located in parks or on individual lots. As mentioned previously, this has been typical of most articles, reports, and sur­ veys . Such information is important, but it also raises certain questions. How do these characteristics vary within the United States? Are the factors associated with mobile home location the same as those associated with the location of mobile home developments? It has been assumed that growth in population and formation of households are important reasons for mobile home sales (Close and Birgell, 1972: 5; Goldblatt and Pitcher, 1972: 7). Particularly important are concen­ trations of younger families with lower incomes. These reasons at least partially help to account for the rapid increase of mobile home sales in the Southeast (Wheeler, Callo, and Brewer, 1973: 95) and the West. Not only has the South led the nation in sales of mobile homes, but it also was the leading region in total numbers of homes in 1970 (Powell, 1975: 28). However, mobile home developments are relatively unimportant in that region. 21 Mobile Home Developments In the United States Data Source No public data sources are available concerning developments for the country as a whole. The most complete source presently available on state-bystate park charac­ teristics is Woodall’s Mobile Home and Park Guide pub­ lished annually by the Woodall Publishing Company of Chicago. Not all parks are listed in this directory. Of the some 24,000 parks (Knight, 1971: 216) In the United States, approximately 13,000 were listed and rated on the basis of quality in the 1972 edition (Woodall*s. 1971). It can be assumed, however, that the vast majority of larger and better quality parks were included. Since all parks were not listed, rankings are used instead of actual numbers in the following discussion. Distribution of Developments Developments, based on numbers of spaces, were concentrated in two regional clusters of states (Figure 5). In the West, California alone contained some 19 per­ cent of the country's more than 1,100,000 spaces listed (Woodall's. 1971) . The four states adjacent to California combined had fewer spaces than in California, but had greater numbers than could be expected on the basis of population. For example, Arizona, Washington, and Oregon ranked fifth, tenth, and twelfth respectively In numbers of spaces and Nevada ranked twentyfirst. In total, these M k- t- M M I I I I t iiiiiiiiiiv iiaananiff ■■■aai , , t h h , ...BH--— « MHk“ t> • M t M i a a i a . , , , |t «««• •••a ||ft| < •••■■ aaaat .... .■aaaaa^H^H^ff I I H I I I xh h ^ I I '•V -■ ■■ ----- hbh»-b«hH||||(i« ^ b|| LJ aaaaaa ---- - I I ' ! ono ft rtHU ti h r , taataatt«taia«xx ■■■■■aataaatttt l i n n ) i • I I I * * .llu ■■III* I" ' lf)( ■ H'jl, M I i«■■ aa> «« K> M » i • ■1 ■at 17-24 25-3? .3- 40 41-46 Figure 5. Number of mobile home development spaces by state, 1970 (Source: computed from data contained in Woodall^. 1971), 23 five Western states contained some 29 percent of the de­ velopment spaces in the United States. A second cluster was apparent in the Great Lakes area. Michigan, Ohio, and Illinois had the greatest num­ bers but when adjacent Midwestern and Middle-Atlantic states were included, the ten states from Minnesota and Missouri to New York and Pennsylvania had approximately 29 percent of the nation's developments. Separated from the two main clusters of states, Florida and Texas ranked second and sixth respectively in developments. The concentration (16 percent) in Florida was particularly striking because of the low numbers found in the remainder of the Southeast. Of the Southeastern states, Georgia had the greatest numbers after Florida, yet ranked fifteenth in the nation. Most of the Southern states (Census definition) had fewer than 10,000 spaces listed. The number of spaces was, as expected, associated with the number of mobile homes In each of the states. The primary exceptions were Southern states where develop­ ments were less numerous than could be expected on the basis of mobile homes. Relative Importance In many respects the relative importance of de­ velopments in the various states is of as much signifi­ cance as total numbers. A simple yet useful way to 24 measure the relative importance of mobile home parks is a ratio of the number of development spaces to the total number of mobile homes in each state. In order to more easily compare the reLative importance among the states, a development quotient was computed. This was obtained by dividing the ratio of spaces to mobile homes for each state by the ratio for the United States which was .53. The potential value of the development quotient for any state would range from 0 to infinity. The actual develop­ ment quotients ranged from a low of .25 for West Virginia to California's 2.00. Development quotients tend to be high in tradi­ tional retirement states such as California, Arizona, and Florida (Figure 6). Yet the values are also high in the North Central states and in southern New England where this retirement aspect would be less important. The low quotients in the South and northern New England where p e r ­ centage increases in mobile homes have been highest suggest an inverse association between mobile home growth rates and developments. Additionally, most of the states with higher than expected quotients have a greater percentage of their mobile homes in metropolitan areas than the national a v e r ­ age. It also could be expected that the income of mobile home households would be highest in these same areas. In order to assess the strength of association, Spearman's Rank Order Correlation Coefficient (rs ) was computed between each of the factors and the development 66 66— 9 0 0 0 0 0------------------- ooooooooouoo 0— 8888 9 0 0 0 0 6 0 9 0 0 0 0— ............... — 00 — 0 0 0 9 6 0 0------------------- oooooooonoooo— 8 9 6 9 9 9 0 0— ----------- ---- ouoooooooooooo— eee 6— 0 60000— 0 9 6 9 9 6 0 0 0 0------------------ ooooooouoooooo— 99600690000--------- — -OOOOOOOOOOQOOO— — at— m b — bibb— • 800— H K I IW O II 00— N M i t i M « N M B « m 99— 00699989— OOOOOU------ 9— 83380— 0000000............ ...... MWHHMM88M8 OOOOCOOOOO......................... — — 80888 80000 000 0000.................... . . . .. 8— 00666 8889806 86 88— OOOOOOOCOno,,..... .. 00000000000000.— 88668008000666 0660— OOOCOi' 060— O000' ir> + + 808 866000000000 00000000000000008888866— 0000 00** 00609— — 88688988— 8668608— 80606666HA066680OOOOOOOO00. — i— — OOOOOOOOOOOOOOOO— OOOQOOOOOQOOOOO— <66666666608— 8888....... . — .............. .— •— 6668866669— 88 • 888— H i — 66— B88I88B8600— ■••••f— 88888— 96668888— 600000000000000+♦+♦♦♦♦+ ♦♦♦++♦♦♦♦*. ♦ ♦+ 000001' + ♦ 60000000000000 + + + + + + + + + + ♦ *♦+++•+♦.+++O0Q00+* 00OOOOOO9OOOQOO + + + + + +++ + + + ♦♦ + ♦♦♦ + ♦ ♦♦♦♦♦OOOiJ** 8060— — 800000000000000 + + + + + + + + + + + + + + + + + + + + + ♦♦♦♦0. . ■ , — 9000— — 06090— 888888880660— 8 ♦♦ ♦ ♦♦♦♦♦♦♦♦OOO— GOOOOO+ + + + + * * * * * .... 888 88— 8888— — 188 — ill— — 88888 — 8818 ♦♦♦0000 000+ ♦♦♦ooooooooo* ♦ ♦♦OOuvJOOOO.+ ♦♦♦ooooooo... ♦♦♦...000...* ^ * t • i t < ■ • I fe ■ — 888680— 88— 8160888— — 8— 818888— 888 8— — 888888—— 88 881 8— 8— 1888881 1888— 1 — 8881 a 4 + 4 4 4 4 4 4 4 4 4 ♦ * + ♦ * 0 0 0 0 0 0 0 0 0 0 0 — 4 4 + 4 4 4 + 4 4 4 4 + * + + +OO00OO0GOOO— 4 4 + 4 + 4 4 4 4 4 + ♦ + ♦ ♦ 400000000 0 0 0 — 4 t + + 4 4 + + 4 + 4 4 + + + 4 4 4 4 + 000000— ♦ ♦ ♦ + ♦ + 4 4 + 4 + + 4 4 + + + 4 4 4 + 4 4 4 4 4 + 4+ + + 4 4 4 4 4 4 4 4 ♦ + ♦ + + + +*.*♦* 4 + O r ---- + 6fl— 6<- ♦0606660 ♦6006681 ►+++84+§|0 ► ♦♦+♦♦♦— ►♦♦♦•♦♦88 - - - ♦ * ♦ > --- R A N K — 8 — a — 6 WP 001 00< < 8 9 16 17 -24 2 5 - 3? 33-40 41-48 ro Ln Figure 6. Development quotient by state, 1970 (Source of data: Woodall*s» 1971 and U.S. Bureau of the Census, 1973b). 26 quotient. For a basis of comparison the coefficient was also computed between each of these variables and the mobile home quotient.^ Younger households make up the largest age group purchasing mobile homes. However, in terms of relative development importance, the association tends to be with older residents (Figure 7). The correlation coefficient between the development quotient and the percent of mobile home heads of households age sixty-five and over was moderately high (rs~.60) (Table 2). On the other hand, there was a slight negative association between older families and the mobile home quotient (rs“-.24). The highest coefficient computed was between the development quotient and percent of mobile homes in SMSA's (Figure 8). This positive association (rs*.62) supports the contention that the relative importance of mobile homes and mobile home developments are not one and the same. Mobile homes are generally rural and nonmetro- politan; developments tend to be metropolitan. Mobile home families generally have lower incomes than families as a whole; however, the higher the Income of these families the higher the development quotient (rs-.50). The greatest departures from this association were for Wyoming where parks were relatively unimportant ^The mobile home quotient was obtained by dividing the percent of mobile homes to year-round housing for each state by the 3.1 percent national figure. »■ -- ------------ -- - - - — -- • - ~ -- - - - - - - { 4 m *4I-I - - ------------------------- -- lllllllll »IIHIVIII1 • iiiit»i i « t ^t i i i » « i* B v a itatiatiaaa+aaattataaai aaatavitiaaaBaataaaafla ♦♦♦♦♦♦ 11 * * * >1' Vi i ■ i. -. i K U ,■ ■ ijOUU " ' m. :: i ■ .y'niHnrtHHt* I H ■■' .■ !. I M» >1 ' • ■ UK II i 1L11 H J: 'UOU! ■ ■ ■ MLJll ",nn." i ► ninii '■.KIOUU. '.■ ■ lO'JU- + ■ ■0 JOOl)'..,OtJ»9»^H(t»»eOOOOOO+ ♦ * * + *t ^ L - j o o y u O s ^ ^ ^ s e e o o o u O ' ) ! ) * * * - ) u o o o a o o o u u n u o y c ‘ 0 0 o c i ' 0 0 ; i t + ♦ * - h , h h h k h * j + + . . * . --------------------------------t)oor'0000wc,'joo+* + * *i t + --------- ()G.)U')UOOGtJL|,jn + + tf'f «4*' ' ,. + + »»+--------UOC-OuOOOOfJ'.ulD * ♦ ♦-----* • • + + ii* -------- GOGOuOOOOi'u-------- . . .***--------- Do'uioooooy--’— ----- . ------------------------ G tn iU iJO O rm n --------------------------------. . . ------------ 1)O0G')()»() i --------------------------------------- 4 M.ijt.Hj ; H>iJ • 1 ' -M, || jijiiii..' -it i---- ii * * , ■ _ . . i'­ ll - - mi •li n n M fH H >-* 11 iH H + ♦ * + * + * + * * ------------------------ O ' l i l 'l ' J u O O --------- +++++..++++++++------- hj.,■, i )Jo**'*■*- ----- 4 ♦ 4- ♦ ♦ * 4 *■ t*4++***4444+41 , . I , .-- taaaaa *■+ ♦ + 44+*- + +4***4f* •Mill. .+++++ ♦ + + + + + + Hill *■■■ •III Mill ■ II ■I •II III n h m ■ - e 9-16 17-24 25-32 7 7 _ 4 r 4 1 - 4 8 Figure 7. Percent heads of mobile home households age 65 and over by state, 1970 (Source of data: U.S. Bureau of the Census, 1973b). 28 Table 2 Development and Mobile Home Location Quotient and Selected Variables Rank Order Correlation Coefficient Variable Development Quotient Mobile Home Quotient Percent head of household age 65 and over Percent mobile homes in SMSA's Median family income Percent increase, mobile homes 1960-1970 .60 .62 .50 - .53 -.24 i itV - .38 .28 Source: Computed from data in Woodall * a . 1971 and U.S. Bureau of the Census , 1972b. *, ► f * M U H k • a t n i a t a t i t i t t t t t t - i " ■••••■a a 1 1 ■ ■ I M l l l t t H • M l * i t l f i l i l i - J - i i - M - u u t J U U i ' i (itatttt-1- i i * * 1 *>-* * * * * * > * * ' i n " I I 1 1 >1 ... ...................................................... * * * * * *• J»f i I V * ■ I M l[ N I " •tit ■ I H I I I I I I I I I I I I H I I - • atttitaataattaaaiita - iiiiaaatatatttaatia •■••taittitttitia - •••■» it I' I , i h-\ > t<}< K f , * lK\t ■• , ' JIJ t J ■ ■ , + . > ittr i-j'tf. >.n )*• ■ Mi>' ■ f|f H ■ pf>* ..\ |b4H***4*"4H 1 'it 'rJU't - " ’ 14.444 + , + * * + * * > * 4 * * + + 4 4 fwiOrn*nr)f)fi . J • • M M H III M . r«4MM>-iMP-f'-4rfpt^**’H -* ►* - • - - * * lf,f >f>----------- rf .-. -*«■ — -*■■ ~ *44 * *444 4* .**+*+» ■•■••• 111 ilJU IIIIIIIIHK^IIUI'')-----------*♦*,*-- ■ .--tit)'.. , ,■ I K i i t i i n n m i i u - - - - - - - - - *****. . . . . ■- -. H | * i ----- lkl'1 •," --- iiiiiiiiiiiiiHiii ******---******---- r i ••■■(■■■•••a i n i i i i H i i i *****- - - - ** + *,..--. . . . -- *************- - - - -- •■■■••aaaiaaaaaaa■■■•■••* * * *~— (***H>*H»+**tH ■ lllllltllMIHIIIIMI* ***** 4t*H*+w ■ ••• I M H I I I I H 1 ♦ *•* • IIIMIIIM I N H I M i l ■ MH**** w*+w* »****•+* ■at 3 - 5 1 7 - 2 4 2 5 - 5 2 5 5 - 4C 4 1 - 4 8 Figure g. Percent of mobile homes in SMSA's by state, 1970 (Source of data: U.S. Bureau of the Census, 1973b). 30 at one extreme and the retirement states of Florida and Arizona at the other (Figure 9). Developments tend to be more Important where per­ centage Increases in mobile homes were low (Figure 10). The negative association was moderately high (-.33) and suggests that areas may go through a sequence of events in regard to mobile housing. For example, as mobile homes become more numerous, household preference, the profit motive, or public regulation create a need for develop­ ments. If this holds true, the relative Importance of mobile home developments should increase in the South. This idea could be better substantiated if past develop­ ment data were available for all sections of the nation. The interstate pattern of mobile home developments adds to the understanding of general regional variation. However, using states as units of observation masks cer­ tain information because one or a very few metropolitan areas may contain the bulk of developments in any one state. Unfortunately, few studies of intrastate develop­ ment location from which comparisons can be drawn have been conducted. An exception was a Massachusetts study conducted in the roid-1960's (El Gammal, 1966). It was found that mobile home parks in Massachusetts were associated with population, metropolitan areas, main roads and highways, and special function areas--meaning military establish­ ments and universities (El Gammal, 1966: 74-75). The •■• i i i i i t i i i n > • • n n t -1 * 11 t + t *■ + + •IHH ‘Bbbbmmbbbibb >ataa**Baaaaa " i i i l i i m i i .. tf»MMe0eft»tte«e»g .. . M * i i a * # e w « H 49»«ft aaaa .,. t a * M a i i » M « H H a < 4»« ataata ... .. aa 'I I I I I I I I H I M M ' ■|>I*1H"1|('I I'll K 1 , . (jO 'iiirijn iK K K iiii ''HilHKWIIttHWUHMHHttWHHHHtiH**, . , ii< w '> ''.i:H )< )fi(iiiiifii!ifii'f|[)i> ()':M M H f4 B ta a a a i« ia g ii... <:i'' I"'.. .I-1 ............ ..................... . r m ■ r ■' nil: in i mu........""Ik mi ini i'.'i.j aaaaiaaaaiaia. ., 1 ^ - , i"i; i■ iw"iv■ 11 u1 ■ '■ hnnj'i!■ ■ ■■ ''Miaiaaiaiaaaaaa iixj mi i.mini ■ iiDiinaaiaiiiiiKii'MiiM f>«*t>4Hf)«4»egBagaaMtaB* * ****8 B aB 8 aB tB iB B a o[> < m uAOBptiBBuaaBaaaai 01MI1M1MI+ + ** * * + f ♦ * * a * * ♦ * *. * 4 + * * ♦ * ♦ eaaaa aaaaaaa bbbbbbb 'in'fjuou(j‘j)iB«e«ei*«BBBMBaaaaa aaaaaaa 'iKJUooouuuQftwtaattfeSHmaiBgaiiaBButfett 'H)()(Mi)Hjii(i|MKioi)ij(jO''juLiHe»fl9«ett«aaBBBaaaBaaaBHBtiHM^^r,,,i,ulju "J'JDul I KJIJCii 'I n)(l(J0(10mj'l(.i' ******* 'I'juuummiH.- '■»)OiiOOt»(.|-;j(i:i(jo* + ** + "UUOM'UtjUU' 'OUOOOOKi'V.'OO* + ♦ * * Mi)(>ij(Kniiim)M"iooo(K)(.",i(,ooii*** + * — --- t 11 t t , it * * * * H U H * -------- - - - , 'util tMHHDlU--- *---- , , t + + * + tHH#ttwu#«e#e------ , uHBftBH----------- U H M H H *BHk#«HHB86flBBHHHM88 * *ni>‘ **0W ++nnr a *nt>* i. + + r ) >pi »*»**. < • « »•■ » • » **.*■*«'' 1 i"‘ : 1 1 m^knh4^H **** + * + *■*♦■*--- -- t| , ^ ^ .- -HHtiHMHHrf. .*»»,***..--------------------------------- -|t(| -U H H t j H H H r i * •****♦*. *-t- h-------------------------------lttlltt-1--' -• -----------------................................... |||||‘« “‘1- H h - . ------ ------- ( # h, ^RHM^AfmiiiitlitMMllllllHllllillflllllH ... ^H^H^HH^HItlMMIIIMItlMMtailMlllttMtlf m b . . . , , ',-■■■ 4fmHHHHHI4t-t4||IM||tt||MlllltltltMllt«M lift '''('OUltH^HtlBtlltlldlltMMMIMMMMMtM ' -r.iooo'OijO'.toodiioittitntdimtttdidudd ' ''.'''''(.)000(IO[jn0u0M«4HHH4t»d*lt»||tt|ldl|| " 'ni;uuriJi:){>OOnooi)t4ftH«rt|*tttd»*»dllHdll . ic(■■'!!■ i 'ijix.irioUOOD(H)OnOuU(Hl4flH*tMtMMut- i i' l l l l t n .............HUHJUOOOOOOOOOoa&WWftW '>"■'> imonooimonno 8«**h ihi.i.j.F 1.1 : ■ i (KjO(ii)O(JUftn uoouun O0l)l> (juo o 'i , i, r , . . ■ ■ ' UHH „M„ 3-iS 17-24 ttt ... 2 b-32 __ 3 3 - 4 0 ;;; 41 - a & u> N3 Figure 10. Percent change in mobile homes 1960-1970 by state (Source of data: U.S. Bureau of the Census, 1962b; 1972b). study also Indicated the metropolitan location of develop­ ments contrasting to the rural location of mobile homes. 3 3 Mobile Home Developments in Michigan Distribution Unlike Massachusetts which had only 10,928 mobile homes in 1970 (U.S. Bureau of the Census, 1972b), Michigan has ranked consistently high in several measures of mobile housing. In 1970, the state ranked ninth in number of mobile homes with 73,012 units (U.S. Bureau of the Census, 1972b). This was only a slight change from its eighth ranking in 1960, and sixth in 1930 (rankings derived from the 1950 and 1960 Census of Housing) . Of more direct con­ cern, however, are developments. Michigan ranked tenth in number of parks, third in number of spaces (rankings derived from Woodall’s . 1971), and sixth in relative im­ portance of developments. In January, 1975, there were over 950 developments with some 87,000 spaces licensed in Michigan (Michigan Department of Health, 1975).