...... FACULTY AND ADMINISTRATOR RERCEPTIONS OF POWER AND INFLUENCE IN UNIVERSITY DECISION - MAKING Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY BARBARA BRADLEY STONEWATER’ 1977 -vaL I (:2. f" 1“ E23061996 ‘05 06 05‘ AUG 052005 ( NOT/31!.)9‘7-ZOUI ABSTRACT FACULTY AND ADMINISTRATOR PERCEPTIONS OF POWER AND INFLUENCE IN UNIVERSITY DECISION-MAKING BY Barbara Bradley Stonewater Purpose The study of the perceptions of power and authority in institutions of higher education has received much attention in recent years, both because of the complexity of the uni- versity as an organization and the unique role that faculty play as professionals in a bureaucracy. This research exam- ined faculty and administrator perceptions of power and in- fluence at a large, public midwestern university. The fol- lowing research questions and hypotheses were explored: 1. How do faculty, as compared with administrators, perceive decision-making in the university? Hypothesis #1 - Compared with faculty, adminis- trators will perceive more faculty influence on academic matters. Hypothesis #2 - The difference between perceived and preferred influence will be greater for faculty than for ad- ministrators. 2. Are there predictable differences in perceptions of the pattern of influence among various sub- groups of faculty? Hypothesis #3 - Faculty in large, complex colleges will perceive more faculty influ- ence on academic matters than those in small colleges. Barbara Bradley Stonewater Hypothesis #4 - Faculty in units determined to be of higher quality will per- ceive more faculty influence on academic matters. Hypothesis #5 - Assistant Professors and In- structors will perceive admin- istrators to have more influ- ence than Associate Professors and Professors will. Hypothesis #6 - The difference between perceived and preferred influence will be greater for the two lower ranks of faculty than for the upper two ranks. 3. What changes in the faculty perceptions of decision- making in the university have occurred since the late 1960's and early 1970's? Methodology The methodology used in the study was a survey sent to 627 faculty from five colleges within the university and 288 academic administrators in the university. Additionally, data from a survey conducted at the same institution in 1970 were used to provide some descriptive comparisons between 1970 and 1977. The primary statistical tests used were the Chi- Square Test of Independence and Yule's Q, to examine relation- ships among frequency data on perceptions of power and influ- ence, and the one-tailed Z-Test, to test differences between means. Results The results of the research supported several, but not Barbara Bradley Stonewater all, of the research hypotheses. The comparisons between faculty and administrators revealed that on several key issues, administrators perceived faculty as having more influence than faculty themselves perceived. Regarding pre- ferred versus perceived influence, there was more difference between the two for faculty than for administrators on vir- tually all comparisons. This indicated higher "dissonance" for faculty, supporting the hypothesis. The next set of hypotheses dealt with comparisons be- tween several subgroups of faculty: large and small colleges, higher versus lower quality units, and the upper two faculty ranks versus the lower two ranks, The predictions with re- gard to differences among the ranks and among the higher and lower quality units were not substantiated to any extent by the results. However, both measures of the differences be- tween large and small colleges showed quite convincingly that faculty in larger colleges perceived themselves and faculty in general as having more influence than do their colleagues in smaller units. In comparing the 1970 and 1977 data, a descriptive analysis indicated tentative support for the prediction that 1977 faculty perceived themselves as having less influence than faculty did in 1970. Additional comparisons revealed that faculty in 1977 are older, at higher rank, less in :favor of collective bargaining, place more emphasis on Barbara Bradley Stonewater applied research, but see a variety of additional university issues in much the same way as they did in 1970. Issues for further study include the difference in perceptions of faculty and administrators, as well as large versus smell college faculty. The implications for effective university governance should be explored. FACULTY AND ADMINISTRATOR PERCEPTIONS OF POWER.AND INFLUENCE IN UNIVERSITY DECISION-MAKING By Barbara Bradley Stonewater A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Administration and Higher Education 1977 Q) (b‘IOQ‘I To Jerry, without whom I would not have begun and could not have finished, and to Jennifer, who made it worthwhile. ii ACKNOWLEDGMENTS A number of people have been most helpful in the com- pletion of this project. I would like to mention a few of them here. Dr. Ray White and Dr. Tim Little, members of my doctor- al committee, made contributions to the final draft that were much appreciated. Dr. Philip Marcus, thesis director, taught me a tremendous amount about survey research and about the study of organizations. I thank him for providing an excellent learning experience for me. To Dr. Lou Stamatakos, Chairperson of my committee, I owe a special thanks. Throughout my tenure as a doctoral student, he helped me clarify my goals, encouraged me as I worked toward them, and most importantly, believed in me. His support has been invaluable to me, personally and professionally. I would also like to express my appreciation to my colleagues in the College of Engineering. The Student Affairs staff encouraged me; the faculty and staff of the Department of Electrical Engineering and Systems Science were interested and gave their support when it was needed. Additionally, I would like to thank Frankie Allen, my typist and colleague. Her excellent skills, her commitment to the project, and her unfailing sense of humor helped me through the rough times. iii The final thanks goes to my family. My parents, Viv and Ray, always taught me to set high goals and to work hard to reach them. Their unrelenting support has meant more than they know. My in-laws, Jean and Stoney, took pride in my accomplishments and supported me through it all. Finally, I thank my husband, Jerry and our daughter, Jennifer for their love, their amazing tolerance, and for all that we share. ‘iv TABLE OF CONTENTS Chapter List of Tables I. II. III. IV. The Problem Introduction and Purpose Need for the Study Limitations of the Study Research Questions and Hypotheses Research Question #1 Research Question #2 Research Question #3 Organization of the Study Review of the Literature Introduction Definition and Sources of Power Organizational Power Power in Higher Education Summary Data Collection and Analysis Procedures Introduction Population and Sample Faculty Sample Administrator Sample The Survey Instrument Questions From 1970 Survey Development of New Questions Pretest Data Collection Procedures Data Analysis Procedures Data Analysis Introduction Questionnaire Response Comparison of Respondents to Population 1977 Survey Results Research Question #1 Hypothesis #1 Hypothesis #2 Research Question #2 Hypothesis #3 Hypothesis #4 Hypothesis #5 Hypothesis #6 Comparison With 1970 Data Research Question #3 Summary V. Conclusions and Recommendations Introduction Purpose and Need for the Study Results and Conclusions Hypothesis #1 Hypothesis #2 Hypothesis #3 Hypothesis #4 Hypothesis #5 Hypothesis #6 Comparisons: 1970-1977 Observations and Speculations Recommendations and Implications for Further Research Summary Appendices Appendix A - Cover Letter to Faculty Appendix B - Faculty Questionnaire Appendix C - Cover Letter to Administrators Appendix D - Administrator Questionnaire - Appendix E - Follow-Up Cover Letter to Faculty Appendix F - FolloweUp Cover Letter to Administrators Appendix G - Comparison of Faculty Respondents to Population for Each College on Variables of Rank, Sex, and Age Appendix H - Perceptions of Who Should and Who Does Have the Mbst Influence for Faculty and Administrators List of References vi 145 146 151 152 Table Table Table Table Table Table Table Table Table Table Table Table .10 .ll .12 LIST OF TABLES Sources of Power Comparison of Faculty Respondents to Population-Rank Comparison of Faculty Respondents to Population-Sex Comparison of Faculty Respondents to Population-Age Comparison of Administrator Respondents to Population-Rank Comparison of Administrator Respondents to Population-Sex Comparison of Faculty Respondents to Population-Position Comparison of Faculty Respondents, Population of Colleges Used, and University Population-Rank Comparison of Faculty Respondents, Population of Colleges Used, and University Population-Sex Comparison of Faculty Respondents, Population of Colleges Used, and University Population-Age Chi-Square and Yule's Q Coefficients for Comparisons of Faculty vs. Administrators for Who Has the Most Influence Z Scores and Rank-Order for Mean Comparisons Between Faculty and Administrators on Amount of Faculty Influence vii 62 63 63 64 64 65 66 66 69 73 Table Table Table Table Table Table Table Table Table Table Table Table Table .13 .14 .15 .16 .17 .18 .19 .20 .21 .22 .23 .24 .25 Rank-Order of Amount of Faculty Influence for Faculty and Administrators Chi-Square and Yule's Q Comparisons for Faculty vs, Administrators on Congruence and Dissonance Size of Faculty and Student Body by College Chi-Square and Yule's Q Coefficients for Comparison of Large vs. Small Colleges for Who Has the Most Influence Rank-Order Correlation for Large vs. Small Colleges on Amount of Faculty Influence Z Scores for Mean Comparisons Between Large and Small College Faculty on Amount of Faculty Influence Comparisons of the Five Colleges on Two Indices of Perception of Quality Chi-Square Coefficients for Comparison of Higher vs. Lower Quality Colleges for Who Has the Most Influence Rank-Order Correlation for Higher vs. Lower Quality Colleges on Amount of Faculty Influence Z Scores for Mean Comparisons Between Higher vs. Lower Quality Colleges on Amount of Faculty Influence Chi-Square and Yule's Q Comparisons of Lower Ranked and Higher Ranked Faculty on Who Has the Mbst Influence Z Scores for Comparison of Lower Ranked vs. Higher Ranked Faculty for Amount of Faculty Influence Chi-Square and Yule's Q Comparisons for Lower Ranked vs. Higher Ranked Faculty on Congruence and Dissonance viii 74 77 80 82 83 85 87 88 89 90 93 94 97 Table Table Table Table Table Table Table Table Appendix G .26 .27 .28 .29 .30 .31 .32 .33 Comparison by Rank for the 1970 vs. 1977 Faculty Respondents Comparison by Age for the 1970 vs. 1977 Faculty Respondents Comparison by sex for the 1970 vs. 1977 Faculty Respondents Comparison of Perceptions of Who Has the Most Influence for Faculty - 1970 vs. 1977 Comparison of Attitude Toward Collective Bargaining in 1970 and 1977 for Faculty and Administrators Rank-Order Correlation and Mean Comparisons for Salary Priorities - 1970 vs. 1977 Rank-Order Correlation and Mean Comparisons for Departmental Emphasis - 1970 vs. 1977 Mean Comparisons for Various University Issues - 1970 vs. 1977 Comparisons of Faculty Respondents to Population for Each College on Variables of Rank, Sex, and Age ix 99 100 100 103 105 107 108 109 146 I. THE PROBLEM Introduction and Purpose For over twenty-five years, social scientists and other researchers have studied power and control in organizations, its nature (Baldridge, 1971(a); Richman and Farmer, 1974; Dahl, 1957; and others), its source (French and Raven, 1960; Filley and House; 1969; Etzioni, 1961), and its distribution (Tannenbaum, 1968; Gross and Grambsch, 1968 and 1974; Blau, 1973; Baldridge, Curtis, Ecker, and Riley, 1973). They have examined a variety of organizations, including business and industrial organizations, voluntary organizations, unions, and educational organizations. But only within the last decade have the organizations of higher education come under careful scrutiny. A number of issues relate to the study of power in higher education, and one, faculty perceptions of power and authority, has received much attention (Blau, 1973; Gross and Grambsch, 1968 and 1974). The university, viewed as a complex organization, (Blau, 1973; Etzioni, 1964) places faculty in a unique position. As professionals, they differ from other workers or members of bureaucratic organizations because they are part of two systems -- their profession 2 and their organization -- which might have very different goals and control structures (Scott, 1966). Within their professions, they have developed a set of internalized standards, which impact their values and goals. In addition, they are subject to peer judgments, which are based on values of a discipline or the academic profession as a whole. As members of organizations, they are subject to the organizational structure and the hierarchy of authority that is part of that structure. Thus, it is possible for faculty to experience role conflicts in terms of their loyalties, priorities, and responsibilities (Blau and Scott, 1962). Also unlike workers in other kinds of organizations, faculty share and have direct involvement in the major goal activities of the institution, primarily teaching and research (Etzioni, 1964). Since faculty are a critical part of the accomplishment of these major institutional goals, and since they are subject to judgment from peers as well as organizational superiors, their place in the authority system of the institution is not all that clear. As a part of that, the place of faculty in the decision-making structure is also unclear, as it is not always the person at the top of the organizational chart who has the expertise or experience to make some of the major decisions. This unique position of faculty in the university hierarchy is certainly one factor that leads students of higher education to examine how that authority system works, ( lull! l UT '1' U U .II III? I l.l‘ll( l ‘1‘ TUTTI-“El! 3 and more specifically, how it is perceived by those most closely involved. Several additional factors have more recently impacted the significance of this issue. In the past several years, the concepts of accountability to funding agencies, systems approaches to budgeting, and quantification of institutional activities have led to different approaches to decision- making. In addition, administrative positions in the university hierarchy which once were held almost solely by former faculty are now being held in greater numbers of professional administrators, whose values and loyalties come from some basic principles of administration, rather than from the institution or a particular academic discipline. The approach of these professionals to the decisions they and others must make adds another factor to the complexities of the university as an organization. Thus, it appears that the university is a complex organization, with professionals in key roles who are subject to scrutiny from several forces and involved in some way in the key decisions of the university. It is also an organization with a variety of ways of setting, accom- plishing, and evaluating its goals, and one with different kinds of professionals making decisions that affect those goals. Consequently, the study of this authority hierarchy, the professionals in it, and the decision-making process is of interest to the student of organizations and of higher 4 education. One way of examining decision-making in the organization is to measure the perceptions of those involved. Thus, this research will examine faculty and adminis— trator perceptions of decision-making at a large public midwestern university. More specifically, the following research questions will be explored: 1) How do faculty, as compared with administrators, perceive decision-making in the university? 2) Are there predictable differences in perceptions of power and influence among various subgroups of faculty? 3) What changes in faculty perceptions of decision- making in the university have occurred since the late 1960's and early 1970's? Need for the Study The concepts of power and authority in higher education have been studied in recent years from a variety of perspec- tives. Probably the first to conduct substantive studies on power, Demerath, Stephens, and Taylor (1967), examined presidents, academic departments, and the state of a univer- sity under two different Chancellors. Subsequently, Gross and Grambsch (1968; 1974) and Blau (1973) have conducted major studies on, among other things, perceptions of power and influence across a large number of institutions. There have been a variety of less elaborate, more focused studies on power and influence in decision-making, but only two recent ones which examined a single institution (Adams, 1974; Stefferud, 1975). Though there have been a 5 significant number of research efforts on the question of power and influence, there have not been many substantial efforts in the last few years. The university has changed, as was alluded to above, in that the pressures of account- ability have given rise to new philosophies and systems of planning, execution, and evaluation of the business of the university. The major concerns of the institution have shifted from.war, peace, and institutional reform to voca- tionalism among students, severe financial problems, declining enrollments, and an over-supply of Ph.D.'s. Jobs, instead of people to fill them, are in high demand. Faculty, though concerned about the issues of the university, are 'also concerned about salaries, job security, and their rights as employees (Kemerer and Baldridge, 1975). With new pressures and changing priorities, the university is a different place than it was several years ago. Thus, it would seem that a study like the one proposed would give additional insight into the university as a complex organ- ization. Limitations of the Study The study will use as a sample faculty from five colleges within a large, complex university, plus those administrators classified as academic administrators. Because the faculty in the sample will not be chosen from the university as a whole, the generalizability of the data [UT lll’ I‘ll I I‘ll! Tll’T‘lU‘T-ll Ul‘l‘ ‘4'" |"{1 U I! IT I [II 6 to the university will be limited, and will depend in part on the similarity of the respondents to the university faculty. Conclusions drawn about any institution other than the one under study would probably not be valid. An additional limitation occurs in the attempt to explore the third research question above, i.e., what changes have occurred in perceptions of faculty since the early 70's. Though this study will replicate parts of a major study done at the same institution in 1970, the items dealing directly with perceptions of power and influence are not easily compared. Consequently, only descriptive analysis can be done on any comparisons. Research Questions and Hypotheses The three specific research questions stated earlier, and the related research hypotheses, were derived from several specific studies found in_the literature on power in higher education. These studies will be briefly reviewed here to develop the research questions and hypotheses for the study. Research Question #1 How do faculty, as compared with administrators, perceive decision-making in the university? One question addressed by students of decision-making in universities is whether faculty and administrators, because of their different orientations and positions in 7 the university, perceive the power structure differently. Certainly, extreme differences, if they exist, might have implications for the effective management of the institution. Faculty and administrators are both involved in the major activities of the institution. Vastly different perceptions of "reality" could certainly result in different approaches to problems. Gross and Grambsch, in their 1964 and 1971 studies, found generally similar perceptions of goals and of the power structure, among faculty and administrators. The major differences they reported were between "insiders", or administrators and faculty, and "outsiders", or legislators, funding agencies, and parents. A limitation on their findings, however, is that they studied comparative power rankings of groups within the university, and did not examine specific kinds of decisions (Gross and Grambsch, 1974). Several other findings or opinions lead to conclusions different from those of Gross and Grambsch. First, Kemerer and Baldridge (1975) have observed that as institutions get more complex and specialized, mpre administrators are required to perform the specialized functions. A recent study has, in fact, shown that the number of administrators per faculty member has increased from one for every five faculty to one for every four since 1972 ("Colleges Adding Administrators", 1977). Thus, faculty may perceive more administrative power solely because of numbers. On the 8 other hand, one of the phenomena that has brought about this specialization is collective bargaining, which many admin- istrators see as increasing certain kinds of faculty power rather than their own numbers or power (Kemerer and Baldridge, 1975). It appears that faculty and administrators may have different perceptions of power or influence based on their position or perspective. Several others have reached conclusions that would indicate that administrators see more faculty involvement, or decentralization of power, in academically related matters than faculty do. Kemerer and Baldridge, in their recent study (1975) of collective bargaining and academic senates, found that college presidents are likely to rate the influ- ence of faculty senates higher than do the faculty. Similar- ly, Hodgkinson found that administrators are generally more "euphoric" about the potential of faculty senates in the decision-making arena than are those who are involved with these senates on a daily basis (1974). Although Gross and Grambsch found, as was mentioned earlier, similar percep- tions among faculty and administrators, the one fairly significant difference they did find was that in 1971, higher level administrators ranked faculty several levels higher in the power structure than faculty rated themselves (Gross and Grambsch, 1974). Finally, in a related study of business organizations, voluntary organizations, and unions, Smith and Tannenbaum 9 found that the supervisors or officers tended to describe the organization in a much more democratic way than the members or workers (Smith and Tannenbaum, 1968). Though certainly a direct comparison cannot be made between faculty and workers or between administrators and officers, the findings can be used in support of the similar observations stated earlier. Thus, for this research, the following hypothesis will be tested: Hypothesis #1 - Compared with faculty, administrators will perceive more faculty influence on academic matters. In business organizations, Smith and Tannenbaum found that the difference between actual (perceived) and ideal (preferred) control was significantly greater for the workers than for the supervisors. This is understandable if in fact the supervisors have more impact on the power structure and a clearer picture of what the structure really is. Also, it is possible that the workers have some ambiv- alence about just what role they really want in the decision process, and thus while dissatisfied with what exists, they are reluctant to assume more involvement. That kind of ambivalence is in fact what both Blau (1973) and the Carnegie Commission Report ("Governance of Higher Education", 1973), ascribe to faculty. Both report that while many faculty indicate dissatisfaction with the power they have, they are reluctant to spend the time necessary in committees or other aspects of the decision process to increase that 10 power or influence. Based on the above, a second hypoth- esis is: Hypothesis #2 - The difference between perceived and preferred influence will be greater for the faculty than for administrators. Research Question #2 Are there predictable differences in perceptions of power and influence among various subgroups of faculty? The first major research question examined faculty and administrators as total groups, and compared perceptions of one group with those of the other. Though such compari- sons provide insight into the similarities and differences perceived by the two groups, it seems also appropriate to determine if within one group (here, the faculty), there are predictable differences in the way they perceive influence in decision-making, or in their satisfaction with it. Two variables of importance will be the unit, or college of the faculty member, and the faculty member's rank. In his book on the organization of work in 115 academic institutions, Blau (1973) reports on two types of authority. Bureaucratic, or legitimate authority, characterizes the central administrators, while professional, or expert authority characterizes the faculty. Blau examined univer- sity conditions that led to centralization, or the predom- inance of bureaucratic authority, and decentralization, or the prevalence of professional authority, in two decision 11 areas: educational policies and faculty appointments. He looked at several structural characteristics of the institu- tion to determine which could predict the kind of authority structure, and found several that indicate decentralization of authority in academic matters: large size, highly qualified faculty, large clerical/faculty ratio, multi- level hierarchy, prevalence of research, high faculty salaries, and high rate of succession of top administrators, (Blau, 1973, Ch. 7). Similarly, Baldridge, et. al. (1973), found, in their study of 249 colleges and universities, that large size and complexity clearly indicated higher faculty autonomy in the decision areas studied. Both of these studies focused on the institution as the unit of analysis, and were based on faculty views of decision-making in their particular institution. Consequently, transferring and applying those findings to one large institution and using the college as the unit of analysis is somewhat difficult, primarily because the