BUREAUCRATIC STRUCTURE. RAHONAL BUREAUCRATS AND BUREAU POLICY; Dissertation! for the Degree of Ph. D. MlCHIGAN STATE UNIVERSITY HAROLD EVANS OLD, JR. 1973 {2313 : , “932. 1 ‘ This is to certify that the thesis entitled BUREAUCRATIC STRUCTURE, RATIONAL BUREAUCRATS AND BUREAU POLICY ‘ ' ' V presented by Harold Evans 01d, Jr. has been accepted towards fulfillment of the requirements for Ph . D . degree in Department of Political Science nugfil,‘é{<524h1&ikdaégn_ i Major professor Date 7‘28'73 * 0-7639 g IIN‘DINB av 1: "MB 8 SUNS' 900K BINDERY INC. LIBRARY BINDE RS srmuerop}. talcum] l ABSTRACT BUREAUCRATIC STRUCTURE, RATIONAL BUREAUCRATS AND BUREAU POLICY By Harold Evans 01d, Jr. Public bureaucracies are playing an ever-increasing role in the functioning of modern society. Whether this is good or bad is a ques- tion which many thoughtful individuals have given attention to. To seek intelligent, reasoned answers to the questions that surround the role bureaucracy should play requires an increase in our knowledge and understanding of bureau behavior. In an effort to expand this knowledge, this dissertation focuses on a particular type of bureau- cracy-~public school administrations--and one associated policy area and suggests a model of bureaucratic decision-making for future research into bureaucratic behavior. In reviewing the appropriate literature, the most directly suggestive and productive approach to the study of bureaucracy found is Inside Bureaucracy by Anthony Downs. Here he attempts to develop a formal theory which explains and predicts bureaucratic behavior. It serves as one of the two principal theoretical sources for this inves- tigation. The theoretical stimulus is Schlesinger's work on political ambition, Ambition and Politics. While tested in an electoral context, Harold Evans 01d, Jr. the concepts of ambition and structure of political Opportunity are most helpful in this research into bureaucratic behavior. The central assumption buttressing this dissertation is the belief that bureaucrats--school administrators--1ike other rational, self-interested political actors have different preference orderings (ambitions) and that they make decisions which reflect these prefer- ences even when acting in their official roles. Extremely important to these rational bureaucrats and their actions is the environment (opportunity structure) of the bureau in which they serve. The Oppor- tunity structure or environment of the bureau helps give direction to a bureaucrat's goals and aids him in obtaining or maximizing those goals. It is important then to understand the nature of the Oppor- tunity structure bureaucrats face if we are to increase our knowledge of bureau behavior. Examining this environment or opportunity structure of school bureaucracies faced by bureaucrats is the principal empirical research conducted here. The Opportunity structure of school bureaus was believed to have a number of characteristics or dimensions which were potentially suggestive to this effort. The dimensions included age of bureau, growth rate of the bureau, administrative turnover rate of bureau, degree of professional specialization, recruitment pattern of bureau, degree of rule formalization, level of financial support and size of bureau. Each of the individual dimensions was hypothesized as having an independent relationship to bureau policy-~in this case educational innovation as measured by use of Title III, Educational and Secondary Education Act of 1965. At the same time, it was also Harold Evans 01d, Jr. believed that these eight dimensions could be combined into two clusters of dimensions which helped give direction to a bureaucrat's goals and select an apprOpriate strategy to maximize those goals--a career cluster and a resource cluster. As with the individual dimen- sions of the opportunity structure, these two clusters were felt to be directly associated with bureau policy. The ten hypotheses formulated for the eight individual dimen- sions and two dimension clusters of the opportunity structure were tested using 508 Michigan school districts which meet the disserta— tion's definition of bureaucracy. Each hypothesis prOposed the relationship between the dimension or dimension cluster and the amount of educational innovation to be found in a school district. With the low level of current knowledge of the environment of bureaucracies, it was sometimes an empirical question as to what the relationship would be between the dimension and policy. The overall results of testing these hypotheses is mixed. 0f the eight original individual dimensions suggested as being related to educational innovation, only four dimensions-~size of bureau, degree of professional Specialization, level of financial support, and recruitment pattern of bureau--display relationships which are in the anticipated direction. Degree of rule formalization and administrative turnover rate of bureau are associated with educational innovation but not in the hypothesized manner. The other two individual dimensions--age and growth rate-~did not exhibit any correlation to the amount of educational innovation in a school district. The career cluster composed of age, growth rate and administrative turnover rate of bureau is not related to educational Harold Evans Old, Jr. innovation as originally suggested. The resource cluster is related to educational innovation but it is not composed of the other five dimen- sions as hypothesized--only size, degree Of professional specialization and level of financial make-up the resource cluster. A number Of alternative indicators for the dimensions are suggested for future research along with additional factors that may impinge upon any rela- tionship between the dimensions of the Opportunity structure and bureau policy. Finally, a bureaucratic model is developed which incorporates concepts from Downs and Schlesinger along with the results of testing the hypothesis formulated for the Opportunity structure. This model attempts to show how bureaucrats select their goals and their goal maximizing strategies in reSponse to the environment--the opportunity structure--they face and how this, in turn, is reflected in bureau policy. An example Of how bureaucrats might respond to one particular strategy to Obtain their goals is elaborated upon. Limited empirical evidence is presented which supports certain portions Of the model as formulated. Research strategies to test the model are prOposed for subsequent investigations into bureaucratic behavior. BUREAUCRATIC STRUCTURE, RATIONAL BUREAUCRATS AND BUREAU POLICY BY Harold Evans 01d, Jr. A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree Of DOCTOR OF PHILOSOPHY Department of Political Science 1973 A. i‘ ‘0 \;§ ACKNOWLEDGMENTS I wish to thank the chairman of my committee, Dr. Joseph A. Schlesinger, and the other members, Dr. LeRoy Ferguson and Dr. Harold J. Spaeth, for their assistance and COOperation 1n the writing of this dissertation. Dr. Schlesinger as chairman and teacher has contributed greatly to my understanding of the nature of the political process, both in the electoral arena and the world of bureaucracy. I appreciate Dr. Spaeth's willingness to fill a void and serve as the third member of my committee after Dr. David B. Meltz joined the faculty of Georgia Institute of Technology. In addition to those who have served on my committee, I would like to acknowledge the help of two colleagues, Dr. Fred Grupp, University of Connecticut and Mr. Gary Rassel, North Carolina State University at Raleigh. Fred Grupp provided a number of insights and suggestions which were most helpful to me in the analysis of the data in Chapters IV and V. Gary Rassel was a most sympathetic listener throughout the long period of time that it takes to bring a disserta- tion to fruition. Finally, a Special thanks is due to my wife, Georgia, and our two sons, Eric and Philip, who have had to bear the brunt of the ii frustration that is inherent in the writing of a dissertation. They have done so with humor and love which has made this task much easier than it might have been. iii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . ix Chapter I. AN INTRODUCTION AND GENERAL DISCUSSION OF BUREAUCRACY AND OPPORTUNITY STRUCTURE . . . . . . 1 II. LITERATURE OF EDUCATION AND BUREAUCRACY . . . . . . 8 Education and Educational Administration Organization Theory and Bureaucracy Summary of Literature III. THE RELATIONSHIP BETWEEN THE OPPORTUNITY STRUCTURE OF A BUREAU AND BUREAU POLICY . . . . . . . . . 37 Theoretical Approach to Bureaucracy Dimensions of the Opportunity Structure The Opportunity Structure and Educational Innovation Data and Measurement IV. TESTS OF THE INDIVIDUAL DIMENSIONS OF THE OPPORTUNITY STRUCTURE . . . . . . . . . . . 88 Hypothesis One Hypothesis Two Hypothesis Three Hypothesis Four Hypothesis Five Hypothesis Six Hypothesis Seven Hypothesis Eight Evaluation of Findings iv Chapter V. VI. TESTS OF THE DIMENSION CLUSTERS OF THE OPPORTUNITY STRUCTURE . Career Cluster Hypothesis Resource Cluster Hypothesis Evaluation of Findings SUMMARY OF RESULTS OF HYPOTHESES TESTING AND PROPOSED MODEL OF BUREAUCRATIC DECISION-MAKING Summary of Results of Hypotheses Testing Proposed Bureaucratic Decision-Model Proposed Research Strategy APPENDICES Appendix I. II. III. IV. V. VI. VII. VIII. IX. Control Control Control Control Control Control Control Control Relationships Relationships Relationships Relationships Relationships Relationships Relationships Relationships of Hypothesis of Hypothesis of Hypothesis of Hypothesis of Hypothesis of Hypothesis of Hypothesis of Hypothesis One Two Three . Four Five Six Seven . Eight . Gammas for Resource Dimension Clusters Actually Found SELECTED BIBLIOGRAPHY . Page 139 170 226 227 228 229 230 231 232 233 234 236 Table 3-1. 3-2. 4-1. 4-2. 4-3. 4-5. 4-6. 4-7. 4-8. 4-9. LIST OF TABLES Career Cluster Patterns of the Opportunity Structure of School Bureaucracy and Associated Levels of Educational Innovation . Polar Resource Cluster Patterns of the Opportunity Structure of School Bureaucracy and Educational Innovation in Bureaus Relation of Level of Financial Support of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . . . . . Relation of Level of Financial Support of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Relation of Age of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Relation of Age of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Relation of Age of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Relation of Age of Bureau and Educational Innovation (Number of Title III PrOposals Submitted) Relation of Growth Rate of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . Relation of Growth Rate of Bureau and Educational Innovation (Number of Title III PrOposals Smej-tted) O I O I O O O O I 0 Relation of Administrative Turnover Rate of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . . . . . . . . vi Page 66 67 92 93 94 95 96 97 98 99 101 Table Page 4-10. Relation of Administrative Turnover Rate of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . . . . . . . . . 102 4-11. Relation of Degree of Professional Specialization of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . . . . . 106 4-12. Relation of Degree of Professional Specialization of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . . . . . 107 4-13. Relation of Degree of Rule Formalization of Bureau and Educational Innovation (Ever Submitted Title III PrOposal) . . . . . . . . . . . . 109 4-14. Relation of Degree of Rule Formalization of Bureau and Educational Innovation (Number of Title III PrOposals Submitted) . . . . . . . . . . . . 109 4-15. Relation of Recruitment Pattern of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . . . . . . . . . . . 111 4-16. Relation of Recruitment Pattern of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . . . . . . . . . 112 4-17. Relation of Size Of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . . . 113 4-18. Relation of Size of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . . 114 4-19. Relation of Administrative Turnover Rate of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . . . . . . . . 127 5-1. Career Cluster Patterns of the Opportunity Structure of School Bureaucracy and Associated Levels of Educational Innovation . . . . . . . . . . . 142 5-2. Relation of Career Cluster Patterns of Bureau and Educational Innovation (Ever Submitted Title III PrOposal) . . . . . . . . . . . . . . 144 vii Table 5-3. 5-4. 5-5. 5-7. 5-8. 5-9. 5-10. 5-11. 5-12. 5-13. Relation of Career Cluster Patterns of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . . Relation of Dimensions of the Career Cluster of Bureau and Educational Innovation (Number of Title III PrOposals Submitted) . . Polar Resource Cluster Patterns of the Opportunity Structure of School Bureaucracy and Educational Innovation . . . . . . . . . . . . Relation of Sample of Resource Cluster Combinations of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . Relation of Polar Resource Cluster Patterns of Bureau and Educational Innovation Relation of Modified Polar Resource Cluster Patterns of Bureau and Educational Innovation Values of Stepwise Multiple Regression of Resource Cluster and Innovation . . . . Relation of Size, Professional Specialization and Financial Support Dimensions of Resource Cluster of Bureau and Educational Innovation (Ever Submitted Title III PrOposal) . . . Relation of Size, Professional Specialization and Financial Support Dimensions of Resource Cluster of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . Relation of Size and Professional Specialization Dimensions of Resource Cluster of Bureau and Educational Innovation (Ever Submitted Title III Proposal) . . . . . Relation of Size and Professional Specialization Dimensions of Resource Cluster of Bureau and Educational Innovation (Number of Title III Proposals Submitted) . . . . . . Internal Dimensions of Bureau's Opportunity Structure and Climber/Conserver Dominance of Bureau . viii Page 146 147 150 152 154 156 157 159 160 162 163 187 LIST OF FIGURES Figure Page 1. Administrative Turnover Rate and Educational Innovation . . . . . . . . . . . . . . . . 103 ix CHAPTER I AN INTRODUCTION AND GENERAL DISCUSSION OF BUREAUCRACY AND OPPORTUNITY STRUCTURE Public bureaucracies play an ever increasing role in the functioning of modern society. Whether this is good or bad is a question that is receiving growing attention by politicians, some political scientists and others interested in the future of the com- munities in which they live. Seeking intelligent, reasoned answers to the questions surrounding the role bureaucracy should play requires an increase in our understanding and knowledge of bureaus and their policies. As an effort in the expansion of our understanding of bureau behavior, this study focuses on a particular type of bureaucracy--public school administrations-~and their associated policy. From this examination, we hOpefully will add to our knowledge of at least one type of bureau and suggest a model of bureaucratic decision-makers useful in future research into bureaucratic behavior. There no longer is any question that education is political by its very nature and worthy of study by political scientists nor are there reservations as to the bureaucratic nature of school administrations.1 However, this does not mean that our task will be an easy one as a perusal of the literature of bureaucracy and education reveals a general paucity of suggestive theory to guide this research. Were one to pick up and examine any recent issue of the American Political Science Review, one would find articles devoted to nearly all facets of political behavior except those dealing with bureaucracy--regardless of its type. Relatively few political scientists have concerned themselves with the nature of bureaucracy and its impact upon modern society and thus made an effort to develOp theory dealing with bureaucratic behavior.2 Of those who have, nearly all begin from a sociological rather than a political perspective. The result has been a comparatively small amount of research under- taken into the political behavior of bureaus and their impact upon public policy. Those political scientists who view bureaucracy as political often involve themselves with the budgetary process of a bureau or take a more descriptive, non-theoretical approach.3 Few significant attempts have been undertaken to develop a positive theory of bureaucracy and those that do exist have been developed principally by economists.4 Examination of the literature dealing with education and educational administration has not been significantly more productive than that mentioned above. The majority of this literature is of the case study type or of a very limited comparative nature. Here, at least, most researchers concern themselves with bureaucrats and the bureaucracy of school districts. There is an awareness of the importance of this institution in the study of education and the educational policy process.5 The theoretical foundations of the majority of these studies are inadequately developed or are sometimes even missing. They do, however, prove suggestive of some of the types of factors that should be considered in examining school bureaucracies. If the literature of bureaucracy and education is as generally theoretically inadequate as it is held to be, where do we turn to find theory to guide this undertaking? The most directly suggestive and potentially productive approach to the study of bureaucracy to date is that taken by Downs in Inside Bureaucracy. Here he attempts to develOp ". . . a useful theory of bureaucratic decision-making. The theory should enable analysts to predict at least some aSpects of bureau behavior accurately, and to incorporate bureaus into a more generalized theory of social decision-making--particu1arly one relevant to democracy."6 While Downs is correctly criticized for his failure to accomplish what he set out to do in terms of scientifically _ developing his theory of bureaucracy, Inside Bureaucracy remains one of the few significant attempts to develop a formal theory which explains and predicts bureaucratic behavior in its totality. It serves here as one of the two principle theoretical sources of this dissertation. The other theoretical stimulus is Schlesinger's work on political ambition, Ambition and Politics.7 While tested in an electoral context, the concepts of ambition and structure of political opportunity are felt to be significant to research into the behavior of bureaus. These concepts provide additional insights which permit the further refinement of the theoretical concepts of Downs selected for use here. The central assumption buttressing this research is the belief that bureaucrats--school administrators--like other rational self-interested political actors have different preference orderings (ambitions) and they make decisions which reflect these preferences even when acting in their official roles. Extremely important to these rational bureaucrats and their actions is the environment (structure of political opportunity) of the bureau in which they serve. The opportunity structure or environment of the bureau helps give direction to a bureaucrat's goals and aids him in obtaining or maximizing those goals. Thus, it is important to understand the nature of the opportunity structure bureaucrats face in order to begin to increase our knowledge of bureau behavior. The Opportunity structure of a bureau is made up of a number of different dimensions which are independently related to bureau policy. The dimensions of this opportunity structure include age of bureau, growth rate of bureau, administrative turnover rate, degree of professional specialization, recruitment pattern of bureau officials, degree of rule formalization, level of financial resources and size of bureau. These individual dimensions combine into two cluster of dimensions which give direction to a bureaucrat's goals and suggest the means, the strategies by which he will attempt to maximize those goals--a career cluster and a resource cluster. As with the individual dimensions, each dimension cluster is expected to be associated with bureau policy. While many potential educational policies could be related to the dimensions of the opportunity struc- ture of a bureau, only policies involving educational innovation are examined. The amount or level of educational innovation present in a bureau is related to the Opportunity structure. It is from the relationship of a bureau's Opportunity structure and its policy that the model Of bureau decision-makers will be drawn. Chapter II presents an overall review of the important litera- ture of educational administration and bureaucracy which aided in the develOpment of this research. The hypotheses relating the Opportunity structure and educational innovation are contained in Chapter III in addition to the theory and design of this study. Chapter IV contains the test of the hypotheses relating the individual dimensions of the opportunity structure to educational innovation. The tests of the dimension clusters and innovation hypotheses are presented in Chapter V. Chapter VI summarizes the results of testing the hypotheses relating educational innovation and the opportunity structure of a bureau and suggests a model Of bureaucratic decision-makers to be utilized in future research of bureau behavior. FOOTNOTES--CHAPTER I 1It was only thirteen years ago that Thomas Eliot was forced to argue that education was political and a suitable subject for political scientists to study. See Thomas H. Eliot, "Toward an Understanding of Public School Politics,” American Political Science Review, LIII (December, 1959), 1032-51. 2In a brief perusal of the American Political Science Review for the past several years, a total of four articles were fOundIto be related in some manner to the bureaucracy--three dealt with areas generally applicable to all organizations and the other involved the budgetary process of state bureaucracy. This seems to be reflective of the lack of attention or importance given to the political nature of public bureaucracy by the most prominent in the discipline. Even the editor of the American Political Science Review noted in the June, 1972, issue that there seemed to be an imbalance in the areas represented in the Review-~public administration and organization theory being one of those under-represented fields. 3Several examples of tne type of literature referred to include: Aaron Wildavsky, The Politics of the Budggtary Process (Boston: Little, Brown 6 Company, 1964); Donald Gerwin, Bud eta Public Funds (Madison: The University of Wisconsin Press, 1969}; JOhn D. MiIlett, Organization for the Public Service (Princeton: D. Van Nostrand Company, Inc., 1966); Samuel Krislov and Lloyd D. Musolf, ed., The Politics of Regulation (Boston: Houghton Mifflin Company, 1964); John J. Corson andIJoseph P. Harris, Public Adminis- tration in Modern Society (New York: McGraw-Hill Book'Company, Inc., 1963); and Francis E. Rourke, Bureaucratic Power in National Politics (Boston: Little, Brown E Company, 1965). 4The principal attempts are Anthony Downs, Inside Bureaucracy (Boston: Little, Brown 6 Company, 1967); Gordon Tullock, The PoIitics of Bureaucracy (Washington, D.C.: Public Affairs Press, 1965); and William A. Niskanen, Bureaucracy and Representative Government (Chicago: Aldine-Atherton, 1971). 5Examples of the type of literature referred to include Marilyn Gittell, Participants and Participation (New York: Fredericklk Praeger, 1967); Mafilyn Gittell and T. Edward Hollander, Six Urban School District§_(New York: Frederick A. Praeger, 1968); Laurence Iannaccone and Frank W. Lutz, Politics, Power and Policy_(Columbus, Ohio: Charles E. Merrill Publishing Company, 1970); Louis H. Masotti, Education and Politics in Suburbia (Cleveland: The Press of Western Reserve University, 1967); John B. Miner, The School Administration and Or anizational Character (Eugene, Oregon: The Center for the AdvancedIStGdy of Educational Administration, University of Oregon, 1967); David Rogers, 110 Livingston Street (New York: Random House, 1968); Alan Rosenthal, Peda o ues and Power (Syracuse: Syracuse University Press, 1969); Russell T. Greff and Stephen J. Knezevich, "The Superintendent: What Makes Him What He Is," The American School Board Journal, CLVIII (June, 1971), 12-17; Donald R. Miller, "An Analysisvof Management Team Characteristics, Educational Characteristics Affecting the Quality of Education, and Selected Cost Factors of Michigan School Districts" (unpublished Ed.D. dissertation, Michigan State University, 1970); and Stuart L. Openlander, "The Development of an Administrative Structure in a Middle Sized City School District" (unpublished Ed.D. dissertation, Michigan State University, 1968). 6Downs, Inside Bureaucracy, p. l. 7Joseph A. Schlesinger, Ambition and Politics (Chicago: Rand McNally 5 Company, 1966). CHAPTER II LITERATURE OF EDUCATION AND BUREAUCRACY In this chapter, we examine the more important literature of education and bureaucracy which has influenced the development of this research. From some of these investigations, the dimensions of the opportunity structure considered important were selected--age, growth rate, administrative turnover, professional specialization, recruit- ment, rule formalization, financial support and size. Other studies contributed to the theory guiding this undertaking. Still other research suggests relationships between dimensions of the opportunity structure and educational innovation. We begin by discussing the significant contributions of education and educational administration. Education and Educational Administration In reviewing this broad field, we first consider several comparative studies which point out potential dimensions of the opportunity structure. John Miner, in his monograph The School Administrator and Organizational Character, comparatively analyzes school districts and their administrators by conducting a concurrent validity study.1 Miner obtains "potential" predictor variables of individual administrators which he correlates with a number of measures of individual job performance criteria gathered at the same time. The "potential" predictor variables primarily involved measures of individual motives and personality characteristics along with some biographical and ability indexes. The job criterion variables include ratings made by supervisors, salary paid and the level achieved by the individual in the school hierarchy. The school districts were classified according to the size of the bureaucracy--large city (population over 300,000), medium city (pOpulation 60,000-80,000), small city (population 20,000-30,000) and consolidated district (large area including rural and suburban). Of particular interest is Miner's examination of the "organizational character" of the bureau- cracy of the variously sized school districts. Administrators in the large city ". . . are not particularly mobile individuals. Job changing is not frequent. Most spend a relatively long time in each position."2 Districts in this type of city are also noted for re- taining intelligent and stable administrators through the effective use of monetary rewards. Medium cities have administrators who receive a salary or pay scale that is on the low side. They are marked by their mobility in moving from district to district. Small cities have districts where hard work, problem solving, originality and ambition are valued. These are stable districts with little job changing taking place and that which does occur is within the district. Miner concludes from the data that this is a district in transition. The consolidated district has a younger group of administrators who are not particularly well paid. The administrators are occupationally mobile with much of the change taking place within the district itself. 10 In Six Urban School Districts: A Comparative Study of Institutional Response, Gittell and Hollander comparatively analyze the fiscal and administrative operations of six large city school systems--Baltimore, Chicago, Detroit, New York, Philadelphia and St. Louis.3 They posit a model to measure one output of a school district--educational innovation. Innovation is ". . . defined as the successful introduction to an applied situation of means or ends that are new to that situation."4 This measure of output was believed to reflect financial inputs (allocation of financial resources), effectiveness of administration (administrative organization of the school district), and the community and political pressure brought to bear upon the school district (extent of participation in policy making). They hypothesize that innovation is related to administra- tive organization in three ways: (1) bureaucratization of the district measured in terms of the relative importance of administra- tion to instruction, (2) internalization measured by the power dis- tribution between central headquarters administration and administra- tion in the schools, and (3) leadership measured by the power of the superintendent in administrating the school system. They also suggest that the level of innovation is associated with the extent of participation since school systems are felt to stay at rest unless external forces such as participation upset this equilibrium.5 Finally, innovation would be more likely if new financial resources were made available to implement change in a school district. As a result of testing these hypotheses, a large number of conclusions were drawn. Several of the more important are suggestive to this research. First, there is little difference in output among 11 the six districts. Of the three inputs examined, only one--public participation--seemed to have a cause and effect relationship with innovation. Four basic groups--school boards, teacher organizations, community participants and governmental officials--make up the participants investigated. Gittell and Hollander found the more open the school system was to the involvement of these groups in the policy process, the more innovative the system was. Other ". . conditions which were originally hypothesized as relevant to innova- tion, and accepted by many to be significant, proved tentatively invalid. Such factors as selection of superintendent from outside the system, fiscal statutes, and method of selection of school board did not vary to any appreciable extent with output."6 Size of bureaucracy, fiscal independence and level of bureaucratization proved to be unrelated to the adaptability/innovativeness of the system. In "The Superintendent: What Makes Him What He Is," Greg and Knezevich develop a portrait of American school superintendents.7 To capture this image, they conducted a mail survey of superintendents from different sized districts throughout the United States in 1970. A number of suggestive relationships were found and reported. The largest school districts have superintendents who receive the highest salaries and are the most mobile of all the superintendents sampled. Although the median age of superintendents is 48, there is a trend toward younger men filling the positions in large districts. The typical administrator serves an average of five years per administrative post regardless of the size of the district. The 12 movement of administrators that does occur is more likely to be from district to district within the same state as Opposed to interstate job changing. Taken as a whole, school superintendents are not a particularly mobile group. Bidwell devotes one section of "The School as a Formal Organization" to schools as bureaucracies.8 He points out that there is presently little empirical material available on the bureaucratic nature of schools. If one is to study schools as bureaus, Bidwell advises the investigator to ". . . attend to variations in the bureaucratization of structural arrangements." Further, he suggests that the functioning Of ". . . school systems can be understood only as an outcome of complex interactions between structural arrangements (degree of bureaucratization), the attitudes and orientations of staff (degree of professionalization) and board members (fiduciary or managerial), and recruitment to organizational roles."9 He does examine these empirical studies available and finds certain common variables to be important--size of school district, recruitment patterns of superintendents and mobility of administrators. Size of district is associated with the willingness of the superintendent to delegate reSponsibility and to support the professional staff, consensus of the school board, and recruitment patterns of administra- tors. Recruitment patterns of superintendents correlates with career patterns and attitude of superintendents, conditions under which administrators work, and their performance in office. Mobility and attitude toward job mobility interact and are reflected in administrative behavior. 13 In a monograph examining the growth, the functioning and the consequences of bureaucratic rules within public schools, Anderson presents suggestive findings about the effects of bureaucracy in junior high schools.10 Drawing his hypotheses from organization theory, Anderson tests them by gathering data from a mailed survey of 152 teachers in 28 junior high schools in a large metropolitan school district. The rules of the schools and the extent to which they are utilized are highly dependent upon the nature of the professional services performed, the size of the school and the relationship established with the public served. Contrary to his expectations, Anderson found that rules appear to mediate authority conflict thus making hierarchical authority more tolerable to the professionals-- the teachers of the school district. He also finds that school size may affect instructional practices in a detrimental fashion. Finally, as student enrollment increases, professional specialization also increases in a school. The final comparative study examined is Rosenthal's Pedagogues 11 In this undertaking Rosenthal deals with the role of and Power. teacher organizations in the making of educational policy. He studies a number of the more impOrtant political characteristics of public teacher groups in five large cities: New York, Boston, Chicago, Atlanta, and San Francisco. Rosenthal concentrates on those factors he considers important in the emergence of teacher organizations (affiliated with either the American Federation of Teachers, the National Education Association or independent teacher groups) as potential members of the educational community during the years 1961- 1965. The theory undergirding his argument is traditional group 14 theory as expounded by Truman. Four major categories of interdependent variables were considered important in light of the concepts he develops and explores. These were: organization--strength based upon a group's internal dimensions, opportunities—-determined by the practices and di5position of school government and diSposition of school teachers, behavior--primarily a product of strength and Opportunity, and influence--related to all of the above factors. Rosenthal sees two sets of findings emerging from his research. One, the influence of teacher organizations varies dependent upon the issue and city studied. Second, a patterned difference by issue, domain and city exists. Overall, teacher organizations had not yet achieved major influence in school decision-making. Their influence must be considered in terms of organizational strength, Opportunities and behavior. He predicts that teacher organizations will grow in ". . . numbers of communities throughout urban and suburban America to participate more decisively than ever before. If the past and present, as explored in this study, serve as an indicator, the years ahead should produce an even greater impact by teacher groups in local governments of public education."12 Turning to the particular for further elaboration of dimensions of the opportunity structure, we look first at Gittell's Participants and Participation: A Study of School Policy in New York, a case study of educational participation and decision-making, which seeks to explore the political forces and the individuals who affect educational policy.13 The author first evaluates the openness of the New York City School system by describing how school policy is made. At the 15 same time, the author reassesses Sayre and Kaufman's assertion that there is not a single elite dominating the political governmental units of New York City. Gittell hopes that ". . . the type of analysis attempted here has broader implications for the conceptual rational- ization of the pluralistic view of power."14 The major conclusion drawn is that rather than being the Open school system Sayre and Kaufman suggest, the New York City system is "narrow, convergent and dominated by a consensual elite." There are only three of four areas in which "any" appreciable influence outside the system can be brought to bear on educational policy. Only those participants inside the school system have a major role to play in setting school policy. These participants are the board of education, the superintendent, the bureaucracy and the teacher union. Overall, these findings differ greatly from Gittell and Hollander's conclusion in Six Urban School Districts. A more encompassing case study than the Gittell research is Politics, Power and Policy.15 In this investigation, Iannaccone and Lutz seek to gain an understanding of the causes, processes and difficulties of local school politics which they see as necessary if one wishes to influence future development in the governing of local districts. To accomplish this task, the authors explore in some depth crucial periods in the develOpment of a school district in which major policy changes take place. They begin by examining the belief system Americans hold about their schools and the environment in which the local school district operates. The school district in the American political milieu is seen as a fairly closed system because of 16 special governmental arrangements and the absence, in most systems, of party politics in school politics. There is also a strong tendency on the part of school districts toward consensus politics which results in a monotholitic power structure for the school district. Iannaconne and Lutz next examine a single school district in some detail to test their concept of school politics. In the process of doing this, they present a model of change for a school district. Using data from an additional school district study and their own data, they further test their model and find that school districts do have the capability to change themselves and their schools. This change occurs when an incumbent school board member is defeated. The new member and the superintendent find themselves in conflict and the superintendent is normally replaced within a three-year period. His replacement is usually drawn from outside the district with a mandate to change the school system. In the latter portion of their work, the authors turn to what happens after this change has taken place in the school district. Another case study involving change is Masotti's Education and Politics in Suburbia: The New Trier Experience. This study of Chicago's New Trier Township High School DiStrict is aimed at showing ". . . how public education becomes increasingly politized in the suburbs as these areas . . . respond to the challenges of dynamic change."16 Masotti examines the school district beginning in 1961, with the defeat of a school building referendum, through 1963 when this referendum was resubmitted and overwhelmingly approved by the -oters. He considers a number of components which he feels make up 17 the political system of the New Trier district. Included among these components is the bureaucracy of the school district. He suggests the significance of bureaucracy to change is dependent upon the size of that bureaucracy. Masotti, however, fails to provide support for this belief in the evidence of his study. 