AUTHORITARIANISM AND ITS RELATIONSHIP TO GROUP STRUCTURE Thais Io: the Degree of Ed D. MI‘SHIGAN 3726‘“ COLLEGE Rah“ A3232“? Peckham" $95-63- {I‘HESIS This is to certify that the thesis entitled AUPFOI‘IPAF’IAIIISI‘I‘ MI”) ITS ‘73} APICNS S'fl? TO ‘xROUP SPRUJ l ' 3 presented by Half A. Peckham has been accepted towards fulfillment of the requirements for ECLD. degree in G'Ilidance (fi‘lcfltionl Major proéésor MW 0-169 i- "a- AUTHORITARIANISM AND ITS RELATIONSHIP TO GROUP STRUCTURE By Ralf Albert Peckham A DISSERTATION Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF EDUCATION 'Department or Guidance and Counselor Training 19Sh {151.3343 ACKNOWLEDGMENTS The author is grateful to the various state directors and supervisors in vocational rehabilitation programs throughout the country for their invaluable cooperation in distributing study materials to specially selected vocational rehabilitation counselors, and in following through on the data gathering phase of the project. Special gratitude is extended to the counselors themselves for their care and consideration in completing the study materials. Acknowledgment is also extended to Dr. Alvin Zander of the Research Center for Group Dynamics, University of Michigan, for certain helpful suggestions relating to the design of this study, and also to President J. Donald Phillips of Hillsdale College for his considerable technical information about the techniques and processes of group- centerod structures. Most particularly, thanks are eXtended to the chairman of the writer's committee, Dr. Walter F. Johnson, for his assistance in formulating the study, and in making important muggestions during the investigation. Thanks are also due to the other members of the committee for their many im- portant contributions: to Dr. Milton Rokeach for his sug- gestions relating to the nature of the personality variable on which this study was predicated; to Dr. Harry Sundwall for his suggestions relating to some of the statistical aspects of the study; and to Dr. Raymond Hatch and Dr. Cecil Millard for their other suggestions as well as their critical reading of the completed manuscript. TABLE OF CONTENTS CHAPTER Page I. THE PROBLEM, BACKGROUND, AND DEFINITION OF TERMS ................................ 1 Introduction ......................... l The Problem ......................... 3 Importance of the Study .............. h Definition of Terms Used ............. 5 Review of the Literature ............. 6 Limitations of the Current Study ..... 22 Organization of the Study ............ 2h II. METHODOLOGY AND PROCEDURE ............... 26 Introduction ......................... 26 Description of the "F" Scale ......... 27 The Experimental Population .......... 29 Selection of the Specific Sample ..... 32 The Selection of Contrasting Groups .. 3h Statistical Description of the Research Population .................. 35- The Research Questionnaire ........... 36 Description of the Questionnaire ..... 39 Group Structure Att itud inal Statements ........................... hO Group Structure Preference Statements ........................... hl CHAPTER Page Method of Questionnaire Construction.. h2 Plan for Analysis of the Data ........ RS III. ANALYSIS OF ATTITUDES IN GROUP STRUCTUREA he Introduction ......................... R6 Presentation of Data on the Distribution of Responses ............ h? Analysis of the Differences Between the Mean Responses ................... 50 Item Analysis of Differences Between the Means OOOOOOOOOOOOOOOOOO00.0.00... 52 Analysis of Performance on Individual Items ................................ 56 Summary .............................. 67 IV. ANALYSIS OF ATTITUDES IN GROUP STRUCTURE B ............................. 69 Introduction ......................... 69 Presentation of Data on the Distribution of Responses ............ 70 Analysis of the Differences Between the Mean Responses ................... 7O Item.Analysis of Differences Between the Means .OOOOOOOOOOOOOOOOOOOOOOOI... 714. Analysis of Performance on Individual Items 0.00.00.00.00.00000000COOOOOOOOO 76 summary 0.0000000000000000eeeoeeeesees 87 V. ANALYSIS OF ATTITUDES IN GROUP STRUCWREC COCOOOOOOOOOOOOOOOO0.0.000... 90 IntrOduction OOOOOOOOOOOOOOOOOO...O... 90 CHAPTER Presentation of Data on the Distribution of Responses ............ Analysis of the Differences Between the Mean Responses ................... Item Analysis of Differences Between the Means 0.0...OOOOOOOOOOOOOOOOOOCOOO Analysis of Performance on Individual Items OOOOOOOOOOOOOOOOOOOOO00.0.0.0... smmary 0.0.0.0000...OOOOOOOCOOOOOO... VI. ANALYSIS OF ATTITUDES IN GROUP STRUCTURED 00.0.0000...0.0.0.0000...O... IntI‘Oduction OOOOOOOOOOeoseOeeeeeeeeeo Presentation of Data on the Distribution of Responses ............ Analysis of the Differences Between the Mean Responses eeeeeeeeeoeeeeeeeee Item Analysis of Differences Between the Means 0.0.0....OOOOOOOOOOOCOOOOOOO Analysis of Performance on Individual Items eeeeeeeeeeeeeeeeeeeoeeeeeeeeeeee summary OOOOOOCOOOOOOOOOOOOOOOOOOOOOOO VII. ANALYSIS OF ATTITUDES IN GROUP STRUCTIIEEOOOOOOOOOOOOOO00.0.0.0...O... IntrOduCtion OOOOOOOOOOOOOOOOOOO00.... Presentation of Data on the Distribution of Responses ............ Analysis of the Differences Between the Mean Responses ................... Item Analysis of Differences Between the Meana OOOOOOOOOOOOOOOOOOOOOOO0..O. vi Page 92 92 96 98 111 113 113 115 115 119 121 133 135 135 136 139 1&1 vii CHAPTER Page Analysis of Performance on Individual Items OOOOOOOIOOOOOIOOOOOOOO0.0...O... Inn-3 Summary .............................. 155 VIII. ANALYSIS OF ATTITUDES IN GROUP STRIJCTUREF.0...‘............0.000...... 158 Introduction ......................... 158 Presentation of Data on the Distribution of Responses ............ 162 Analysis of the Differences Between the Mean Responses ................... 162 Item Analysis of Differences Between the Mean! 0.0...OOOOOOOOOOOOOOOOOOOOOO 166 Analysis of Performance on Individual Items .0000...OOOOOOOOCOOOOCOOOOOOOOOO 168 summary eeeeeeeeeeeeeeeeeeeeseeeeeeeee 179 IX. ANALYSIS OF THE DATA FOR RANK ORDER OF GROUP PmmmCES COOCOOCOOCOCCOCOOOOOOCC 182 Introduction ......................... 182 Distribution of Stated Preferences for the Various Group Structures ......... 183 Analys 18 O C O O O O O O C C O O O O O O O O C O O O O O O O O C C 183 summary C C . 0 C O C C O O C O C O O O C O O O C O O O O C C O O C 1914. X. SUMMARY, CONCLUSIONS, AND IMPLICATIONS FOR RESEARCH OOOOOOOOOOCOOOOOOOOOOOOOOOOO 197 Summary .............................. 197 Findings ............................. 200 Conclusions .......................... 207 Recommendations ...................... 210 Implications for Research ............ 212 BIBLIOGRAPHY O0.00.0...OCOOOOOOOOOOOOOOO00...... APPENDIXES A. Copy of Original Data-Collecting Letter.. B. ”F" Test of Authoritarianism ............ C. Group Structure Questionnaire . . . . . . . . . . . viii Page 215 221 22h 227 TABLE I. II. III. VII. VIII. XI. XII. XIII . "XIV. LIST OF TABLES Distribution of Responses on Group Structure A by Non-Authoritarians ....... Distribution of Responses on Group Structure A by Authoritarians ........... Mean Responses to Group Structure A by Authoritarian and Non-Authoritarian samples 0.0.0.0000.........OOOOOOOOOO.... Critical Points for One-Tailed and Two- Tailed T‘TeSt. eeeeeeeeeeeeeeeeeeeeeeeeeo Critical Ratios for Group Structure A ... Distribution of Responses on Group Structure B by Non-Authoritarians ....... Distribution of Responses on Group Structure B by Authoritarians ........... Mean Responses to Group Structure B by Authoritarian and Non-Authoritarian samples ......CCOOOOOOOOO0.00000000000000 Critical Ratios for Group Structure B ... Distribution of Responses on Group Structure C by Non-Authoritarians ....... Distribution of Responses on Group Structure C by Authoritarians ........... Mean Responses to Group Structure C by Authoritarian and Non-Authoritarian smploa 0.00.00.00.00......OCOOOOOOOOQOOO Critical Ratios for Group Structure C ... Distribution of Responses on Group Structure D by Non-Authoritarians ....... Page h8 h9 51 53 55 71 72 73 75 93 191+ 95 97 116 TABLE XVI. XVII. XVIII. XIX. XXI. XXII. XXIII. XXVI. XXVII. XXVIII. XXIX. Distribution of Responses on Group Structure D by Authoritarians ........... Mean Responses to Group Structure D by Authoritarian and Non-Authoritarian samples ......OOOOOO.......OOOOOOOOOOO0.0 Critical Ratios for Group Structure D ... Distribution of Responses on Group Structure E by Non-Authoritarians ....... Distribution of Responses on Group Structure E by Authoritarians ........... Mean Responses to Group Structure E by Authoritarian and Non-Authoritarian samples 0......00.0.0...00.000.000.000... Critical Ratios for Group Structure E ... Distribution of Responses on Group Structure F by Non-Authoritarians ....... Distribution of Responses on Group Structure F by Authoritarians ........... Mean Responses to Group Structure F by Authoritarian and Non-Authoritarian samples 0.000.000.0000...00.0.0.0...O.... Critical Ratios for Group Structure F ... Distribution of Preferences for Various Group Structures by Non-Authoritarian Group ......OCOOOOOOOOOO00..........0.... Distribution of Preferences for Various Group Structures by Authoritarian Group.. Comparison of Authoritarian and Non- Authoritarian Mean Preferences for various Group Sthtures eceeeoeoeeeaeeoe Comparison of Authoritarian and Non- kuthoritarian Mean Preferences for Paired Group Structures ................. Page 117 118 120 137 138 1&0 1&2 163 16h 165 167 18h 18u 186 188 CHAPTER I THE PROBLEM, BACKGROUND, AND DEFINITION OF TERMS Introduction This is a study that seeks to approach the Problem of group structuring or conference planning from the standpoint of the influence of a certain personality syndrome which may be particularly represented in the membership. The variable in question is that of "authoritarianism". To the degree that authoritarian personalities constitute in a greater or lesser degree the membership of the group, then by so much should that group be characterized by needs and preferences expressive of either the one kind of personality grouping or the other. If this indeed be so. this variability may be expressed in the following hypotheses: 1. M31 2 Eypothesis. Preferences for structured and unstructured groups are a function of the degree of authori- tarianism. 2. Mgggg'fiypotheseg. Highly structured groupings are more preferred by the authoritarian personality. a. Organized groupings with a minimum of structure are more preferred by the non-authoritarian personality. b. Unorganized or spontaneous groupings engender the greatest ambivalence of attitude among the authoritarian personalities. c. A greater unity of attitude in favor of spontaneous grmlpings is more characteristic of the non-authoritarian personality. The substantiation of these hypotheses should offer an important clue as to the manner or degree in which program planners might structure group processes for a particular membership. The population that was selected for this study con- sisted of the more experienced counselors who were employed by Vocational Rehabilitation Bureaus in each of the various states throughout the country at the time of the data- collecting activity, or more precisely, the Period July 1 to December 1. 1953. This is a causal comparative study of two groups selected from that population who were pre- sumed to be selectively similar in all aspects except that of the degree to which the one was imbued with authoritarian characteristics and the other to be relatively free of such. They were contrasted with each other in terms of their atti- tudes towards the various kinds of group structures with which they had acquired experience at local. state, regional, and national conferences of their professional association, the National Rehabilitation Association. This professional association is one which had enabled the experimenter to 3 become personally known to much of the membership by means of the various office-holding opportunities which he has had. A more detailed description of the nature of the popu- lation, their conference experiences, and the instruments used to evaluate them will be found in Chapter II. The Problem Statement g§_§h§_prob1em. From a historical point of view, conference planning has followed more or less an end- less repetition of a few relatively unchanged models. These have consisted essentially of an arrangement wherein a pas- sive audience is immobilized before a rostrum from which is presented a bill of fare preplanned by a program committee as being of a suitable tOpical nature. The manner of pre- sentation has usually consisted of individual talks or the readings of papers, or through the medium of a "panel dis- cussion" which in itself is sometimes observed to be so formal as to consist of a series of read papers. At the conclusion of either type of presentation, or at strategically interspersed intervals, the moderator usually asks the audience for questions. This sometimes does and sometimes does not produce a response. Intermissions are scheduled at periodic intervals and the assemblage then splinters into spontaneous groupings throughout the corridors and lounges. Memberships in these unstructured groups are often transient and fluctuating. h For a certain proportion of individuals, spontaneous par- ticipation and membership in dynamically changing groups seldom occurs, and for them the intermissions are but op- portunities for solitary pursuits. These represent the typical range of group structures at a large meeting, from the most informal or spontaneous groupings to the most structured and preplanned groupings. That some of these kinds of experiences are not wholly satisfying to many segments of a convention membership is an old.and familiar problem. The purpose of this study is (l) to compare the view- points of a selected authoritarian group with that of a non- authoritarian group toward these various levels of group structuralization, and (2) to show characteristic differences in attitude and preference for these differing levels of structuralization in order (3) to present a basis for more intelligent planning of group structuring