study will examine faculty perceptions of decision-making in the university, and not specifically in their unit or college. However, the findings cited lead to the generation of hypotheses for testing in a different setting. Consequently, the following hypothesis emerges: Hypothesis #3 - Faculty in large, complex colleges will perceive more faculty influence on academic matters than those in small colleges. Another characteristic of units where faculty perceived greater involvement in decision-making in the Blau study was 12 higher quality faculty, a characteristic most difficult to measure. Blau used several indicators of faculty quality, including high salaries and research involvement. Since in this study five units representing vastly different disci- plines with very different emphaSes will be examined, comp parisons on salary levels and amount of research would be compounded by other variables. It is most difficult to obtain an external index of quality. Consequently, for this research, an internal assessment of quality will be determined, based on perceptions of the faculty and admin- istrators in the study, for the five units involved. The description of this assessment of quality will be described along with the research results in Chapter IV. Though the determination of quality of an academic unit will be some- what different from.Blau's determination, in order to gain additional information from the data, a second hypothesis regarding differences among units will be tested: Hypothesis #4 - Faculty in units determined to be of higher quality will perceive more faculty influence on academic matters than those in units determined to be of lower quality. Regarding the other characteristic under consideration, faculty rank, several studies address perceptions of the power structure. Blau (1973), Ladd and Lipset (1973), and Kemerer and Baldridge (1975) have documented that faculty at the higher ranks generally are more influential and involved in key decisions compared to their colleagues at 13 lower ranks. Ladd and Lipset discuss the separateness that younger, lower ranked faculty feel from their older colleagues, particularly when they have not achieved tenure, and have shown that these more alienated faculty give the strongest support for collective bargaining. Though Kemerer and Baldridge (1975) describe ways in which it is possible for a faculty government and a union to coexist, it is clear that certain kinds of faculty, usually the younger and lower ranked ones, are the strongest advocates for collective bargaining and are those who feel most alienated by the faculty governance system. Consequently, two hypotheses are: Hypothesis #5 - Assistant Professors and Instructors ‘will perceive administrators to have more influence than Associate Professors and Professors will. Hypothesis #6 - The difference between perceived and preferred influence will be greater for the two lower ranks of faculty than for the upper two ranks. Research Question #3 What changes in the faculty perceptions of decision- making in the university have occurred since the late 60's and early 70's? Since the violent and vocal period on American campuses of the late 1960's and early 1970's, a variety of obvious changes have taken place. Concerns over Vietnam.and involve- ment in the decision process of the university have given way to increased vocationalism.among students, and adminis- trative problemp have shifted from keeping peace on the l4 campus to keeping students and trying to maintain quality in the severest of financial situations. In addition, faculty unions have brought a variety of changes to the campus. Collective bargaining is clearly a phenomenon of the 70's. In 1966, eleven campuses were unionized, a figure which grew to 160 in 1970 and 430 in late 1975 (Kamerer and Baldridge, 1975). These changes have no doubt affected faculty and their involvement in the decision- making process. The conclusion of Kemerer and Baldridge is that though the recent collective bargaining movement has complicated the decision-making process, administrators do not appear to have lost power, and potentially may be rewarded with more control. They point out that financial stringencies have forced many governing boards to take a more active role in campus affairs, and that centralization of policy-making at many institutions, whether unionized or not, has moved from departments, to colleges, to central administration and governing boards. In addition, faculty collective bargaining helps to push power upward, as trust- ees and administrators act as "employers", and reclaim some of the power once granted to faculty senates in order to have an effective bargaining position with faculty unions (Kemerer and Baldridge, 1975). In addition, Baldridge, et. a1. (1973) explain the growth, and now possible decline, of faculty power. In the 1960's, expanding enrollments, public support of higher 15 education, and a demand for qualified faculty put faculty in an excellent bargaining position with administrators. In very recent years, however, leveling of enrollments, reduced public support of higher education, and the over- supply of Ph.D.'s have given us reason to predict a weak- ening of the strong and powerful position held by faculty. For the above reasons, it would seem appropriate to compare the perceptions of faculty and administrators at the institution under study as they have changed since 1970. As was mentioned earlier, portions of a study done at that time will be replicated in the proposed study. However, the items used in 1970 to assess perceptions of power and influence are not comparable to questions in that area that will be used in the current study. Consequently, direct statistical comparisons will not be made. There will be other items on attitudes and perceptions of other issues in the university that can be compared, however, so that some non-statistical comparisons will be made. It is hoped that some descriptive conclusions can be obtained about perceptions on power and influence. Because of the methodological difficulties, no specific hypotheses will be examined, though the above discussion of recent changes in higher education would suggest that faculty will not see themselves as having as much influence now as in 1970. The data will be examined in an exploratory way and observations made will be reported. 16 Organization of the Study Building on these research hypotheses and expanding on the literature briefly referred to here, Chapter II will contain a review of the literature on power and authority in general and power in higher education. Chapter III will be a description of the methods employed and organization of the reserach. Chapter IV will contain the results of the study, and in Chapter V, the researcher will summarize the study and present some interpretations, observations, and suggestions for further study. II. REVIEW OF THE LITERATURE Introduction Two important goals guided this literature review: to understand more about the concept of power, and to survey the research done on power and authority in higher education. Since much of the work on power in higher education has roots in previous work done by social scientists on power in non-educational organizations, a thorough review must also include a more general look at organizational power. Consequently, the three major divisions of this review are: 1) Definitions and Sources of Power 2) Organizational Power 3) Power in Higher Education Definitions and Sources of Power The key to the understanding of power and authority in institutions of higher education is the survey and under- standing of the nature of power as a concept: its definition and its sources. Many theorists and researchers over the years have attempted to define the concept of power, and although the perspective of these definitions may vary slightly, the definitions in their basic form are very similar. In an 17 18 early classic on the theory of organizations, Weber (1947) defined power as "the probability that one actor within a social relationship will be in a position to carry out his own will despite resistance, regardless of the basis on which this probability rests" (p. 153). Dahl's (1957) definition, "A has power over B to the extent that he can get B to do something B would not otherwise do" (p. 202-3), captures an essential characteristic of power - that it is a relational concept, i.e., that it involves interaction between two or more persons or groups, and that it is mean- ingless unless exercised (Hall, 1972). Though other writers have developed alterations on these definitions, and some have introduced the terms control, authority, and influence, 1 most of the definitions are similar. In addition to merely stating definitions of power or its related terms, another valuable way of viewing power is by the descriptions of where power comes from, or its sources or bases. Again, one of the basic management-administration con- cepts comes from Weber (1947), who differentiates power from authority. Power implies a kind of coercion, or force, whereas authority possesses legitimacy, or the right to give directives. Weber's three-fold typology of authority is of most interest here: 1) traditional, or adherence to and 1For other definitions of power and similar terms, see Filley and House (1969), Tannenbaum (1968), Etzioni (1964), Peabody, (1962), and Kaplan (1964). l9 belief in long-held traditions, and the legitimacy of status determined by such traditions; 2) charismatic, or devotion to a particular person and recognition of normative patterns of behavior resulting from this devotion; 3) legal- rational authority, with roots in legitimate patterns of rule. Since Weber, several writers have attempted to develop fairly simple typologies of the kinds or sources of power useful in organizations. Perhaps the best known are the five bases of social power described by French and Raven (1968). These are: l) Reward power - based on the ability to reward; 2) Co- ercive power - involves the ability to manipulate sanctions; 3) Legitimate power - based on the assumption of a legiti- mate right to prescribe behavior; 4) Referent power - based on a feeling of identity with another person; and 5) Expert power - based on the perception of expertness or extent of knowledge. Etzioni's typology, somewhat similar to French and Raven's, describes three types of power: coercive, remun- erative, and normative (Etzioni, 1961). Coercive, like the similar base of power of French and Raven, is based on the ability to utilize sanctions. Ramunerative, or utilitarian power is similar to reward power. Normative, or social power is based on the manipulation of symbolic reward and the influence of a particular person, and is close to the 20 referent power described above. Etzioni eliminates expert and legitimate power, and says that legitimate authority is based on one of the three power bases (Hall, 1972). Two other similar descriptions of bases of power are those of Kaplan (1964) and Peabody (1962). Kaplan's four kinds of power are almost identical to French and Raven's: coercion, reward, reference or identification, legitimacy, and expertness. Peabody, like Weber, uses the term author— ity in his breakdown of types, and labels two formal bases of authority and two functional bases. The formal.types of authority are legitimate and position, with position authority being a kind of combination of the reward and coercive power described by others. His functional types are competence (expert) and person (referent) authority. To review the four similar typologies described, it is useful to look at the labels given the types or sources of power in a table with similar types listed horizontally. Table 2.1 Sources of Power French & Raven Kaplan Peabody TEtzioni l. Reward l. Reward l. Remunerative 2. Coercive 2. Coercion 1' Position 2. Coercive 3. Expert 3. Expertness 2. Competence 4. Le itimate 4. Legitimacy 3. Legitimate 5. Re erent 5. Reference 4, Person 3. Normative In addition to these fairly simplistic concepts, Filley and House (1969) review the different kinds of power described ITIEII I'lII Illl'l'vlll'll'l 21 by a number of other writers, some of which designate as many as ten or twelve kinds, or bases. Some of these are: formal position, rewards and sanctions, social approval, physical power, manipulation, control of information, personal affection, control of property, prestige, popular- ity, bureaucratic rules, etc. In reviewing all the designa- tions given, however, it seems that they could quite easily all be placed in one or the other of the categories of French and Raven. In summary, then, of definitions and sources of power, it seems that power is a relational concept involving two or more persons or groups and some kind of exercise of that power. Some definitions use control, influence and authority synonymously, while others see distinctions, particularly 'with the term authority. Power is seen as coming from.many sources, but the most common seem to be legitimate position, expertness or knowledge, reference or identification with a particular person, and ability to manipulate rewards and punishments. With the background on the nature and sources of power, influence, and authority, the second topic for review was to examine studies of power in an organizational setting. Organizational P0wer This section will review some of the various approaches to studies of power and authority in organizations. The 22 research on organizational power is critical to the under- standing of power and authority in institutions of higher education, but because power in universities is central to this research, studies relating directly to power in higher education will be reviewed in a separate section. Several books written about organizations in general have sections on power most helpful in forming basic con— ceptions about organizational power. Etzioni (1961), as mentioned earlier, distinguished three kinds of power - coercive, remunerative, and normative. He then describes the kinds of organizational involvement that accompany these three kinds of power. In organizations using coercive power, the involvement of the lower participants is class- ified as alienative. In those using remunerative power, their involvement is called calculative, and in those utilizing normative power, the involvement is moral or social. Some consider Etzioni's description of power, and the involvement characterizing it, too simplified, saying that it does not deal with the complex nature of power held by both elite and lower participants in an organization (Hall, 1972). A collection of useful studies is found in Tannenbaum's book on organizational control (1968). It is the first research-based volume on control, and has part of its value in the fact that it is a collection of related studies utilizing the same underlying concepts. Although much of 23 the book deals with control in organizations as it relates to other variables, such as satisfaction and performance, it is useful both conceptually and empirically, The "control graph", used and applied in studies throughout the book, illustrates both the total amount of control in an organization as well as the distribution of control, and is helpful in the quantification and description of the research data. Several of the studies done by other research- ers and contained in the book have been or will be referred to elsewhere in this research. That the French and Raven (1968) typology mentioned earlier has been important in subsequent work on organiza- tional power is evident in the fact that several studies have been based on their five "bases". Bachman, Bowers, and Marcus (1968) summarize the results of five separate studies designed to determine why organizational members comply with the wishes of their supervisors. The data were obtained from.2,840 respondents in the following settings: 1) salesmen in 36 branch offices in a firm selling intang— ibles: 2) faculty in 12 liberal arts colleges; 3) agents in 40 insurance agencies; 4) production workers in 40 appliance company work groups; and 5) semiskilled workers in 20 utility company work groups. Satisfaction and organ- izational effectiveness were also studied, but are not of prime importance here. In all units, the most important reasons for complying with organizational superiors were 24 legitimate and expert power, with referent and reward power of lesser importance and coercive being the least likely reason. The units in which the greatest differences in kinds of power were found were the colleges and utility companies. In the former, expert power was the most important reason for complying, with coercive the lowest. In utility companies, legitimate power ranked highest and referent the lowest. warren (1968) also used the French and Raven typology in his study of power and conformity of school teachers. He found that in most settings, more than one form of power was used, with expert and referent most often found together and coercive standing alone. He also found that in highly professional settings, coercive power was weak with legitimate the strongest, and also that certain kinds of individual conformity are linked to differ— ent bases of power. All of the above studies indicate that in most settings, more than one kind of power is utilized to bring about certain behaviors in organization members, with a variety of factors determining the effects of the kinds of power. An additional study on the relative importance of different kinds of power was done by Peabody (1962) who attempted to distinguish between formal authority (legit- imacy and position) and functional authority (competence and person). These categories were described previously. He studied 76 members of three public service organizations: 25 welfare workers, school employees, and police officers. In all three organizations, considerable importance was attached to legitimacy and position, though police singled out person authority and school workers, competence. Though all of the above studies have focused on interpersonal power, Hall (1972) indicated that we ought also to consider the power of subunits in an organization. One study which examined power among groups attempted to determine the bases of power of twenty-eight units within seven manufacturing organizations (Hinings, et. a1., 1974). This was an attempt to test Hickson, et. al.'s "strategic contingencies" theory of intraorganizational power (1971). This theory suggested four bases of subunit power: 1) coping with uncertainty - the more a unit copes with uncertainty, the higher its power; 2) substitutability - the lower the substitutability of the activities of the unit, the greater its power; 3) centrality - the higher the pervasiveness and immediacy of the workflow of a unit, the greater its power; and 4) control of contingencies - the more contingencies that are controlled by the unit, the greater its power. The test by Hinings, et. al. of this theory revealed that coping with uncertainty is the variable most critical to power and the best single predictor of power, but not the only factor. The next most important variables, in order, were immediacy of workflow, non-substitutability, and pervasivenss of workflow. A similar result was found in Perrow's (1970) 26 study of four departments in each of twelve companies - sales, production, research and development, and accounting or finance. Perrow was not dealing with all the complexi- ties of the power question, but was asking only which depart- ment had the most power. His results showed that sales dominated in eleven of the twelve organizations, and he concluded that it was because sales absorbs most of the un- certainty about the diffuse and changing environment of customers. Customers, he says, determine organizational success, since they largely determine the nature and quantity of the product to be sold. Though all units contribute to customer satisfaction, sales has the most contact with this critical group. Several additional studies dealing with power of subunits in institutions of higher education will be described in a subsequent section. In summary, it appears that the study of organizational power has been facilitated by the efforts of the various writers to distinguish different kinds of power, and in a number of studies it appeared that more than one kind of interpersonal power often surfaced in a particular organiza- tion, with the nature of the personnel a determining factor in which kinds were employed. In studies of power among subgroups in an organization, the consistent finding was that units performing functions most critical to the organ— ization are the ones with significant amounts of power. Before turning specifically to research on power in higher education, some comments are in order regarding the 27 difficulties in transferring the conceptualizations of organizational power to collegiate institutions. Though many of the writers cited previously have acknowledged the complexities of organizational power and have distinguished several different kinds of power, in most of the organiza— tions studied, there is a clear hierarchy within the formal organization. What this means is that in these organizations, the legitimate authority clearly rests with those at the top of the chart, though there are, of course, other kinds of power and influence throughout the organization. When universities are examined, it is apparent that the largest group of organizational employees, the faculty, do not fit neatly into such a hierarchical system. Indeed, Blau and Scott (1962) and Scott (1966) have clearly indicated that faculty, as professionals, must not only respond to the organizational control structure, but also to the controls and standards of their profession. When professionals enter a bureaucracy, they must face the issue of identity with the organization or with their profession, an issue which often brings about role conflict. Also, as Etzioni (1964) has pointed out, faculty are clearly involved in the major goal activities of the university, teaching and research. Cere tainly the professional, or expert power held by faculty is of critical importance because of the significant place they hold in the organizational structure. Both Etzioni (1964) and Blau (1973) emphasize the differences and potential con- flicts between bureaucratic power, or legitimate authority, 28 and professional power. Nowhere are these two kinds of power in closer contact or potential conflict than in institutions of higher education (Blau, 1973). It is within this framework of the significance of professionals in a bureaucracy and professional power that a review of power and authority in higher education is in order. Power in Higher Education In reviewing the literature on power in institutions of higher education, one finds several attempts to describe power relationships in terms of a model of university gov- ernance, several major studies done on power and influence, and a variety of studies focusing on some particular aspect of power in higher education. Until the late 1960's, there were two major approaches to university governance that dealt specifically with power and authority. One was Max weber's bureaucratic model, which focused on the formal organization, hierarchies, rules, and procedures (Weber, 1947). Indeed, Weber's focus on authority as being legitimized power as mentioned in the previous section, is indicative of this approach. The other, the collegial model, was based on the idea of a community of scholars and full participation in decision—making (Baldridge, 1971a) and became popular with those who were disenchanted with bureaucracy. This model advocated decision-making within the congenial professional group of faculty. Though 29 both of these approaches have their merits, they also have limitations. Baldridge (1971a) points out that the bureau- cratic model, while encompassing many of the features that indeed do exist in the university, only explains legitimized power, or formal authority, and does not deal with informal power, or power based on expertise or personal influence. On the other hand, Baldridge says, the collegial approach is really a utopian approach that is fairly ambiguous and neg- lects to deal with "what is," or with conflict and the "politics" of academic decision—making. Consequently, Baldridge attempted to develop a model of university govern- ance that he felt more appropriate to decisionwmaking in the complex, political arena of the university. Thus, in con- nection with his analysis of decision-making at New York University in 1968, he developed a new "political model" of academic governance (Baldridge, 1971a; 1971b). Essentially, this model assumes conflict in the university to be a natural phenomenon, and takes into account informal interest groups and power blocks. Baldridge has also found that for- mal authority can be severely limited by political pressure, bargaining, and negotiations between competing groups. The development of this model is seen by many as an important contribution to the literature (Richman and Farmer, 1974). A fourth contribution to the conceptualization of decision-making in the university is the model of "organized anarchy,” developed by Cohen and March (1974). They studied 30 the presidencies of forty-two colleges and universities of different types, and concluded that power is ambiguous, diffused, that participants vary from one issue to another, and are often more concerned with their own importance than with the content of decisions. They see institutional goal systems as ambiguous and ill-defined, and are skeptical about how much effective long-range planning can be done. To enable the administrator to survive in an organized anarchy, Cohen and March present eight "tactical rules" designed to aid in decision—making in the university. Demerath, Stephens, and Taylor (1967) present the results of another group of studies on universities. The first study represents qualitative material on the presi- dencies at forty-five major institutions. The major findings were that there were several diverse and often inconsistent roles to be played by the president, and the adjustments necessary to conform to these roles often strain the