110 Livingston Street is the final case study to be examined.17 Rogers is concerned with the political and social forces that affect the New York City school system. In particular, he focuses upon innovation and reform in the school system. To do this, the author looks at the way school desegregation was handled in New York City. The conceptual scheme chosen by Rogers analyzes the formal bureaucratic organization of the district in terms of: the tasks it performs, the structure and administration of this bureaucracy, the codes and traditions of the bureau, the internal politics of the bureau and the bureau's outside clientele. The author begins by considering the significance of "ecological and situational" factors to the failure of public education in the city. The major portion of Rogers' study involves the administration and politics of educational decision-making and the sources of pressure and constraint that face the city's board of education. These sources are the outside interest groups, the school system itself and the other governmental agencies (city, state and national) the district must deal with. The period studied (1963- 1968) provides Rogers with extensive material which he incorporates in a very lengthy work--little of which will be discussed here except those portions dealing directly with the professional bureaucracy. In considering the results of his research, one should keep in mind 18 that Rogers takes ". . . the position that analysis of complexity is not enough, and the social scientist has a moral obligation to devote much of his energies to suggestive remedies for the many social problems in the cities of mid-twentieth century America."18 Three particular areas of difficulty pointed out by ROgers are of interest. He finds that the overcentralization present in the bureaucracy and the proliferation of specialized administrative units have been major contributors to the "pathologies" present in the New York City schools.19 This centralization has had its impact upon school services, staffing and local innovation as well as a number of other factors. Recruitment and promotion practices in the bureaucracy restrict major change unless they can be altered. The bureaucracy is extremely in-bred resulting in little innovation. The third factor hampering the system is the compulsive rule following syndrome associated with civil service mentality--the tendency to invoke particular policy as an attempt at personal and organizational sur- vival. This syndrome is found to be most strongly correlated with the more tradition-bound bureaucrats. While we have certainly not exhausted the research undertaken in this area, we have discussed those studies most useful in our research of the opportunity structure. Whether it is good or bad for the educational policy process, bureaucracy is clearly a most important part of that process. Many of the dimensions of the opportunity structure of school districts were considered at least potentially useful explanatory variables. Size of the school district bureaucracy receives considerable attention in both comparative and case studies. 19 Size is found to be associated with the mobility of administrators, the incentive structure of the bureau, the recruitment pattern of adminis- trators, the bureaucratization of the rules and the professional Specialization of the bureaucracy. Other dimensions found important include recruitment pattern of bureaucrats and the mobility of administrators (administrative turnover rate). Variations in recruit- ment pattern is related to the ultimate career patterns of those individuals and their administrative performance. Performance of school administrators is also tied to the mobility of administrators. Two dimensions of the opportunity structure were found to be unrelated to the level of innovation present in a district after they were originally hypothesized as being important. Financial resources of a district (level of financial support) and size of bureau did not appear to be associated with innovation in the Gittell and Hollander research. Educational innovation was found to be related to the hierarchical arrangement of a bureau. Bidwell suggests that it is important to examine the interaction of certain dimensions of school systems--professionalization, recruitment, bureaucratization (size and rule formalization)--to understand the functioning of school districts. The exact use given to these dimensions of the Opportunity structure and their relationship to educational innovation is developed in Chapter III. Organization Theory and Bureaucracy While organization theory and that specifically relating to bureaucracy is not overly helpful theoretically, certain empirical 20 studies do contribute to the overall development of this research. In a short article on bureaucracy, Hall and Title find that organizations can be ranked according to their degree of bureaucratization.20 ‘They measure bureaucratization by combining an organization's score on six dimensions of bureaucracy (hierarchy of authority, division of labor, complexity of rules, Specificity of procedures, impersonality of Operation, and technical competency as the criterion for hiring and promotion) into a single Guttman scale. This measure is found to be somewhat associated with the Size of a bureau (increased size, in— creased bureaucratization),tjuahierarchy of a bureau and several other less important variables. Hall and Tittle conclude by stressing the importance of studying goals as well as structural elements of a bureaucracy. In "A Multidimensional Approach Toward a Typology of Bureau- cracy," Samuel and Mannheim study a number of dimensions and sub- dimensions of an organization and develop a typology of bureaucracies based upon these dimensions.21 They begin by looking at two basic dimensions of a bureaucracy--contextual and structural. The con- textual dimension is made up of a number of subdimensions including age of an organization (length of time in existence) and Size of the enterprise (number of participants in the organization). The struc- ture of a bureau is made up of the subdimensions of hierarchy of authority, functionalization (functional differentiation, diversifica- tion and Specialization), interlevel personality and formalization of organizational rules. A number of associations between contextual and structural dimensions were found to exist. Size of organization was 21 found to be negatively related to intensity of control exercised in the chain of supervision. Size is positively related to formalization of rules--the larger the bureau, the more formal the rules and pro- cedures. Age is positively associated to impersonality of the bureau. From relationships such as these and others, the authors develop a typology of six basic types of bureaucracy which they feel indicates ". . . the feasibility of establishing empirically and inductively certain consistent patterns, contextual as well as structural, which are associated with bureaucratic organizations."22 Meyer attempts to link the formal structure of a bureau to the decision-making process, in particular, to the centralization or decentralization of authority in a bureau in "The Two Authority Structures of Bureaucratic Organization."23 In this research, Meyer 1 finds that Size of bureaucracy is related to functional differentia- tion and hierarchical differentiation--the larger the bureau, the greater the functional and hierarchical differentiation. At the same time functional differentiation and hierarchical differentiation are alternative ways of dividing the work of a bureaucracy. Holding size constant, the greater the number of subunits (functional differentia- tion), the fewer the number of levels of supervision (hierarchical differentiation) and vice versa. The author also notes that with increased hierarchical differentiation, regulations are more likely to be formalized by being written. In an article dealing with career patterns in business organizations, Dill, Hilton and Reitman attempt to answer two questions: "Why do some men move ahead more rapidly in management 22 during their first five or ten years in industry than others with essentially the same education and experience? To what extent do the men who move ahead rapidly act as agents of their own prOgress?"24 To answer these questions, the authors conducted in-depth interviews with graduates of a master's program in industrial administration. The individuals selected were not representative of all college graduates as they were distinguished by their high level of intellectual ability and their specialized training in engineering and management. While the researcher acknowledged that they could not precisely test hypotheses about factors in career progress with those being studied, they could present their preliminary findings and what they felt to be interesting strategies for further research. Dill, Hilton and Reitman's central finding was that career progress ". . . depends greatly on the strategies or heuristics that an individual employs in "his" game with the organizational environment that he encounters at work."25 The fact that career progress depended upon strategies is seen as promising by the authors because: (1) the individual's inter- action with his environment is stressed over other possible relation- ships, (2) the strategies are expected to be more directly related to his performance than the biOgraphical and personality variables that influence the development of the strategy, and (3) the focus on strategy has been fruitful in past research. Dill gt_al: go on to point out that men who advance rapidly are individuals whose goal is to advance. Their behavior is governed by this desire to achieve their goal of advancement. Those individuals who do not progress as rapidly, if at all, when compared 23 to their counterparts profess less desire for specific benefits such as power, autonomy, income or status; less tolerance towards the costs of achieving these benefits by being a manager; and a greater commit- ment to and an identification with their present jobs and working conditions. In analyzing the managers who have made rapid prOgress, they find that these individuals are eSpecially sensitive in their strategies to Specific features of the environment in which they work. They also maintain a wide range of independent contacts to evaluate their present position and develOp alternative Opportunities. These individuals are seen as "active agents of their own progress." With this brief review of some of the appropriate organization literature, we will turn to one of the principal theoretical sources of this research--Downs' Inside Bureaucracy. Downs bases his theory on the fundamental premise that bureaucratic officials are signi- ficantly, although not solely, motivated by their own self-interest. He posits three central "hypotheses or axioms" from which he purports to deduce a set of propositions that can be empirically tested in the future; i.e.,not in the book itself. These axioms briefly stated are: (l) bureaucrats seek to attain their goals rationally—~they are utility maximizers, (2) bureaucratic officials have complex goal sets, and (3) bureau social functions influence a bureau's internal struc- ture and behavior and vice versa. Unlike economic or game theory's "perfectly informed world," Downs sees the environment the bureaucrat Operates within as one in which information is costly, the decision- maker has only limited capabilities when making decisions and he operates under conditions of uncertainty.26 24 Of particular interest is the set of "ideal" bureaucrats developed from the first and second axioms. Downs identifies five different "types" of bureaucrats based upon their different sets of goals and the fact that they each attempt to maximize their own utility, i.e., goals. These five types can be grouped according to who receives the benefit from the goals they seek. ”Purely self- interested" officials are those bureaucrats who are motivated by goals which benefit themselves directly rather than their bureau or society as a whole. The second category of official are those with "mixed motives" who have goals that combine self-interest with altruism. Purely self-interested officials can be divided into two types-- climbers who consider goals of income, power and prestige as nearly all important and conservers who see convenience and security as being all important. The mixed motive officials are differentiated by the breath of the altruistic goals they seek. Zealots are officials who are loyal to fairly narrow policies or programs. Having a somewhat greater breath to their loyalty than zealots are advocates whose loyalty is to a broad gauge prOgram or the bureau itself. Those who measure their loyalty in terms of society as a whole are known as statesmen. Three principle factors determine which type of bureaucrat a particular member of a bureau will be. One is the psychological predispositions inherent in the individual which are viewed as a given. These predispositions are subject to change, however, by the other two factors-—the nature of the bureaucratic position and the probability of attaining the goals associated with a particular type. 25 Having constructed this typology of bureaucrats, Downs points out that "one of the purposes of this typology of officials is to enable analysts to forecast behavior by identifying the particular types of officials occupying key positions."27 To aid in doing this, two classes of propositions associated with each type of bureaucrat are presented-~one describing the psychOIOgical proclivities of the type and the other suggesting the likely behavior of a type under varying conditions. Downs further offers ways of empirically testing each type of proposition. To test the first, the psychological attributes of actual bureaucrats would have to be measured and compared to those established by Downs for his different types of bureaucrats. However, this would not allow prediction of the type of behavior associated with the differing bureaucrats. To accomplish this, Downs suggests testing the second type of preposition which permits predict- ing behavior without specifically identifying individual types in a particular bureau. If, as he prOposeS, bureaus having certain func- tions or internal structural dimensions are dominated by a Specific type of official (i.e., slow growing or stagnant bureaus will tend to be dominated by conservers), then prepositions about the type- dominated bureau can be drawn which will hopefully predict the actions of that bureau. Unfortunately, Downs steps short of this and fails to suggest directly what type of policy would be expected to emerge from a particular type—dominated bureau. He does, however, point out a number of factors or dimensions that imply the type of official expected to dominate a particular bureaucracy and the type of actions these officials would likely take. One of the more important factors 26 suggested is the age Of the bureau itself. Older bureaus should have a greater proportion of conservers than would a younger bureau and hence be less likely to undertake innovative action. The growth taking place in an agency should indicate the more dominant official. Downs also argues that the amount of turnover in the bureau, the level of specialization present, the degree Of rule formalization and the pattern of bureau recruitment will all be associated with the type of dominant official. The other principal theoretical stimulus of this research is Schlesinger's Ambition and Politics. Here it is argued that "Ambition lies at the heart of politics."28 and is a most important concept in political analysis. Ambition has a most elementary relationship with political action which must not be hidden by "subtle" political explanations. While acknowledging that there has been some concern as to why men have ambition, Schlesinger points out that "What is needed . . . is a theory of politics which explicitly accepts the assumption that politicians respond primarily to their office goals, in effect an ambition theory of politics, rather than a theory which explains personal ambitions."29 The central assumption of ambition theory rests on the belief that a politician's action or behavior are in response to his own Office goals--the politician is involved in political activity in seeking office and making decisions designed to gain that office. Thus, the politician is faced with two problems: (1) defining his office goal/goals and (2) relating his activity to them. Ambition theory is then concerned with alternative strategies of advancement for the politician. 27 Identifying political ambitions concretely is a difficult assignment because ambition is first of all an abstraction which cannot explain a politician's behavior in its totality. Office holding by a politician is not always a totally encompassing assign- ment which means other motivational forces may be present and operating at the same time as ambition. A politician's career is also partial because of its potential intermittent character. Thus, it is impossible to assign specific ambitions except when a politician is openly seeking office. To develop a theory of ambition, it is neces- sary to adopt certain assumptions about when ambitions occur. Ambition develOps in a Specific situation--as a reSponse to the possibilities confronting the politician. Office goals are not assumed to be con- stant in their intensity nor in their direction for the same individual or for different individuals. Thus, ". . . the ambitions of any politician flow from the expectations which are reasonable for a man in his position."30 The major difference existing between office-seeking ambitions is in terms of their direction. Ambitions vary in three directions-- discrete, static and prOgressive. A discrete ambition is one which guides a politician to seek a particular office for a particular term and then voluntarily choose to withdraw from public Office. For the politician who has a static ambition, a career commitment is made to holding a particular Office over the long run. Finally, the politician with a progressive ambition is one who aSpires to an office more important than the one he now holds or is seeking. The t direction a politician's ambition takes is not assumed to be fixed 28 over a lifetime. Ambition theory is primarily concerned ". . . with taking a man's current ambitions and predicting from them this political behavior."31 To bring order to the ambitions of politicians, it is necessary to find order in their chances or opportunities for office-- to discover the structure of political opportunities. To order ambitions, a certain stability of opportunities must exist since "The direction Of the ambitions fostered by an office depends upon the way in which office-holders and the electorate typically treat the office, as well as upon the route it provides for political advance- ment."32 The structure of opportunities does not include all the opportunities that face a politician nor does it include all the opportunities possible in the social system. What the exact Struc- ture of political opportunities looks like in a particular Situation is a matter of empirical observation which is the major focus of the rest of Schlesinger's book. He does very little to develop the strategies of political advancement. In the only study directly relating ambition to bureaucracy, Swinerton applies ambition theory to upper echelon state bureaucrats in "Ambition and American State Executives."33 He attempts to identify among 46 appointed state executives in six comparative states: ". . . 1) the direction of ambition; 2) the relationship of ambition to background factors; and 3) the effect of ambition on a given set of executive behavior patterns generally considered to be prerequisites of effective coordination with the governor and the . . . . 4 . governor's party in the conduct of state administration."3 Hav1ng 29 ascertained the direction of ambition of the executive under study, Swinerton found that age, party activity, years of service and salary incentives were related to ambition. He then undertakes the explana- tion of strategies that these executives would select to maximize the potential of moving in the direction of their ambition given their own particular situations. Quite different perceived patterns of behavior were obtained from executives with "prOgressive” and "discrete” ambitions than from those with "static" ambition. This difference was felt to be due to the importance of security because it is a crucial factor to those with static ambition and not to those state executives with progressive or discrete ambitions. The studies selected from organization theory also support the importance of the dimensions of the opportunity structure selected. Size is found to be associated with bureaucratization of an organiza- tion, intensity of control, and functional and hierarchical differen- tiation. Bureaucratization was found earlier to be related to educa- tional innovation and thus size may be correlated to innovation through its relationship to bureaucratization. The impersonality of a bureau is correlated with the age of the bureau. Downs suggests that the rate of growth of a bureau is an important variable in ascertaining which type of official will dominate a bureau. Other dimensions he posits as being useful in this process include administrative turnover, professional specialization, rule formalization and recruitment pat- tern of the bureau. Other studies and works examined contribute to the development of the theory presented in Chapter III and the decision model discussed in Chapter VI. 30 Summary of Literature In this chapter, we find that the suggested dimensions of a bureau's environment or Opportunity structure have been utilized by researchers in many different ways in exploring the nature of bureau- cracy. The investigators report differing conclusions as to the relationships between these dimensions and other variables of a bureau- cracy and bureau behavior. As an aid in clarifying these diverse results, a summary for each dimension of the opportunity structure is made from the various investigations reported. The first dimension of the opportunity structure of a bureau-- level of financial support--is only considered by Gittell and Hollander. They originally hypothesize that innovation would be more likely if new financial resources (increased resources) were made available to implement change in a school district. Gittel and Hollander did not find this to be the case. Innovation was unrelated to fiscal resources. Samuel and Mannheim and Downs feel that age of the bureau is a useful dimension of a bureau. The former view age of a bureau as one of the components making up the contextual dimension of a bureaucracy. They found age of bureau to be positively related to the impersonality present in a bureaucracy. Downs sees age of bureau related to the type of official who will dominate an agency and, ultimately, the type of policy to emerge from the bureau. Older bureaus will have more con- servers and will be less innovative than younger bureaus. The only one to consider growth rate of a bureau is Downs. As in age of bureau, he suggests that the growth rate of a bureau is correlated to the domi- nant type of official in a bureau and the policy that is associated with the bureau. 31 The administrative turnover rate of the bureau is related to the size of a bureau by both Greg and Knezevich and Miner but their conclusions differ. Greg and Knevevich find that large school dis- tricts have the greatest mobility in their administrators while Miner suggests that large districts have very little mobility. Small dis- tricts have mobility similar to that of the large districts. The medium sized districts have the greatest mobility of administrators. Bidwell reports research results that suggest that the mobility or administrative turnover is related to the administrative behavior of school bureaucracies. The type of official dominating a bureau is felt by Downs to be associated with the administrative turnover rate. This, in turn, should be related to bureau policy. Anderson finds that professional specialization influences the rules of a school and the extent to which they are used. He also points out that professional specialization in a school increases as student enrollment increases. Professional specialization is felt by ROgers to be a negative influence on performance of the New York City school system. It contributes to the "pathologies" he finds in this system. Finally, professional Specialization is seen by Downs as being associated with the type of official who dominates a bureau. The degree of rule formalization was found to be correlated with a number of different variables. Both Anderson and Samuel and Mannhiem find size of bureau to be related to rule formalization. The larger the school or bureau, the greater the rule formalization present. Anderson also finds that rules may make hierarchical authority more tolerable to teachers. ROgers feels that rules and 32 compulsive rule following are related to the "pathologies" he finds in New York City. Rule formalization is one component of Hall and Title's concept of bureaucratization which they use to rank organizations. Hierarchical differentiation is found by Meyer to be associated with rule formalization--increased hierarchical differentiation, increased rule formalization. Downs also believes degree of rule formalization is important as it is related to the type of dominant official in a bureau. The source of officials or the recruitment practice of a bureau was suggested by Gittell and Hollander as being related to innovation. Where superintendents were recruited from outside the bureau, more innovation would take place. No relationship was found in the six districts studied. Bidwell reported that the recruitment practices of school bureaus was associated with size of district, the administra- tive behavior of a district and the career patterns of those selected. A recruitment pattern from within the district resulting in in-breeding was felt by Rogers to be another major contributor to the inadequate performance of the New York City schools. Where bureaus recruit offi- cials is seen by Downs as another factor in determining which type of official will dominate a bureau and the type of policy associated with that bureau. The last dimension of the opportunity structure--size Of bureau--is also the most studied variable. Many different types of relationships are found. Miner and Greg and Knezevich find differences in mobility associated with size but they disagree as to what size school district is the most mobile. Miner finds medium size districts have the most mobility while Greg and Knezevich find that large 33 districts have the greatest mobility. Gittell and Hollander did not find size of bureau related to innovation as they had originally hypothesized. The behavior of school superintendents and the recruit- ment pattern of superintendents is related to the size of the district in Bidwell's article. Masotti suggests that the size of a district influences how reSponsive a district will be to change. Hall and Title find the bureaucratization of an organization increases with the size of the organization. Size of bureau is another component of Samuel and Mannhiem's contextual dimension of bureaucracy. They found that the size of a bureau was negatively related to the intensity of control in the bureau. Size is positively correlated to the formalization of rules/procedures in a bureaucracy. Meyer finds that functional and hierarchical differentiation of a bureau displays a positive associa- tion with the size of the bureau. Finally, Anderson found that size of a school can have a detrimental influence on instructional practices in a school. The range of utilization of these various dimensions of a bureau's opportunity structure is quite large. Gittell and Hollander are the only researchers who consider a financial dimension while nine different investigators study size of bureau. The roles these eight dimensions of the Opportunity structure play in this research are develOped in Chapter III. FOOTNOTES-~CHAPTER 11 1Miner, o . cit. A concurrent validity is one which is charac- terized by prediction to an outside criterion and by checking the measuring instrument (now or in the future) against some outcome. See Fred N. Kerlinger, Foundations of Behavioral Research (New York: Holt, Rinehart and Winston, Inc., 1967), pp. 447-48. 2Miner, p. 70. 3Gittell and Hollander, op. cit. 4Ibid., p. 3. 5While it is possible for internal forces to upset the equilibrium of the system, Gittell and Hollander feel that most pressures will be exogenous to the system and originate in the community in which the system functions. 6Gittell and Hollander, p. 52. 7Greg and Knezevich, op. cit. 8Charles E. Bidwell, "The School as a Formal Organization," The Handbook of Or anizations, ed. by James G. March (Chicago: Rand McNally 6 Company, 1965), pp. 972-1022. 91bid., p. 996. 10James G. Anderson, Bureaucracy in Education (Baltimore: The John Hopkins Press, 1968). 11Rosenthal, op. cit. 121bid., p. 185. 34 35 13Gittell, op. cit. 14Ibid., p. 3. 15Iannaconne and Lutz, Op. cit. 16Masotti, op. cit., p. i. 17 . Rogers, op. cit. lslbid., p. 14. 19Overcentralization refers here to the extreme amount of decision-making that takes place at a central headquarters several administrative layers removed from the individual school and the problem area. 20Richard H. Hall and Charles R. Tittle, "A Note on Bureaucracy and Its Correlates," The American Journal of Sociology) LXXII (November, 1966), 267-72. 21Yitzhak Samuel and Bilha F. Mannheim, "A Multidimensional Approach Toward a TypOIOgy of Bureaucracy," Administrative Science ggarterly, XV (June, 1970), 216-28. 22Ibid., p. 227. 23Marshall W. Meyer, "The Two Authority Structures of Bureau- cratic Organization," Administrative Science Quarterly, XIII (September, 1968), 211—28. 24William R. Dill, Thomas L. Hilton, and Walter R. Reitman, "How Aspiring Managers Promote Their Own Careers," California Management Review, II (Summer, 1960), 9-15. 251bid., p. 10. 26Downs, pp. 2-4. 271bid., pp. 90-91. 28Schlesinger, Ambition and Politics, p. 1. 291bid., p. 4. 36 30Ibid., p. 9. 31Ibid., p. 10. 321b1d., p. 11. 33Nelson Swinerton, "Ambition and American State Executives," Midwest Journal of Political Science, XII (November, 1968), 538-49. OthEr studies utilizing ambition theory include Gordon S. Black, "A Theory of Political Ambition: Career Choices and the Role of Structural Incentives," American Political Science Review, LXVI (March, 1972), 144-59; Gordon S. BlaCk, “A Theory of Professionaliza- tion in Politics," American Political Science Review, LXIV (September, 1970), 865-78; Kennéih Prewitt, The Recruitment of Political Leaders: A Study of Citizen Politicians (New York: BOEbs-Merrill Co., 1970); Kenneth Prewitt and William Newlin, "Political Ambitions and the Behavior of Incumbent Politicians," The Western Political Quarterly) XXII (June, 1969), 298-308; Kenneth Prewitt, "Political Ambitions, Volunteerism, and Electoral Accountability," American Political Science Review, LXIV (March, 1970), 5—17; Richard L. Engstrom, "Polit- icaI Ambitions and the Prosecutorial Office," The Journal of Politics, XXXIII (February, 1971), 190-94; John W. Soule, "Future Political Ambitions and the Behavior of Incumbent State Legislators," Midwest Journal of Political Science, XIII (August, 1969), 439-54; Jeff FiShel,i“Ambition and the Political Vocation: Congressional Challengers in American Politics," The Journal of Politics, XXXIII (February, 1971), 25-56; Michael L. Mezey, "AmbitionTheory and the Office of Congressman," The Journal of Politics, XXXII (August, 1970), 563-79; Paul L. Hain, "The American State’Legislators' Ambitions and Careers: The Effects of Age and District Characteristics" (unpublished Ph.D. dissertation, Michigan State University, 1971). 34Swinerton, pp. 539-40. CHAPTER III THE RELATIONSHIP BETWEEN THE OPPORTUNITY STRUCTURE OF A BUREAU AND BUREAU POLICY Theoretical Approach to Bureaucracy The manner in which this research is being approached is dif- ferent from most other studies of bureaucracy. Niskanen succinctly summarizes our approach when he argues that My method is that of economics, although the subject [bureau- cracy] has been the almost exclusive domain of sociology. In this book I develop a theory of "supply" by bureaus, based on a model of purposive behavior by the manager of a single bureau. . . . The "compositive" method of economics, which develops hypotheses about social behavior from models of pur- posive behavior by individuals, contrasts with the "collect- ivist" method of sociology, which develops hypotheses about social behavior from models of role behavior by aggregative ideal types. The individual consumer, entrepreneur, or, in this case, bureaucrat is the central figure of the charac- teristic method of economics. He is assumed to face a set of possible actions, to have personal preferences among the outcomes of the possible actions, and to choose the action within the possible set that he most prefers. He is a "chooser" and a "maximizer" and, . . ., not just a "role player" in some larger social drama. The larger environment influences the behavior of the individual by constraining his set of possible actions, by changing the relations between actions and outcomes, and, to some extent, by in- fluencing his personal preferences. The economist develops models based on purposive behavior by individuals, not to explain the behavior of individuals (that is necessarily the task of psychology), but to generate hypotheses con- cerning the aggregative consequences of the interaction among individuals.1 37 38 As a first step in this process, the concepts bureau and bureau- crat must be defined. A bureau is a large, non-market organization whose members are full-time employees hired, retained and promoted upon the basis of expected or actual performance in their organiza- tional roles. Defining a bureau in this way suggests that much of organization theory is inapplicable because it is grounded in the market place where organization output can be evaluated by means of quid pro quo transactions. This also leads to a different view of the individual employed in a bureaucracy. A bureaucrat could be defined as someone who works in a bureaucracy but this is far too simplistic. Rather, a bureaucrat is an individual who is employed full-time and receives the majority of his income from a bureaucracy where personnel policies are based upon internal employee role performance and whose own job output cannot be measured in market terms. This definition suggests that a bureaucrat is not subject to the same fiduciary relationship that other political actors have with their constituency because the bureaucrat is not held directly accountable through periodic elections by outside constitutencies nor is his continuance in office tied directly to the individual office of someone else who is elected.2 In a practical sense, this excludes from the definition of bureaucrat those top members of a bureau who owe their position to some elected official or group of officials.3 Nevertheless, school administrators, i.e., school superintendents, are considered bureaucrats even though they serve at the pleasure of a political sovereign. The norms associated with school administration dictate that administrators will not normally be replaced just because new members are elected to the 39 body making up the political sovereign of the bureau--the school board.4 Even though the fiduciary relationship is not the same for the bureaucrat as the elected official, there is a relationship between the bureaucrat and the bureau he serves because of the ultimate accountability of the bureau to the public through the political sovereign. Thus, the individual bureaucrat is held, at least in- directly, accountable for his actions in his bureaucratic role. Having defined the concepts of bureau and bureaucrat, we now turn to the individual who is a bureaucrat and the assumptions being made about him. These assumptions differ somewhat from those of Downs because they are drawn more strictly from utility theory.5 The school administrator or any bureaucrat is seen to: (1) be rational. A bureaucrat maximizes his expected utility. He does so even when acting in his official capacity. (2) have a complex set of goals/preferences. A bureaucrat has a set of goals/preferences for which a utility schedule could be constructed. (3) have imperfect information. Regardless of how much data are available, the bureaucrat has imperfect information as information is intrinsically costly because it takes time, effort and normally money to obtain data and understand them. Bureaucrats have only limited personal capability, even when aided by modern technology, in absorbing and utilizing information. These factors insure that a bureaucrat makes decisions in "ignorance"--in an environment where he lacks all the necessary information to make perfectly informed decisions. 40 (4) make decisions under risk. A bureaucrat knows that each action he takes leads to one of a set of possible outcomes, each outcome occurring with a known probability. The complete set of goals/preferences in assumption two are not those derived from the performance of his social function but rather they are goals derived from his personal motives for carrying out that function. Some of these goals may be identical but the two sets never coincide identically.6 Although a bureaucrat's goals could be made up of an almost unlimited number of items, they are restricted here to just nine different types--power, money income, prestige, convenience, security, personal loyalty, pride in performance of work, desire to serve the public interest and commitment to a specific prOgram of action.7 It makes no difference to the develOpment of this research whether or not these preferences are intermediate means or ultimate ends to the bureaucrat. Five of these goals are seen as "pure" manifestations of self- interest. They are power, money income, prestige, convenience and security. Pure self-interest is defined as the benefit from the goal which accrues directly and only to the individual bureaucrat. Desire to serve the public interest is often thought of being altruistic and not a self-interested goal but it makes no difference when examined as a part of a bureaucrat's preference ordering. If the bureaucrat seeks this particular goal then it is in his own self-interest to act to serve the public interest as he himself perceives it. The other three goals (personal loyalty, pride in performance of work and commitment to a Specific program) can be thought of as containing elements of 41 both pure self-interest and altruism (benefits from the goal which accrue directly to others in addition to the benefits the individual receives). A brief elaboration of these goals will aid in understanding the nature of goals that bureaucrats seek. Money income, prestige and pride in performance are the common sense ideas associated with these terms. Power can be defined simply as the ability of an indi- vidual to enforce compliance with his wishes. Power as a goal can include power within the bureau and/or outside it. Convenience is resistance to any change in official behavior which requires increased personal effort (increased costs incurred in modifying behavior) and a willingness to accept any change which would reduce his personal effort expended in Official behavior. Security is defined as a low probability of suffering future losses in power, prestige, income and convenience. Personal loyalty is a bureaucrat's allegiance to some unit--his work-group, the bureau itself, the government or to the nation-State. Commitment to a Specific program of action is that strong personal attachment some Officials have toward a specific policy which can motivate their actions, i.e., Vice Admiral Rickover and the atomic submarine program. Desire to serve the public interest is not some sort of moral commitment to an absolute value called the "public interest" but it is rather what each individual bureaucrat believes the bureau ought to do to carry out its social function. The importance of these goals is the manner or method by which a bureaucrat selects or orders his preferences and the strategies he adapts to maximize those preferences. Downs argues that the key to 42, identifying his ideal types is to understand the preference ordering of the goals the ideal types seek to maximize. This, in turn, Should enable the researcher to forecast bureau behavior by the identifica- tion of the particular bureaucrats occupying key dominant positions in the bureau. One way of doing this would be by examining the psycho- logical attributes of each ideal type and classifying dominant bureau officials according to these psychological dimensions. Having identi- fied the type of bureaucrat, it should then be possible to predict the bureaucrat's behavior. Alternatively, Downs suggest that predictions of bureau behavior can be better developed by examining the dimensions of a bureau because bureaus having certain characteristics will tend to be dominated by a specific type of official who will act in a particular manner.8 The basis of this alternative approach is the belief that it is only possible for an official to maximize his goals in a bureau having certain characteristics--either internal dimensions or social functions. An official who wishes to maximize his money income would not be expected to be dominant in a bureau characterized by low salaries throughout the bureau or a bureau whose social function was expected to be performed at a very low cost to the public receiving the benefit of that social function. It is this second approach which Downs feels will prove more useful in empirically testing his theory. This alternative approach by Downs is not unlike the argument Schlesinger makes in ambition theory. In this case, Schlesinger is dealing with differing ambitions of politicians rather than goals of bureaucrats but he, like Downs, places great importance on the environment facing the political actor. The central assumption made 43 by Schlesinger is that a politician's behavior is in response to his office goalS--his ambition. Or as Schlesinger argues ". . . the politician as office seeker engages in political acts and makes decisions apprOpriate to gaining office. His problem consists, first, in defining his office goal or goals and, secondly, in relating his current activity to them."9 To ascertain which office goals a poli- tician selects, it is first necessary to ". . . find order in their choices or opportunities for office. It is therefore necessary to discover, if possible, the structure of political opportunity."10 The structure of political opportunity is also important if one is to discover the strategies of advancement a politician selects to attain his office goals. Knowledge of the structure of political opportunity and hence the ambition and strategy of the politician should permit the prediction of his political behavior since this behavior is in response to his ambition.11 Dill, Hilton and Reitman also suggest that professional advancement is related to the goals of the individual and his strategies to obtain those goals. Their research is another example of purposive behavior based upon the personal preferences/goals of the individuals. The individual maximizes his goal of advancement by selecting strategies that relate his goal to the organizational environment in which he works. Thus, to understand his behavior in advancing himself in the organization, it is necessary to examine the larger environment facing the individua1--his structure of organ— izational opportunity. 