COMMUNITY DECISMN MAKERS’ ATTETUDES TGWARD AGENCY PROGPAM EFFECTIVENESS Dissertation for the Degree of Ph. D. MICHIGAN 3mm UNIVERSITY mom JAMES KUBiAK 1973 My“ #2 00+. Jan 1 1 @883»: 80 K011 WW w ,« at \‘027 {E I E fl _, I It". '3... v3 EEBMM ABSTRACT COMMUNITY DECISION MAKERS' ATTITUDES TOWARD PROGRAM EFFECTIVENESS BY Timothy James Kubiak Little is known of the relationship between the success of a deve10pment program and the attitudes toward that program held by the participating local decision makers. Program achievement may be a function of local decision makers attitudes toward program effectiveness. Negative attitudes may hinder actual program outcomes while positive attitudes may, in fact, lead to or promote program success. Moreover, to determine and explain the attitude forming process based upon antecedent socio- economic characteristics, roles, membership and other beliefs and opinions would be of value to agency performance in program implementation. Attitudes are considered as one measure of program outcome. Other more precise measures are recognized but not used within the sc0pe of this study. As part of a larger evaluation project sponsored by the Economic Research Service (ERS) of the U.S. Department of Agriculture, this study has as its principle objective to determine the overall relationship between various - socio-economic characteristics and positive attitudes toward Timothy James Kubiak program effectiveness. Dealing specifically with the Northwest Michigan Resource Conservation and Deve10pment Program (RC&D) the attempt was made to quantify the socio- economic indicators of local decision makers within the thirteen county project region and find causal relationships between antecedent variables and attitudes. To accomplish those objectives the data used in this study was provided by the Economic Research Service. The ERS gathered the data through individual personal inter- views with the seventy-two county RC&D leaders in the Autumn of 1971. These local decision makers were members of county steering committees and represented each of the thirteen counties within the region. The results of the questionnaire survey yielded a total of three-hundred and sixty-eight variables that were used in further analysis. It was hypothesized, based upon a review of pertinent literature, that there is a positive relationship between socio-economic characteristics of the decision makers or resPondents, their role relations, reference groups, their environmental perceptions and positive attitudes toward program effectiveness. In order to test the hypothesis several statistical techniques were used. To reduce the number of variables to meaningful size simple cross-tabulation and correlation analysis were performed upon all the data. From this, significant variables were chosen and further analyzed as to their relationship to positive attitudes by the method Timothy James Kubiak of canonical correlation. Canonical correlation involves the maximum correlation of sets of variables. One set, the dependent variables, consisting of seventeen closely related attitudes representing opinions of program effectiveness, was correlated to five different sets of independent variables representing socio-economic and other characteristics and Opinions. A sixth independent set of variables was chosen from the results of the first five canonical correlations and computed again with variables in the constant dependent set. As a check on the method of variable selection for canonical correlation A analysis, factor analysis was employed as a means of isolating significant variables from the total of three- hundred and sixty-eight variables. The results of factor analyzing the variables were then used as an independent set of variables of a final canonical correlation. The results of both methods were similar. The method of canonical correlation analysis provided insight into the complex concept of attitudes toward program effectiveness by yielding results in weighted coefficients or measures of association between variables and sets of variables. From the results, equations relating positive attitudes toward program effectiveness and the variables representing socio- economic and other characteristic indicators were constructed. The models suggests that positive attitudes toward program effectiveness held by the steering committee decision makers in the Northwest Michigan RC&D Program are most closely Timothy James Kubiak associated with the characteristics of length of: residence within the State (the longer the period of residence; the likelihood of positive Opinions increases; the occupation of farming; roles related to the Soil Conservation Districts; lower than average educational achievement (did not complete high school): higher than average income (over $11,000); and, membership in organizations focusing upon transportation and land related problems. The study calls for further research to test the model construction through the use of canonical analysis. Further it is suggested that additional research correlate actual program outcomes to attitudes, perceptions and problem recognition held by the decision makers in order to test‘the validity of the decision makers' attitudes in relation to program achievement. COMMUNITY DECISION MAKERS' ATTITUDES TOWARD AGENCY PROGRAM EFFECTIVENESS BY Timothy James Kubiak A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Resource Deve10pment 1973 F ACKNOWLEDGMENTS W Major gratitude is extended to: my major professor and graduate advisor, Professor William J. Kimball, for his continued advice and counsel throughout the preparation of this report and my period of residence at Michigan State; Professors Milton Steinmueller and Manfred Thullen for their advice and suggestions; and Professor Stanley Brunn for his continued counsel concerning the many aspects of. methods of analysis used in this study. Special appreciation is extended to the Economic Research Service, U.S. Department of Agriculture, both for ' financial support and interest to make this study possible. Very special acknowledgment is due to Mr. David Carvey, Project Leader of the RC&D evaluation project, ERS, USDA, for his suggestions and constructive criticism throughout the entire study.w Further thanks are given to Mr. John Putman, Leader, North Central Resource Group, ERS, and Ms. Priscilla Prophet, Computer Programmer, ERS, for their aid. To all those peOple who provided ideas, criticisms and thought to directly influence the study I offer my thanks. And certainly, my most sincere appreciation is extended to an understanding and encouraging women - my wife, Sandy. ii TABLE OF CONTENTS CHAPTER PAGE I. IntroauctionCOOOOOOOOOOOOOOQOOOOOOO... 1 Problem Statement................... 1 Problem Setting and Historical . PerspectiveOOOOOCOOOOOOOOOOOOOOOO. 9 Purpose of This Study............... 24 II. Literature Review..................... 26 III. Research Procedures.......... ......... 34 ObjectiveSOOOOOOO000......0.0.0.0... 34 Data COllectionOOOA.IOOOOOOOOIOOOOOOO 35 HypotheseSooooooococo-coco...coco-o. 38 Methods of Statistical Analysis..... 39 Factor Analysis With Canonical MaIYSiSOOIOOOOOOOOOOOOOOOOOOOOOOO 47 IV. Results.......................... ..... 51 The Preliminary Results............. 51 Socio-Economic Characteristics.... 52 Attitudes and Perceptions......... 60 Summary of Preliminary Findings... 57 Results of Statistical Analysis..... 69 Results of canonical Analysis..... 72 Summary of Sixth Canonical Analysis........................ 97 Factor Analysis With Canonical Analysis........................ 97 smarYOOOOOQOQooooooooooooooooooooo 101 V. Results of Analysis In Equation Form.. 103 Models for Positive Attitudes Toward Program Effectiveness...... 105 VI. Summary, Conclusions and Recommendations.............. ....... 111 Summary........... .......... .,...... 111 conCluSionS.OOOOOOOOOOOOOOOOOOOOOOOO 114 Recommendations ..... ........... ..... 115 iii Page Bibliography................................ 120 AppendiceSOOOOOOOOIOOOOOOOOOOOOOO0.0... ..... 124 A. Questionnaireooooooooocooooooooooooooo 124 B. List of Variables..................... 133 C. Cross Tabulation of A11 Variables Used In Subsequent Analysis......... 142 D. Correlation Matrix of Positive' Opinions of Program Effectiveness and Predictor Variables............. 145 iv TABLE 2. 3. 4. 5. 6. 7. 10. 11. 12. 13. 14. 15. 16. 17. 18. LIST OF TABLES PAGE Socio-Economic Characteristics of the Respondents..0.0.0.000...0.00.00.00.00 53 Respondent Membership................... 55 Respondents' Initial RC&D Contact....... 56 Influential Organizations by Category... 58 Key Leader Identification and RC&D Involvement By Occupation....... ..... . 59 Key Leader Participation By Roles....... 60 Respondents' Opinions of RC&D Effectiveness......................... 61 Respondents' Attitudes Toward Program Contribution To Human and Natural Resource Problems..................... 63 Respondents' Attitudes Toward Program Contribution To Economic Development.. 64 Respondents' Perception of Economic conditionSOOOOOCQIOOOOOOOOO ..... O ..... 65 Respondents' Perception of Natural and Social Environmental Quality.......... 66 Respondents' Opinions of Community Awareness of the RC&D Program.... ..... 66 Respondents' PrOpensity to Change the RC&D ProgramOOOOOOOOOOOOOO0.... ....... 67 Correlation Matrix of Criterion variableSOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 73 Results of First Canonical Analysis..... 76 Results of Second Canonical Analysis.... 77 Results of Third Canonical Analysis ..... 79 Results of Fourth Canonical Analysis.... 31 V Page 19. Results of Fifth Canonical Analysis...... 82 20. Results of Sixth Canonical Analysis...... 85 21. Results of Canonical Analysis Using the Results of Factor Analysis as Predictor Set.......................... 99 vi LIST OF FIGURES FIGURE PAGE 1. Predictor and Criterion Sets Used In canonical AnalYSiSOOOIOOOOOOOOOOO0.0. 48 vii CHAPTER I INTRODUCTION Problem Statement In the past, local develOpment, growth or decline were affairs primarily of the locale. These matters were endemic in nature and their solution most often rested upon the activities of the community itself. Now, however, development inadequacies are no longer considered to be essentially a local concern. Recently it has been recognized that healthy, viable communities or regions are in the total national interest. James L. Sundquist recognized three phases through which public attitude passes as internal development problems come to public attention. At first, the problem is viewed by the community as theirs alone and not of national 'concern. Later, as the problem persists and local communities or states realize that they are unable to solve the problem, federal aid is proposed. Nevertheless, in this phase the various communities still see the issue as essentially local. In the final stage, the locus of responsibility can shift from local to the national level with all concerned realizing that a solution requires a national effort.1 Since the early 1960's it is not rare that extra- local agencies have made entry into what were originally community issues. Questions arise, however, as to whether these outside agencies are working in the locale or gf_the locale. Several studies have suggested that there is a significant difference between an outside agency in the community or an outside agency of the community. Dasgupta and Wilkinson, when speaking specifically of watershed develOpment, but of equal applicability to all development programs, claim that widespread participation of local residents in program planning, organization and imple- mentation is critical in gaining positive local support and linking development to the social structure. They further claim that such support and linkage are often ignored by technical and planning agencies.2 The problems associated with an agency merely working in the community focusing primarily upon the project with 1James L. Sundquist, Making Federalism Work (Washington: The Brookings Institution, 1969), ll. 2Satadal Dasgupta and Kenneth P. Wilkinson, "Local Participation in Watershed DevelOpment: A Comparative Study of Two Communities," Proceedings of Third Annual American Water Resources Conference (San Francisco, 1967), 396-404. little or no commensurate knowledge of the perceptions and attitudes of local people appear to be extremely important as more and more responsibility is ceded to state and federal agencies. With a development program deeply involved as an integral part of the community, chances of positive association with that community are much greater than in situations in which the program is partly or wholly organized and implemented by outside agencies with a minimum of local participation. Dasgupta and Wilkinson have shown that the consequences of agency solicitude or disregard can spell the success or failure of development programs through local attitudes.1 Positive or negative views of local residents toward the objectives and goals as well as the effectiveness of the program may, in fact, determine actual program outcomes. This study will attempt to critically analyze the attitudes of local participants within an on-going devel- cpment program (The Northwest Michigan Resource Conservation and Development Program). Specifically, the attitudes of [program effectiveness will be studied as a measure of program achievement. The assumption is made that such attitudes toward the effectiveness of the program can be used as indicators of program outcomes but it is also recognized that such an assumption is not based upon an lIbid. independent measure of program achievement. Other more direct measures of program achievement, although not within the scope of this study, could include: the actual number of project achievements as measured by proposal adoption, implementation and completion of specific projects within the overall framework of RC&D: the achievement of stated goals set forth by the Project Steering Committee and the sponsoring agency, and: precise measures of achievement such as the number of new jobs, number of recreation sites added, amount of acreage converted to woodlots and many other measures of similar nature. Attitudes of program effectiveness, however, are the focus of this study and should, at this point, be clearly defined in order to establish a conceptual base of definitional constructs used throughout the remainder of the report. The term "attitudes" can be used interchange- ably with "belief" or "Opinion" to describe some particular preference held by an individual or collectively in a group toward an object, concept or outcome of some action.1 For .the purposes of this study this definition can be further expanded to include the idea that attitudes are a "... system of three components centering about a single object: 1Gilbert F. White, "Formation and Role of Public Attitudes," Environmental Quality in a Growing Economy (Baltimore: The Johns HOpkins Press for Resources For the Future, Inc., 1966), 108. “ the beliefs about the Object - the cognitive component; the affect connected with the object - the feeling component: and the disposition to take action with respect to the Object - the action tendency component."1 Such preferences and beliefs are rooted in core values, the basic values determining the individual's perception of all things. In turn, the individual or group reacts to all things - the environment - in a way commensurate with those perceptions. The result is an environment, program or setting that reflects attitudes and perceptions.