^ Developments were highly concentrated in the southern portion of the state, par­ ticularly in the seventeen counties comprising the state's SMSA's (Figure 11). These metropolitan counties contained approximately 72 percent of Michigan's total developments **Data for this section were compiled from the files of the Michigan Department of Health, Lansing, Michigan. This agency licenses all parks in the state. 3 4 i . j... M I C H I G A N S PACES • 100 5 0 0 • A 1000 5 0 0 0 * 0,000 I f . V • I s Figure It. Humber of mobile home development spaces In Michigan by county, 1975 (Source of data: Michigan Department of Health;. 35 (spaces). This is similar to the distribution of popula­ tion, i.e., slightly less than the 77 percent of the state's Inhabitants were In metropolitan counties in 1970 (U.S. Bureau of the Census, 1971b). Unlike mobile homes in general which were 51 per­ cent metropolitan in 1970 (U.S. Bureau of the Census, 1973b), developments were concentrated in SMSA's in 1960 as well. Using the same seventeen counties,** 78 percent of the state's 21,900 spaces were metropolitan in 1960. The changes that have occurred between 1960-1975 were essentially Increases in numbers (Table 3). For ex­ ample, there was an increase of over 65,000 spaces and 550 developments. The average size of developments increased from 54 spaces in 1960 to 91 in 1975 and only three coun­ ties contained no developments in 1975 compared to nine­ teen in 1960. Relative Importance It was generally in metropolitan counties that mobile home developments were of greatest relative impor­ tance in Michigan (Figure 12). Three counties (Berrien, St. Clair, and R o s c o q d ot i ) in which the development quo­ tient exceeded one were not in SMSA's. Roscommon County contains few mobile homes or development spaces and the high quotient can partly be accounted for by its Importance ^There were fourteen counties in SMSA's in 1960. The three counties not included in 1960 were Lapeer (Flint S M S A ), Monroe (Toledo), and Ottawa (Grand Rapids). Summary of Michigan Development Characteristics, 1960-1975 Table 3 No. of Developments No. of Spaces Avg. Size of Devpt. X Spaces in Met. Counties 407 511 747 956 21,877 28,716 56,071 87,259 54 56 75 91 78 73 72 72 Year 1960 1965 1970 1975 Increase over prev. five years total % - 6,839 31.3 27,355 95.3 31,188 55.6 Source: Michigan Department of Health *1970 Definition 37 M I C H I G A N d e v e l o p m e n t q u o t i e n t Figure 12. Development quotient in Michigan by county, 19 70 (Source: computed from data obtained from Michigan Department of Health and U.S. Bureau of the Census, 1972c). 38 as a recreation area. St. Clair and Berrien both have comparatively large numbers of developments and are ad­ jacent or nearby SMSA's. Counties with low quotients are, with few exceptions, in the northern portions of the state where mobile homes on individual lots predominate. Local Studies Local studies are more numerous, but these too have limited comparative value. Some studies have looked at all mobile homes, others only at developments. There has also been a wide range of methodology employed. The most numerous local studies are those pre­ pared by planning agencies. Some local agencies have prepared reports dealing exclusively with mobile homes (Maricopa County Planning Commission, 1963; Detroit Region Transportation and Land Use Study, 1969; and Macomb County Regional Planning Commission, 1969). How­ ever , most have devoted a small section to mobile housing in general housing reports (examples Include: Tri-County Regional Planning Commission, 1971; Genesee Metropolitan Planning Commission, 1968: 40-50). The concern of these reports has been a description of local features, fore­ casts of future demand, and regulation of mobile homes and parks within their specific jurisdiction. Studies of mobile home developments within metro­ politan areas have tended to emphasize zoning as the roost important factor influencing development location but other 39 factors have also been noted. Included have been employ­ ment opportunities particularly In "blue collar" occupa­ tions (Field, 1970: 24), proximity to highways (El Gammal, 1966: 72-73; Freed, 1969: 202), and land availability in fringe locations (El Gammal, 1966: 113; Freed, 1969: 202; Field, 1970: 24). However, zoning restrictions together with the manner In which they are applied is usually con­ sidered the most important single factor influencing de­ velopment location (El Gammal, 1966; Field, 1970). It appears that the location of the mobile home parks within a particular community has been more the result of chance and the presence or absence of public regulation than of choice (Wehrly, 1972: 11). Zoning restrictions clearly eliminate the mobile home alternative from the spectrum of choice for many urban and suburban residents . . . . (Knight, 1971: 216). One of the main reasons why most mobile home parks have been located in the less desirable residential areas is the reluctance of zoning and planning officials to permit this type of land use in residential areas reserved for con­ ventionally built housing (Goldblatt and Pitcher, 1972: 7). There is no doubt that mobile home parks are the most highly restricted residential land use. Industry sources state that obtaining favorable zoning rulings is a major problem in park construction (Wehrly, 1972: 29) and restrictive zoning practices are common (Freed, 1969: 150-155). But municipal zoning ordinances are only one of many factors which influence park location. There are two reasons for this. First, zoning 40 ordinances are not final. Restrictions against mobile home parks can be and have been changed (Goldblatt and Pitcher, 1972: 7). Even though public officials often hold mobile home developments in low esteem (El Ganroal, 1966: 127), the absolute prohibition of developments is decreasing (Business W e e k , 1970: 74; Drury, 1972: 137-138). M u ni c i ­ pal ordinances are guidelines which affect the location of roost land uses. They are not, however, the only factor. The second reason for believing municipal decisions have been overemphasized is perhaps more pertinent to the problem at hand. The m o d e m mobile home park is a size­ able investment often amounting to hundreds of thousands of dollars.^ With an investment of this size, it is u n ­ likely that the developer would jeopardize an investment by locating at submarginal locations merely because zoning allowances were easily obtained (Jacobson, 1963: 348). Q Mobile home parks are normally located at the 7 Wehrly (1972) found from a survey of mobile home developers that the cost of developing a park (ready for occupancy) ranged from approximately $3,000 per site plus land costs in the South to almost $4,000 per space plus land costs on the West Coast (Wehrly, 1972: 36). The average total cost for FHA insured developments in 1971 was $3,600 per space (U.S. Department of Housing and Urban Development, 19 72). O °This would not necessarily hold true for smaller parks such as those constructed behind the owner's house; a type referred to by one writer as "parks in the pasture" (Wehrly, 1972: 19). The trend, however, has been toward larger and more expensive developments. For example, FHA insured developments in 19 71 averaged 186 spaces (U.S. Department of Housing and Urban Development, 1972: 254). 41 outskirts of metropolitan areas or even in rural portions where population densities and land values are lower (Knight, 1971: 216). Perhaps it follows that zoning re­ strictions are absent or more liberal in these areas as well. But it should be noted that mobile home develop­ ments are important and numerous in areas such as the M i d ­ west and West Coast where zoning restrictions are strin­ gent . Even though parks are a moderate density residen­ tial use, the m o d e m park requires a considerable amount of land (Freed, 1969: 202), coononly exceeding twenty-five acres (Knight, 1971: 216). As can be seen on the graph of housing density ranges (Figure 14), mobile home parks over­ lap density ranges of single family, duplexes, and town- house s . Unlike apartments which have density figures allowing them to compete almost anywhere in an urban area (Bourne, 19 71: 325), mobile home developments would be much more limited. Regardless of zoning ordinances or the way the restrictions are enforced, the peripheral location is a normal outcome. Within the context of the general peripheral loca­ tion, are parks located anywhere in the less developed portions of metropolitan areas? Do parks display "no particular pattern of location" as was noted in one report (Macomb County Planning Commission, 1968: 11)? Are zoning restrictions the important factor that has so often been assumed? Or are there a large number of site and area HIGH DENSITY MOOE RA T E D E N S I T Y LOW DENSI TY 8 10 14 _L_ U N I T S / A C R E S I N G L E FAMtLV DUPLEX MOBILE HOME T O W N H O U S E A P A R T M E N T F ig u r e 13. Density ran g e s by type o f residential development (Source: adapted from Genesee County Metropolitan Planning Commission, 1971; Figure III-l). 43 factors that Influence development location? Are differ­ ent types of parks influenced In different ways by these site and area factors? Has the location of parks changed through time? These are representative questions that have been inadequately addressed in previous studies. An Approach to the Investigation Most studies relevant to mobile home development have viewed public regulation, or lack thereof, as the most important factor of location. This is contrary to most research of other residential forms which have nor­ mally focused on the actions of the household consumer (Kaiser, 1968: 351). In reality, as is true with virtually all activi­ ties, a large number of factors would be necessary to account for the location of mobile home developments. Kaiser (1966) found that the spatial distribution of single family developments was associated with the spa­ tial distribution of what he termed site characteristics within the context of the type of developer, and price range of the home. Three classifications of site char­ acteristics were devised in Kaiser's study. These were termed: 1) physical, 2) locational, and 3) institutional (Kaiser, 1966: 63-64). 1. Physical Characteristics a. Size of the tract of raw land b . Topography c . Soil conditions d . Ground cover 2, Locational Characteristics 44 a. Social characteristics of location b. Proximity to transportation arterials c. Accessibility to schools d. Accessibility to shopping e. Accessibility to employment centers f. Proximity to existing development g. Visual quality of the approach route to the site h. Proximity to incompatible uses 3. Institutional Characteristics a. Governmentally imposed boundaries and con­ tent of regulations, services, and poli­ cies for; 1) Water and sewer service 2) Zoning regulation 3) Subdivision regulation 4) School districts b. Ownership patterns including: 1) Size of individual parcels under separate ownership 2) Whether or not the raw land parcel is on the market and 3) The terms of availability (terms of payment price) c. Marketability rating by financial in­ stitutions determining the availability and the terms of financing available to developer. Bourne (1968 and 1971), in a Toronto study, enumer­ ated six major kinds of area and site factors expected to influence apartment location. These factors included mea­ sures of: 1) the existing housing inventory, 2) accessi­ bility, 3) clusters of existing apartment developments, 4) the physical and social environment of the immediate area, 5) the cost and availability of land, and 6) avail­ ability of municipal services (Bourne, 1971: 328). 4 5 The factors influencing the location of single family subdivisions and apartment development are not un­ like many of those found in a recent extensive publication on mobile home and recreational vehicle park management. In the report, the authors suggest the following items as being significant in the site location of mobile home de­ velopment (Nulsen and Nulsen, 1971: 259-260): 1) Cost of land 2) Size of parcel 3) Terrain 4) Soil condition (if septic system is used) 5) Likelihood of flooding 6) Accessibility (to major streets and highways) 7) Nearness to shopping facilities, employment, and schools 8) Availability of public services (water, sewer, gas, phone, etc.) 9) Favorable zoning 10) Nearness to public transportation This similarity is not surprising for: There is usually little disagreement among the various segments of the land development industry regarding the major items (aside from land cost) which should be considered in selecting a piece of land for any given use (Wehrly, 1972: 37). The results of a survey of selected mobile home park owners and managers,^ conducted by the Urban Land Institute regarding important site location factors can be summarized in the following list (Wehrly, 1972: 37): 1) Location ^Only parks with Woodall's rating of "four" and "five" stars were included in the survey. These are the highest quality parks listed and rated in Woodall's Mobile Home and Park Guide. Of the over 12,840 parks listed in the 1972 edition (Alaska - 65 parks and Hawaii - 0 were not counted), slightly more than 1,100 were in this cate­ gory (Woodall*s, 1971). 2) Public utilities and services 3) Access 4) Terrain 3) Surrounding land uses 6) Amenities These are the kinds of site and area characteris­ tics found to be important to the location of single family subdivisions and apartment development and suggested con­ siderations for mobile home parks. From this literature base, a set of site and area factors thought to be impor­ tant in the location of mobile home developments were enumerated. Because of the nature of mobile home parks, certain modifications to this foundation were necessary. These modifications, the rationale for the changes, and the list of factors used in the study area will be pre­ sented in Chapter Three. CHAPTER THREE RESEARCH DESIGN Introduction Evidence Indicates that mobile home parks are in­ creasing In importance as a form of residential develop­ ment particularly in metropolitan areas. If it is assumed that this will continue, then it would be desirable to be able to anticipate where within these areas parks are likely to occur. The first step is a better understanding of the present spatial distribution. Evidence also Indicates that mobile home parks are evolving. Changes, such as larger size, decreased densi­ ties, and more amenities within recently constructed parks, have been documented (Newcomb, 1971; Wehrly, 19 72). What has not been fully documented but can be assumed is that location is also evolving. Determining the degree of change, if any, should aid the understanding of mobile home developments. While recently constructed mobile home develupmtnt s resemble more the single family subdivision of the 19^0'a than those of the 19 7 0's (Wehrly, 19 72: 12), it is assumed that they have evolved to where simllar sets of site a n d area factors would influence the location of both types 47 48 of residential development. The set of site and area fac­ tors found to Influence the location of single family development and those thought to Influence the location of apartment construction discussed in Chapter Two form the base for those used in this Investigation. However, mobile home parks have features unlike other forms of residential development and certain modifications to these sets are necessary. The reasons for these changes are presented below. There Is some question as to just where mobile homes fit in the nation's housing supply. They are "de­ tached" and are almost always owner occupied. Therefore, they could be considered a form of single-family dwe11- i n g ^ and it is in this manner that they are usually portrayed (Bair, 1965; El Gammal, 1966; Cheek, 1967; Drury, 1972; Freed, 1969; and Newcomb, 1971). However, the majority of all new homes are located In parks (U.S. Department of Housing and Urban Development, 1968: 76, 111) . In a sense, a mobile home park can be considered a 'horizontal apartment complex" (Bartley and Bair, 1960: 13-14. See also Scheff and Bleck, 1973: 15). In a park ^ T h e Department of Commerce does not include mobile homes when compiling new housing starts (in Con­ struction Review and Cons truet ion Reports). InsteaT] they are Listed separately using figure s supp1ied by the Mobile Homes Manufacturers Association. An associated problem Is that the government reports only national sales figures, not state, county, or city breakdowns. State sales data can be obtained in certain trade publleations. 49 setting, they represent a moderate density housing form; usually seven to ten spaces (homes) per acre. Also, mobile home park lots are normally rented, not sold, with the home Itself usually purchased Independently of the site.