44 Considered together, Downs, Schlesinger and Dill g£_gl, provide sufficient justification to explore and study the environment the bureaucratic actor confronts. They all argue that the goal/goals an individual selects and the strategies he adopts to obtain those goals are greatly influenced by the environment he faces. Thus, if we are to explain and predict the behavior of the bureaucratic actor, it is first necessary to increase our knowledge of a bureau's environ- ment. This is exactly what we are attempting in this research. We explore the environment--the Opportunity structure of the bureaucrat-- and attempt to define this Structure. The exact nature of a bureau's Opportunity structure is an empirical question just as it was for Schlesinger when he identified the environment political actors face-- the structure of political opportunity.12 Downs leads us to believe that the Opportunity structure reflects differences due to variations in a bureau's social function as well as the bureau's structural dimensions.13 Since we are dealing with bureaus having the same social function-~public education in grades kindergarten through twelve--our concern is directed exclusively toward the Structural dimensions of a school district's Opportunity structure. Identifying that a particular dimension of a bureau is part of the opportunity structure for the bureaucrat presents some special difficulties in this research. Unlike Schlesinger's investigation of politicians which found that "The order which emerges from these data on career patterns becomes 'the structure of political opportunities'.", we do not have large reservoirs of public data on the careers of bureaucrats.14 Thus, it is not possible to look for variations in 45 career patterns to identify those important characteristics/dimensions which are a part of a bureaucrat's opportunity structure. However, we can follow Downs' suggestion and look at variations in public policy, in this case educational innovation, to ascertain whether a particular dimension of the bureau is potentially important to the calculations of the bureaucrat by determining if this dimension is related to variations in the amount of innovation taking place in bureaus. Each of the dimensions of the opportunity structure suggested is considered potentially important in a theoretical sense and associated with educational innovation. Dimensions of the Opportunity Structure The environment of a school bureaucracy has many characteristics which can be used to describe it. However, the actual opportunity structure of the bureau which helps give direction to a bureaucrat's goals and aids him in obtaining or maximizing those goals is made up of only a few dimensions. This Opportunity structure can refer to either the internal, the external or both contexts of a bureau. The external environment of the school bureaucracy is important to the administrator because this is where the bureau's social function actually takes place. Education is the very reason for the bureau's existence. Additionally, the bureau's social function influences the bureau's internal structure and vice versa. The most important element in the external environment of the bureau is the political sovereign who is the individual official or body of officials that has legal authority over the bureaucracy. School bureaucracies often have more than one political sovereign-~the local school board and the state 46 board of education both have areas of legal authority over a school b 15 ureaucracy. The authority of the local school board is directly administered while that of the state is indirect through the local _ board. Other groups in the external environment that the administrator must deal with include clientele, allies and/or enemies. Clientele are those individuals who receive direct benefit from the performance of the social function of the bureau--i.e., the school children and their parents. Allies/enemies are individuals or organizations who either support the school bureaucracy in a conflict situation or are in Opposition to the bureau in that conflict. The position of ally may change to that of enemy and enemy to that of ally on different issues.16 Examples of potential allies/enemies in most school environments include teacher associations or unions, ethnic parent groups, civil rights groups, taxpayer groups, anti-busing groups and parent-teacher associations. Their importance to any school bureaucracy has been Shown in several of the studies previously cited-~Gittell and Hollander, Rosenthal, Masotti and Rogers.17 A possible indicator of the external environment that potentially taps the overall level of support or lack thereof for the school district is the level of financial support given to that district. This dimension may reflect how well the political sovereign and the school bureaucracy are getting along with the other external actors in the bureau's environment. The internal environment or structure of a bureaucracy has been extensively studied and numerous dimensions have been suggested as being useful. Only those few found suggestive in Downs and other 47 research discussed in Chapter II will be examined in this research. These internal dimensions are the age of the bureau, the growth pattern of the bureau, the administrative turnover rate of the bureau, the degree of professional specialization, the degree of rule formalization, the recruitment pattern and the size of the bureau. We begin this discussion of Specific dimensions of the oppor- tunity structure by considering the only dimension of the bureau's external environment selected--level of financial support. This dimension has received relatively little attention in relationship to bureau behavior. One measure of financial support--additional resources available to implement change-~was found by Gittell and Hollander to be unrelated to the degree of educational innovation present in large urban diStricts. Most financial Studies of public bureaucracy are examinations of revenues, expenditures and the budgetary process.18 In Downs' development of factors influencing the performance of officials, he never considers the effect financial factors might have upon their behavior. He does discuss in general terms what he calls "waste" or as it is better known "organizational slack" but he does not relate variations in slack to individual behavior.19 As envisaged here, the level of financial support of a bureau is indicative of the amount of resources available to the bureau to perform its social functions. The first internal dimension of a bureau considered-~age of a bureau--is an important factor in a bureau's opportunity structure for several reasons. Samuel and Mannheim find a relationship between age and impersonality of the bureau.20 Downs devotes a great deal of 48 attention to a bureau's age and predicts a number of relationships between age and the type of bureaucrat expected in a bureau. The basis of his predictions rests on the potentiality of a bureaucrat obtaining his goals within the agency he serves. One example of the type of relationship suggested is developed in his discussion of the proportion of conservers to be found in bureaus where he concludes that "The older a bureau is, the higher the likely prOportion of conservers therein."21 Another factor in the use of age is seen in Downs' discussion of the life cycle of bureaus where he identifies a number of what he describes as "predictable ways" bureaus change as they grow older. Examples of this behavior include the improved performance of the bureau's social function as bureaus gain experience with age; the more formalized rule system that tends to develop as bureaus grow older; and the increase, both proportionately and numerically, of the administrative levels of a bureau with age. The growth pattern of bureaus has not been extensively studied by those in educational administration. Growth has been of some concern to those interested in organization theory but not in terms of individual behavior.22 Downs is the principal supporter of the use of this dimension. AS with age, the growth in the size of the bureau has a significant influence upon an official's potential for obtaining his goals. Downs suggests that: (l) the more rapid the growth rate of a bureau, the greater the proportion of climbers present and the more likely they will dominate the bureau. With this increased growth comes more innovation Since climbers are more inno- vation prone than the conservers they replace. (2) Growth normally 49 provides bureaucratic leaders (dominant officials) with increased power, income and prestige. (3) Rapid growth weakens the influence of conservatism on'a bureau as it grows older.23 The administrative turnover rate reflects the mobility of the bureaucrats within an agency. As previously shown by Bidwell, mobility of administrators is reflected in their administrative behavior.24 Turnover in a bureau is also reflective of the promotion Opportunity in that bureau. Bureaus where turnover is taking place at least have the opportunity for promotion to take place depending upon which official/officials leave. Downs sees the turnover rate serving as an influence upon the degree of conservatism present in a bureau and indirectly upon the type of bureaucrat that would dominate the agency. Internal turnover in a bureau is felt to counteract the increased conservatism that accompanies increased age in the bureau. As bureaus grow older a number of effects occur which can be mitigated by increased turnover--the greater the turnover, the smaller the proportion of conservers and the greater the prOpor- tion of climbers.25 The extent to which a bureaucracy has divided the tasks of performing the bureau's social function into units of specialized activity is reflected in the degree of professional Specialization in that bureau. A bureau which is highly Specialized has divided the work load into many more discrete units than one which is not specialized. A highly Specialized bureau is also more decentralized than a non-specialized one in terms of control. A number of studies have shown that professional specialization is related to behavior in 50 a bureau. Rogers argues that specialization is very much correlated to the "pathologies"--the bureaucratic behavior--of the New York City schools. Specialization has a negative impact on a number of factors in the behavior of that bureau.26 Meyer finds professional special- ization associated with the hierarchy of a bureau-~the greater the 27 The extent of specialization, the fewer the levels of supervision. an individual's Specialization is seen by Downs as a factor in his ability to realistically evaluate the importance of his own specialty to the bureau. The more Specialized his activity, the less he knows of the other parts of the bureau and its outside environment and the more he exaggerates the importance of his own specialty. Thus, specialization is an important factor in a bureaucrat's decision--it has an impact upon his information.28 An argument different from Downs can be made that the more professionally specialized the bureau, the greater the total information that is readily available to the bureau about the broad range of activities making up the bureau's social function and the lower the cost of gathering this information. However viewed, the degree of professional Specialization is an important consideration in determining the availability and cost of information in a bureau. The flexibility or rigidity of the rules of a bureau (degree of rule formalization) is the Sixth dimension of the opportunity structure. If the rules are written and extensive in nature, then they are more rigid than if the rules are brief and unwritten. Rules have been found to be important by Rogers because of their detrimental influence upon the behavior of bureaucrats.29 Downs also finds formal 51 rules play an important function in the behavior of bureaucrats. Rules have an impact upon the amount of discretion available to the bureau- crat--the greater the degree of formalization, the less the amount of discretion available to the bureaucratic decision maker and the less he is able to further his own personal goals.30 Recruitment patterns of bureaus have been found to be associated with bureaucratic performance in office in a number of investigations. Bidwell finds that the recruitment pattern of school administrators is correlated to attitudes of superintendents and their performance in office.31 Rogers also finds that recruitment has an influence on the behavior of bureaucrats.32 Downs argues that recruit- ment patterns of a bureau are important since the effectiveness of a bureau is dependent upon the nature and capabilities of its members. Further, the level of recruitment has an impact upon the performance of the bureau's social function.33 A bureau's recruitment pattern is important because it reflects the source of new officials and the promotion patterns of a bureau--Are all promotions from within the bureau and hence, all recruitment takes place at the lowest levels? Does the bureau promote both from within and from without for top level appointments? Are all top level promotions filled from outside the bureau and thus there is no upward mobility for lower level officials? The last individual dimension of the opportunity structure-- size of bureau has received the greatest attention. Size is found by Miner to be related to the mobility of school administrators, salary paid to administrators and type of administrators found in school 52 districts. Greg and Knezevich also find that size of school district is associated with mobility of administrators and the salary level of the district. Other educational studies find recruitment patterns, iconsensus levels and willingness to delegate vary with the size of the bureau.34 The organization theory literature also finds Size to be a very suggestive variable--Hall and Tittle, Samuel and Mannheim, and Meyers all find size to be associated with important variations in bureaus. The extent of‘bureaucratization, the intensity of control, the degree of rule formalization and the degree of hierarchical and functional differentiation vary with the size of the bureau. Downs suggests that size is important in a number of different contexts. He argues that size of a bureau is related to the total "wasted motion" taking place within the bureau. He also posits three prin- ciples of organizational control based upon size: (1) No one individual can fully control a large organization. (2) The larger any organization becomes, the weaker is the control over its actions by those at the top. (3) The larger the organization becomes, the poorer is the coordination among its actions.35 These individual dimensions of the opportunity structure which were suggested as being important variables have all received some consideration by other researchers. Certain dimensions have received greater theoretical support for their use than have others. A few dimensions draw their principal support from empirical evidence. The lexact relationship each single dimension has to educational innovation is detailed in the next section. Before turning to those relation- ships, we must first examine the two dimension clusters of the Opportunity structure. 53 The first dimension cluster is the career cluster made up of the individual dimensions of age of bureau, growth rate of bureau and turnover rate of bureau. This cluster of dimensions is identified as the "career cluster" because of the importance Downs gives to the individual dimensions that compose it. These dimensions-—age, growth rate and turnover rate--are seen by Downs as important in ascertaining which type of bureaucrat will be found in a particular bureau and which will most likely dominate that bureau. AS previously shown, more conservers are expected in older bureaus; more climbers should be found in bureaus that are growing rapidly; and more climbers are predicted in bureaus with high personnel turnover.36 These three individual dimensions are alSo theoretically related by Downs and can be considered tOgether in determining which bureaucrat should be found in a particular bureau--i.e., younger, rapidly growing, high turnover bureaus will have more climbers than older, stagnant, low turnover bureaus. This, in turn, suggests which type of goals the bureaucrat has selected to maximize. These goals will be reflected in what type of career a bureaucrat has--his bureaucratic behavior in achieving those goals. Thus, when considered together, age, growth rate and turnover rate are seen as an important determinant in an individual's career as a bureaucrat. Support for the second dimension cluster-~the resource cluster made up of size, professional specialization, recruitment pattern, financial support and rule formalization--is drawn principally from Bidwell and Downs. Neither suggests that all these dimensions be considered together but they do agree that certain of these dimensions 54 could or Should be examined together. Bidwell states that if the researcher is to understand the functioning of school systems, he must study the complex interactions of professional specialization of staff, recruitment patterns of staff, degree of bureaucratization and school board attitudes and orientations. Degree of bureaucratization has been shown in other research to be positively associated with the Size of the bureau and the degree of rule formalization. Thus, one can substitute size and rule formalization for bureaucratization and remain consistent with Bidwell's exhortation. Believing that Bidwell is correct in his advice is then one reason for considering size, rule formalization, recruitment pattern and professional specialization together. Downs' support for examining these individual dimensions together (except for financial support) is entwined in his overall development of Inside Bureaucracy. At various points in the book, he points out the relationship between the dimensions and the type of official expected and his potential for maximizing the goals he seeks. The way Downs treats professional specialization, recruitment pattern, rule formalization and size is as if these dimensions are resources in a bureaucrat's calculations of how to maximize his preferences or goals. If they are indeed indicators of resources available to the bureaucrat, then they should be examined in relation to educational innovation. Neither Bidwell nor Downs includes level of financial support in their recommendations. Nevertheless, we can not forget to include financial support with these other dimensions of the resource cluster 55 since financial support is "the" essential component of the operation Of any bureau. Without some basic level of support or allocation of resources by the political sovereign, a bureau would be unable to carry out its social function. This, in turn, would severely affect a bureaucrat's chances for achieving the goals he wants. The level of financial support a bureau receives must be viewed as a part of the resources available in a bureau. The Opportunity Structure and Educational Innovation The hypotheses in this section suggest the predicted relation- ships between the dimensions of the opportunity structure (individually and together) and educational innovation. While a directional relation- ship between the dimension and innovation is stated, this direction is given with the awareness that the relationship is not always based upon firm theoretical grounds. At times, the direction of the relation- ship is inferential having been drawn from studies previously cited. In several of the hypotheses, plausible arguments could be made to support a relationship in the opposite direction to that given. Only the actual testing of these hypotheses will resolve what is funda- mentally an empirical question at this time. Educational innovation is defined very much as Gittell and Hollander do in their research of urban school districts. Innovation is the introduction to an applied situation of means and/or ends that are new to that situation.37 This definition of innovation means that we are concerned with the attempt at introducing new means and ends rather than any measure of the success of the introduction of these new means and ends. Success of innovation is extremely difficult 56 to measure in any social function and especially so in education. Education is a social function noted for its inability to adequately measure achievement or success of educational endeavors. This is not unexpected Since Downs argues that it is almost impossible to measure anything on the output side of a bureaucratic organization. We also subscribe to this position and have incorporated it into our definition of bureaucracy as a large, non-market organization whose output cannot be evaluated by means of quid pro quo transactions. Another reason for not using success of introduction of new means and ends is the large number of intervening external groups who have significant influence on the level of success in a school district; groups over which the bureaucracy has little or no control--the national government, state government, interest groups, etc. The first dimension is the level of financial support of the bureau which is indicative of the amount of resources available to the bureau in the performance of its social function. This overall level of support will be reflected in the resources a bureau can allocate to innovative action. Where the level of financial support is low, a bureau has fewer resources available to perform its basic social function. These bureaus will be less able to divert a portion of their scarce resources to innovative undertakings since innovation is not a cost-free process. Bureaus must expend resources of time, material and personnel if they are to attempt to be educationally innovative. This argument is very similar to Downs' view of organizational slack and the belief that non-programmed actions (innovation, long range planning, basic research) will be cut back or eliminated when slack is 57 reduced for whatever reason. The more organizational slack is re- duced, the more these nonprOgrammed actions will suffer to the point where they will be completely eliminated by the bureau. We suggest that the greater the level of financial support of a bureau, the greater the potential for innovative action being undertaken. l. The higher the level of financial support a district receives, the greater the educational innovation taking place in that district. Ho: Innovation(High Support)=lnnovation(Low Support) H1: Innovation(High Support)>Innovation(Low Support) The age of a bureau is expected to be reflected in the amount of innovation taking place in that bureau. Downs discusses the effect of age on a bureau in his development of the life cycle of a bureau. In this dynamic model of bureau change over time, Downs suggests that there will be differences in bureau behavior due to the age of the bureau. Bureaus grow more rigid in their behavior as they grow older. The rules become more formalized and detail more com- pletely what sort of actions can be taken. The age distribution of those serving in a bureau also changes when the bureau grows older--the prOportion of older bureaucrats increases unless there is rapid growth in size of the bureau, a high rate of personnel turnover, or both. The increase in the age of the bureaucrats tends to strengthen the influence of conservers in the bureau because other types of Officials are likely to become conservers as they grow older with the bureau. Since conservers are interested more in maintaining the status quo than changing it, less innovation is expected to be initiated by older 58 bureaus dominated by conservers. The effect of age upon an indi- vidual's behavior has also been suggested by Schlesinger and others when they argue that age is extremely relevant to the type of career 38 the individual has in politics. As Schlesinger points out: The age cycle also restricts a man's political chances. A man's reasonable expectations in one period of his life are unreasonable at another time. A man can fail to advance in politics as much because he is at the wrong age at the wrong time as because he is in the wrong office.39 The behavior of the political actors Should reflect the differences in their ages. Taking all of these factors into consideration suggests that age will be a depressant upon the innovativeness of school districts--costs of innovativeness will increase with the greater rigidity in the system and the decrease in discretion available to the decision-maker arising from more formalized rules. We thus hypothesize that 2. The older the school district, the less the educational innovation taking place in that district. Ho: Innovation(Older)=Innovation(Younger) H1: Innovation(Older)Innovation(Slower Growing) The administrative turnover rate of a bureau reflects the mobility of those serving within that agency. Bidwell argues that mobility of administrators is associated with administrative behavior. He points out research which shows that administrators who are mobile are more innovative than those who are not mobile. At the same time, he also reports findings which suggests mobile oriented superintendents are not all that innovative when compared with other less mobile superintendents.42 Downs clearly believes in the importance of mobility of bureaucrats. He feels that it is definitely related to their ability to achieve desired goals. "Jumping"--leaving a position in one bureau for a new and better position in another bureau--is one of the principal ways in which a climber can increase his power, income or prestige. An individual bureaucrat's ability to jump is a function of the type of skills he possesses and their transferability to other bureaus.43 Where there is considerable turnover of personnel in a bureau, Downs feels there is an increased likelihood Of innovation 60 taking place because of the new ideas, new information, new perSpectives and need for success that these individuals bring with them. Thus, a direct relationship between the administrative turnover rate and innovation is expected. 4. The higher the rate of administrative turnover in a district, the greater the educational innovation taking place in that district. Ho: Innovation(Higher Turnover)=Innovation(Slower Turnover) H1: Innovation(Higher Turnover)>Innovation(Slower Turnover) The extent to which a bureau divides the tasks of performing its social function into units of specialized activity is the degree of professional specialization of a bureau. This division of labor along lines of professional training and expertise has been shown to be related to variations in bureau behavior. While most find professionalization to be functional in education, Rogers finds it correlated to the patholOgies he finds in the New York City schools. Degree of professional specialization is an important determinate of how much information is available to a bureau and the cost of that information. Where there is greater specialization, there is greater information about the broad range of activities that make up the bureau's social function. This increased level of information intro- duces new ideas and techniques to areas of difficulty at lower cost to the bureau. In turn, the costs of undertaking innovative action are less. Thus, we hypothesize that: 61 S. The greater the degree of professional specialization in a district, the greater the educational innovation taking place in that district. Ho: Innovation(Highly Specialized)=lnnovation(Less Specialized) H1: Innovation(Highly Specialized)>Innovation(Less Specialized) The degree of rule formalization can be a serious inhibitor of educational innovation. Rules do not necessarily have a detrimental effect on a bureau as both Anderson and Downs point out. Anderson finds that bureaucratic rules make hierarchical authority more tolerable to teachers which, in turn, influences their performance favorably. Downs suggests a number of positive contributions rules make to the functioning Of a bureau.44 On the other side of the argument is Rogers' view that extensive rules and rule dependence is used by bureaucrats in their efforts at personal and organizational survival which results in many of the difficulties facing the New York City system. The extent to which rules exist and are used is important in ascertaining the freedom of action that exists within a bureau. Bureaus which have a high degree of rule formalization (Specificity and complexity), have less discretion available to officials when they make decisions. Their freedom of choice is limited by these extensive rules. They have less latitude of action in making decisions and less opportunity in introducing innovative solutiOns to problems of the bureau. An inverse relationship is expected between the degree of rule formalization and innovation. 62 6. The greater the degree of rule formalization in a district, the less the educational innovation taking place in that district. Ho: Innovation(More Formalized)=Innovation(Less Formalized) H1: Innovation(More Formalized)Innovation(Recruitment Inside) Size of bureaucracy is the last and probably the most important single dimension of a bureau's Opportunity structure. Size of bureau- cracy or the number of members of the bureau is related to many important variations in bureaus--both structural and behavioral. Un- fortunately, the only attempt at relating size directly with educa- tional innovation was unsuccessful. Gittell and Hollander did not find any variation in the amount of innovation when they examined the differences in size of the six large urban school districts. The weight that should be given to their finding is Open to question--the sample of schools is numerically small and environmentally the same, in that they were all big city urban districts. The major criticism that can be levied against their use of size is that there is no real variation in this independent variable. All of the schools studied are extremely large when contrasted to the possible range of school district sizes in the United States. For example, the range in size of Michigan school districts is from only two employees to over 19,000 total employees. Because of this lack of real variation in size, we reject the Gittell and Hollander finding. We continue to feel that size of bureau is an important determinant of the amount of innovation that will be found in a bureau. Size of bureau is related to innovation in two principal areas--need for innovation and the capability of being innovative. 64 In the first case, need for innovation arises out of the relationship of bureau size and the number of students being served in the per- formance of the bureau's social functiOn.4S While there is a certain economy of scale that applies in handling larger number of students, education remains a labor intensive endeavor in which the magnitude of the problems reflect the size of the student enrollment. Education researchers clearly are aware of this relationship. This is apparent from these investigators intensive interest and research into the problems of the large school districts. The fact that the problem areas are larger also suggests that there is a greater need for solutions because of the larger number of students being affected. If larger school districts are confronted with more serious problems, at least in terms of the numbers of students involved, they also are blessed with larger bureaucracies to develop innovative responses. As mentioned many times already, innovation is costly to accomplish because it requires the diversion of scarce resources from the regular performance of the bureau's social function. One of these scarce resources is personnel to discover and implement innovative action. Only larger bureaus have sufficient personnel to divert to innovation and at the same time carry out the day to day performance of education. Thus, we hypothesize that the larger the size of the bureau, the greater the amount of innovation taking place in the bureau. 8. The larger the Size of the bureaucracy of a district, the greater the educational innovation taking place in that district. Ho: Innovation(Larger Bureau)=Innovation(Smaller Bureau) H1: Innovation(Larger Bureau)>Innovation(Smaller Bureau) 65 The relationship of the career cluster (composed of the dimensions of age, growth rate and turnover rate) to educational innovation is shown in Table 3-1. As can be seen in this table, certain combinations of age, growth and turnover are expected to be related to higher levels of innovation than are other combinations. These interactions are consistent with the hypotheses develOped for each separate dimension and Downs' discussion of their interaction in the life cycle of a bureau. Age of bureau should be manifested in the level of innovation present unless modified by changes in growth, turnover or both. How much weight one should give to each separate dimension of the cluster in suggesting the final anticipated relation- ship is not clear--will the presence of a high bureau growth rate overcome the effects of age and low or no administrative turnover? We have decided that unless two of the three conditions suggested for greater amounts of innovation are present, a lower innovation level will be hypothesized for that particular combination of age, growth and administrative turnover. 9. School districts having career cluster dimension patterns 1, 2, 3(if exists) or S will be more innovative than districts having career cluster dimension patterns 4(if exists), 6, 7 or 8. Ho: Innovation(Pattern 1, 2, 3[if exists] or S)=Innovation (Pattern 4[if exists]. 6, 7 or 8) H1: Innovation(Pattern l, 2, 3[if exists] or S)>Innovation (Pattern 4[if exists], 6, 7 or 8) Unlike the career cluster, the exact relationship of the resource cluster to educational innovation is much less apparent. While we expect variations in innovation in bureaus with different 66 .u66wpsm many mo cowmmsomflv OuOHQEoo a now ofluod .mm .mcaoa mom .nuzoam oflumuosmOHSD mo mufiewcxv one we mfimxfimcm was :a poonuoo ma wagon m“ uaooo ou xfioxfia «on ma pa .Ham um ma .xfizoam mcfizoum m“ umnu xumuosmonsn so: u o>mn ou manfimmom xHHmuHmoH m“ aw oHfl:3e hemzogu “nozouu fleozosv fieoemuzv fiepzoau fleeemsz Aeoemumv heoemuzv Ho>og :ofium>oc:H HmuHm zm>mm wa m>Hm maom mmmzh 02% mzo a HOHmSHU 303 case 304 swam 303 swam 303 seem mm>oszH mm>oszH mm>ozmab mm>ozmsb madcaflooa mcacflaooo .1 . . i . . \ucmcmmum\ommuo>< nmfim \ucm:MMOm\ommno>< cm“: mhoccu Hecowumosem mo mHo>oa nopawuomm< ecu xoeuoamonsm Hoonom mo ouauoahum xuficsphomno any we mcuouumm Houmsau uooumu--.