rela- tionships between the president and those around him. The second study, an analysis of administrative style at the University of North Carolina under two different Chancellors, included a survey of faculty administered under both, and showed that faculty ratings of the general excellence of the university, their own professional and personal satisfaction, and their influence over educational policies all increased during that time period. Though there were several factors that could have caused those changes, the authors concluded 31 that the change in the chancellor's style, which included streamlining policies and procedures as well as increasing faculty participation, was a major factor, The study showed that it is possible to combine hierarchical authority and faculty participation into an effective administrative style. The final part of the Demerath, Stephens and Taylor study was done on thirty academic departments at five major universities. The 211 respondents were asked questions dealing with power and esteem of their colleagues. The findings revealed that while department chairmen had the most overall power in fourteen policy areas described, the more important the policy area in the minds of the faculty, the fewer the chairmen that occupied the tOp power positions. In a study related to this last part of the Demerath, Stephens, and Taylor studies, Schuh (1975) studied deans of liberal arts colleges in state and land-grant schools. Res- pondents were asked to indicate what kind of influence various people in the academic hierarchy had on twenty-one specific issues. The results showed a vital role played by faculty and department chairmen in the phase of initiation and implementation of policy, but the dean having the primary role in policy approval. This study did not classify issues by importance, as did the previous study, but both studies indicated that perhaps the power of the chairmen may be limited in key policy decision areas. Hill and French (1967) also studied the perceived power 32 of department chairmen, but from the perspective of its relationship to faculty satisfaction and professional output. They found that the greater the perceived power of the chair- man, the greater the satisfaction of faculty, with little relationship of chairman power to professional output. Another important collection of data on university goals and academic power is the work done by Gross and Grambsch (1968; 1974). In 1964, they conducted a survey of faculty and administrators at 68 universities on institution- al goals and academic power. After the disruptive events on American campuses of the late 60's, Cross and Grambsch decided to replicate their study to determine if those events had caused any significant changes in the goal or power structures. Since the research on power is of more relevance here than the data on goals, only those results will be described. The faculty and administrators in the sample were asked "how much say" each of sixteen groups had over major decisions in the university. As was expected, presidents came out on top, with regents, vice-presidents. and deans of professional schools next in that order. Students ranked fourteenth out of sixteen in the 1964 study. In 1971, the researchers found that the relative positions had not changed much. Of the top six power holders, only the dean of liberal arts (fifth in 1964) and the dean of the graduate school (sixth in 1964) had changed positions. Several other groups 33 changed slightly, with students moving up to twelfth (see Gross and Grambsch, 1974, p. 122). Another interesting finding is that when the results are broken down by kind of control (public vs. private institution) it is clear that groups like legislators and the state government rank con- siderably higher in public institutions, with sources of grants and endowments ranking high in private universities. In fact, in 1971, legislators ranked fourth in overall power in public universities. One last finding is that in 1971, the respondents were again given the same sixteen groups and asked, on a five—point scale, to what degree the group's power had increased or decreased since 1964. When trans- lated into numbers, the results showed that the respondents perceived most groups to have increased in power since 1964, and few to have decreased. What this indicated is that in a quantitative sense, respondents saw an overall increase in the amount of power. That question also showed that those whose power was seen to have increased markedly were legis- lators, regents, state and federal governments, as well as students and faculty. The power of most administrators remained the same, except that the power of department chairmen decreased. A study that provided a good deal of data on how structural characteristics affect power and authority is the one cited earlier by Blau (1973). He studied 115 colleges and universities to determine the general 34 relationship of the administrative process to the nature of academic work. Blau had three sources of data for his study, which included a representative sample of all four— year schools granting liberal arts degrees in the 1960's. His data sources were: 1) interviews with two central administrators, usually the president and vice president or assistant to the president, plus self-administered question- naires sent to academic deans; 2) published compilations of quantitative information on the various institutions; 3) a survey of 2577 faculty members from the same 115 institutions conducted by Talcott Parsons and Gerald Platt in 1967 through the University of Michigan Survey Research Center. As was stated in Chapter I, Blau was interested in determining what structural characteristics of the institu- tion indicated the prevalence of bureaucratic, or adminis- trative power, and professional, or faculty power. He did this by an analysis of two particular decision areas: educational policies and faculty appointments. Blau used a number of variables, including size, verti- cal differentiation, horizontal differentiation, formal division of labor, and administrative apparatus. He decided to look at distribution of power in the areas of educational policies and faculty appointments because they are areas where power is not clearly assigned to any one particular group. Though several factors were different for the two 35 decision areas, his general conclusions were that large size, high clerical/faculty ratio, high quality faculty, and visible research accomplishments were the major factors leading to decentralization, or the prevalence of profession- al authority. The main conclusions of the study were that large institutions tend to be less bureaucratic than small ones, and often, because they are more decentralized, offer faculty more opportunities for significant authority. Blau says that several so-called bureaucratic characteristics actually reduce centralization of authority, and if there are negative effects of bureaucracy, they affect the teach— ing, rather than research function. A similar study was done by Baldridge, et. a1. (1973). Under the auspices of the Stanford Center for Research and Development in Teaching, they undertook a study of decision- making and academic autonomy in 249 colleges and universi— ties. Utilizing questionnaires and a variety of printed material, they examined the areas of l) peer evaluation, 2) department autonomy, 3) freedom from administrative work regulations, and 4) overall decision centralization. They had previously correlated various measures of size and complexity, and so used faculty size for simplicity. Their results showed large size to clearly indicate high faculty autonomy in the first three areas. In the fourth, overall decision centralization, all sizes of institutions indicated generally decentralized decision-making. However, large 36 institutions more often than smell ones, indicated central- ized decisions at the departmental level. At first thought to be a surprising conclusion, the investigators, on further examination, offered three explanations for this departmental decision centralization at large schools. First, because decisions are generally highly decentralized in large institutions, departments need to be making important deci- sions. Central administrators may be more willing to relinquish these decisions if there is a strong department head, for accountability. Secondly, large schools generally have larger departments, thus necessitating the centraliza- tion of decisions in a smaller percentage of the department members. Third, it is possible that departments are willing to invest a good deal of power in a chairperson in order to protect the departmental autonomy from central administrators. Even with the similar results on overall decision central- ization, the general conclusions were that large size does indicate more faculty autonomy, and that large, complex institutions may give faculty more opportunities to partici- pate in academic decision-making. The results of these two recent studies perhaps give students of the decision-making structure in higher education a new perspective on the effects of large complex institutions on faculty involvement. Turning from structural effects on power and authority, three studies will be reviewed which, like several mentioned 37 in the previous section, use, at least in part, the French and Raven typology as the base of interpersonal power (French and Raven, 1968). The first is a study by Bachman (1968), which was included in the synthesis of five similar studies mentioned earlier (Bachman, Bowers, and Marcus, 1968). Bachman obtain- ed data from 685 faculty members in twelve liberal arts colleges. He examined amounts of administrative influence, bases of influence for the dean and faculty, and levels of satisfaction. His general results indicated that faculty satisfaction was greatest in those colleges having the high- est total influence across all levels of the academic hierarchy, and also is greatest with deans who have relative- ly strong influence in academic decision-making. Regarding the bases of power and influence, the strongest reasons for faculty compliance with requests of the dean were expert, legitimate, and referent power, with reward and coercive the weakest. The bases of faculty power over the deans, on the other hand, were legitimate, expert, and referent, in that order. Apparently, the deans felt strongly that faculty had a legitimate right to influence them. Additionally, two similar studies were done at individ— ual large universities that used French and Raven as a basis (Adams, 1974; Stefferud, 1975). Adams studied the bases of power and amount of control exercised by the faculty, dean, and department chairmen in 20 large departments at a 38 ‘midwestern university and the effects of the various control structures on job satisfaction and professional output of faculty. (The relationships to satisfaction and professional output will not be reviewed here, as they are not central to the purposes of this research.)' Adams found that all three (faculty, chairmen, dean) have effective control over the department and over one another. Chairmen are perceived to have the greatest overall influence, but not at the expense of the other two, as the influence is definitely tied to decision areas. In teaching and research activities, for example, faculty have the most influence, and with department faculty personnel policies, all three share power. An over- all conclusion reached by Adams is that the most typical control structure seems to permit effective decisions to be made by those who have primary interest vested in them. Regarding the power bases, Adams found expert, referent, and. legitimate to be the most important. with expert and referent the most important where all three groups had high levels of influence. Reward and coercive power, not sur— prisingly, were important where faculty input was low. In another study done at a single large university, Stefferud (1975) studied two separate questions. First, he examined the sources and intensity of power or influence perceived by faculty and students. A large number of separate hypotheses were tested, with the most important ones for the present research being those involving the 39 perceptions that faculty have of the kind of power that deans, chairmen, and other faculty have over them. The results showed that faculty see deans as primarily having reward power over them” with legitimate power second, then coercive. Faculty see chairmen as using coercive and reward power the most, with referent third, and legitimate and expert last. Finally, faculty see their faculty colleagues as having expert power primarily, with reward second and legitimate last. Contrary to the previous studies (Adams, 1974; Bachman, 1968) where reward and coercive power seemed relatively weak, in this study, reward appears to be the most important kind of power used as perceived by faculty, with legitimate power the least utilized. Indeed, when the . responses to the three questions are averaged, reward and coercive power rank first and second. In the second part of the study, Stefferud attempted to determine where faculty, students, and administrators differed in perceptions of who does and who should have authority to make certain decisions, and also to determine a general level of satisfaction with the decision-making process. Regarding who does and who should have power, Stefferud designated 16 different decision situations and 12 different individuals or groups who could have power over the decisions.‘ Respondents were asked to mark all that were appropriate for each category and each individual or group. He reported only percentages, as the data were too 40 cumbersome to determine statistical significance. Looking at overall percentages. it did not appear to Stefferud that faculty and administrators saw the decisionwmaking process very differently. On several specific issues. the two groups perceived some difference in the involvements of certain peOple or groups, with the most significant being in the determination of course content. Faculty clearly felt faculty are and should be involved, while administrators indicated that the department chairman and the dean should have as much to say as faculty. In areas such as promotion and salaries, hiring and firing faculty, and determining department budgets, faculty and administrators had similar perceptions overall of who does and should have power. Additional information gained by Stefferud in this second part of his research was an overall index of satis- faction, obtained by determining the overall differences or similarities for students, faculty, and administrators between who does have power and who should. The indexes obtained were 70.3% for administrators, 66.7% for faculty, and 42.4% for students. These figures were interpreted by Stefferud to mean administrators were the most satisfied with the decision-making structure, with faculty a close second, and students the least satisfied. Again, it must be empha- sized that results for this second section of Stefferud's research were nOt obtained by any statistical procedures. Turning from.the bases of individual power, there are 41 two studies which deal with the power of subunits within the institution. Salancik and Pfeffer (1974) hypothesized that subunits will possess relatively more power to the extent that they provide resources valued or critical to the organization. After assessing subunit power, as well as establishing the relative importance of certain resources, they concluded that quality graduate education and acquisi- tion of outside research funds are the most valued resources, and the leading departments in those areas were the units with the most power. Thus, those two resources were deter- mined to be the best predictors of subunit power. In a related study, looking more specifically at the university budget process, Pfeffer and Salancik (1974) examined the power of various subunits, using measures similar to those used by Perrow (1970). They asked department heads to rate departments on overall quality and also looked at memberships on key committees. The results generally showed that the more powerful the subunit, the larger portion of the budget it received, indicating perhaps the complex, political nature of budget decisions in institutions of higher education. No analysis of literature on power in higher education would be complete without a brief mention of several of the current books on faculty collective bargaining. Though the detailed conclusions are not relevant here, it is worth reviewing the implications of unionism for faculty governance and perceptions of power and influence. 42 Three of the major books, Garbarino (1975), Ladd and Lipset (1973), and Kemerer and Baldridge (1975) all made reference to the inherent conflict between unions and the faculty senate type of governance, indicating that a clear separation of territory is going to become more and more difficult, and that faculty influence on unionized campuses has taken a very different form than on those campuses where senates, or some other similar body, prevail. It is very possible, according to Kemerer and Baldridge, that unions will force administrators into the positions of employers of faculty, with the result being the reclaiming of authority by administrators in order to possess a strong bargaining position with faculty. All agree that though unionism may not be all-pervasive, and up to this time has mostly affected campuses with.weak faculty government, it is likely to increase in the 1970's, and is likely to have further impacts on faculty in the governance process. In summary, then, of the research on power in higher education, it appeared that in most cases, administrators and trustees generally hold the most power over university- wide decisions, particularly with relationship to financial matters. Faculty, particularly senior faculty, do maintain influence over academic-related matters, particularly in large, complex institutions, and the power of department chairmen appears to be related to the decision area and particular situation. The major sources of interpersonal 43 power within higher education appear in most cases to be legitimate, referent, and expert power, with acquisition of research money and quality graduate programs the subunit characteristics leading to power. Faculty and administrators see the institution generally the same way, but that view is quite different from legislators and other "outsiders", who appear to be gaining in influence at public institutions. Summary Power, and its related concepts of influence, control, and authority involves some kind of interaction between two or more persons or groups, and comes from a number of sources. The most common appears to be legitimate position, expertness, identification with a person or group, and ability to mani- pulate rewards and punishments. When applied to organiza- tions, it is apparent that more than one kind of power often operates in an organization with the personnel and the situa- tion determining what kinds of power or influence are employed. When power in universities was examined, a major issue was that faculty, as professionals, do not easily fit into the organizational hierarchy. Consequently, the dynamics of decision-making may be different than in non-educational or- ganizations. The research on power and influence in univer- sities showed that while administrators hold the most power over university-wide decisions, faculty have significant influence over academic matters. The differences in 44 perceptions of the university appeared between the "insiders" and "outsiders". Power is a complex concept, and the present study will attempt to gain further insight into the perceptions of power and influence in a university setting. In Chapter III, the data collection and analysis procedures used in the current study will be reviewed. 45 III. DATA COLLECTION AND ANALYSIS PROCEDURES Introduction As stated in Chapter I, the purpose of this research was to study the perceptions of faculty and academic admin- istrators of the power structure at a large public midwest— ern university to answer the following questions: 1) How do faculty, as compared with administrators, perceive decision-making in the university? 2) Are there predictable differences in perceptions of power and influence among various subgroups of faculty? 3) What changes in the per- ceptions of the decision—making in the university have occurred since the late 60's and early 70's? This chapter will include a description of the sample used in the research, the development of the survey instru- ment, the data collection procedures, and the procedures used for data analysis. Population and Sample Since the objectives of this research were to test several hypotheses regarding perceptions of faculty and administrators at a large public midwestern university, representative samples of those populations were used for the study, and are described below. 46 FacultySample In order to have larger numbers within a given aca— demic unit to test some comparisons between the units, five colleges, from a total of seventeen within the university under study, were used. The colleges were chosen on the basis of four structural variables determined by Blau (1973) to have an effect on faculty perceptions on their power or influence. The variables used were size (of faculty and student body), faculty salary level, differentiation, or number of units within the college, and clerical-faculty ratio. Data on these variables for all seventeen colleges was obtained from the Office of Institutional Research. The objective in selecting the five colleges was to choose colleges that were different on as many of those variables as possible. When the data on all colleges were examined, it was obvious that because the colleges were very different kinds of units, it would be difficult to find any that ranked very high or very low on all variables. Consequently, the major variables used were size and differentiation, as those were the variables found to most consistently impact perceptions of faculty influence by Blau (1973) and Baldridge, et. a1. (1973). Consequently, colleges that differed on the variables of size and differentiation were sought. Since the researcher was interested only in degree-granting units with both graduate and undergraduate programs, the three medical schools, the three residential colleges, and the 47 lower division, non-degree college were eliminated. Of the remaining ten colleges, an attempt was made to select units different not only on the two desired variables, but in type. For example, the researcher attempted to choose at least one professional college and one research—oriented college. Using these guidelines, then, five colleges were selected for the study, which appeared to meet the criteria of Blau and Baldridge, et. a1. and of the researcher. For this research, those colleges were designated as Colleges A, B, C, D, and E. After the five colleges were chosen, mailing labels were obtained from Data PrOceasing through the Office of the Provost for every full-time faculty member at the ranks of Professor, Associate Professor, Assistant Professor, and Instructor, both tenure track and non-tenure track. Excluded from these sets of labels were any faculty with administra- tive titles who were included in the University's list of titles used to designate administrators. Those administra- tors were included in the administrator sample. (See the following section for details.) The total number of faculty in the sample was 627. Administrator Sample To obtain the list of administrators from.Which to draw the sample, labels were requested, again through the Office of the Provost, for all those individuals with titles designated as administrators by Michigan State University. 