** A distinct possibility is that mobile housing represents a compromise between the traditional importance of home ownership and the smaller Initial outlay for mul­ tiple family housing. There is validity in considering mobile homes as single family housing when the dwelling is located on an Individual lot. However, when consider­ ing mobile housing within a park setting, it is the de­ velopment, not the unit, which becomes Important. Unlike conventional subdivisions, the park de­ veloper must construct and maintain utilities, streets, recreational facilities, and other features of the park. There is a continuing relationship between the household and management similar to multi-family developments. But unlike apartments, the housing unit is not part of this re 1stionshrp. Thus it is possible that the locational characteristics of mobile home parks, while similar in many respects to other housing alternatives, will have ce rta ir. uni que fea ture s as we 11 . - ■'"There £ re motile home subdivisions where the Jot is Stic instead cf rented, but these are not common . Of the approximateiy 1J,000 perks listed in Woodall's Mobile H o y and Park Guide (1V71), only 217 parka selI Tots. Fifty-nine cT thest art ir. Arizona with «noti»er fifty- one ir Florida. Even fewer parks, most of which art 1ocatec tr Florida, sell the mobile home and lot as a p a c k a g e . 50 Improvements in mobile home parks have lagged b e ­ hind certain improvements in the production of the mobile home itself. For example, the introduction of ten foot wide units in the mid-1950's and even larger units during the 1960's and 1970's made obsolete many existing parks. The sites in these parks, designed for the often transient homes no wider than eight feet, were too small. However, the general shortage of mobile home parks meant that crowd ing was a common result of locating the larger homes in parks ill-equipped to serve them. This lag between pro­ duction advancements and improvements in parks has been one of the more serious problems of mobile housing. One way in which this lag has affected mobile home parks is that municipalities were caught unprepared for contempo­ rary mobile housing (Knight, 1971: 218). Ordinances de­ signed for transitory trailers are often not appropriate for m o d e m manufactured (mobile) housing, but change is often slow. 1 2 Determination of Site and Area Factors Literature Base Research has shown that three major categories of 12 In a recent publication by the Urban Land Insti­ tute, Newcomb (1971) lists several examples of inappropri­ ate regulations. These include: the requirement of some municipalities that parks have comnunal restroom, bathing, laundry, and even cooking facilities; confining parks to commerical districts ; and establishing time limits on t tie length of stay in the municipality (Newcomb, 1971: 37-38). As the author states, these may be appropriate for recrea­ tional vehicle (travel trailer) parks, but not for m ob i le home developments (Newcomb, 1971: 37). site and area factors influence the location of single family subdivisions (Kaiser and Weiss, 1969). These are: 51 1) Accessibility: a) to downtown b) to employment c) to an elementary school d) to the major street system 2) Environment of the immediate area: a) socioeconomic prestige level b) amount of contiguous residential develop­ ment 3) Institutional factors: a) zoning protection b) availability of public water and sewerage (Kaiser and Weiss, 1969: 76-77). Modificat ion The principal modifications of this set involve the inclusion of additional factors. Mobile home develop­ ments require relatively large parcels of land and the availability of undeveloped land should be an important factor. Additional items should be included with the list of environmental factors because of the supposed negative locations of parks in general. Also, due to greater public restrictions on mobile home parks, additional zoning m e a ­ sures were included. The modified set of site and area factors include: 1) Accessibility: a) to the urban center b) to employment c) to major highways 2) Availability of undeveloped land 3) Environment of the immediate area: a) Surrounding land use: 1) residential development 52 2) nonresidentlal development b) Adjacent land use c; Proximity to an elementary school d) Socioeconomic prestige level e) Number of Incompatible uses A) Institutional factors: a) zoning 1) surrounding residential zoning 2) adjacent zoning 3) zoning of the site b) type of sewer system c) type of water system This set of site and area factors thought to in­ fluence the location of mobile home parks is based on two considerations: 1) the assumed similarities between m o d e m mobile home parks and other forms of residential development, and 2) the responses of mobile home develop­ ers to a national survey conducted by the Urban Land In­ stitute (Wehrly, 1972). It is not the purpose of this research to conduct an additional survey of park owners and managers. Instead, these factors were enumerated to specifically answer the following questions: 1) What combinations of site and area factors influence the location of mobile home de­ velopments? 2) What are the similarities between site and area factors influencing other forms of residential developments and mobile home parks ? 3) What changes in the combinations of site and area factors have occurred between parks opened during different construction periods? In the remaining sections of this chapter, the factors are operationalized, the sources of data are given, and the procedure of data analysis is described. 53 Description of Factors Used in Study Accessibility Since they (mobile home parks) are a residen­ tial use, they should follow the general criteria of other residential land uses and consider the park site's location in relation to employment, schools, commercial activities, recreation facili­ ties and the transportation network (Macomb County Planning Commission, 1969: 11). The developers, desiring to quickly obtain tenants and to keep the park as fully occupied as possible, will attempt to locate the mobile home development as close as possible to those factors thought to be important. Mobile home parks are usually constructed in the less built up portions of metropolitan a r e a s . Within the context of this generalization, however, it is expected that differ­ ences exist in the combinations of these factors influ­ encing parks opened during different construction p e r i o d s . Distance to central business district. Distance was measured along the most direct major street from the park entrance to the center of the central business dis­ trict of Flint. This point was defined as the intersec­ tion of Saginaw and Detroit streets (U.S. Bureau of the Census, 1967) and measured in tenths of a mile. It Is expected that more recently constructed parks will be located at greater distances from the central business district than older ones. Distance to employment c e n t e r . The major source of employment in the Flint SMSA is manufacturing. Of the 54 area's 186,616 person non-farm labor force In 1970, 45.8 percent were employed in manufacturing (U.S. Bureau of the Census, 1973a: 560). Thus, for Flint, it can be assumed that access to manufacturing employment would be the most important. The center of manufacturing employment was ob­ tained by plotting all manufacturers employing 500 or more workers (Consumers Power Company, 1965), and using the computer routine CENTRO 13 to obtain the mean geographic center. The measurement was in tenths of miles along the most direct route from the development entrance to the mean geographic center of manufacturing employment. It is expected that newer parks are located at greater dis­ tances from the center of employment than those constructed during the early period. Distance to nearest state or federal highway. This variable was measured in tenths of a mile from the develop­ ment boundary to the nearest state or federal highway. Since recently constructed parks serve a more permanent residential function than older ones, it is expected that newer parks are located at greater distances from highways. Availability of Undeveloped Land The variable used to measure availability of land ^ T h e CENTRO routine was written by John F. Hult- quist and is on permanent file at the Southwest Missouri State University Computer Center. 55 was Che percent of undeveloped land in Che area surround­ ing the park. The data were obtained from 1957 and 1972 aerial photography. Because of the increased size, newer parks are expected to be located In areas of greater un­ developed land. Environment of the Impedlate Area Though important to developers of most land uses including mobile home parks, accessibility measures are insufficient to account for the location of mobile home developments. Of the numerous other factors, those in­ volving amenities or the attributes of the iomediate environment are becoming increasingly important. Today, residential development emphasizes a e s ­ thetic or site amenities, not access. The a u t o ­ mobile has made most suburban sites almost equal in access terms, people are free to look for other things like site qualities (Abler, Adams and Gould, 1971: 361). Often the setting in which parks are found is listed as a major problem to mobile home park acceptance by municipalities and the general public as well. "Much of the poor image of the mobile home parks in the past has revolved around their setting . . . ." (Wehrly, 1972: 19). As stated in a recent planning report: Some ccminunities feel, or have felt in the past, that the best locations for mobile homes are next to railroads or within Industrial or comnercial areas (Macomb County Planning Com­ mission, 1969: 11). Mobile homes are an important part of our housing supply, yet in terms of location, there has been a 56 tendency In the past for mobile home developments to be more similar to commercial and industrial development than residential. Often adjacent and surrounding land use would not have been considered desirable for other residential forms. The trend in recent years has been the location of parks within residential areas (Wehrly, 1972: 19), but this trend is slow. In a recent survey 74 percent of four and five star (highest rating) parks were " . . . partially or wholly surrounded . . . " by residential developments (Wehrly, 1972: 1 9 ) . ^ This quality rating includes a very small percentage of mobile home parks for the nation as a whole (10 percent) and for Michigan (7 percent) (Percent­ ages derived from listings in MoodaI1 1s . 1971). It is believed that as mobile home parks have become an increas­ ingly Important part of residential land use, the influ­ ence of the environment of the immediate area on the loca­ tion of parks has also increased in importance. Surrounding land u s e . Surrounding land use was divided into two categories and ascertained from air photographs taken in 1957 and 1972. The categories were: 1) Percent of residential development (ex­ cluding mobile home parks) in the immedi­ ate area. 2) Percent of nonresidentla1 land use in the immediate area. In that report, single-family housing was de­ fined as residential while apartments were included in commercial figures. It was reported that 50 percent of the parks were adjacent to single-family developments and 24 percent were adjacent to apartments (Wehrly, 1972: 19). 57 For all parka constructed prior to 1957, Soil Con­ servation Service aerial photographs taken in August, 1957 were used to measure land use in the Immediate area. NASA RB-57 high altitude photography taken in June, 1972 was used in measuring land use around parks constructed from 1965 to 1972. Since the acreage of the sixty mobile home develop­ ments in Genesee County varied considerably, it was neces­ sary to devise a method which assured consistency in mea­ suring surrounding land use. This was accomplished by the following procedure: 1) The study area was divided into units of one-quarter mile square. 2) A circle with a one-half mile radius was drawn from the center of each unit con­ taining land in mobile home parks. a) Each circle contained sixty-eight one- tenth mile square cells. b) Several parks were large enough to be found in more than one quarter-mlle square unit. There was, therefore, the possibility of multiple counting of the cells. Each tentn mile square cell was counted only once. 3) Surrounding land use was measured as the proportion of cells totally devoted to the categories of land use in the one mile circles from the centers of the units for each park. It Is expected that the land use of the immediate area will be more residential in character for recently constructed parks. 58 Land use adjacent park boundary. A third land use variable was that Immediately adjacent to the park boundary. This was measured to elicit the effects of mixed land uses on park location. Data were divided into five categories and were: 1) Mainly residential with no mixed uses. 2) Mainly undeveloped with no mixed uses. 3) Mixed but majority residential. 4) Mixed and majority nonresidentlal. 5) Mixed but majority undeveloped. As with the other land use measures, data were obtained from 1957 and 1972 air photography. Newer parks are ex­ pected to be located at sites where adjacent unmixed land use predominates. Distance to nearest elementary school. This fac­ tor, measured in tenths of miles, was along the most direct route from the development entrance to the nearest elementary school. School district boundaries were recog­ nized and nearest elementary school at the time of the park's construction was used. The influence of proximity to an elementary school is expected to be of greater Im­ portance for newer parks. Socioeconomic index. An additional variable mea­ suring features of the site and area environment was socio­ economic index. This is an index of four measures of each census tract population in relation to the population of 59 Che metropolitan area as a whole. An occupation ratio, an education ratio, median family income, and the median value of single-family homes were u s e d . ^ Each was standardized on a scale of 0-100 and the four ratios averaged to obtain an index for each census tract. Data were obtained from the 1960 census (U.S. Bureau of the Census, 1962a) for all parks constructed prior to 1957 and the 1970 census (U.S. Bureau of the Census, 1972a) for those opened after 1965. It is expected that recently constructed parks are located in tracts with higher socioeconomic index values. Incompatibility index. Newer parks should be located in areas of fewer incompatible uses. In order to determine If this is true, a crude index of incompatible uses was computed. This was a scale ranging from 0 to 8 and was obtained b y : 1) Railroad -- adjacent score: 2 within one-quarter mile score: 1 2) Industrial -- adjacent score: 2 in surrounding area score: 1 3) Highway -- adjacent score: 1 it) CooBnercial -- adjacent score: 1 5) Other incompatible uses -- adjacent score: 2 (salvage yards, sewer plants, etc.) in area score: 1 Data were obtained from 1957 and 1972 aerial photography. ^ T h e procedure for this measurement was outlined in Kaiser (1966: 223-226) who adapted the methodology from Eshref Shevky and Wendell Bell, 1955, Social Area Analysis, Stanford Sociological Series, No, 1; Stanford, California, Stanford University Press. Institutional Variables 60 Municipalities do much to encourage or retard de­ velopment within and adjacent to their boundaries. In order to take into account the importance of the municipal role, the following institutional variables were used: Zoning. "Zoning protection" was found to be sig­ nificant in the location of single-family developments (Kaiser, 1966). Mobile home parks are often for what the "protection" is Intended (Babcock, 1966). Restrictive zoning practices are listed as a major concern to the industry (Wehrly, 1972: 29). One of the main reasons for the more rural loca­ tion of mobile home developments than other forms of medium density housing may be due to restrictive municipal zoning practices. If municipalities prohibit mobile home develop­ ments or maintain unduly restrictive ordinances, it is reasonable that developers will seek more rural locations which generally have less restrictive ordinances. However, it is assumed that like other land developers, operators of more recently constructed parks would seek those areas which have formalized policy regarding their investment. Three measures of zoning were obtained. These were : 1) Percent of surrounding area zoned residential. 2) Zoning lnm»edlately adjacent to the park boundary. 3) Zoning of the specific site where the develop­ ment was located. 