~um m4mInnovation(Pattern Two) Data and Measurement The hypotheses developed are tested by examining the public school districts of the State of Michigan. There are over 600 K-lZ districts in this state that could have been included in this research. The student enrollment ranges from under 100 to over 25,000 pupils in the Detroit Public School system. The size of the bureaucracies of these districts is reflective of the variations in student enrollment. Some of the small districts employ as few as five or six personnel which includes the superintendent, principal and teachers. Detroit, on the other hand, has a total employment of over 19,000 (professional and non-professional) employees. Because a bureau is a 123323 non- market organization, there must be some minimum Size administration to qualify as a bureau. The smallest school administration that meets the definition of bureaucracy is one which employs at least 30 staff members.46 This eliminates approximately 100 of the smallest districts 69 from consideration leaving slightly over $00 for study. The measures of the dimensions of the Opportunity structure and educational innovation are gathered from these 508 school districts. The principal sources of the aggregate data of the dimensions and innova- tion are the Michigan Department of Education and the Michigan Educa- tion Association. Each of these organizations publishes a number of documents containing characteristics of the school districts needed to test the hypotheses. The quality of this information varies because the original sources of the data are the districts themselves. Some districts have at times failed to provide the requesting agency the information being sought. The universe of 508 districts is reduced in several instances to 437 bureaus by this randomly dis- tributed missing data. The time frame of this data, depending upon the variable, extends from before the Second World War until the 1970-1971 school year. Operationalization of the dimensions of the opportunity structure and educational innovation did present certain difficulties. Empirical research into the characteristics of bureaus and their behavior is not particularly well developed. This inadequate develop- ~ment does not provide a great deal of expertise to draw upon. While Downs suggests a number of the dimensions of the Opportunity struc- ture, he does not posit ways of measuring these variables. Schlesinger's theory is developed in the electoral arena and the measures of the structure Of political Opportunity are not directly applicable to this research. For these reasons and because of the exploratory nature of our research, we have attempted to keep the 70 measures of the dimensions of the opportunity structure and innovation as simple as possible. The first independent variable Operationalized is the age of the bureau.47 Age of the school bureaucracy is the length of time that the school bureau has administered. the school district it serves--the year of the district's creation or the year of its most recent major reorganization to the end of the 1970-1971 school year. This permits differentiating school districts which have existed for a relatively long period of time from those which have been recently formed or consolidated. Annexation of one district by another is not the creation of a new bureaucracy. Rather, it is a measure of growth for the annexing district. The division between older and younger bureaus is based upon the latest revision of the law governing Michigan K-12 public education. This revision took 48 All districts fOrmed or place in the 1958-1959 school year. consolidated prior to this change in the basic law governing school district organization are classified as being older districts--a total of 454. The 54 districts formed or consolidated since then are the younger districts. The reason for selecting this dividing point is that it marks a major change in district organization requirements. This revision required the K-8 district to be phased out as a separate entity with only K-12 districts to remain. This change in state law produced a significant alteration to the overall organization of education in the state and, in particular, to the districts affected by the required consolidation with other districts or to the new districts formed under the new standards. We believed 71 these younger districts would be distinct from other districts created earlier. Growth rate of bureau can refer to change in either the number of clientele served or the number of bureaucrats employed by the bureau. The change in bureau employment is potentially the better measure of growth in a bureaucracy. However, there are difficulties in using this indicator of growth when considering the possibility Of stagnant or negative growth rates. There is a natural reluctance on the part of administrators to let employees go which suggests that the total number of bureaucrats tends to remain fairly constant and does not reflect stagnant or negative growth in employment until much later if at all.49 Because of the unreSponsiveness of bureau employment to negative change, we have turned to the other measure of bureau growth for educational bureaucracies--change in student enrollment (clientele served) over time which is the more reSponsive to fluctuations since school districts are basically unable to in- fluence the number of children enrolled in the district they serve. The growth rate of school districts in Michigan is measured by ascertaining the average yearly growth rate for each district for the period 1966-1967 to 1970-1971.50 While this is only a five-year period, it does begin to capture declining enrollments in some school districts especially in the urban areas. Up until the mid-nineteen Sixties, enrollments were generally increasing in all districts but the last half of the decade found an out migration from the central cities to the suburban districts. This shorter period better captures the changing patterns of growth that is taking place than a period 72 encompassing the entire decade. Bureau growth ranged from a positive 38.4 per cent per year to a negative 4.8 per cent per year. The overall state-wide average for growth in student enrollment is +1.8 per cent per year. Administrative turnover rate of bureaus is reflective of the mobility of the employees of the bureau. This mobility could be based upon the turnover of all of the bureau's employees or upon only those employees who have most to do with the administration of the district. Since we are concerned with innovation, we examine the mobility of only those bureaucrats who are potentially involved in making deci- sions about committing the district to innovative programs. In the actual determination of administrative turnover, we were limited to those administrative positions reported by school districts to the Michigan Reading Circle, publishers of Michigan Education Directory and Buyer's Guide. This is the only source of data on administrators on a statewide basis which is available over time.51 The actual turnover rate for a bureau is the turnover per administrative position per year for the period from school year 1958-1959 to school year 1969-1970--which was the last year published when the assembling data. For example, the Lansing Public Schools had 14 changes in personnel during the time frame for the 8 positions examined--an average of 1.8 changes per position or .15 changes per position per year. The administrative turnover rate could potentially vary from .00 which reflects no personnel change in the period studied to 1.0 which indicates perfect mobility with change in personnel in every position in every year. The turnover rate did, in fact, vary from 73 .00 to .60 for the districts studied--505 of the original 508 districts. The statewide average turnover was .21 for the eleven years considered. The degree of professional Specialization present in a bureau is measured by the extent to which a school district has divided up the administration of the district. The Michigan Department of Education requires each district to submit on the fourth Friday of every school year a report of student enrollment and personnel employed in order to obtain their state aid allotment.52 In the breakdown of employees, the districts indicate the numbers and types of professional and non-professionals working for the district. From this breakdown, it is possible to ascertain the degree to which the administrative tasks have been divided into specialized areas. To be useful, this data has to be transformed so a comparison can be made as to how professionally specialized the 508 districts are. A Guttman Scale was created to accomplish this for the 1970-1971 school year. The presence or absence of six administrative positions in a district are combined into the scale reflecting professional specialization. The positions are business manager, personnel manager, research director, attendance officer, maintenance engineer and professional health. The resulting cumulative scale meets the usual criteria for being considered an approximate uni-dimensional scale. The co- efficient of reproducibility is .95 with a minimal marginal repro- ducibility of .82. The degree of rule formalization is reflective of the flexi- bility or rigidity of the rules of the bureau--informal as well as formal. To measure this flexibility or rigidity, it is necessary to 74 find a set of formal rules which all districts potentially have since ascertaining the existence of informal rules and their nature is extremely difficult because only by intensive contact with an organ— ization can one gather information about its informal structure and rules. To do this for 508 districts is a formidable task even if unlimited money and time is available for such a project. We were able to find data on a set of formal rules which all districts have-- rules dealing with the teacher-administration contractual relationship. By examining teacher contract provisions, we develOp at least one potential measure of rule formalization.53 Five teacher contract provisions relating to restrictions on the freedom of action Of administrators are combined into a Guttman Scale. The presence or absence of the five contract provisions make up the scale--class size limitations, grievance procedures on class size, grievance procedures for probationary teachers, consultation with teachers prior to adaption of Operational millage requests, and binding arbitration as the terminal step in grievance procedures. The coefficient or repro- ducibility for the scale of rule formalization is .90 with a minimal marginal reproducibility of .72. To measure the recruitment pattern of bureaus, we examine the sources of replacement personnel when administrative turnover takes place in a bureau. There are two sources of replacement for a bureaucrat when he leaves his current position. The bureau can promote/demote some one already in the bureau to fill that position or the bureau can go outside itself to find a replacement. In rare situations, the bureau may decide to do away with the position or leave it empty fer a long period of time. Since we are interested in 75 the source of replacement, this unusual event is not incorporated into the determination of a bureau's recruitment pattern. As in the measurement of administrative turnover, this measure of a bureau's recruitment pattern is limited to those administrative positions potentially involved in undertaking educational innovation and reported by the school district.54 The actual measure is the percentage of recruitment that takes place outside the bureau for the period from school year 1958-1959 to school year 1969-1970. The range of outside recruitment can be from zero per cent where all recruitment is from within the bureau to 100 per cent where all recruitment is from out— side the bureau. The actual recruitment patterns do vary from zero to 100 per cent. The average recruitment pattern for the 505 districts with data available is 40 per cent from outside. Level of financial support of a bureau can be measured by a number of different fiscal figures. The total instructional expendi- ture per pupil, current operating expenses per pupil, total general fund expenditure per pupil and various millage rates (Operational, building and Site, debt reduction) all measure the financial condi- tion of a school district.55 Because we are trying to capture something of the nature of the local support for the district, it is necessary to select a fiscal measure that the local community can directly influence without intervention by other governmental agencies--i.e., county or the state. The only measure of support which does this and is available statewide is extra operating millage voted--the tax rate in mills applied to the state equalized valuation . . 56 of the district to produce revenue for the operation of schools. 76 Unlike millage for building and Site and expenditure measures, this millage is a reasonably current indicator of support since the time frame for the millage normally runs from one to three years. Extra operating millage is much more responsive to fluctuations in support than are other millages. The expenditure measures include monies received from the state and federal governments which contaminates them as indicators of local support. The range of extra Operating millage voted is from 3.0 mills to 29.0 mills. The statewide average is 13.6 mills. The final measure of the opportunity structure is size of the bureau. This is a very simple measure Since size can be considered as nothing more than the number of bureaucrats working in the bureau. This is a more than adequate indicator of size. However, in this re- search, we have determined that a stronger measure of size is gained by limiting size to only those bureaucrats involved in administrative functionS--leaving out those working in instructional and non- administrative assignments.57 The range of size of the districts studied is from 8 to over 2,000 administrative employees. The dependent variable--educational innovation--was defined earlier as the introduction to an applied situation of means and/or ends that are new to that situation. Finding an operational measure of innovation which meets this definition and is comparable across districts was difficult. After exploring various alternatives, the only measure of innovation readily available statewide involves educa- tional activity associated with Title III, Elementary and Secondary Act (ESEA) of 1965 and 1967 (also identified as PACE--projects which 77 advance creativity in education). The Title III program was designed to (1) encourage the development of innovations, (2) demonstrate worthwhile innovations in educational practice through exemplary programs, and (3) supplement existing programs and facilities.58 Federal funds became available to local school districts that applied and were able to survive the screening of the state departments of education and the Office of Education, US Department of Health, Education and Welfare. Our measure of educational innovation is then drawn from the efforts of Michigan school districts to obtain funds under Title III, ESEA. Because of the limited funding that has taken place fer programs under Title III in the years since its passage, it is necessary to limit the indicator of innovation to only attempts of school districts to receive funds rather than their success in obtaining funds. Many more programs have been suggested by the local districts than have ever been funded. Educational innovation is measured by examining the efforts of the 508 school districts to receive Title III funding. Since the original opportunity to apply for assistance in 1966, there have been seven other times that districts could have applied for funding. Two possible alternative indicators of school district activity under Title III can be constructed. In both measures, all districts in Michigan were potentially capable of undertaking the innovative activity. The first measure of innovation under Title III is a dichotomous indicator in which we simply determine if a school district has ever submitted a proposal to receive Title III funding in any of the possible submission periods. From 1966 to 1971, a total of 78 116 school districts have made application for funds to support what they believed to be innovative programs in their own districts.59 The second possible measure is based on the number of Title III prOposalS submitted by a school district. The number of proposals submitted by a district has varied from a single attempt to receive Title III funding to as many as eleven different proposals. Arguments can be made supporting the use of either measure of innovation. The simpler dichotomous indicator has two principal factors supporting its use. In examining the development of Title III, ESEA, we found that a large proportion of the districts who have ever applied did so in the first submission period only to find out that the funds were not as readily available as they had been led to believe. They apparently became discouraged and never applied again after their initial rejection. The second factor is that the number of attempts made/proposals submitted is more than just educational innovation. It taps an element of "gamesmanship" or "grantsmanship" which is not directly a part of educational innovation. It includes the prestige associated with gaining the funds and the recognization that goes to the bureaucratic leadership fer obtaining these funds. The argument for the use of the number of proposals submitted rests on this last point raised supporting the dichotomous indicator. The reason the interval measure should be used is because it does incorporate the element Of "gamesmanship" and "grantsmanship." These secondary factors/influences are an important part of-basis for being innovative in a bureau. The prestige associated with having a grant approved and the recognition that goes to the leadership of a bureau 79 for obtaining the extra resources are essential components of the overall innovative process and must not be excluded if one is to really measure innovation. The unanswered question is which alter- native indicator of innovation is the better or more useful. To discover the answer, both measures are utilized in testing the hypotheses. A question that cannot be resolved is whether or not ESEA Title III prOposals are a "good" indicator of education innovation. Are the programs being suggested in these Title III proposals "truly innovative" educationally? The criteria used by the state to define what is educationally innovative changed from submission period to submission period. There are no objective standards by which one can evaluate plans/programs for their innovativeness. To the district submitting the proposal, the program seemed innovative whether or not the state officials agreed. We cannot really answer the question about the quality of the programs submitted in the proposals as to their "innovativeness" and it is not really important to this research. AS far as our definition of innovation, the Title III proposal is an acceptable indicator of education innovation. In this chapter, we set forth the theoretical foundations guiding this research and explore the nature of the bureaucratic opportunity structure. The hypotheses to be tested are developed and discussed. Finally, we treat the procedures by which the hypotheses are tested and the operational measures of the independent and dependent variables used to test the hypotheses. Chapter IV presents the results of testing the first eight hypotheses. The final hypotheses are tested in Chapter V. FOOTNOTES-~CHAPTER III 1Niskanen, Bureaucracy and Representative Government, pp. 5-6. 2William H. Riker, The Theory_of Political Coalitions (New Haven: Yale University Press, 1962), pp. 24-27. 3Defining a bureaucrat this way suggests that there are no studies relating ambition, the structure of political opportunity of bureaus and bureaucrats. Swinerton's research deals with top level administrators who are either directly elected or appointed by an elected official thus binding their continuance in office to that appointing official. This research is the first real attempt to identify a bureaucratic structure analogous to the politician's structure of political opportunity. 4The finding of Iannaccone and Lutz's research that the school superintendent is replaced within three years after the election of a new school board member is not seen as being typical. For this event to take place, there must normally be a conflict situation that has produced serious disagreement over educational policy along with a major shift in school board membership due to recall action or a large scale election defeat of incumbent members. 5See Downs, pp. 2-4 for a complete discussion of his assump- tions. The source of the modifications to Downs' assumptions is R. Duncan Luce and Howard Raiffa, Games and Decisions (New York: John Wiley 6 Sons, Inc., 1957), pp. 12438T‘ 6A more complete discussion of this distinction is made by Downs, pp. 82-83. These are the same goals or motives used by Downs to establish his typology of bureaucrats. They were selected for use here because they are considered as a part of a bureaucrat's official role motivation and lend themselves to different suggestive preference orderings. There are a number of other possible ways to 80 81 look at possible preferences or goals (i.e., material/non-material, financial/non-financial, or maintenance/motivation goals) but they can all be divided into the nine different type goals of Downs. This typology of goals is seen as being exhaustive. It is also viewed as being basically mutually exclusive. If one were to view these goals as being continuous in the sense that the law of diminishing return would apply, then these goals would be considered as only three different types. Power, prestige and money income would be one type of goal. Security and convenience would be the second and the third type would be composed Of personal loyalty, pride in performance of work, commitment to a program and service of the public interest. These three goals could then be thought of as continuous but it would not later the basic argument being made. See Downs, pp. 84-89 for his use Of these nine goals. 8Downs, pp. 90-91. 9Schlesinger, p. 6. 1oIbid., p. 11. 11Unfortunately there has not been a great deal of work done in this direction to the present time. Gordon Black makes a limited attempt to do this for city councilmen in "A Theory of Professional- ization in Politics," 0 . cit. R. Engstrom suggests the effect of ambition on prosecutors in a research note titled "Political Ambitions and the Prosecutorial Office," 0 . cit. Swinerton also undertakes to relate ambition to potential behav1or in "Ambition and American State Executives," Op. cit. 12The bulk of Ambition and Politics is devoted to identifying the structure of political Opportunity in the United States. There are some general characteristics of this country's opportunity structure which are potentially suggestive to an examination of the environment bureaucrats Operate within. The most salient charac- teristic of the structure of political opportunity to Schlesinger is the proliferation of outlets for political ambition. This is certainly a characteristic of the general milieu bureaucrats face. Bureaucracy is a rapidly growing activity--especially at the State and local levels--providing many different agencies in which a bureaucrat may maximize his goals. Another feature of the political opportunity structure is the openness of the lines of political advancement. In a similar fashion, the lines of advancement within the bureaucracy are open in that there are not prescribed charac- teristics such as religion, race, social class, age, etc. for individual advancement in bureaucratic office. At the same time, there is not the openness of the political Opportunity structure since there are requirements as to education, professional training, and experience which must be met by a bureaucrat wishing to advance. The degree of risk present in the structure of political opportunity is a 82 consideration that the politician must always keep in mind. This feature is less important to the bureaucrat although it cannot be totally ignored. Most bureaucrats are provided with some system of job security--i.e., tenure or merit system--which is designed to remove them from certain types of political pressures. However, the bureaucrat is still faced with a degree of risk in maximizing his personal preferences. There is no certainty when it comes to his next promotion. Another dimension of the political opportunity structure-- the two party system--is not directly important to most bureaucrats. The majority of bureaucrats are shielded from the impact of the party system. Only those tOp level officials who deal directly with those actors making-up the political sovereign need to take the effects of the party system into their immediate calculations. And even here there are certain norms which act to reduce the effect of partisanship upon bureaucrats. The identification that change in the structure of political opportunity is slow and predictable is also a reasonable conclusion for the structure of Opportunity for bureaucrats. There are seldom major revisions in the short run that have direct impact on the opportunity structure. The bureaucrat basically faces a stable environment overtime. Even when change occurs, the bureaucrat can reasonably estimate the impact upon his opportunity Structure. While we have presented examples of Similarities between the struc- ture of political opportunity and the Opportunity Structure of bureaucrats, we feel that the utility of these bureaucratic features is greatest when comparing the opportunity structures of bureaus with different social functions rather than when examining bureaus with the same social function as in this research. 13Downs' argument in Inside Bureaucracy rests on three central hypotheses--one of which assumes that an organization's social function Strongly influences its internal Structure and behavior and vice versa. This suggests the social functions influence behavior through their impact upon individual officials whose behavior is in reSponse to their environment. Hence, social function does impinge on the opportunity structure facing the bureaucrat. This is demonstrated at various points in the book. See pp. 41-48 and 92-111 as examples. 14Schlesinger, O . cit., p. 15. What little data that does exist is basically related to the federal bureaucracy. The major published study on federal bureaucrats is W. Lloyd Warner, et al., The American Federal Executive (New Haven: Yale University Press, ). A new major large scale study of state executives--"The American State Executive"--is presently being conducted by Fred W. Grupp, University of Connecticut and Alan Richards, Louisiana State University. The results of this ten-State study Should be published within the next year or so. 15A third unit could be a city or county government. This is a relatively rate situation and will not be considered here as it is not appropriate in the State under study, Michigan. 83 16See Downs, pp. 44-47 for a more complete discussion of the external power setting of a bureau. 17Gittell and Hollander, Six Urban School Districts; Rosenthal, Peda o ues and Power; Masotti, Education andiPoiitics in Suburbia; and Rogers,’i10 Livingston Street. 18Examples of research into the budgetary process include John P. Crecine, Governmental Problem-Solvipg (Chicago: Rand McNally E Co., 1969); Fremont J. Lyden and Ernest G. Miller, Plannin , PrOgrammin , and Bud etin (Chicago: Markham Publishing Co., 1967); Aaron Wil avsky, The Politics of the Budgetagy Process; and Donald Gerwin, Budgeting Pfiblic Funds. 19Downs, pp. 136-40. He does suggest that bureaus with less organizational Slack will be less able to undertake innovative action. 20Samuel and Mannheim, p. 225. 21Downs, p. 99. 22William H. Starbuck, "Organizational Growth and Development, Handbook of Organizations, ed. by James G. March (Chicago: Rand McNally, 1964). 23Downs, pp. 263-64. 24Bidwell, p. 1000. 25Downs, pp. 11-15. 26Rogers, 110 Livingston Street. 27Meyer, "The Two Authority Structures of Bureaucratic Organizations." 28Downs, p. 103. 29 . Rogers, Op. cit. 30Downs, pp. 59-61. 84 31Bidwell, "The School as a Formal Organization." 32 . Rogers, op. cit. 33Downs, pp. 228-31. 34Miner, The School Administrator and Organizational Character; Greg and Knezevich, "The Superintendent: What Makes Him What He Is"; and Bidwell, Gittell and Hollander do not find a relationship between size and innovation even though one was predicted. 35Downs, pp. 140-43. 36Ibid., pp. 5-23. 37This definition is very similar to Jack L. Walker's definition of innovation in "Innovation in State Politics," Politics in the American States, ed. by Herbert Jacob and Kenneth N. Vines (2nd ed.; Boston: Little, Brown and Co., 1971), p. 355. Walker sees innovation as simply a program or policy which is new to a situation. It makes no difference that the program is old or has been used before by others. AS long as it is new to the unit adopting it, then it is innovative. 38Swinerton found ambition of state executives varies with their age in "Ambition and American State Executives." Hain also finds that ambition varies with the age of State legislators in "The American State Legislators' Ambitions and Careers." 39Schlesinger, p. 174. 40Starbuck, op. cit., p. 509. 41Downs, p. 11. 42Bidwell, p. 1000. 43Downs, pp. 95-96. 44Ibid., pp. 59-61. 85 45The simple correlation in 1971 between the total number of students in a school district and the number of employees in the schools studied in Michigan is .994. Size of bureau is almost perfectly reflective of the number of students it must deal with in carrying out its social function. The fact that there is not a "perfect relationship" is very likely due to the lag in reflecting student enrollment when there is a decline in that enrollment. 46Large organizations are defined by Downs as organizations in which top officials know less than one-half the total employees of the organization. To find out what large meant empirically, we turned to organizational theory. While all classifications are somewhat arbitrary, Caplow suggests that small organizations are those organizations with fewer than thirty members where there exists the possibility of forming a primary group. This is exactly what Downs desires to eliminate by defining bureaucracy as large--bureaucracies are secondary groups to those who serve in them. This, then, is the basis Of selecting thirty staff members as the cutting Off point for defining an organization as a bureaucracy. Theodore Caplow, Prin- ci les of Or anization (New York: Harcourt, Brace 5 World, Inc., , p. 2 . 47The data used in determining the age of bureaus was obtained from information supplied by Roger Boline, Director of School Management Services, Michigan Department of Education. 48Michigan, General School Laws of Michigan, Statutes (1959). 49Magnifying this problem are the teacher tenure laws of the state and the unionization of the non-professional staff. While it is relatively easy to add staff in an expanding district, it is extremely difficult to reduce staff levels unless the school district is in extreme financial difficulty. Michigan law states that tenured teachers may only be laid off when the district is experiencing financial difficulties and only then if the prOper procedural steps are taken. Even then, the remaining employees may threaten "to withhold their services" if their fellow employees are let go. Downs also discusses decline in bureau employment, see pp. 13-14. 50The source of the data of the growth rate of bureaus is drawn from Bulletin 1012 1966-1967 and 1970-1971, Rankin of Michi an Public Hi h School Districts by Selected Financial Data iLansing, Mich.: Michigan Department of Education, December, 1967 and 1971). The 1966-1967 1012 Bulletin was the earliest one available for use. In all of our longitudinal measures, we attempted to use the 1958-1959 school year as the base point Since this was the school year in which the major change in school laws took place. Unfortunately, the data is very often incomplete or unavailable back to this school year, so we have been forced to use different base years in determining these 86 measures. Every effort was made to obtain the longest time frame possible because of the slowness of change being manifested in bureaucratic organizations. SlMichigan Reading Circle, Michigan Education Directory and Buyer's Guide (Lansing, Mich.: Michigan ReadingiCircIe, 1959-1970). 52The actual source of the data used is a report 1970-71 Employee_£ersonnel Data in Michigan Public Schools (E. Lansing, Mich.: Michigan Education Association, 1971) prepared by the Michigan Education Association from information supplied by the Michigan Department of Education. 53The contract provisions are contained in the Michigan Education Association's Summary of Selected A reement Provisions 1970-1971 (E. Lansing, Mich.: Michigan Education Association, 1971). Oily 441 of the 508 districts are reported in this document. This is due to: (l) the failure of a number of local teacher associations to provide their contract provisions to the Michigan Education Association, and (2) the fact that some district's teachers are represented by the American Federation of Teachers which does not provide MBA with data about those districts. 54Michigan Reading Circle, Michigan Education Directory and Buyer's Guide . 55These various fiscal measures are defined in the Ranking of Michi an Public Hi h School Districts by Selected Financial Data 1975-1971 Bulletin I012. 561bid., p. 4. 57The breakdown of personnel is from the 1970-1971 Employee Personnel Data in Michigan Public Schools, 9p3 cit. 58Office of Education, U.S. Department of Health, Education and Welfare, Manual for Project Applications and Granters, Title III, ESEA (Washington, D.C.: Government Printing Office, 1967). For a more complete discussion of Title III and ESEA in general, see Stephen K. Bailey and Edith K. Moser, ESEA: The Office of Education Administers a Law (Syracuse: Syracuse University Press, 1968); Stephen K. Bailey, The Office of Education and the Education Act of 1965 ICP #100 (IndianapOiis: )The Bobbs-Merrill Co., Inc., 1966); and Marcia M. Boznanzo, "An Evaluation of Selected Title III Inservice Projects in Michigan as Perceived by Participating Teachers on Thirteen Variables" (unpublished Ph.D. dissertation, Michigan State University, 1968). 87 59The data on school district requests for Title III, ESEA funds was obtained with the cooperation and assistance of Eugene Paslov, Coordinator, ESEA Title 111 Unit, Instructional Innovation Program, Michigan Department of Education. CHAPTER IV TESTS OF THE INDIVIDUAL DIMENSIONS OF THE OPPORTUNITY STRUCTURE Before presenting the actual tests of the first eight hypo- theses, several points should be briefly discussed. As was pointed out in Chapter III, the directional relationships Stated for these hypotheses are not always grounded upon firm theoretical foundations. In fact, very plausible arguments could be made in several cases supporting the opposite position. Thus, we are facing here what is really an exploratory test for these hypotheses. Hopefully, the results of the tests of these hypotheses will clarify the situation. A more important consideration is the nature and source of the data used in testing the hypotheses. This investigation relies almost entirely upon information about school districts originally collected by the Michigan Department of Education and the Michigan Education Association. While there are many advantages in using aggregate data of this type-~ready availability, relative low cost, basic compatabil- ity with the needs of this research--there are also certain inherent disadvantages. The purposes for which the data are gathered are rather heterogeneous thus providing us with data that are not always the most suitable for this research. Another feature of this data is 88 89 the fact that these information gathering organizations did not gather the data directly but rather depended upon "the good will" of the districts to provide complete and accurate information. Overall, the districts seem to have done this. At the same time, one is prudent to remain slightly skeptical of data collected in this manner. To this point, there has been no mention of the statistical techniques being used in this research. While many are potentially acceptable, we attempt to strike a balance between under-utilization of Statistical methods and an overdependence upon inappropriate techniques which are not justified given the data and the assumptions necessary for their correct use. Before touching upon the Specific statistics chosen, the applicability of significance tests to this research will be examined. Our basic position, like that of Morrison and Henkel, is that significance tests are very often misused and indiscriminately applied in behavioral research.1 Without going into an elaborate discussion of the use of significance tests in statistical inference, we suggest that the assumptions made in using Significance tests are violated by the data.2 First, the data are not gethered by any type of random sampling prOcedure. We have the complete universe of Michigan school districts meeting the definition of a bureaucracy. Thus, having a universe or population we are faced with the question of what inference are we to make by testing for Significance. Even if we were to consider every district in the state, the data for the 508 districts selected cannot be considered a random sample of the total population of districts. These districts would either be a purposive or quota sample-~not a random sample--which is a violation 90 of one of the major assumptions of significance tests. Another dif- ficulty in using significance tests is the relationship of the size of the sample to the power of the test. As Kish points out, " . . . if the sample is large enough the most insignificant relationship will appear 'Statistically significant.'"3 Having actually performed significance tests on several of the weak relationships fOund, we find support for this position in our research. We conclude for these reasons that Significance tests are inappropriate to this research and will not be reported here. A number of statistical tests are utilized depending upon which measure of educational innovation is used. When educational innovation is considered simply as the presence or absence of ever having submitted an ESEA Title III proposal during the seven possible submission periods, the primary method of analysis is the contingency table. The coefficient of differentiation (theta) and the correlation ratio (eta) are applied where appropriate to demonstrate the strength of association and the variance accounted for by the independent variable.4 The other measure of innovation--an interval measure based upon the total number of Title III proposals submitted by a district--permit the use of more "powerful" statistical tests in addition to the examination of the contingency tables. Coefficients of correlation (Pearson's r) demonstrate the Strength and direction of the relationship between the independent and dependent variable. Regression analysis is also performed in one hypothesis to explore a suSpected curvilinear relationship. The statistical tests used in testing the cluster hypotheses are discussed in Chapter V. 91 Hypothesis One In discussing the test of each hypothesis relating an indi- vidual dimension of the Opportunity structure to educational innova- tion, a brief synopsis of the development of that hypothesis is given as an aid in interpreting the findings. The first hypothesis suggests that the level of financial support of a district is related to educational innovation. The overall level of financial support reflects to a large extent the amount of resources available to per- form a bureau's social function. Those districts with greater resources will be better able to divert a portion of those funds to support innovative purposes than those districts which have fewer resources available to them. Thus, it is suggested that the higher the level of financial support a district receives, the greater the educational innovation taking place in that district. The measure of financial support utilized here is the extra operating millage voted by the citizens of each district. We expect districts with higher extra operating millage rates will be more educationally innovative than districts with lower millage rates. Table 4-1 presents the relationship found between the level of financial support and the dichotomous measure of educational innovation--presence or absence of ever having submitted a Title III proposal. In examining these results, one of the first things noticed is the relatively low proportion of districts which have ever sub- mitted innovative proposals. Only 116 of the 508 districts investi- gated (23 per cent) have attempted to receive funding in the seven potential submission periods. This seemingly low level of activity 92 TABLE 4-l.--Relation of Level of Financial Support of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Level of Financial Support (Mills) Educational Innovation Low(15) NO 87.2% 81.1% 66.3% 392 YES 12.8 18.9 33.7 116 Total 100.0% 100.0% 100.0% N 148 + 164 + 196 = 508 Y’Miliage = 13.6 n = .20 n2 = .045 under Title III might reflect the fact that Michigan districts are not particularly innovative as a whole. The opposite Situation might also be true. Michigan districts may really be innovative when contrasted with school districts of other states. Unfortunately, comparative state data are not available so this question cannot be resolved here. We can, however, discuss the hypothesis. The hypothesis is supported by the findings in Table 4-1. As the level of financial support of a district increases, the percentage of districts who have submitted Title III proposals also increases. The relationship is not par- ticularly strong as only 4 per cent of the variance in this innovative measure is accounted for by knowing the level of financial support. To insure that this association is not being influenced by the other dimensions of the opportunity structure, each dimension is introduced as a separate control variable. None of them has an influence on the relationship found in Table 4-1.6 The control relationships are fOund in Appendix I. 93 When using the other indicator of educational innovation-- number of prOposals submitted by a district, we are able to improve upon the first results. As shown in Table 4-2, districts with a high level of financial support submit more proposals than do school dis- tricts with lower levels of support. The amount of variance accounted for increases slightly to 5 percent. In this hypothesis, the number of Title III proposals submitted is a better measure Of educational innovation. While the strength of support is only moderate, districts with higher levels of financial support are associated with greater amounts of educational innovation. TABLE 4-2.--Relation of Level of Financial Support of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Level of Financial Support (Mills) Educational Innovation Low (<10) Medium(lO-15) High(>15) 0 87.2% 81.1% 66.3% 392 1 10.1 11.0 14.3 61 2 or more 2.7 7.9 19.4 55 Total 100.0% 100.0% 100.0% N 148 + 164 + 196 = 508 r = 23 r2 = 05 Hypothesis Two In this hypothesis, we postulate that the age of a bureau will have impact upon the amount of educational innovation taking place in that bureau. Borrowing from both Downs and Schlesinger, we 94 argue that age is a depressant upon the capability of a school district to be innovative--as districts become Older, their rigidity increases and the discretion available to the decision-maker decreases. The older the district, the less innovation taking place in that district. Our measure of age of bureau is the length of time that a school's bureaucracy has administered the district it serves. The number of years of administration is measured from the time of the creation of the district or its most recent major reorganization to the end of the 1970-71 school year.7 The relationship between age of bureau and the presence or absence of innovation is Shown in Table 4-3. The relationship exhibited is in the correct direction but it is extremely weak. Younger bureaus are only 3.4 per cent more innovative than the older bureaus. Knowledge of bureau age does not account for even 1 per cent of the variance in innovation. TABLE 4-3.--Relation of Age of Bureau and Educational Innovation ((Ever Submitted Title III Proposal) Educational Innovation Age of Bureau* Younger(<13 years) Older(>13 years) NO 74.1% 77.5% YES 25.9 22.5 Total 100.0% 100.0% N 454 + 54 = 508 7' Age = 30.8 n = .04 n2 = .00 *Division based upon the change in school district organ- izaigog requirements caused by the revision of Michigan school laws in 5 95 Control variables were introduced to ascertain if any of the other independent variables might be suppressing an association between age and innovation--none were found. These results are presented in Appendix II. Table 4-4 diSplays the relation of age of bureau and innova- tion considered as an interval measure (number of prOposals submitted). We continue to find that age of bureau is only slightly correlated to educational innovation. Younger districts are a little more inno- vative when submitting one proposal only. When considering submission of two or more proposals, older and younger districts are not really different. We must again conclude that age is apparently not related to innovation. This continues to holduafter controlling for the other independent variables (Appendix II). TABLE 4-4.