‘ In other words, individuals acting collectively in a group create the environment in which they Operate based upon. the components of attitudes. fEffectiveness" can be defined as the state Of being. effective or producing an efficient or decisive effect. Attitudes of program effectiveness, then, are beliefs or Opinions describing the relative efficiency or effect Of the program in terms of its stated Objective. The above definition of attitudes toward program effectiveness is the basis for further consideration Of attitude explanation. To determine and explain the attitudes toward program effectiveness can be of value in program analysis. Explanation of attitudes toward the 1David Krech, Richard Crutchfield and Egerton L. Ballachey, Individual in Society (New York: McGraw—Hill, Inc., 1962), 146. ‘ I effectiveness of a development program can also be of value to agency performance in prOgram implementation. The approach Of determining public attitude and adjusting agency policy to public Opinion is one way of utilizing public Opinion for agency benefit. Such an approach is not new. It has been shown that there has been a long tradition of using survey data in relation to social policy. In fact, the results of attitude surveys are very commonly seen as appropriate when decisions or recommendations are to be made on matters of social policy.1 Attitudes can also play a less obtrusive role in decision making. White said that attitudes enter into- decision making in three ways. There are the personal attitudes of the peOple sharing in the decision. There are Opinions as to what others prefer and there are Opinions as to what others should prefer.2 These three influences upon decision making rarely coincide but it is recognized that there is a tendency for both personal attitudes and normative, i.e., what ought to be, to merge. The idea Of "what ought to be" is not only a component Of attitudes but is Often the concern Of the sponsoring agency in develOpment programs. Such concerns, from the point-Of-view Of this study, involve considerations 1Jennifer Platt, "Survey Data and Social Policy," The British Journal of Sociology, Vol. 23 (1972), 77-92. 2White, 109. of organization both within the community and within the agency. When the use of attitudes as measures Of program achievement do not coincide with the agency's concept Of what ought to be, organizational change should be considered. Alvin Toffler aptly pointed out, "...organization change - a self renewal, is a necessary and unavoidable response..."1 In addition, he suggested that any organization is nothing more than a collection of human objectives, expectations and Obligations, or a structure of roles filled by humans. It follows that a rearrangement Of these roles can create a new structure.2 It seems reasonable to assume that rearranging roles, readjusting social and economic variables to enhance attitudes Of program effectiveness is feasible if the correct combination of variables can be determined. Here again, according to Toffler, the approach is not new. He cites many examples of "project management teams" - ad hoc groups assigned to solve particular problems in specified time periods.3 Meeting today's demands for organizational structures _amenable to the task of problem solving arises directly out of the pace at which new and first-time problem situations arise. Achieving local participation at a high level of 1Alvin Toffler, Future Shock (New York: Random House, 1970), 130. 2Ibid., 127. 31bid., 132. involvement in development program decision making, it is agreed, is of prime importance to the technical or planning agency. However, structuring the local decision making body to enhance positive Opinions and hence outcomes is not as easily accepted as the pace Of change would suggest. Vested interests, social values and tradition are the inter- vening variables that make organizational restructuring at the local decision making level more difficult. Yet such a systems management approach can provide a social setting for program success. On the other hand, it can easily be argued that this form of social management strategy can suggest a depreciative form Of social manipulation. Never- theless, finding the critical variables is possible with relatively simple analysis and prediction techniques. Adjusting these variables to effect attitudes, then, is allowed through program organization. This study recognizes the problems Of agency concern for local participation in development programs and will seek to discover the relationship between the Opinions held by the decision makers within an on-going development 'program and the socio-economic characteristics Of those decision makers. The results Of this research could be implemented by development agencies tO help solve problems of local participation and negative attitudes related to the program goals and actual program achievement. In addition, the results of this study can provide development agencies with the quantification of useful information never before available to personnel responsible for the iguidance Of an onegoing development effort. Problem Setting and Historical Perspective The number of people in rural areas of the United States has been declining for many decades. In the 1950's more than half of the counties in the nation lost population and the majority of those showing population loss were rural.1 The main thrust Of rural—urban migration was prominantly noticeable on a nation—wide basis after 1935 when total numbers and proportions of farm population to total pOpulation steadily drOpped to the present low Of less than 5%. The reasons, Of course, are many. Not the least Of which was an unresponsive market and consequent unmet demands for higher farm profit margins. Meanwhile, mechanization Of agriculture freed a large portion Of the agricultural labor force and the new entrants into the job market turned to the growing industries. The non-agricul- tural industries were, for the most part, located in the cities and, as a result, the peOple moved to those growing metrOpOlitan areas. Census figures show that during the 1950's some 6.7 million people moved from nonmetropolitan areas to metrOpOlitan areas. This trend has continued throughout the 1960's. 1Sundquist, 130. 10 This rural-urban migration might seem to have reduced employment pressure in rural areas but it did not solve the larger problems being experienced in most rural counties. In fact, rural America, particularly the South and Appalachia, has the highest proportion of unemployed and underemployed. Moreover, rural areas, both farm and non-farm have, for the past few decades, had a higher proportion Of families existing at the poverty level than in the much more Obvious poverty areas in the urban sector. Commensurate with the statistical changes have been significant social changes. Increasing concentration Of production and ownership, higher production and marketing costs and shifts to larger land holdings have been reshaping rural society to a degree not yet understood on a natiOnal scale. At the other end Of the spectrum these rural changes have played a major role in reshaping urban society and in turn these same societal forces are at work in rural America but at a much different scale. Responding to the appeal from the declining areas for help, coupled with the realization that regional and area development is in the total national interest, the national government in the past 15 years has been grOping toward a policy Of intervention to stimulate economic growth. Such wide range programs as will be mentioned later can have serious effect upon rural society. As life styles and social structure seem to be intrinsically tied to the economic structure Of the group, any broad based efforts 11 in economic develOpment will have commensurate effects in other non-economic realms. The Objective of community develOpment at the federal level in rural areas has been the preparation and adoption of a comprehensive plan for simultaneously alleviating a wide range of community shortcomings and mobilizing the resources of many agencies at all levels both public and private. Several approaches have been attempted from stabilizing rural society to "keep'em down on the farm," to the promotion Of growth centers to absorb the unemployed from depressed areas. The efforts of the U.S. Department Of Agriculture to promote local planning and development organizations aimed at comprehensive programs for economic growth and rural social stability extends back as far as 1955 when President EiSenhower first authorized the establishment Of a rural development program. This is not tO say that federal interest in rural areas did not exist prior to 1955. TO the contrary, such federal programs as the Resettlement .Administration (1934), Rural Electrification (1936), Rural Land Use Planning (1938) and the Farmers Home Administration (1946) were involved in rural development but Operated within a specific program framework as Opposed to comprehensive programs for social and economic growth. The Extension Service, too, has been involved in rural development since the Smith-Lever Act Of 1914 created it. 12 But Extension's role has expanded over the years to meet changing needs. By the 1950‘s it too was and is now involved in comprehensive rural development and community development. .By 1955, then, the focus of federal effort in rural development had shifted from specific program emphasis to that Of local initiative to promote comprehen- sive development. The new rural development programs attempted to draw upon the resources Of all concerned federal agencies in achieving their goals of economic growth and social stability. The new program to grow from the new emphasis in the mid 1950's was known simply as the Rural Development Program. It was seen, at that time, as primarily a local program carried out by local leadership within the existing social structure. The federal agencies were to supply advice and technical assistance but local committees and social institutions were at the heart of any progress that might be forthcoming. As a result of the startling changes experienced in _rural areas in the previous decades the rural development program generally focused upon agriculture and land use. At the same time it was also concerned with industrial develOpment in an attempt to ameliorate the growing unemployment problems. The Kennedy Administration took rural development from its infancy and placed it in a position of highest priority in the Department Of Agriculture. The new 13 administration renamed Rural Development to Rural Areas Development (RAD) and created a new Office to carry out its function. During the same period, the rural areas were also made eligible to receive funds under the Area Redevelopment Act passed in 1961. Administering this new authority was complicated. After jurisdictional disputes between departments it was finally decided that the U.S.D.A. would provide technical assistance to the program while the Department Of COmmerce would review loans and grant proposals. By 1963, the Secretary of Agriculture, Orville L. Freeman, reported that rural develOpment groups had been organized in 2/3 of the nation's counties with over 65,000 local peOple engaged in problem solving in their communities. But during these early years Freeman rejected the idea Of establishing a new agency to handle the multiplicity of new funds, directions and areas. Rather, he organized the field personnel of the U.S.D.A. agencies already represented in most counties to handle the new task. Agencies such as .Farmers Home Administration, Soil Conservation Service, Extension Service, and the Agricultural Stabilization and Conservation Service were organized into technical action panels with the job Of coordinating the services Of all department agencies and making these services available to the people. Even as the Area RedevelOpment Act was being passed in 1961 the U.S.D.A. concluded that the Act alone would not 14 provide the needed measures to revive most impoverished rural counties. It was shown that about 800 counties (k of all) in the nation were economically lagging behind the remainder Of the nation. It was in such counties that commercial, community and private facilities had deteri- orated and farmland was underutilized or abandoned. The situation in these counties was analogous to city slums and an effort was made tO design a program of rural renewal.l It was the Food and Agricultural Act of 1962 which gave the Department Of Agriculture the authority for the program it called “rural renewal". The power Of the new authority would be comparable to that of urban renewal agencies in planning, acquiring, developing and reselling. land.- This program, Secretary Freemand told Congress, "would aid in developing new uses for land and water, create industrial parks, assist small farmers in farm consolidation ...and develop needed public facilities..."2 Most states, however, did not respond with necessary enabling legislation for rural renewal authorities as they lIbid., 139-40. 2U.S. Congress, House, Committee on Agriculture, Food and Agricultural Act of 1962 Hearings, 87th Cong.a 2d sess., 1962, p. 68. 