61 Data were obtained from municipal zoning maps. Because of the increasing residential function of parks, as well as recent changes in municipal ordinances, newer parks are expected to be more often located at sites and in areas zoned for residential purposes. Type of sewer system used. This variable was mea­ sured on the following schedule: 1) if the park used septic tanks 2) if the park used a lagoon system 3) if the park used a private sewer treatment plant 4) if the park was connected to a municipal sewer system Data on the type of sewer systems used by the parks in the study area were obtained from the Michigan Department of Health. Availability of municipal sewer service is ex­ pected to be a greater influence for the location of newer parks. Type of water system used. This variable was mea­ sured on a dichotomy of: 1) private water system 2) municipal water system This information was obtained from the Michigan Department of Health. As with availability of municipal sewers, newer parks should be more often connected to municipal water systems than older ones. 62 Type of Development It was expected that different types of parks would be influenced in varying ways by these factors. While mobile home parks are not usually differentiated, differ­ ences have been observed for single-family subdivisions (Kaiser and Weiss, 1969: 77) and apartments (Bourne, 1968), For example, research findings suggest that large devel­ opers tend to locate single-family subdivisions in a dif­ ferent spatial pattern than small developers. Small de­ velopers tend to locate in portions of the urban area which have fewer services and less desirable accessibility than large developers (Kaiser and Weiss, 1969: 77). There is a growing trend for large corporations to enter mobile home park development (Business Week, 1970: 74-75; Cloos and Birgells, 1972: 10). Often these firms will construct parks with several hundred spaces. How­ ever, mobile home park construction is normally not a con­ tinuing activity. There are a few firms which specialize in mobile home park construction and those which develop and manage several parks, but the majority are owned by individuals or firms owning only one park. For example, the listing of mobile home parks in Michigan indicates that the vast majority are proprietorships^® and the ^^There was one corporation with nine parks in seven counties containing a total of 810 spaces. Addi­ tionally another firm (listed as an individual) controlled nine parks containing over 4,000 spaces located In five counties (Michigan Department of Health, 1974). 63 average size is only 91 spaces (Michigan Department of Health, 1975) . The type of development used in this research was a dichotomy based on age of development. The literature as well as careful observation substantiates this classi­ fication. Older parks generally lack the amenities of m o d e m developments. They were smaller both In number of spaces and acreage, the densities were greater, and fea­ tures such as recreation facilities, open space, and other design features were lacking. Additionally, older parks were designed for smaller often transient trailers. The types of developments were: 1) Developments constructed prior to 1960. 2) Developments constructed 1960 or after. Data Analysis The area and site characteristics enumerated in this chapter were examined by the use of crosstabulation tables. Crosstabulation tables are most often used In the social sciences for survey research, but this technique has applicability for other forms of research as well (Baker and Lee, 1975). Some of the reasons for using this technique of analysis are elaborated below. First, it was felt that research into mobile home park location has not progressed to the level where mean­ ingful predictive statements can be made. This was essen­ tially a preliminary study where a descriptive base, as accurate as possible, was needed. 64 Secondly, many of the variables were obtained from information sources such as high altitude imagery, aerial photography, and small scale maps. Often it was possible to obtain only nominal or ordinal scale data from these sources. Crosstabulation provides a way of presenting and analyzing such data. Also, this research was concerned with not only measuring site and area factors influencing the location of mobile home parks in general, but also the different ways these factors influence types of parks. By categoriz­ ing parks and factors, a way was provided to better discern these differences. What was needed, then, was a technique that allowed numerical description and one in which relationships be­ tween variables could be examined without the assumptions of inferential statistics. In this research, the type of park (classified on the basis of age) was entered as a row variable and the area and site factors as column variables (Figure 14) . Four pieces of information are printed in each cell of the table. The upper number in each cell is the actual cell count or absolute frequency. The second number is the percent of the total row frequency in the cell, and the third is the percent of the total column frequency in the cell. The fourth or bottom number is the cell 's per­ centage of the total number of parks. Additionally, column I II, I J i ■ • i J ' 1 i *■ 1 - ' 4 t J ! . ' ■■■ 1 " • 1 1 • ' . , ' . r , ■ t k I / r , i ' . ■ " ' . ' i ■ ! . * ' ' - i . -i " . 1 1 ' . ' 1 | I 1 1 1 | 1 1 J > ' J. ' 1 * f • 1 ' 1. 1 J . . , 0 ' i . ' 1 . ' . 0 . 0 '■ 4 . < 1 / . :- / , , ■ v . . I ' , , . ■, ] 0 0 . 0 > , . r • t ’ V \ > ' 4 J * ! ■■ . ( 1 < • " . < . — j . ■ - ------------- - - - - ----------------- -------------------------------------- 1' ] ' L ^ t 1 / 1 ' * 1 J ■ a * i ■ ■ ■. ■■ , $ , > * Figure 14, An example of a crosstabulation table illustrating kinds of information included from CROSSTABS routine. 66 and row totals In both actual count and percent of total are printed at the bottom and right of the table respec­ tively.*7 The specific computer routine used was CROSSTABS. CROSSTABS is fully described in Statistical Package for the Social Sciences (Nie, Bent, and Hull, 1970: 115-128). The program is on file at the Computer Center at Southwest Missouri State University where the data were processed. The format of the table display printout from CROSSTABS (Figure 14) was used in typing the tables presented in this dissertation. CHAPTER FOUR MOBILE HOME DEVELOPMENT IN GENESEE COUNTY: DESCRIPTION Introduction Mobile home developments are associated with metropolitan areas and Michigan is no exception. For example, approximately 48 percent of Michigan's parks and 72 percent of park spaces were found in the state's Standard Metropolitan Statistical Areas. Genesee County of the Flint SMSA is one of the more important units of the state for mobile home developments. Sixty mobile home developments constructed in Genesee County prior to December, 1973 were used in this study (Figure 15). IB These parks contained 8,570 licensed spaces by the end of 1974 making Genesee second only to Oakland County of Metropolitan Detroit among Michigan's eighty-three counties (Michigan Department of Health, 1975) . Growth of Mobile Home Developments The Early Period There were two distinct phases of park construction 1 ft°Two additional parks were opened during 1974. f>7 r 68 G-TOC*-''. - Figure 15. Location of mobile home developments, cities, villages, and townships in Genesee County, 1973 (Source: field data). 69 in Genesee County. The first of these was between 1951- 19 57 and thirty-five parks with 2,285 spaces were con­ structed during those seven years (Table 4). While these parks ranged in size from 5 to 251 spaces, most were small. The average size of parks constructed during this period was sixty-four spaces. Two things are readily apparent from the pattern of parks constructed during this period (Figure 16). First, twenty-five of the thirty-six developments were located either in the City of Flint or within two miles of the city limits. The City of Flint alone contained twelve of the early developments all in the eastern part. In contrast, nine of the remaining parks were constructed at sites at least seven miles from the city limits of Flint. Secondly, precisely one-half of these early parks were located adjacent to or within one-half mile of one highway, what is now M-54. In addition to the small number of spaces, parks constructed during this time had a high density by m o d e m standards. The ten-foot wide mobile home did not become widely accepted until the roid-1950's. The older parks were built for eight-foot wide units, many of which were small enough to be pulled by automobile. Averaging only eight acres, the typical park's density was approximately fourteen units per acre. Additional characteristics of parks of this earlier period should be mentioned. Aerial photography clearly Mobile Home Development Construction in Genesee County by Year: 1950-1973 Table 4 Year 1950 51 52 53 54 55 56 57 58-64 65 66 67 68 69 70 71 72 73 1* 4 2 2 2 1 1 1 Political Unit City of Flint City of Davison City of Flushing 1 Burton Town­ ship (City) 2 2 1 Clayton Township Davison Township Fenton Township Flint Township 1 1 1 1 1 2 1 1 Total 14 1 1 5 1 3 3 1 1 1 4 1 1 10 Political Unit Year 1950 51 52 53 54 55 56 57 58-64 65 66 67 68 69 70 71 72 73 Total Table 4 - Continued Flushing Township Genesee Township Grand Blanc Township Montrose Township Mt. Morris Township Thetford Township Richfield Township V ienna Township Total 1 1 L 1 1 1 1 1 1 1 1 1 1 1 1 3 1 1 1 1 2 6 2 3 4 1 2 2 1* 5 10 2 7 7 3 1 0 1 1 1 2 1 6 4 3 5 60 ★Licensed in 1947 Source: Michigan Department of Health 72 \ ; ' ! \ : r. . 4* <> ■r' Si t - - - I- 4 • N(V \ # 9 I H Q B ’ L C H O M f E > t V C lO r M I « T B C M I T f l V C T f » I M T - I H 7 •■««•(! t^atT, Figure 16. Location of mobile home developments con- strueted prior to 1957 (Source: Michigan Department of Health and field data). 7 3 indicated two things both relating to park design and lay­ out. The first is straight alignment of park streets with the resulting diagonal or perpendicular "parking" of the mobile homes . The second design feature is the lack of open space and recreation facilities. While some had laundry facili­ ties and perhaps small playgrounds, "community” buildings, recreation facilities, and open space were not characteris­ tic of parks in this early stage of mobile home develop­ ments . The Intervening Years In the seven year period between 1957 and 1964, no parks were developed in the Flint area. However, expan­ sion of existing parks consisted of 359 spaces added dur­ ing this time period. By 1964 there were 2,700 spaces in Genesee County, and the average size of the developments had increased to seventy-five spaces. The Latter Period The second period of park construction began in 1965 and was still in progress at the end of 1973. The twenty-four parks constructed during this latter period were larger, had a lower density per acre, and contained more amenities within the park. The 5,929 spaces constructed during this time brought the average size of all parks to 141 spaces. The average size of the newer parks, however, was 204 spaces. 7 4 Developments were also larger in total acres; sixteen c o n ­ tained more than twenty acres. Density of the newer parks was greatly reduced over the older parks. With the average of twenty-eight developed acres, the average density of the newer parks was seven units per acre. This brought the density of all sixty parks in the county to eleven units per acre. In addition to larger mobile homes (12-foot wide units were introduced in the mid-1960's and 14-foot wide homes were allowed in Michigan beginning in 1972), an additional reason for decreased densities was the increased amount of open space in newer parks. Larger recreation facilities were common in these parks and a greater dis­ tance between units was apparent.^ The pattern of the recent phase of construction differed from that of the earlier period (Figure 17). A more dispersed pattern is apparent for the more recently constructed developments. For example, only two of the newer parks were constructed in the City of Flint com­ pared to twelve of those opened prior to 1957. There was also a shift in the direction of development growth. Dur­ ing the first period the construction of developments was in the eastern part of the county. Dividing the study area into eight sectors, twenty-three of the thirty-six ^Significant also in reducing overall densities was the needed space for sewer treatment lagoons in four of the suburban parks not served by municipal sewer. Figure 17. Location of mobile home developments con- Hiath*SLi9g;is72.is"c’‘ Michigan D*p,irtm*nt °f 76 older parks were located In the four eastern sectors (Fig­ ure 18). While sector one, and to a lesser extent sector two, continued to be Important directions of growth during the latter period, none of the newer parks were constructed in areas to the southeast (sectors three and four) (Figure 19). In fact, only five of the twenty-four newer develop­ ments were constructed in the southern part of the county. Location by Political Type Maps presented previously In this chapter (Figures 15, 16, and 17) reveal a general fringe location of mobile home parks in Genesee County. It is significant, however, that sixteen were located within incorporated cities . Us­ ing a modification of a classlficatory scheme used by the Genesee County Metropolitan Planning Commission (1971: 36), the twenty-six political entities were grouped into five municipal types. The eighteen townships were grouped into three categories: Rural, Suburban, and Urban. Using this categorization, there are seven rural, six suburban, and five urban townships. The seven smaller cities^® com­ prised the fourth group and the City of Flint by itself made up the fifth one. Mobile home developments were most numerous in the five urban townships (Table 5). There were twenty-seven developments with 3,496 spaces accounting for almost 41 ^^There were in addition five villages in the county. These were included in the township figures. 77 G E N E S E E COUNTY, MICHJGAN Figure 18. Location of pre-1957 mobile home develop­ ments by sector (Source: field data). 78 Figure 19. Location of mobile home developments con­ structed after 1965 by sector (Source; field data). Mobile Home Developments in Genesee County by Political Type: 1951-1974 Table 5 Political Type 1951 Sp X Pk 1955 1960 Pk Sp X Pk Sp X Pk 1965 Sp X Pk 1970 Sp % Pk 1975 Sp X Flint 2 120 29 10 975 48 12 1402 52 13 1723 55 14 1920 35 14 1920 22 Small City - - - 1 26 1 1 26 1 1 26 1 2 161 3 2 424 5 Urban Township Suburban Township Rural Township 5 208 50 14 805 39 15 995 37 15 1087 35 20 2107 38 27 3496 41 2 - 85 21 6 158 8 6 211 8 6 211 7 9 609 11 12 1245 15 - - 2 79 4 2 79 2 2 107 2 4 707 13 5 1485 17 Total 9 413 - 33 2043 - 36 2713 - 37 3154 - 49 5504 - 60 8570 - Source: Computations based on data from the Michigan Department of Health. 80 percent of all spaces in Genesee County. Of least impor­ tance were the county's seven smaller cities. Only two, Davison and Flushing, contained any parks, each with only one development. Combined, these had approximately 5 per­ cent of the total development spaces. While only five developments were found in the seven rural townships, two were the largest in the county. In fact, these two parks with 1,206 spaces accounted for 14 percent of all development in the county. Four of the seven rural townships contained no parks. This pattern has changed through time. In 1955 the City of Flint had ten of the thirty-two parks and 48 percent of the 2,043 spaces in the county. The peak pro­ portion for Flint was reached in 1965 when the city's thirteen parks contained 55 percent of the 3,154 spaces in the county. Since that time only one park has been constructed in the city and Flint's share had decreased to 22 percent by the end of 1973. The five urban townships have remained important, with the share never below 35 percent (1965) but construc­ tion has fluctuated in these townships. In 1955, the share was approximately the same as in 1973 (39 percent compared to 41), with proportionally less construction in the inter­ vening years. The greatest amount of park construction in the urban townships has occurred since 1965 with an addi­ tion of twelve parks and 2,405 spaces. 81 Changes in Parks The first period saw development in the outer por­ tions of the central city or at great distances. Later development has concentrated in the urban and suburban townships. The parks have changed; they are larger in total acres, In lot size, and in total numbers of spaces. Parks have also changed in the function they serve. For example, all parks in Flint were adjacent to or within one- quarter mile of M - 54 (or the Dort Highway). The highway orientation of smaller trailers can partially account for this as well as the zoning ordinances of the City of Flint which specified that parks be located in commercial dis­ tricts . One of the greatest differences between older and newer parks is size (Table 6). Two-thirds of the older parks had fewer than 100 spaces, and only three of the thirty-six parks in this category exceed 200 spaces. While four newer parks were small, these were in incomplete stages of development. Thirteen of the newer parks were medium sized but seven exceeded 200 spaces. Related to size are the number of acres in the parks and the density of units per acre. These measures also point to changes in mobile home developments. Mobile home parks are becoming significantly larger (Table 7). Twenty-six of the older parks have fewer than nine developed acres. A full one-third of the newer parks exceed thirty developed acres compared to none of the older ones. 82 Table 6 Size of Parks in Number of Spaces Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row X Column % % of Total Number Row % Column % % of Total Total number Column % of total Number of Spaces 1-99 100-199 200-608 Total 24 66.7 85.7 40.0 4 16.7 14.3 6.7 28 46.7 9 25.0 40.9 15.0 13 54.2 59.1 21.7 22 36.7 3 8.3 30.0 5.0 7 29.2 70.0 11.7 10 16.7 36 60.0 24 40.0 60 Source: Michigan Department of Health. Densities are also decreasing. All but two of the older parks exceed nine units per gross acre and nineteen exceed fourteen units per acre (Table 8). Only two of the twenty-four newer parks exceed nine spaces per acre. Summary In the twenty-three years between 1950 and 1973 fifty-nine parks with 8,514 spaces were constructed in Genesee County. During this time, parks have Increased in both acres and number of spaces, have lower densities, and have greater amenities than older parks. There has been fluctuation in park construction through the years in Genesee County. Since 1951, there have been seven years (1958-1964) in which no parks have Table 7 Size of Parks in Number of Developed Acres Number of Acres 3-9 10-19 20-29 30-39 40-85 Total 26 72.2 100.0 43.3 0 0 0 0 26 43.3 8 22.2 50.0 13.3 8 33.3 50.0 13.3 16 26.7 2 5.6 20.0 3.3 8 33.3 80.0 13.3 10 16.7 0 0 0 0 3 12.5 100.0 5.0 3 5.0 0 0 0 0 5 20.8 100.0 8.3 5 8.3 36 60.0 24 40.0 60 Parks Constructed Prior to 1937 Parks Constructed 1965-1973 Number Row 7. Column 7, X of Total Number Row X Column X % of Total Total number Column 7. of Total Source: Field data. Table 8 Density of Parks in Spaces Per Acre Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row 7. Column % % of Total Number Row % Column 7. 