--Relation of Age of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Age of Bureau Educational Innovation Younger(<13 years) (Older(>13 years) 0 74.1% 77.5% 1 14.8 11.7 2 or More 11.1 10.8 Total 100.0% 100.0% N 454 + 54 = 508 2 96 In a last attempt to discover the existence of a relationship between age of bureau and educational innovation, another measure of age is tried. In a recent article by Ripley, Moreland and Sinnreich, they suggest using the number of years since the replacement of the head of a bureau by a new administrator as an alternative way of conceiving a bureau's age.8 Since this information had already been gathered, this new indicator of age was tested with innovation. The results are contained in Tables 4-5 and 4-6. The relationship between age of bureau and innovation is slightly strengthened when this alter- native measure of age is used. Even though this improvement is not terribly large, school districts having the newest superintendents are a little more innovative with either measure of innovation than are districts having chief administrators with longer service. TABLE 4-5.--Relation of Age of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Age of Bureau (Years Since New Superintendent) Educational Innovation 0-4 Years 4-8 Years 8-12 Years NO 73.8% 80.2% 81.0% YES 26.2 19.8 19.0 Total 100.0% 100.0% 100.0% N 249 + 131 + 126 = 505* n = .05 n2 = .00 *Missing Data. 97 TABLE 4-6.--Relation of Age of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Age of Bureau (Years Since New Superintendent) Educational Innovation 0-4 Years 4-8 Years 8-12 Years 0 73.8% 80.2% 81.0% 1 12.7 10.7 11.1 2 or More 13.5 9.1 7.9 Total 100.0% 100.0% 100.0% N 249 + 131 + 126 = 505* r = -.09 r2 = .01 *Missing Data. The basic conclusion about age and innovation is not changed however by this new measure. The data only marginally supports the hypothesis developed. There is inadequate evidence here to conclude that the age of a district is related to the level of educational innovation in that district--regardless of the indicator of age chosen. Hypothesis Three Hypothesis Three suggests that the growth rate of a bureau is directly related to the level of innovation taking place in that district--the faster the growth rate of a school district, the greater the educational innovation taking place in that district. Downs' argument about rapidly growing bureaus being faced with new Situations and problems is the principal source of support for this hypothesis. Because of the new Situations and/or problems, bureaus 98 are faced with conditions for which the existing rules and guidelines are inadequate. Therefore, bureaus must come to grips with the difficulties arising of rapid growth through the adoption of new or innovative solutions. Actually measuring growth in school bureaus does present some difficulties which are resolved by using change in pupil enrollment in a district as the indicator of growth. Growth rate of a bureau is the average yearly change in student enrollment for the five-year period, 1966-1967 to 1970-1971. As is shown in Table 4-7, the growth rate of a bureau is not related to educational innovation in the manner hypothesized. The slight relationship exhibited is negative-~the higher the growth rate of a bureau, the lower the level of educational innovation taking place. Controls were introduced to ascertain if one of the other independent variables was distorting or suppressing this association. TABLE 4-7.--Relation of Growth Rate of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Growth Rate of Bureau Educational Innovation Stagnant/ . Declining (1‘3fiaéé) (gzgg%) (<+1.0S%) ' ' ' NO 70.5% 78.7% 81.7% YES 29.5 21.3 18.3 Total 100.0% 100.0% 100.0% N 139 + 249 + 120 = 508 X.Growth = +1.8% per year n = .05 n = .00 99 As in the earlier hypotheses, these variables do not influence the relationship found. (See Appendix III.) In Table 4-8, we find similar results to those in Table 4-7. School districts with stagnant or declining enrollments submit more innovative proposals than do districts with higher growth rates. The relationship remains very weak with only 1 per cent of the variance being accounted for with knowledge of the growth rate of the district. No suppressor or distorter variables were discovered when controls were introduced (Appendix III). The evidence found in Tables 4-7 and 4-8 is inadequate to conclusively reject the original hypothesis and conclude that there is a negative association between the growth rate of a bureau and the amount of innovation in a bureau. At this point, we neither reject nor accept the proposed hypothesis. TABLE 4-8.--Relation of Growth Rate of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Growth Rate of Bureau Educational Innovation ggifgggfié Average High (<+1.0%) (1.0-4.0%) (>4.0%) O 70.5% 78.7% 81.7% 1 13.2 11.5 10.9 2 or More 16.3 9.8 7.4 ___,___ .______ ______. Total 100.0% 100.0% 100.0% N 139 + 249 + 120 = 508 2 r: —.10 r -'-' .01 100 Hypothesis Four The relationship between the administrative turnover rate of a bureau and the level of educational innovation is formulated in this hypothesis. Administrative turnover rate of a bureau reflects the mobility of bureaucrats within a bureau which, in turn, has been previously found to be associated with administrative behavior. Downs feels that where there is a high level of turnover or mobility, there is a heightened or increased potentiality of innovation taking place because of the fresh ideas, information, perceptions and need for success which these new people bring with them. Thus, we propose a direct relationship between the administrative turnover rate of a bureau and the level of innovation--the higher the turnover rate, the 1 higher the level of innovation. The measure of administrative turn- over is based upon the mobility of those bureaucrats who are poten- tially involved in making decisions about innovative activities in the school district. The administrative turnover rate is the change in personnel per administrative position per year for the eleven year period ending in the 1969-70 school year. The range of turnover can be from zero or no turnover per position per year to 1.0 or complete turnover for every position for every year. The results in Table 4-9 are rather interesting when innova- tion is measured as the dichotomous dependent variable. Rather than the expected linear relationship, a nonlinear association seems to exist. As the administrative turnover rate of the bureau increases from zero, educational innovation also increases until the turnover rate exceeds .30 when there is a rather large drOp in the innovation 101 taking place. The test of association eta (a linear test) also supports the possibility that there may be a nonlinear relationship. Eta is zero which would be the expected result if curvilinearity exists between the two variables. TABLE 4-9.--Relation of Administrative Turnover Rate of Bureau and Educational Innovation (Ever Submitted Title III PrOposal) Administrative Turnover Rate Educational Innovation Low(<.lO) Medium(.lO-.30) High(>.30) NO 80.0% 74.9% 84.8% YES 20.0 25.1 15.2 Total 100.0% 100.0% 100.0% N 50 + 356 + 99 = 505* xTurnover Rate = .21 n = .00 n2 = .00 *Missing Data. A similar result is obtained in Table 4-10 when using the interval measure of innovation. 'As the turnover rate increases, the number of prOposals submitted by a district also increases until a certain level of turnover is reached at which time a decline in innovation takes place. Controls were introduced for both measures of innovation and the relationship remains the same (see Appendix IV). Very much like the first test, the correlation between turnover rate and innovation is nearly zero as one would expect if the association is nonlinear. Because of this apparent nonlinearity, regression analysis is conducted to determine more about the nature Of this 102 relationship between administrative turnover rate of a bureau and educational innovation. TABLE 4-10.--Relation Of Administrative Turnover Rate of Bureau and Educational Innovation (Number of Title III PrOposals Submitted) Administrative Turnover Rate Educational Innovation Low(<.10) Medium(.lO-.30) High(>.30) 0 80.0% 74.9% 84.8% 1 12.0 13.0 8.1 2 or More 8.0 12.1 7.1 Total 100.0% 100.0% 100.0% N 50 + 356 + 99 = 505* r =: -.06 r2 = .00 *Missing Data. As a first step in this process, the administrative turnover rate is plotted against educational innovation for each five points of increase in turnover. In Figure 4-1, educational innovation is the simpler dichotomous measure converted to a percentage of all districts who have ever submitted a Title III proposal-~the number of yes districts divided by the total number of districts at the points of change in the administrative turnover rate.9 An examination of Figure 4-1 does not really clarify the problem because in a two- dimensional Space it is always possible to fit both a straight line and a curve to a given set of points as is done here. To discover which line describes the set of points better, some sort of criterion 103 :oHum>o::H Hecowueoaem one can: uo>oeuab o>fiuaupmwcfiavocn=h o>wuauumwcwau< oo. mm. om. mv. ov. mm. on. mN. on. ma. 0H. mo. 1 A d J - d u - q . A Home“; /. newmmouuom\\\x :IIIIIIIIII // / / \ . / I I / \ III/III: \\ cowmmoumom aflumuve=0.\\\\\\\w _II I O 0-4 mm ON tasodOid 111 01111 a penitmqns Bataan IOAQ sietxista 30 iueeied In N on 104 is necessary--which line provides the better least square fit for the data. In this case, the second degree polynomial provides a fairly close fit to the data contained in Figure 4-1 and does explain more variance than the linear regression.10 The simple linear equation for this collapsed data accounts for 20 per cent of the variance while the nonlinear equation increases the variance.explained to 56 per cent. Other and possibly better fits could be obtained with other models but they have not been explored since this simple quadratic equation amply demonstrates that the relationship is curvilinear. From the evidence here, we cannot conclude that the linear hypothesis between educational innovation and administrative turnover rate is supported. A different kind of relationship exists--a curvilinear association where innovation is at zero with zero turnover rate, increases as the turnover rate increases until an excessive level of turnover is reached at which time innovation returns to zero, and remains there as the turnover rate increases until the maximum is reached. The strength of this association is fairly strong with over 50 per cent of the variance of the collapsed data being accounted for. Hypothesis Five The extent to which a bureau divides the tasks of performing its social function into units of Specialized activity--the degree of professional specialization--is hypothesized as being related to the level of educational innovation in a bureau. The greater the degree of professional Specialization, the greater the educational innovation taking place in that district. The degree of professional specialization in education has been previously found to be 105 associated with administrative behavior by many researchers. Of particular utility here is the importance of specialization to the availability and cost of information in a bureaucracy. With increased professional specialization, greater amounts of information become available as to the broad range of activities making up the bureau's social function. This heightened level of information provides those in the bureau with new ideas/techniques at a lower cost which, in turn, reduces the costs of undertaking innovative activities. The degree of professional Specialization in a school district is reflected by the score that the district receives on a Guttman scale combining the presence or absence of six administrative positions in‘a district. This measure of professional Specialization is properly considered an ordinal level measure. Interval level statistics would normally be inapplicable.11 However, an argument can be made for their use and has been made by some statisticians. Hays suggests "There are very many instances where the level of measurement may not reach the level supposed by the statistical technique, and yet the method itself may be quite adequate for showing what the experimenter wants to know."12 Blalock also argues in this direction when he points out that ". . . it may turn out that it is no more misleading to make use of dubious assumptions about the level of measurement than it is to make use of data involving . . . ordinal scales that obscure differences in the amount of variance."13 In this analysis when we are faced with an ordinal scale (professional specialization/ rule formalization) as a measure of an independent variable, the "dubious assumption" about the correct level of measurement is made 106 with the ordinal scale being considered an interval level measure. This will hOpefully reveal greater information about the relationship being tested and will permit the use of more SOphisticated statistical techniques when necessary in testing the cluster hypotheses in Chapter V. In Table 4-11, we find strong support for the hypothesis developed when innovation is measured by the presence or absence of ever having submitted a Title III proposal. As Specialization increases from a low level to a high level, the percentage of innova- tive districts increases from only 14 per cent to almost 68 per cent. The strength of this association is reflected in the theta of .38, eta of .39 and eta2 of .15. The strong relationship exhibited in Table 4-11 continues in Table 4-12 with the other indicator of innovation. This is the strongest:linear association so far found-- more than 25 per cent of the variance can be accounted for with TABLE 4-ll.--Relation of Degree of Professional Specialization of Bureau and Educational Innovation (Ever Submitted Title III PrOposal) Degree of Professional Specialization Educational Innovation Low Average High (Score 1-3) (Score 4-5) (Score 6-7) NO 85.5% 56.4% 32.2% YES 14.5 43.6 67.8 Total 100.0% 100.0% 100.0% N 393 + 78 + 37 = 508 2 O = .38 n = .39 n = .15 107 TABLE 4-12.--Re1ation of Degree of Professional Specialization of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Degree of Professional Specialization Educational Innovation Low Average High (Score 1-3) (Score 4-5) (Score 6-7) 0 85.5% 56.4% 32.2% 1 10.7 19.2 10.8 2 or More 3.8 24.4 57.0 Total 100.0% 100.0% 100.0% N 393 + 78 + 37 = 508 e = 70 r = 51 r2 = 27 knowledge of the degree of professional Specialization in a bureau. This hypothesis receives strong support from the data in either table--more professionalized districts are more educationally innovative. As with the other hypotheses, controls were introduced for both measures of innovation. The original relationships are unaltered by the introduction of these independent variables (see Appendix V). Hypothesis Six An inverse relationship between the degree of rule formal- ization and level of educational innovation is expected in this hypothesis. The degree of rule formalization in a bureau is reflective of the flexibility or rigidity of the rules of a bureau-- both formal and informal. While rules do not necessarily have a 108 negative influence on bureau behavior, the extent to which rules exist and are made use of is important in discovering the freedom of action existing in a school district. Those districts with a higher degree of rule formalization (Specificity and complexity) give less discretion to their deciSion-makers, thus limiting the bureaucrat's freedom of choice. This reduced latitude suggests a reduced potentiality for introducing innovative solutions. Bureaus with greater rule formal- ization will be less educationally innovative. The measure of rule formalization used here is the Guttman scale score a district receives on a set of formal rules that all districts possess--teacher- administration contractual relationships. This indicator of rule formalization is an ordinal measurement but we are going to treat it in the same manner as professional Specialization. Interval statistics are utilized along with theta. The results of testing rule formalization and the dichotomous measure of innovation do not support this hypothesis. The opposite relationship is found in Table 4-13--bureaus with a high degree of rule formalization are 11 per cent more innovative than districts with a low degree of rule formalization. With innovation measured by the number of proposals submitted, the'relationship is again in the opposite direction to that suggested. AS Shown in Table 4-14, the evidence produced in this test is not overly strong since only 3 per cent of the variance is accounted for with information as to the degree of rule formalization. Controls were introduced with both measures of innovation and no distorter- suppressor variables were discovered (Appendix VI). The data supports 109 TABLE 4-13.--Relation of Degree of Rule Formalization of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Degree of Rule Formalization Educational Innovation Low Average High (Score 5-6) (Score 3-4) (Score 1-2) NO 80.4% 76.7% 68.8% YES 19.6 23.3 31.2 Total 100.0% 100.0% 100.0% N 138 + 223 + 80 = 441* 0 = -.12 n = .12 n2 = .01 *Missing Data. TABLE 4-14.--Relation of Degree of Rule Formalization of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Degree of Rule Formalization Educational Innovation Low Average High (Score 5-6) (Score 3-4) (Score 1-2) 0 80.4% 76.7% 68.8% 1 11.6 13.0 12.5 2 or More 8.0 10.3 18.7 Total 100.0% 100.0% 100.0% N 138 + 223 + 80 = 441* 2 O = -.18 r = -.18 r = .03 *Missing Data. 110 to a limited extent the assertion that there is an association between degree of rule formalization and educational innovation but this relationship is not in the predicted direction. At least with this indicator of rule formalization, educational innovation increases with an increase in the degree of rule formalization instead of decreasing with an increase in rule formalization. Hypothesis Seven The recruitment pattern of a bureau--the seventh dimension of the Opportunity Structure--is hypothesized as being related to educational innovation. Recruitment practices have been found to influence the administrative performance of a bureau. In this hypothesis, the recruitment pattern of the school district is con- sidered important because the source of bureaucrats should be indi- cative of the availability of new information and the credence given to that information. When the source of new bureaucrats at all levels is from outside the bureau, the bureau receives a valuable new input of fresh ideas and information at a reduced cost. If the bureau only recruits new officials at the bottom of the hierarchy and all other promotions are from within, there is less likelihood of getting this information and if it is obtained, the costs will be higher. This, in turn, is reflected in the costs of undertaking innovative activities. It is then suggested that innovation will increase as the proportion of new officials recruited outside the bureau increases. The measure of the recruitment pattern of the bureau selected here is based upon the source of replacement personnel when administrative turnover takes place in a bureau. Their source can be either from inside the bureau 111 when a bureaucrat is promoted-demoted or from outside the bureau itself. The percentage of total recruitment that occurs outside the school district for positions potentially involved in educational innovation is the actual measure of recruitment pattern (0 to 100 per cent outside). Examining the results of Tables 4-15 and 4-16 provides some support for this hypothesis. Districts which perform the greatest percentage of their recruiting outside the bureau are more innovative than those who recruit very little from outside, at least for those positions potentially connected with innovative programs. As is shown with the other hypotheses, the relationship is stronger when the interval measure of innovation is used. Knowledge of the recruitment pattern accounts for 6 per cent of the variance in educational innova- tion. Appendix VII contains the negative results of introducing control variables in both tests of the hypothesis. TABLE 4-lS.—-Relation of Recruitment Pattern of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Recruitment Pattern of Bureau (% Outside) Educational Innovation Low(<33%) Medium(33-67%) High(>67%) NO 87.2% 74.1% 65.9% YES 12.8 25.9 34.1 Total 100.0% 100.0% 100.0% N 179 + 233 + 91 = 503* Y . 2 Percent OutSide = 40% n = .22 n = .05 *Missing Data. 112 TABLE 4-16.--Relation of Recruitment Pattern of Bureau and Educational Innovation (Number of Title III Proposals Submitted) Recruitment Pattern of Bureau (% Outside) Educational I““°Va‘1°" Low(<33%) Medium(33-67%) High(>67%) 0 87.2% 74.1% 65.9% 1 9.5 13.4 13.2 2 or More 3.4 12.5 20.9 Total 100.0% 100 0% 100.0% N 179 + 233 + 91 = 503* r = 24 r2 = 06 *Missing Data. Hypothesis Eight In this last hypothesis relating an individual dimension of the opportunity structure to educational innovation, we postulate that the size of a bureau is directly associated with the amount of educa- tional innovation taking place in that district. Large bureaus are more innovative than small bureaus. Size of bureau is expected to be important to innovation for two principal reasons--need and capability. As the Size of the clientele served increases, the magnitude of the problems associated also increase. This creates a greater need for innovative undertakings to deal with these problems--a fact well appreciated by the educational researchers in their investigations. If large school districts have larger problems, they also have a greater capability to develop innovative responses. They have one major resource which small school districts do not have--a larger 113 staff. Only large districts have the ability to divert personnel to engage in innovative programs and at the same time carry on their normal social functions. Size of bureau is measured here by taking the number of bureaucrats involved in administrative functions only-- instructional and non-administrative assignments are not included in this indicator. The correlation between size of bureau and educational innova- tion is strong. In Table 4-17 where innovation is the dichotomous measure, large districts are clearly more innovative than the small districts--almost five times as innovative. The relationship is even stronger in Table 4-18, where knowledge of size of bureau accounts for 36 per cent of the variance. This is the strongest linear relation- ship found between any of the dimensions of the Opportunity Structure and educational innovation. As in the seven other hypotheses, controls TABLE 4-17.--Relation of Size of Bureau and Educational Innovation (Ever Submitted Title III PrOposal) Size of Bureau Educational Innovation Small(30) NO 87.5% 78.5% 38.6% YES 12.5 21.5 61.4 Total 100.0% 100.0% 100.0% N 313 + 107 + 88 = 508 n = .25 n2 = .06 114 TABLE 4-18.--Relation of Size of Bureau and Educational Innovation (Number of Title III PrOposals Submitted) Size of Bureau Educational Innovation Small(<13) Medium(lS-SO) Large(>30) 0 87.5% 78.5% 38.6% 1 9.9 14.0 17.1 2 or More 2.6 7.5 44.3 Total 100.0% 100.0% 100.0% N 313 + 107 + 88 = 508 r = .60 r2 = 36 were introduced to ascertain if any of the other independent variables were influencing this association. None were found (see Appendix VIII). The hypothesis is strongly supported by the evidence in these two tables. Large districts are very much more innovative than small districts. The overall results of testing each of the individual dimen- sions of the opportunity structure and educational innovation are mixed. The hypotheses relating degree of professional specialization and size of bureau to educational innovation are strongly supported by the data. The strength of the associations are strong and the amount of variance accounted for is considerable. Exhibiting weaker relationships to educational innovation are the level of financial support and the recruitment pattern of the bureau. In each case, the correlation and variance explained is much less than those of size and professional Specialization. Two of the remaining four hypotheses 115 receive some support but not in the anticipated direction. The degree of rule formalization is weakly related to educational innovation but not in the hypothesized manner. The level of innovation increases with increased rule formalization instead of decreasing. The rate of ad- ministrative turnover is not a linear relationships as originally suggested. This dimension exhibits a curvilinear association with innovation--innovation increases with turnover until turnover exceeds 25 per cent. Innovation then begins to decrease and returns to zero when the turnover rate reaches 50 per cent. The last two variables-- age of bureau and growth rate of bureau--mainfest such a low level of correlation with innovation that there is inadequate evidence to con- clude that they are related either positively or negatively to educa- tional innovation. The overall level of innovation varies little with differences in the age of the bureau or the growth rate of the bureau. Evaluation of Findings In our discussion of these results, we will evaluate each dimension of the opportunity structure, the two indicators of innova- tion and suggest possible improvements for further research.' Let's begin by first examining the two measures of educational innovation utilized here. These indicators of educational innovation are based upon the definition of innovation as the introduction to an applied Situation of means and/or ends that are new to that Situation. This indicates that our concern is with the attempt at introducing new means and/or ends rather than any measure of the success of the intro- duction of these means and/or ends. We also pointed out the 116 difficulty in evaluating the "success" of the introduction of the new means and ends. To actually operationalize innovation in school bureaucracies, two possible alternative measures based upon educational activity associated with Title III, ESEA of 1965 were constructed. In both measures, all districts in Michigan were potentially capable of under- taking the innovative activity. The first way of measuring innovation under Title III was simply to determine if a school district had ever submitted a Title III proposal during the seven possible submission periods. This dichotomous indicator was felt to fairly well exclude the element of "gamesmanship/grantsmanship" that the second measure of innovation includes. This other measure--the number of Title III proposals submitted by a school district--does capture better the secondary factors/influences that come with an increased number of prOposal submissions such as the prestige associated with having a grant approved and the recognition that goes to the leadership of a bureau for obtaining the extra resources. The important question was which of the two indicators is the better or, at least, the more useful measure of educational innovation. Clearly, the latter indicator has greater utility. In every hypothesis, the number of Title III pro- posals submitted exhibits a stronger association with the dimension of the opportunity structure. More variance is accounted for when using this measure of innovation. This finding supports the argument that the effects of "gamesmanship/grantsmanship" are a part of the process and/or basis for undertaking innovation as we have defined it here. Whether or not ESEA Title III proposals are "good" or even "adequate" 117 indicators of educational innovation can not be resolved here. They meet the definition of educational innovation used here and are con- sidered an acceptable measure of innovation. Finally, only 116 of the 508 districts have ever submitted a single Title III proposal with only 55 of the 116 submitting two or more proposals. There is no way of telling without comparative state data whether or not Michigan school districts are different from districts in other states in their use of the Title III section of ESEA of 1965. We know of no other measure of innovation available for comparison across districts in Michigan and gathered by the Department of Education. Considering all the factors relating to our measures of innovation, we believe that while these indicators of innovation are imperfect they do capture something of the essential quality of innovation-~especially the interval measure of innovation. The first dimension of the opportunity structure tested is the level of financial support a district receives from the community. While only receiving moderate support (r=.23 and r2=.05), the amount of innovation does increase with an increase in the level of financial support--extra Operating millage voted. This finding differs from that of Gittell and Hollander in Six Urban School Districts where they discover innovation to be unrelated to fiscal questions in the large urban districts studied. This is an interesting contrast with our results where there is some relationship between the level of financial support a district receives and educational innovation. The Detroit public school district, which is included in both studies, is the most innovative district in Michigan with eleven Title III proposals 118 submitted. However, it does not have the highest extra Operating millage rate nor does it have the lowest. While it is difficult to evaluate their findings and our findings about Detroit together, we question the overall applicability of the Gittell and Hollander re- sults due to their limited number of schools and the slight variations in contextual detail. The limited relationship between financial support and innova- tion provides some evidence for Downs' discussion of the influence of organizational slack upon non-programmed actions. Innovation is costly in terms of time, material and personnel which suggests that only those bureaus having a certain level of slack will be able to engage in non- prOgrammed (innovative) activities. When slack is at a low level, bureaus will be unable to undertake innovation. Where slack is greater, more innovation should occur. This seems to be the pattern here. As the level of extra Operating millage increases, the amount of educational innovation also increases. While we did find support for the hypothesized association, a possibly better indicator of financial support might have been achieved by creating an Index of Tax Effort based upon the original millage measure. Extra operating millage does not take into account differences in State Equalized Valuation (SEV) which exist between the school districts.15 Different amounts of resources will be ob— tained with the same extra Operating millage given different SEV'S. An Index of Tax Effort might better capture the support a district has by contrasting how much a district taxes itself against its SEV. Districts which have low SEV'S and vote a low extra Operating millage 119 would have a lower Index of Effort than a district which had the same low SEV and taxed themselves at a higher millage rate, and so on. This Index of Tax Effort would be created by using SEV as the inde- pendent variable and extra operating millage as the dependent variable. Taking standardized scores for their reSpective values (normalized on size of district and/or the area the district encompasses such as urban/suburban/rural), a simple regression analysis would then be performed to obtain the beta weights. Next, the SEV Z-scores would be put into the regression model with their corresponding beta's and the predicted extra Operating millage rates in Z-sceres will be found. The actual Z-score extra operating millage rates minus the predicted Z-score extra operating millage rates would be the Index of Tax Effort (negative values would indicate a high tax effort and vice versa).16 The second hypothesis suggested that age of bureau is related to level of innovation. The results of the tests do not support this preposition. Younger bureaus are not appreciably more innovative than the older bureaus regardless of the measure of age or innovation used. The principal support for the inclusion of age as a dimension in a bureau's Opportunity structure is drawn from Downs' discussion of the dynamic life cycle of a bureau. A number of effects on bureau behavior were expected with increasing age of bureau. The most important here being the reduction in the capability of older bureaus to be innovative. To test this hypothesis, we gathered data on the age of the school districts Studied--the length of time that a bureau 120 had administered its district. The base period was either the year the district was created or the year of its most reorganization. At that point, difficulty arose in the use of age. The Department of Education did not have records of district organization/ reorganization prior to 1935. Most districts were organized prior to this date, 350 of the total 508 districts. We were forced to use this information but it did not permit the real differentiation of school districts on age. When this original indicator of age did not appear related to innovation, we turned to an alternative way of viewing age suggested first by Ripley e£_al: This measure of age is the number of years since the replacement of the head of the bureau by a new administrator. There is a Slight improvement using this indicator over the number of years since a bureau has been in existence/reorganized but not sufficiently large enough to conclude that age is related to innovation or that this new measure is a better indicator of age. In fact, one can question whether or not this is really a measure of bureau age. The number of years since the replacement Of the head of the bureau by a new administrator may very well be related to the administrative turnover rate of the bureau rather than the age of the bureau. Regardless of this point, the results of testing might have been altered if data for the original indicator of age been available for a longer period of time. Perhaps with greater variation in age such as districts of 75 to 100 years old, younger districts might have been found really more innovative. One other element could be included in evaluating the utility of age of bureau beyond the use of a better indicator of age. The 121 effect of a bureau's social function may play an important role when considering the age of a bureau and its influence on bureau behavior. If we accept Downs' proposition that the social function of a bureau influences a bureau's internal structure and behavior and vice versa then it just may be that the argument on the impact of age on a bureau is mitigated in bureaus with different social functions. Bureaus differ in their social function as to the substantive content and structural a5pects of the function. The specific things a bureau does, i.e., providing national defense, collecting taxes, removing snow, educating children, forecasting weather, etc., make up the sub- stantive content of a bureau's social function. The structural aspects of a bureau's social function incorporates more general characteristics such as: clarity with which the functions can be defined, ease by which bureau results can be perceived and effective- ness evaluated, stability of the bureau's internal technological environment and external environment over time, Operational inter- dependence of various functions, scope of functions, and nature of power setting--nature of its institutional surroundings.17 The fact that bureaus do differ greatly in terms of substan- tive content and structural aspects suggests that the age argument may hold with a bureau having a particular combination of substantive content and structural aspects characterizing its social function and not hold for a bureau with a social function having a different combination of elements. Educational bureaucracies certainly have a very different substantive content to their social function than do the regulatory agencies. The effect of age on these bureaus may 122 differ greatly but there is no way of knowing whether or not this is factual Since there are no studies of the influence of age on bureaus where substantive content Of social function is controlled for. There are difficulties however when making comparisons based upon sub- stantive context. There is a great diversity of substantive functions, so great as to possibly preclude devising categories of functions narrow enough to be usable. More importantly is the difficulty arising from the fact that bureaus are often assigned combinations of sub- stantive contents that logically fall into different categories, i.e., DOD which combines national defense with education of children and adults. Finally, the process of dividing substantive functions into categories suggests that functions have unique characteristics unto themselves which tend to resemble the structural aspects to some extent. Thus, we believe that if the argument does not held across bureaus with different social functions it is because of the variations in structural aspects found not substantive content. Educational bureaucracies are clearly different from other types of bureaus in terms of these structural aspects. For example, there is a great difference between the clarity of definition of function of a fire department and that of a school district. Variation in these struc- tural aSpectS are likely to be the elements which influence the effect of age on a bureau and its ability to be innovative. Before rejecting age as a dimension in the Opportunity structure of a bureau further investigation needs to be made into the effect of social function (structural aspects) on the life cycle of a bureau (the influence of age on the behavior of a bureau). 123 The growth rate of a bureau was posited as being directly related to the amount Of innovation in a bureau. An examination of the data fails to support the supposition that the faster the growth rate of a bureau, the higher the level of educational innovation in that district. In fact, the little correlation that does exist between growth rate and innovation is negative (-.10). As in the age hypo- thesis, the major support for the suggested relationship comes from Downs' model of a bureau's life cycle where he argues that any rapidly growing bureau is more innovative than a slower or a non-growing bureau (either in terms of absolute growth or growth relative to that taking place in other bureaus with a Similar social function since there is an inherent tendency for any bureau to grow, however, slowly).18 There is no distinction made by Downs about differential rates of innovation for rapidly growing bureaus having different social functions--i.e., variations in structural aspects of the bureau. Bureaus which grow rapidly are more innovative regardless of the factors that can influence the reasons for the growth taking place in those bureaus. While the results of our test are inconclusive, it appears that the school districts with no growth/declining enrollments are slightly more innovative than the rapidly growing bureaus. The question then is: "Why don't school bureaus conform to Downs' model?" A plausible explanation may rest on our previous discussion of the influence of social function on age in the life cycle of bureaus. We suggested that the effect of age on a bureau's behavior may be modified by the bureau's social function--its particular set of structural aspects. 