in relation to the authoritarian factor present in any group to be assembled. Importance 2; Egg gtggy. Some eXperimentation has already been attempted with regard to a wholesale revision in the entire approach to conference planning. Phillips and Taylor (73zh56-h62) have moved in the direction of an entire democratization of all group processes. It is not the purpose of this study to assume that all individuals or groups are alike in their preferences for one kind of struc- ture or'another but rather to assume that they differ along S the continuum, some preferring highly formalized settings, and some greatly rejecting them. If this study succeeds in establishing the basis for this differential attitude, it will furnish to the program planner or convention chair- man an additional element for consideration not elsewhere mentioned in the literature. Definition of Terms Used Authoritarian Personali_y. Summarizing the description of the authoritarian personality as it appears in the re- search of Adorno, Frenkel-Brunswik, Levinson, and Sanford (13227-229), it may be defined as one in which there is a rigid adherence to conventional, middle-class values, an uncritical attitude toward idealized moral authorities of the ingroup, a tendency to condemn, reject, and punish those who violate conventional values, and to oppose empathizing with tender-mindedness. It includes a disposition to think in rigid categories, to be preoccupied with the dominance- submission dimension, and to be vulnerable to make outward projections of unconscious emotional impulses. In the sexual area there is an exaggerated condemnatory attitude towards delinquencies and transgressions of the moral code. Towards people in general, there is a variable degree of generalized hostility. This study shall particularly employ an operational definition of authoritarianism using performance on the 6 scale designed by Adorno, Frenkel-Brunswik, Levinson, and Sanford as the criterion by which to define population samples possessing high and low degrees of this general constellation of traits known as authoritarianism. This has been titled by them as the "F" scale (1:255-257). Ml _c_>_f_ £222 Structure. Structure refers funda- mentally to level of organization. The greater the degree of structuralization, the greater the evidence of organiza- tion. Analagous to organization is formalism, procedure, and regulation. The more unstructured the group, the more plastic and fluid is its organization. In the perspective of this study, a highly structured group may be seen as one in which the rule of each member has been well defined and there is no confusion about the lines of authority. In the unstructured group, authority lines are de-emphasized and each individual plays more or less of an equalitarian role. Review of the Literature Literature ggflthg nature 2; the authoritarian personal- ‘igy. The present concept of the authoritarian personality and the research on it has been of relatively recent origin. ' The social-psychological studies of Paul Messing in his work ”Rehearsal for Destruction" (Sh) were among the very first to call attention to the role of the bigot and the extremely authoritarian-centered personalities comprising . the leadership of Nazi Germany. This research marked a 1, shift of focus away from the controversial aspects of hereditarian and environmental factors with regard to their influences on certain ethnic, and usually minority, groupings and centered it squarely upon the remarkable kinds of per- sonalities who conceived and carried out such horrifying social phenomena as mass extermination and the most exces- sive of pegroms against innumerable relatively defenseless peoples. Bettelheim and Janowitz (9) in their psychological and sociological study of veterans isolated some of the basic structures that delineated democratic from non-democratic personalities and.in this they furnished considerable sup- port for the remarkably comprehensive studies of Adorno, Frenkel-Brunswik, Levinson, and Sanford (l) in their full treatment of the authoritarian personality from the purely analytical viewpoint. These authors succeeded in not only isolating traits that comprise the authoritarian personality but also succeeded in producing an instrument for measuring these traits among various strata of the population. Their "instrument, known as the "F" scale (1:255-257), is in fact ’,\that which is to be employed in the current problem to separate the two population samples to be studied. Zilboorg (78) points out a paradoxical psychological state on the part of the more authoritarian leaders in that the greater the authority of a given leader, the more sheltered he is from the impacts of reality. This is partly due to