15 had in authorizing urban renewal. As of 1968 only five rural renewal areas had been designated by the Secretary Of Agriculture. These five areas comprised a mere twelve counties with a small amount of federal loans actually issued. Under the same general language Of the 1962 Food and Agricultural Act which had authorized rural renewal, the U.S.D.A. organized a second program: Resource Conservation and Development (RC&D). RC&D was a relatively late starter as compared tO rural renewal. The Soil Conservation Service which provided the technical staff for each RC&D project had no difficulty finding sponsors for a new program that Offered a new source of funds and Opportunity for depressed rural areas. A Although the 1962 Act did not actually mention the program by name but simply gave the U.S.D.A. the authority to proceed at will, the final definition of RC&D was provided in a memorandum issued by the Secretary Of Agriculture in 1962. The Secretary Offered this definition: "RC&D: A locally initiated and sponsOred project designed to carry out a prOgram Of conservation and utilization of land in.areas where acceleration Of current conservation activities plus the use Of new authorities will provide additional economic Opportunities to the peOple..."1 The basic Objective Of all Resource Conservation and Development projects is the focusing of various governmental 1U.S. Department of Agriculture, Resource Conservation and Development Projects, Sec. Memorandum NO. 1515 (Nov. 2, 1962). ' 16 efforts, both federal and state with primary federal responsibility, in certain problem areas. The aim of such attention is to bring about land use adjustments and commensurate economic development in the best interest of the local rural pOpulation. The Soil Conservation Service, which administers the program, sees an orderly development and utilization of all resources as basic to the principle Objective of RC&D. All Of this is accomplished through the framework Of local leadership. The local decision makers are placed in the position Of coordinating the local human and natural resources in such a way as to facilitate the comprehensive development and utilization of these resources in keeping with the long-range community ambitions. The S.C.S. as the departmental administrator, had the major responsibility for program implementation. S.C.S. wasn't alone in its efforts however. Further, the Secretary directed other agencies to assist an needed or as their contributions were feasible. More specifically, the work to be undertaken in this new federal effort borne so inconspicuously in the Secretary's memorandum was to complement and accelerate the related regular programs of all U.S.D.A. agencies as provided in Section 102 Of the Food and Agricultural Act of 1962. ”RC&D projects will include two or more contiguous counties in the same area... 17 to be develOped and carried out under local initiative and leadership, and organized to assure a balance in economic growth and stability..."1 The above justification statement from the ApprOpriations Committee hearings added that the on-going regular programs will be continued at their normal rates. It was suggested that these new project activities would be in addition to and not in lieu of active programs and that apprOpriate land use adjustments in accordance with the needs Of local people would be stressed wherever the program was instituted.2 Five points stand Out in the various testimony and justification for the new authority's goals. They are: l. Accelerated adjustments in land use and . ownership to improve the economic stability Of family farms: 2. Shift use of land from the production of crops now in over-abundance to suitable uses for which there are unmet demands, such as recreation, industry, roads and water supply; ' 3. Speed up the planning and application of sound soil, water and plant conservation treatments to protect and improve those resources for future use; 1U.S. Congress, House, Committee on Appropriations, Hearings, before a subcommittee on appropriations for the U.S. Department Of Agriculture, 88th Cong., 1964, Part 2, p. 1007. 21bid. 18 4. Provide additional employment in rural areas and thereby reduce undesirable migration to pOpulation centers already facing unemployment problems; 5. Above actions in rural areas will enhance the economy Of the nation as a whole. Underlying all of these aims, ends and Objectives was the premise that RC&D projects would be locally initiated and locally sponsored. The success Of each project would depend upon synchronized planning and action provided by local people with administrative and financial assistance provided by participating federal, state and local agencies. Corrollary to the public project activities, farmers, ranchers and other Operators on private lands had available such things as technical assistance to facilitate physical improvements on their land. These same people also found local credit readily available to finance capital improvements. A Soil Conservation Service review in 1962 resulted in a preliminary inventory Of possible project areas. The inventory indicated 108 potential projects within 38 ‘states. NO explanation was given for the choice of the 108 potential project areas but the idea of local initiative together with other socio-economic considerations played a role in delineating them. Convincing Congress Of the need for such a program to be added to U.S.D.A. efforts ‘undoubtedly also played a role. This local initiative factor was demonstrated in the 19 early days of the program even prior to appropriations. Letters from soil conservation district Officials and other leaders indicated that local interest in the new form of develOpmental assistance was quite substantial.1 The Soil Conservation Service acted upon this interest and potential and estimated a budget for fiscal 1964 of $6,275,000 in support of RC&D. These monies were to cover the basic Objectives as outlined in the program proposals. In turn, this investment would create additional economic and employment Opportunities within each project area. Within the appropriations sub-committee and the Bureau of the Budget where budget restrictions were being imposed for 1964, the RC&D program was not on firm ground. Skepticism with regard to the usefulness of such a program within the cluttered field Of area development left the fledgling program with an appropriation Of only $1.5 million for its first year Of Operation. Further restraints were eventually placed upon the SOOpe Of the program itself when the budget was finally amended in Congress. In reality, RC&D had a mere $425,000 tO Operate in 1964 (the remainder Of the $1.5 million was held for use in fiscal 1965). The thinking in Congress was that the public interest might better be served by a pilot program approach at least for the first few years. As a result, the $425,000 was tO be utilized in working with local leadership tO develop long-range program proposals. lIbid., 1055. 20 The pilot project stage was an interesting phase of the RC&D approach. The choice Of the ten pilot projects is somewhat Obscure but it followed this format: initially information pertaining to the new program was disseminated through the U.S.D.A.‘s various agencies. Eventually about 20 applications were received from local areas representative Of various parts of the country. Each Of these applications was submitted through the Governor of the state in which the project was to take place. From the applications, the final designations Of the ten pilot projects was made through a joint effort Of S.C.S. and the Secretary of Agriculture's Office. The criterion of choice was based primarily upon the strength Of local initiative. As Secretary Freeman stated in his announcement Of January 31, 1964, "These are local projects with Federal assistance, and our decision to approve applications was determined in large part by the readiness Of local people to provide leadership and direction in the use of this new development tool."1 The ten pilot projects were characterized by several common features. Each consisted of two or more counties and all were sponsored by such local organizations as Soil and Water Conservation Districts, County Commissions, Town lU.S. Congress,House, Committee on Appropriation, Hearing before a subcommittee on appropriations for the U.S. Department Of Agriculture, 88th Cong., 1965, p. 489. 21 Councils, State Parks Commissions and many other previously established develOpment organizations. Today there are over 70 Operational RC&D projects in the fifty states. These projects are all locally initiated, sponsored and directed projects, "...designed to carry out a program of land conservation and land utilization, accelerated economic development, reduction Of chronic unemployment or underemployment in an area where these activities are needed to foster a local agency."1 Specific Objectives of the U.S.D.A. in the RC&D program include: 1. "The orderly develOpment, improvement, conservation, and utilization of natural resources Of the project area and thereby to provide employment and other economic Opportunities to the people Of the area." 2. "To provide to local leadership the Opportunity to more fully coordinate and utilize the facilities and techniques available under current agricultural programs,...and any applicable new programs as may be instituted to aid in planning and carrying out a balanced program Of development and conservation Of natural resources to meet local, state, and national needs." 3. "The orderly extension of this Program, where needed, project by project as local leadership is able to effectively plan and carry out the activities necessary to achieve the goals Of the Program."2 lU.S. Department of Agriculture, Soil Conservation Service, Resource Conservation and Development Projects Handbook, June 1972. 21bid. 22 This study, being concerned with the RC&D effort, will focus upon one Of the RC&D areas - Northwest Michigan. In keeping with the U.S.D.A. Objectives the Northwest Michigan RC&D Project was conceived by a small number Of interested Soil Conservation Districts in the Summer Of 1967. This nucleus fostered local interest and the RC&D region expanded to nine counties in the Northwest corner of Michigan's lower peninsula. An application for federal assistance was prepared under Soil Conservation Service guidance during January of 1968. Later that year, April 5, the application was endorsed by Governor George Romney and forwarded to the U.S. Department Of Agriculture in Washington. The original application was approved in October of 1968. Again interest in RC&D expanded to four adjacent counties. These four counties were added tO the RC&D region by an amendment to the original application and. endorsed by Governor William Milliken on March 21, 1969. .The RC&D area now comprises a total area of 4,081,280 acres including the thirteen counties of Antrim, Benzie, Charlevoix, Emmet, Grand Traverse, Kalkaska, Lake, Leelanau, Manistee, Mason, Missaukee, Osceola and Wexford.1 1U.S. Department of Agriculture, Soil Conservation Service, Northwest Michigan Resource Conservation and DevelOpment Project Plan, Lincoln, Nebraska: SCS, 1969. 23 Basically, the Northwest Michigan RC&D Project has as its goals to: help agriculture make its greatest contribution to the regional economy through wise land use planning and management of agricultural land: encourage woodland management in order to reduceerosion, improve woodland quality and increase local processing of forest products: provide watershed protection, flood control, reduce pollution and encourage wildlife habitat; and, assist communities in solving local problems through land use planning and to aid in the provision of such services as health and medical, housing, transportation, employment and education. The program is expected tO be in Operation for a period Of fifteen to twenty years. During that time the project steering committee will set priorities and initiate action with the aid of local, state and federal technical and financial assistance to achieve the above goals. The Economic Research Service of the U.S. Department of Agriculture has estimated that one hundred and sixty-one million dollars Of gross income and nine hundred and seventy man-years of employment could result if the project plans are achieved.1 The Northwest Michigan RC&D Project is a product Of the Department Of Agriculture's effort to stimulate overall rural develOpment. The success of that effort is now the 11bid. 24 concern Of two agencies within the Department: The Soil Conservation Service which manages all RC&D projects and; The Economic Research Service which is evaluating the RC&D effort in Northwest Michigan. Purpose Of This Study The larger project, Of which this study is a part, was established in 1970 by the Economic Research Service of the U.S. Department of Agriculture for the evaluation of the Soil Conservation Serviceis Resource Conservation and Development Program. The overall Objective Of the parent project is to secure information that would be helpful to the Soil Conservation Service and other project sponsors in future development efforts of the RC&D type. Within that framewOrk, several sub—Objectives have been proposed and are being carried out. They are: to examine the major classes Of project measure proposals for their complementarity with project goals in the NW Michigan RC&D program; to identify proposal and project priorities; to determine the under- lying socio-economic characteristics affecting decision iresults and; to suggest a policy strategy for improving the efficiency Of the program. The primary focus Of the entire study and the specific focus Of this study has been directed toward an analysis Of crucial social and economic influences on local RC&D decision making as applied to the NW Michigan RC&D project area consisting Of thirteen contiguous counties in the northwest 25 portion of Michigan's lower peninsula. The decision makers are specifically the RC&D county steering committee members. These persons generally selected and appointed to their steering committee posts by S.C.S. representatives, i.e., district conservationists within the area, will be the subjects of intensive study throughout this and the ERS project. This study, in keeping with the Objectives Of the overall ERS parent project, will through certain Objectives tests correlate certain social and economic variables tO attitudes toward program effectiveness. Through this process it will be possible to determine the combination of variables that contribute most to positive attitudes. CHAPTER II LITERATURE REVIEW Preliminary data from the results of the Economic Research Service's initial study has shown that fully 82 percent Of the respondents feel that the Resource Conser- vation and Development Program has thus far been effective in achieving stated goals. The underlying characteristics Of the respondents, their role relations to the program and the characteristics of the program itself may be related to such attitudes. 'A review of the pertinent literature reveals that such relationships may indeed be the case and that knowledge of those relationships is Of increasing importance to development efforts. The literature also reveals both through the lack Of research dealing with this subject matter and through specific references that a study Of ‘this nature will fill a significant gap in development program analysis. Further, it will provide knowledge Of those variables that significantly affect program outcomes either through their indirect influence upon attitudes or directly upon the decision making process. Green and Mayo, for example, point out that community studies generally concern themselves more with structure 26 27 than with social action. Although structure studies are important, little can be accomplished in predicting thel attitudes and actions of organized groups within the communities.1 Many methodological studies have been undertaken that have indicated various techniques for analyzing decision making processes within the community organization. In a 1957 study, Freeman and Mayo, using very basic analytic techniques and measures Of decision making, gathered data through personal interviews. The results were indicative Of ways in which to identify leadership action within the community situation but, unfortunately the authors did not concern themselves with the concept Of attitude.2 Using the Green and Mayo framework Of analysis, Folkman tried to analyze the decision making process with- in a farmer's cooperative organization. He suggested certain socio-economic characteristics as well as differing roles and role conflicts as the major characteristics that 0 U C 3 influence decisions. 1James W. Green and Selz C. Mayo, "A Framework for Research in the Actions of Community Groups," Social Forces, Vol. 31 (1953), 320-327. 2Charles Freeman and Selz Mayo, "Decision Making in Rural Community Action," Social Forces, Vol. 35 (1957), 319-322. 3William S. Folkman, "Board Members as Decision Makers in Farmer's Co-Operatives," Rural Sociology, Vol. 23 (1958), 239-252. 