124 This may also be true for the influence of growth on the behavior of school districts eSpecially when considering a behavior like innova- tion. Innovative activity, as we have defined it, presumes that it is possible to introduce to an applied Situation means and/or ends that are new to that situation. Innovation is then very much related to at least one of a bureau's structural aspects-~the clarity with which the function/functions of the bureau can be defined. Bureaus that have very explicit, well-defined and agreed upon functions have a greater potentiality for having innovation introduced into them than do bureaus where the social function is abstruse, ill-defined and lacks consensus. The variation of clarity of function can easily be seen by contrasting the functions of NASA and that of a local school district. One of NASA'S functions was to land a man on the moon and safely return him to earth in a certain time frame. This function has the clarity that lends itself to innovation. The function of education being performed by a local school bureau has never been clearly defined or agreed upon to date. If one would ask the political sovereign of a school bureau to define what education is, one would very likely receive as many definitions as there are members of that school board. Education as a social function lacks the clarity or precision which is likely an essential component in being able to be innovative in all situations. The ability or inability to clearly define a bureau's social function may then effect how bureau's resend in periods of rapid growth. Bureaus with explicit functions may find it easier to adept innovation in periods of very rapid growth as did NASA when it utilized extremely innovative technolOgy to perform its social 125 function. When educational bureaus with their rather ill-defined functions are faced with very rapid growth, they seem to respond differently based upon our results. The rapidly growing bureaus appear to use the old solutions in order to just try and meet the needs of the clientele during the period of rapid growth. Periods of stagnant/declining growth may present the bureau with an Oppor- tunity to explore more fully exactly what their function is (they may be forced to do this by other actors in their external environment). Having better defined or at least attempted to define what their social function is, the problems they are facing should be clearer. This, in turn, Should increase the possibility of the bureau engaging in some innovation in trying to resolve the problems the bureau faces. The only real way of evaluating this possibility is to per- form comparative research with bureaus having different structural aspects and experiencing different levels of growth. As discussed in Chapter III, change in Student enrollment was selected as the indicator of the growth rate of a bureau since it was felt this would more quickly reflect actual growth or change. Change in bureau employment was seen as being less responsive to fluctuations in the bureau's environment arising out of either posi- tive or negative growth. While there is almost perfect correlation ‘between total bureau employment and student enrollment, it may be that change in the administrative Staff in innovative activities is masked by the use of change in student enrollment. An alternative measure based upon growth in administrative employees might capture periods of very rapid growth in a bureau better than our present 126 indicator. However, there remains the problem of reflecting no change or negative change in the administrative staff. Only further research can resolve this empirical question. The relationship that exists between the administrative turnover rate of a bureau and educational innovation is not the one prOposed. It was originally suggested that the amount of innovation would increase directly as the administrative turnover rate increased-- a linear association. The data supports the assertion that there is a relationship between these two variables but it is curvilinear not linear. A curvilinear relation does not conform to Downs' development of the influence of turnover in a bureau nor to Bidwell's report that mobile administrators are more innovative but neither does it completely negate their basic arguments.19 Bidwell points out that research into the mobility of administrators shows mobile administra- tors to be more innovative than those who were not mobile. He also acknowledges that other research shows that mobile administrators are not all that much more innovative than the nonmobile. Without the original data of the respective studies, it is difficult to tell but the latter finding may be somewhat supportive of our results. If the original research was not carefully conducted, it may have been possible that the most mobile were included with the somewhat less mobile administrators. This could have reduced the overall level of innovativeness when contrasted with the least mobile officials if a curvilinear relationship existed with the most mobile administrators being non-innovative. For example, if the original data in our test is collapsed into dichotomous categories of administrative turnover, 127 Extreme mobility reflected in the turnover rate of a district is a real depressant upon the amount of innovation taking place in the district (see Table 4—19). TABLE 4-19.--Relation of Administrative Turnover Rate of Bureau and Educational Innovation (Ever Submitted Title III Proposal) Administrative Turnover Rate Educational Innovation Low(<.22) High(>.22) NO 75.1% 79.2% YES 24.9 20.8 Total 100.0% 100.0% N 265 + 240 = 505* *Missing Data. Another factor to be considered in analyzing these results in light of Bidwell's discussion is the nature of the measure of turnover in school districts. The indicator of administrative turnover rate reflects the change in personnel serving in a particular district--the change per administrative position per year for that district. It does not directly reflect the mobility of those who fill the positions. There can be situations where a high rate of turnover exists while the individuals who actually fill the positions are not personally mobile. Many possible combinations may exist. The indirect indi- cator used may not then adequately describe the effect that mobility has on behavior in reSpect to innovation. 128 Downs' argument rests on the influence mobility has upon the chances an administrator has in achieving his personal goals. Where there is considerable turnover in a bureau (reflective of high mobility), Downs expects to find an increase in the likelihood of innovation taking place because of the new ideas, information, perSpec- tives and need for success (goal attainment) that these individuals bring with them. This need for success could very well be the factor that causes a decrease in innovation in bureaus with high levels of mobility. These mobile bureaucrats realize that it takes a great deal of time and effort on their part to introduce innovation in their bureau. If they are not going to be there to reap the rewards which go with this activity, what is the point of these officials expending that extra effort? Therefore in bureaus with very high rates of turnover, lower levels of innovation may be found because of the decreased potentiality of gaining the rewards that go with innovative activity. At the same time, a moderate amount of turnover has a posi- tive effect on innovation because it stimulates the bureaucrat to be innovative Since he realizes there is no certainty that he will be promoted from this turnover or even continue to be employed without such efforts. With little or no turnover, the bureaucrat may feel little or no pressure to be innovative since he likely is obtaining his goals in other ways. The results of the fifth hypothesis clearly support the assertion that the greater the degree of professional Specialization in a district, the greater the educational innovation taking place in that district. This dimension of the Opportunity structure exhibits 129 the second strongest linear association to innovation of all of the individual dimensions (r=.51 and r2 =.27). Professional Specialization is not the negative influence on innovation suggested by Rogers when he argues that professional specialization contributes to the "patholOgies" of the New York City School system and its difficulties in being innovative. While we can not tell directly, it seems likely that professional specialization is an important factor in the overall cost of information in the innovative process in a district. Districts which are highly Specialized in their administrative staff do have greater information readily available to them about the broad range of activities they perform in carrying out the bureau's social function. This greater level of knowledge should make it easier and cheaper to actually undertake innovation as measured here. The only reservation one might have with the indicator of professional specialization is that it is based upon the assumption that all innovative proposals are generated by the administrative staff. The instructional staff is assumed to play little or no role in this activity. This may or may not be true in fact. It might be possible to improve on this measure by including the degree of professionalization of the instructional staff as well as that of the administrative staff. This new index might prove more adequate in capturing the overall professional specialization in a district. Whether or not it would exhibit a stronger relationship to innovation is an empirical question for future research. The degree of rule formalization was expected to adversely affect the level of innovation in a district--the greater the degree 130 of rule formalization (Specificity and complexity), the less the educational innovation taking place in a district. The results of this test do not support this hypothesis. While it is not a strong association (r=.l8 and r2 =.03), the degree of rule formalization is in the Opposite direction to that originally suggested. AS the rules become more formalized, the amount of innovation increases. This result directly contravenes the original arguments made in this hypothesis. Downs suggests that rules influence the amount of dis- cretion available to the bureaucrat. Greater rule formalization leaves less discretionary power available to the bureaucrat. This reduction limits their freedom of choice which, in turn, should inhibit them in introducing innovative solutions to problems in the bureau. A cOunter argument can be made here to Downs. The fact that there are extensive and formalized rule systems existing in a bureau may mean greater freedom to the bureaucratic officials in that bureau. Those most involved in the decision—making process no longer would have to involve themselves with the day-to-day, routine decision-making because their subordinates could turn to the rules for their guidance in answering questions they might have. The top-level officials would then be free to undertake the important non-programmed activities that require innovative solutions. Thus, districts where there is a high degree of rule formalization might be more innovative than other districts with a lower degree of formalization. Our results would tend to support this alternative argument. Another factor influencing the results may be the measure of rule formalization developed. This indicator is based upon teacher 131 contract provisions which supposedly relates the freedom of action of administrators. We know from Anderson's research that rules appear to mediate authority conflicts thus making hierarchical authority more tolerable to the teachers. This same phenomenon may be going on for the administrators. These teacher contract provisions may also mediate authority conflicts for the administrators--less direct conflict with teachers. This, in turn, may increase the adminis- trator's freedom to involve himself in other problem areas which require innovative solutions. Our present indicator captures some- thing of the quality of this relationship. Districts having these contract provisions score high on rule formalization. These more formalized districts are somewhat more innovative which may be re- flective of the greater potential freedom from hierarchical authority conflict. The only real way to resolve the difference between our finding and the arguments being made by others is to obtain a better measure of rule formalization which incorporates all the rules of a bureau--beth written and unwritten. The indicator in use here may just be too restrictive to really measure the degree of rule formal- ization in a bureau. The seventh dimension of the opportunity structure--recruit- ment pattern of the bureau--receives moderate support in relation to educational innovation (r=.24 and r2=.06). The more often a district recruits new officials from outside the bureau itself, the greater the educational innovation taking place in that district. This result supports the findings of most researchers that the source of new officials is related to administrative behavior. Just as Rogers 132 suggests, districts which are "in-bred"--recruit primarily from inside--are less innovative. A difference does exist between the Gittell and Hollander research and this investigation. They conclude that district output (innovation) is unrelated to the recruitment source of district superintendents. Their conclusion is not supported here even when using an earlier indicator of recruitment pattern which, like their own, was based solely upon the source of new superintendents for a district. The present indicator provides an even better basis for rejecting their conclusion Since it incorporates most of the officials who are potentially involved in innovative activities. We continue to differ as to the Detroit system. Detroit, the most innovative Michigan district, fills 75 per cent of its vacancies from outside the bureau itself. Bidwell continues to re- ceive support in this research. Superintendents from outside the bureau do act differently from those recruited from inside the bureau. The overall utility of this dimension may be somewhat limited Since the present indicator reflects the source of those most intimately involved in innovation and yet accounts for only 6 per cent of the variance. This variable may be more important when dealing with types of administrative behavior other than innovation. Size of the bureau diSplays the strongest relationship with educational innovation of all the dimensions of the opportunity structure (r=.60 and r2=.36). AS the size of the school bureau increases, the amount of educational innovation also increases. Size of bureau was suggested by all the researchers as being an 133 important variable but only Gittell and Hollander attempt to relate size of bureau and innovation. They find them unrelated. From the evidence here, we disagree with this conclusion. Size of bureau is clearly an important variable in studying innovation. Detroit is the largest district studied here and also the most innovative in the state. When studied by Gittell and Hollander, it was found to be non-innovative. It is difficult to accept their overall conclusion when we find size to be related to innovation in one of their own cities. It is even harder to reject size based on their research because of the lack of variability in size in their cities. The evidence found here strongly supports the argument that size of a bureau is related to innovation. This may be due to two principal factors-—need and capability of large bureaucracies. Large bureaus are faced with larger student enrollments (r=.99) which reflect the magnitude of the problems these districts must deal with in per- forming their social functions or the greater need for solutions to problems because of the sheer numbers of students affected. In addition to the serious problems faced and the needs of the students involved, the large district has the greater capacity to develop innovative solutions. An important factor in the cost of innovation is the expenditure of personnel resources to carry out the innova- tive activity. Only large bureaucracies have sufficient personnel on the staff to divert a portion of them to undertake and carry out the innovative activity and at the same time perform their normal day-to-day education function. The indicator of size of bureau is the number of personnel involved in administrative functions full-time. Those engaged in 134 instructional and non-administrative assignments are excluded from size of bureau. This figure is the best of the different possible ways of determining size--non-instructional staff only, total Staff including instructional and non-administrative staff, instructional staff only or non-administrative staff. Other types of organizational research may call for the use of another measure of size but this measure proves to be the most helpful in examining size and innovation. In this chapter, we have examined the results of the tests of each individual dimension of the opportunity structure, discussed some of the implications of the findings and offered possible improve- ments to be incorporated in future research. As one might realis- tically expect in an exploratory study, not all the hypotheses received strong support or even any support from the data. The next chapter will evaluate the remaining two hypotheses dealing with the dimension clusters and educational innovation. FOOTNOTES--CHAPTER IV 1Denton E. Morrison and Roman E. Henkel, eds., The Si ificance Test Controversy (Chicago: Aldine Publishing Company, 1970), p. ix. 2Discussions of significance tests are readily available in most basic statistical text books. A very understandable introduction to Significance tests is contained in Linton C. Freeman, Elementary_ Applied Statistics (New York: John Wiley 8 Sons, Inc., 1965), pp. 142-61. Other sources containing discussions of significance tests include Hubert M. Blalock, Social Statistics (New York: McGraw Hill Book Company, 1960); Jerome L. Myers, Fundamentals of Experimental Desi (Boston: Allyn and Bacon, 1966); William MenHEnhall, Introduc- tion to Probability and Statistics (Belmont, Calif.: Wadswort PubliShing Co., Inc., 1967); and Morrison and Henkel, op. cit. 3Leslie Kish, ”Some Statistical Problems in Research Design," in The Si ificance Test Controversy, ed. by Denton E. Morrison and Roman E. Henkel (Chicago: Aldine Publishing Company, 1970), p. 139. 4Freeman, Elementary Applied Statistics. The various tests of association presented and the levels of variance accounted for in each of the tables are based upon the un- collapsed data relating the dimension and innovation. n (eta) is a measure of association and n2 is the amount of variance explained. They are discussed in Freeman, pp. 120-130. The magnitude of n and n are comparable to Pearson's r and r2. The value of n is from O to +1.0. 2 6For a discussion of the introduction of a control variable into a two-variable relationship and the interpretation of the results, see Morris Rosenberg, The Legic of Survey Analygis (New York: Basic Books, Inc., 1968). See also Hubert M. Blalock, "Correlated Independent Variables: The Problem of Multicollinearity," Social Forces, XLII (December, 1963), 233-37, for a discussion of multicollinearity and how this problem Should be treated among independent variables. The levels of multicollinearity in the table below reflects the values between the 135 136 eight dimensions of the Opportunity structure. Where the correlation is below .60, Blalock argues that the multicollinearity be ignored and the variables be treated as independent. Dimension Ago. Growth Turnover Size Profess. Rule Recruit. Finan. Age of Bur. 1.000 - .114 .079 .069 .133 .005 .098 .106 Growth Rate 1.000 .024 - .084 - .105 .000 - .053 - .129 Admin. 1 Turn. Rate 1.000 - .021 - .105 - .008 .226 - .043 Size ' of Bur. 1.000 .325 - .179 - .024 .090 Prof. Spec. 1.000 - .210 .040 .236 Rule Formal. 1.000 .014 - .126 Recruit. Pattern 1.000 .075 Finan. Support 1.000 7A practical problem arose in determining the actual age of the school districts. The information available from the Department of Education does not extend back before the 1935-36 school year. We were forced to use this school year as the base date of organization for all districts which were not created or reorganized since then. This may be a major contributing factor in the lack of variability in innovation with differences in age of bureau. 8Randell B. Ripley, William B. Moreland and Richard E. Sinnreich, "Policy-Making: A Conceptual Scheme," American Politics, I (January, 1973), p. 39. Their primary approach to age is similar to Downs but they do suggest this as an alternative way of conceiving an agency's age. 9The basis for approaching the problem of curvilinearity in this manner was drawn from S. Sidney Ulmer's paper on Supreme Court Justice Black's voting behavior from 1937-1967. Ulmer was faced with the problem of ascertaining whether or not Black's support of civil liberty claims was linear over the 31 terms studied. The process which he used in resolving the issue is borrowed in this effort at determining the exact nature of the relationship between administrative turnover rate and innovation. See 8. Sidney Ulmer, "Black, Longitudinal Behavior," a paper presented to the American Political Science Association in Chicago, 1971. 137 10This is the one hypothesis in which statistical tests are performed on collapsed data. The data was collapsed because it demonstrates better the shape of the relationship as shown in Figure 4-1. With this collapsed data comes stronger associations so no comparisons will be drawn with the tests of the other hypotheses. 11For a discussion of those who argue this point, see Bela 0. Baker et al., "Weak Measurements vs. Strong Statistics," Education and Psychological Measurement (Summer, 1966), 291-309. 12William Hays, Statistics for Psychologists (New York: Holt, Rinehart, and Winston, 1963), p. 74. 13Hubert Blalock, Causal Inference in Non-experimental Research (Chapel Hill: University of Norih Carolina Press, 1964), p. 94. 14Theta is the coefficient of differentiation which is an extension of the Wilcoxon signed-ranks test used in describing association between a nominal and ordinal scale. See Freeman, Elementary Applied Statistics. 15State Equalized Valuation is the taxable value of real estate (assessed value as finally equalized by the State Tax Commission) in a school district. The SEV is the value upon which the extra operating millage is applied to produce revenues for the school district. 16The Index of Tax Effort is patterned after an Index of Success created to measure relative achievement among state executives by Fred W. Grupp, University of Connecticut as a part of a larger study of state executives being conducted by Grupp and Alan Richards, Louisiana State University. 17Downs, pp. 42-44. 18Anderson found in his study of rule formalization that as the student enrollment increased, the professional specialization also increased. Our results differ from his. Rapidly growing bureaus are the least professionalized while stagnant/declining bureaus are the most professionalized. 138 Growth Rate of Bureau Professional Stagnant/ . T 4- . . Average High Specialization Declining (<+I.O) (1.0-4.0) (>4.0) Low 67.6% 81.4% 80.3% Average 18.0 14.2 14.8 High 14.4 4.4 4.9 Total 100.0% 100.0% 100.0% N 139 + 247 + 122 = 508 19While Downs and Bidwell are the principal sources of support for the use of this indicator of mobility, Greg and Knezevich and Miner find mobility associated with other factors of a bureau. Miner suggests administrators in large and small districts are less mobile than those in medium Size districts. Greg and Knezevich argue that administrators from the large districts are the most mobile of all the administrators. The data from this investigation basically supports Greg and Knezevich. Large districts do have the highest rate of administrative turnover (high and medium turnover). Large and small districts have higher total mobility (high and medium turnover rate) than the medium sized cities which is counter to Miner's suggestion. Size of Bureau Admin. Turnover Rate Small (<13) Medium (13-30)_ High (>30) Low (<.lO) 8.1% 12.8% 5.3% Medium (.10-.30) 63.8 68.9 86.2 High (>.30) 28.1 18.3 8.5 Total 100.0% 100.0% 100.0% N 160 + 251 + 94 = 505* *Missing Data. CHAPTER V TESTS OF THE DIMENSION CLUSTERS OF THE OPPORTUNITY STRUCTURE The hypotheses being tested in this chapter relate the dimen- sion clusters (career and resource) of a bureau's Opportunity structure to educational innovation. The basis of the anticipated relationships of the dimension clusters to educational innovation rests upon the assumption that the career cluster gives direction to a bureaucrat's goals and the resource cluster suggests the means or strategies the bureaucrat adopts to maximize those goals. The eight combinations of the career cluster are hypothesized as being associated with the levels of educational innovation given. The second hypothesis states that the level of innovation is not the same for all combinations of the re- source cluster found. A subordinate hypothesis gives the level of innovation expected for the polar combinations of the resource cluster. These hypotheses are tested in a manner similar to the testing of the individual dimension hypotheses in Chapter IV--a brief review of the development of the hypothesis is made and this is followed by the results of the actual tests. After testing these hypotheses, an evaluation of the findings is made. 139 140 As with the individual dimensions of the opportunity struc- ture, we use both measures of educational innovation--presence or absence of ever having submitted a Title III proposal and the number of proposals submitted by a district. The primary method of analysis in this chapter continues to be the contingency table for the dichotomous indicator of innovation. The contingency coefficients reported are for information only.1 When using the interval measure of innovation, gamma, multiple correlation coefficients and Stepwise multiple re- gression analysis are determined where applicable in addition to pre- senting the appropriate contingency tables.2 The multiple correlation coefficient demonstrates the strength of the association of the independent variables acting together to the dependent variable. Stepwise multiple regression analysis provides information about the relative importance of each of the dimensions making up the resource cluster and the overall relationship of the resource cluster to innovation. The first hypothesis tested asserts that certain combinations of the dimensions of the career cluster are correlated with higher levels of innovation than are other career cluster combinations. The second major hypothesis proposes that the level of innovation will not be the same for all combinations of the dimensions of the resource cluster. These hypotheses are tested in manner Similar to the testing of the individual dimensions in Chapter IV--a brief review of the development of the hypothesis followed by the results of the actual tests. After testing the cluster hypotheses, an evaluation of the findings is made. 141 Career Cluster Hypothesis The dimension cluster in this hypothesis is identified as a "career" cluster because of the use that Downs makes of the individual dimensions of age, growth rate and turnover rate in his discussion of the life cycle of a bureau. Each dimension is seen by Downs as being important in the determination of which type of bureaucrat will be present in a particular bureau and which type will likely dominate that bureau, i.e., more conservers are expected in older bureaus, more climbers are to be found in rapidly growing bureaus or ones with high personnel turnover. Downs also theoretically integrates these indi- vidual dimensions in proposing which type bureaucrat should be feund in a particular context. Younger, rapidly growing bureaus with high administrative turnover should have an increasing proportion of climbers and a decreasing prOportion of conservers. In a bureau with opposite characteristics, the reverse situation would be expected--an increasing percentage of conservers and a decreasing percentage of climbers. From an awareness of the type of dominant official, Downs argues that one can determine the types of goals these bureaucrats seek to maximize--climbers wish more power, income and prestige while conservers desire convenience and security. These preferences would then be reflected in the type of career a bureaucrat has--his bureaucratic behavior in achieving these goals. The predicted level of innovation fer the career cluster combinations of this hypothesis are indicated in Table 5-1. The level of innovation proposed there is consistent with the original hypotheses developed for each of the individual dimensions and Downs' 142 .uoohpsm many mo :onmaomHv ouoHQEOO a you bated .mm .mczoo oow .guzoum afiumuoseoHSD mo mofiemcxv one we mwmxamce was cw poouuoo ma meson m“ usooo ow xaoxfia we: ma ya .HHm on ma .xfionm mcflzohm m“ was» xoewosaousn so: a o>es ou oHnfimmom xHHeowmoH ma ufi onnzr awesome fluoroSU “Rescue “dreamed fluorosv fleoemuxu fieoemumv fluoememv H6>63 cowum>oqca armor zm>mm me m>Ha «sod mmmre ore mzo nopesz encoded 363 can: 363 emu: 363 rum; 364 emu: mm>ozm=e rm>oz¢=e mm>ozm=e mm>ozm=e madcafiooo madcaaooo a. . . a. . . \ucmcwmum\ommuo>< cm“: \pnacmmum\ommno>< cm“: maoecH aocofiumosum mo m~o>oq eoumMOOmm< ecu xumuusmousm Hoocom mo ousuushpm xuficauAOQQO on» me maneuumm Houmsau uooneu--.~-m mam

om me o>Hm used couch 03H Ono Hones: :uouuam no. u 0 OH. u o me. u u mo. u 0 «com u cum + on + em 4 em + mm + m + Na + N z ao.oo~ ao.ooH ao.ooH wo.oo~ *o.ooH wo.oo~ ao.oo~ ao.oo~ Hooch n.m~ H.n~ N.o~ m.w «.ma m.mn n.He o.o . mm» am.en am.mm mm.mn an.~m ww.Hm $5.00 «n.mm mo.co~ oz 8... .8 33 8m .a 23: 8m .8 :3 8m .313: 8n .3 :3 8... .3 reel 8n .8 :3 8m .0 e3: e3. 226:5 Maccaudusvm ovum .chsh .:«Ee< comm .cwse .cfiav< oumm .cush .cwav< ovum .nush .cwao<. “wo.va acacwauoo fiwo.evu unwcwaooo \ucecmmum\omeuo>< AwO.VAV an“: \ucmcmeum\owauo>< nxo.eAu nu“: awoken mo comm guzouu awaken mo oumm :pzouu mmuao> MHAV noefio Amado» mHvV Humane» season me out mammOQOHS HHH OHHHH vouufienzm uo>mv :ofium>o::H ”anewpousvm one amousm we mahouumm Roomsau uoouou mo cofiuamomuu.~um mamom xvm o>wm usom oouzh 039 one Honasz :uovvom .eom u OAN 4 on 4 er 4 em 4 mm 4 n 4 NH 4 N z ro.oo~ oo.oo~ so.oo~ so.oo~ ro.ooH ro.oo~ ro.oo~ ao.oo~ Hedda 4.MH A.m H.a m.m H.m n.nn 4.» o.o one: to N m.NH 6.HH u.mu 0.0 ~.m o.o ”.mn o.o a rm.4a ra.~w rm.da 4K.Hm vw.um «5.66 vm.rm so.ooa o fiom.vvros flom.xmcmw: nom.erOS flow.xmeer “on.eroS mom.xqmmw= “on.vvroa non Mucus: eomue>oeeu Hocovvooavm ovom .chsh .cweem ovum .cvSH .cflso< ovom .cuoh .cflad< ovom .cuzh .cwav< “so.evV ueueuadoa floo.vvv meaeuuuoo \vcocmovw\omovo>< mwo.ekv cm“: \vcocuovm\omovo>< Avo.VAV mum: omoHSm mo ovmm cvzovu soonsm mo ovum szouu Amend» ALAS edeuo Amado» mHvV traced» sooHOm mo om< flvovvwansm mammomoud HHH ovab mo venezzv cavvo>oacH HocoHvooaom ecu zoovsm mo mcuovvom uovmnau voonou mo cemvoaomuu.mum mqm<9 147 .uvea «mammal. N0. 4 N0 00. 4 0 00000002 00. 4 0 00. 4 0 00. 4 0 00. . 0 00.- 4 0 00. 4 0 .000 n 00 4 00 4 00 4 00 4 000 4 N0 4 0 4 00 4 0N 4 N 4 0 4 0 4 0 4 00 4 0 4 0 4 0 4 N 2 00.000 00.000 00.000 00.000 00.000 00.000 00.000 00.000 00.000 00.000 00.000 00.0 00.000 00.000 00.000 00.000 00.000 00.000 00000 N.00 0.00 0.00 0.0 0.N0 0.0 0.N0 0.0 0.0 0.00 0.00 0.0 0.0 N.0 0.00 0.0 0.00 0.0 606: 00 N 0.0 0.00 0.00 0.00 0.00 0.0 0.N0 N.00 0.0 0.0 0.0 0.0 0.0 0.0N .0.0 0.00 N.NN 0.0 0 00.N0 00.00 0N.N0 00.00 0N.00 00.00 00.00 00.00 00.00 00.00 00.00 00.0 00.000 00.00 00.00 00.00 00.00 00.000 0 .>o::0 .0000 000. 00 000. -00..0 000. 00 000. VV 000. -0110.000. 40 000. 00 000. -00..0 000. 40 0001w0 000. -0040 000. 00 000. vV 000. -00. 0.000. 0 000.00 000.40040 000.00 :00 00000: 000: 060 00006: 000: 80 00000: 000: 060 000062 000: 060 00006: 000: . :60 00000: 000: 0000 .0000 .0000< 0000 .0000 .0000< 0000 .0000 .00000 0000 .0000 .00000 0000 .0000 .e0ae< 0000 .0000 e0ae< 000.0v0 . - . . 000.0v0 . - . . 000000000000000000 000 0 0 00000uo>< 000 0400000 000000660000000000 000 0 0 00ouauo>< 000 040000: aeovnm we ovum zvxovo Amuse» MHAU hovao soovsm we ou< uneven mo ovum nvxouo anvao> n0vm humane» nvovvfiaonm nannoaoum 000 o0v0h mo woneszv newvo>occn Hoseavmosvm one awoken we nov0300 uoouau 04v we mcofincoava mo newvoaom--.vum m4no::H 304 Hmcoflumosvm ZOHHmv :ofipm>oc=H Sawhfim mo mCOMHmv—MDEOU HOHMSHU OUHSOmmm mo ode—aw MO GOMHQHOKII.OIm m4m<9 153 *o.ooH 0.0 0.0 o.o 0.0 o.o *o.ooH o.o 9.0 cm“: same»: so; ZOHH0 fifl H .mwmm agave: .umm Hmeowuaosem onHoccH gun: semen: :04 can: agave: :04 gm“: sauce: so; Annapuaosem eaoaaam 4< 304 onHmv sawuw>oacH Hacowuaunvm van saoazm mo noumsau condoned mo mcoflmcoawo uuoamzw Hmwocmcwm cad coHuwuwaufloomw Hmcofimmomoum .oufim mo cowuaflom--.oaum m4meeen can: eonoe: :eo non: eenoez :eo zonz gonzo: sea decenneueom non: Pmommaw 4< hmommam Aocnm Hmcowpauzfim we: :aousa mo noumsflu oun50mom mo mcofimcoano uuomnsm Hawucdcnm can cowuuuwauwuomm aacoflmmomonm .ounm mo :owuamomu-.unum m4m< 300 m + 0N + mm + 0 + NH + How $0.00H w0.00H $0.00H 0.0 $0.00H $0.00H 0.0V 0.5N 0.5H 0.0 0.0” n.NH $0.00 wv.mn wm.~0 .0.0 wv.m0 $5.50 nun: ommno>< 300 an“: ommuo>< 300 ZOHHocca Mononumuavm mammomonm HHH ofinfih wouuwansm uo>m0 canuo>occu Hmcowpmusvm one summon mo noumsflu condemom mo mcowmcoeflo cowumunflofioomm Hmcofimmomonm one oNflm mo cowumaomuu.maum mamoczm Hoconuoosum one noonsm mo noumsHu condemom mo mcowmcosflo cowpmNnHofluoam Hmcowmmowoumnxmwoufim mo coflumaomuu.nflum mgmnomcou con» .mczoo xn poswno on we denuncwmoo x0 muonsfiao x0 vouonflaov no: one mseonsn so: Ham mHH mmm>mmmzou u mm mammzmqo n << “opoz mm mm mm mm << mm Hmm << Hmm << . << << 304 onenopoz swam 300 oponoooz an“: 300 oumuovoz sun: :04 ouonopo: awn: am>OZMDH m>HHozmah m>HHoz¢=H m>HH02¢3h m>HHn0mcou\nonaflfio one onsnosunw xnncsnnoamo m.:monsm mo mcoflmaoafio Hangoucu-u.ano mqm<9 188 jump into that bureau from another bureau. When these two favorable conditions exist in a new bureau, then climbers are clearly expected to dominate. When only two of the three conditions exist together-- new bureau, high growth rate or moderate administrative turnover-- climbers are still expected to dominate since they continue to have a fairly reasonable set of conditions to maximize their goals.12 When the opposite conditions are present (old bureaus, low growth rate and high or low administrative turnover), conservers will dominate since these are bureaus where conservers can maintain the status quo. Conservers will be found more often in older bureaus because bureaucratic preferences seem to be at least partially a function of the age of bureaucrat (all bureaucrats have a tendency to become conservers with increased age) and older bureaus are made up of older bureaucrats unless rapid growth has overcome the age-lump phenomenon in a bureau.13 A low growth rate is also indicative of the presence of conservers because the opportunity for promotion through growth is reduced thus forcing climbers to either jump out of the bureau or change their preferences to conform to the conditions he faces (only in the long run). When turnover is low, security is provided the conserver because he is maintaining his same position and no one new is coming into the bureau or moving up in the bureau to upset the status quo. If turnover is high, this is reflective of climbers jumping out of the bureau into other bureaus leaving an in- . creasing prOportion of conservers who do not engage in this activity. Conservers will dominate older bureaus with low growth, high/low turnover or combinations of two of these three conditions. 189 It is simply not enough to know under what conditions (dimen- sions of the opportunity structure) climbers or conservers will dominate and the goals the officials seek to maximize but one must also know something about the strategies these dominant officials select to maximize their respective goals. Or at least, one must have further knowledge of a bureau's opportunity structure to suggest which dimensions they take into consideration in formulating these strate- gies. The actual formalization of a complete set of strategies at this point is a most difficult if not impossible assignment. Our knowledge in the area of strategy selection is most limited and there is little to turn to in other research to guide us. Neither Downs nor Schlesinger really get involved in defining explicit strategies for their rational actors. What we will attempt to do then is to sug- gest that certain dimensions of the opportunity structure are poten- tially important to strategy selection (set of possible actions) and discuss one possible strategy/activity that bureaucrats might engage in to obtain their goals. Although a large number of dimensions of a bureau's Oppor- tunity structure may be related the strategy selection process of bureaucrats, we have limited these dimensions to just five dimensions of a bureau's environment. Four of them describe a bureau's internal structure (size of bureau, degree of professional Specialization, degree of rule formalization and recruitment pattern of bureau) and the fifth (level of financial support) is a part of the bureau's external environment.14 The importance of these dimensions is in how climbers/conservers reSpond to these dimensions in maximizing their 190 goals--i.e., what sort of a strategy will a climber or conserver adopt in a large, professionally Specialized bureau with very formalized rules that recruits primarily from outside and has a high level of financial support in order to Obtain more power, income and prestige or security and convenience. Downs would argue that climbers and conservers would select different strategies because they are different types of bureaucrat. As he points out in a number of dif- ferent contexts, climbers should be innovative in problem solving in a particular situation while conservers would not be innovative because of the risks involved to his personal status. This is not necessarily a correct evaluation of the possibilities. In fact, given the same set of conditions they might very well adopt the same strategy or plan of action even though their goals are quite different. The opportunity structure may demand of them a similarity of behavior that one might not expect given the differences in preference orderings. To obtain more power, money income and prestige may require the same strategy as maintaining what one already has of these goals.15 If climbers and conservers depend upon other actors (the political sovereign, professional peers, clientele, etc.) to satisfy the goals they seek, then they must, at least, attempt to meet the expectations of these other actors. If the political sovereign provides a high degree of financial support to the bureau, both climbers and conservers are faced with a situation where the sovereign has indicated that he wants a "good" program--something beyond the bare bones minimum. Failure to provide such a program would be very costly in terms of the bureaucrat's personal goals. Thus, even though climbers and 191 conservers have very different goals, they could easily adopt the same strategy--being programmatically innovative for example--to utilize the high level of resources and meet the expectations of the sovereign. The relationship of the five dimensions of the opportunity structure to strategy can best be seen in the context of a particular strategy available to bureaucrats. The strategy to be examined here is the use of innovation by bureaucrats to gain their goals. As already mentioned, Downs argues that only climbers will be innovative. We propose that if the right conditions of size, professional Specialization, rule formalization, recruitment pattern and financial support are present, bureaucrats will be innovative and if these conditions are not present, then neither climbers nOr conservers will be innovative. Innovation is defined as it was previously as the introduction to an applied situation of means and/or ends that are new to that situation. The use of innovation as a strategy or plan of action is interesting since it incorporates an element of change which conservers might be inherently reluctant to invoke but will do so if the opportunity structure requires it to obtain their goals. Size of bureau influences the bureaucrat in deciding whether or not to be innovative in two principal ways. First, size affects the need to be innovative in a bureau and second, the capability of being innovative. The size of a bureau is related to the number of clientele served or the scope of the activity being performed by a bureau. In either case, the magnitude of the problems to be solved and the need for their solutions are reflected, at least 192 indirectly, by the size of the bureau. At the same time that large bureaus also have a greater inherent capability for developing innova- tive solutions to those problems. Innovation is a costly activity since it requires the diversion of scarce resources from the perform- ance of the bureau's designated social function. However, large bureaus have more of one of those essential scarce resources-~personnel to discover and implement innovative activity. Climbers or conservers examine the size of the bureau they dominate as one step in determining whether or not innovative activity is called for on their part. Professional specialization is also important in this deter- mination since it is another resource in the innovative process. Information is another scarce resource which is costly to obtain. Where there is greater professional specialization, there is also greater information available at lower costs about the broad range of activities that make up the bureau's social function. This reduced information cost is an important factor in deciding about innovation. If the bureau is professionally Specialized, this is a positive influence on climber or conserver deliberations to be innovative. If the bureau is not