48 This group included all those classified as faculty, but with administrative titles, and did not include those in the Administrative-Professional classifications. There are twenty-nine titles used to designate administrators, with the total number of administrators being 427. Since the major purposes of this research deal with academic matters, it was decided to include only those administrators specifi- cally related to the academic units of the University. Also, since the issues included in the research were particular to the main campus, those whose primary functions were off- campus were excluded. Thus, those excluded were: County Extension Directors Program Directors Residence Hall Managers Student Affairs Staff Off—Campus Continuing Education Staff President's Staff Vice President for Research Development's Staff Vice President for Business and Finance's Staff Vice President for University Development's Staff Vice Pfigsident for University and Federal Relations' Sta Executive Vice President's Staff Those included were those reporting to the Office of the Provost, or those who are primarily academic administrators. The final sample included 288 administrators with the titles listed below: Vice-President Assistant Vice President Assistant to the Vice President Associate Vice President Provost Associate Provost Assistant Provost Assistant to the Provost Registrar 49 Associate Registrar Assistant Registrar Dean Associate Dean Assistant Dean Director Associate Director Assistant Director Chairman Associate Chairman The Survey Instrument Several steps were involved in the development of the survey instrument. First, since there was some interest on the part of the researcher to make some comparisons of current attitudes and perceptions of faculty with those from 1970, several questions were used from a study done at the same university in 1970. Then additional questions deemed necessary were developed, and the entire instrument was pre-tested. After suggestions from the pre-test, the final version of the questionnaire was developed. Each of these steps will be described in detail below. Questions From 1970 Survey The survey done in 1970, the MSU Omnibus Survey, was administered by the Urban Survey Research Unit to approxi- mately 2500 students, 500 faculty, and 500 administrators. The questionnaire was designed to seek information, attitudes and opinions on a variety of issues important at the time at Michigan State University. The question from that survey which directly assessed perceptions of influence was asked in Va 50 such a way that the data needed for the current study could not be obtained if the question were replicated. The problem with the format in 1970 was that respondents to that question were asked to indicate which group or combination of groups (Administrators. Faculty, and Students) should have and d9 have the most influence over a variety of issues. Since respondents were able to mark combinations of groups, they did not have to choose one group over the other. The current research hypotheses, which required an indication of the perceptions of faculty and administrators of their own and the other group's influence, could not really be tested unless the respondent picked one group over the other. Consequently, when designing the replication, the formet was changed to get the infonmation needed. Because the questions on the two surveys were not comparable, it became obvious to the researcher that only descriptive comparisons and interpretations could be made. To aid in that, and also gather some additional data that the researcher felt would give further insight into faculty attitudes, several other questions were chosen for replication. These did not speci- fically relate to power and influence, but examined other perceptions about the university. Included were questions on departmental priorities, salary increase determinants, collective bargaining, and other university issues. Some of the wording was edited slightly to make the items rele- vant to 1977, but content remained the same. Some of the DE it 51 background questions used for the current study were also taken from the 1970 study, so comparisons of the samples could be made. Development of New Questions Since only the one question described above dealt dir- ectly with perceptions of power and influence, two addition- al questions were developed to provide additional data to be used to test the hypotheses dealing with current atti- tudes and perceptions. These questions were designed to assess perceptions of the influence of the faculty in general and the perception of the faculty member of his or her own influence. In addition, a question was developed to assess attitudes about a related issue, collective bar— gaining, and its potential effects upon faculty. For each of these new questions, the important issues were defined, then put into the desired question format. Another question, which asked faculty about their general satisfaction, and their perceptions of the satisfaction of other faculty members, was adapted from a recent survey used by the Mich- igan Education Association. Finally, several additions to a question which asked about attitudes on university issues were made to a question from 1970. The faculty question- naire can be found in Appendix B. The new questions are items 4, 5, 6, 7, and parts of 9. Since administrators were also included in the sample, and since slightly different information was needed from 52 them, a separate questionnaire was developed for them. There were several questions appropriate for only one group or another, but otherwise, the questionnaires were quite similar. The final questionnaire sent to administrators is Appendix D. As in the faculty questionnaire, several new questions were used. These are items 4, 5, 6, and parts of 7 in Appendix D. Pretest After the questionnaire was developed, approximately thirty faculty and administrators from academic units not included in the final sample were asked to complete the pretest questionnaire. In addition, they were asked to record the time the survey required and to make any comments, criticisms, or suggestions. At the same time, representa- tives of the Academic Council (governing body), two faculty organizations (AAUP and Faculty Associates) and University administration were consulted, to review and react to the questionnaire, as well as to make suggestions on additional items that should be included. 0f approximately 36 pretests sent out, 21 were returned, 14 from faculty and seven from administrators. Based on the responses and comments from the pretest, and ideas and suggestions from others consulted, several items were eliminated or revised, and several new items were added. The final questionnaire, with a cover letter, was then prepared for mailing. 53 Data Collection Procedures Prior to the distribution of the survey to the faculty in the five colleges chosen, each of the Deans was notified that their faculty would be receiving the questionnaire, and asked if they had objections. As none did, the question- naires were distributed to all faculty in each of the five colleges. The 627 faculty questionnaires, with a cover letter and addressed return envelope, were mailed on wednes- day, March 9, 1977. The 288 administrator questionnaires, also with a cover letter and return envelope, were mailed on Thursday, March 10, 1977. The cover letters are Append- ices A and C. Returns were requested by Wednesday, March 16, 1977. Since the March 16 date was during final examination week, it was anticipated that returns might be lower than desirable. Consequently, a follow-up mailing was planned. The original questionnaires were not coded in any way for identification, so a reminder letter, with another copy of the questionnaire, was sent to all 915 in the sample on Friday, April 1. The second letters are Appendices E and F. Though no return deadline was indicated, an immediate return was requested. Any questionnaires, regardless of when they were mailed, which were returned after Tuesday, April 5, were considered as responses to the second mailing, and were coded as such. 54 Questionnaires returned on or before Friday, April 22 were included in the analysis. Those returned later are reported in the following chapter, but not included in the analysis. The questionnaires were coded and key punched. The coding reliability was checked and the key punching verified. After a duplicate deck of data cards was made, the data were analyzed. Data Analysis Procedures The statistical technique used to analyze the data in the study was somewhat similar across hypotheses, as in several cases, the hypotheses asked the same questions, but on different subsets of respondents. For Hypotheses #1, 3, 4, and 5, the question used was Question 1 of the questionnaire. For this, the initial statistic used to examine perceptions of who has the most influence over a particular matter for whatever two groups are involved was the Chi-Square Test of Independence, where 2 2 = z (O-E) X E (Schmidt and Ripstra, 1973, p. 71) That test indicates whether the perception of who actually has the most influence is independent of subgroup membership (i.e., faculty versus administrators, or large college faculty versus small college faculty), The Chi-Square was chosen because the data available is frequency data, and there was a need for a statistic that would easily show wh wc ir va ta be ti Cd in ha Wh th 011 f0 55 whether the frequencies obtained differed from those which would be expected by chance. Since the Chi—Square only indicates whether one variable is independent of another, and not what the relationship is, an additional test was done to determine strength and direction of the relationship. The test used was the Yule's Q Coefficient, where = bc - ad BC + ad Q (Davis, 1971, p. 40) The Yule's Q was used because it is a simple way to obtain a good deal of information about a relationship between two variables with four cells and with frequency data.. For the tables with more than four cells, statistical tests cannot be done, but observations can be made on strength and direc- tion. There were very few comparisons in which that is the case. The other question used in the data analysis to direct- ly test the hypotheses is Question 5, which asks how much influence the respondent perceives university faculty to have over eight issues. To determine if the subgroups on which comparisons were made saw strong or weak influence on the same issues, a Spearman Rank-Order Correlation was done on the means of each of the eight items. The Spearman formula is: rS = l = Efifi§f¥%3 (Glass and Stanley, 1970, p. 174) The Spearman does not give any comparisons on the values 56 of the means, but is an easy way to examine the relative importance of issues for two groups of respondents. In order to determine if the amount of influence ascribed to faculty (as indicated by the mean) was differ- ent between two subgroups, a test of differences between population means was used. The assumptions made and met were large sample size, unknown population variances, and known sample variances. Because in each case the hypothesis to be tested suggested that one mean will be greater than the other, a one-tailed, two-sample test was appropriate. The formula used, then was the formula for the one-tailed two-sample Z Test, where il'i2 Z = (McSweeney and Olejnik, 1974, p. 7) 2 2 S 8 f.1_.+_2_ n1 n2 The alpha level sought was to be .00625, so that the overall significance level for the eight-item question will be .05. For Hypotheses 2 and 5, four-celled tables were again generated, comparing the number of respondents who saw "congruence" on an item to the number seeing "dissonance" for whatever subgroups are involved.2 For these, also, the 2For this research, congruence exists if the respondent per- ceives the same group (Administrators, Faculty, or Students) actually has the most influence over an issue that should have it. Dissonance exists if some other group has it than the one the respondent feels should. 57 Chi-Square and Yule's Q were calculated as before. For Hypotheses 7 and 8, no statistical comparisons were made, as the data gathered in 1977 were not presented in the same fashion as in 1970. Consequently, some data will be presented, and observational statements made about the results for the two time periods. The following chapter will present the results of the data analysis and some conclusion statements about them. IV. DATA ANALYSIS Introduction The first part of this chapter will describe the response to the questionnaire and compare the respondents to the population on several key variables. The next sec- tion will analyze the survey data with regard to Hypotheses 1-5, or those which deal with only the 1977 data. The final section will deal with the comparison of the 1970 to the 1977 data. Questionnaire Response As was stated in Chapter 3, a total of 915 question- naires were sent, 627 to faculty in five selected colleges, and 288 to those holding administrative positions. Of the questionnaires sent, 412, or 45%, were returned on or before April 5, 1977 and were considered results of the initial ‘mailing. After April 5, an additional 17% were returned, bringing the total to 563, or 62%. Of the 563 total return- ed, four were returned after April 27, the last day for return to be used in the study. In addition, eleven were returned blank, with some indication of unwillingness or inability to complete the instrument, such as illness, 58 59 sabbitacal, lack of time available, and others. An addi- tional four questionnaires had insufficient identifying information to make them usable, making the total of usable questionnaires 544, 347 from.faculty and 197 from adminis- trators. A summary of the return of the usable question- naires is found in Table 4.1. Table 4.1 Questionnaire Response Usable Questionnaires Sent lst Mailing 2nd Mailing Total Group Out Returned Returned Returned i i 7.: i Z: i 72 College A 245 91 37 36 15 127 52 College B 55 22 40 9 16 31 56 College C 81 37 46 12 14 49 60 College D 61 27 44 10 17 37 61 College E 185 70 38 33 18 103 56 Total Faculty 627 247 39 100 16 347 55 Total Adminis- trators 288 155 54 42 14 197 68 TOTAL 915 402 44 142 15 544 59 Comparison of Respondents to PopulatiOn To determine the extent to which inferences could be Inade about the population from which the respondents came, 60 a comparative analysis was done on several key demographic variables to determine similarity. For the faculty, these variables were faculty rank, sex, and age. For the admin- istrators, the variables were faculty rank, sex, and admin- istrative position. There was one problem involved in the comparison of the faculty respondents to the faculty in the population. The faculty in the five colleges to whom questionnaires were sent included the ranks of Instructor, Assistant Professor, Associate Professor, and Professor, both tenure track and non-tenure track. Consequently, the group of respondents included some faculty not in tenure track positions. The demographic data available for the population, which was obtained from the Office of the Provost, included only faculty in tenure track positions. Approximately 11% of the respondents indicated they were not in tenure-track positions. Consequently, they were excluded in the come parisons reported below. The faculty comparisons were made on each of the five colleges individually, and on the faculty as a whole. The administrator comparisons were made on the respondents and population as a group. Generally, the respondents appeared to be very similar to the population on all variables examined (see Table 4.2 through Table 4.7). For the faculty, the percentages in each category of the rank and sex variables were extremely 61 close. On the age variable, it appeared that the respond- ents were slightly younger than the population, but they still were quite similar, For the administrators, the distribution on the rank and sex variables were also very similar. The comparison on administrative position showed slight differences, but again, generally appeared to be very similar. The consistency between the respondents and the popula- tion proved to be high enough overall that inferences about the population could easily be made from the respondents' data. Although, as was stated above, 11% of the faculty, the non-tenure track people, were not included in the come parisons, the similarity between the respondents and the population for the tenure track faculty was high enough such that the addition of the non—tenure track faculty would probably not make a meaningful difference in the com- parability of the two groups. Table 4.2 through Table 4.7 below reveal the comparisons described for the faculty as a whole and for the administrators. The comparative data for the individual colleges is found in Appendix H. 62 Table 4.2 Comparison of Faculty Respondents to Population—Rank Rank Populationa Respondents a Z a Z. Professor 289 47 143 46 Associate Professor 169 27 78 25 Assistant Professor 149 24 82 27 Instructor l3 2 6 2 TOTAL ezob 100 309c 100 aData obtained from the Office of the Provost, April, 1977. bDifference between this total and the total in the faculty mailing is accounted for by the exclusion of non-tenure track faculty and the inclusion of some administrators. CTenure track faculty only. Table 4.3 Comparison of Faculty Respondents to Population-Sex Sex Population Respondents a Z 9. 7o. Male 510 82 249 81 Female 110 18 59 19 No Response 1 “TOTAL 620 100 209 100 Comparison of Faculty Respondents to Population-Age Table 4.4 63 Age Group :opulation fiespondents Under 30 21 3 ll 4 30-39 189 30 113 37 40-49 193 31 87 28 50-54 63 10 30 10 55-59 75 12 36 12 Over 60 79 13 31 10 No Response 1 TOTAL 620 99 309 101 Table 4.5 Comparison of Administrator Respondents to Population-Rank Rank Professor Associate Professor Assistant Professor Instructor Other TOTAL Population E a 195 68 48 17 22 8 5 2 l8 6 288 101 Regpondents ‘E a 133 68 35 18 18 9 4 2 7 4 197 101 64 Table 4.6 Comparison of Administrator Respondents to Population—Sex Sex Population Respondents "'— a 7 a Z Male 252 88 171 87 Female 36 12 26 13 TOTAL 288 100 197 100 Table 4.7 Comparison of Faculty Respondents to Population-Position Position Population Respondents __ a 72 a E Dean 21 7 l3 7 Associate Dean 15 5 9 5 Assistant Dean 33 12 17 8 Director 68 24 46 23 Associate Director 14 5 12 6 Assistant Director 22 8 l4 7 Chairperson 85 29 57 29 Associate Chair- person 16 6 15 8 Other 14 5 l4 7 TOTAL 288 101 197 100 and that coul isoc tior The the coun CO‘ 65 The high similarities between the respondent group and the population of the five colleges studied indicates that generalization to that population is legitimate. In order to determine if the data from this research could be generalized to the university as a whole, compar- isons were also ma _ tween the respondents, the popula— tions of the five c lel es, and the university-wide faculty. The latter populati ncluded all tenure track faculty in the university. Tables 4.8 through 4.10 indicate these comparisons. Table 4.8 Comparison of Faculty Respondents, Population of Colleges Used, and University Population-Rank 1 Five College University Rank Respondents Population Population a To. a E e. 7: Professor 143 46 289 47 1014 50 Associate Professor 78 25 169 27 570 28 Assistant Professor 82 27 149 24 432 21 Instructor 6 2 13 2 32 2 TOTAL 309 100 620 100 2048 101 66 Table 4.9 Comparison of Faculty Respondents, Population of Colleges Used, and University Population-Sex Sex Male Female No Response TOTAL Respondents 2. IT 249 81 59 19 l 309 100 Five College Population 2 ‘Z 510 82 110 18 620 100 University Population n 7% 1919 87 297 13 22168 100 aThe n used here is different than the n in Tables 4.8 and 4.10 because the information was obtained at a different time. Table 4.10 Comparison of Faculty Respondents, Population of Colleges Used, and University Population-Age Ass Under 30 30-39 40-49 50-54 55-59 60 and over No Response TOTAL ‘ Respondents a Z 11 4 113 37 87 28 30 10 36 12 31 10 l 309 101 Five College EPopulationZ 21 3 189 30 193 31 63 10 75 12 79 13 620 99 University Population ‘2 ‘Z 42 2 624 30 660 32 232 11 242 12 248 12 2048 99 67 It appears from the above information that the respond- ents fit the university population about as well as they fit the population of their own colleges, except on the sex variable, where there is some difference. There are some slight differences on the other variables, which are similar to the ones mentioned earlier. Again, the respondent group is younger, with 41% under 40 years of age, compared with 32% in the university population. Also, the distribution by rank revealed fewer Professors and Associate Professors and more Assistant Professors in the respondent group. Con- sequently, statements about the university faculty as a whole from these data can possibly be made, but with some caution. The following section will report the results of the data relating to each of the research questions and related hypotheses for the 1977 survey. 1977 Survey Results Research Question #1'- How do faculty as compared with administrators, perceive decision- making in the university? Hypothesis #1 - Compared with faculty, administrators will perceive more faculty influence on academic matters. The parts of the questionnaire used to determine the relationship between faculty and administrator perceptions of faculty influence were the columns of Question 1 which asked, "Who actually has the most influence?" for each issue, plus Question 5, which asked "how much influence" faculty have on each of eight‘matters. 68 For the twelve items in Question 1 which asked who actually has the most influence, a Chi—Square test of infependence was used, where x2 = z (O-E22 , to determine if the perceptions of who has the most influ- ence are independent of position, i.e., whether the respond- ent is a faculty member or administrator. For several of the items, though most of the respondents indicated either Faculty or Administrators as a group had the most influence, a few selected Students as having the most. Consequently, on those items, there occasionally were cell sizes of zero or of less than five, which meant that those items did not meet the assumptions for the Chi-Square test. For those items, the Chi-Square was recalculated, leaving those who selected students out, as in all cases it was less than 1.5%. Of the twelve items in the question, four appeared to be significant at the .05 level, which indicated that per- ception of actual influence is dependent in some way on position. To try to determine more about that dependent relationship, a Yule's Q coefficient was calculated for each four-cell table (Davis, 1971, p. 47). The sign and magnitude of the Yule's Q coefficient gives some indication of the direction and strength of the relationship. For one of the four significant items, a sufficient number of respondents selected Students as having the most influence to make the table 6-celled. As the Yule‘s Q can only be calculated on 69 four-celled tables, it was not done on that item. Observations about that and the other items in which there is a signifi- cant Chi-Square will be discussed. Table 4.11 below indicates the Chi-Square and the Yule's Q for each item. Two items, determining internal university budget allocations and determining university tuition and fees, have been omitted, as both of them are 6-ce11ed tables with four of the cells having frequencies less than five. Table 4.11 Chi-Square and Yule's Q Coefficients for Comparisons of Faculty vs. Administrators for Who Has the Most Influence Item Chi-Square Yule's Q Appointing a department chairperson 3.33 .18 Appointing an academic dean 3.05 .21 Appointing a provost 7.26* .57 Creating new educational programs 18.47* .41 Determining faculty salaries 1.44 .17 Determining if a temporary faculty 1.62 —.14 member should be rehired Determining tenure for faculty members 3.03 .17 Determining undergraduate university 6.20* .35 admissions policy Disciplining a student for cheating 10.28* cannot be on an exam calculated Hiring new faculty members 1.06 .10 *Significant at the .05 level or better with 1 df 70 In each of the above items in which the Chi-Square was significant, the Q coefficient is positive, indicating that administrators are in fact more likely to feel that Faculty, as contrasted with Administrators, actually have the most influence on that particular item. 0n the item "disciplining a student for cheating on an exam”, where the number of cells rules out the Yule's Q, observation shows that 80% of the administrators and 67% of the faculty felt Faculty had the most influence. Consequently, it appeared that that partic- ular item follows the pattern of the other three. The four decision areas in which administrators do perceive more faculty influence than faculty perceive for themselves, are appointing a provost, creating new education- al programs, determining undergraduate admissions policy, and disciplining a student for cheating on an exam. It should be noted, of course, that of the ten measurable items, only four resulted in a significant Chi-Square, indicating that for the other six, the perception of actual influence did not appear to be dependent on whether the respondent was a faculty member or administrator. Additionally, if all the items in Question 1 are to be taken as an overall indication of actual influence, with a question-wide alpha of .05, each item should be examined for an alpha level of .005, as there are ten measurable parts. With that in mind, only the perceptions of faculty influence over the appointing of a provost and creating new educational programs appear to be 71 dependent on classification. Therefore, overall, it could probably be said that administrators and faculty perceive faculty influence differently in some, but not all, decision areas. Question 5 of the questionnaire was the second question used to examine Hypothesis #1. The question asks how much influence university faculty have on each of eight depart- ment matters, with the following responses possible: 4 - A great deal of influence 3 - A moderate amount of influence 2 — A slight amount of influence 1 - No influence at all Means for faculty and for administrators on each of the eight items were calculated and two types of analysis done. First, an assessment was made to determine if the rank order of the matters was similar, that is, if the faculty saw themselves as having the strongest influence over the same issues as administrators did. The statistic used to test the rank order is the Spearman Rank-Order Correlation, where rs = 1 _ 6Z(x-y)2 n(n -1) The rank-order coefficient, when calculated, was rs=.90, which indicates an extremely high agreement between faculty and administrators on which areas faculty have the most influence over. 72 The rank-order correlation does not, if course, indicate anything about the magnitude of the differences between the means for any given item. Consequently, to determine if the faculty perception of their influence on a given matter differed from that of administrators on the same matter, a comparison of means on each of the eight items was done. Because the n is large, the Z test was done, where xl‘xz To keep the level of significance at .05 for the entire question, an alpha level of .00625 (.05/8) was used for each item. The test showed that of the eight matters, six means were significantly different at the level of significance demanded. The issues on which faculty and administrators did not see faculty influence differently were the hiring of new faculty and the criteria for graduate student admission, while the means on the other six issues were significantly different. Table 4.12 indicates the mean, variance, Z-score and rank-order for each item. 73 Table 4.12 Z Scores and Rank-Order for Mean Comparisons Between Faculty and Administrators on Amount of Faculty Influence Item Administrators Facult Z i s 2 Rank i s 2 Rank _ _l 1 Order _2 2 Order Curriculum 3.89 .141 l 3.66 .296 1 5.75* Hiring of new faculty 3.39 .413 4 3.32 .463 2 1.19 Development of faculty personnel policies 3.27 .506 5 2.96 .673 5 4.42* Selection of a depart- ment chairperson 3.43 .400 3 3.21 .577 4 3.61* Department budget allocations 2.30 .581 8 1.97 .590 8 4.85* Faculty load deter- minations 2.83 .554 6 2.36 .751 7 6.06* Criteria for graduate student admission 3.44 .594 2 3.30 .703 3 1.97 Selection of an academic dean 2.80 .604 7 2.39 .687 6 5.77* *Significant at the .00625 level n1 = 197 I12 = 347 One additional way of examining these