28 Other studies have been concerned with the process Of leadership involvement in develOpment programs. Sower and Freeman have proposed that the researcher recognize the existence Of a traditional set Of beliefs and relation- ships among local participants which can and should be utilized when activating prople in a program defined for the common good.1 Here, for the first time, one can see the recognition of convergent variables within the community and the decision making body that in combination can lead to success or failure Of a program. How does a researcher approach the study Of such variables within a community develOpment program situation? As we are concerned with the action organization itself - the decision making body, the focus should be at the inter- actional level - the program. This, according to Kaufman and Cole is the "interactional field." 'In contrast tO the neat systematization Of local society in the social system approach, the interactional field is a highly dynamic Open system which focuses upon interaction and process.2 Within .that frameworka host Of studies have been carried out. Many Of these have focused upon water resources but the lChristopher Sower and Walter Freeman, "Community Involvement in Community DevelOpment Programs," Rural Sociology, Vol. 23 (1958), 26-33. 2H. F. Kaufman and L. W. Cole, "SOciological and Social Psychological Research for Community Development," International Review Of Community Development, NO. 4 (1959), 293. 29 principles and techniques are equally applicable to related non-water develOpment. Few were concerned with the analysis Of attitudes toward program effectiveness per se. The one notable exception was, however, John D. Photiadis' 1966 study of attitudes toward water resources development in South Dakota.1 Photiadis was concerned with attitudes Of local residents toward various programs as well as the characteristics Of the respondents, their knowledge Of the programs and other variables. He tried to identify factors relevant to attitude formation. Age, education, residence and ownership were among those variables he found to be of most significance in attitude formation. Earlier, Wilkinson had reported his research effort. dealing with the influence of community structure upon the course and outcome Of watershed development programs in the South.2 Pointing out the scarcity Of such research efforts, he indicated that the greater the linkages and the stronger the lines Of communication within the interactional field, ‘the greater the likelihood of project accomplishment. 1John D. Photiadis, Attitudes Toward the Water Resources Development Program in Central South Dakota, Department of Sociology Extension Service (Brookings: South Dakota State College, 1966). 2Kenneth P. Wilkinson, Local Action and Acceptance Of Watershed DevelOpment, Water Resources Research Institute (State College, Miss.: Mississippi State University, 1966), 9. 30 In a later study Wilkinson and Cole delved deeper into the concept of attitude and social variables.1 Their Objective was to assess the influence of community structure on the effectiveness of local watershed develop- ment. They clearly point out that attitudes are a field theory concept having to do with the qualitative relation- ship between an individual's inner feelings and some Object in the environment. Differences in socio-economic character- istics influence attitudes to a significant degree. They further suggested that the extent Of the individuals' knowledge Of and participation in the program will also influence attitudes. Other literature diverges from the focus Of this study.. In Price's study of organizational effectiveness the conclusions center not upon the explanation of attitudes but rather the measure Of such attitudes Of effectiveness.2 The importance Of attitudes in influencing program outcomes must be regarded as significant. Attitudes, according to Robert Lauer, influence various psychological 1Kenneth P. Wilkinson and L. W. Cole, Sociological Factors in Watershed Develgpment, Water Resources Research Institute (State College: Mississippi State University, 1967). 2James L. Price, "The Study Of Organizational Effectiveness," The Sociological Quarter1y, Vol. 13 (1972), 3-15. 31 processes and govern perceptions so that the perception Of any phenomenon is congruent with attitudes.1 The paucity Of literature in the field of attitude explanation explicitly related to decision making and program effectiveness is quite apparent. Again, Wilkinson and Singh, leaders in the field of social science applications to develOpment program analysis, suggest that social- psychological problems in develOpment efforts have been virtually ignored. ‘YIn the long run," they say, "these may prove to be the most significant problems of all. Programs are Operated, blocked, salvaged or abandoned by people. While the environmental situation and social structural‘ conditions Obviously contribute to human behavior, in the final analysis man behaves or fails to behave in a particular manner on the basis of his own more-or-less unique perception of reality..."2 More recently, Wilkinson said that theoretical concepts Of community develOpment also have been merely ignored and it _ 1Robert N. Lauer, "The Problems and Values Of Attitude Research," The Sociological Quarterly, Vol. 12 (1971), 247-252. 2R. N. Singh and Kenneth P. Wilkinson, Social Science Studies Of Water Resources Problems: Review Of Literature and Annotated Bibliography, Water Resources Research Institute (State College: Mississippi State University, 1968), 23. 32 would seem to be important to consider the dominant forces affecting community structure in order to develOp more valid criteria for evaluating programs. He said that consideration of community structure is being ignored by external agencies and submits that very little is known of the consequences Of such ignorance.1 The literature suggests both directly and indirectly by the lack Of relevant literature that there is a need for an evaluation of social factors implied in develOpment. Beyond that there is a dearth of knowledge and research concerning the role Of decision makers' attitudes and how these complex multivariate attitudes relate to social,' economic and perceptual variables. In general, adequate models to describe the intricacies Of decision making and, thereby, critical points in the process, are lacking.2 Although several studies have dealt with attitudes and their relation to program outcome, they provide little evidence to show just how much of a role attitudes play in 1Kenneth P. Wilkinson, "A Field-Theory Perspective for Community DevelOpment Research," Rural Sociology, Vol. 37 (1972), 43-52. 2White, 108. 33 determining final outcomes.1 Further, there is as yet no adequate and explicit model upon which development efforts or organization can be structured to enhance positive attitudes and program outcomes. 1Several studies have noted the relationship between favorable or positive attitudes and actual behavior within an organization. Some examples are: James H. COpp, "Perceptual Influences on Loyalty in a Farmer Cooperative," Egral Sociology, Vol. 29 (1964), 169; Irwin Deutscher, "Words and Deeds: Social Science and Social Policy," Social Problems, Vol. 13 (1966), 235; John Harp, "A General Theory Of Social Participation," Rural Sociology, Vol. 24 (1959), 380; Rensis Lickert, New Patterns of Management, (New York: McGraw—Hill, Inc. 1961). CHAPTER III RESEARCH PROCEDURES The literature review Of the preceeding chapter points out a need for research concerning the relationship between social factors and attitudes. It also suggests that previous research dealing with attitude formation has isolated such factors as age, education, residence and tenure as significant antecedent contributors to attitudes. Yet, there is no real indication Of the degree to which such sociological indicators are related to attitudes. It is therefore important to investigate such relationships in a manner that will yield results to help explain attitude formation. Objectives The Objective of this study involves the quantification 'Of various socio-economic characteristics of the Northwest Michigan RC&D steering committee members, their attitudes and opinions in order to: 1. Determine the overall relationship between such characteristics and attitudes. 34 35 2. Attempt to find causal relationships between specific antecedent socio-economic characteristics and attitudes toward program effectiveness. .It is, however, recognized that causal relationships are merely assumed and not proven in this study. 3. Apply techniques of analysis not previously used in the analysis Of development programs to: A. provide more precise measures Of relationship between variables; B. suggest further application of similar techniques to development program analysis. 4. Develop a model upon which to analyze attitudes in other related development programs to: A. provide sponsoring agencies with an additional resource for assessing program achievement; B. provide new insight into attitude formation within the local decision making framework.‘ Data Collection TO accomplish the above Objectives the data tO be quantified and analyzed was provided by the Economic Research Service (North-Central Resource Group). Although .the author was not directly involved in the data collection process a description of the procedure will nevertheless be provided. The data was gathered through individual personal interviews with the entire population Of seventy-two county RC&D leaders (county steering committeemen). As indicated earlier, these steering committee members representing the 36 thirteen counties in the NW Michigan RC&D Project were generally selected and appointed to the county steering committees by Soil Conservation Service representatives within the region. Each county has its individual steering committee which in turn is responsible to the Project Steering Committee. Some county steering committee members can simultaneously serve at the Project level. Because of the relatively small number Of persons in the steering committee population a census rather than a sample was determined to be feasible within the time and funding limits established by the Economic Research Service evaluation project. The individual personal interviews were administered in the Autumn of 1971. Two interviewers intimately involved in the questionnaire design were utilized in the field to contact each of the steering committee members on an individual basis. It was reported that each interview required an average of one and one-quarter hours to administer. The questionnaire itself (see Appendix A) was designed, pretested for weaknesses and addition of new questions as well as revised under the direction Of Mr. David G. Carvey, Project Leader, during a six month period prior to its actual administration in 1971. The following persons outside Of. the Economic Research Service - Natural Resource Economics Division participated in the review and criticism of the questionnaire: Dr. William J. Kimball, Department of Resource DevelOpment, Michigan State University; Dr. J. 37 Allen Beegle, Department Of Sociology, Michigan State University; Dr. A. A. Schmid, Department of Agricultural Economics and Department Of Resource Development, Michigan State University; Dr. James Copp, Economic Development Division, Economic Research Service. The pretest of the instrument indicated several changes to be incorporated into the final version. For example, the personal data questions were relocated from the beginning of the interview schedule to the end Of the schedule. The decision was made based on the assumption that such questions could be construed by the respondents as threatening to their privacy and as a result could hinder the relationship between interviewer and interviewee thereafter. The exclusion Of questions related to the respondents' length of present employment, the educational achievement of his dependents, his dependents' mobility and his estimate of the pOpulation of the nearest village, town or city, were deemed to be superfluous and Of little value to the goals Of the interview survey. The addition .of the Open ended question allowing the respondent to comment on any aspect Of RC&D at the end of the interview schedule was found to be necessary as a relief valve to ideas not covered in the interview per se. In addition, the interview survey did not coincide with any similar data collection effort by the ERS or other federal agency during that time. The survey resulted in thirty-nine quantifiable questions from which there are a 38 possible three-hundred and sixty-eight responses or variables available for further analysis. Hypotheses The principle Objective Of this study is to identify and attempt to explain the relationship between certain socio-economic characteristics of the decision makers, their roles, perceptions and beliefs and positive attitudes toward program effectiveness. In an attempt to hypothesize the direction of the relationships between each of the possible three-hundred and sixty-eight variables from the attitude survey, it became apparent that such a task would itself require a volume of pages to record. For example, it could be hypothesized that each age group, income group, occupation, memberShip affiliation and so on, is positively or negatively correlated to affirmative attitudes toward program effective- ness. Instead of such an exhaustive listing of hypotheses the literature review suggested that certain characteristics Of the respondents would be positively correlated tO attitudes of effectiveness. As a result, the following 'general working hypotheses were constructed to include broad groupings Of characteristics without identifying individual categories within the groups. The hypotheses are: 1. There is a positive relationship between the socio-economic characteristics of the respondents, their role relations, reference groups, their environmental perceptions and positive attitudes toward program effectiveness. 39 2. The respondents' Opinions of RC&D contribution to the solution of both human and natural resource related problems are highly correlated to positive Opinions of program effectiveness. The above general hypotheses are aimed at identifying those variables that offer explanations of the steering committee members' positive attitude toward the RC&D program effectiveness and hence, its possible outcome. Although this study does not deal with program outcomes or measure adoption per se, the basic assumption in the measurement of attitudes toward program effectiveness is that such attitudes are a representation or surrogate measures for program outcomes as Of the time period in which the questionnaire was administered. It is further assumed that the attitudinal patterns are, in an aggregate sense, true representations_ Of actual program effectiveness. Methods of Statistical Analysis When analyzing attitudes toward program effectiveness two tasks must initially be performed. First, clusters of .related events to be included within concept of ”effective- ness" must be identified. Secondly, the procedure for analyzing the relationship between the multivariate responses making up the concept Of effectiveness and the multivariate responses accounting for the variance in "effectiveness" must be selected in order to explain those relationships. 