specialized, this dimension is a negative factor in deciding to engage in innovative activity to get one's goals. Rule formalization is believed by Downs to be a negative influence in the innovation process. Conservers are seen as compulsive rule followers which reduces their inclination to be innovative while climbers find extensive rules restrictive on their freedom of choice. Climbers are less likely to be innovative since this activity requires greater latitude in decision-making. However, this is not necessarily 193 the correct evaluation of the impact of an extensive fOrmal rule system. Rules may really make innovation easier since they may resolve many, if not most, of the routine decisions which interfere with the bureaucrat's undertaking innovation. We are suggesting that a high degree of rule formalization may be a positive influence since it can give the climber or conserver bureaucrat more time to engage in this time consuming activity. Where a bureau does not have a high degree Of rule formalization, the bureaucrat will have less time to be innovative since he will personally have to make many more of the routine decisions that must be made if the bureau is to function. The source Of new personnel also suggests to the bureaucrat whether or not he should adopt innovation as a suitable strategy. Where new personnel come from is a potential factor in the resources available for innovative activity. When a bureau recruits for all positions, high and low, from outside itself, the bureau exposes itself to fresh and new information that the new officials bring with them. These new bureaucrats are likely to take different approaches to solving problems confronting the bureau since they are originally not as intimately involved with those problems. The cost of their new information is lower than the cost of information that would have to be gathered Specifically to undertake innovation.' The final dimension is the level of financial support which reflects the amount of resources available to perform the bureau's social function. The level of support is also indicative of the resources that can be diverted to innovative action. Where the financial support level is low, few or no resources can be diverted to 194 innovation since they are all needed to perform the basic social function/functions. Innovation requires the expenditure of time, material and personnel which just are not available at lower levels of financial support. Where financial support is stronger, there is a greater capability to support innovation and the social function at the same time. Having identified the dimensions and their suspected influence as a resource in the selection of innovation as a strategy, we can now suggest how climbers and conservers respond to these individual dimensions. They basically make their decisions in terms of how the dimension influences their chances of meeting or fulfilling the expectations of those who can satisfy/provide the goals the bureaucrat wants. Those who can satisfy the needs of a bureaucrat, either directly or indirectly, include many different actors in his environ- ment. The most directly important of these actors is the political sovereign of the bureau. The sovereign directly controls the granting of those goals most desired by our bureaucrat-~the sovereign can give more of them, he can continue to give what he is presently giving or he can reduce what he is giving. This direct Control provides an extremely strong movitation for the bureaucrat to meet the demands of the sovereign. The political sovereign has the strongest influence upon the conserver since he is not particularly interested in further- ing his personal status by jumping. The conserver must necessarily meet the expectations of his present sovereign if he is to obtain the goals of security and convenience. The climber has other expectations to fulfill in addition to those of the sovereign. He knows that if 195 his own sovereign does provide him with the goals he wants, he has the option of jumping to another bureau. Thus, he must also fulfill the expectations of others such as professional peers if he is to create the personal reputation necessary for jumping. The first individual dimension to be considered is size of bureau. If climbers and conservers dominate large bureaus, they will be more likely to adOpt innovation to obtain their goals than if they were dominant in a small bureau. They both respond to the need and capability for innovation of a large bureau but for different reasons. The climber recognizes that if he is innovative in response to the greater need and capability of a large bureau, he is improving his chances for further promotion by the political sovereign. He could also be aggrandizing his position by develOping innovative solutions which increase the people assigned to him or the number of functions he is responsible for. By being innovative, he may also be increasing his capability for jumping in the future by creating a personal reputation for innovative action which is independent of the bureau. In each case, the climber is responding to facts--need and capability of large bureaus—-which other actors are well aware of. The climber knows that they will expect innovative behavior given these conditions--failure to be innovative would be very costly. The con- server reSponds to Size differently. In a large bureau, he recognizes the need and capability to be innovative. By failing to act correctly in such a situation, he could lose what he presently has because the political sovereign expects this type of behavior. The political sovereign in this situation is likely responding to pressures from 196 those who elect them. Thus, to insure his security and convenience, the conserver will make use of the greater capability of the large bureau to respond to the element Of need and be innovative to meet the expectations of the sovereign. The small bureau forces neither the climber nor the conserver to be innovative to obtain their goals so they will be less innovative in this sized bureau. A high degree of professional specialization in a bureau contributes to the capabilities of the climber to achieve his goals through promotion and jumping. The higher levels of information at reduced costs which occur with a highly professionalized bureaucracy, make it easier for the climber to engage in activities which will either win him promotions or improve his personal reputation which is important if he is to jump to another bureau. Thus, the climber will take advantage of this condition and select innovation as a suit- able strategy to maximize his preferences by meeting the expectation of those who control his potential for achieving his goals. Where there is not a high level of professional Specialization in a bureau, the information level is reduced and the costs are increased which is detrimental in the climber's calculations to be innovative. Con- servers find professional specialization another factor which may, more or less, force them to be innovative. With the higher levels of professional specialization, information is cheaper and more readily available which means the conserver has fewer reasons/excuses for not being innovative when dealing with his political sovereign. The political sovereign can reasonably expect different solutions to problems when he provides fOr an increased number of professionals. 197 If he does not receive them, the conserver is in difficulty for his failure to use this resource provided by the sovereign. The conserver also has the pressures of the professionals below him who want to utilize their skills which often requires innovative action on the part of the conserver and the bureau. Since conservers dominate, they have to deal with these pressures and the easiest and least upsetting to their personal status can be the innovation the lower level pro- fessionals suggest. In bureaus with low professional Specialization, these pressures are not present to force the conserver to engage in innovation. The way that climbers/conservers react to rule formalization in a bureau is not as clear as it is with some of the other dimensions of the Opportunity structure. When climbers face a high degree of rule formalization, they are in a favorable environment as far as their promotional Opportunity is concerned. An extensive formalized rule system frees them from having to devote a major portion of their time to making the commonplace decisions day-to-day operations require. They can devote their attentions to non-programmed activities, like innovation, which take much time and effort on their part. Promotions are often based, in large part, upon how the climber performs the non-programmed activities of the bureau. There- fore, a high degree of rule formalization allows the climber to be innovative and fulfill the requirements of the sovereign for promo- tion. The Opposite effect would result from a low degree of rule formalization--much of the climber's time and effort would be devoted to routine decision-making which would reduce his Opportunity to gain 198 promotion by the quality of his non-programmed activities. Conservers are faced with a different situation in a bureau with extensively formalized rules. This is not necessarily to their liking. When the bureau does not have a formalized rule system, conservers can spend their time in routine decision-making with its decreased chances of making serious errors for which they would be held accountable by the sovereign. They also do not have the time and capability to engage in non-programmed activity since their attention is focused upon the routine. However, where rules can handle much of the necessary pro— grammed decision-making, conservers can no longer claim to the sovereign that they do not have the time to engage in non-programmed activity like innovation (which is inherently more threatening to them in terms of their goals because of the increased chance for making mistakes). While the conserver does not undertake innovation to be rewarded with future promotion, he does so to insure that the political sovereign is satisfied enough with his performance to main- tain that which he presently receives. Climbers who dominate bureaus where the majority of the recruiting of new bureaucrats is from outside are doubly motivated to be innovative. First, these new officials bring with them new information at reduced cost which in itself contributes to a decision to be innovative. It is cheaper than normal to be innovative which, in turn, improves the climbers' chances for promotion by meeting the sovereign's expectations for promotion. A second factor is that these new officials bring with them fresh perspectives or approaches to solving the problems confronting the bureau. They, themselves, are 199 introducing innovation in the bureau's decision-making by taking part which may be the intent of the sovereign when he approves the hiring of these officials from outside the bureau. As we have already pointed out, the act of innovation by a climber may contribute to future promotion or the creation of a reputation necessary for jumping. If the climber already in the bureau wishes to compete in the future, he must make use of these new officials to insure his own future by also being innovative. Where bureau recruitment is primarily internal except at the lowest ranks, the new information is not as readily available and new ideas are not introduced by new personnel. In this situation, less innovation will be performed by the climbers of that bureau. When conservers dominate a bureau with a high degree of outside recruitment, they reSpond more to the pressure being brought by the new officials than the increased information levels.16 These new bureaucrats will introduce new ways of solving the problems they face. They will be innovative at least for some initial period of time just by being new. If the conserver is to maintain what he presently has, he must deal with the threat these new officials raise. What if they are successful and he opposed their activity? How does this affect his relationship to the political sovereign? The easiest and possibly least costly way of dealing with the new bureaucrats is to let them go ahead and attempt to be innovative--acquiesce to their activity but not strongly support it. If the new bureaucrats are successful, the conserver can take credit for showing the insight of a good leader by letting these new officials have "their head" and develOping ZOO innovative solutions. If they fail, the conserver can always say he was reluctant but he felt he should let them at least try. It did not work, but he is not to blame for their incompetency. This puts the new officials in a tenuous position for the future and reduces their immediate threat to the conserver. Conserver dominant bureaus where outside recruiting for new officials is the common practice will be more innovative than bureaus with little outside recruitment. Level of financial support is clearly important since the resources of a bureau are important considerations in any bureaucrat's deliberations of how to maximize his goals. Innovation as a strategy is dependent upon resource level since the decision of whether or not to be innovative must be at least partially based upon the capability of the bureau to support this activity. Climbers know that the successful introduction of innovation (or at least attempt at intro- duction) often results in promotions, aggrandizement or independent reputations important for jumping. They also know that failure may be very detrimental to these activities. Therefore, an adequate level of resources is necessary to even begin to engage in innovation. The greater the resources of the bureau, the greater the likelihood that climbers will be innovative and vice versa. This is also true for conservers but for a somewhat different reason. Conservers in bureaus with greater resources are in a situation where the political sovereign expects more results from the bureau Since he has provided more or greater capability to perform the bureau's social function. The sovereign by this manifestation of support has demonstrated that he wants a social function well performed. If he contrasts the results 201 of his bureau with those of other bureaus and finds that his bureau's outputs are not as "good" as the other bureaus, the sovereign will likely be unhappy to the point of replacing the top officials who dominate the bureau. Therefore, conservers in bureaus where the resources permit innovation must be innovative to insure that the political sovereign is satisfied with the outputs produced by the bureau. This, in turn, helps insure that the status quo remains unchanged for the conservers. When the bureau has a low level of financial support, the conserver can convincingly argue, if asked by the sovereign why the bureau is not innovative, that there is no capability within the bureau to be innovative if the basic social function is to be performed. The more likely condition in bureaus with low support is that the political sovereign never asks this question since he knows he is not providing enough resources (for whatever reason) to do much more than the minimum necessary. To this point nothing has been mentioned about the possible effects of the dominant bureaucrats deciding whether or not to be innovative to maximize their preference orderings. What impact does their decision have upon the behavior of the bureau they dominate? Since dominant officials make decisions based upon their personal preferences even when engaging in their official activities, the behavior of the bureau will be influenced by their personal preferences. Thus, we expect bureaus possessing the favorable condi- tions of the five dimensions to be marked by greater innovation in their output than bureaus without those characteristics. In this discussion of the dimensions of the opportunity structure and strategy selection, we have suggested various 202 conditions which contribute to the determination of whether or not innovation will be adopted as a strategy. As far as the five dimen- sions considered here, climbers and conservers respond to these dimensions (given the same conditions) in a similar fashion although they do so for very different reasons. In a large, highly pro- fessionalized, well financed bureau with extensive rules which re- cruits outside itself, climbers and conservers will select innovation as a strategy to Obtain their respective goals. Given these same conditions and another strategy, the results might be very different-- they may or may not adopt the same strategy with the same environment. How they would reSpond to innovation with other possible dimensions of the Opportunity structure may be very different from that prOposed here. The only way of knowing is to examine the strategy in terms of the dimension and its possible effect upon the bureaucrat's chances of getting his goals. This completes the basic develOpment of the model. We have specified the goals that bureaucrats seek to maximize, the conditions under which climbers or conservers will dominate a bureau and the dimensions that influence the selection of a strategy suitable to maximize the preferences of the dominant officials. One possible strategy is examined in relation to the five dimensions suggested as being important to strategy selection. While this model is developed in rather general terms, it is possible to bring some limited empirical evidence to bear upon Specific portions of the model. This evidence suggests that the model has a certain potential 203 utility and is worthy of further testing in subsequent research into bureaucratic behavior. The evidence that we do have is in two principal areas: bureaucrats do exist who seem to have goals which conform to those posited for climbers and conservers; and the five dimensions of the opportunity structure do appear useful in selecting innovation as a strategy. This first evidence dealing with the existence of bureau- crats who fit the climber/conserver distinction comes from two studies discussed in Chapter II'S review of organization theory and research reported in Bidwell's article on school bureaucracy which was also partially reviewed in Chapter 11.17 The studies we are referring to in the discussion of organization theory are the Dill, e£_gl: article on career patterns in business and Swinerton's application of ambition theory to state bureaucrats. The fact that these investigations are in very different contexts is more supportive of the model than if they had taken place only within bureaus performing the same social function. However, we must also acknowledge the limitations of this evidence. The Dill, g£_pl: research is not conducted within bureau- cratic organizations. It involves individuals who are associated with business organizations which have a very different environment from that facing the bureaucrat--at least in terms of measurability of organizational and individual output. What is interesting is that these individuals have a great similarity of goals even though they work in very different environments from bureaucrats. The Swinerton research deals with individuals who are classified as bureaucrats although they do not exactly fit the definition of bureaucrat as 204 developed here. There is enough resemblance though to be at least suggestive of support for our position. The studies in Bidwell do not share these limitations since they report directly on educational bureaucracies and bureaucrats. The Dill, e£_gl: article suggests that individuals in a business environment have two basic goals or preference patterns. There is one pattern in which the goal is personal advancement. It incorporates a preference for doing managerial work and/or achieving managerial income and status. These individuals want positions where they can act independently and be judged by the results of their efforts. They are eager and want new jobs. The men who hold the second pattern have more limited aSpirations--they do not actively seek advancement as a goal. Their goal seems to be maintaining what they presently have. They have less desire for the benefits (power, autonomy, income, status) that are conferred by a managerial position. These men are less willing to accept the costs of being a manager and have more of a personal commitment to their present position. In terms of achieving their goals, those seeking advancement must pay particular attention to the expectations and standards of the men who govern their chances for promotion. Those who have been successful in making rapid progress seem to be especially sensitive to specific features of the environment in which they are working. Theyrelate their goals to their environ- ment to determine the strategies they will use to obtain those goals. These two preference orderings found are not dissimilar to those suggested for climbers and conservers. While described somewhat differently, the advancement goal is much like the climber's goals of 205 more power, income and prestige. The conserver's goals of convenience and security basically describe the goals being sought in the other preference pattern. It is also interesting to note the importance given the environment by the managers in determining what strategies they will select to achieve their desired goal. In "Ambition and American State Executives," Swinerton attempts to relate ambition theory to upper echelon State executive in Six comparative states. He is able to find state executives who express each direction of ambition and correlate four background characteristics with those ambitions-~age, number of terms of service in present posi— tion, salary incentives and party activity. Swinerton attempts to identify strategies that these executives might use to maximize the possibility of moving in the direction of their ambition given their reSpective structure of political opportunity. The strategies were expressed in terms of the executive's theoretical response to require- ments for political behavior on their part. Very different perceived patterns of behavior were obtained from those executives expressing a prOgressive or discrete ambition than the executives stating a static ambition. Swinerton believes personal security is the influencing variable which differentiates the two patterns of behavior. While his research is not completely supportive of our model, it does point out differences in the goal structures of state executives and in their anticipated behavior in response to these ambitions. Executives with static ambition are against engaging in activity which threatens their present position while those executives with progressive ambi- tions are very strongly in favor of this type of activity since they 206 view it as important in obtaining higher office. For these two groups of officials, there appears to be real differences in their attitude toward change-~much like that posited for climbers and conservers. Executives having progressive ambition are willing to undertake activity which alters their personal status quo to gain their goal of higher office while those officials with static ambitions do not particularly desire to engage in this type of activity if they have a choice. The responses from executives with discrete ambition is not helpful here since there is no equivalent set of bureaucratic goals incorporated in the model. The research reported in Bidwell is directly applicable to the model since it is dealing with bureaucrats (school superintendents) and bureaucracies (school districts). The first investigator, Carlson, argues that there are two types of school superintendents distinguished by differences in career line characteristics--place-bound and career- bound superintendents. These bureaucrats differ systematically in attitude and in characteristics. Place-bound administrators desire to advance their careers without running the risks or incurring the costs of inter-system mobility or of entering a new situation. They are not particularly oriented toward career mobility. They did not actively seek their job. Career-bound superintendents are quite different-- they tend to equate intersystem and career mobility, they are mobility oriented and they actively sought new jobs according to their own career needs. Career-bound officials are well known to fellow professionals while place-bound administrators are not. Carlson also believes that career-bound individuals are oriented toward professional 207 peers and to professional ideology. The place-bound superintendent supports the existing policies of the school and the values of the local community. Other differences between place-bound and career- bound bureaucrats are identified. Career-bound administrators have more numerous job alternatives available to them and are less dependent upon the local school board. This, in turn, is reflected in salary levels where place-bound officials receive considerably lower salaries than do the career-bound. These superintendents behave quite differ- ently. Career-bound superintendents make new rules and rapidly add new members to their personal staffs early in their tenure. The place-bound administrator emphasizes established rules and makes few changes in the central staff. The other useful study is Seeman's research that finds two basic kinds of officials or preference pat- terns. The less mobile/non-mobility oriented superintendents are characterized by maintenance of the status quo. The other superin- tendents are highly mobile and mobility oriented and they emphasize status advancement. In both studies, we find a basic distinction being drawn between bureaucrats based upon the type of goals they are personally seeking. One set of goals requires an alteration in the status quo for the individual while the other set is designed not to introduce change. While neither Carlson nor Seeman Specify what the exact goals are for each of their superintendents, it is not too broad an inference to make to suggest that the "career-bound or mobile" superintendent seeks goals similar to those proposed for climber bureaucrats and "place-bound or non-mobile" superintendent desires 208 goals like those identified for the conserver. Other similarities can also be noted: career-bound officials have alternative job openings available to them--opportunities for climbers to jump to obtain their goals; place-bound administrators do not take "unnecessary" risks as is the natural inclination of conservers. When considered together, these four studies provide limited support for the belief that the preference patterns developed here do exist and that the individuals who seek them conform to the climber/ conserver distinction raised. It is clear from this research that the individuals studied actively reSpond to their environment-~their opportunity structure--in maximizing their preference orderings. They adopt strategies of action which seem to them to be appropriate to gain their goals given the environment each faces. None of the research provides support for the dimensions of the opportunity structure proposed to identify bureaus where climbers or conservers could be dominant. However, some of the variables identified with behavior are considered potentially helpful in identifying an indi- vidual as either a climber or a conserver in a bureau. These variables are treated in the final section on research strategies. The utility of the five dimensions of the opportunity structure in selecting a strategy can be seen from the research conducted earlier in this dissertation. One of the major purposes of this research is to increase our knowledge of a bureau's environment by examining Specific dimensions which were believed to be, at least, potentially important to the bureaucrats who served in those bureaus. We suggested that if a dimension of the Opportunity structure was important, this 209 importance would be reflected by variations in the public policy that emerged from bureaus. This position is basically in conformance with the model since we argue in the model that the behavior of a bureau, expressed in policy, will be influenced by the preferences of the officials that dominate the bureau. The Strategy that the domi— nant bureaucrats select to obtain their goals is chosen because of the bureau's opportunity structure. If a strategy of innovation is selected, then the behavior of that bureau as reflected by its policy will be more innovative than if a non-innovative strategy was chosen. We will now demonstrate that each of the five dimensions of the opportunity structure is related to variations in public policy (innovation) as it should be if our model is correct and bureaucrats select their strategies based upon consideration of these dimensions. While this evidence is inferential and limited to one type of bureau, it does provide further support for the continued development and testing of the model. In considering innovation as a strategy, we posited that the dimensions of size, professional Specialization, financial support, rule formalization and recruitment pattern would all influence the decision-making of climbers and conservers. If the favorable condi- tion is present in a dimension, innovation would be selected as useful in Obtaining the desired goals. If this is true, then the policy in the bureau will reflect the choice of innovation as a strategy. A bureau with a favorable condition of a dimension should be more innovative than if the unfavorable condition of the dimension is present. The results of relating these five dimensions to innovation supports the model's use of the dimensions. 210 Size of bureau exhibits a strong relationship to innovation when tested in an educational bureaucracy. The large school districts are much more educationally innovative than are the small districts. This is exactly the direction the model suggests. Climbers and conservers Should both select innovation as a strategy in large bureaus. Large bureaus' policy Should then reflect more innovation. Another strong association is found between degree of professional Specialization of school districts and educational innovation. This also conforms to the model since the most highly Specialized districts are the most innovative districts. Climbers/conservers select innovation as a strategy in bureaus with a high degree of pro- fessional specialization. Level of financial support and school district recruitment pattern display only a moderate association to educational innova— tion. However, the relationships found are in the direction the model suggests. School districts with higher levels of financial support are more educationally innovative than districts with fewer resources. This is exactly the pattern of innovation one would expect if the model is correct. Climbers/conservers choose innova- tion as a strategy in bureaus with greater resources and do not select it in bureaus with fewer resources. In the model, it is argued that innovation is a suitable strategy for dominant bureau- crats where the bureau recruits from outside itself. We find that school districts where recruitment of Officials is predominantly from outside are more innovative than districts where the recruitment is internal (except at the lowest levels) which again is supportive of 211 the model. The final dimension-~rule formalization--exhibits the weakest relationship with educational innovation. Districts that have a high degree of rule formalization are somewhat more innovative than districts with a low degree of formalization. This weak rela- tionship is the direction of the prediction of the model. Dominant bureaucrats will be more innovative in bureaus with a high degree of rule formalization than in bureaus with low rule formalization. Each of these individual dimensions is related to innovation and in the direction the model suggests. No attempt is made in the model to answer the question of what the result would be if the dominant officials considered these dimensions together in their deliberations.* So little is known about strategy selection at this point that it is difficult to hypothesize what the impact would be. But if forced to, we believe that the bureaucrat would select from the various dimensions only those that are most important in giving him clues for correct strategy selection. The bureaucrat has only a limited capability in responding to his environment therefore he must include in his calculations only those dimensions which are highly reliable in terms of obtaining his goals. Exactly how many dimensions he can deal with at the same time is an empirical question which cannot be answered here. We do suggest, however, which of the five dimensions are relatively more important to strategy selection when considered together. When the five dimensions were clustered together as the resource cluster and related to innovation, we determined that only three of the dimensions continued to exhibit an association with innovation. Size, professional specialization and financial support 212 continued to diSplay a relationship to innovation while the other two dropped out. If we can make the same inference about the grouped dimensions as we have for the individual dimensions, then climbers and conservers could very well consider only these three dimensions in their strategy selection process. These three dimensions together may be the best indicators of the expectations being placed upon the bureaucrat by the political sovereign. The political sovereign knows of the greater need of the large bureau and its greater inherent capability to respond to those needs. At the same time, the sovereign has provided greater capability to respond by having a highly pro- fessionalized staff and by the high level of resources available. Together they indicate a strong expectation for a particular type of behavior--innovation--and the climber and conserver respond by being more innovative. Thus, districts which combine the favorable condi- tions of the grouped dimensions are more innovative than districts which do not. Having examined the empirical evidence that can presently be brought to bear on the model, we feel that it supports the probability of the existence of bureaucrats and goals similar to those develOped in the model and the importance of the dimensions of the opportunity structure to the strategy selection process for at least one strategy--innovation. We can now touch upon some research strategies that may be applicable in testing the model. Proposed Research Strategy_ In this final section, research strategies are suggested which will hopefully aid in ascertaining the usefulness of the model 213 in future research into bureaucratic behavior. For purposes of discussion, we will assume that resources such as time and money are basically unlimited and thus not an important consideration in developing a suitable research strategy. We also assume that the model is being tested within bureaus having the same social function/ functions. So little is known about the real influence of social function that it is necessary to restrict ourselves, at this point, to bureaus where social function is controlled. Once knowledge of bureau Opportunity structure and behavior is expanded for similar type bureaus, then comparative research across bureaus with varying social function can be undertaken. The actual strategies being pro- posed are in general terms and not designed fOr any particular type of bureau--hopefully this model will prove suitable for all different types of bureaus. In the model, two basic preference orderings were posited. Climber and conserver bureaucrats were believed to seek these prefer- ence orderings. They could only maximize their preferences in bureaus with certain types of opportunity structures--where they would domi- nate the bureau. To test for the existence of bureaucrats with these preferences in the appropriate bureaus requires that the bureaucrats be identified according to their goals and the Opportunity structure of the bureaus they serve within. Are there really two preference patterns being maximized and are those seeking them dominant in bureaus with the appropriate opportunity structure? The most obvious way Of identifying preference/goals is to ask the individual bureaucrat the goals he has set for himself. A number of different 214 psychological testing devices are available and could be used to test the reliability and consistency of their responses. This would tell us something about their goals and whether or not these goals corres- pond in any way to those of the model. However, there is always the question of whether these bureaucrats actually do seek the goals they express for themselves. A preferred way to identify the existence of bureaucrats that fit the model is to draw upon the general strategy of those who conduct research in ambition theory and examine mani- festations of their overt behavior in addition to asking the indi- vidual his personal goals-~his political ambition. With knowledge of the bureaucrat's behavior, one can determine if these bureaucrats act consistently with the goals they have set for themselves.18 What sort of overt bureaucratic behavior should one then examine in this process? There are any number of variables that reflect behavior on the part of the individual bureaucrat. Since public data on the careers of bureaucrats is basically unavailable, information about these variables will have to be obtained from the bureaucrat himself either through a personal interview and/or a questionnaire. The type of data to be sought should include his: educational level; efforts to improve his professional training since his formal educational training ended (seminars, institutes, etc.); individual pattern of job mobility; present position in the bureau; length of time in present position; professional activity expressed in number of organ- izations he is a member of and the Offices held if any; salary scale; total years of service in the bureau; activity in adding new men to his personal staff (number added and how rapidly has this been done); and failures, if any, at being promoted in his bureau. 215 While this list is likely incomplete and may include un- related behavior, it does suggest the types of manifested bureau- cratic behavior that are seen as being related to the preferences a bureaucrat holds. If a bureaucrat expresses a preference for climber goals, then his behavior as measured by these variables should be reflective of these goals. For example, climbers seek more power, income and prestige. If a bureaucrat had actually pursued those goals, his behavior should exhibit, to some degree, activity that Shows that he at least attempted to obtain those goals somewhere along the way. If he has a low educational level, failed to ever get pro- moted, receives a very low salary, has held his present position for thirty years and is totally inactive professionally, it would be very hard to conclude that this individual was a climber even if he expressed goals like those of a climber. Such an inconsistency is not expected often if individuals do, in fact, reSpond to their environment in a rational manner in the long run where it is possible to alter goals. Such a behavior pattern would be more typical of the conserver bureaucrat. It is reCOgnized that distinguishing between climbers and conservers through expressed behavior may be an exact process at times eSpecially if only a short period of behavior is available to study. This is why we have suggested that expressed goals be obtained along with manifested behavior rather than simply inferring the existence of climbers and conservers through behavior. Each serves as a check on the other to better insure accuracy in the identification process. 216 Once the bureaucrats are classified according to preference pattern (expressed goals and overt bureaucratic behavior), we can attempt to discover whether or not they dominate the bureaus with the opportunity structures suggested. Since dominance in a bureau is achieved when the bureaucrats filling the key positions in a hierarchy hold the same preference orderings, the bureau's hierarchical authority structure must be examined to determine if the key authority positions are filled by bureaucrats having the same goals. The number of posi- tions that have to be considered is a function of the tallness or flatness of the hierarchy. Fewer bureaucrats need the same preference ordering to dominate a bureau with a tall hierarchy while more bureau- crats with the same goals are required to dominate a flat hierarchy-- it is all a question of organizational authority and control. Having obtained the necessary knowledge about the bureaucrats who fill the dominant hierarchical positions, we then investigate the opportunity structure of the bureau to discover if it has the characteristics pro- posed in Table 6-l. How important each dimension is in identifying the bureau where climbers or conservers will dominate is not known. Like other questions dealing with the opportunity structure, it is an empir— ical question which can only be resolved by further research. When operationalizing the dimensions prOposed, the recommendations in Chapter IV should be considered since we now know some of the diffi- culties associated with their use. This is likely one of the easier parts of testing the model. Once preferences are known and dominance determined, it is a relatively simplé assignment to measure the Oppor- tunity structure to see if the three dimensions correspond to the characteristics suggested. 