results, which does not yield any particular statistics but gives another perspec- tive, is to look at matters in which the faculty appear to have the most influence and on which issues the greatest discrep- ancy occurs between the perceptions of faculty and adminis- trators. Table 4.13 below reviews the rank-order for the 74 two groups. Table 4.13 Rank-Order of Amount of Faculty Influence for Faculty and Administrators Item Administrative Rank Faculty Rank Curriculum 1 1 Criteria for graduate student admission 2 3 Selection of a department chairperson 3 4 Hiring of new faculty 4 2 Development of faculty personnel policies (reappointment, promo- tion, tenure) Faculty load determination Selection of an academic dean QVOUI WGVUT Department budget allocations An indication of the magnitude of the difference between the mean for faculty and the mean for the administrators is the Z score, which was calculated above. The three items with the largest Z scores are faculty load determination, the selection of an academic dean, and curriculum” An overall ivew of the data on whether administrators perceive more faculty influence than faculty do reveals that in certain decision areas, there is some indication that 75 administrators do perceive faculty to be more involved. Those areas include curriculum matters, or creating new educational programs, faculty load determinations, selection of a provost and selection of an academic dean. The evi- dence is not overwhelming, but probably sufficient to lead the researcher to tentatively accept the hypothesis and to suggest further investigation of the issue. Hypothesis #2 - The differences between the perceived and preferred pattern of influence will be greater for faculty than for administrators. The question used to explore this hypothesis was, as with the previous hypothesis, Question 1. The method used here, however, was a comparison on each of the twelve items of the column asking "who should have the most influence" with the column asking "who actually has the most influence". If the respondent indicated the same group should have the most influence that actually does, the response was considered "congruent". If the indication was that a group other than the one with the most influence now should have it, the answer was considered "dissonant". For each of the twelve items in Question 1, then, the responses were collapsed into a four- celled table to compare the dissonance of faculty and admin— istrators, The tables were constructed as follows: Congruence Dissonance Administrators I Faculty [ 76. Those respondents in the Congruence cell were those indicating the same group "should" and "does" have influence; those in the Dissonance cell indicated different groups for "should" and "does". The initial test on the four-celled tables was, as in the previous hypothesis,the Chi-Square Test of Independence, to determine whether Congruence or Dissonance was independ- ent of classification (i.e., faculty or administrator). A Yule's Q coefficient was then calculated to determine the direction and strength of the relationship. The Chi-Square analysis showed all comparisons to be significant at the .025 level or better. There was one table, however, containing a cell size of less than five, so the calculations were not performed on that item. The significant comparisons indicated that a congruent or dis- sonant response on a particular item is not independent of whether the respondent is a faculty member or administrator. The Yule's Q coefficients calculated indicated all the rela- tionships to be negative, which, for the way the tables were developed, meant administrators were likely to indicate congruence on the items compared and faculty were likely to indicate dissonance. Table 4.14 below indicates each decision area, or item, the Chi-Square coefficient for congruence versus dissonance, and the Yule's Q coefficient. The item, "determining university tuition and fees", is omitted, because of a cell size less than five. 77 Table 4.14 Chi-Square and Yule's Q Comparisons for Faculty vs. Administrators on Congruence and Dissonance Item Chi-Square Yule's o Appointing a department chair- 25.82** -.46 person Appointing an academic dean 43.28** -.56 Appointing a provost 27.86** —.47 Creating new educational programs l9.82** -.38 Determining faculty salaries 51.48** -.65 Determining if a temporary 29.95** -.50 faculty member should be rehired Determining tenure for faculty 15.50** -.37 members Determining undergraduate 5.34** -.22 university admissions policy Determining internal university l7.75** -.64 budget allocations Disciplining a student for 7.93** -.31 cheating on an exam Hiring new faculty members 15.06** -.36 **Significant at the .025 level or better with 1 df In examining the data in Table 4.14, it appears that faculty do in fact perceive a greater degree of dissonance between "what is" and "what ought to be", or "who should" and who does". This could also be interpreted as dissatis- faction with the decision patterns on university issues. If 78 so, it appears that faculty are more dissatisfied than administrators. On the basis of the above finding, it appears that the greatest differences between faculty and administrator dissonance are in the following areas: determining faculty salaries, appointing a dean, determining if a temporary fac- ulty member should be rehired, determining internal university budget allocation, appointing a provost, and appointing a department chairperson. Though nothing was said in the above reporting of statistical results about "direction" of the dissonance, some descriptive comments are in order here. In examining what faculty were reporting when they indicated dissonance, by far the most common pattern, as might be expected, was that of faculty saying that administrators actually had the most influence, but faculty should. For the selection of a dean, 67% of the faculty expressed this kind of dissonance, and for the selection of a provost, 59% of the faculty responded in this fashion. For all of the other issues, with the exception of budget allocation, disciplining a student for cheating, and determining tuition, between 46% and 50% of the faculty responded that they should have the most influence where administrators currently do. Though administrators generally indicated congruence between who should and does have influence, there were a few issues where a fairly large number felt faculty should 79 have influence but do not. In the determination of universi- ty admissions policy, 40% of the administrators indicated this kind of dissonance, and for appointing a dean, 33% did. For appointing a provost, 31% indicated faculty should have the most influence but administrators actually do. It should be noted that on each of these three issues a larger percent indicated administrators do and rightly should have the most influence (43%, 43%, and 61% respectively). A complete breakdown of the combinations of responses of who should versus who does for each issue is found in Appendix I. The overall conclusion, then, for the comparison of dissonance between faculty and administrators is that faculty do in fact perceive a greater difference between who should and who does have the most influence. The data indicates acceptance of the hypothesis. Researchoguestion #2 - Are there predictable differences in perceptions of the pattern of influence among various subgroups of the faculty? Hypothesis #3 - Faculty in lar e, complex colleges will perceive more aculty influence on academic matters than those in smaller colleges. Of the five colleges surveyed for the study, College A and College E were designated as large colleges for evaluating this hypothesis, and the remaining three, Colleges B, C, and D, were considered small colleges. Faculty size and student enrollments of the five are indicated in Table 4.15. 80 Table 4.15 Size of Faculty and Student Body by Collegea FTE Undergraduate FTE Graduate College FTE Faculty Students Students College A 243 3002 724 College B 48 1808 339 College C 84 2028 280 College D 53 1889 304 College E 192 4403 975 aData obtained from the Office of Institutional Research, 1974-75 and 1975-76 Annual Evaluation and Report The same two questions on the survey instrument used for Hypothesis #1 were used for this hypothesis. In question #1, the columns asking “who actually has" the most influence were analyzed by college. For each item on that question, the faculty member was asked to indicate whether Administrators, Faculty, or Students actually have the most influence over that particular decision. Those responses were analyzed to determine if there were differences other than those attributable to chance between the large and small colleges. The test used was the Chi-Square Test of Independence. This test, as indicated for the previous two hypotheses, tests whether responses to "who actually has the most influence" are independent of college affiliation. 81 For the twelve items on the question, most respondents indicated either faculty or administrators, rather than students, as having the most influence. Consequently, most calculations were on a four-celled table: Large Small Colleges Colleges Administrators have most influence [ I 7] Faculty have most influence [ I ] A few, however, indicated students, which resulted in six- celled tables, but with frequencies less than five. Conse- quently, the Chi-Square assumptions were not met. However, in one item the responses in the student category were .6% of the total, so they were eliminated and a new Chi-Square calculated. For several other iteme, cell frequencies of zero or less than five occurred other than in the student category, so those iteme were not used. Of the twelve items, meaningful Chi-Squares were calculated on seven. A Yule's Q coefficient was also calculated for each item to determine the strength and direction of the relationship. Table 4.16 indicates the Chi-Square and Yule's Q coefficients. Table 4.16 indicates that perception of who has the most influence is not independent of size of college on five of the seven measurable items: hiring new faculty, determining tenure for faculty members, appointing a department chair- person, determining whether to rehire temporary faculty, and creating new educational programs. The Yule's Q on those 82 Table 4.16 Chi-Square and Yule's Q Coefficients for Comparison of Large vs. Small Colleges for Who Has the Most Influence Item ' Chi-Square Yule's Q Appointing a department chairperson 13.58*** .43 Appointing an academic dean 2.87 .30 Creating new educational programs 7.66*** -.32 Determining if a temporary faculty 9.73*** .41 ‘member should be rehired Determining tenure for faculty 18.34*** .51 ‘members Determining undergraduate university .19 .14 admissions policy Hiring new faculty members 20.71*** .53 ***Significant at .01 level or better with 1 df five items indicates the relationship to be moderately or substantially positive on four of the five, which means that faculty in large colleges to tend to perceive more faculty influence on these issues than those in the smaller colleges. On the fifth item, creating new programs, the Q is negative, meaning that faculty in the smaller colleges are the ones to see stronger faculty influence. Even with the one relationship being the reverse of what was hypothesized and two non-significant Chi-Squares, the data still support the hypothesis that faculty in larger 83 Table 4.17 Rank-Order Correlation for Large vs. Small Colleges on Amount of Faculty Influence Item Large Colleges Small Colleges Mean Rank Mean Rank Curriculum 3.75 l 3.67 l Hiring of new faculty 3.53 2 3.10 3 Criteria for graduate 3.51 3 3.16 2 student admission Selection of a department 3.48 4 2.94 4 chairperson Development of faculty 3.27 5 2.77 5 personnel policies (reappointment, promo- tion, tenure) Selection of an academic 2.62 6 2.34 6 dean Faculty load determinations 2.59 7 2,13 7 Department budget alloca- 2.18 8 1.91 8 tions r = .99 colleges will perceive more faculty influence than faculty in smaller colleges. The other question used to investigate this hypothesis is Question 5. For each of the items, respondents were asked to indicate how much influence faculty have on eight individ- ual matters, on a 1-4 scale, as explained earlier under 84 Hypothesis #1. As in Hypothesis #1, the first comparison done was a Spearman Rank-Order Correlation, to determine whether faculty in small colleges perceived high faculty influence on the same items that faculty in large colleges perceived high faculty influence. The rankings are shown in Table 4.17, and are quite similar for the two groups of faculty. To analyze the differences between the means for each of the issues in Question 5, a one-tailed Z test was com- puted. As in Hypothesis #1, in order to keep the level of significance at .05 for the entire question, the significance level sought for each item was .00625. The calculations showed the differences between the means for five of the eight items to be significant at the .00625 level or better. The significant differences between the two groups of faculty were found on the issues of selec- tion of a chairperson, hiring of new faculty, faculty load determinations, development of faculty personnel policies, and criteria for graduate student admission. On those five items, the faculty in larger colleges perceived greater amounts of faculty influence than the faculty in the smaller colleges. Table 4.18 indicates the mean, variance, and Z score for each item. Table 4.18 85 Z Scores for Mean Comparisons Between Large and Small College Faculty on Amount of Faculty Influence Item Curriculum Hiring of new faculty Development of faculty personnel policies (reappointment, pro- motion, tenure) Selection of a depart- ment chairperson Department budget allocations Faculty load deter- minations Criteria for graduate student admission Selection of an aca- demic dean Large College Small College Faculty FaCulty - 82 — 2 .;1 .41. .JE 32. 3.75 .806 3.67 .256 3.53 .947 3.10 .454 3.27 .828 2.77 .669 3.48 .184 2.94 .598 2.18 .641 1.91 .966 2.59 .548 2.13 .542 3.51 .457 3.17 .767 2.62 .980 2.34 .648 Score .97 4.75** 4.24** 5.31** 2.24** 4.30** 3.04** 2.41** **Significant at .025 level or better n2 = 119 When all the data are examined, it appears that part- icularly in the areas of appointing a department chairperson, hiring new faculty, determining faculty loads, and develop- ment of faculty personnel policies, faculty in large colleges 86 do, in fact, perceive greater faculty influence than do faculty in smaller colleges. Though there are a few issues where the comparison has not proved significant, the results suggest acceptance of Hypothesis #3. Hypothesis #4 - Faculty in colleges assessed to be of higher quality will perceive more faculty influence than those in colleges of lower quality. As was mentioned in Chapter I, an internal assessment of quality was used to categorize each of the five colleges in the study as either higher or lower quality. That assessment was obtained from data on two of the questions in the survey. The first, Question 11 of the Faculty Questionnaire, asked each faculty respondent to give a relative ranking, on a six- point scale, of his/her department compared with similar departments nationwide. Though the self-rating is of the department rather than college, it gives one index to self- perception of quality. The second question used was Question 19 on the Admin- istrator Questionnaire, which asked the respondents to rate, on a similar six-point scale, each of the five colleges in the study. The responses for the first question were broken down by the college of the respondent with the higher means indicating higher self-perception. The means for each college on the second question were then compared with the first, to determine if the rankings and differences in means were similar enough to easily separate the colleges into 87 lower and higher quality. The means and rankings of the colleges appear in Table 4.19. Table 4.19 Comparisons of the Five Colleges on Two Indices of Perception of Quality College EEEEF1§£§:_ ggfistiogafi College B l 4.97 2 4.13 College D 2 4.82 l 4.28 College E 3 4.71 3 3.78 College A 4 3.81 4 3.57 College C 5 3.53 5 3.35 It appears that the perceptions of quality on the two questions used are quite similar, and that it would be legit- mate to consider Colleges 8, D, and E as higher quality and Colleges A and C as lower quality. To determine if differences exist between these two groups of colleges, the same type of analysis was done on Questions 1 and 5 of the questionnaire for the two groups as was done for large versus small colleges for Hypothesis #3. For Question 1, the Chi-Square Test of Independence was used to determine if perception of who has the most influence is independent of quality of the unit. As in the previous hypothesis, some of the tables used for the Chi-Square 88 calculations had frequencies of less than five, and could not be used. In others, if the percent of those indicating that students had the most influence was less than 1%, those responses were eliminated and a new Chi-Square calcu- lated. When the Chi-Squares were examined, results showed that none of them were significant at the .05 level or better, indicating that the response to who has the most influence is independent of quality of unit, insofar as the indices used measure quality. The Chi-Squares are given in Table 4.20. Table 4.20 Chi-Square Coefficients for Comparison of Higher vs. Lower Quality Colleges for Who Has the Most Influence logo Chi-Square Appointing a department chairperson .10 Appointing an academic dean 1.06 Appointing a provost .00 Creating new educational programs 4.39 Determining faculty salaries 1.34 Determining if a temporary faculty member should be rehired .00 Determining tenure for faculty members .93 Determining undergraduate university admissions policy .02 Disciplining a student for cheating on an exam 5.83 Hiring new faculty members .02 89 The second test was on Question 5, where again, the Spearman Rank-Order Correlation and the one-tailed Z-Test were used. The Spearman Rank-Order Correlation showed the rankings, based on mean scores, to be almost exactly the same between lower and higher quality colleges. The means and ranks are shown in Table 4.21. Table 4.21 Rank-Order Correlation for Higher vs. Lower Quality Colleges on Amount of Faculty Influence Item. Lower Quality Higher Quality M252. Rank 3222. Rank Curriculum 3.77 l 3.67 1 Hiring of new faculty 3.41 2 3.35 3 Criteria for graduate 3.39 3 3.40 2 student admission Selection of a depart- 3.35 4 3.24 4 ment chairperson Development of faculty 3.18 5 3.01 5 personnel policies (reappointment, promo- tion, tenure) Selection of an academic 2.60 6 2.45 6 dean Faculty load determinations 2.48 7 2.39 7 Department budget alloca- 2,21 8 1.97 8 tions When the Z test was done, again keeping the alpha level at .00625 for each individual test, the difference between 90 means was not significant for any of the items, and only significant at the .05 level for one of the eight items. Table 4.22 below indicates the means, variances, and Z scores for each item. Table 4.22 Z Scores for Mean Comparisons Between Higher vs. Lower Quality Colleges on Amount of Faculty Influence Item Lower Quality Higher Qualipy, Z Score _ 2 - 2 iii .1 *2 :2. Curriculum 3.77 .767 3.67 .466 1.18 Hiring of new faculty 3.41 1.018 3.35 .637 .63 Development of fac- 3.18 1.608 3.01 1.348 1.32 ulty personnel . policies (reappoint- ment, romotion, tenure? Selection of a de- 3.35 1.177 3.24 .914 1.03 partment chairperson Department budget 2.21 1.729 1.97 1.092 1.94* allocations Faculty load deter- 2.48 1.464 2.39 1.030 .70 minations Criteria for grad- 3.39 1.257 3.40 1.241 -.07 uate student admis- sion Selection of an aca- 2.60 2.011 2.45 1.063 1.19 demic dean *Significant at .05 level 91 Since there is no indication that the indices used are a valid measure of quality of faculty, it is probably best to consider the interpretation of this data exploratory. However, insofar as the measures used give us some sense of an internal perception of quality, it appeared that there were no significant differences on perceptions of who has influence and how much influence faculty have that is attrib- utable to the measures used for higher versus lower quality. Consequently, the data suggest rejection of Hypothesis #4. Hypothesis #5 - Assistant Professors and Instructors will perceive Administrators to have more influence than Associate Professors and Professors will. As in the case of the two previous hypotheses, this hypothesis was used to examine perceptions of who has the most influence for two groups of respondents. For this hypothesis, the two groups compared were the higher ranked faculty (Professors and Associate Professors) and the lower ranked faculty (Assistant Professors and Instructors). The same tests used before, i.e., the Chi—Square and Yule's Q were used to analyze Question #1, which was the only question which provided data to directly explore this question, since the researcher was examining perceptions of the influence of administrators, not faculty. However, to elicit additional information, Question #5 was examined, assuming that if it could be expected, that lower ranked faculty would perceive more administrator influence, they might also be expected to SE 115 Or ie VE be 1') TN p I tI WI 01 fa the lib: 92 see less faculty influence. The one-tailed Z test was again used. For this particular hypothsis, the Spearman Rank- Order Correlation was not appropriate. Again examining the items in Question #1, cell frequenc- ies of less than five caused several comparisons to be in- valid. For this particular analysis, four items could not be used at all. Of the reamining eight, two had less than 1% in the student category, so those responses were eliminated and a new Chi-Square calculated. On the basis of the eight calculations, only one of the Chi-Squares was significant at the .05 level or better, and thus was the only item where perception of who has the most influence is not independent of faculty rank. That item was appointing a department chair- person. As in previous procedures, a Yule's Q was calculated on the items to determine strength and direction of the relation- ship. The Q coefficient on the one significant item showed the relationship to be negative, which, for the way the table ‘was constructed, meant that the lower ranked faculty were more likely to perceive faculty as having the most influence over the appointment of a chairperson than the higher ranked faculty. This result was contradictory to the predicted one. Thus, since the remaining comparisons were not significant, there was no evidence on this question to provide support to the hypothesis. Table 4.23 reports these results. 93 Table 4.23 Chi-Square and Yule's Q Comparisons of Lower Ranked and Higher Ranked Faculty on Who Has the Most Influence Iggy, Chi-Square Yule's Q Appointing a department chairperson 6.68*** -.32 Appointing an academic dean 2.45 -.29 Creating new educational programs .21 .05 Determining faculty salaries .32 .15 Determining if a temporary faculty 2.46 -.22 member should be rehired Determining tenure for faculty members .52 -.10 Determining undergraduate university .12 .11 admissions policy Hiring new faculty members .13 .06 ***Significant at the .01 level with 1 df For Question 5, the one-tailed Z test comparisons on the means of each part of the question were computed, again seeking a .00625 level of significance. As was mentioned above, this question measures amount of faculty influence, not administrator influence as is suggested by the hypothesis. Consequently, the results are only informational, to see if, in fact. lower ranked faculty perceive any less faculty influence than higher ranked faculty. The results revealed no significant differences on amount of faculty influence as perceived by the lower ranked versus the higher ranked 94 faculty. Table 4.24 indicates the means, variances, and Z scores for each items Table 4.24 Z Scores for Comparison of Lower Ranked vs. Higher Ranked Faculty for Amount of Faculty Influence Lower Ranked Higher Ranked Z Item Faculty ,_Faculty, Score - ’2 - 2 .1. 11.12. i2_ Curriculum 3.70 3.73 -.30 Hiring of new faculty 3.38 3.39 -.03 Development of faculty 3.19 3.05 .80 personnel policies (reappointment, pro- motion, tenure) Selection of a depart— 2.29 3.30 -.06 ment chairperson De artment budget 2.23 2.01 1.38 a locations Faculty load deter- 2.46 2.42 . .29 minations Criteria for graduate 3.35 3.41 -.45 student admission Selection of an aca- 2.62 2.48 .90 demic dean From the above data, there is no evidence to support the hypothesis that lower ranked faculty perceive more administrator influence than higher ranked faculty. 95 Hypothesis #6 - The difference between perceived and preferred pattern of influence will be reater for the two lower ranks of aculty than for the upper two ranks. The investigation of this question was conducted in the same way as for the second hypothesis. That is, the number of faculty indicating congruence on a particular issue (i.e., the same group has the most influence that should) was com- pared with the number indicating dissonance (a different group having the most influence from.Who should) for the two groups under consideration. Here, the groups are Professors/ Associate Professors and Assistant Professors/Instructors. As in the previous hypothesis that asked this question for Faculty versus Administrators, four-celled tables were developed, in the following form; Congruence Dissonance Professors/Associate Professors I I I Assistant Professors/InstructorsIo I #J The initial test statistic computed for the resulting four-celled tables was, again, the Chi-Square Test of Inde- pendence, to determine if Congruence or Dissonance was independent of faculty rank. A Yule's Q coefficient was then calculated to determine direction and strength of the rela— tionship. The Chi-Square coefficients resulting from.the compari- son by rank showed significance on only two items: appoint- ment of a department chairperson and appointment of a 96 provost. The Yule‘s Q coefficients were then computed, and the two on the significant items resulted in fairly IOW’Co- efficients, one of which was positive and the other, negative. The first coefficient, on the issue of appointing a chair- person, was negative, indicating, for the way the table was constructed, that the higher ranked faculty were more likely to have congruence on the item, and lower ranked faculty, dissonance. This result is in support of the hypothesis. On the second significant item, however, appointing a provost, the positive coefficient indicated that lower ranked faculty were more likely to have congruence and higher ranked faculty, dissonance. This is, of course, in opposition to the hypoth- esis. Table 4.25 indicates the Chi-Square and Q coefficients. 