40 Ad discussed in Chapter I, concepts Of attitude represent many related and individual forms of behavior. As a result, the researcher can no longer be satisfied with a simple and singular predictor and criterion relationship. Rather, it is more valuable, in terms Of explanation, to determine the relationship between several variables eventually reducing these to the principle factors of ’ explanation. In pursuing that objective, the researcher can identify unifying principles and at the same time discard irrelevant phenomena.1 The first task, that Of identifying related events to be included within the concept Of effectiveness, was performed in two related Operations. All of the responses from the quantifiable questions on the attitude survey were cross-tabulated with one another. (See Appendix C) The result of this cross-tabulation specifically showed the number and nature of all other responses when one particular response was given. For example, of those respondents that gave an "effective" response to the question relating to their Opinion of RC&D effectiveness, 16 respondents indicated Itheir income to be between five and ten thousand dollars. In this way each of the 368 responses can be cross-checked 1Paul B. Koons, Jr., "Canonical Analysis," in Computer Applications in the Behavioral Sciences, ed. by Harold Borko (Englewood Cliffs: Prentice-Hall, Inc., 1962), 267-279. 41 with each of the others in a like manner. This step enables the selection of meaningful variables to be further analyzed in light Of the hypotheses. As a second check on the relevance of variables chosen, the correlation coefficients of each of the 368 variables was calculated to determine each variable's measure of relationship to all others. Again, the technique provides -a means of isolating those variables that would be further analyzed with more powerful explanatory techniques. (See Appendix D) Using the variables selected from the above procedures another simple cross-tabulation was performed. A matrix of cross-tabulations was constructed for ease of analysis. From these tables the final categories of variables were selected in terms of the criterion and predictors to be used in the final analysis. The second major task of attitude analysis is the selection of the analytic technique by which to test the hypotheses. Considering the elements of the problem, the loosely defined and not easily quantifiable variables involved in attitudes of program effectiveness coupled with several measurable or observable quantitites such as socio-economic data, role relations and others, it was clearly determined that no single multiple regression equation would provide an adequate solution to explain such attitudes. As a result, two analytical methods were chosen: Canonical Analysis and Factor Analysis for factor isolation. 42 Canonical Analysis was chosen as the principle technique for the task. To more clearly define canonical analysis and its explanatory power the closely related technique Of Multiple Regression Analysis will be reviewed while pointing out the similarities and dissimilarities. The problem at hand involves a large amount of data. What is required is an equation for one of the quantities in terms of its relationship to all the others. The technique for arriving at such an equation is called "regression". Regression identifies the strength and direction of relationships, summarizes the data and predicts new or future situations. In other words, regression finds the "best" equation relating Y to X1 and. x2...xn. Pictorially, a regression equation locates a plane passing closest to a cluster of points in space (a point being a Y corresponding to x1, x2...xn). A formula based on the data is easily computed and prediction of new values for a given X value can be made.1‘ Similarly, [canonical correlation can be considered as a measure of the extent to which individuals in sets occupy the same relative 1International Business Machines Corporation, Concepts and Applications of Regression Analysis (White Plains, N.Y.: International Business Machines Corporation, 1966), 1-3. 43 relative position in the predictor (independent variables) space as they do in the criterion (dependent variables) space.1 Canonical analysis is similar to regression in the sense of finding the best equation relating the variables but it goes much further in another sense. Whereas in regression analysis there can only be one Y variable (dependent) and multiple X (independent) variables, in canonical analysis the data are organized and analyzed in sets of variables. 'That is, both the dependent and independent variables are multivariate. Each set can theoretically contain as many variables as the researcher finds necessary. In reality there are limits placed on) the number of variables in each of the sets by the computer capacity. The use canonical analysis can involve any or all of the following: 1. Determining which variable in each of the two sets contributes most to the between set association. 2. Finding linear combinations of the predictor and criterion sets that maximize correlation when linear combinations are correlated in a two variable sense. 1William W. Cooley and Paul R. Lohnes, Multivariate Procedures for the Behavioral Sciences (New York: John Wiley & Sons, Inc., 1962), 36. 44 3. Predicting linear combination scores of objects in one set of variables from those in the other set.1 The similarities between canonical analysis and multiple regression analysis are apparent. For example, in multiple regression the researcher is attempting to find a linear combination predictor variable that is most highly correlated with the criterion or dependent variable. When using canonical correlation the researcher finds linear combinations made up of sets of variables on both sides of the equation. Rhea Das has pointed out: "A variety of research problems call for the investigation of the relations between two sets of variables. Given a set of predictor variables and a set of criterion variables with measurements on the same individuals, the empirical question is one of obtaining the most effective prediction... To explore a new domain, several standard reference tests of one set of factors and a number of standard reference tests of another set of factors can be employed. These illustrative situations concern the relations between two sets of variables and as such should be amenable to the statistical method of canonical analysis." 1 Paul E. Green and Donal Tull, Research For Marketing Decisions, (Englewood Cliffs, N.J.: Prentice-Hall, Inc. 1970), 377-378. 2Rhea S. Das, "An Application of Factor and Canonical Analysis to Multivariate Data," The British Journal of Mathematical and Statistical Psychology, Vol. 18 (1965), 57. 45 In further detail, the technique of canonical correlation begins with the determination of the inter- correlations of the variables within each of the two sets and a cross-correlation between those two sets. The result of the first intercorrelation, utilizing one set as the "predictor" and the other set as the "criterion", yields a matrix of correlation between the predictor and the composite criterion as well as a matrix of regression weights for the set of predictors. It also yields the reverse of the above. In other words, it yields the matrix of correlation between the criterion set and the composite predictor and matrix of regression weights for the criterion.1 Most importantly, we now have the numerical values known as canonical coefficients. The resulting numerical values for each of the variables in both sets show their relative predictive power. The weights derived for each of the variables in this maximum correlation of both sets show their relative importance in both the concept of effective- ness and the formulation of the attitudes toward program effectiveness. These weights are interpreted in terms of .the magnitude of their values in combination with the direction Of their sign. Canonical analysis yields the following data from which conclusions can be drawn with regard to the general hypotheses: lKoons, 269. 46 1. Means and standard deviations of all variables. 2. Correlation matrix. 3. The Chi-squares and degrees of freedom. 4. Factor pattern for both sets. 5. Canonical correlations with weights associated with each variable in each set. The computation of canonical correlations is accomplished by using the procedure developed by Paul R. Lohnes at the University of Buffalo and modified for use on Michigan State University's CDC 3600 computer by A. V. Williams of the Computer Institute for Social Science Research. CISSR'S CANON routine requires that no more than 80 variables be computed during any single run. As a result, I have arranged all of the pertinent variables into five "predictor" sets and one "criterion" set (listed below). Each computer run will perform canonical analysis upon each of the five predictor sets in combination with the single criterion set. Computationally, it is irrelevant whether the variables on the left or on the right are considered as criterion or predictor. However, computer time is vminimized if the data are arranged so that the variables on the right are less in number than those on the left. A sixth predictor set was constructed after the results Of the first five canonical correlations were analyzed to determine those variables most influential in explaining attitudes toward program effectiveness. From that, a composite predictor was maximally correlated 47 again with the original criterion set to obtain results that were originally unavailable because of the limited computer capacity. Following is a list of variables (Figure l) grouped into predictor and criterion sets. The sets were established in such a way as to minimize computational time and maximize ease of analysis. All sets are categorized according to types of variables contained within each set and are so indicated by their respective titles. The variables were grouped together within sets based upon ease of identification and conceptual similarity. Factor Analysis With Canonical Analysis In an attempt to further understand the relationship between favorable attitudes toward the RC&D program and . the socio-economic and perceptual structure of the decision makers another canonical correlation was computed this time using as the independent or predictor variables the factors isolated from the total Of 368 variables. The means by which these factors were isolated was "factor analysis." The Economic Research Service, Natural Resource Economics Division data file provided the results of a previously computed factor analysis of all variables derived from the attitude survey. The ERS data yielded a total of fifty factors which could be used as the predictor set in canonical analysis. This new predictor set was correlated with the previously established criterion set 48 Predictor Sets Set Number of Set Number of Number Variables Number Variables 1. Socio-Economic 4. Membership Age 4 Influential Education 6 Human Res. Proximity to Oriented Econ. Place 3 Organization 15 Occupation 20 Influential Income _§_ Natural Res. Total 39 Oriented Organization 15 2. Problem Perceptions Membership in Human Economic Res. Oriented Perceptions 3 Organization 10 Nat. Environment Membership in.Nat.' Perceptions 4 Res. Oriented Social Envir. Organization ._Q Perceptions _§_ Total 49 Total 25 5. Goal Relations 3- Role Relations Availability of Residence & Tenure 12 Goals 1 Related Roles & Efforts to Interest ~Tenure 3 Local People 1 Current Roles 4 Proposal Forwarding 1 Initial Contact 7 Community Awareness 3 Key Leaders .11 Leadership Involve- Total 37 ment 5 Propensity to Change _;1 Total 12 Criterion Set Attitudes Toward Number of Program Effectiveness Variables Opinions of Effectiveness 2 RC&D Contribution to Problem Solution (HRO) 7 RC&D Contribution to Problem Solution (NRO) 7 RC&D Contribution to _ Economic Development ._1 Total 17 Figure l. - Predictor and Criterion Sets Used in Canonical Analysis 49 used in prior canonical analysis. This additional model or technique of analysis was attempted for several reasons. First, the results of factor analysis were used to compare those variables isolated by that technique to the results of the initial canonical analysis. This would provide, it was thought, additional insight into the explanation and prediction of positive attitudes toward program effectiveness. It was further reasoned that such a comparison would provide a check on the reliability of canonical analysis as well as the cross-tabulations and correlation coefficient method of variable selection used to establish criterion and predictor sets in the first five canonical correlations. The technique of factor analysis as employed by the ERS was an attempt to find order and regularity in phenomena. ”As phenomena co-occur in space or in time, they are patterned; as these co-occurring phenomena are independent of each other, there are a number of distinct patterns.1 We assume patterns for such things as political systems, develOpment schemes as well as decision makers' attitudes. Factor analysis handles all measurements both quantitative and qualitative and resolves them into distinct patterns of 1R. J. Rummel, "Understanding Factor Analysis," Conflict Resolution, Vol. 11 (1967), 445. 50 occurrence. "It makes explicit and more precise the building of fact-linkages going on continuously in the human mind."1 Factor analysis, like canonical analysis has the capability of considering a large number of variables or characteristics and reduces these into a smaller number of factors for further conceptual analysis. This study, then, utilized factor analysis in the selection of independent or predictor variables to be used in canonical analysis. The resulting weights and equation as well as the comparison to previous methods will be reported in the following chapter. Ibid 0 CHAPTER IV RESULTS To accomplish the Objectives of the study the data derived from the personal interview survey will be discussed and analyzed at two levels: first, the primary data are reviewed to illuminate the basic socio-economic, perceptual and attitudinal constructs of the respondents; second, the results of the statistical analytic techniques are discussed for their explanatory content and modelling implications. The Preliminary Results The data derived directly from the questionnaire survey reveal interesting generalizations that can be drawn concerning the overall socio-economic attributes of the steering committee or respondent pOpulation. From the data portrayed in Table l, a hypothetical average or 'composite steering committee member can be characterized. Generally, this respondent is male, about 51 years Of age and likely to be a farmer or self-employed in other business activities. He has completed high school and has had some college training. His present income is about eleven thousand dollars and he has lived in his present county of residence for more than twenty years. 