217 The third and last step in evaluating the model is probably the most difficult. Strategy selection based upon the opportunity structure is difficult since our knowledge is extremely limited in this area. The five dimensions Of the Opportunity structure tested here were found to be individually related to innovation. With this as a basis and our other knowledge of dominant officials in their bureaus, we could go ahead and determine if both climber and con- server dominated bureaus with the apprOpriate features of these dimensions are more innovative than other dominant bureaus that lack these features. We could also ask the dominant officials if they consciously made an effort to be innovative and contrast their re- Sponses with the characteristics of the opportunity structure to see if they correctly align--are those who try and be innovative found in bureaus with the most favorable characteristics of the dimensions of the Opportunity structure. The difficulty arises when considering other strategies. Obviously, many other strategies are available to climbers and conservers to Obtain their goals. An important question is what dimensions of the opportunity structure are important to the calculations of the bureaucrat in selecting the appropriate action of the general strategy. The dimensions suggested here may or may not prove useful with other strategies. This means that a great deal Of further research into the Opportunity structure is necessary if we are to test a broad range of strategies in the model. Another important consideration is what sort of strategies do bureaucrats adOpt to maximize their goals. This is also an area of high ignorance. Little is known about this phenomenon in bureaucracy 218 or in other organizational settings including the political arena where ambition theory is tested. One example of the type of strategy that might be incorporated is found in Carlson's research into career-bound and place-bound school superintendents who adopted different strategies to achieve their respective goals--their professional orientations differed. Career-bound officials adopted policies that reflected professional ideology while place-bound superintendents supported existing policies--ones which mirrored the values of the local community. Several different broad classes of strategy can be offered as being potentially important to the bureau- crat. How the individual climber or conserver responds must be left for future consideration. One broad area of strategy that has been explored as a policy area for bureaucracy is the field of budgeting. Budget strategies are of vital concern to all bureaucrats since the bureau's budget is the source of many of his personal goals. Wildavsky in The Politics of the Budgetary Process discusses a number of different budget strategies (increase the base, maintain the base, etc.) which might be tested in the context of the model. Another broad budget strategy proposed by Niskanen for rational bureaucrats is a strategy of budget maximization for the bureau. There is suffi- cient previous concern to conclude that budgeting is definitely one strategy that should be incorporated in future research. Another class of strategies could include the whole area of leadership techniques used by bureaucrats in their bureaus. How dominant bureaucrats provide leadership to other bureaucrats should effect their ability to gain their desired goals. Other strategies 219 that might be examined include: the communication techniques used in bureaus, the different ways control is maintained in bureaus and the use of bureau ideolOgy. Each of these strategies has the potential for influencing the ability of bureaucrats to gain their preferences. Other strategies having this capability obviously exist but those suggested provide a broad range of possible strategies, one of which should be applicable to any type of bureau environment. Our concern in this final section has been to propose a few tentative thoughts on how one might test the decision-model. This has been done in general terms which permits its application in any type of bureaucratic setting. Another reason for this lack of specificity is the low level of knowledge of the bureau's Opportunity structure and maximizing strategies which would permit a more complete treatment. We believe the way to test the model is to identify the goals that bureaucrats seek and ascertain if their overt bureaucratic behavior reflects the goals they have selected for themselves. Once this is accomplished, it is necessary to find out if these bureaucrats dominate bureaus with the opportunity structure characteristics hypothesized. Finally, we can test to see if these bureaucrats adopt the strategy suggested as being associated with the five dimensions of the bureau's Opportunity structure. The task we set out to do in Chapter I is now complete. We prOposed to expand our knowledge of bureaucracy and its behavior expressed in public policy and to develop a bureaucratic decision-model which would attempt to explain bureau behavior. To accomplish this, a limited examination of the bureau's environment and policy was first 220 made. We found that six of the eight individual dimensions of the bureau's opportunity structure were related to educational innovation in some fashion--not always in the direction hypothesized nor as strongly as one would like but nevertheless associated. The resource dimension cluster was also found to be correlated to innovation. The career cluster did not diSplay a relationship to innovation. Sugges- tions were made to improve these dimensions and their use in future research. Finally, a model was develOped which attempts to demonstrate how bureau behavior is a product of rational bureaucrats responding to their environment-—their opportunity structure--to obtain their stated personal goals. How successfully all of this has been accomplished must be the decision of the reader. FOOTNOTES--CHAPTER VI 1Examples of this latter type of proposition are contained in Chapter III in the development of the hypotheses dealing with the individual dimensions and the cluster dimensions of the Opportunity structure. However, the suggested relationships between behavior (manifested in innovation) and condition (individual dimensions and cluster dimensions) were not always presented in terms of explicit types of officials as it was felt that this was not always necessary for the basic argument relating the condition and behavior. Chapter XXII of Inside Bureaucracy_contains a summary of these prOpositionS that are develOped’throughout the book. 2Downs, pp. 92-96. 3These definitions and assumptions are the same as those originally developed in Chapter III, pp. 38-40. The model being proposed is not specifically tied to educational bureaucracies although portions of it are discussed in an educational context later in the chapter. 4These goals are the same as those used by Downs to develop his ideal types of bureaucratic officials. They were also defined earlier in Chapter III, pp. 40-41. 5We have already suggested one way of distinguishing between goals. While all goals are viewed as being self—interested, one can characterize goals on the basis of who receives the benefit. In the "pure" self-interested goals, all benefits accrue to the individual seeking and obtaining the goal. The other self-interested goals have benefits which are divided between the individual seeking the goal and others not directly seeking the goal. Another way of making a distinction between goals is by determining whether or not the goals involve change in one's present personal status--does obtaining this goal alter the status quo for the individual. This sort of dis- tinction is most suggestive when considered together with the benefit characteristics of the goal. In fact, this is the basis for the preference orderings developed. The goals contained at the top of both preference orderings are "pure" self—interested goals which is 221 222 in keeping with the assumptions of the model. The division between the pure self-interested goals is based upon personal change require- ments of the goals. Security is by definition a goal where main- tenance of already established levels of power, income and prestige is the value sought. Convenience also involves resistance to change. In this case, behavior requiring more personal effort on the indi- vidual's part. The other three pure self-interested goals all require change in the bureaucrat's personal status if they are to be obtained. The distinction between preferences for climbers.and conservers is between pure self-interest and self-interest goals with the pure self-interested goals being divided according to the requirement for change in the status quo of the individual seeking the goal. 6It is assumed that all nine goals suggested must be included somewhere in the preference ordering of each bureaucrat as does Downs, pp. 84-89. 7One advantage of identifying bureaucrats in terms of their preference orderings is the relative ease in accounting for change. If there were really different types of bureaucrats as suggested by Downs, then an official changing from one type of bureaucrat to another would be difficult to account for. When all bureaucrats are the same, only individual preferences change. The only.things that need to be examined to account for change then are those factors which influence the bureaucrat to alter his preference ordering. In this model, we suggest that change be examined in light of the opportunity structure of the bureau the official serves within. No assumption is being made that a bureaucrat's preferences will remain constant over time. However in the short run, the bureaucrat does not alter them without a very high cost (sunken capital investment in present preferences and associated behavior) and thus they are considered fixed. Additionally, the opportunity structure of a bureau is relatively stable in the short run. There is no com- pelling need to modify preferences or strategies in the short run. 8There is nothing equivalent to a discrete ambition among bureaucratic officials. A politician who has a discrete ambition seeks a particular office for its Specific term and then voluntarily withdraws from that office. By the definition of bureaucrat, we exclude any one with an analogous goal of this type from being a bureaucrat. The only individual in a bureaucracy who might correSpond to a politician with a discrete ambition are the tOp members of a bureaucracy who owe their positions to the political sovereign and they are not bureaucrats as we have defined them. 9Downs, p. 11. The reason for the dominance by a few indi- viduals over a bureau's complete Operation is due to the hierarchical authority structure of a bureau which tends to concentrate power diSproportionately at the top. The number of bureaucrats with a particular preference ordering needed to dominate is related to the 223 tallness or flatness of the hierarchy. In a very tall hierarchy (greater number of levels of conflict settlers), fewer bureaucrats with the same ordering are needed while a flat hierarchy requires more bureaucrats with the same preference to dominate. For simplicity sake, bureaucrats are often identified throughout the remainder of themodel as simply climbers or conservers. 10Downs, pp. 92-96. Promotion is achieved in two principle ways: please your superiors and score well on whatever objective standards are used in appraising your promotion potential/ qualifications. Aggrandizement is increasing the amount of power, income and prestige associated with one's present position. A bureaucrat has a number of ways of aggrandizing, i.e., do more than the previous office-holder, do the same things better or both. Jumping is the practice of leaving one's present office in a bureau and taking a new position in another organization for the purposes of gaining increased power, income and prestige. 11These three dimensions are the same as those discussed in Chapter III and IV. Their use in this model is quite different from that suggested previously. This difference in use may be a possible reason why no relationship was found to exist between two of the three individual dimensions and innovation nor when they were considered together as a cluster with innovation. It was originally hypothesized that they would be directly related to innovation since they were a part of the opportunity structure and a bureaucrat's behavior is in response to that environment. The evidence did not support this position. While these dimensions are likely a part of a bureaucrat's opportunity structure, he does not respond to them in a way that reflects itself directly in behavior. In the model, we suggest that these dimensions help determine which bureaucrat will dominate a bureau but not the strategies he will select to obtain his goals. We would then not expect these dimensions to be directly related to bureau behavior in the model. The evidence tends to support this inference since no relationship to innovation is found except for the individual dimension of administrative turnover rate and that relationship disappears when all three dimensions are con- sidered tOgether. The use of these three dimensions should be restricted to suggesting where dominant bureaucrats will be found in bureaus not the strategies they will adopt. 12Downs, pp. 10-22. 13The age-lump phenomenon is the tendency of the average age of the members of a bureau to rise tOgether, a "lump" of personnel, since they all joined the bureau at approximately the same time during the period of original rapid growth. Periods of rapid growth are needed to break down this mass aging process by bringing younger individuals into the bureau. Downs, pp. 18-22. 224 14Like the first three dimensions, these dimensions have been discussed completely in earlier chapters and will not be expanded upon here. How these dimensions relate to the previous research will be treated in the discussion of the empirical evidence supporting the use of the model. 15This strategy selection process seems to be somewhat similar to the discussion that has been raised about parties domi- nated by those seeking office and parties dominated by those seeking benefits. While they have very different goals, the two types of parties often seem to end up adOpting a similar strategy in pursuit of their reSpective goals. The reason why involves the vote- minimizing and vote-maximizing Strategy question that is raised for the two types of parties. The structure of political Opportunity may dictate that they both select a vote-maximization strategy even though the goals may theoretically call for different strategies. Professor Schlesinger is working with this issue and is attempting to resolve the strategy problem. He first pointed out the possi- bility that climbers and conservers might adopt the same strategy even though they sought different goals. ~’ “may" F 16Downs argues that there is little, if any, outside recruit- ing going on in a bureau dominated by conservers except at the lowest entry levels. Downs, p. 230. Whether or not this is true remains' to be verified by empirical evidence. If it is found that conserver dominated bureaus do not recruit from the outside at higher levels, we would not expect them to be innovative as our argument now suggests. However, if there are conserver dominated bureaus that do recruit from outside for all levels of employment, we believe then our own position will hold for the reasons given. 17Dill, et al., "How ASpiring Managers Promote Their Own Careers"; Swinerton, "Ambition and American State Executives"; and Bidwell, "The School as a Formal Organization." 18Factors such as psychological makeup (psychological pre- diSpositions), personality traits, social system, etc., are not included in our considerations since these assume that a bureaucrat's behavior is a function of the social system or individual psycholOgy rather than a reSponse to the institutional framework/environment where he seeks his own personal preferences. Personal charac- teristics such as race, religion, social class are also ignored since these features do not enter into behavior if the bureaucrat serves in the type of bureaus defined here. Ascribed characteristics do not, theoretically, enter into bureau personnel policies since they are based upon bureaucratic role performance. However, there is one variable of the individual which should be incorporated in any identification of preferences and that is the characteristic of 225 age. Both Schlesinger and Downs show that age is an important con- sideration in determining what goals the individual seeks to maximize. There is definite variation in goals reflecting the differences in the age of individuals. While Downs only discusses this in general terms, Schlesinger, Swinerton and Hain all find specific age groupings associated with ambition. This would be the expected Situation with the preference patterns of the model. The actual age groupings favoring one or the other pattern is an empirical question but we can suggest that younger men will be more likely to pursue climber goals while older men will seek conserver goals. The dividing point is very likely around fifty years of age if it is consistent with the broad pattern of other findings in this area. 19Niskanen, Bureaucracy and Representative Government. APPENDICES APPENDIX I CONTROL RELATIONSHIPS OF HYPOTHESIS ONE (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN LEVEL OF FINANCIAL SUPPORT AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Level of Financial Support Other Dimensions G C Age of Bureau - Younger .10 .26 - Older .47 .25 Growth Rate of Bureau-Stagnant/Declin. .46 .24 ' ' -Average .44 .26 t-High .26 .13 Administrative Turnover Rate - Low .53 .32 - Medium .42 .25 - High .59 .24 Degree of Professional Special. - Low .24 .ll - Average .42 .25 - High .59 .24 Degree of Rule Formalization - Low .44 .30 - Average .29 .17 - High .58 .29 Recruitment Pattern of Bureau - Low .17 .07 - Medium .43 .25 - High .57 .34 Size of Bureau - Small .30 .13 - Medium .07 .ll - Large .11 .20 Gamma for Level of Financial Support and Educational Innovation .43 226 APPENDIX 11 CONTROL RELATIONSHIPS OF HYPOTHESIS TWO (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN AGE OF BUREAU AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Other Dimensions Level of Financial Support - Low - Medium - High Growth Rate of Bureau - Stagnant/Declin. - Average - High Administrative Turnover Rate - Low - Medium - High Degree of Professional Special. - Low - Average - High Degree Of Rule Formalization - Low - Average - High Recruitment Pattern of Bureau — Low - Medium - High Size of Bureau - Small - Medium - Large G - .62 - .20 .02 .17 - .17 - .39 - .38 .01 - .22 - .35 .24 -1.0 .17 .18 - .48 - .38 - .23 - .46 - .46 - .04 - .36 Age of Bureau C .22 .09 .05 .11 .07 .15 .16 .02 .13 .12 .18 .15 .36 .09 .15 .11 .15 .17 .17 .OS .16 Gamma for Age of Bureau and Educational Innovation 227 -.13 APPENDIX III CONTROL RELATIONSHIPS OF HYPOTHESIS THREE (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN GROWTH RATE OF BUREAU AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Growth Rate of Bureau Other Dimensions . G C Level of Financial Support - Low .11 .09 - Medium - .04 .08 — High - .17 .12 Age of Bureau - Younger .16 .27 — Older - .20 .12 Administrative Turnover Rate - Low 0.00 .40 - Medium - .15 .10 - High - .38 .22 Degree of Professional Special. - Low .09 .06 - Average - .04 .25 - High - .ll .22 Degree of Rule Formalization - Low - .20 .27 - Average - .17 .15 - High - .04 .19 Recruitment Pattern of Bureau - Low - .23 .15 - Medium - .12 .08 - High - .18 .16 Size of Bureau - Small .19 .09 - Medium - .31 .18 - Large - .22 .22 Gamma for Growth Rate of Bureau and Educational Innovation -.17 228 APPENDIX IV CONTROL RELATIONSHIPS OF HYPOTHESIS FOUR (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENTS, BETWEEN ADMINISTRATIVE TURNOVER RATE AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Administrative Turnover Rate Other Dimensions G C Level Of Financial Support - Low - .25 .14 - Medium - .20 .12 - High - .22 .12 Age of Bureau - Younger - .14 .19 - Older - .13 .11 Growth Rate of Bureau - Stagnant/Declin. - .07 .09 - Average - .01 .16 - High - .43 .22 Degree of Professional Special. - Low - .14 .05 - Average .27 .14 - High 0.00 .16 Degree of Rule Formalization - Low - .28 .32 - Average - .12 .10 - High - .04 .17 Recruitment Pattern of Bureau - Low - .13 .06 - Medium - .18 .ll - High .07 .14 Size of Bureau - Small - .05 .09 - Medium .10 .08 - Large - .10 .18 Gamma for Administrative Turnover Rate and Educational Innovation -.14 229 APPENDIX V CONTROL RELATIONSHIPS OF HYPOTHESIS FIVE (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN DEGREE OF PROFESSIONAL SPECIALIZATION AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Degree of Professional Specialization Other Dimensions G C Level of Financial Support - Low .59 .25 - Medium .63 .43 ~ High .66 .44 Age of Bureau - Younger .46 .44 - Older .72 .45 Growth Rate of Bureau - Stagnant/Declin. .79 .53 - Average .68 .39 — High .46 .36 Administrative Turnover Rate - Low .51 .33 - Medium .70 .45 - High .80 .43 Degree of Rule Formalization — Low .62 .40 - Average .61 .38 - High .80 .47 Recruitment Pattern of Bureau - Low .56 .20 - Medium .63 .40 - High .77 .53 Size of Bureau - Small .19 .07 - Medium .31 .29 - Large .38 .30 Gamma for Degree of Professional Specialization and Educational Innovation 230 .70 APPENDIX VI CONTROL RELATIONSHIPS OF HYPOTHESIS SIX (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN DEGREE OF RULE FORMALIZATION AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Degree of Rule Formalization Other Dimensions G C Level of Financial Support — Low — .55 .25 - Medium - .12 .ll - High - .13 .17 Age of Bureau - Younger 0.00 .44 - Older - .20 .16 Growth Rate of Bureau - Stagnant/Declin. - .34 .23 - Average - .04 .15 - High - .22 .22 Administrative Turnover Rate - Low - .52 .38 - Medium - .17 .14 - High - .10 .23 Degree of Professional Special. - Low - .ll .07 - Average .09 .21 - High - .01 .16 Recruitment Pattern of Bureau - Low .20 .11 - Medium - .18 .15 - High - .47 .29 Size Of Bureau - Small .08 .08 - Medium - .12 .15 - Large .06 .22 Gamma for Degree of Rule Formalization and Educational Innovation -.18 231 APPENDIX VII CONTROL RELATIONSHIPS OF HYPOTHESIS SEVEN (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN RECRUITMENT PATTERN OF BUREAU AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Recruitment Pattern of Bureau Other Dimensions G C Level of Financial Support - Low .11 .16 - Medium .24 .14 - High .48 .29 Age of Bureau - Younger .61 .48 - Older .34 .20 Growth Rate of Bureau - Stagnant/Declin. .30 .21 - Average .44 .24 - High .34 .18 Administrative Turnover Rate - Low .19 .20 - Medium .39 .23 - High .31 .18 Degree of Professional Special. - Low .19 .10 - Average .11 .14 - High .61 .38 Degree of Rule Formalization - Low .56 .33 - Average .43 .26 - High .02 .09 Size of Bureau - Small .19 .13 - Medium .08 .12 - Large .19 .15 Gamma for Recruitment Pattern of Bureau and Educational Innovation .37 232 APPENDIX VIII CONTROL RELATIONSHIPS OF HYPOTHESIS EIGHT (CORRELATIONS, GAMMA AND CONTINGENCY COEFFICIENT, BETWEEN SIZE OF BUREAU AND EDUCATIONAL INNOVATION, BY OTHER DIMENSIONS OF OPPORTUNITY STRUCTURE) Other Dimensions Level of Financial Support - Low - Medium - High Age of Bureau - Younger - Older Growth Rate of Bureau - Stagnant/Declin. - Average - High Administrative Turnover Rate - Low - Medium - High Degree of Professional Special. - Low - Average - High Degree of Rule Formalization - Low - Average - High Recruitment Pattern of Bureau - Low - Medium - High Gamma for Size of Bureau and Educational Innovation 233 Size of Bureau G C .73 .44 .54 .37 .63 .45 .22 .20 .69 .48 .79 .51 .65 .48 .33 .33 .69 .56 .67 .46 .51 .41 .34 .22 .65 .42 .44 .24 .76 .53 .59 .40 .54 .49 .47 .28 .67 .45 .62 .47 gr" . i‘iqm # “ . 2? GAMMAS FOR RESOURCE DIMENSION CLUSTERS ACTUALLY FOUND APPENDIX IX PROFESSIONAL RECRUITMENT FINANCIAL RULE SIZE SPECIALIZATION PATTERN SUPPORT FORMALIZATION G N Small Low Low Low Low/Medium/High .56 19 Medium " .73 39 High " - .71 25 Medium Low " -- 6 Medium " .67 12 High " - .57 16 High Low " -- 3 Medium " -- 6 High " -- 5 Average Low Low " -- 1 Medium " -- 1 High " -- 1 Medium High " -- 1 High Medium " -- 1 Medium Low Low Low Low/Medium/High .33 20 Medium " - .69 42 High " .09 26 Medium Low " .33 11 Medium " .03 26 High " - .33 34 High Low " -- 7 Medium " .20 10 High " .60 5 Average Low Low " -l.0 2 Medium " -— 2 High " .67 10 234 1‘...“ J... _-. SIZE Medium 235 PROFESSIONAL RECRUITMENT FINANCIAL RULE SPECIALIZATION PATTERN SUPPORT FORMALIZATION N Average Medium Low Low/Medium/High 1.0 3 Medium " - 4 High " - 2 High Low " - 2 Medium " -- 1 High " -l.0 4 High Low High " - 1 High Medium " - 1 Low Low Low Low/Medium/High - 1 Medium " - 2 High " - 2 Medium Medium " 1.0 3 High " 1.0 7 High Medium " - 1 High " 1.0 4 Average Low Medium " -l.0 3 High " 1.0 2 Medium Low " 1.0 3 Medium " -l.0 5 High " - .12 18 High Medium " -- 2 High " 0.0 6 High Low High " - 1 Medium Medium " -- 1 High " - 52 15 High Medium " - 2 High " - 10 SELECTED BI BL IOGRAPHY SELECTED BIBLIOGRAPHY I. Articles Baker, Bela 0.; Hardyck, Curtis 0.; and Petrinovich, Lewis F. "Weak Measurements vs. Strong Statistics." Education and Psycho- lOgical Measurement, XXVI (Summer, 1966), 291-309. Black, Gordon S. "A Theory of Political Ambition: Career Choices and the Role of Structural Incentives." American Political Science Review, LXVI (March, 1972), 144-59. "A Theory of Professionalization in Politics." American Political Science Review, LXIV (September, 1970), 865-78. Blalock, Hubert M. "Correlated Independent Variables: The Problem of Multicollinearity." Social Forces, XLII (December, 1963), 233-37. Blankenship, L. Vaughn, and Miles, Raymond E. "Organizational Structure and Managerial Decision Behavior." Administrative Science Quarterly, XIII (June, 1968), 106-20. Bonjean, Charles M., and Grimes, Michael D. "Bureaucracy and Alienation: A Dimensional Approach." Social Forces, XLVIII (March, 1970), 365-73. Brown, Julius 5. "Risk Propensity in Decision Making: A Comparison of Business and Public School Administrators." Administra- tive Science Quarterly, XV (December, 1970), 473-812 Carzo, Rocco, Jr., and Yanouzas, John N. "Effects of Flat and Tall Organization Structure." Administrative Science Quarterly, XIV (June, 1969), 178-91. Clark, Terry N. "Institutionalization of Innovations in Higher Education." Administrative Science Quarterly, XIII (June, 1968), 1-25. Dill, William R.; Hilton, Thomas L.; and Reitman, Walter R. "How Aspiring Managers Promote Their Own Careers." California Management Review, II (Summer, 1960), 9-15. 236 237 Eliot, Thomas H. "Toward an Understanding of Public School Politics." American Political Science Review, LIII (December, 1959), 1032-51. Engel, Gloria V. "Professional Autonomy and Bureaucratic Organiza- tion." Administrative Science Quarterly, XV (March, 1970), 13-21. Engstrom, Richard L. "Political Ambitions and the Prosecutorial . Office." The Journal of Politics, XXXIII (February, 1971), 190-94. Fishel, Jeff. "Ambition and the Political Vocation: Congressional Challengers in American Politics." The Journal of Politics, 25-56. Gregg, Russel T., and Knezevich, Stephen J. "The Superintendent: A Good Look at Him." The American School Board Journal, CLVIII (June, 1971), 12-17. Hall, Richard H., and Tittle, Charles R. "A Note on Bureaucracy and Its Correlates." The American Journal of Sociology, LXXII (November, 1966), 267-72. Meyer, Marshall W. "The Two Authority Structures of Bureaucratic Organization." Administrative Science Quarterly, XIII (September, 1968), 211-28. Mezey, Michael L. "Ambition Theory and the Office of Congressman." The Journal of Politics, XXXII (August, 1970), 563-79. Niskanen, William A. "Nonmarket Decision Making, the Peculiar Economics of Bureaucracy." American Economic Review, LVIII (May, 1968), 293-305. Prewitt, Kenneth, and Nowlin, William. "Political Ambition, Volunteerism, and Electoral Accountability." American Political Science Review, LXIV (March, 1970), 5-17. Ripley, Randall B.; Moreland, William B.; and Sinnreich, Richard H. "Policy-Making: A Conceptual Scheme." American Politics, 1 (January, 1973), 4-43. Rosengren, William R. "Organizational Age, Structure, and Orienta- tions Towards Clients." Social Forces, XLVII (September, 1968), 1-11. Samuel, Yitzhak, and Mannheim, Bilha, F. "A Multidimensional Approach Toward a Typology of Bureaucracy." Administrative Science Quarterly, XV (June, 1970), 216-28. 238 Soule, John W. "Future Political Ambitions and the Behavior of Incumbent State Legislators." Midwest Journal of Political Science, XIII (August, 1969), 439-54. Swinerton, Nelson. "Ambition and American State Executives." Midwest Journal of Political Science, XII (November, 1968), 538-49. II. Books Abbott, Max G. "Hierarchical Impediments to Innovation in Educational Organizations." Organizations and Human Behavior: Focus on Schools. Edited by Fred 0. Cower and Thomas J. Sergiovanni. New York: McGraw-Hill Book Co., 1969. Abrahamson, Mark. The Professional in the Organization. Chicago: I Rand McNally 8 Co., 1967. Alexis, Marcus, and Wilson, Charles 2., eds. Or anizational Decision- Making. Englewood Cliffs, N.J.: Prent1ce-Hall, Inc., 1967. American Association of School Administrators. Profiles of the Administrative Team. Washington, D.C.: American Association of school Administrators, 1971. School Administration in Newly Reorganized Districts. Washington, D.C.: American Association of school Administrators, 1965. Anderson, James G. Bureaucragy_in Education. Baltimore: John Hopkins Press, 1968. Ayars, Albert L. Administering the Pe0ple's Schools. New York: McGraw-Hill Book Co., Inc., 1957. Bailey, Stephen K. The Office of Education and the Education Act of 1965 ICP #100. Findianapolis: The Bobbs-Merrill CO., Inc., 19662 , and Moser, Edith K. ESEA: The Office of Education Administers a Law. Syracuse: Syracuse University Press, 1968. Bartky, John A. Administration as Educational Leadership. Stanford, Calif.: StanfOrd University Press, 1968. Bernstein, Abraham. The Education of Urban POpulations. New York: Random House, 1967. Bidwell, Charles E. "The School as a Formal Organization." Handbook of Or anizations. Edited by James G. March. Chicago: RandiMcNally E Co., 1965. 239 Blalock, Hubert M. Causal Inference in Non-experimental Research. Chapel Hill, N.C.: UniverSity of North Carolina Press, 1964. . Social Statistics. New York: McGraw-Hill Book Company, 1960. Blau, Peter M., and Scott, W. Richard. Formal Or anizations. San Francisco: Chandler Publishing Company, 1962T5 Campbell, Roald F.; Cunningham, Luvern L.; and McPhee, Roderick F. The Or anization and Control of American Schools. Columbus, 0510: Charles E. Merrill Books, Inc., 1965. , and Gregg, Russell T., eds. Administrative Behavior in Education. New York: Harper and Row, 1957. Caplow, Theodore. Principles of Organization. New York: Harcourt, Brace 6 World, Inc., 1964. Carver, Fred D., and Sergiovanni, Thomas J., eds. Or anizations and Human Behavior: Focus on Schools. New York: McGraw-Hill Book Company, 1969. Cody, Wilmer S. "The Administrator Looks at His Practice." Freedom,lBureaucracy and Schoolipg, Edited by Vernon F. Haubrich. WaShington, D.C.: Association for Supervision and Curriculum Development, National Education Association, 1971. Corson, John J., and Paul, R. Shale. Men Near the Top, Baltimore: The John Hopkins Press, 1966. , and Harris, Joseph P. Public Administration in Modern Society. New York: McGraw-H111 BoOk Co., Inc., 1963. Crecine, John P. Governmental Problem-Solving, Chicago: Rand McNally 8 Co., 1969. Crozier, Michel. The Bureaucratic Phenomenon. Chicago: The University of Chicago Press, 19641 Dill, William; Hilton, Thomas L.; and Reitman, Walter R. The New gpppgers: Patterns of Behavior and Development. EngIewood 1 S, N}J}: *Prentice-Hall, Inc., 1962. Downs, Anthony. Inside Bureaucracy. Boston: Little, Brown and Co., 1967. 240 Etzioni, Amitai. Complex Or anizations. New York: The Free Press of Glencoe, Inc., 1951. , ed. Complex Or anizations: A Sociologlcal Reader. New York: Holt, Rine art and Winston, 1961. Fensch, Edwin A., and Wilson, Robert E. The Superintendency Team. Columbus, Ohio: Charles E. Merrill BOOks, Inc., 1964. Freeman, Linton C. Elementary Applied Statistics. New York: John Wiley 5 Sons, Inc., 1965. Gawthrop, Louis C. Bureaucratic Behavior in the Executive Branch. New York: The Free Press, 1969. Gerwin, Donald. Bud etin Public Funds. Madison, Wisc.: The University of Wiscons1n Press, 1969. Gibson, R. Oliver, and Hunt, Herold C. The School Personnel Administrator. Boston: Houghton Mifflin Company, 1965. Gittell, Marilyn, ed. Educatin an Urban Population. Beverly Hills: Sage Publications, Inc., I967. Participants and Participation. New York: Frederick A. Praeger, 1967. , and Hollander, T. Edward. Six Urban School Districts. New York: Frederick A. Praeger, l Golembiewski, Robert T. Behavior and Organizations. Chicago: Rand McNally 8 Co., 1962. . Organizing Men and Power: Patterns of Behavior and Line-Staff Models. Chicago: Rand McNally 8 Co., i967. Gore, William J. Administrative Decision-Makin : A Heuristic Model. New York: John Wiley ESOns, Inc., 196 . , and Dyson, J. W., eds. The Makin of Decisions: A Reader in Administrative Behavior. New York: The Free Press, 1964. Gross, Berthram M. Or anizations and Their Managing. New York: The Free Press, 1968. Hage, Jerald. "An Axiomatic Theory of Organizations." Organizations and Human Behavior: Focus on Schools. Edited by Fred 0. Carver andiThomas J. Sergiovanni. New York: McGraw-Hill Book Co., 1969. ' 241 Hays, William. Statistics for Psychologists. New York: Holt, Rinehart and Winston, 1963. Iannaccone, Laurence, and Lutz, Frank W. Politics, Power and Policy, Columbus, Ohio: Charles E. Merrill Publishing CO., 1970. Jacob, Charles E. Policy and Bureaucra_y, Princeton, N.J.: D. Van Nostrand COmpany, Inc., 1966. Kaufman, Herbert. The Forest Ranger. Baltimore: John Hopkins Press, 1960. Kepner, Charles, and Tregae, Benjamen B. The Rational Manager. New York: McGraw-Hill Book Co., 1965. Kerlinger, Fred N. Foundations of Behavioral Research. New York: Holt, Rinehart and Winston, Inc., 1967. Kimbrough, Ralph B. Political Power and Educational Decision-Making, Chicago: Rand McNally 6 Company, 1964. Kish, Leslie. "Some Statistical Problems in Research Design." The Si ificance Test Controversy. Edited by Denton E.- Morrison and Roman E. Henkel. Chicago: Aldine Publishing Co., 1970. Lindblom, Charles E. The Policy-Makin Process. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 19 . Lyden, Fremont J., and Miller, Ernest G. Plannin , Pro rammin , and Budgeting. Chicago: Markham Publish1ng Co., 1567. Luce, R. Duncan, and Raiffa, Howard. Games and Decisions. New York: John Wiley 8 Sons, Inc., 1957. March, James G., and Simon, Herbert A. Organizations. New York: John Wiley 8 Sons, Inc., 1958. Masotti, Louis H. Education and Politics in Suburbia. Cleveland: The Press of Western Reserve University, 1967. McNemar, Quinn. Ps cholo ical Statistics. 3rd ed. New York: - John Wiley E Sons, Inc., 1962. Mendenhall, William. Introduction to Probability and Statistics. Belmont, Calif.: Wadsworth Publishing Co., inc., 1967. Meranto, Philip J. School Politics in the Metropplis. Columbus, Ohio: Charles E. Merrill PEbliShing Co., 1970. 242 Meyers, Jerome L. Fundamentals of Experimental Desigp, Boston: Allyn and Bacon, 1966. Michigan Education Association. 1970-71 Employee Personnel Data in Michi an Public Schools. E. Lansing, Mich.: ’Michigan Educa- tion Association, 1971. Summary of Selected Ar ent Provisions 1970-71. E. Lansing, Mich.: Michigan Education Association, 1971. Michigan Reading Circle. Michigan Education Directory_and Buyer's Guide. Lansing, Mich.: Michigan Reading Circle, 1959-1970. Miner, John B. The School Administrator and Or anizational Character. Eugene, Ore.: The Center for the AdvancediStudy of’Educa- tional Administration, University of Oregon, 1967. Morrison, Denton E., and Henkel, Roman E., eds. The Si ificance Test Controversy. Chicago: Aldine Publishing Co., 1970. Niskanen, William A. Bureaucracy and Representative Government. Chicago: Aldine-Atherton, 1971. Peabody, Robert L., and Rourke, Francis E. "Public Bureaucracies." Handbook of Organizations. Edited by James G. March. Chicago: Rand McNally 8. Co., 1965. Press, Charles, and Arian, Alan. Empathy and Ideology: Aspects of Administrative Innovation. Chicago: Rand McNally 8 Co., 1966. Prewitt, Kenneth. The Recruitment of Political Leaders: A Study of Citizen Politicians. New York: BObbs-Merrili Co., 1970. Riker, William H. The Theory of Political Coalitions. New Haven: Yale University Press, 1962. Rogers, David. 110 Livingston Street. New York: Random House, 1968. / Rosenburg, Morris. The Lo ic of Survey Analysis. New York: Basic Books, Inc., 1968. Rosenthal, Alan, ed. Governin Education. Garden City, N.J.: Doubleday 6 Company, Inc., 1969. Peda O ues and Power. Syracuse: Syracuse University Press, 69. 243 Rourke, Francis E. Bureaucracy, Politics, and Public Policy. Boston: Little, Brown 6 Co., 1969. Bureaucratic Power in National Politics. Boston: Little, Brown 6 Co., 1965. Schlesinger, Joseph A. Ambition and Politics. Chicago: Rand McNally 5 Co., 1966. Schram, Hugh, ed. A Look at Your Job. Lansing, Mich.: Michigan Association of SChool Administrators, 1955. Stanley, David T. The Higher Civil Service. Washington, D.C.: The Brookings Institution, 1964. ; Mann, Dean E.; and Doig, Jameson W. Men Who Govern. Washington, D.C.: The Brookings Institution, 19671 Starbuck, William H. "Organizational Growth and DevelOpment." Handbook of Organizations. Edited by James G. March. Chicago: Rand McNally, 1964. Thompson, James D. grganizations in Action. New York: McGraw-Hill Book Co., 19 7. Thompson, Victor A. Bureaucracy and Innovation. University, Ala.: University of Alabama Press, 1969. Tullock, Gordon. The Politics of Bureaucracy. Washington, D.C.: Public Affairs Press, 1965. Umans, Shelley. The Management of Education. Garden City, N.J.: Doubleday E Co., Inc., 1970. Van de Geer, John P. Introduction to Multivariate Analysis for Social Sciences. San Francisco: W. H. Freeman and Co., 1971. Walker, Jack L. "Innovation in State Politics." Politics in the- American States. 2nd ed. Edited by Herbert Jacob’and Kenneth N. Vines. Boston: Little, Brown 8 Co., 1971. Walton, John. Administration and Policy-Making in Education. Rev. ed. Baltimore: John Hopkins Press, 1969. Warner, W. Lloyd, et al. The American Federal Executive. New Haven: YaIe University Press, 1963. Wildavsky, Aaron. The Politics of the Buggetary Process. Boston: Little, Brown 8 Co., 1964. 244 Zimmer, Basil G., and Hawley, Amos H. Metropolitan Area Schools. Beverly Hills: Sage Publications, Inc., 19681 111. Public Documents Michigan. General School Laws of Michigan. Statutes (1959). Supplement to the 1959 Revision of the General School Laws of Michigan. Statutes (1964). Michigan Department of Education. Rankin of Michi an Public Hi h School Districts by Selected Financial Data, 1965-67: Lansing, Mich.: Michigan Department of Education, 1967. Rankin Of Michigan Public High School Districts by_ Selected Financial Data,_l970-7l. LanSing, Mich.: Michigan Department Of Education, 1971. U.S. Office of Education, Department of Health, Education and Welfare. Manual for Project Applications and Granters, Title III, ESEA. WaShington, D.CT: GovernmentPrinting Office, 1967. IV. Unpublished Materials Boznanzo, Marcia M. "An Evaluation of Selected Title III Inservice Projects in Michigan as Perceived by Participating Teachers on Thirteen Variables." Unpublished Ph.D. dissertation, Michigan State University, 1968. Hain, Paul L. "The American State Legislators' Ambitions and Careers: The Effects of Age and District Characteristics." Unpublished Ph.D. dissertation, Michigan State University, 1971. Miller, Donald R. "An Analysis of Management Team Characteristics, Educational Characteristics Affecting the Quality of Education and Selected Cost Factors of Michigan School Districts." Unpublished Ed.D. dissertation, Michigan State University, 1970. Ulmer, S. Sidney. "Black, Longitudinal Behavior." Paper presented at meeting of the American Political Science Association, Chicago, Ill., September, 1971. MICHIGAN STATE UNIV. LIBRARIES \H\lNWUH“!WWImmWI‘HWIWIWI‘HIWll 31293103962555