97 Table 4.25 Chi-Square and Yule's Q Comparisons for Lower Ranked vs. Higher Ranked Faculty on Congruence and Dissonance Item Chi-Square Yule's Q Appointing a department chairperson 6.56* -.30 Appointing an academic dean 1.16 -.14 Appointing a provost 4.26* .24 Creating new educational programs .001 -.01 Determining faculty salaries 3.20 .22 Determining if a temporary faculty 1.04 -.06 faculty member should be rehired Determining tenure for faculty .03 -.02 members ' Determining undergraduate university .06 .07 admission policy Determining internal university 1.09 .14 budget allocations Determining university tuition 1.34 -.24 and fees Disciplining a student for cheating 1.20 -.15 on an exam Hiring new faculty members .17 .05 *Significant at the .05 level or better with 1 df In reviewing the results of the data, there appeared to be no evidence to support the hypothesis that lower ranked faculty see more difference between who should and who does have the most influence than higher ranked faculty. There 98 do not appear to be any real meaningful differences based on the higher-lower division of ranks. Comparison With 1970 Data Research Question #3 - What changes in faculty perceptions of decisiondmaking in the university have occurred since the late 1960's and early 1970's? As was mentioned earlier, no specific research hypoth- eses were developed for this comparison over time, since the key items from the 1970 MSU Omnibus Survey were not useful for the current study and thus could not be replicated. The format was changed enough that statistical comparisons were not possible. Consequently, only descriptive comments will be presented here. This section will first describe some characteristics of the two faculty respondent groups and how they compare. Secondly, observations made concerning percep- tions of influence will be summarized. Finally, observations made on the comparisons of other related items for the two groups will be described. For the purposes of comparison, the faculty members in the sample in 1970 who were affiliated with one of the five colleges used in the 1977 study were used. The total number in that group was 118, compared with 347 in the later sample. The same variables (rank, sex, and age) used to compare the respondents to the population in the present were examined here for comparisons. The distribution by rank showed a larger percentage of Professors and a smaller percentage of 99 Instructors now than in 1970. In a period of financial stringency and little job mobility, that change is not a surprising one. Along with that is an expected age change since 1970. This year's sample was somewhat older, with 56% over 40, compared with 50% in 1970. At the other end of the age scale, only 24% were under 35 this year, as compared with 34% in 1970. The comparison by sex showed very little change, which is interesting in light of the nationwide efforts in affirmative action. Tables 4.26, 4.27 and 4.28 reflect these comparisons. Table 4.26 Comparison by Rank for the 1970 vs. 1977 Faculty Respondents Rank 1219 1211 Professor 33% 42% Associate Professor 21% 23% Assistant Professor 26% 28% Instructor 20% 8% TOTAL 100% 101% 100 Table 4.27 Comparison by Age for the 1970 vs. 1977 Faculty Respondents Age Groop 1919 1911 Under 25 0% 0% 25-29 16% 7% 30-34 18% 17% 35-39 16% 20% 40-44 16% 14% 45-49 7% 12% 50-54 11% 9% 55-59 7% 11% 60 and over 9% 9% TOTAL 100% 99% Table 4.28 Comparison by Sex for the 1970 vs. 1977 Faculty Respondents sea L919. 91.7. Male 76% 78% Female 24% 22% TOTAL 100% 100% lOl Regarding the questions directly relating to perceptions of influence, very little conclusive information could actually be gained. One way the data were examined was to compare the responses on the question of "who should" have influence over particular issues. Though the question was worded differently in 1977 than in 1970, it was similar enough to compare. The same seven response categories were available on both surveys; A (only administrators), F (only faculty), S (only students), AFS (all three groups), AF, AS, and FS (only the respective two groups). In 1970, the question was "What group or combination of groups should have the most influence over each of the following decisions?" In the present study, the question was "What group or groups should be involved in the following decisions?" Since the same response categories were possible, the data appeared to be comparable. On examination, the only difference of any consequence was that on several academic issues in 1977, faculty were less inclined to say that faculty alone should be involved in the decision than was the case in 1970. Faculty in the current sample were more likely to say Administrators and Faculty (AF) or Administrators, Faculty, and Students (AFS) should be involved than were their 1970 counterparts. The differences were not great, however, so it would be diffi- cult to say that faculty preferred less involvement now than they did in 1970. 102 On the question asking "Who actually has the most influence?", the difficulties in comparison are even greater. As was mentioned above, faculty in 1970 were given the same seven choices previously described: A, F, S, AFS, AF, AS, or F8. They were not asked to necessarily select one group over another, whereas in the present study, respondents had to select A, F, or S. In 1970, on the academic issues, from 33% - 67% of the faculty marked the AF category. With such a large percent in that category, it would be most difficult to allocate that percent to either A or F without the risk of severely misinterpreting the intent of the respondent. One comparison that was made on the two sets of data was between those who did select A or F alone as having the most influence. Though percentages were considerably lower in 1970 because of the other options available, the relation- ship of the percentage who chose A to the percentage who chose F was quite similar on four of the seven academic re- lated issues. For example, on the issue of creating new educational programs, the percentage selecting A and F alone in 1970 were 16% and 13% respectively, and in 1977, were 52% and 48%. In each of the four cases, if more saw A as having the most influence in 1970, more also saw A as having the most influence in 1977. The exceptions to that similar- ity were on the issues of appointing a chairperson, appoint- ing an academic dean, and determining tenure for faculty 'members. For the chairperson and tenure issues, in 1970, 103 more faculty perceived that F had the most influence than perceived A to have the most influence. In 1977, the reverse was true. For the academic dean issue, in 1970, three times as many faculty saw A as having the most influence as saw F as having the most influence. In the current study, five times as many selected A. A summary of these results can be found in Table 4.29. Table 4.29 Comparison of Perceptions of Who Has the Most Influence for Faculty — 1970 vs. 1977 Issue 1970 1977 Aa Fb Aa Fb Appointing a department 11% 17% 56% 45% chairperson Appointing an academic dean 32 10 83 17 Creating new educational 17 13 52 48 programs Determining faculty 64 2 94 6 salaries Determining if a temporary 36 ll 67 33 facultv member should be rehired Determining tenure for 18 21 60 40 faculty members Hiring new faculty members 18 17 55 45 aIndicates percentage perceiving Administrators to have the most influence bIndicates percentage perceiving Faculty to have the most influence 104 The responses indicate that on those three issues, fewer faculty in 1977 perceived themselves as having the most influence than did in 1970. The interpretation is, of course, tentative, but it could be said that faculty perceive them- selves as less influential now than in 1970, at least on those three issues. This interpretation would lend some sup- port to the suggestion from the literature cited in Chapter II that faculty do, in fact, perceive themselves to be less influential now than in 1970. Several other variables which were included on both survey instruments were also compared for informational purposes. Though the data are almost an exact replication on those items, no statistical analysis was done, since those variables were not central to the thesis of this study. The use of the comparison will be to integrate any findings that seem appropriate into the discussion of conclusions in Chapter V. The first of these questions was: "To what extent do you favor the establishment of a collective bargaining unit for faculty?" The scale was a four-point scale, as follows: 4 - To a great extent To some extent 3 2 To a slight extent 1 - To no extent at all The results of the comparative analysis revealed quite a visible change, both for faculty and administrators. For 105 faculty, 33% in 1970 said they favored collective bargaining to a great extent, while only 19% said so in 1977. The percentage in the categories of "do not favor at all" or "favor to a slight extent" was 43% in 1970 and 56% in the present study. For administrators, 46% did not favor col- lective bargaining at all in 1970, a percentage which in— creased to 77% in 1977. In contrast, 30% said they favored it to a great or to some extent in 1970, while only 13% checked those two categories in the current study. Table 4.30 reveals the results of the analysis. Table 4.30 Comparison of Attitude Toward Collective Bargaining in 1970 and 1977 for Faculty and Administrators Favor To Favor To Favor To Do Not Great Some Slight Favor Group Extent Extent Extent At All TOTAL 2 L E. L E. L. E. L E. L Faculty-1970 38 33 28 24 16 14 33 29 113 100 Faculty-1977 65 19 84 25 68 20 123 36 340 100 Administrators -l970 24 7 76 23 76 23 151 46 327 99 Administrators -1977 12 6 13 7 20 10 147 77 192 100 With the knowledge that the current faculty respondent group is somewhat older and at higher rank than the 1970 group, one might hypothesize that the change in attitudes toward collective bargaining are associated with older age 106 or higher rank. When the responses to the collective bar- gaining question were cross-tabulated with both age and rank, it appeared that there were very minor differences by age. With regard to rank, it appeared that Professors do in fact seem less inclined to favor collective bargaining than the other three ranks. Since there are a higher per- centage of Professors now than in 1970, that higher percen- tage could account for some of the shift. The other items in the current study that were compared with 1970 responses were those dealing with priorities for faculty salaries, departmental emphasis, and attitudes on certain university issues. Generally, the responses indicat- ed that faculty attitudes and perceptions had changed very little on those issues. On the issue of priorities for salaries, the rank-order correlation appeared quite high, with teaching effectiveness, publications, and research continuing to be most important. The means on the items appeared quite similar, with research and public service up somewhat. On the departmental emphasis question, the respon- ses were again quite similar, with some increase over the seven-year period on applied research, and slight decreases of the emphasis on the advising of students, both graduate and undergraduate. The applied research item was the only one that had a somewhat different mean score and also changed several places in rank of importance. Faculty ranked it seventh, or last in 1970, and it was ranked third 107 of seven in 1977. (The reader will note that two of the items in the 1977 version of the question were not included in the comparison, as they were not on the 1970 questionnaire.) Tables 4.31 and 4.32 indicate these comparisons. Table 4.31 Rank-Order Correlation and Mean Comparisons for Salary Priorities - 1970 vs. 1977 Item 1970 1977 Rank Mean Rank Mean Academic advisement of 4 2.71 6 2.74 students Job counseling and career 7 2.59 8 2.55 guidance of students Personal values and ethical 6 2.63 5 2.75 standards Popularity with students 9 2.24 9 2.30 Publications 3 3.03 3 3.35 Research activities 2 3.21 2 3.54 Service activities in the 8 2.37 7 2.71 community Service activities in the 5 2.70 4 2.76 university Teaching effectiveness 1 3.89 1 3.83 108 Table 4.32 Rank-Order Correlation and Mean Comparisons for Departmental Emphasis - 1970 vs. 1977 Item 1970 1977 Rank Hean Rank Kean Instruction of under- 1 3.71 l 3.78 graduate majors Instruction of graduate 1 3.71 2 3.64 students Advisement of graduate 3 3.29 5 3.22 majors Advisement of under- 5 3.25 6 3.18 graduate majors Basic research 3 3.29 4 3.28 Applied research 7 3.00 3 3.31 Instruction of under- 6 3.06 7 3.02 graduate non-majors Continuing/life-long Not included ' 9 2.65 education Implementation of a Not included 8 2.66 strong affirmative action program The comparison on the attitudes about various university issues was difficult, because there were only four items that remained the same from 1970 to 1977. Those items, which were to be answered on a scale indicating extent of agreement, are listed in Table 4.33 with the appropriate means. 109 Table 4.33 Mean Comparisons for Various University Issues - 1970 vs. 1977 Item ' 1970 1977 Mean Mean In making student admissions deci- 3.12 3.39 sions, academic aptitude should be given the greatest weight. The percentage of graduate students 2.30 2.12 at MSU should be increased consid- erably above its present 20% level. Two-year community colleges would 3.15 2.72 probably better serve the needs of most disadvantaged students. This university should admit dis- 2.64 2.35 advantaged students who appear to have potential, even if they do not meet normal entrance standards. The biggest change was that there was less agreement in 1977 that two-year colleges would better serve the needs of disadvantaged students. The means on the other three items changed slightly, with agreement increasing on the question of giving academic aptitude the greatest weight in admissions decisions, and decreasing on the other two, increasing the graduate student percentage and admitting disadvantaged students who don't meet criteria. Generally, then, the comparison of the 1970 data with similar current information revealed that the respondent group was somewhat older, at higher rank, and less in favor 110 of collective bargaining. There was some tentative evidence of faculty perceiving themselves with less influence in 1977. They continued to see teaching effectiveness, research, and publications as highest priorities for salary considerations and indicated an increase in departmental emphasis on applied research. With the exception of the collective bargaining issue, the two groups appeared to be fairly similar. Summary The overall examination of the data proved to support some of the hypotheses but not others. For the first two hypotheses, which were that administrators will perceive faculty to have more influence than faculty themselves will, and that faculty will perceive mmre dissonance than adminis- trators will, there was some evidence of predicted results. On several key issues, administrators did tend to see faculty as having more influence than faculty themselves saw. This was more apparent when looking at perceptions of relative amounts of influence on particular items than on the question of which group has the most influence. Regarding the degree of "congruence" and "dissonance" for faculty and administra- tors, virtually all comparisons were significant in the direction predicted. The next set of hypotheses dealt with comparisons be- tween several subgroups of faculty: large and small colleges, higher versus lower quality units, and the upper two faculty lll ranks versus the lower two ranks. The predictions with regard to differences among the ranks and among the higher and lower quality units were not substantiated to any extent at all by the results. However, both measures of the difference between large and small colleges showed quite convincingly that faculty in larger colleges see themselves and faculty in general as having more influence than did their colleagues in smaller units. In comparing the 1970 and 1977 data, the only inferences regarding perceptions of influence that could be made gave tentative suggestion that faculty perceive themselves as less influential now than in 1970. Overall, the 1977 respondent group appeared to be older, at higher rank, less in favor of collective bargaining, and saw applied research as a higher priority. Chapter V, the final chapter, will contain a discussion of the findings with regard to conclusions, interpretations, and recommendations. V. CONCLUSIONS AND RECOMMENDATIONS Introduction This chapter will include a summary of the purpose and need for the study, a review of the results as they relate to the individual research hypotheses, observations and spec- ulations, recommendations for further study, and a summary. Purpose and Need for the Study The purpose of this study was to examine the perceptions held by faculty and administrators of power and influence in the decision-making process at a large, public midwestern university. More specifically, the following research ques— tions were explored: 1) How do faculty, as compared with administrators, perceive decision-making in the university? 2) Are there predictable differences in perceptions of power and influence among various subgroups of faculty? 3) What changes in faculty perceptions of decision- making in the university have occurred since the late 960‘s and early 1970's? The issue of faculty perceptions of power and influence in higher education has recieved a good deal of attention in recent years (Blau, 1973; Gross and Grambsch, 1974). One reason for that attention is the fact that faculty, as 112 113 professionals in a bureaucratic organization, hold a unique position. As professionals, they have loyalties and goals related to their profession, while being a part of an organ- ization which may have very different goals. They are subject to peer judgment within their professional area, and to supervision and evaluation by their superior in the organiza- tional hierarchy (Scott, 1966). In addition, faculty share and have direct involvement in the major goal activities of the organization, teaching and research, and in fact, often have an expertise held by no one else in the organization (Etzioni, 1964). This factor alone makes universities dif— ferent from other bureaucratic organizations, and makes the study of decision—making in the university of interest to students of higher education. Gross and Grambsch (1968; 1974), Blau (1973), and Demerath, Stephens, and Taylor (1967), among others, have studied perceptions of power and influence in universities in recent years. In the last several years, however, a number of changes have occurred which make the university a different place. The pressures of accountability and finan- cial stringency have brought about new kinds of management systems and new ways of planning, accomplishing andevaluating institutional goals. The emphasis has shifted from keeping peace on the campus to keeping students in the face of declining enrollments. Thus, in a complex organization such as the university, with professionals as a major part 114 of the bureaucratic structure and new priorities and problems, it would seem that a study such as the present one would lead to a greater understanding of the perceptions of power and influence in a modern university. The population studied included faculty members in five of seventeen colleges in a major public midwestern university and academic administrators in that same university. There were 627 faculty members in the sample and 288 administrators. A survey, asking questions about perceptions of influence in decision-making as well as other university issues, was sent to these faculty and administrators, with a return rate of 59%, or a total of 544. In addition, data from a similar study done at the same institution in 1970 were used for comparative purposes. Results and Conclusions In this section, each hypothesis will be discussed, with the intent of briefly summarizing findings and presenting some interpretive comments. Hypothesis #1 - Compared with faculty, administrators will perceive more influence on aCademic matters. In developing this hypothesis, several previous studies led to the prediction that administrators would perceive more faculty involvement than faculty would perceive for themselves. Though they found that faculty and administrators often have similar perceptions Gross and Grambsch found that 115 high level administrators saw faculty as higher in relative power or influence than faculty themselves saw (1974). Similarly, Hodgkinson (1974) and Kemerer and Baldridge (1975) found that administrators were much more optimistic about the power of faculty senates than faculty were. The data from the first question indicated that on only four of ten items did administrators in fact see faculty as more involved than they perceived themselves to be. On the second question analyzed, which measured amount of faculty influence, data on six of the eight items indicated that faculty perceived significantly less influence for themselves than did adminis- trators. Since one analysis obviously indicated more consis- tent differences than the other, perhaps the issue is that when asked who has the most influence, or who really makes the final decision, both groups are similarly realistic about areas in which faculty do and do not have major impact. There were not many issues on which faculty or administrators felt faculty had the most influence. When asked to assess, on an arbitrary four-point scale, how much influence faculty have, administrators clearly saw the institution as more democratic, or with a greater degree of faculty involvement, than faculty did. It is possible that while administrators are willing to admit that they, rather than faculty, have' the most influence over an issue, they overestimate the amount of influence faculty have in some of those same or similar areas. 