51 52 Socio-Economic Characteristics More specifically, 62 percent of the steering committee members were fifty years of age or older. Only 14 percent were less than forty years of age. Nearly seventy percent Of the respondents had lived in Michigan for forty years or more, while about 35 percent had lived in their present county of residence for forty years or longer. Slightly less than 30 percent of the committeemen had completed high school only. But over 37 percent had some college training and 18 percent were college graduates. Income was varied but the estimated average income was over $11,000 in 1970. Yet, nearly 10 percent of the respondents earned less than five thousand dollars; more than half earned over ten thousand dollars. To further determine factors associated with attitudes, perceptions and priorities, each respondent was asked to specify the distance from his home to the village, town or city (economic place) where his family does most of its shopping. Thirty-eight percent live within the limits of an economic place while another 30 percent live within five .miles of an economic place. The data also reveal that 22 Percent live ten miles or more from their economic center. (See Table 1) Each respondent was asked to indicate his membership in groups or organizations according to the categories listed in Table 2. Additional data was derived by asking 53 Table l Socio-Economic Characteristics of the Respondents Percent Percent Categories of Total Categories of Total Age Education 20-29 2.8 Completed H.S. 29.2 30-39 11.1 H.S. Tech. Tr. 2.7 40-49 23.6 Some College 37.5 50-59 33.7 College Graduate 18.0 60 plus 27.8 Special training* 29.2 Residence & Income Tenure less than $5,000 9.7 Years in State $5,000 to $10,000 37.5 under 20 1.3 $10,001 to $15,000 26.4 20-29 11.1 $15,001 to $20,000 12.5 30-39 18.0 over $20,000 8.3 40-49 22.2 non-response 4.2 50-59 29.2 over 60 18.0 Proximity to Economic Place Years in County 0 miles 37.5 under 20 23.6 l-5 miles 29.2 20-29 25.0 6-10 miles 11.1 30-39 11.1 more than 10 miles 22.2 40-49 18.0 50-59 16.7 over 60 5.6 *In some cases a response to "special training" was received in conjunction with a response in another category. 54 each respondent to indicate whether or not he is an active member or officer in each group or organization. Of the fourteen types of categories of organizations listed, seven were human resource oriented and the remaining seven were natural resource oriented.1 Within the natural resource grouping, membership totaled 129. Of those, 114 were active participants. The highest number of members recorded for all groups was in the "Agricultural" category in which 33, slightly less than half of the respondents, indicated participation. The human resource oriented categories had 95 respondents indicating their membership. Ninety-one of those said they were active participants. (See Table 2) Respondents were asked: "How did you first hear of the RC&D program?" Fifty percent of the steering committee members specified an agriculturally related group or organization as their initial contact. The principle agricultural sources were Extension Agents and District Conservationists. In all, very nearly 75 percent of the decision makers indicated initial contact with RC&D through ‘natural resource oriented groups (see Table 3). In fact, only about 22 percent had first heard of the program through human resource related groups. Within that category, 1All problem areas, memberships, priorities and other concerns were divided into either categories of human resources, e.g., housing, medical, education, etc. or natural resources oriented, e.g., water, forestry, recreation, etc. 55 Table 2 Respondent Membership Natural Resource Oriented Membership Environment 17 Land 28 Water 4 Agriculture 33 Forestry 6 Recreation 23 Planning and Development _18_ Total 129 Human Resource Oriented Membership Education 24 Health and Medical Service 18 Industry 4 Employment 6 Transportation 8 Housing 12 Community Facilities & Service 29 Total 101 56 Table 3 Respondents' Initial RC&D Contact Number of Percent Categories Respondents of Total Natural Resource Oriented Environment 1 1.4 Land 13 18.0 Water .. ... Agriculture 36 50.0 Forestry .. ... Recreation 4 5.5 Planning and Development .. ... Human Resource Oriented Education .. ... Health and Medical Serv. l 1.4 Industry .. ... Employment .. ... Transportation 1 1.4 Housing .. ... Community Facil. & Serv. 14 19.4 19 percent of all respondents had made their initial contact through "Community Facilities and Services" groups. "Land" related groups ranked third as a source of contact and primarily included SCD personnel as the contact source. In order to further identify the social structure and leadership framework within the RC&D area, each committeeman was asked to identify up to three groups or organizations which he considered to be influential within the community and the RC&D program. They were also asked to identify, by occupation, the individuals each considered as key leaders within the community. 57 The respondents listed 192 groups or organizations they felt were influential in the RC&D region (Table 4). There is a relatively even distribution between human and natural resource categories. A closer look, however, reveals that the category, "Community Facilities and Services“, accounted for 77 responses. This represents 40 percent of the 192 total. Included within that category are all social, civic, and religious organizations (other than church membership per se) as well as some governmental organizations and specific interest groups. Agriculture accounted for 22 percent of the responses principally because of the committee memberships association with the Extension Service and the Soil Conservation Service. Other organizations included within this category were, Farm Bureau and local Grange. Planning and DevelOpment organizations were also mentioned frequently (31). Most mentions represented county planning commissions and the Economic Development District (EDD) . Key community leaders were identified by occupation. The five major occupation groups identified include: Managers, Farmers, Extension Agents, Foresters-Conservationists and, Public Officials. Respondents were further asked to indicate whether or not these key leaders were active or involved in the RC&D effort. 58 Table 4 Influential Organizations by Category Categories Mentions Human Resource Oriented Education 3 Health and Medical Services 1 Industry 4 Employment 0 Transportation 5 Housing 3 Community Facilities and Services _11_ Total 93 Natural Resource Oriented Environment ‘ 1 Land 14 Water 1 2 Agriculture 42 Forestry 3 Recreation 6 Planning and Development _31 Total 99 The results, again, reflect the natural resource orientation of the RC&D committeemen. The three occupational groups with the highest incidence Of RC&D ‘involvement are: Farmers, Extension Agents and Forester- Conservationists (see Table 5). Within the same context, the respondents were asked to identify the roles that the various leaders they had singled out by occupation now filled in the RC&D program. Of the 101 key leaders involved in RC&D, half were steering committee members (Table 6). Obviously, then, the RC&D 59 Table 5 Key Leader Identification and RC&D Involvement By Occupation RC&D Involvement Occupation Mentions Yes No Undecided Manager 50 15 7 33 2 Farmer 34 23 10 1 Public Official 17 8 9 0 Teacher 9 2 Extension Agent 22 19 0 Forester- Conservationist 18 17 1 0 Construction 7 2 4 1 Physician 5 l 4 0 Engineer 6 2 3 1 Insurance Agent 3 0 3 0 Retired 9 6 3 0 Other ' l9 6 12 1 Total 199 101 90 8 committeemen tended to identify members of their own steering committees as key leaders. About 18 percent were identified as advisors, while prOposal sponsorship accounted 'for 5 percent. Indirect involvement, implying that the leaders were working toward results similar to those desired by the RC&D project itself but not through personal RC&D involvement, accounted for 28 percent of the key community leaders. 60 Table 6 Key Leader Participation By Roles Percent RC&D Role Mentions of TotaF County Steering Committee 50 25.1 Advisors 18 9.0 Project Sponsors 5 2.5 Indirect 28 14.1 *Percentages based upon 199 key leaders identified Attitudes and Perceptions This portion of the data from the results of the questionnaire survey concerns the respondents’ attitudes toward the RC&D program and its contribution to the solution of community problems. In addition, the findings are concerned with the steering committee members' perceptions and beliefs about various aspects of the local environment. All committeemen were asked: "Which response best expresses your Opinion of RC&D effectiveness in achieving .goals of the RC&D project for your county?" The respondents' choices were: effective, slightly effective, not effective, undecided. More than half (54.2%) chose "effective" to describe their feeling toward the RC&D program. The combination of both positive responses (i.e., "effective" and "slightly effective") accounted for the attitudes of 82 percent of 61 the steering committee members.‘ Only about 15 percent chose to characterize the program as "ineffective" while two respondents were undecided (see Table 7). Table 7 Respondents' Opinions of RC&D Effectiveness Slightly Not Effective Effective Effective Undecided Number of Respondents 39 20 ll 2 Percent of, Total 54.2 27.8 15.3 2.8 Table 8, is divided into the two broad categories of human resource orientation and natural resource orientation. The data portrayed in the table is the result of the question asked of all the reSpondents regarding their Opinion of RC&D's potential contribution to the solution of universal community concerns. It is interesting to note that the steering committee membership generally felt that RC&D is more likely to achieve valuable progress in the natural .resource oriented problem areas than in the human resource oriented problem areas. The highest positive response in any of the problem areas was that dealing with "Land." Here, 89 percent of the 72 respondents felt that RC&D can make valuable contributions to problem solution. Not far behind in the number of positive responSes is the problem category "Water" (86%). Considering all categories dealing 62 with natural resource oriented problems, the lowerst number of positive responses is greater than the highest number of positive responses in the human resource oriented categories. The greatest number of positive responses in the human oriented problem areas lies in the "Community Facilities and Services" category. Seventy-six percent of the committee members felt RC&D could make a valuable contribution to problem solution within that realm. Negative Opinions were few in the natural resource areas, while as many as 32 percent of the respondents registered a negative opinion for the "Health and Medical Services" category. The problem areas of "Housing" and "Transportation" drew 31 percent and 29 percent negative response respectively. (See Table 8) In answering a question concerning whether the RC&D project has helped to improve economic Opportunities for the peOple of their counties, the steering committee members holding the Opinion that it did were in the majority (58.3%). Some respondents (30.6%) felt that the _project is not making any noticeable improvement and there- fore see no RC&D results in their counties (see Table 9). Comparing the results of this question to the responses regarding opinions of RC&D effectiveness a definite connection can be surmised. In fact, this same distributional pattern of opinions is evident throughout the survey results. 63 Table 8 Respondents' Attitudes Toward Program Contribution to Human and Natural Resource Problems Percent Percent of of Categories Total* Categories Total* Education Environment Yes 42 Yes 80 No 26 No 7 Undecided 32 Undecided 13 Health & Medical Land Yes 33 Yes 89 NO 32 No 3 Undecided 35 Undecided 8 Industry Water Yes 56 Yes 86 No 21 No 3 Undecided 24 Undecided 11 Employment Agriculture Yes 67 Yes 67 No ‘ 15 No 12 Undecided 18 Undecided 21 Transportation Forestry Yes 47 Yes 81 No 29 No 7 Undecided 24 Undecided 12 Housing Recreation Yes 47 Yes 83 No 31 No 3 Undecided 22 Undecided 14 .Community Facilities Planning and and Services Development Yes 76 Yes 85 No 7 No 4 Undecided 17 Undecided ll *Percentages are rounded to nearest whole number. 64 Table 9 Respondents' Attitudes Toward Program Contribution To Economic Development Yes No Undecided Number of Respondents 42 22 8 Percent of Total 58.3 30.6 11.1 Questions were asked concerning the respondents' Opinion of the economic situation within his county, quality Of the natural and social environment, community awareness of the RC&D project and, the individual's prOpensity to change the program. Responses to the question concerning economic conditions ranged from "Growing Rapidly" to "Declining Rapidly". Generally, the RC&D leaders perceived economic conditions as "growing". None felt that his county was declining rapidly and only one committeeman felt that his county was declining at all. Comparing this data to the responses from Table 9, it is assumed that the respondents 'attribute some of the perceived economic growth to the RC&D effort. 65 Table 10 Respondents' Perception of Economic Conditions Number of Percent Choices Responses of Total Growing Rapidly 21 29.7 Growing Slowly 40 55.6 Stabilized 10 13.8 Declining Slowly l 1.4 Declining Rapidly .. ... Undecided .. ... The steering committee members' perception of the natural environment generally favored the Opinion that the natural environment is improving slowly. Extreme perceptions of rapid deterioration or rapid improvement of the environment accounted for only 12 percent of the responses (Table 11). The perception of the social environment was primarily that of "improving slowly" and generally these responses were more positive than the responses to perceptions of the natural environment. (See Table 11) When asked to indicate the degree to which the RC&D program is known and understood by the peOple of their county, over three-fourths of the committeemen expressed the belief that the program was not very well known. (Table 12) 66 Table 11 Respondents' Perception of Natural and Social Environmental Quality Natural Environment Social Environment Number of Percent Number of Percent Choices Responses of Total Responses of Total Improving Rapidly 3 4.1 11 15.3 Improving Slowly 25 34.7 44 61.1 Stabilized 19 26.4 12 16.7 Deteriorating Slowly 19 26.4 3 4.2 Deteriorating Rapidly 6 8.3 1 1.4 Undecided .. ... 