7, of Total Total number Column % of Total Spaces Per Acre Not 10-13 14-17 18-21 Complete Tot; 15 41.7 100.0 25.0 1 4.2 7.7 1.7 16 26.7 12 33.3 92.3 20.0 0 0 0 0 7 19.4 100.0 11.7 0 0 0 0 12 20.0 7 11*7 0 0 0 0 5 20.8 100.0 8.3 5 8.3 36 60 24 40 60 5-9 2 5.6 10.0 3.3 18 75,0 90*0 30.0 20 33.3 Source: Michigan Department of Health and field data* 85 opened. The leading year for park construction was in 1952 when ten parks were opened. Seven parks were opened in 1954 and 1955, with six opening in 1970. There has been a shift in both municipal type and direction of growth. From a peak of 55 percent of total county development in 1965, the proportion in the City of Flint decreased to 22 percent in 1973. The urban town­ ships , particularly Flint and Genesee Townships, have be­ come most important with some 41 percent of the developed spaces. The direction of growth has shifted to the northern and western portions of the county. CHAPTER FIVE MOBILE HOME DEVELOPMENTS IN GENESEE COUNTY: ANALYSIS OF SITE AND AREA FACTORS Introduc tion The previous chapter provided a base for the analy­ sis of site and area factors influencing the location of mobile home developments in the Genesee County study area. It is the purpose of this chapter to: 1) identify loca­ tional factors influencing park construction in the study area, 2) note similarities between factors Influencing the location of other forms of residential development, and 3) elicit changes in combinations of site and area factors Influencing the location of mobile home develop­ ments constructed in different time periods. Distance to Urban Center Accessibility Factors Mobile home developments in the study area were located at sites with a wide range of distances from the urban center. The range extended from a low of 2.8 road miles to the CBD to a maximum observed distance of 20.8 miles. The median distance from the CBD for all sixty parks was 5.4 miles. For early parks the median distance 86 87 was 4.6 miles compared to 8.1 miles for the more recently constructed ones. However, in both periods of construc­ tion there was a wide range between minimum and maximum distances. Even though the ranges were considerable, there was moderate concentration of parks at locations of inter­ mediate distance from the CBD (Table 9). For example, 68 percent were located within nine miles and over 53 percent within six miles of the urban center. Unlike early parks which were most often constructed at distances under six miles, recently constructed parks were more evenly dis­ tributed through the categories. It is apparent, however, that even parks of the recent period were not constructed in great numbers in the more distant portions of the county. Eighteen of the twenty-four newer developments were within twelve miles of the urban center. Distance to the Center of Employroen~t Distance to the mean geographic center of manufac­ turing employment revealed differences between the two categories of developments (Table 10). However, for the study area, this measure adds little to the overall spa­ tial understanding of mobile home developments because the center of employment was only 0.4 miles from the urban cen­ ter. The median distance to the center of employment for all sixty parks was identical to that for the distances to the CBD (5.4 miles) and similar to that measured for early Table 9 Distance to the Urban Center Road Miles 2,8- 5.9 25 69.4 78,1 41.7 7 29.2 21.9 11.7 32 53.3 6.0- 8.9 2 5.6 22.2 3.3 7 29.2 77.8 11.7 9 15.0 9.0- 11.9 3 8,3 42.9 5.0 4 16.7 57.1 6.7 7 11.7 12.0- 14.9 15.0 or more Total 2 5.6 40.0 3.3 3 12.5 60.0 5.0 5 8.3 4 11.1 57.1 6.7 3 12.5 42.9 5.0 7 11.7 36 60.0 24 40.0 60 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column 7. X of Total Number Row % Column 7, X of Total Total number Column % of Total Source: Field data. Table 10 Distance to Center of Employment Road Miles 2,8- 5.9 25 69.4 78.1 41.7 7 29.2 21.9 11.7 32 53.3 6.0- 8.9 2 5.6 22.2 3.3 7 29.2 77.8 11.7 9 15.0 9.0- 11.9 3 8.3 42.9 5.0 4 16.7 57.1 6.7 7 11.7 12.0- 14.9 15.0 or more Total 3 8.3 50.0 5.0 3 12.5 50.0 5.0 6 10.0 3 8.3 50.0 5.0 3 12.5 50.0 5.0 6 10,0 36 60.0 24 40.0 60 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row 7. Column % 7, of Total Number Row 7, Column % % of Total Total number Column 70 of Total Source: Field data. and recent parks (A.4 miles and 8.5 miles respectively). 90 Distance to Highways Proximity to state or federal highways was an im­ portant factor for development location in the study a r e a ^ (Table 11). Twenty-five of the parks were located adjacent to highways at the time of their construction and the median distance was only 0.2 miles. Only seven of the sixty parks were located at distances exceeding one mile from a highway with 3.9 miles the maximum distance ob­ served. While similar percentages of both early and recent parks were located within one half mile of a highway, a higher percentage of newer developments were located ad­ jacent to highways. Influence of Accessi­ bility Factors It is apparent that of the three accessibility fac­ tors measured, proximity to highways had the strongest in­ fluence on the location of mobile home p a r k s . It is also apparent that this was an important factor in both con­ struction periods. While not as Important as highway proximity, distance to the urban center and to employment was more important for parks in the study area than the literature suggests. This conclusion is based on the c o n ­ centration of parks at locations of intermediate distance 2 1 Only six parks were not located on major thorough­ fares as defined by the Genesee County Metropolitan Planning Commission (1970c). Five of these were older parks. Table 11 Distance to State or Federal Highway Adja­ cent 13 36.1 52.0 21.7 12 50.0 48.0 20.0 25 41.7 0.2- 0.5 16 44.4 69.6 26.7 7 29.2 30.4 11.7 23 38.3 Miles 0.6- 1.0 3 8.3 60.0 5.0 2 8.3 40.0 3.3 5 8.3 1.1- 2.0 2 5.6 100.0 3.3 0 0 0 0 2 3.3 2.1 or more 2 5.6 40.0 3.3 3 12.5 60.0 5,0 5 8.3 Total 36 60.0 24 40.0 60 Parks Constructed Prior to 1957 Parks Constructed 1965*1973 Number Row % Column % % of Total Number Row X Column % X of Total Total number Column X of Total Source: Field data. 92 and that only seven of the developments were located at distances exceeding fifteen miles from the CBD. Since most developments were constructed at intermediate dis­ tances and in close proximity to major highways, the com­ muting distance from the majority of the newer parks would not seem excessive. Availability of Undeveloped Land Related to the general peripheral location of mobile home developments is the percentage of land not in urban development in the immediate area of the park. The range for the fifty-two developments for which data were available was considerable; from a low of 13 percent for one development in the City of Flint to 99 percent for two rural parks . The median for all parks was 60 percent and the medians for both early and recent parks were similar (60 and 63 percent respectively). Approximately four- fifths of the developments were located where at least 40 percent of the immediate area was undeveloped and approxi­ mately 58 percent of the parks where the value exceeded 60 percent (Table 12). As might be expected, recently constructed parks were found more often in areas of greater percentages of undeveloped land. Only three were in areas where the value was less than 40 percent of surrounding undeveloped land while thirteen recent parks were in areas where the p e r ­ centage exceeded 60. Table 12 Percent of Surrounding Area Not in Urban Development Percent 20-39 40-59 60 or more Total 8 19.4 85.7 11.5 1 4.8 14.3 1.9 7 13.5 7 22.6 58.3 13.5 5 23.8 41.7 9.6 12 23.1 17 54.8 56.7 32.7 13 61.9 43.3 25.0 30 57.7 31 59.6 21 40.4 52 1-19 1 3.2 33.3 1.9 2 9.5 66.7 3.8 3 5.8 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column X % of Total Number Row % Column 7, X of Total Total number Column X of Total Source: Field data. 94 Area Environmental Factors Socioeconomic Prestige A socioeconomic index was computed for all census tracts in the study area for both 1960 and 1970. The index based on the 1960 census figures was used for parks con­ structed prior to 1957 and data from the 1970 census were used for those constructed after 1965. Both indexes had a potential range of 0-100 with 100 indicating the highest socioeconomic prestige level. The actual range for tracts where mobile home parks were located was from 15 to 62 with a median index of 39. The majority of all parks (88 percent) were located in tracts with a value between 25 and 54 (Table 13). The socioeconomic prestige of the immediate area was a stronger influence for recently constructed parks. This was re­ flected in a higher median (43 compared to 33) and a higher modal interval; none of the recent parks were in tracts which had an index below 28. Interestingly, the Index in­ creased between 1960-1970 for tracts in which twenty-one of the early parks were located. Of the thirteen early parks in which the index was lower in 1970, eleven were in the City of Flint. Land Use in the Immediate Area An important feature of the area environment is surrounding land u s e . Three factors were used in this measurement: 1) surrounding residential land use Table 13 Socioeconomic Prestige of Area Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column % % of Total Number Row % Column % % of Total Total number Column % of Total Index 15-24 25-34 35-44 45-54 8 22.2 100.0 13.3 0 0 0 0 8 13.3 14 33,9 77.8 23.3 4 16.7 22.2 6.7 18 30.0 3 8.3 21.4 5.0 11 45.8 78.6 18.3 14 23.3 9 25.0 60.0 15.0 6 25.0 40.0 10.0 15 25.0 55 or higher 2 5.6 40.0 3.3 3 12.5 60.0 5.0 5 8.3 Total 36 60.0 24 40.0 60 Source: Computed from: U.S. Bureau of the Census, 1962a; 1972a. 96 (excluding mobile home developments), 2) surrounding non* residential land use, and 3) land use immediately adjacent to the park boundary. Land use data were obtained from 1957 and 19 72 aerial photography and were available for fifty-two developments. Surrounding residential land u s e . All but six of the parks for which data were available had some residen­ tial development in the immediate area. The median per­ centage was 16 and the range was from no surrounding resi­ dential land use to 68 percent. There were only slight differences between the two age groups. For example, the median value of 15 percent for newer parks was similar to the 16 percent median for parks constructed prior to 1957. Also, the modal interval for both age categories was the 1-19 percent range and the differences between early and recent parks in the other intervals were not great (Table W ) . Surrounding nonresidential land u s e . Fewer parks in the study area were in areas of nonresidential land use than might be expected on the basis of the general litera­ ture. While the overall range was from 0 to 59 percent, fifteen developments had no surrounding area devoted to nonresidentia1 uses. The median for all parks was 4 per­ cent. Early parks were more often located in nonresiden- tial areas than those constructed after 1965. The median for newer parks was 2 percent compared to 8 percent for Table 14 Residential Land Use In Surrounding Area Percent None 1-19 20-39 40-59 60 or more Total 3 9.7 50.0 5.8 3 14.3 50.0 5.8 6 11.5 13 41.9 56.5 25.0 10 47.6 43.5 19,2 23 44.2 12 38.7 63.2 23.1 7 33.3 36.8 13.5 19 36.5 1 3.2 50.0 1.9 1 4.8 50.0 1.9 2 3.8 2 6.5 100.0 3.8 0 0 0 0 2 3.8 31 59.6 21 40.4 52 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row X Column X % of Total Number Row % Column % X of Total Total number Column X of Total Source: Field data. 98 early parks. Only one recent park was In an area exceed­ ing 19 percent nonresidential uses compared to six older developments (Table 15). Land use adjacent park boundary. Only slightly over one-half of the parks had no contiguous mixed land use (Table 16). This reveals the general tendency for mobile home parks to be located in areas less desirable than is generally true for single family subdivisions. There were, however, apparent differences between older and newer parks. One example is the greater percentage of older parks with adjacent mixed land use. Over one-half of the parks constructed prior to 1957 had contiguous mixed land use compared to fewer than one-third of the more recently constructed ones. For the study area, the absence of mixed adjacent land use was a more important influence on the location of newer parks than those opened during the earlier period. Index of Incompatible Uses A crude index of uses generally considered incom­ patible with residential development was computed for fifty-six parks. The potential range was from 0 (no nega­ tive features in the immediate area) to 8 (high negative index). The actual range was from 0 for sixteen develop­ ments to 7 for one early park. The primary difference between early and recent parks was at the high negative end of the scale (Table 17). Table 15 Nonresidential Land Use in Surrounding Area Percent 20-39 6 19.4 100.0 11.5 0 0 0 0 6 11.5 1-19 16 51.6 53.3 30.8 14 66.7 46.7 26.9 30 57.7 None 9 29.0 60.0 17.3 6 28.6 40.0 11.5 15 28.8 40-59 Total 0 0 0 0 1 4.8 100.0 1.9 1 1.9 31 59.6 21 40.4 52 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row X Column X % of Total Number Row % Column X X of Total Total number Column X of Total Source: Field data. Table 16 Land Use Adjacent Park Boundary Residen­ Undevel­ tial Type Mixed1 Mixed'1 Mixed Total 0 0 0 0 4 19.0 100.0 7.7 4 7.7 6 19.4 85.7 11.5 1 4.8 14.3 1.9 7 13.5 11 35.5 84.6 21.2 3 14.3 21.4 5.7 14 26.9 31 59.6 21 40.4 52 oped 11 35.5 47.8 21.2 12 57.1 52.2 23.1 23 44.2 3 9.7 75.0 5.8 1 4.8 25.0 1.9 4 7.7 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column % 7, of Total Number Row X Column % X of Total Total number Column X of Total Source: Field data. ^Majority residential ^Majority nonresidential ^Majority undeveloped Table 17 Index of Incompatible Uses Index 0 1 - 2 3 - 4 5 - 6 O O1 Total 9 28.1 56.3 16.1 7 29.2 43.8 12.5 16 28.6 12 37.5 48.0 21.4 13 54.2 52.0 23.2 25 44.6 6 18.8 60,0 10.7 4 16.7 40.0 7.1 10 17.9 4 12.5 100.0 7.1 l 3.1 100.0 1.8 0 0 0 0 4 7.1 0 0 0 0 1 1.8 32 57.1 24 42.9 56 Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row 7. Column % % of Total Number Row % Column % % of Total Total number Column % of Total Source: Field data. 102 None of the recent parks had a score exceeding 4 compared to five early developments. Recently constructed parks tended not to be located adjacent to railroads or In in­ dustrial areas as was true for many older ones. Location near freeways was the major contributor to negative scores for newer parks. Only sixteen of the parks were located in areas free from any nearby negative features indicating that the absence of these factors has not been an especially strong influence on park location. Also indicated by this measure, however, is that there has been an upgrading of sites dur­ ing the recent period of construction. Distance to Elemen­ tary School It was thought that newer parks would be more greatly influenced by proximity to elementary schools. This factor, however, appears to be of little importance to development location (Table 18). The overall range varied from 0.4 to 7.9 miles and the median distance to an elementary school was 1.8 miles. The median for older parks was 1.6 miles compared to 2.0 miles for parks con­ structed after 1965. Only eight of the developments were within walking distance of schools which was defined as three-quarters of a mile (Genesee County Metropolitan Planning Conmission, 1970a: 6). Table 18 Distance to Nearest Elementary School in District Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column X % of Total Number Row X Column 70 X of Total Total number Column X of Total Source: Field data. 0.8 or less 5 13.9 62.5 8.3 3 12.5 37.5 5.0 8 13.3 0.9- 1.5 10 27.8 62.5 16.7 6 25.0 37.5 10.0 16 26.7 1.6- 3.0 11 30.6 52.4 18.3 10 41,7 47,6 16.7 21 35.0 3.1- 4.5 4 11.1 50.0 6.7 4 16.7 50.0 6.7 8 13.3 4.6 or more 6 16,7 85.7 10.0 1 4.2 14.3 1.7 7 11.7 Total 36 60.0 24 40.0 60 1 0 3 Influence of Environ­ mental Factors 104 When considering the general location of all parks in the study area, the environmental quality of the immedi­ ate area was not a strong influence on the location of mobile home parks. This conclusion is based on the fol­ lowing : 1) Most parks were located in areas where the residential land uses were not in the major­ ity. 2) Most parks were located in tracts which had a relatively low socioeconomic level. 