116 Hypothesis #2 - The difference between perceived and preferred influence will be greater for facultythan for administrators. There has not been a great deal of research done on this particular issue, and in fact, the major literature source that gave support to this idea came from a study of a business organization, where the difference between per- ceived and preferred power was greater for workers than for supervisors (Smith and Tannenbaum, 1968). The results of the survey clearly supported the hypothesis, as on every item, faculty indicated more dissonance than administrators. And, as might have been expected, the dissonance most often occurred where faculty perceived that administrators had the most influence but the faculty themselves should. The dissonance was the most pronounced in determining faculty salaries and internal budget allocations. These two issues are ones that have impact at the department level and that probably are very important to faculty. The dissonance expressed by faculty may indicate any of several things. It could, in fact, mean that they really feel that some other group, in most cases, themselves, should make decisions than the group that actually does. However, Blau (1973) and the Carnegie Commission (1973) observed that while faculty are often dissatisfied with the power or influence they have, they are often reluctant to be more involved themselves. Perhaps it is just the "appropriate" thing for faculty to indicate, in whatever way, that someone 117 has power that someone else should have. Or, perhaps admin- istrators are really not as satisfied as they seem, but for them, the appropriate thing is to indicate that "all is well" with the decision-making apparatus. Part of the role of the administrator is to implement whatever structure there is for making decisions, regardless of who has the influence or power, so it is possible that they either have been condi- tioned to perceive or want to perceive that things exist as they should. Examining the results of the first two hypotheses together, data indicated that faculty perceive themselves to have less influence than do administrators, and perceive more dissonance with the relationship between who has and who should have power and influence. Whether faculty really want more influence in the system than they have or just want to express dissatisfaction with what exists is a question that remains to be answered. Hypothesis #3 - Facultyin largeycomplex colleges will perceive more aculty influence on academic matters than those in small colleges. The major support for the generation of this hypothesis was from Blau (1973) and Baldridge, et. a1. (1973), both of whom found that large size and complexity led to perceptions of greater faculty involvement in decision-making. Their studies were conducted across a number of universities, which meant that the units under study were somewhat similar. Consequently, conducting the same kind of study with 118 individual (and very different) colleges as the units of analysis seemed to be a logical extension of Blau's and Baldridge, et. al's. findings. Of all the comparisons between subgroups of faculty, this one most clearly showed significant differences. On the two questions used, faculty in the two large colleges perceived themselves to be more influential than did their colleagues in the small colleges on almost all items. This finding goes against prevalent beliefs that as institutions get larger and more complex, the bureaucracy gets larger, and people feel more alienated and less involved. These findings seem to reveal just the opposite perceptions, and may, along with the studies cited above, help counteract the myth that if faculty want to have impact on decisions that affect them and feel involved in the decision-making struc- ture, the place to be is at a small college or in a small academic unit. Actually, there is a logical explanation to a finding such as this, as was mentioned by Baldridge, et. al. As institutions get larger and more complex, it becomes more difficult for a central administration to maintain a high level of power and control. Consequently, decentraliz- ing, or having more decisions made at the college or depart- mental level, becomes necessary. It may be the case in this study that the faculty in the larger colleges do not have any more influence at the department level than the faculty in small colleges, but rather that the departments themselves 119 have more key decisions in their hands in the two large colleges. Consequently, the result is the faculty perceiv— ing a high degree of influence. Another possibility, which should be mentioned at this point in the discussion of the results, relates to the fact that of the seventeen colleges in the university under study, the five used in the sample were selected somewhat arbitrar- ily, though there were guidelines based on previous research. Consequently, it is possible that the differences attributed to size might in fact be related to the nature of the colleges involved. Perhaps these results should be inter- preted with some caution until that issue can be explored further. In any event, the finding is worth noting. To attempt to gain some perspective on the concern men- tioned above, several other variables were examined to deter- mine if there might be other conditions affecting this per- ception along with size of the unit. The large and small colleges were compared on the question which asked for a rating of the respondents' own department, thinking that self-image may have an impact on perception of influence. As will be explained more with the next hypothesis, that comparison did not yield any significant results. In fact, the two large colleges were not at all close on the self-rating variable. A last comparison made was to examine large versus small colleges on the variable that asked the faculty members to rate their own satisfaction. Again, there was no 120 clear cut evidence that size was related to satisfaction, or that satisfaction was related to high perception of influence. There is a great deal more examination that should be done on the size variable, to try to determine if there are other conditions which affect perceptions of influence. Even with the necessary caution, the finding is a most significant one. Before going on to the remaining hypothesis that relates to differences among the five colleges in the study, one additional problem in data interpretation should be mentioned. As was stated above, the fact that the five colleges may have some differences due to the disciplines represented might have introduced some biases into the research results. In addition, it is also possible that the deans of the particular colleges and their leadership styles or person- alities might have an impact on faculty perceptions. That could be true regardless of which units were involved. No attempt was made to assess or control for these variables. Consequently, it should be stated as an additional lflmita- tion of the study that comparisons between colleges might be compounded by the variables mentioned. Hypothesis #4 - Faculty in units determined to be of higher quality will perceive more faculty influence on academic matters. Again, the basis for this hypothesis was the 1973 study by Blau on universities. When using high salaries, research involvement, and several other variables to determine quality, 121 he found that high quality, like large size, was related to high perceptions on the part of faculty of their own influ- ence. His study was done on a number of universities, so comparisons on variables such as high salaries and research involvement were made on similar units. In the present study, however, the nature of the colleges involved was sufficiently different such that those comparisons were difficult. Consequently, an internal index of quality was determined, as explained in the previous chapter, on the basis of data from two survey items: faculty rating of their own department and administrator rating of the five colleges. The results revealed no significant differences on the basis of quality determined in that manner. A pos- sible explanation of that lack of significance would be that the index used did not really measure quality. The best way to have determined high or low quality would be to have acquired some external comparative rating of the five colleges, but that was not easily possible. It is possible that there really is no valid way to measure comparative quality, at least not with units as different as the ones in the study. It does not appear that this variable is clearly enough defined to warrant further reserach, at least not on vastly different units, unless a better measure can be determined. Hypothesis #5 - Assistant Professors and Instructors will perceive administrators to have more influence than Associate PrOféSsors and Professors will. 122 Hypothesis #6 - The difference between perceived and pre- wferred influence will be greater for the two lower ranks offacultythan for the two upper ranks. These two hypotheses are reviewed together here, as comments about both are similar. Both Blau (1973) and Kemerer and Baldridge (1975) found evidence in their re- search that the two upper ranks of faculty were more involved in the decision-making process than the lower ranks, and also, that lower ranked faculty often felt more alienated. The prediction that faculty in the two lower ranks would perceive themselves to be less influential and would see more difference between who has and who should have influence was a logical extension of the previous findings, with regard to faculty ranks. When analyzed, however, the results indicated virtually no differences on the basis of rank. Of the three individual items where differences did appear, two were contrary to the direction predicted. One factor that could explain why the current study revealed no differences by rank, while research done five or six years ago did indi- cate such differences, is that because the financial situa- tion in universities has reduced faculty turnover, many of those who were at the lower ranks five to seven years ago may still be faculty members but at higher ranks. The current younger faculty at the lower ranks may be different than the lower ranked faculty of several years ago. If the young Instructors and Assistant Professors of the campus dissent era were the dissatisfied, alienated ones, and they 123 have maintained their perspectives and philosophies, that dissatisfaction and alienation has, if nothing else, prob- ably moved up a rank. And, with new problems and university priorities, the dissatisfaction may have tempered. Many of the younger, lower ranked faculty today were students during the campus dissent years, and it would be difficult to pre- dict where they are on some of the variables. It is entirely possible that there are conditions affecting these percep- tions other than rank, and in fact, size of academic unit seems to be one. The variable of academic rank and its im: plications for perceptions of power and influence probably warrants further investigation. Comparisons: 1970-1977 Of the various data comparisons made between the data gathered in 1970 and the current study, one of the most notable results was the change that occurred over the seven year period on attitudes toward collective bargaining. The entire respondent group, and particularly the administrators, indicated they were considerably less in favor of collective bargaining for faculty now than in 1970. Regardless of the changes in faculty attitudes, it might be useful to again look at the change in percentages at each faculty rank in the last seven years. The most obvious difference is that the percentage of Professors has increased considerably. When the collective bargaining issue is examined by rank, 124 the only group different than the others is the group of Professors, and they are obviously less in favor of collec- tive bargaining. Consequently, if there are more Professors now than in 1970, it is understandable that there might be less inclination toward collective bargaining. Regarding the change in administrative attitudes, one factor might be that in the past seven years, those administrators have had the opportunity to observe other institutions with unionized faculty, and also to experience the first few years of a clerical-technical union on the campus under study. If reac- tions to either of those observations were negative, admin- istrators' attitudes may have become more decidedly negative. Two additional findings merit some discussion. One is the tentative support to the notion that faculty perceive themselves as having less influence now than in 1970. An issue that was not investigated, but which might have some importance is that even if faculty do not perceive themselves as having as much influence, it may not be quite as much of a concern as in 1970. The changes in priorities of the university may have changed priorities of faculty, and it is possible that academic governance is no longer the hot issue it once was. It would have been interesting to have secured data on "dissonance for both time periods. Since that was not possible, the decrease in the importance of the govern- ance issue is only speculation. The final item in the 1970-77 comparison that merits 125 discussion is the increase in importance of applied research. In units with a significant research component, the question of applied versus basic research is an important one. With the environment and the energy situation becoming increasing- ly critical problem areas, it is a predictable result that research leading to solutions of these problems must increase in importance. Consequently, that change in departmental priorities is understandable. Observations and Speculations In reviewing the entire study, several subjective and speculative comments and observations of the researcher are in order. First, in one sense, the results of the study were somewhat disappointing, in that very little actual compari- sons could be made between the 1970 and 1977 data. When the study was begun, the comparison between the two groups was a major purpose of the study. However, no method for com- paring the data on power and influence, to any reasonable extent, was found. In a purely subjective sense, the re- searcher would still hypothesize that faculty do in fact perceive themselves to be less influential in 1977 than they did in 1970, but that is not as much of a critical issue. A second speculation is that administrators would also per- ceive themselves to have less influence in 1977. Consequent- ly, if each group (and probably students would be included) 126 perceive themselves to have lost power, the question of where the power is is a logical one. One additional comparison between the two sets of data on the collective bargaining issue merits an observation. The changes in attitudes on that question were perhaps the most dramatic of the issues investigated. Before that change is assumed to be significant, however, the reader should be reminded that only five of the seventeen colleges within the university were studied. The relationship of those colleges to the university as a whole in 1970 on that particular issue needs to be examined in more depth. When compared as a group to the university faculty as a whole in 1970, the attitudes were quite similar. The numbers were much smaller in 1970, however, and there probably still ought to be further exploration. Turning to an issue related to one mentioned earlier, i.e., that both faculty and administrators may perceive themselves to have lost influence, it should be noted that a similar pattern emerged with the 1977 data. Though the measurement was not over time, it was in fact the case that both administrators and faculty indicated that not they, but the other group, has strong influence over a number of issues. Consequently, everyone appears to perceive that someone else has the power. It would be interesting to speculate on the implications of this perception for uni- versity governance. 127 First, if each group says they do not have the power to decide an issue, then it is much easier for that person or group not to take responsibility for that decision. Sec- ondly, if individuals in the governing process all attribute the power and influence to someone else, the whole process becomes muddled and difficult to understand. If that happens, then it would seem to be most difficult to establish a clear governance system and especially, to determine responsibility and accountability. Further study on the issue might help in determining whether those who perceive themselves not to have influence but indicate that they should, i.e., the faculty,really want to assume more responsibility or whether they prefer to assign the responsibility to someone else and merely express dissatisfaction with the situation. A final observation relates to the study as a whole. Though some of the findings may add to the knowledge of faculty and administrator perceptions of power and influence, there is some speculation on the part of the researcher that only the surface of some of the issues has been examined. Perhaps that is the case for many who are somewhat inexper- ienced in the area of survey research. In any event, there appears to be another level of analysis that should be attempted before thorough understanding of the issues is possible. The following section will indicate some of the specific areas that would lend themselves to this additional analysis. 128 Recommendations and Implications for Further Research The following issues appear to merit further investiga- tion: 1. The issue of differences between faculty and administra- tggg. This merits further exploration both because the evi- dence for the predicted differences was not overwhelming and because, as was stated previously, such differences as were explored could have important implications for university governance. 2. The issue of dissonance for faculty members, or the difference between who should and who does have influence. This issue needs further exploration as it relates to satis- faction and also to real desire to be more involved. The term "satisfied" was used in connection with congruence, but there may be no real justification for using that label. Also, to gain more information on the meaning of the expres- sed dissonance, a future study could perhaps differentiate between a faculty member's perception that faculty in general should be more influential and the desire of that individual faculty member to assume that responsibility. It may be that faculty say they are not involved enough in key deci- sions, but do not want the responsibility themselves. 3. The issue of size of unit. As was mentioned earlier, this finding is an important one, but perhaps before any broad conclusions are made, the question ought to be explored further, controlling for other variables that might have an 129 impact on perceptions of influence, such as inherent dif- ferences in the nature of the unit or college. The results of the analysis, which are consistent with several previous studies, certainly merit refinement and further study. 4. The issue of faculty rank. Since faculty rank has been shown in some previous studies to have an impact on issues related to power and influence, and in the current study was not a significant variable, perhaps some additional research could determine what changes in the university msy have brought about changes in the attitudes of faculty at certain ranks. The final comment is a recommendation related to the methodology that might be worthy of consideration if further study in the area is attempted. The use of faculty from five colleges for the sample rather than drawing from the entire university was useful in that a larger number of faculty from each unit was in the respondent group. However, as was noted previously, it is possible that there were differences among the colleges that were not controlled for in the measurement of the variables of interest. Consequent- 1y, if further research involves the same type of comparisons, some attempt should be made to control for, or at least assess, characteristics of the unit or of the chief academic officer in which might affect perceptions of the faculty. 130 Summary The purpose of this study was to examine the perceptions held by faculty and administrators of power and influence at a large university. The results showed that faculty perceive themselves to have somewhat less influence than administrators ascribed to them and perceive more dissonance between who does and who should have influence than do administrators. Among the faculty, those who perceived themselves with high influence were those in large colleges, a most significant finding in that the common belief is that large size leads to bureaucracy and alienation. Quality of the unit, as measured by an internal index of quality, was not related to perceptions of influence, nor was academic rank. One pos- sible explanation for the latter finding is that some of those faculty who were the younger, lower ranked faculty five to seven years ago might now be at higher ranks, blur- ring the distinctions between the ranks. The most meaningful changes that have occurred on related items since 1970 are that faculty and administrators are less in favor of collec- tive bargaining and faculty perceive applied research as a higher department priority. The first finding could be explained in that there are more Professors now than in 1970, and they are as a group less inclined toward faculty unionism- There is some speculation on the part of the researcher that the study did not explore the issues in sufficient depth, and recommendations for further study ought to attempt another 131 level of analysis on the issues of importance. Those issues that merit further consideration include the differences between faculty and administrators, the dissonance expressed by faculty, and the effects of size of unit and faculty rank on perceptions of power and influence. APPENDICES 132 MICHIGAN STATE UNIVERSITY APPENDIX A - COVER LETTER TO FACULTY WAWWWWY MW’W'M March 7, 1977 Dear Faculty Member. Barbara Stonewater and I plan to replicate a study of facultybadministrator relationships conducted at Michigan State University in 1970. (Copies of the original report are available upon request.) The research will yield data about current perceptions and attitudes, as well as changes from a time of turmoil to one of relative calm. As in the previous study, both faculty and administrators will receive questionnaires. Summaries of our findings will be available to all in- terested parties: no person, department or college will ever be identified. we consider all data confidential. we have consulted members of the Faculty Associates, AAUP, Academic Council and University Administration to assure that questions reflect many different perspectives. rive colleges, including yours. have been selected for the faculty portion of this study. we hope you will cooperate by completing the attached questionnaire: our pretest shows it will take less than fifteen minutes. we would like to have the completed forms by Rednesday, March l6. A self- addressed campus mail envelope is enclosed for your convenience. Thank you very much for your help. Sincerely, I) ~ ’ . ‘._ -...__ ,« 49, /7,. /2; (W..._,. QLWIW Philip M. Marcus Barbara A. Stonewater Professor Specialist Department of Sociology Department of Electrical Engineering 8 Systems Science ID 9'. 133 APPENDIX B - FACULTY QUESTIONNAIRE FACULTY-ADMINISTRATOR RELATIONSHIPS QUESTIONNAIRE-1977 l. The following alphabetical list of university policies require decisions by some combination of Administrators, Faculty, and Students. which group or groups should be involved in some way in that policy decision. Mark as many as are applicable. In the next column, indicate which pp; group should have the most influence. Then, in the far right column, please indicate the group that actuaII has the most influence. For example, Ia Ea I shou nvolved in the decision, if for decision 1, students should have the most influence, but faculty actually have the most influence, you would’respond? Should Have Please indicate with an "X” faculty and students Actuallv Has Administrators Faculty Students Most Influence Most Influence A F S IA, F, or 8) IA. F, or 51 3. __§_ x S F Who should be involved in decision? CHECK AS MANY AS APPLY MARK ONLY ONE MARK ONLY ONE Should Have Actuall Has University Administrators Faculty Students Most Ianuence Most In luence Policy Matters A F S IA, F, or S) A, P, or S) Appointing a department chairperson ILJ I I I I I I I I Appointing an academic dean IL 1, 111 11.1, I 1 I 1 Aepointing a provost I 1 111 I l I J I 1 Creating new educational 2£99rams I J I I I l I I I 1 Determining faculty salaries [.1 II I I II 1111 I 1 Determining if a temporary faculty member should be rehired I I I I I I I I I I Determining tenure for faculty members L,L_ I I I I I I I I Determining undergraduate university admissions policy I 1 [41 I l [ l I I Determining internal university budget allocations Determining university tuition and fees I I Disciplining a student for cheating on an exam Hiring new faculty members GO TO NEXT PAGE 134 2 2. How much priority should each of the following be given when considering faculty salaries? CHECK ONE ALTERNATIVE FOR EACH LINE A high Some Little Hp priority priority priority priority at all Academic advisement of students I ] I 1 I I [ I Job counseling and career gpidance of students I,] III I I IIJL Personal values and ethical standards I I I I I I I I Popularity with students I 1 I71 [41‘ L41 Publications I j I I I l I I Research activities I_1 I l I I I 1 Service activities in the community I J I I I L I1J_, Service activities in the university (etg,, committees) I ] I I I I I 1 Teaching effectiveness I 1 I l 4I_I, IAI I 3. To what extent do you favor the establishment of a collective bargaining unit for M.S.U. faculty members? CHECK ONE To a great extent _.I_J___ TO some extent .L_1.