a reduction of his free channels of communication. Rokeach (61) has contributed research on the social dynamic of ethnocentrism and has indicated a factorial re- lationship between that phenomenon and the general char- acteristic of mental rigidity in the thinking processes of affected individuals. Literature 23.the authoritarian personality lg g_grggp setting. Titus (7h) has contributed some research with re- Q gard to the kinds of individuals who become leaders of groups. ‘ He finds such authoritarian motives as the need to dominate, the quest for survival, and the urge for ego-appeasement. ’ C,Sanford (63) has shown in his study of follower phenomena that a high degree of acceptance on the part of the follow- ers varies somewhat proportionately with the strength of 4 / [the command position of the leader, and that a very strong leader can obtain a very strong acceptance on the part of a group. Whyte (77:297'312) has studied group leaders with regard to their methodology in initiating group action. Democratically oriented leaders rely on inter-personally aroused cues coming from within the group whereas auto- cratically oriented leaders direct the group towards ob- tainable goals. Edward Jones (A6) has done a study that more nearly parallels the design of the contemplated experiment, but 9 differing in focus. Rather than relating the preferences of his experimental groups toward some aspect of rigid or non-rigid group structure, he concerned himself with a study of their attitudes toward the leaders. He contrasted a group low in "authoritarianism" (N - 100) with a group high in ”authoritarianism" (N a 95), and found that the ”low” group preferred the more democratic type of leader whereas the ”high" group preferred the more authoritarian type of leader. This would seem to suggest corroborative implica- tions for the basic hypothesis of the instant study under consideration. The setting, however, was that of a training center for naval recruits rather than a group conference situation. Authoritarianism and leadership choice i 3 military setting. Though.military organization is based fundamentally on the function of command, subordination, and discipline, the literature interestingly enough indicates that leader- ship choices among peers are not related to the extent with which one is imbued with the authority dimension. Hollander (AZ) contributes a study in this area in which 268 naval cadets completed a leadership nomination form for their own group in which they indicated those individuals that might be most fitted for "student commander" and those individuals that might be least fitted for this role. At about the same time they were administered the "F" scale. Leadership and authoritarianism scores were then derived for all cadets. 10 It was found that the individuals who were nominated for leadership positions were "low" in authoritarianism, whereas individuals rejected for leadership roles were found to be "high" in authoritarianism. This study seems somewhat paral- lel to that conducted by Jones in the preceding citation. Sewell in his "Structures of Leadership" (66) reviews the basic fundamentals upon which leadership in the United States Army rests at the present time. The traits he de- scribes as those desired for a leader selectee seem to be similar to those preferred in Hollander's discussion sum- marized in the foregoing. Jenkins (MS) implicitly confirms this in his review of leadership studies with particular reference to military problems. Literature‘ 2.2E2 nature gf'thglggggp. Bernstein (8) concludes that every group structure is a complex of factors and forces which give it a special character and that ef- fective working together is dependent on an understanding of the whole situation. This requires an analysis of the role of the group leader, program development within groups, and psychological factors, Which affect group performance. Bion (10) analyzes the function of groups from the stand- point of psychoanalysis. He previews the conventional dis- tinctions between individual and group and finds that the essence of groupness lies in the effects which other people have upon beliefs, ideas, emotions, and motives of the in- dividual. McCurdy and Lambert (SS) examine the efficiency ll of the small group process and assert that such groups do not work more successfully with problems than do individuals. Martin, Gross, and Darley (53) examined the behavior of groups and concluded that they tend to divide themselves into leaders, followers, and isolates. Festinger (26) looked at some of the social pressures developed in the smaller groups and noted similar features of inter-personal communication, deviation, and rejection. Hyman (uh) addressed himself to the