3) Most parks were located at considerable dis­ tances from elementary schools. 4) Most parks were in areas where nearby nega­ tive features were present. However, four of the six environmental factors used in this study indicated changing locational considera­ tions for newer parks. Parks constructed after 1965 were more often located in areas which had a moderate instead of low socioeconomic level as was true for older parks. Mixed land use, both in terms of surrounding area and that adjacent to the park boundary, was not as common for recent mobile home developments. Additionally, absence of nearby features which might distract from the residential quality of the area was a stronger influence for modern parks. These factors indicate transition, but this transition has not progressed to the point where the quality of the im­ mediate environment is the same location influence as is generally true for single family subdivisions. 105 Institutional Factors Zoning As presented in Chapter Two, zoning is a critical factor to establishing mobile home developments. The as­ sumption has been that zoning permitting mobile home de­ velopment would be more easily obtained in rural areas. Related to this is the absence of rural zoning in many parts of the nation. When parks are allowed in urban areas, the assumption has been that they would more likely occur in those portions zoned for nonresidential purposes. All jurisdictions in Genesee County have zoning ordinances and only ten parks were established prior to zoning regulations in three of the townships. Three items pertaining to zoning were measured. These were: 1) per­ centage of residential zoning in the surrounding area, 2) zoning immediately adjacent the park boundary, and 3) zoning of the specific site. Zoning data were avail­ able on fifty-seven of the sixty developments. Surrounding residential zoning. Almost three- quarters of the parks were located where at least 40 per­ cent of the immediate area was zoned residential and only seven were located where none of the surrounding area was zoned residential. There were few differences between early and recent developments (Table 19). The range for all parks varied from 0-95 percent of the surrounding area zoned residential with a median Table 19 Residential Zoning in the Surrounding Area Percent None 1-19 20-39 40-59 60 or more Total Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column X X of Total Number Row X Column X X of Total Total number Column X of Total 3 8.6 42.9 5.3 4 18.2 57.1 7.0 7 12.3 1 2.9 100.0 1.8 0 0 0 0 1 1,8 5 14.3 62.5 8.8 3 13.6 37.5 5.3 8 14.0 14 40.0 70.0 24.6 6 27.3 30.0 10.5 20 35.1 12 34.3 57.1 21.1 9 40.9 42.9 15.8 21 36.8 35 61.4 22 38.6 57 Source: Municipal Zoning Maps^, ^Cities--Davison, Flint; Townships--Burton, Clayton, Davison, Fenton* Flint, Flushing, Genesee, Grand Blanc, Montrose, Mt, Morris, Thetford, Vienna. 107 percentage of 51. The two age groupings were similar. Older parks had a median of 54 percent compared to 48 per­ cent for those constructed after 1965. Zoning adjacent to boundary. Examination of the zoning designation of land immediately adjacent to park boundaries provides additional insights. Most had mixed contiguous zoning and only eight had all commercial or all industrial zoning adjacent to the park (Table 20). Thirty had mixed contiguous zoning, two-thirds of which were older parks. Precisely one-half of the recent parks had all the adjacent land zoned residential compared to 23 per­ cent of the early developments. Specific site zoning. Considerable differences were apparent between older and newer parks when specific site zoning was examined (Table 21). Ten early parks were established prior to zoning regulations in their jurisdic­ tions, but there were other differences as well. For ex­ ample, only three recent parks were located on sites zoned for nonresidential purposes compared to fifteen older de­ velopments. Additionally, a greater proportion of recent parks were located on parcels zoned residential. Type of Sewer System A recent national survey reported that 69 percent of higher quality mobile home parks were connected to municipal sewer systems (Wehrly, 19 72: 23). Within the Table 20 Zoning Adjacent park Boundary Type Residen­ tial Mixed ft Commer­ Indus­ trial cial Mixed^ Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column X X of Total Number Row % Column % X of Total Total number Column 7. of Total 8 22.9 42.1 14.0 11 50.0 57.9 19.3 19 33.3 8 22.9 61.5 14.0 5 22.7 38.5 8.8 13 22.8 12 34.3 75.0 21.1 5 22.7 38.5 8.8 17 29.8 4 11.4 100.0 7.0 0 0 0 0 4 7.0 3 8.6 75.0 5.3 1 4.5 25.0 1.8 4 7.0 Total 31 61.4 22 38.6 57 Source: Municipal Zoning Maps^. ^Residential and comoercial ^Residential, commercial, and industrial ■^Cities--Davison, Flint; Townships--Burton, Clayton, Davison, Fenton, Flint, Flushing, Genesee, Grand Blanc, Montrose, Mt. Morris* Thetford, Vienna. Table 21 Specific Zoning of Site Type Residen­ tial Mobile Home Nonresi- dential No Zoning 8 2 2 . 9 4 7 . 1 1 4 . 0 9 4 0 . 9 5 2 .9 1 5 . 8 17 2 9 . 8 2 5 . 7 1 6 .7 3 . 5 10 4 5 . 5 8 3 . 3 1 7 . 5 12 2 1 . 1 15 4 2 . 9 8 3 . 3 2 6 . 3 3 1 3 .6 1 6 .7 5 . 3 18 3 1 . 6 10 2 8 . 6 1 0 0 . 0 1 7 . 5 0 0 0 0 10 1 7 . 5 Total 35 6 1 . 4 22 3 8 .6 57 Parks Constructed Prior to 1957 Parks Constructed 19 65 -1973 Number Row X Column X X of Total Number Row X Column X X of Total Total number Column X of Total Source: Municipal Zoning Maps*, *Cities--Davison, Flint; Townships--Burton, Clayton, Davison, Fenton, Flint, Flushing, Genesee, Grand Blanc, Montrose, Mt. Morris, Thetford, Vienna, 110 study area, however, 77 percent of all parks were on munic­ ipal systems. Almost Identical proportions were found for early or recent developments (Table 22). The major dif­ ferences between the two groups were that only one newer park utilized septic tanks and that four newer parks used lagoon sewer treatment systems. Type of Water System The availability of municipal water supply was not a critical factor of location for parks within the study area. Only 27 percent were connected to municipal water systems. An almost identical proportion of recent and early parks utilized municipal supplies (Table 23). Influence of Institu­ tional Factors Parks within the study area were most often located at sites and in areas zoned for some form of residential use. This partially reflects changes in the ways munici­ palities designate mobile home parks in zoning ordinances. Beginning in the mid-1930's, Michigan courts have generally ruled against municipalities prohibiting mobile home parks outright or excluding them by making no provision for parks in their ordinances (exclusionary zoning). This has been noted frequently in the literature (examples include: Newcomb, 1971: 38; Gibson, 1972: 36-37; Galetschky, 1974: 12-13; Elias, 1974: 171). In only four political units containing parks in Genesee County are they presently Table 22 Type of Sewer System Used Type Septic Lagoon Private Municipal Total Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Number Row % Column % X of Total Number Row X Column % 7. of Total Total number Column % of Total 8 22.2 88.9 13.3 1 4.2 11.1 1.7 9 15.0 0 0 0 0 4 16.7 100.0 6.7 4 6.7 1 2.8 100.0 1.7 0 0 0 0 1 1.7 27 75.0 58.7 45.0 19 79.2 41.3 31.7 46 76.7 36 60.0 24 40.0 60 Source: Michigan Department of Health 112 Table 23 Type of Water System Used Parks Constructed Prior to 1957 Parks Constructed 1965-1973 Total number Column X of Total Type Private Municipal Total Number Row % Column % X of Total Number Row X Column % X of Total 20 55.6 60.6 33.3 13 54.2 39.4 21.7 33 55.0 16 44.4 59.3 26.7 11 45.8 40.7 18.3 27 45.0 36 60.0 24 0•0 60 Source: Michigan Department of Health. restricted to comnercial districts in their zoning ordin- ances. In the remaining eleven municipalities, parks are allowed in mobile home, mobile home and apartment, or residential districts. Most of these (eight) presently have specific mobile home districts. It should be noted, however, that the majority of these changes in designation are recent. Within the study area, regardless of the ways parks were designated at the time of construction, they were most often located at sites and in areas zoned for some form of residential u s e . The high percentage (77 percent) of parks connected to municipal sewer systems may be unique to the study area. consulted is found in Appendix C. A list of municipal zoning ordinances and maps 113 Genesee County began a program In 1961 to provide sewerage service to all developed areas of the county (Banks, 1972: 16). Additionally, state law requires connection to a pub­ lic sewer system " . . . where a public sewer system Is available and accessible" (Michigan Department of Health, 1971: 12). Summary The idea that mobile home developments display "no particular pattern of location," at least for parks within the study area, should be modified. It Is felt that state­ ments as the above indicating a "random" pattern of loca­ tion result from failure to analyze park location in de­ tail. Seven of the site and area factors measured in this study were associated with general park location. Distance to the urban center and to employment opportunities was of greater importance than the literature suggests. There was no doubt of the general peripheral location, yet 68 percent of all parks were within nine miles of the urban center and the center of employment, and 80 percent were within twelve miles. Proximity to highways has been emphasized in other studies (El Gammal, 1966: 72; Freed, 1969: 202) and this was also an important factor for parks in Genesee County. Slightly over 80 percent were adjacent to or within one-half mile of highways. Related to the peripheral location, an additional important factor for all parks was the availability of 114 undeveloped land. Approximately 58 percent were located In areas where undeveloped land exceeded 60 percent and 81 percent where the surrounding area exceeded 40 percent undeveloped. When all parks were considered, area environmental factors tended not to be of great Importance. For example, the presence of mixed adjacent land use was not a major inhibiting factor. Only the absence of excessive numbers of nearby noxious elements seemed to be an important in­ fluence for all sixty parks. Two institutional factors--surrounding residential zoning and availability of municipal sewers — we re of im­ portance to the location of all parks. Approximately 72 percent were located where surrounding residential zoning exceeded 40 percent. Approximately 77 percent were con­ nected to municipal sewers. The findings also show that certain site and area factors were not of great importance to park location in general. This was Indicated by low socioeconomic levels of areas where parks were located, the variety of surround­ ing land use, and low importance of proximity to elementary schools. It was also found that site and adjacent zoning and availability of municipal water supplies were among the least important factors influencing general mobile home park location. Also important were the varying combinations of factors influencing the location of parks constructed 115 during the early and recent phases. There were differ­ ences between the two categories In two of the distance factors and in availability of undeveloped land. As could be expected, the larger recently constructed parks were located at greater distances from the urban center and em­ ployment opportunities. The median distance for recent parks on both measures was 8.1 miles compared to 4.6 miles for early parks. While the ranges and medians for both categories of parks were similar for the percent of un­ developed land, a slightly higher percentage of recent parks were In areas where undeveloped land exceeded 40 percent. The greatest differences between early and recent parks were found in the measures of the immediate environ­ ment. Newer parks were more likely to be located in areas where the socioeconomic level is higher, where the surround­ ing nonresidential land use is lower, and where unmixed ad­ jacent land use is present. Also an important factor in­ fluencing the location of recently constructed parks was the absence of nearby incompatible features. This last characteristic would have shown even greater differences were it not for the tendency for newer parks to be located adjacent to freeways. There were also differences apparent in two insti­ tutional measures. Zoning of the specific site for newer parks was most often residential or specifically mobile home designation. Only three recent parks were located at 1X6 sites zoned nonresidential compared to fifteen early ones. The importance of residential zoning was also found in the measure of the type of zoning immediately adjacent the park boundary. Older parks were most often located where mixed adjacent zoning was present and particularly where commer­ cial or industrial zoning was predominant. For newer parks, however, the absence of mixed zoning was the most important. Site and area factors influencing the location of mobile home developments are not identical to those found to Influence the location of single family subdivisions. However, the findings of this study show certain changes in location of parks opened in the two construction phases. These changes Indicate a convergence toward locations which are more typically residential in character for newer parks and these changes have several implications which are pre­ sented in Chapter Six. CHAPTER SIX SUMMARY AND IMPLICATIONS Introduc Cion The rapid increase In the use of mobile homes is easily documented. From approximately 8 percent of all housing additions in 1960, mobile homes comprised some 22 percent in 1974. This rapid increase has been attributed to a variety of factors Including the cost of single family housing, improvements in mobile homes, smaller family size, and changing housing preference. Regardless of the rea­ sons, the more widespread use of mobile housing presents challenges to those concerned with housing and the growth of metropolitan areas. One of the challenges concerns the topic of this research--mobile home development location. Development construction has not kept pace with mobile home manufactur­ ing, and presently about one-half of all mobile homes are located on individual lots. However, within metropolitan areas and increasingly elsewhere, the normal setting for mobile homes is in mobile home developments. Development construction has been most rapid in recent times. For example, in Michigan 47 percent of all parks and 67 percent of all spaces have been constructed 117 118 since 1965. Within the Genesee County study area, the fig ures for the same period are 40 percent and 63 percent, respectively. This recent rise in construction has often resulted in inadequate policy, or at least charges of in­ adequate policy, related to mobile home developments. Much of the criticism directed toward public offi­ cials, in regard to development policy, is only partially justified. Experience with mobile home parks has been limited and many of the early parks provided considerable negative experience. Older parks tended to be small, over crowded, and had few amenities considered basic for resi­ dential uses. Densities in parks constructed prior to 1957 in the study area ranged up to 21 and averaged some 14 spaces per acre . Many early parks were also poorly located. Industrial areas, heavy commercial districts, adjacent railroads and salvage yards offer few positive residential features. The poor location of many early parks was no doubt partially due to decisions (or lack of decisions) of pub­ lic officials. There was a tendency to exclude mobile home developments from residential areas. Within the study area, for example, four municipalities containing mobile home parks presently allow parks only in commercial areas. There are, however, other factors influencing park location. The influence public policy, such as zoning, has on the location of mobile home developments may have 119 been overemphasized. For one thing, the role and function parks perform have changed and what are now considered poor locations was a typical result. Early parks, for example, were constructed for a small, highly portable, often transient dwelling. There was less need for the amenities, internal or locational, basic for more permanent housing. M o d e m parks, however, are residential developments designed for manufactured housing that is usually permanent. They are larger with the twenty-four parks in the study area constructed since 1965 averaging 29 developed acres and 204 spaces. Average densities have also decreased to seven homes per acre. These newer parks approximate more the densities of town- houses than the much higher densities of older parks. The curvilinear street design, the variety in home placement, increased open space, and recreational facilities are other examples of internal amenity features in many newer parks. Failure