1 1.4 Table 12 Respondents' Opinions of Community Awareness of the RC&D Program Number of Percent Choices Responses of Total Very Well 2 ' 2 .8 ' Somewhat 14 19.4 Not Very Well 55 76.4 Undecided l 1.4 67 In view of the above responses, the results of the question concerning the respondents' willingness to change the program if they had the authority are not surprising. Forty-six percent of the decision makers felt they would change the program while 32 percent would not (see Table 13). Table 13 Respondents' Propensity to Change the RC&D Program Number of Percent Choices Responses of Total Yes 34 46.5 No 23 32.4 Undecided 15 21.1 Summary of Preliminary Findingg The preliminary results of the questionnaire survey indicate that local RC&D leadership in the NW Michigan RC&D project has a strong natural resource orientation. This favoritism of natural resource concerns is shown by ' such indicators as organizational membership, occupation and perceptions of RC&D contributions to problem solution. Over one-third of all membership was in Agriculture, Land or Recreation organizations. Over half of the steering committee members were farmers. The social system identi- fied by the respondents also pointed to the natural resource orientation. Influential organizations and key 68 leaders as identified by the RC&D leadership principally focused upon natural resources as well. Committee members were generally positive in their Opinions concerning RC&D effectiveness although such opinions were tempered by their more frequent choice of "Slightly Effective" as opposed to the more positive response. Further, the majority of respondents felt that program effectiveness is reflected in improved economic conditions. Most felt, however, that the program was not very well known within their communities. Since the objective of this study is to determine the underlying structure of positive opinions toward program effectiveness the data derived from the questionnaire was coded and organized for further analysis using the techniques described in Chapter III. From the above description of the raw data it can be observed that there are some relation- ships between the various Opinions, perceptions and socio- economic characteristics of the respondents. Yet there is no indication of the degree of relationship or the magnitude of variance or covariance between these factors to account for the formation of attitudes toward program effectiveness. The following description of the results of the statistical analysis will attempt to define those relationship more precisely. 69 Results of Statistical Analysis Noting earlier that the concept of attitudes represents many related and individual forms of behavior that are often intrinsically tied together by the same attitude forming processes, it is difficult to single out an isolated response from the entire attitude survey and assume it to be representative of closely related and interwoven attitudes and beliefs. Further, to identify the various relationships between attitudes it is more valuable from an explanatory point of view to scan all available data in order to discern relationships among actual responses rather than assume relationships on an a priori basis. In this study, initially all of the data were viewed as a single matrix in which all variables were correlated to all others. The purpose was- to identify unifying principles related to the respondents' characteristics and belief patterns. Concurrently, it was possible to eliminate obviously irrelevant phenomena from‘ further consideration. The task of identifying related variables was accomplished by employing simple cross-tabulation and simple correlation analysis. The cross-tabulation procedure involved compiling the number of mentions for each possible .response choice on the questionnaire in relation to all OHoEm ooo.H «an. Nae. mom. mNe. one. mad. men. mom. nmm. mHH. oHo. muumsocH ooo.H mom. con. man. MNN. mam. qu. omN. mod. coo.l wHH. .omz a Suammm ooo.H omm. mum. «om. oMN. mmN. ooN. nae. OHN.I NHN. coaumospm ooo.H Non. one. mmm. moo. mom. Nun. moo. NmH. .>oo w .cmHm ooo.H one. one. mun. Nho. wNo. mNo. th. .umouomm ooo.H Hue. mam. moo. «mm. 000.! Ave. huumouom ooo.H Mme. com. com. moo.l “NH. .Ofiuw< no ooo.H omw. «do. oNo. mmH. Houm3 HI ooo.H CNN. NNo. NON. pond ooo.H moo.l NwH. .GOHH>Gm ooo.H «no.1 .uomwwm .Hm ooo.H .uomwmm P n morn». m . e .& H... w V. 8 CW m. mm .4..an m nae CC rev:&0 .1pr mev Enummuhuasun mm...“ o iSOStaoeBCI rEC rmuwflmummumummu.fi mmmnmmemmehsmsmmu moanoaum> cowuouwuo mo genus: cowuoaouuoo «H manna 74 derived for each of the indicators in the maximum correlation show their relative importance in both the concept of "effectiveness" and the formation of the attitudes toward that concept. The technique of canonical analysis yields results ..1 in so-called "dimensions in which certain variables are isolated in relation to the variables in both sets. Only the first dimension of a possible seventeen in each of the first five canonical analyses will be reported here as the first extraction yields the maximum correlation for inter- pretative purposes. In addition, only those variables with weights, .4500 are reported. It is from this first and maximum correlation in each of the first five computations that the variables used in the sixth and final canonical model were chosen (see Tables 15 - 19). In the first canonical model, where the criterion set is correlated with the predictor set consisting of socio- 1"Dimensions" may be thought of as defining a multi- dimensional space with "attitudes of effectiveness" being 'defined by coordinates determined by the interaction of the independent variables within that space. Dimensions can also be thought of as seventeen distinct relationships of "effectiveness". Each of these relationships is a stage for a relatively distinct type of attitude toward program effectiveness. 75 economic indicators,1 the over-all canonical coefficient, R equals .9598. This means that there is near perfect agreement between the two sets of variables. The highest weight from the total of seventeen variables in the criterion set was assigned to "Water" as a problem area where RC&D contributes to problem solution and, hence, to program effectiveness. The predictor variables standing out were those related to education and occupation (see Table 15). These results can be interpreted as meaning that there is a negative association between the higher levels of education and positive opinions of program effectiveness in general, and to Opinions of RC&D's contribution to problem solving in water related areas specifically. This can be explained by assuming that higher levels of education tend to promote higher levels of critical judgement, longer spans of decision making or the suspension of judgement until further evidence can support positive conclusions. Of all the occupations indicated by the RC&D committee- men (20), the farming profession sees the RC&D effort as least effective. The other three occupational groups 1Predictor variables consisted of: age, education, jproximity to economic place, occupation, income. The total :number is thirty-nine. 76 Table 15 Results of the First Canonical Analysis 1. Canonical Coefficient, R = .9598, p I nnonmue3< .asoo I aux OHmm. Oman.I «cos. woueanmum I : : I max «Non. enmm.I mace. soaoam .DeaH I .>cm Hmooom I max came. oeun. new I woos I x uom uouoaooum anN.I .uuoooo .soom popouoaH I “fix ammo. quH.I .omm .soo I : : I max mmoo. wcemsom I : : I «ax Hoem. .nanne I z = I max mome.I mmmo. unmasooeam I z = I New osae.I seam. memo. snunseaH I = = I Hex mama. meam.I .emz e assume I = = I oax emm~.I momm. cowumosem I : : I ax mmmo.HI ameH.I .>mo e .anam I = = I we mama. ammo. coeunmnomm I z : I he omHh.l MONO. nahummhonw I : z I ON “Neo.I .uenm4 I = = I mm moo~.H mme~.H nouns I = z I ex w-m.oI mmoe.I sumo I = = I ma moms. comm.I .ne>am I .nonueoo mannnon> I Nx seeo.I mpeuummem souawoom I ox owes. m>euummom I x umm cowuouauo moaomflum> ommm. mmnm. moww. Home. omNm. coca. «Hoe. Amo.uVoV unmeoemmmoo Hmoecocmo n o m e m N H Auoomv coamcmseo mfimhamo< HOOfioosoo nuxwm mo muHOmom oN OHan 86 won. onm.I omNm. manm.l oomq. onN.MI oqmw.NI smhw.NI omoH.NI Hoom.HI onmo.l woes. mowo.l onm. oemm.HI NHOQ.HI mem.HI NmoH.HI nwow.l oomo.l «How.I Noam.l ahon.l oHHm. ONHN. 5F wnnm. Home. smoo.l moon. omam. oHom. Noon. ooom.l mmao. «mom. Hoco.l Neon. omoa. omwN. moom.l mNoN. homo. omen. momm. Nmmm.l HomH.I mama. ano. oNHN. mNNN. Hooa.l meH.I noHH. NHNm.I memo.l NmoN.HI NooN.HI «Noo.NI oeoo.al mmma.HI Hema. ooNq.I ono. Nona.l Noon. mmwN.I mMNo. meN.I Hmom.l oqam.l onlom I mNION I oN OOOH I huosou ca .mmm mumow cowumooom I cowuooouom Hmoo omu I ucoam>ao>cH smegma mo some soaumouoom I coaumuuoomsmua I ownmuonaoz .moomlme : .mdoml: : .ooMINe : .oom .BOUI: : Q a onl: : .Huwcm I uomusou HafiuecH smooawom Ha>fio I .- 535 one: uoow< coamcouxm I soaumosooo HQUWQHOW l HOEHMW I made oo I : mmlom l : G€l0¢ l : ¢MIOM l : mNION I .£OM2 5 wcwoflmufi umnuo I oumooouo owoaaoo omOHHoo oaom Sago Hoonom swam .m.m .Hoaoo uoc can emerge ou zuamcoooum O>Huo¢ “momma Hmuocoo .>HO>OH ofinmuoooog Hooog mum> uoz I nogamaom I mumcmhmaw..aaou -— .- = = = _— = .- .mom memo» doaumooom NMx'Tl-n Int-filfiln d)O‘C>r4 <><><><><><><><>Fi NNMM NNNNNNNNNNNNNNNNNNNMNNN h- N I mmoooum3¢ .aaoo IONx fiONHHfiDMUW I : : IaHN «New. seem. eeHe.I soaoom .neaH I .>nm annuom Imox mom I oaom I x uom uouoaooum NNoo. ONHm.I .uuoooo .ooom oo>ouoaH Imwx neNe.I .une .aoo I = = Imox wchoom I : : Iqax mmmn.| .msmue I : : Imax usoahoaoam I : : I HN Haas. AnnmaeaH I = = IMHN omom.l .ooz e anamom I : : IOHN GOHUMUGUM I = : I GM moon. .>oo w .smHm I : : I ox W comm. cowumouoom I : : I fix eomn.I snunmnos I = = I x mmmm. maew.I .onnwa I = = I mm kum3 I = z I fix Nmom. ooma I : : I mx omom. OHON. .HH>cm I .AHHOOOU OHnmsHm> I Nx m>euomoom seeaweom I fix o>Huoowmm I N mom coauoufiuu moanmeuo> Honm. «moo. mode. Home. mNmo. moon. moan. powwowwmooo Hmoacocoo «a ma NH HH oH a m Auoomv oowmooafio Ae.uao0o om manna 88 oomm. moon. owmh.I amoo.l aNmo.I Nnmm.l oeam.al onno.l anon.l m¢mm.l nwem.l mono.l owno.l «NNo.I Noam. QONw.I oon.I nnoN.aI wN¢N.I anwm.l cmmm.l mmom.I ¢¢o0.I mono. neon. ooom. maoo. amen. oooo.l ommo. NmoN.aI omNN.NI mmN¢.aI mmnm.al ommm.al mmNn. mmcm.l aomm. owao. nomm.al cmoa.N woMN.aI oooa.m oaoa.al mNao.N NmNm.aI aNm¢.N momm.al aoow.a Non. mNNm.I sumo. aces. Namq.l ooam. menu. omma.a qum.a mamm.a awqm.a mmmq.a mmlom I z : : : GNION I : : = : ON mmoa I mussou ca .mom mumow coaumoopm I coauooouom amoo omu I usoaopao>oa Hoomoa mo ooha coaumouoom I : mono. coauouuoomcmua I oanmumnsoz . mdomlm x» : = : .mfioml: : : : .UOMINfi : : : .omm .EOUI: : : : Q Q .Hml: : = : ..nHw4I: : : vomAIae .omwuo amauoosamoa umoz usoaooua>sm I uoousoo amauaoa noooawam aa>ao I : umumouom I : awaken I : uoow¢ ooamoouxm I coaumooooo MSHQ 00 I : : : : mmIOm I : : : : QwONol OQIOG I z : : : Hhmw.l mmlom I : : : : mnauo< smegma aouosoo .>ao>ca oanmuoooma aoooq Hwy? ...—02 I = : umnsmsom I encompass .aaoo >4 MN?!“ Inmtn N Ln OI—I mm NMNNNNNNNNNNNNNNNNNNxNNN NNNNNNNNN 0‘ \1' (D \‘f [x cm aoaoom I ax mom I oaom Iw x uom uouoaooum .uuoooo .ooom oo>ouoaa Immx .UQW .500 I : : IMHN wGHmDOm I : : IQHN .mGMHH I : : IMHX ucoaaoaoam I : : INax hhumDGGH I z : IHHN .Umz w SUHQOE I : : IOHN GOHUQUDUM I : : I 0% .>wQ w .Gme I : : I wx oqnm. coaumouoom I : : I Nx muumouom I : : I ox .UHHw< I : : I WM Hmum3 I : : I QM WMHNoHI figmg I .- 2 I M“ .ua>sm I .nauusoo oanmoam> I x m>nuumoom nonsmoom I we o>auoowwm I x mom soauouauo moanmaum> maqm. ucoaoamwoou amoasocmo ma Auoomv coamcmaao Ae.uaouO oN manna 9O 4.. ohms. NmNN.a whom. wamo.a ooam. mooo. maNm. ooow. .m.m .Heaoo no: use I consensus I owcmno Ou Nuandgooum I o>auo< smegma awesome I .>ao>ca magmuoomoa amooa I hhm> UOZ I : : I a'me-rmo NNNNNN mmx mmlom I = : : : Ion GNION I = = : : Immx ON mmoa I husoou ca .mmm mumow I doaumooom I coauooouom amoo Ime umu I usoao>ao>sa umomoa mo dose Iamx coaumouomm I : Iomx ooaumuuoomsmue I oanmuonaoz Imwx . mflomlm «x : : : IN QM . mSOMI: : = : I®¢N . UOMIN «x = : = I N .Oo Jam—GUI: : , : : IMWN D Q . HmI: : : = Im QN .HHMcm I uumuooo amauasa Iwwx someones Ho>oo I : Iamx Houwouom I = I N Hookah I : Iwmx usow< soamoouxm I coaumooooo Immx mfiflfl 09 I : = : : ImmN wmlom I : = : : IdmN m¢I0¢ I : : : : ImmN @MIOM I : : : : INMN mNIom I .noaz ea .mom mumow Iamx waoaonne nurse I : Iomx muooomuo owoaaoo I : I N owoaaou maom I : Imwx mono Hooaom snow I = I x mm x N N x N amx unsaoaom I : News I mmmcm~ma< .-GUI Ia 89 I IIIlll Iom vuuHHNAMum I : : aHN haaoam .HQEH I .>am Hmaoom ImHN com I maom I N umm wouuavmum .uuoaao .aoom vm>ouaaH ImwN 00““ oaoo II = = lme wcfimaom I : : IqHN .meHH I z : IMHX unmahoamam I : : INHN muumawcH I : : IHHN .wmz a nuammm I : : IOHN acaumoswm I : : I mN .>OQ ¢ .GMHm I z : I mN oqmm. cowummuomm I : : I hN huummuom I : : I oN oUHHw< I : : I fix H0um3 I : : I «N mmHN.HI wand I : : I mN .HH>:m I .nwuucoo manmsam> I NN m>fiuummmm Naunmfiam I fix m>fiuommmm I N umm aOHumuHHo mmHAMflum> “Hem. ucmfiofimmmoo Hmoaaoamu ma Auoomv seawamaao Au.uaouV om manna 90 {ii mums. NmnN.H whom. meo.H ooam. mooo. MHNm. omow. mmlom I : : : : IMMN mNEON I : z : : Immx ON mama I hudaou aw .mmm mummw I N defiumusvm I coaummoumm Hmou INmN omo I unmam>ao>aH umwmmg mo mama IHmN aowummuowm I : IomN nowumuuommamue I awnmumnamz IMMN .mflomlm* : : : INQN .m50MI: : : : IQQN .UQMINfi : : = I N .um .meOUI: : : : IMWN Q Q .HmI: : = : IMGN .HHwum I nonuaoo HMHuHGH IHcN nomaflwam HH>ao I = qux “mummuom I : ImmN Hmaumm I : Ime uamw< coamamuxm I coaumasouo ImmN MSHQ CO I : : :. : ImmN mMIOm I : : : : I¢MN m¢l0¢ I : : : : ImmN QMIom I : : : : INMN mNIoN I .nowz afi .mmm mummw I N wawaamuH umnuo I : IHmN muwavmuo mwmaaoo I : ImmN mmeHou meow I : IwNN haao Hoonum swam I : Inmx .m.m .anou uoa wan I defiumonvm I N mwamnu ou hufimdmaoum IoNN 0>Huu< umvmma Hmumamo InNN .>Ho>uH aaamumvmmg Hmuoq IemN hum> uoz I : = ImuN ungamaom I = = INmN .Nuo> I mmodmum3< .aaou IHNN 91 opinions the correlates indicating their role as SCD director (.5260), their perception of a slowly improving social environment (.6615), their relatively low educational achievement (.5402), i.e., did not complete high school, and their membership in transportation related organizations (.7486) are positively weighted. Moreover, these attitudes are negatively weighted in relation to a feeling that the RC&D program in "somewhat" and "not very well" known within the community (-.5146 and -.5051). It also appears that such opinions are highly correlated to years residence in Michigan. With increasing length of tenure there exists increasing dissatisfaction with the program's effectiveness. Farmers, likewise, view the program as ineffective (-.5217) in terms of its contribution to water and transportation problem solution. Such relationships appear to make sense. A soil conservation district affiliated role would likely be concerned with water resources and would judge his role as decision maker within the RC&D effort in a positive way .in relation to such contributions. This type of relation- ship becomes important when Speaking to the notion of positive attitudes functioning to favor positive outcomes. It also follows that the decision maker who sees his community's social environment as improving would find reason to assess the program, not only in natural resource develOpment but in human resource development, in a 92 positive way. Membership in a transportation oriented organization (e.g., county road commission) would also tend to correlate positively with attitudes toward valuable contributions in the field of transportation. Further, it can be noted in Table 20 that as education levels rise there is a general decrease in the strength of association to positive Opinions of program effectiveness. Here again, those persons to whom the program theoretically should provide the most benefits in human resource develop- ment areas regard the program effort more favorably, while those more highly educated possess a more critical judgement and longer periods of assessment in light of actual out- comes. Residence and tenure, likewise, seems to play an important role in attitude formation. Although all categories dealing with length of residence in Michigan are negatively correlated, it is interesting to again note in this sixth model that increasing length of residence within the state up to "60 plus" years evokes increasingly higher negative weights. This is assumed to be the result of, as mentioned earlier, a witnessing, on the part of the long-time residents, of what they perceive as a continuing deterioration of their environment and a negative view of any significant program achievement. The shorter term Iresidents, on the other hand, have witnessed no similar Ilong-term deterioration and can look to the future with 93 perceived longer planning horizons and therefore, at least, somewhat less skepticism with regard to the program and its effectiveness. It can also be noted from Table 20 that "Land" and "Agriculture" organizations are weakly correlated to positive attitudes. Although farmers are well represented in the decision making body, their perception of influential organizations representing their vested interests is weak. They, it is assumed, then feel that agriculture in general is receiving little attention and view the program from that perspective. This same feeling is noted in the criterion set as well where the program's contributions to problem solution in agriculturally related areas is weakly weighted (-.0727). The second dimension with a canonical coefficient of .9406, emphasizes the RC&D contribution to the solution of "Land" (-l.3278) and "Water" (1.2015) related problems.' These variables are associated with the composite predictor variables of: education - high school only (-.6401), . years residence in Michigan - 40-49 years (.5057), 50-59 years (.6198), and years residence in county - less than 20 years (-.5964). This dimension can be interpreted as meaning that those respondents having lived within the State longer periods of time will View the RC&D program an ineffective. 