psychological aspects of the individual status of each group member. He found that as the individual varied his membership from group to group, his own status evaluation changed with each change in the reference group. Sterling- and Rosenthal (70) found that changing phases of group ac- tivity exerted changes in the relationship between leader- ship and followership authority, thereby implying that no one leadership "personality" or "process" is suitable for all groups at all times. Livingston (h9) stated this same general conclusion in a somewhat inverse fashion. Varying kinds of group organi- zations are perceived by the individual, including the leader, as varying kinds of opportunity for power and benefit. Some have a high valence for the power value; some, by virtue of their organization, a low valence. Marquis, Guetzkow, and Heyns (52) examined the nature of the decision-making conference. They found that member satisfaction with this type of meeting increases with the 12 cohesiveness of the group and the amount of procedural structuring; but decreases with the incidence of self- oriented need behavior on the part of the participants. Group productivity increases with the urgency of the prob- lem, and with the power of the group to deal with the prob- lem; it decreases with the incidence of self-oriented need behavior. Residual disagreement is high when there is dis- agreement as to goals, a good deal of self-oriented need behavior, and a feeling of inadequate power to deal with the problem on the part of the participants. Phillips and Taylor (73) describe a democratic method of organizing group conferences in Which authoritarian leadership is completely de-emphasized. Giving it the popular title of "Phillips 66" plan, they fractionate any large assemblage into groups of six individuals each. Each such group is instructed to choose from among themselves a recorder and a chairman. The chairman is charged with the responsibility of having each.member in the group contribute his best thinking. The recorder makes a record of each state- ment and the group then votes on which one it desires to pre- sent before the entire assemblage. Convention hosts collect the records of every delegate's individual thinking for sub- sequent evaluation. Thus each individual makes some sort of an immediate and personal contribution, and all of them collectively rather than a handful of program planners deter- mine what shall be heard and discussed at the conference. 13 Moreno (57) concerns himself with a study of group dynamics and sets forth three principles of sociometric action. The first is to obtain full group and audience par- ticipation leading to collective self-expression. The sec- ond is to have the participants approach the subject in the midst of actual life situations, and the third is to obtain a group concensus or integration which is the end result of the meeting. Festinger, Pepitone, and Newcomb (25) examined aspects of group dynmnics that occur when members lose their in- dividuation and do not regard each other as individuals as such. (A role-playing structure would be a common illustra- tion.) They advance the theory that this results in re- duction of inner restraint or fear of personal threat, and thus each member is free to indulge in behavior from which he is usually restrained. This facet of group structure in- creases its attractiveness to individuals possessing such needs. Festinger and Thibaut (2h) also made a study of the inter-personal communication that operates within groups: they disclosed a dynamic of pressure to achieve group uni- formity directed toward the center of the opinion range. The bulk of group communications were primarily directed toward those individuals who manifested extreme opinbns at either end of the opinion range. 1h Attitudes toward structure. Hartley (39) indicates that as the individual acquires memberships in various groups, he is confronted with such psychological problems as identification, group role, group status, ethnocentrism, and the concepts of self and ego, and that each of these problems vary with each group. Bradford (13) defines typical behaviorisms found in \ group processes such as the "audience" phenomenon in which the membership accepts a completely passive role, the ”anonymity" phenomenon in which the listening group in- creases the feelings of individualism, and frustrates the basic human need for affiliation or belongingness. Each individual is more or less completely uninvolved, and has no responsibility. Other behaviorisms include "identifica- tion” with the speaker and complete allowance for the speaker to speak for them; the "rejection" phenomenon in which the listening group resists the speaker's ideas but does not