to recognize these changes is one policy criticism that is justified. It is still common to find statements in policy positions which reflect this failure to recognize change: The relatively transient and non-permanent housing characteristics of the modern mobile home require carefully planned site facilities which are designed for the unique group living needs of the mobile home dweller (Tri-County Regional Plan­ ning Commission, no date, 3a). The argument is not with the requirement for "carefully 120 planned site facilities." This should be true for any land development. The question is with "relatively transient and non-permanent characteristics" of mobile housing and the '‘unique" needs of mobile home residents. These fea­ tures were no doubt true in the past, but not necessarily so now. In the present study, locational change In mobile home parks was less apparent than those internal to the park. However, it was apparent that different combinations of site and area factors influenced the location of early and recent parks. Summary of Factors Parks constructed prior to 1957 were located closer to the urban center and to employment opportuni­ ties than those opened after 1965. The majority of modern mobile home developments were within reasonable commuting distance, but the range extended to over twenty miles. Proximity to highways, however, remained an important factor with only slight differences between older and newer developments. One-half (twelve) of the newer parks were adjacent to highways; six of these were adjacent to expressways. Availability of undeveloped land was a strong in­ fluence in park location. With the relatively large per- cels necessary for recent parks this was an expected char­ acteristic and all but three of the never developments 121 were constructed where surrounding undeveloped land ex­ ceeded 40 percent. This factor, however, should be ex­ amined in combination with other measures. The most dis­ tant and most rural portions of the study area were not the areas of greatest park construction. Instead parks were located in the less developed portions of urban and suburban townships in moderate proximity to the City of Flint. Recent mobile home developments were influenced by availability of undeveloped land, but proximity to the urban center and particularly automobile transportation were also of importance. Several measures of the immediate environment in­ dicated changes in park location. For example, the trend was for newer parks to be constructed in areas of moderate socioeconomic prestige levels. Mobile home developments cater to moderate income households so this characteristic is important primarily because of the contrast to the areas where older parks were located. The trend is clearly toward location in areas where the general socioeconomic level is higher. The surrounding developed land for the majority of parks was predominantly residential. Few of the recent parks were in areas where the surrounding nonresidential land use exceeded 5 percent and only one where it exceeded 19 percent. This general residential location was further indicated when land use adjacent to the park boundary was examined. Only eight recent parks had mixed adjacent lane 122 use compared to seventeen (over 50 percent) of the early developments. The trend is away from location In areas of mixed land use. Considerable concern has been expressed about ad' verse effects parks may have on the land use surrounding them. Since this concern is seldom based on actual mea­ surement , it was thought appropriate to determine actual changes that had occurred surrounding parks opened during the early phase of construction in the study area. It was possible to measure land use changes surrounding thirty of the thirty-six mobile home parks in existence at the end of 1957. This was accomplished by comparing measurements taken from aerial photography for 1957 and high altitude RB-57 imagery taken in 1972. Fifteen years is sufficient time to make at least tentative statements. For the thirty pre-1957 parks, an average of 61 percent of the surrounding area had no urban development. Obviously there has been a reduction in this amount during the fifteen years. Overall, there was a median reduction of 17 percent of vacant land. The important question is, however, for what use? The 19 57 aerial photographs revealed that the greatest proportion of developed land was devoted to resi­ dential uses (excluding mobile home parks). The breakdown is as follows: 1) Mobile home developments, median--2 percent 2) Residential (single and multiple family), 123 median--16 percent 3) Nonresidential, median--8 percent. The remaining proportion was lakes and reservoirs in the immediate area of seven parks and ranged to a high of 27 percent for one park and highways and railyards which ranged to a high of 22 percent. The greatest conversion in surrounding land use was for residential purposes. During the fifteen years, the median increase was 11 percent but the range was consider­ able . For one park there was a 3 percent reduction and the greatest increase was 51 percent. For only four parks was there no change in surrounding residential uses. For nine of the parks there was no increase in the amount of surrounding industrial and commercial uses. However, the figure ranged to a high of 25 percent. Much of the total increase was accounted for by two parks. One park located south of Flint in Grand Blanc Township on M-54 had an increase of 21 percent in primarily coimnercial uses. The other development, situated in an industrial area ad­ jacent to a railyard, the intersection of two interstates, and immediately north of Bishop Airport, registered an in­ crease of 25 percent. This park was also one of two that registered a decrease in surrounding residential uses. The location of parks is sometimes criticized b e ­ cause of adjacent railroads and f r e e w a y s . For seven of the older parks in Genesee County, freeways were constructed through the immediate area after the parks were opened. 124 Even though only seven parks were affected, this provided an overall median increase in land devoted to highway de­ velopment of 2 percent. Due to both expansion of existing parks and con­ struction of new ones, there was a median increase of 1 percent in mobile home park land uses surrounding the pre- 1957 parks. Eighteen recorded some Increase ranging to a high of 6 percent. For eight of the parks this increase was due to construction of new parks. When land use changes were compared on the basis of municipal type, only slight differences were noted in most categories. Increases in residential uses were rather uniform. Increases in mobile home uses also varied little from the overall median but the greatest change occurred around the parks located in Flint. This was due mainly to expansion of existing parks and the construction of two new ones in the Kersley Reservoir vicinity In the northeast part of the city. Greater differences between municipal types, how­ ever, were recorded for cocmnercial-industrial, highway, and decrease in vacant land. The urban townships and one small city registered the greatest changes in all categories ex­ cept adjacent mobile home developments. These were also the only municipalities recording increases in highway development. Additionally, in these six jurisdictions were found the greatest increase in comnercia1-industrial uses and the greatest decrease in vacant land surrounding their fourteen parks. 125 The changes that have occurred around pre-19 57 mobile home developments have several significant implica­ tions. First, the fact that the median increase in indus­ trial-commercial uses was only 2 percent should reduce fears that mobile home parks necessarily adversely affect adjacent land uses. Much of the increase was accounted for by parks located in nonresidential areas at the be gin­ ning of the period. This is particularly true for the sixteen developments located adjacent or near M-59. When the one park located at the intersection of M-7B (now 1-69) and 1-75 is adJed to these, the increases in c o m ­ mercial- industrial uses for the remaining thirteen parks are small by comparison. Secondly, the greatest conversion of undeveloped land was for residential p u r p o s e s . This accounted for approximately 61 percent of the converted land and was some 7 0 percent when increases in mobile home land use were included. Obviously, residential uses account for most of the land conversion in metropolitan a r e a s . Of importance is that the figure was also high around mobile home p a r k s . The discussion of actual changes that have oc­ curred around the thirty pre-1957 parks during the fifteen year period indicates that mobile home parks are not necessarily incompatible with other forms of residential development. Much depended on the type of area where the 126 park was initially located. The greatest increases In non- res idential uses were recorded around those parks initially located in nonresIdential areas. Related to this was the measure of features de­ tracting from the residential character of areas where developments were located. For the study area at least, surrounding negative features were not as pronounced as the literature would lead one to believe. Over 73 percent of all parks were In areas with a negative score of 2 or less. The highest negative score for newer parks was h compared to 7 for early developments. The greatest con­ tributing factor for negative scores for newer parks were adjacent freeways. One area environment factor that had very little influence on park location was proximity to elementary schools. Based on past experience, this was expected. Mobile home households tend to have few school-age chil­ dren. However, the lack of influence of school proximity on park location should be a matter of concern. Assuming that the cost of single family housing continues to in­ crease at a rate higher than mobile housing and that mobile homes will increase in size and number of conve­ niences, it is possible that a greater proportion of families will remain in or purchase mobile homes than has been true in the past. The implication, of course, is that even if school proximity does not become an important factor for development location, developments will have a 127 greater Impact on the school districts in which they are found. The impact will not only be due to increased en­ rollments but also to higher transportation costs. Of the institutional factors measured in this re­ search, availability of municipal sewers was of consider­ able importance. Approximately 77 percent of all parks were connected to public sewer systems. On the other hand, less than 50 percent of the developments within the study area were connected to municipal water systems. The order of importance of these two factors was reversed in the study area compared to one national survey of high quality parks. Zoning, the institutional factor so often empha­ sized in the literature, was inconclusive as a factor of location in this research. Taken as a whole, parks were found in sites zoned residential, specifically for mobile home parks, commercial, industrial, and agricultural. But, as with other factors, there were apparent trends. The differences between older and newer parks were not so much in surrounding residential zoning, where some 72 percent of all parks were located in areas of at least 20 percent of the immediate area was zoned residential, as with specific site and adjacent zoning. For example, more than four-fifths of recent parks were located at sites zoned residential or specifically for mobile home parks. In terms of adjacent zoning, mixed zoning was more of a deterrent to recent developments than older ones. One-half 1 2 8 of the newer parks had no mixed adjacent zoning compared to fewer than one-quarter of the early developments. It is apparent from this research that zoning pro­ tection was not the important factor of location for mobile home developments as had been indicated for single family subdivisions. But it was also apparent that park location was more than a simple "either-or" proposition of obtaining zoning variance just anywhere. The trend is toward location in areas which are not only predominantly residential in terms of land use but also those which have residential zoning. There is no question that mobile home parks are highly restricted compared to other residential develop­ ment. Literature for the nation as a whole and regula­ tions of municipalities in the study area reveal that special use permits or some form of variance is normally required regardless of the type of districts in which parks are allcwed. The developer who must appear before zoning boards, municipal governing bodies, and courts to obtain zoning permits would logically consider this an important factor. Using survey results only, however, leaves out important information pertaining to land zoning of the site and area where the development was located. It was the intention in this research to examine this institu­ tional factor in terms of the specific site and area zon­ ing. The important finding revealed in this study was 129 that mobile home parks have been located at sites and in areas predominantly residential and that this is of par­ ticular importance to newer p a r k s . Suggestions for Future Investigations This research concerned the location of an in­ creasingly important form of residential land use. The primary focus was the enumeration and examination of a set of site and area factors thought to Influence the intra­ metropolitan location of mobile home developments. Mobile home park location has often been examined in rather sim­ plistic terms. The intent of the approach used in this investigation was to recognize and incorporate a certain degree of complexity by utilizing a large number of fac­ tors including those of accessibility, land availability, area environment, and institutional. The approach also provided a method of identifying change in the way these factors influence park location in different periods of construction. While the combination of site and area factors influencing early and recent parks did not vary to the extent expected, there were nonetheless indications of change. For example, more recently c o n ­ structed parks were more influenced by characteristics of the area environment as indicated by area socioeconomic levels, adjacent and area land use, and the absence of nearby negative features. Institutional factors, on the other hand, proved less conclusive. There was evidence, however, that newer parks were more influenced by such fac­ tors as site and area residential zoning and availability of municipal services. A major implication centers on evidence that mobile home park location is becoming less unique. There has been a convergence toward similar site and area factors influ­ encing other forms of residential land use. Mobile home developments are evolving. By incorporating this transi­ tion into analysis, the spatial understanding of this resi­ dential form is aided. The generalizations presented in this research are limited, for it was based primarily on sixty mobile home developments in one metropolitan county in the upper- Midwest. It is hoped that these findings will generate additional research in other areas. A considerable amount of controversy has accompanied the more widespread con­ struction of mobile home parks . Additional research and understanding may provide less controversial decisions in the future. Finally, it is hoped that this study has added to the spatial understanding of mobile home parks. The intent throughout was the application of geographic inquiry on a problem of present concern. In this regard, the present author shares the opinion expressed in a recent publication The translation of basic research into usable material for public and private development and decision making is an important responsibility of the academic community (Horton, 1973: i i i ) . APPENDICES APPENDIX A United States Data Table A1 - Mobile Home Data by State No. of Mobile Homes 1960 9,932 23,243 4,880 101,601 12,979 6,456 3,569 65,087 12,689 6,763 32,470 27,994 11,555 11,783 10,076 9,445 6,180 Rank 1960 25 10 43 1 16 37 45 2 18 36 5 9 20 19 24 28 39 No. of Mobile Homes 1970 51,407 52,247 29,666 197,358 31,147 9,605 9,001 172,100 76,435 15,939 73,757 67,983 24,285 26,690 43,291 38,305 16,250 Mobile Home Quotient 1970 Rank 1970 Rank 1970 13 12 25 1 22 44 47 2 8 36 10 11 30 29 19 20 35 1.51 2.94 1.44 .92 1.37 .32 1.68 2.26 1.70 2.18 .39 1.29 .83 1.10 1.33 1.09 1.56 18 2 20 34 22 45 13 5 12 6 44 24 36 28 23 29 17 Table Al - Continued No. of Mobile Homes 1960 9,521 6,766 29,400 10,782 6,327 16,613 7,077 7,155 8,026 2,896 9,156 12,937 31,306 19,133 5,017 42,892 8,086 14,090 31,434 1,513 11,072 Rank 1960 27 35 8 23 38 13 33 32 31 46 29 17 7 11 42 3 30 15 6 48 21 No. of Mobile Homes 1970 20,343 10,928 75,012 29,740 30,581 50,878 16,935 14,838 20,520 12,621 15,025 18,911 77,560 98,474 9,645 85,824 27,600 37,801 87,571 2,338 50,211 Mobile Home Quotient 1970 Rank 1970 Rank 1970 32 41 9 24 23 14 34 38 31 39 37 33 7 3 43 6 27 21 5 48 16 .54 .19 .86 .79 1.44 .99 2.30 ,95 3.91 1.66 .21 1.91 .41 1.99 1.57 .81 .96 1.68 .74 .25 2.04 42 48 35 39 