'rhose respondents with one or more of the above attributes 94 or Opinions will have lived in their county of residence for less than twenty years and will have completed high school. The third dimension with a canonical coefficient of -9296, emphasizes RC&D's valuable contribution to the solution of Recreation proglems (.9993). Planning and Development is also highly weighted (-l.0953). Associated predictor correlates include: opinions that the community is only "somewhat" aware of RC&D (.5667), high school education level (.5019), and a perception that an influential organization within the community is identified with Housing (.5196). Dimension four (.9241) interestingly isolates "Industry" (.5346) and "Employment" (-.6219). Their predictor correlates include: a perception of an improving social environment (-.5874) and a stabilized social environment (-.7390). Although the decision makers holding the perception of the social environment as stabilized to improving would logically view the RC&D contribution to employment as .favorable, they, on the other hand, are less likely to view RC&D as contributing to industrial development within the region. The respondents holding such opinions, in addition, will tend to be long-time residents of both the State and county. Farmers (.5093) tend to view employment negatively but view the industrial develOpment contribution jpositively. Those respondents with the above views and 95 characteristics also recognize housing type organizations (-.6697) as influential and see at least one of their community leaders involved with RC&D at the county steering committee level (.6301). Dimension five shows an increasingly diminishing canonical coefficient (.8805). When the canonical coefficients decrease rapidly the researcher should be cautious in further interpretation of following dimensions. In this case, even though "Education" (-.7536) and "Health and Medical Services" (.5383) are highly weighted as positive contributions to program effectiveness, the predictor correlates with high canonical weights are less emphatic. The results of this fifth extraction can cautiously be interpreted as meaning that those respondents not having completed high school or with some college or college degree tend to increasingly hold the Opinion that RC&D has contributed to the solution of education problems. The decision makers with those characteristics are, however, negatively associated with opinions concerning RC&D's . contribution to problem solution in the field of Health and Medical Services. These same persons view the social environ- ment as stabilized and tend to feel that the program, if not very well known, is at least somewhat known to the residents of the RC&D area. I; 96 In the sixth extraction (canonical coefficient, .8583) the factor related to the program's valuable contribution to the solution of "environment“ problems is isolated. ”Forestry" is nearly equal in weight but possesses the Opposite sign. The predictor correlates most highly associated with this dimension are: the occupational category - extension agent, years residence in Michigan, opinions Of a stabilized social environment, and a feeling that the program is not very well known in the community. Logically the occupation of agricultural extension agent would find RC&D as a valuable contributor to the solution Of environmental problems as the agricultural extension role is more concerned with natural resource problems. The respondents holding those Opinions are increasingly correlated in an inverse way with longer periods of residence within the state. Beyond the sixth extraction (see Table 20) the relationships between criterion and predictor variables appear to become more obscure and tend to be less realistic. .The relationship between weights as well as direction of their signs appear without a recognizable pattern. The sixth dimension, then, is a logical departure from further analysis and interpretation. The remaining dimensions are reported in Table 20 for comparative purposes only. 97 Summary of Sixth Canonical Analysis In summary, the occupation of farming, the role of SCD director as related to RC&D, years residence in Michigan, education and opinions regarding the community's awareness of the RC&D effort are variables most closely associated with Opinions Of program effectiveness. From these results of canonical analysis an equation for relating positive Opinions of program effectiveness to the predictor variables will be constructed and discussed in the following chapter. Factor Analysis With Canonical Analysis The Economic Research Service data file provided a listing of all factors isolated by factor analyzing all 368 variables from the questionnaire survey (see Appendix B). The list of fifty factors was used in canonical analysis as the predictor set. In this seventh canonical correlation the criterion set consisted of the same seventeen variables used in previous computations. Interestingly, the results of the factor analysis singled out nearly the same variables for further analysis as did the methods of preliminary analysis used in the first six Icanonical computations. Factor analysis did, however, identify a series of variables related to the respondents' recognition of certain problem categories. These twenty ‘variables were not included in the original five canonical correlations as predictor variables. 98 The use of the Economic Research Service data derived from factor analyzing all variables provided further data from which to assess the formation of positive attitudes toward program effectiveness. The results of this seventh canonical correlation is reported in Table 21. Only the first dimension is reported and interpreted. The results are then compared in equation form to previously discussed canonical correlation results. The first dimension of this seventh canonical correlation possesses a canonical coefficient relating both sets of: .9991. The first dimension isolates RC&D's valuable contribution to the solution of Health and Medical Services problems (.7516). Those respondents holding that particular view of RC&D effectiveness also see the natural environment as improving (.6967), see their program leaders as active (.4992) and tend to be in the higher income groups (.8756). There is an inverse association between the positive view of program effective- . ness and the Opinion regarding a lack of industrial development (-.6276), formal county goals (-.6219), familiarity with RC&D (-.6855 and -.5191), and membership in Land and Agricultural organizations. The associations identified by canonical analysis logically pattern themselves. Those respondents holding a positive opinion of program effectiveness from the point €39 cm mema. msauoxumz I mamaboum huumouom I N «Hoe. oaooaH I maOHnoum ousuasowuw< I oNN 0Nm~.I sowusaaom I mapooum nous: I wNN oo-.I ucoaooam>mo Hmwuumsocu I = = I NNN oomH.I cowmoum I Emaboum vs I bNN swm~.I mm: sang magmamno I = = I mNN mama. xusmmm Nonnumz I mamaboum Hmucmasouw>cm I qNN mmso.I .uoqumuma I = = = I mmx mem~.I wcw>ouoaH I sofiummouwm ucmaaouw>cm Hmfioom I NNN mqmq. .noaumumo I = z = I mwx womb. wcw>ouaEH I cofluomoumm unmacouw>cm HODODOZ I N OMQN.I wCfiCwHUQQ I : : I OHX Noam.I wcwzouo I cowuaoouom owaocoom I mHN noboamoum bmom.l maficsusocmo owaocoom po>ouaBH I NAN oon.I .nmm s .ums .aoo I = = I osx smmm.I wcflmsom I : z I max namo.l cowumuuoamcmue I : : I qHN Nowa. osmazoaoam I : : I MHN amoo.I suumacca I = z I max cam“. .86: a gunmen I z = I mwx obo~.I OOHumosom I : : I mN omoa. .>oa w .cmam I : : I N mwmm. cowummuomm I : : I MWN qwmc. >uumwuom I : : I NN “moo.I musnasogumc I = = I ox Home. umumz I : : I mN anom.I cams I = = I «x Hmmn. ucwscouw>cm I cofiusbfluucou mabmsam> I mN NNwH.I o>fluumcom saunmfism I Nx mmqm. m>auomwwm I Ax cowuwuwuo Naomm. u uOOMOflwwmoo Hmowcocmov munwwm3 amuwcocmu moabmwum> Dom HOuONOOOm mm mfimzamc< HOuomm mo muasmmm mnu wcwm: mamzams< HOOficocmu mo muaammm Hm canoe l()0 mo..vm mmqo. N0NH.I nmbm.l Nomn.I eomfl.l mma0.I mmmq. oomm. emnm. mmom. wam~.I coma. amaq. ommH. oomm. ammo.l mmwm.l owHN. oqmc. Haoq. Nome. Hoam.l mmwo.l oamo.l omoo.l mmmm. ommo. cqu.I bmmo.l wwom. mmqm.I moon. mnnm. swoo.l Hmuu. moan. oomH.I .Hm>oo a wcflssmHm I magmHObSOZ coaummuomm I ousuasuauw< I puma I : .>umm a .Omm .3800 I : wcwmsom I mfinmuwbamz .>pmm w .omm .6600 I : : musuasowuw< I uomucou Hmfiuwcu mafia ooo.omm cu Hoo.mflm I : ooo.me cu ooo.mm can» mmmH I maooaH umcao no umwmamz I : umaumm I cowumaoooo OOHHB moan OH I momam .soom ou wocmumfln .vmuo wwmaaou w mwmaaoo meow I : HOO£om swam muoaaaoo no: can I cowumoscm OOHO ObIom I : msIOM I mw< .5800 .um .00 .mHW I : : : mom .mu» I OHON vmumamm a oussme m>fiuo< NH m>oumEH m>wuu< “momma Hmumcmu HH03 hHm> UOZ : : : unnamaom I Qwom :ufia zuaumwafiamm mHmOU mucsoo Hmauom .>QQ .UHDU @ .umHm I : : : : .umfio w .Hddnm umumS I .boum .>uom w .omm .EOU .mom a .OOH mo Nomq I : : wcflcfimue mo 30mg I mamfiboum ucmakoaaam .HO>OQ mo Nona I mamaboum %uumsvcH .HHUmh I : : : : .muom .uoz I .Hboum .>umm .Omz w nuammm Hmsoaumoo> I mamaboum sowumospm swam .aaou I maoaboum .Hw>wo w .cmam wcwocmcwm I unmaaoam>mvum>o I unmaaoao>ooumusa I : : mmoou< owasm I mamaboum COHummuomm NI \0 r-INMQU‘ND \O\O\O\D\D\O (DO‘O Int-00 ViXIX N:K:K=<><>¢fl:¥:x h- In OHNMQ‘U‘O mmInLnInInIn Q'U'IONGDO‘ -I>J>-I>«I> O‘J-‘NONH LOI-‘I-‘VUI ...: I—‘NN “MN I ' Variable Name Effective Effective Slightly 143 Variable Names Slightly Effective Effective OCCUPATION Public Adm. Extension Agent Teacher Construction Professor Farmer Mgr. & Owner Librarian Machinist Forester Postmaster Civil Engineer Store Keeper Plumber Real Estate Painter Auctioneer Salesman Labor Relations Retired Work in Co. of Res. INCOME Less than $5,000 $5,001-10,000 $10,001-15,000 $15,001-20,000 $20,000 plus INITIAL CONTACT Environment Land Agriculture Plan. & Devel. Industry Transportation Comm. Fac. & Serv. MOST INFL. ORGAN. #1 Environment Land Water Agriculture Forestry Recreation Plan & Devel. Education Industry Transportation H O‘U'IOOOOI-‘I-‘OOI-‘OHI-‘OVI-‘NWON w H WUIQO‘UI CDCOJ-‘NOO OI--'I-‘mOI'-‘OI-‘CJ'IHI l-" w-I-‘J-‘OO ”O‘HOOHWOHOONOONHOI—‘ONO WOI-‘OWI-‘I-I I-‘I—‘OUOI-‘OWOI-‘O Comm. Fac. & Serv. MOST INFL. ORGAN. #2 Land Agriculture Forestry Recreation Plan. & Devel. Education Transportation Housing Comm. Fae. & Serv. MOST INFL. ORGAN. #3 Land Water Agriculture Forestry Plan. & Devel. Health & Medical Industry Transportation Housing Comm. Fac. & Serv. MEMBERSHIP Education Health & Medical Industry Employment Transportation Housing Comm. Fac. & Serv. Environment Land Water Agriculture Forestry Recreation Plan. & Devel. Civic Group Social Group Religious Group TYPE LEADER Involv. #1 Co. Steering Comm. Advisor Sponsor Indirectly \INI-‘I-‘tfioowlv ...: F'P‘CDCDP‘hDhJ\Jk‘P‘ OOHOI—‘OOOI—‘I—‘I—‘OOOOI—‘H U'IOOO QONONI-‘I-‘UN NOOHONOUOH p I-' I-‘ h‘h‘ ha hi hihiP'\J$~UIG>C>O\O\hJP‘~JCD 12 Variable Name Slightly 144 Effective Effective Variable Name Slightly Effective Effective TYPE LEADER INVOLV. #2 Co. Steering Comm. Advisor Sponsor Indirectly TYPE LEADER INVOLV. #3 Co. Steering Comm. Advisor Indirectly GOAL PERCEPTIONS Environment Land Water Agriculture Forestry Recreation Plan & Devel. Education Industry Employment Transportation Housing Comm. Fac. & Serv. UINI-‘m b-I-‘m N WOWONH-L‘mNHOO‘N L‘ONW 9H9.) H O‘I—‘I—‘NOOI—‘CDNNNUJW APPENDIX D Correlation Matrix of Positive Opinions of Program Effectiveness and Predictor Variables Variables 20 21 Variables 20 21 1 .099 .069 244 .099 -.125 2 .338 -.215 245 -.003 .031 3 .058 -.107 246 .052 .020 5 -.023 .148 247 -.l77 .143 6 -.O94 -.007 248 -.115 .121 7 .128 -.159 249 .1419 -.211 10 .203 —.062 250 .1870 -.028 11 -.208 .121 252 .114 -.019 12 -.018 .051 253 -.169 .006 13 -.025 -.075 254 .155 -.105 14 .003 .081 255 -.129 .191 15 .009 .049 256 .196 -.075 16 .037 -.111 257 -.151 -.032 17 -.087 -.129 258 -.023 .079 18 .109 -.O74 260 .0523 -.129 142 .009 .171 261 -.227 .181 144 —.003 .091 262 .192 -.129 146 -.O45 .019 263 .052 .026 148 -.014 .084 264 .109 -.074 149 .1702 .009 265 .345 -.329 150 —.183 .020 266 -.048 .143 152 -.099 .021 267 .109 -.074 154 -.O45 .024 268 .109 -.174 156 .022 .032 269 -.227 .181 158 .022 .032 270 .109 -.O74 160 .049 -.135 271 -.129 -.O74 161 -.302 .159 272 -.184 .084 220 .155 -.075 273 .109 —.074 222 '.199 -.172 274 -.l42 .256 226 .115 -.114 275 -.129 .191 230 —.029 -.022 276 -.129 -.O74 231 -.014 .020 277 -.129 -.O74 232 .027 —.062 278 -.129 .191 233 .010 .100 279 -.148 .193 234 .208 .099 280 .025 -.037 235 .038 .079 281 .114 -.204 236 .155 -.105 282 .079 .096 237 -.094 -.032 283 -.082 -.089 238 -.075 .031 284 -.074 .081 239 -.084 .011 285 -.025 .149 241 -.118 .175 287 —.129 .191 242 -.003 .031 288 .142 -.210 243 -.045 -.033 289 -.139 .186 145 146 Variables 20 21 Variables 20 21 290 .223 -.150 337 .045 -.019 291 -.129 .191 338 -.O45 .019 292 -.129 .074 339 .156 -.108 293 .029 .069 340 -.027 .041 294 .109 -.O74 341 -.144 .262 295 .208 —.O99 342 .016 -.O72 296 .109 -.O74 343 -.087 -.026 297 .089 -.108 344 .019 -.099 298 .109 -.O74 345 -.O87 .026 299 -.264 —.015 346 -.O34 .199 300 .029 -.069 347 .029 -.069 301 .1092 -.O74 348 —.227 .037 302 —.087 .026 349 .155 -.105 303 -.129 .191 350 .010 .141 304 -.114 .239 351 .064 -.O49 305 -.O78 .015 352 .075 -.O75 306 .099 -.069 353 -.O74 .141 307 -.129 .191 354 -.085 .349 308 -.184 .094 355 .032 .117 309 -.152 -.091 356 .003 -.005 310 -.014 -.105 357 -.046 .211 311 —.087 .181 358 -.087 .181 312 .155 -.105 359 .027 .062 313 .137 .032 360 -.111 .231 314 —.014 .084 361 .127 -.O92 315 .109 -.074 362 .109 -.O74 316 .037 -.028 363 .155 -.105 317 .109 -.O74 364 -.174 .258 318 —.126 -.137 365 .101 -.015 319 .109 -.O74 366 -.129 .191 320 —.129 .191 367 -.208 .230 321 -.129 -.O74 322 .109 -.O74 323 .042 .042 324 -.084 .011 325 -.112 .041 326 -.142 .120 327 -.126 .149 328 .208 —.204 329 -.208 .204 330 —.037 .055 331 -.207 .134 332 .029 .165 333 .053 -.014 334 .223 -.150 335 .137 -.224 336 —.187 .196