__ To a slight extent ___I_l_ Do not favor at all __I_I_____ 4. What impact do you feel the establishment of a collective bargaining unit would have on the following changes? CHECK ONE ON EACH LINE A great impact Equalizing faculty salaries across units I I Improving the overall economic status of the faculty I I Reducing the merit basis of salary increases I 1 Acquiring additional funds from the legislature I I Giving faculty greater involvement in decision-making Providing greater job security, I l I I GO TO NEXT PAGE Some impact I I A slight impact I I No impact at all I I a. b. c. d. 135 3 S. In general, how much influence do universityyfaculty have over the following matters? CHECK ONE ON EACH LINE A great A moderate A slight Hp deal of amount of amount of influence influence influence influence at all Curriculum I I I l I I I Hiring of new faculty IIJ I I I I I Development of faculty personnel policies (reappointmenty promotion, tenure) I,I I,l‘_ I I I Selection of a department chairperson I I I I I I I Dgportment budget allocations I 1 I | I l I I Faculty load determinations I I I I I I I I Criteria for graduate student admission I 1 I I If] I I Selection of an academic dean I l 5 I I I I I I 6. Compared to other faculty in your department, how much influence do you feel you have on the following departmental matters? CHECK ONE ON EACH LINE More than About the Less than other faculty same other faculty Curriculum I I I I I I Hiring of new faculty I l I I I I Development of faculty personnel policies (reappointment, promotion, tenure) I I I I I I Selection of a department chairperson I I I I I I Department budget allocations II I I I I I Faculty load determinations I I I I I I Criteria forygraduate student admission .1 1 I 1 I I, Selection of an academic dean II 1, IA] I I 7. How would you describe your personal satisfaction as a faculty member and the satisfaction of other faculty members you know? My satisfaction is: Very high . . . . Fairly high . . . Fairly low. . . . Very low. . . . . O O O O I I O O C O O O O O O O O O I I 8. How much emphasis should be placed on each of the following areas within your particular department? CHECK ONE ON EACH LINE CHECK ONE IN EACH COLUMN The satisfaction of other faculty seems to be: A great Some A slight Hp emphasis ppphasis emphasis emphasis at all Instruction of undergraduate majors I 1 III I I I I Instruction of graduate students I I I I I I I I Advisement of graduate mdjors III II], I I I I Advisement of undergraduate majors I 1 [4I_, I I I I GO TO NEXT PAGE 136 4 A great Some A slight Np emphasis emphasis emphasis gmphasis at all e. Basic research I I III I I III f. Applied research I I I l I I I I 9. Instruction of undergraduate non-majors II II, I 1 I I I J h. Continuing/life-long education I ] I l I l I l i. Implementation of a strong affirmative action pregram II 1 I 1 III III 9. To what extent do you agree with each of the following statements about current university issues? CHECK ONE ON EACH LINE Agree to a Agree to Agree to a Do 225 agree great extent some extent slight extent at all a. Procedures for reappointment of I faculty are generally fair to the faculty members involved III I I, I l I I b. In making student admissions decisions, academic aptitude should be given the greatest weight I l I I III I I c. Giving college credit for remedial courses is justified III I I I I I_I I, d. Eliminating academic programs or depart- ments is a legitimate means of budget reallocation II I III. I ] I I e. The percentage of graduate students at MSU should be increased considerably above its present 20s level ij I I Igj I 1 f. The current grievance procedures for faculty are adequate I l I I I I II_I g. Two-year community colleges would probably better serve the needs of most disadvantaged students II I? I I I ] I I h. The university is dealing with its current budgetary problems in the most reasonable way possible I I III, I l ._I II i. University rules are often ignored by faculty I l 1,1 [,1 I I j. Life-long education is important enough to compete with other university programs for resources I 1 I4] I ] _Iy] k. There should be greater university coor- dination of programs, even if it means loss of unit autonomy _I I I I LJII I I l. The soon-to-be-appointed President's Planning Council will be a good means of assessing Universityipriorities I I I I I 1 II I, m. This university should admit disadvantaged students who appear to have potential. even if they do not meet normal entrance 2££ndards I I I I _I I I I GO TO NEXT PAGE 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 137 5 Which academic unit are you affiliated with? (If you have a joint appointment, where is your primary appointment?) CHECK ONE Arts 8 Letters Communication Arts Engineering Human Ecology Social Science Other Compared with other academic departments in your discipline nationwide, how do you rate your department overall? CHECK ONE ____In the top St ____Top 10‘ ____Top 25‘ ____Top 50‘ ____Eottom 50‘ ____Bottom 25% What do you consider your primary responsibility? CHECK ONE Administration _____ Research _____ Teaching What is your academic rank? CHECK ONE ____Professor ___;Associate Professor ____Assistant Professor ____Instructor ____Other How long have you been on the faculty/staff at Michigan State University? CHECK ONE ____less than 1 yr ____l-4 years ____s-1o years ____l0-20 years ____more than 20 years How long have you been in your current position or rank? CHECK ONE less than 1 yr 1-4 years 5-10 years 10-20 years more than 20 years Do you have a tenure track appointment? CHECK ONE Yes No Do you have tenure? CHECK ONE Yes No What is the highest college degree that you now hold? CHECK ONE N3 college degree A degree based on less than 4 years work (e.g., A.A.) Bachelor's degree (e.g., E.A.. 8.5.) Master's degree (e.g., H.A., H.S.) Professional degree requiring at least 3 years post graduate work (e.g.. L.L.B.. M.D.. D.V.M. Doctorate (e.g., Ph.D., Ed.D.) What is your sex? CHECK ONE _____ Hale _____ Female How old were you on your last birthday? CHECK ONE (a) Under 25 _____ (b) 25-29 _____ (c) 30-34 ______(d) 35-39 _____ (e) 40-44 _____ (f) 45-49 _____ (g) 50-54 _____ (h) 55-59 _____ (i) 60 and over Thank you for participating in this study. Please enclose the completed questionnaire in the envelope provided and place in campus mail. If you have any comments, criticisms, or suggestions, please write them in the space below. We appreciate your help. 138 MICHIGAN STATE UNIVERSITY APPENDIX C - COVER LETTER TO ADMINISTRATORS mmossoaowor mums-mamm-«m March 7, 1977 Dear Administrator, Barbara Stonewater and I plan to replicate a study of faculty-administrator relationships conducted at Hichigan State University in 1970. (Copies of the original report are available upon request.) The research will yield data about current perceptions and attitudes, as well as changes from a time of turmoil to one of relative calm. As in the previous study, both faculty and administrators will receive questionnaires. Summaries of our findings‘will be available to all in- terested parties: no person, department or college will-ever be identified. we consider all data confidential. we have consulted members of the Faculty Associates, AAUP, Academic Council and university Administration to assure that questions reflect many different perspectives. Faculty from five colleges and a sample of administrators have been selected to participate in this study. We hope you will cooperate by completing the attached questionnaire: our pretest shows it will take less than fifteen minutes. we would like to have the completed forms by Nednesday, March 16} A self- addressed campus mail envelope is enclosed for your convenience. Thank you very much for your help. Sincerely, ph Barbara A. Stonewater Professor Specialist Department of Sociology Department of Electrical Engineering 6 Systems Science 139 APPENDIX D - ADMINISTRATOR QUESTIONNAIRE FACULTY-ADMINISTRATOR RELATIONSHIPS QUESTIONNAIRE-1977 The following alphabetical list of university policies require decisions by some. combination of Administrators, Faculty, and Students. Please indicate with an "x” which group or groups should be involved in some way in that policy decision. Mark as many as are applica e. In the next column, indicate which one group should have the most influence. Then, in the far right column, please indicate the group that actualI has the most influence. For example, if for decision a, faculty and students shouIa Es Involved in the decision, students should have the most influence, but faculty actually have the most influence, you wouldirespond: Should Have Actuallv has Administrators Faculty Students Most Ianuence Most Influence A F 5 IA, P, or S) (A, F, or S) a. x x s F Who should be involved in decision? CHECK AS MANY AS APPLY University Policy Matters Appointing a department chairperson Administrators A I_l. Faculty P I 1 Students 5 girl MARK ONLY ONE Should have Most Ianuence IA, P, or S) I!) MARK ONLY ONE Actually Has E225 In luence AL P, or S) __L L Appointing an academic dean I 1 Appointing a provost I.l Creating new educational programs I l Determining faculty salaries [rL I l Determining if a temporary faculty member should be rehired I I Determining tenure for facultyymembers L,L I l Determining undergraduate university admissions policy I l I l Determining internal university budget allocations I l Determining university tuition and fees I l Disciplining a student for cheating on an exam I ) Hiring new faculty members GO TO NEXT PAGE I l Do 140 2 2. How much priority should each of the following be given when considering faculty salaries? CHECK ONE ALTERNATIVE FOR EACH LINE A high Some Little N2 priority priority priority priority at all Academic advisement of students [ 1 [ 1 1417 [Ajy Job counseling and career guidance of students I I III I I I I Personal values and ethical standards I I 1,] I l I I Popularity with students 1 1 1 1 [41 1_1 Publications 1 1p_ 1 1 1 1 [71 Research activities 1 1 1 1 41 1 1 1 Service activities in the communitx I l I L I 1 I 1 Service activities in the university (e.gg, committees) .[ 1 [A1 [ 1 1 1 Teaching effectiveness [,1 [g] 1 1 1 1 3. To what extent do you favor the establishment of a collective bargaining unit for N.S.U. faculty members? CHECK ONE To a great To some To a slight Do not favor extent _ extent extent at all __I_L___ -I_L__ ___I__I__ ____I__1____ 4. What impact do you feel the establishment of a collective bargaining unit would have on the following changes? CHECK ONE ON EACH LINE A great Some A slight No impact impact impact impact at all Equalizing faculty salaries across units I I I I I I I I Improving the overall economic status of the faculty I I I I I I I I Peducing the merit basis of salary increases I I I I fifiI I I I Acquiring additional funds from the legislature I I I I I I I I Giving faculty greater involvement in decision-makingg I I I I I I I I Providing greater job securityg, I I I I I I I I GO TO NEXT PAGE u... ,. HI! " we» I» 141 3 S. In general, how much influence do university faculty have over the following matters? CHECK ONE ON EACH LINE A great A moderate A slight No deal of amount of amount of influence influence influence influence at all Curriculum I I I I II I I I Hiring of new faculty I l I I I I III, Development of faculty personnel policies (reappointment, promotion, tenure) 41 1 [ 1, 41 1 I 1 Selection of a department chairperson 441 1 I 1 I 1 I 11 Department budget allocations 1 1 I 1 I 1 I ] Pacultygload determinations I 17 I 1 I 147 I I Criteria for graduate student admission I 1 I 1 I I 1.11 Selection of an academic dean L 1 I J I )‘ I417 6. Compared to other administrators in similar positions, how much influence do you feel you have over the following departmental matters? CHECK ONE ON EACH LINE A great A moderate A slight No deal of amount of amount of influence influence influence influence at all Curriculum I!) I I I J, I I Hiring of new faculty I I I III I_I I,I Development of faculty personnel policies (reappointment, promotion, tenure) I 1 II) I I I I Selection of a department chairperson 1111 I 1 I 1 I I Department budggt allocations I I I I I I I 1 Faculty load determinations I 1 141 I I I I Criteria for graduate student admission I I I 1 11 I I J Selection of an academic dean I 1 L 1 I I I I 7. To what extent do you agree with each of the following statements about current university issues? CHECK ONE ON EACH LINE Agree to a Agree to Agree to a Do not agree great extent some extent slight extent at all Procedures for reappointment of faculty are gEnerally fair to the faculgy members involved 1 1 I 1 1 ] I,I In making student admissions decisions, academic aptitude should be given the b greatest weight I I III I I I I Giving college credit for remedial courses is justified AL,] I 1 I 1 I 1 Eliminating academic programs or depart- ments is a legitimate means of budget reallocation I I J_] I I I I The percentage of graduate students at MSU should be increased considerably above its present 20% level I I _J I I I I I GO TO NEXT PAGE 142 4 Agree to a Agree to Agree to a Do not agree great extent some extent slight extent at all The current grievance procedures for faculgy_are adequate I I I I I I I I Two-year community colleges would probably better serve the needs of most disadvantaged students 171 1 1 1 1 [ 1 The university is dealing with its current budgetary problems in the most reasonable wgy_p9ssible I 1 [ 1 [111 1 1 University rules are often ignored by faculty I 1 I I II,1 I I Life-long education is important enough to compete with other university programs for resources I 1 [ 1 111 I ] There should be greater university coor- dination of pregrams, even if it means loss of unit autonomy 1117 I 1 I 1 I 1 The soon-to-be-appointed President's Planning Council will be a good means of assessipg,Universityppgiorities 1 1 1_1 I 1 I 1 8. What is your title? (Dean, Director, Assistant Provost, etc.) 10. ll. 12. l3. 14. 15. 16. 17. Are you an academic administrator (reports to Office of the Provost) or non-academic administrator (does not report to Office of the Provost)? cagcx one Academic Administrator Non-Academic Administrator Do you have faculty rank? CHECK ONE Yes No If yes, what is your rank? CHECK ONE Professor Associate Professor Assistant Professor Instructor Other Do you have tenure? CHECK ONE _____ Yes _____ No What is your 88X? CHECK ONE _____ Female _____ Hale How old were you on your last birthday? CHECK ONE (a) Under 25 ____ (b) 25-29 ____ (c) 30-34 ____ (d) 35-39 ____ (e) 40-44 ____ If) 45-49 ____ (9) 50-54 ____ (h) 55-59 ____ (i) 60 and over ____ How long have you been on the faculty/staff at Michigan State University? CHECK ONE ___less than 1 yr ___1-4 years ___5-10 years ___lO-ZO years ___more than 20 years How long have you been in your current position? CHECK ONE ___less than 1 yr ___1-4 years ___5-10 years ___10-20 years ___more than 20 years What do you consider your primary responsibility? CHECK ONE Administration Research Teaching GO TO NEXT PAGE 1(53 5 18. What is the highest college degree that you now hold? CHECK ONE Np college degree A degree based on less than 4 years work (e.g., A.A.) Bachelor's degree (e.g., B.A., 8.5.) Master's degree (e.g., H.A., H.S.) Professional degree requiring at least 3 years post graduate work (e.g., L.L.B., M.D., D.V.H.) Doctorate (e.g., Ph.D., Ed.D.) 19. In comparison with similar academic units nationwide, how would you rank the following colleges in this university? CHECK ONE ON EACH LINE In the Tap Top Top Bottom Bottom ggg_§1 £91 321 Egg sog 25$ a. Arts 5 Letters I I I I I I I I I I I I b. Communication Arts 11 I I I I I I I I I I I c. Engineering I I I I I I I I I I I I d. Human Ecology I I I I I I I I I I I I e. Social Science I I I I I l I I I I I I Thank you for participating in this study. Please enclose the completed questionnaire in the enve10pe provided and place in campus mail. If you have any comments, criticisms, or suggestions, please write them in the space below. We appreciate your help. llwh MICHIGAN STATE UNIVERSITY APPENDIX E - FOLLOW-UP COVER LETTER TO JAKE-TY ”WNW MW'W°“ REMINDER REMINDER April 1, 1977 Dear Faculty Member, About two weeks ago Barbara Stonewater and I sent you a brief questionnaire replicating a 1970 study of administrator-faculty relationships at MSU. If you have completed that questionnaire and returned it, we thank you very much and offer the enclosed for your files or other amusement. For those of you who have not completed the fifteen minute questionnaire, we hope you will do so as soon as possible so we can start analyzing the data and provide a public report before the end of this term. In order to make the comparison between 1970 and today meaningful, we need a very high return, and your completed questionnaire will help greatly. Only five colleges were selected for the study (to conserve costs) and each additional response takes on greatly weighted importance. Please remember the data are anonymous, and no person, department or college will ever be identified in the report. We have consulted members of Faculty Associates, AAUP, Academic Council and University Administration to assure that questions reflect many different perspectives. If you have not completed the questionnaire, please do so immediately. We need your answers to help us understand current perceptions and attitudes as well as changes that have occurred over the past seven years. Only those of you who actively work in the university can provide the data we need to answer important policy questions. All interested parties will receive a full report of the findings. We are sure we can count on your cooperation. Please mail the completed ques- tionnaire back to us in the enclosed envelope. Thank you again for your help in this important academic study. Sincerely, “4%,, W Warm Philip M. Marcus Barbara A. Stonewater Professor Specialist Department of Sociology Department of Electrical Engineering G Systems Science IZIS MICHIGAN STATE UNIVERSITY APPENDIX F - FOLLOW-UP COVER LETTER To WM ”Am (I WY EAST MNSIM ' MICHIGAN ° m REMINDER REMINDER April 1, 1977 Dear Administrator, About two weeks ago Barbara Stonewater and I sent you a brief questionnaire replicating a 1970 study of administrator-faculty relationships at MSU. If you have completed that questionnaire and returned it, we thank you very much and offer the enclosed for your files or other amusement. For those of you who have not completed the fifteen minute questionnaire, we hope you will do so as soon as possible so we can start analyzing the data and provide a public report before the end of this term. In order to make the comparison between 1970 and today meaningful, we need a very high return, and your completed questionnaire will help greatly. Faculty from only five colleges and a sample of administrators were selected for the study (to conserve costs) and each additional response takes on greatly weighted importance. Please remember the data are anonymous, and no person, department or college will ever be identified in the report. We have consulted members of Faculty Associates, AAUP, Academic Council and University Administration to assure that questions reflect many different perspectives. If you have not completed the questionnaire, Please do so immediately. We need your answers to help us understand current perceptions and attitudes as well as changes that have occurred over the past seven years. Only those of you who actively work in the university can provide the data we need to answer important policy questions. All interested parties will receive a full report of the findings. We are sure we can count on your cooperation. Please mail the completed ques- tionnaire back to us in the enclosed envelope. ’ Thank you again for your help in this important academic study. Sincerely, Philip M. Marcus Barbara A. Stonewater Professor Specialist Department of Sociology Department of Electrical Engineering a Systems Science 146 APPENDIX C Comparisons of Faculty Respondents to Population for Each College on Variables of Rank, Sex, and Age College A Rank Population Respondents _ a Z a Z Professor 110 45 53 46 Associate Professor 81 33 31 27 Assistant Professor 50 21 30 26 Instructor 2 l 0 0 TOTAL 243 100 114 99 Sex Population Respondents — a 1 a To. Male 203 84 95 83 Female 40 16 l9 17 TOTAL 243 100 114 100 Age Iliopula t L972. IPiespondeIEYLZ Under 30 7 3 3 3 30-39 61 25 37 32 40-49 87 36 37 32 50—54 30 12 15 13 55-59 24 10 8 7 60 and over 34 14 14 12 TOTAL 243 100 114 99 147 College B Rank Population Respondents a Z a 7 Professor 20 42 9 35 Associate Professor 12 25 7 27 Assistant Professor 14 29 8 31 Instructor 2 4 2 8 TOTAL 48 100 26 101 Sex Population Respondents —_ a 7 a 7: Male 43 9O 23 88 Female 5 10 3 12 TOTAL 48 100 26 100 Age gopulatigi Seapondenti Under 30 2 4 1 4 30-39 17 35 ll 42 40-49 14 29 7 27 50-54 4 8 2 8 55-59 8 l7 4 15 60 and over 3 6 1 4 TOTAL 48 99 26 100 a; 148 College C Rank Population Respondents — a :Z a 7: Professor 39 46 18 41 Associate Professor 25 3O 15 34 Assistant Professor 20 24 11 25 Instructor 0 0 0 0 TOTAL 84 100 44 100 Sex Population Respondents — a Z a. Z Male 84 100 44 100 Female 0 0 O 0 TOTAL 84 100 44 100 Age Population Respondents a Z a Z Under 30 2 2 2 5 30-39 28 33 18 41 40-49 28 33 13 30 50-54 7 8 2 4 55-59 9 11 7 16 60 and over 10 12 2 4 TOTAL 84 99 44 100 149 College D Rank Population Respondents a Z a 1 Professor 19 36 10 37 Associate Professor 10 19 4 15 Assistant Professor 20 38 11 41 Instructor 4 8 2 7 TOTAL 53 101 27 100 Sex Population Respondents — a Z a 1 Male 11 21 5 18 Female 42 79 22 82 TOTAL 53 100 27 100 Age EliopulatioIrT“; gespondelgi Under 30 2 4 2 7 30-39 15 28 7 26 40-49 16 30 8 30 50-54 7 13 3 11 55-59 6 11 3 ll 60 and over 7 l3 4 15 TOTAL 53 99 27 100 150 College E Rink Zopulatioi rl—Eeepeondenei Professor 101 53 53 54 Associate Professor 41 21 21 21 Assistant Professor 45 23 22 22 Instructor 5 3 2 2 TOTAL 192 100 98 99 Sex Population Respondents _ a Z a 7 Male 169 88 82 84 Female 23 12 15 15 No Response 1 1 TOTAL 192 100 98 100 Age IZOBUIati‘O‘tZ: :e spondenti Under 30 8 4 3 3 30-39 68 35 39 40 40-49 48 25 22 22 50-54 15 8 8 8 55-59 28 15 14 14 60 and over 25 13 10 10 No Response 2 2 TOTAL 192 100 98 99 151 eumomaum I m MUHmomh I h euouewueHmHam< I 4 H N He +3 NM 9» m H mm a among—08 muHaomw 30a 95.35 c. H o n H N n N ms co m mH H u H n w some me no meueeno wow anemone m mmHmHHAHoeHn n H H n mm mm been use moHuHmu huHeuo>Ham wcHGHauouoa N a on an ac emOHueooHHe uewvon huHeuo>Hm5 3533 PRESS H em a as on H as on Rammed umoHeeHave huHewe>Hm= euaseeuumevmm usHmHEHeuen H H oe on «N he s N on NH chooses huHoomw you enamou mchHaueuon H H H «a an on as n H m mH veuHAew on ammone nomads huHmomw mumwo auu m uH memdfiueuoo n n «H as n H on we eeHweHee huHmoem mchHaueuon H H H «0 he mm as m H o m easemoua HemoHueomvo 3mm waHueoHo n N Hm an m Ho on unopoma m meuoHoan< «H 0H am so n H as 0H meow UHEUveom me mmHumHong< «we Rue saw so“ an RN aha um moeuoawH o usuauweaev e meucHo m< M M m M a. M m M m M m M m M m M m M 83 ouIm ovum ovu< ovum ovum ovu< ovum ovum ovu< vaonunm vHSosnnm vHsonaum vaonenm vaonu-u vaoneum vaonnn< vHooneu< vaoseu< euoumuueHcHEu< can huHmoem you eoaosHmcH umox one e>e= moon 0&3 use vHsonm 053 we «GOHuaoouem m Nanmmmd LIST OF REFERENCES LIST OF REFERENCES Adams, D. R. Organizational control structures and bases of power in university departments and their faculty job satisfaction and professional output correlates. Unpublished doctoral dissertation, University of Kentucky, 1974. ‘ Bachman, J. G. Faculty satisfaction and the dean's influence: An organizational study of twelve liberal arts colleges. Journal of Applied Psychology, 1968, 52, 55-61. Bachman, J. G., Bowers, D. G., and Marcus, P. M; Bases of supervisory power: A comparative study in five organizational settings. In A. S. Tannenbaum, Control in orgenizations. New York: McGraw-Hill, 1968. Baldridge, J. V. Academic Overnance. Berkele , Calif.: McCutcheon Publishing orporation, 1971 (ag. Baldridge, J. V. Power and conflict in the university. New York: John Wiley and Sons, Inc.,’1971 (b7) Baldridge, J. V., Curtis D. V., Ecker G. P., and Riley, G. L. The impact of institutional size and complexity on faculty autonomy. Journal of Higher Education, 1976, 31, 391-411. Blau, P. M. The organization of academic work. New York: John Wiley and Sons, 1973. Blau, P. M., and Scott, W. R. Formal organizations. San Francisco: Chandler PublishingCo., 1962. Bucher, R. Social process and power in a medical school. In N. M. Zald (Ed.), Power in or anizations. Nashville: Vanderbilt University Press, I978. Cohen, M. D., and March, J. G. Leadership and ambiguity: The American college president. INew York: ‘Mc raw- Hill, 1974. 152 153 Colleges adding administrators, study reveals. The Chronicle of Higher Education, July 11, 1977, pp. 7. Dahl, R. A. The concept of power. Behavioral Science, 1957, A, 201-215. Davis, J. A. Elementary survey analysis. Englewood Cliffs, N. J.: Prentice-Hall, Inc., 1970. Demerath, N. J., Stephens, R, W., and Taylor, R. R., Power, presidents, and professors. New York: Basic BrooEs, Inc., 1967. Etzioni, A. Co arative analysis of camplex organiZations. New York: e Free Press, 1961. Etzioni, A. Mbdern organizations. Englewood Cliffs, N. J.: Prentice-Hall. Inc.. I964. Filley, A. G. and House, R. J. Managerial process and organizational behavior. Glenview, 111.: Scott, Foresman and Company,l969. French, J., and Raven, B. The bases of social power. In D. Cartwright and A. Zander (Eds.), Group dynamics. Evanston, Illinois: Row, Peterson, 19601 Garbarino, J. W. Faculty bargaining: Change and conflict. New York: McGraw-Hill, 1975. Glass, G. V., and Stanley, J. C. Statistical methods in education and s cholo . Englewood Cliffs, N1 J.: Prentice-Hall, Inc., 1670. Governance of higher education: Six priority problems. Carnegie Commission. New York: McGraw-Hill, 1973. Gross, E., and Grambsch, P. V. University goals and academic power. Washington, D. C.: American Council on Education, 1968. Gross, E., and Grambsch, P. V. Chan es in universit orgenization, 1964-1971. New YorE: McGraw-HIII, 1974. Hall, R. H. Opganizations: Structure and process. Englewood Cliffs, NT J.: Pfentice—Hall, Inc.,l972. Helsabeck, R. E. The relationship between decision-making structure and organizational effectiveness: A com ar- a tive case study in college governance. Uh ublis ed doctoral dissertation, Indiana University, 1 71. 154 Hickson, D. J., Hinings, C. R., Lee C. A., Schneck, R. E., and Pennings, J. M. A strategic contingencies' theory of intraorganizational power. Administrative Science Quarterly, 1971, A9, 216—229. Hinings, C. R., Hickson, D. J., Pennings, J. M. and Schneck, R. E. Structural conditions of intraorganizational power. Administrative Science Quarterly, 1971, lg, 22-44. Hill, W. W. and French, W. L. Perceptions of the power of department chairman by professors. Administrative Science Quarterly, 1967, 1;, 548-575. Hodgkinson, H. L. The campus senate: Experiment in democracy. Berkeley: GentefIfor Research and Devel- opment in Higher Education, University of California, 1974. Hodgkinson, H. L., and Meeth, L. R. (Eds.). Power and authority. San Francisco: Jossey-Bass, Inc., I975. Kaplan, A. Power in perspective. In R. L. Kahn and E. Boulding, Power and conflict in organizatiOns. New York: Basic BoOks, I964. Kemerer, F. R., and Baldridge, J. V. Unions on campus. San Francisco: Jossey-Bass, Inc., 1975. Ladd, E. C., Jr., and Lipset, S. M. Professors, unions, and American higher education. Carnegie Commission. New' or : Mc raw- , . Ladd, E. 0., Jr., and Lipset, 8. Mr The divided acade . Carnegie Commission. New York: McGraw—HIII, I975. McSweeney, M., and Olejnik, 8. Notes, exercises, and diagnostic measures for Education 9693. MIChigan State University, 1974. Mortimer, K. P., Gunne, M. G., and Leslie, D. W. Perceived legitimacy of decision making and academic governance patterns in higher education: A c arative analysis. Research in Higher Education, 1976?mg, 277-290. Peabody, R. L. Perceptions of organizational authority: A comparative analysis. Administrative Science Quarterly, 1962, 9, 463-482, Perrow, C. Departmental power and perspectives in industrial firms. In N. M. Zald (Ed.), Power in organizations. Nashville: Vanderbilt University Press, 1970. 155 Pfeffer, J., and Salancik, G. R. Organizational decision- making as a political process: The case of a university budget. Administrative Science Qparterly, 1974, A2, 135—151. Polsby, N. W. Community power and political theory. New Haven: Yale Universitfoess, 1963. Richman, B. M., and Farmer, R. N. Leadership, goals, and power in higher education. San Francisco: Jossey- Bass PubliShers, 1974. Salancik, G. R., and Pfeffer, J. The bases and uses of power in organizational decision making: The case of a university. Administrative Science Quarterly, 1974, _l_9_, 453-473. Schmidt, W. H., and Ripstra, C, Syllabus for Education 869. Michigan State University, 19731 Schuh, J. H. The deciSion-making process in liberal arts colleges of large universities. College and Universit , 1975, so, 177-1 2. Scott, W. R. Professionals in bureaucracies - areas of conflict. In H. M. Vollmer and D, L. Mills, (Eds.). Professionalization. Englewood Cliffs, N. J.: PrentICe Hall, Inc., 1966. Slakter, M. J. Statistical inference fer educatiOnal researchers. Reading, Massachusetts: Addison- WesIey Publishing Co., 1972. Smith, C. G., and Tannenbaum, A. S. Organizational control structure: A.comparative analysis. In A. S. Tannen- baum, Control in organizations. New York: McGraw- Hill, I968} Stefferud, J. A. Perceptions of power and authority at the university of Arkansas. Unpublished doctoral dissertation, University of Arkansas, 1975. Tannenbaum, A. S. Control in organizations. New York: McGraw-Hill, 1968. Warren, D. J} Power, visibility, and conformity in formal organizations. American Sociological Review, 1968, 6, 951-970. Weber, M. Thegtheory of social and economic organization. New York: The Free Press, 1947.