21 30 4 33 1 15 47 10 43 9 16 37 32 14 40 46 8 18, Maryland 19. Massachusetts 20. Michigan 21. Minnesota 22. Mississippi 23. Missouri 24. Montana 25. Nebraska 26, Nevada 27. New Hampshire 28. New Jersey 29. New Mexico 30. New York 31, North Carolina 32. North Dakota 33. Ohio 34. Oklahoma 35. Oregon 36. Pennsylvania 37. Rhode Island 38. South Carolina Table Al - Continued No. of Mobile Hones 1960 6,929 9,792 36,878 4,858 2,335 1.7,257 14,940 5,245 11,064 6,165 Rank 1960 34 26 4 44 47 12 14 41 22 40 No. of Mobile Homes 1970 11,637 48,418 94,687 9,189 9,364 50,421 43,978 27,123 28,474 10,256 Mobile Hone Quotient 1970 Rank 1970 40 17 4 46 44 15 18 28 26 42 1.72 1.22 .81 .96 2.04 1.11 1.19 1.49 .66 2.92 Rank 1970 11 25 38 31 7 27 26 19 41 3 39. South Dakota 40. Tennessee 41. Texas 42. Utah 43. Vermont 44. Virginia 45. Washington 46, West Virginia 47. Wisconsin 48. Wyoming Source: U.S. Bureau of the Census, 1962b; 1972b. Table A2 - Mobile Home Development and Mobile Home Household Data by State Rank No, of Spaces Rank Develop- ment Quotient Rank % Age 65 and Over Rank % in SMSA's Rank Md, Income Alabama Arizona Arkansas California Colorado Connecticut Delaware Florida Georgia Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan 25 5 33 1 19 35 37 2 15 30 7 11 17 24 28 29 43 27 32 3 40 4 37 1 23 9 8 2 38 22 18 29 12 27 41 39 36 28 10 6 41 3 21 2 28 11 42 1 45 13 22 26 16 37 33 43 14 39 4 15 25 4 34 1 17 3 33 16 35 42 12 20 28 29 39 27 46 11 7 13 42 36 47 16 20 2 20 48 34 20 5 7 13 20 44 27 36 U 7 3 4 34 2 39 29 47 26 20 3 31 32 28 36 33 10 11 21 35 45 25 Table A2 * Continued Rank No. of Spaces Rank Develop- ment Quotient Rank 7, Age 65 and Over Rank X in SMSA's Rank Md. Income Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee 13 42 14 40 34 21 38 23 36 9 20 41 4 22 12 8 46 26 47 31 7 44 26 33 24 11 20 5 31 30 47 19 17 21 15 32 3 42 45 46 24 38 20 35 23 27 8 7 29 9 46 47 10 34 5 19 25 48 18 36 24 40 23 41 36 2 43 5 38 18 30 44 9 21 19 10 8 22 45 31 7 45 32 14 20 1 19 7 30 14 39 29 11 34 20 20 32 39 43 46 18 7 14 30 37 24 9 44 48 27 5 40 38 13 19 17 46 8 42 6 Table A 2 - Continued Rank No, of Spaces Rank Develop­ ment Quotient Rank % Age 65 and Over Rank % in SMSA's Rank Md. Income Rank % Change 1960- 1970 41, Texas 42. Utah 43. Vermont 44. Virginia 43. Washington 46. West Virginia 47. Wisconsin 48. Wyoming 6 39 45 18 10 44 16 48 25 14 35 34 13 48 16 43 32 30 17 44 6 31 12 40 6 15 47 26 14 37 32 47 16 30 27 39 5 36 16 3 23 41 12 16 15 1 22 43 Source: Woodall's, 1971; U.S. Bureau of the Census, 1973b. 137 APPENDIX B Michigan Data Table Bl - Michigan Mobile Home Data by County Number Mobile Homes 1970 Mobile Homes as a % of Housing Mobile Home Quotient 1 . Alcona Alger 2. Allegan 3. Alpena 4. Antrim 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Arenac Baraga Barry Bay Benzie Berrien Branch Calhoun Cass Charlevoix Cheboygan Chippewa Clare Clinton Crawford Delta Dickinson Eaton Emmet Genesee Gladwin Gogebic Grand Traverse Gratiot Hillsdale 125 124 1463 457 279 344 103 779 1679 198 1468 684 1378 1051 298 365 481 491 1063 178 386 166 902 385 4175 450 117 1037 809 786 3.48 3.61 6.87 4.62 4.62 8.79 3.55 5.94 4.66 5.55 2.66 5.40 2.96 6.74 4.75 6.06 4.39 4.44 7.55 7.64 3.13 1.92 4.36 5.60 3.09 6 .31 1.50 7.97 6.66 6.41 1.31 1.37 2 .60 1.75 1.75 3.33 1.34 2.25 1.77 2.10 1.01 2 .05 1.12 2.55 1.80 2.30 1.66 1.68 2.86 2.89 1.19 .73 1.65 2.12 1.17 2 .39 .57 3.02 2.52 2.42 138 Table Bl - Continued Number Mobile Homes 1970 Mobile Homes as a X of Housing Mobile Home Quotient 369 471 1980 755 346 171 906 2270 1429 147 3562 17 186 1061 169 1038 1327 133 159 4122 205 925 305 793 264 1355 127 1011 1022 116 1029 809 5295 399 308 2.92 3.93 2 .44 5.45 3.22 2.81 7.46 5.15 2 .30 5.61 2.73 1.74 4.51 7.00 3.34 4.03 7.15 6.62 3.80 2.34 2.62 4.61 3.26 8 .86 2.96 6 .94 4.52 2 .96 7 .32 3.33 2 .09 7 .50 1.93 6.22 3.97 1.11 1.49 .92 2.06 1.22 1.06 2.83 1.95 .87 2 .13 1.03 .66 1.71 2.65 1 .27 1.53 2.71 2.51 1.44 .89 .99 1.75 1.23 3.36 1.12 2.63 1.71 1 .12 2.77 1.26 .79 2 .84 .73 2.36 1.50 31. 32. 33. 34. 33. Houghton Huron Ingham Ionia Iosco Iron Isabella Jackson 36. 37. 38. 39 . Kalamazoo 40. Kalkaska 41. 42. 43. 44. 45. Kent Keweenaw Lake Lapeer Leelanau Lenawee 46. 47 . Livingston 48. Luce 49 . Mackinac 50. Macomb 51. 52. 53. 54. 55. Manistee Marquette Mason Mecosta Menominee 56 . Midland 57 . Messaukee 58. 59 . Montcalm 60. Monroe Montmorency Muskegon 61. 62 . Newaygo 63 . Oakland 64. 65. Oceana Ogemaw 139 Table Bl - Continued Number Mobile Homes 1970 Mobile Homes as a % of Hous ing Mobile Home Quotient 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. Ontonagon Osceola Oscoda Otsego Ottawa Presque Isle Roscommon Saginaw St. Clair S t . Joseph Sanilac Schoolcraft Shiawassee Tuscola Van Buren Washtenaw Wayne Wexford State 208 319 51 359 644 272 221 1553 1449 895 715 70 1094 1096 1708 1341 5804 411 75,012 5.33 5.69 1.64 8.21 1.73 6 .11 2.93 2.37 3.56 5.45 5.73 2.17 5.65 7.21 8.60 1.88 .64 5.60 2.64 Source: IKS. Bureau of the Census, 1972c. 2.02 2.16 .62 3.11 .66 2.31 1.11 .90 1.35 2.06 2.17 .82 2.14 2.73 3.26 .71 .24 2.12 — Table B2 - Michigan Mobile Home Development Data by County No* Spaces 1960 No. Spaces 1970 No. Spaces 1975 Average Size of Parks Development Quotient 1970 1. Alcona 2. Alger 3. Allegan 4. Alpena 5. Antrim 6. Arenac 7, Baraga 8. Barry 9. Bay 10. Benzie 11. Berrien 12. Branch 13. Calhoun 14. Cass 15. Charlevoix 16. Cheboygan 17. Chippewa 18. Clare 19. Clinton 20. Crawford 15 0 85 82 0 0 5 100 198 9 604 95 230 23 19 21 24 26 159 0 15 23 1038 192 14 82 44 357 1481 66 1313 264 932 451 220 117 189 133 949 0 15 32 1432 212 30 125 54 543 2121 73 2454 588 1304 619 220 164 279 172 1107 71 15 16 42 38 30 31 27 54 88 24 85 42 69 48 44 33 35 34 101 36 .16 .25 .95 .56 .06 .32 .57 .61 1.17 .44 1.19 .52 .91 .57 .99 .43 .52 .36 1.19 0 Table B2 - Continued No. Spaces 1960 No. Spaces 1970 No. Spaces 1975 Average Size of Parks Development Quotient 1970 21. Delta 22. Dickinson 23. Eaton 24. Emmet 25. Genesee 26. Gladwin 27. Gogebic 28. Grand Traverse 29. Gratiot 30. Hillsdale 31. Houghton 32. Huron 33. 34. 35, Ingham Ionia Iosco Iron Isabella 36. 37. 38. Jackson 39. Kalamazoo 40. Kalkaska 3 15 95 0 2630 60 0 212 0 39 96 57 736 108 164 0 116 588 550 0 43 15 767 66 5723 199 14 587 180 230 109 153 1775 316 195 0 514 942 1735 25 165 173 953 182 8965 293 72 863 304 302 131 213 2465 444 265 12 789 2039 2839 100 33 43 106 36 145 49 14 86 43 34 22 21 107 32 44 12 72 89 123 50 .15 .12 1.13 .23 1.83 .59 .16 .76 .29 .39 .40 .43 1.20 .56 .75 0 .76 .55 1.61 .23 Table B2 - Continued No. Spaces 1960 No. Spaces 1970 No. Spaces 1975 Average Size of Parks Development Quotient 1970 1388 0 0 38 0 210 373 0 15 2062 60 331 70 75 0 145 0 501 102 0 3462 0 33 192 0 734 696 13 33 4621 60 409 144 242 65 443 0 1630 198 0 5894 0 33 677 58 938 1356 13 40 6270 60 624 318 505 101 763 0 2381 297 0 120 0 33 113 29 61 75 13 20 179 60 39 53 56 50 76 0 132 33 0 X ? * Z 1.29 0 .24 .24 0 .95 .69 .13 .28 1.49 .39 .59 .63 .41 .33 .44 0 2.15 .25 0 41. Kent 42. Keweenaw 43. Lake 44. Lapeer 45. Leelanau Lenawee 46. 47. Livingston 48. Luce 49. Mackinac 50. Macomb 51. Manistee 52. Marquette 53. Mason 54. Mecosta 55. Menominee 56. Midland 57. Messaukee 58. Monroe 59. Montcalm 60. Montmorency Table B2 - Continued No. Spaces 1960 No. Spaces 1970 No. Spaces 1975 Average Size of Parks Development Quotient 1970 61, Muskegon 62. Newaygo 63. Oakland 64. Oceana 65. Ogemaw 66. Ontonagon 67. Osceola 68. Oscoda 69. Otsego 70. Ottawa 71. Presque Isle 72. Roscommon 73. Saginaw 74, St. Clair 75. St. Joseph 76. Sanilac 77. Schoolcraft 78. Shiawassee 79. Tuscola 80. Van Buren 547 0 2569 25 15 0 0 9 27 224 56 138 183 371 116 105 28 233 62 62 926 249 6435 96 28 45 42 9 49 672 94 255 1014 1222 400 515 28 672 273 626 1462 383 10454 194 105 59 42 9 111 1655 104 288 2357 2084 676 630 39 894 636 966 104 43 183 49 35 59 14 9 22 92 35 29 139 91 42 63 20 75 58 54 1.20 .41 1.63 .32 .12 .29 .16 .24 .19 1.39 .47 1.53 .87 1.12 .60 .96 .53 .81 .33 .49 Table B2 - Continued No. Spaces 1960 No. Spaces 1970 No. Spaces 1975 Average Size of Parks Development Quotient 1970 81. Washtenaw 82. Wayne 83. Wexford 815 3737 21 1146 6622 210 1616 8640 300 State 21,877 56,071 87,259 1.13 1.52 .68 95 157 100 91 Source: Michigan Department of Health, 1975. 145 APPENDIX C Zoning Ordinances and Maps Consulted Burton Township - 1965 Burton (City) - 1974 Clayton Township - 1967 Davison (City) - 1971 Davison Township - 1952 and 1972 Fenton Township - 1953 (Revised 1962) and 19 71 Flint (City) - 1927 and 1968 Flint Township - 1950 and 1971 Flushing Township - 1966 Genesee Township - 1955 and 1975 Grand Blanc Township - 1954 and 1964 Montrose Township - 1955 Mt. Morris Township - 1955 and 1972 Thetford Township - 1954 Vienna Township - 1955 LIST OF REFERENCES List of References Abler, Ronald, Adams, John S., and Gould, Peter. 1971. Spatial Organization: TheGeographer's View of the World. Englewood Cliffs, New Jersey: Prentice- Ha11, Inc. Abu-Ayyash, A. Y. 1972. "The Mobile Home: A Neglected Phenomenon in Geographic Research." The Geographi­ cal Bulletin. 5:28-30. Babcock, Richard F- 1966. The Zoning Game: Municipal Practices and Policies. Madison: University of Wisconsin Press. Bair, Frederick H., Jr. 1965. Mobile Homes and the General Housing Su p p l y . Chicago: Mobile Homes Manufacturers A s s o c i at io n. Baker, Earl J., and Lee, Yuk. 1975. "Alternative Analy­ sis of Geographical Contingency Tables." Profes­ sional Geographer. 27:179-188. Banks, Price T. 1972. "Zoning Policy and its Relation­ ship to Urban Sprawl in the Flint Metropolitan Area." Master's thesis, East Lansing: Michigan State University. Bartley, Ernest R., and Bair, Frederick H . , Jr. 1960. Mobile Home Parks and Comprehensive Community Planning. Gainsville: University of Florida. Bourne, L. S. 1968. "Market Location and Site Selection in Apartment Construction." Canadian Geographer, 12:211-226. ------------ Bourne, L. S. 1971. "Apartment Location and the Housing Market," in Internal Structure of the City: Read­ ings on Space and Environment J 1971. Larry S . Bourne, ed. New York: Oxford University Press, 321-328. Business Week. 1970. "Mobile Home Sales Roll Toward $3 Billion." January, 24:73-76. 146 147 Cheek, William H. 1967. "A Geographic Analysis of the United States Mobile Home Industry with Special Emphasis on Arizona.’1 Master's thesis, Tempe: Arizona State University. Clawson, Marion. 1971. Suburban Land Conversion in the United States : An Economic and Gwernmental Pro­ cess . Resources for the Future, Inc. Baltimore: TheJohns Hopkins Press. Cloos, George W., and Birgells, Edward W . , Jr. 1972. "Mobile Homes and the Housing Supply." Business Conditions, Federal Reserve Bank of Chicago, November: 2-16 . Consumers Power Company. 196 5. Data on Flint, Michigan. Area Development Department: Jackson, Michigan. Cowgill, Donald 0. 1941. Mobile Homes: A Study of Trailer Life. washington, D .C .: American Coun­ cil on Public Affairs. Detroit Region Transportation and Land Use Study. 1969. A Survey of Mobile Homes in the Detroit TALUS Region Detroit, Michigan. Drury, Margaret J. 1972 (Revised Edition). Mobile Homes The Unrecognized Revolution in American Houslng. New York: Praeger (Publishers. Edwards, C. M. 1965. A Survey of the Mobile Home Con­ -965. A Survey of the Mobile Home Chicago: Trailer topics Magazine. sumers . El Gammal, Tarouk M. 1966. "Theory and Reality in the Use of Land for Residential Purposes: The Case of the Mobile Home Parks in Massachusetts." Ph.D. dissertation, Worcester, Massachusetts: Clark University. Elias, Erwin. 1974. "Significant Developments and Trends in Zoning Litigation," in Land Use Controls: Pres ent Problems and Future Reform. David LIstokin, e d . New Brunswick, New Jersey: Center for Urban Policy Research, Rutgers University. Federal Housing Administration. 1970. Mobile Home Court Development Guide. Washington, D . C . : tJ. S . Govern­ ment Printing Office. Field, Thomas P. 1972. Mobile Homes of Kentucky and the Lexington Hexagon: A Study in Areal Distribution. Lexington, Kentucky: Fayette County Geographical Society. 148 Freed, Michael D. 1969. "An Analysis of the Municipal Role in Mobile Home Land Development." Master's thesis, East Lansing: Michigan State University. French, R. M., and Hadden, J. K. 1965* "An Analysis of the Distribution of Mobile Homes in the United States." Land Economics. 41:131-329. Fuller, Richard C., and Meyers, Richard R. 1947. "The Natural History of a Social Problem." American Sociological Review. 6:320-329. Galetschky, Henry. 1974. Mobile Homes: Challenges for Today and Tomorrow. Missouri Local Government Administrative Guide Series. Columbia: University of Missouri--Columbia. Genesee County Metropolitan Planning Comnlsslon. 1970a. Genesee County Educational Facilities: Conclusion “ Report ‘ Flint, Michigan. Genesee County Metropolitan Planning Conmisslon. 1970b. Genesee County Land Use: Conclusion Report. F l in t, Mic h l g a n . Genesee County Metropolitan Planning Commission. 1970c. Genesee County Transportation Facilities: Con­ clusion fteportl Flint, Michigan. Genesee County Metropolitan Planning Commission. 1971. Genesee County 1990 Land Use-Transportation Plan. Flint, Michigan. ~ Gibson, Constance B. 1972. Policy Alternatives for Mobile Homes. New Brunswick” New Jersey: Center for Urban Policy Research, Rutgers University. Goldblatt, Abraham, and Pitcher, Charles B. 1972. "Mobile Homes--A Growing Force in the Housing Sector." Construction Review. U.S. Department of Commerce, Washington, b.C.: U.S. Government Printing Office. Horton, Frank E. (ed.). 1973. Geographical Perspectives and Urban Problems. Washington, D .C .: National Academy of Science. Huntoon, Maxwell C . , Jr., and Kirk, John. 1971. "It's Time to Take the Low-Price Market Back from the Mobiles." House and Home, April: 62-71. Jacobson, Charles E. 1965. "The Development of a Mobile Home Park." The Appraisal Journal, July: 347-350. 149 Kaiser, Edward J. 1966. Toward a Model of Residential Developer Locational Behavior. Thesis Series, C h ap e1 H i 11, Worth C a r o l i n a : Center for Urban and Regional Studies, University of North Carolina. Kaiser, Edward J. 1968. ,rLocational Decision Factors In a Producer Model of Residential Development." Land Economics. 44:351-362. Kaiser, Edward J., and Weiss, Shirly F. 1969. "Sane Com­ ponents of a Linked Model for the Residential De- velopment Decision Model." Proceedings of the AAG. 1:75-79. Knight, Robert L. 1971. "Planners and Mobile Home Re­ search." Socio-Economic Planning Sciences. 5: 213-220. Lemert, Edwin M. 1951. "Is There a Nautral History of Social Problems?" American Sociological Review. 16:217-223. Macomb County Planning Comnisslon. 1969. Mobile Home Parks. Mount Clemens, Michigan. Maricopa County Planning and Zoning Department. 1963. A Report on Mobile Home Parks and Subdivisions. Mioenix, Arizona: Maricopa County Board of Super- visors. Michigan Department of Health. 1971. Michigan's Mobile Home Park Law. Lansing: Michigan Department of Hea1t h . Michigan Department of Health. 1975. Lansing, Michigan: (Files). Mobile and Recreational Housing Merchandiser. 1974. M o b i l e Home Statistics." March: 15. Newcomb, Robinson. 1971. Mobile Home Parks: Part 1, An Analysis of Characterlstlc5~^ Washington, D.C. : l*He Urban Land Institute. Nie, Norman H., Bent, Dale H., and Hull, C. Hadlal. 19 70. Statistical Package for the Social Sciences. New Vork : McGraw-Hill Book Company J Nulsen, David R., and Nulsen, Robert H. 1971. Mobile Home and Recreation Vehicle Park Management (Two V o l s .). Beverly Hills, California: Trail-R-Club of America. 150 Powell, William E. 1975. "Residences on Wheels: A Geo­ graphical Analysis of Mobile Homes in America." Ecimene. 7:25-34. Scheff, S. Leonard, and Bleck, Erich H. 1973. "Why the Apartment House is a Failing Competitor." Arizona Review. Tucson: University of Arizona. Swaback, Vernon D. 1971. "Production Dwellings: An O p ­ portunity for Excellence." Land Economics. 47: 321-338. Tri-County Regional Planning Commission. 1971. An Ap­ proach to Regional Area Identification and Factors influencing Housing for Lower Income Families. Lansing, Michigan. Tri-County Regional Planning Conmisslon. (No Date). Zon­ ing Guide A i d : Mobile Home Park Provisions. Lana - lng, Michigan. Wehrly, Max S. 1972. Mobile Home Parks: Part 2. An Analysis of Connunlties. Washington, D.C.: The Urban Land Institute. Wheeler, James 0., Callaghan, Gale, and Brewer, Gordon. 1973. "Locational Factors in the Growth of Mobile Home Manufacturing in the Southeastern United States." Southeastern Geographer. 13:92-104. Woodall's Mobile Home and Park Directory. 1972 Edition. I97T: highland ParE, Illinois: tioodall'a~Pub- lishlng Company. U.S. Bureau of the Census, 1953. Detailed Housing Char­ acteristics . 1950 Census of Housing, Washington, D.C.: U.S. Government Printing Office. U.S. Bureau of the Census. 1962a. Census Tracts, Flint Michigan SMSA. 1960 Census of Population and Housing, Washington, D.C.: U.S. Government Print­ ing Office. U.S. Bureau of the Census. 1962b. Detailed Housing Char­ acteristics , 1960 Census of Housing, Washington^ D . G . : U.S.Government Printing Office. U.S. Bureau of the Census. 1967. Retail Trade: Major Retail Centers. 1967 Census of Business, Michi­ gan, Washington, D.C.: U.S. Government Printing Office.