communicate to the speaker these areas of resist- ance so that he will know of their existence; and the "non- " phenomenon in which communication is typically participation one way from the platform to the audience. Leslie (h8) re- ports an experiment on attitudinal change occurring as the result of group interaction. Four steps in the process were noted. First was manifested a testing-out period; then, in the second phase, resistance to the leader. Then occurred acceptance with understanding; and finally, personal respon- 15 sibility in which the leader becomes only a resource person. Cantor (15) reports on common attitudes toward the role of the leader as he attempts to guide the group. When the con- ference leader develops an accepting atmosphere, the usual defenses against change can be broken down and anxieties relieved so that learning can take place. The leader may then clarify the unconscious feelings of the members of the group. Scheidlinger (6h) discusses attitudes toward various phases of group structure in psychoanalytic terms. The role of the leader is described as symbolic of the parental image, and he will be perceived and accepted in relation to the re- jective or acceptive feelings or needs that are correspond- ingly aroused. He also discusses the nature and origin of gregariousness, with the implication that this need is necessarily unmet in certain types of group structures. Likewise, various emotional processes (identification, object-ties, relative loss of personal identity, regression, growth-promotion) are operative in various kinds of group structures some more and some less. Gilchrist (31) studied the attitudes that govern the selection of a group by a prospective member. These were noted to be largely confined to the area of level of aspira- tion and expectation. Bovard (12) carried out an experiment to prove that inter-personal affect was heightened in a group-centered process wherein verbal interaction was l6 maximized, and that inter-personal affect would be decreased under a leader-centered process when verbal interaction was largely restricted. This led to the inference that certain attitudes are a function of the discussion process that is employed by the group. Literature gg_£hg_nature g; the leader. After examin- ing some of the literature on the nature of the group, one might well review some of the materials that pertain to the individual who leads the group. Gordon (33) made a study on some of the personal factors that comprise the leadership personality. In examining such personality traits as sociability, ascendency, responsibility, hyper- sensitivity, and refinement, he found that a sense of re- sponsibility showed a substantial relationship to leadership, whereas ascendency and absence of hypersensitivity showed only a slight relationship. The relationship of intelli- gence to leadership was indicated to be of about the same level of correlation as these last two traits. Richardson and Hanawalt (60) attempted to discover some relationships between the leadership personality and the Bernreuter personality measure. Highest correlations were found in relation to the traits of dominance and self- confidence, with only a mild correlation in the areas of self-sufficiency and sociability. Wittenberg (76) believed that the leader must be able to recognize his own needs and l7 limitations and must have the ability to accept some hos- tility from the group; in short, the leader must possess self-discipline but not necessarily conformity. Holtzman (M3) made studies on the relationships between leadership and personal adjustment as measured by the Ror- schach Test. He finds a significant correlation between leadership and sound adjustment "scores" as interpreted from the materials. Goode (32) made a comprehensive study of a number of reports concerned with research on leader- ship. He found the following qualities to be characteristic of the successful leader: mental ability; breadth of inter- ests and aptitudes; language facility; maturity; strong motivation; social orientation; and administrative skill. Carter (16) made a study of a number of factors that might be related to the leadership personality and reduced these to two significant clusters: insight (initiative, interest, activity); and cooperation (friendliness, efficiency, loquac- ity, insight). Chowdhry and Newcomb (18) found that leaders are significantly superior to followers in their ability to judge group Opinion on familiar and relevant issues. They concluded that the ability to be a good judge of other's attitudes is a necessary condition of being chosen for leadership. Stone (72) classified leaders into three types. He contrasts the ”autocratic paternalistic leader", the "democratic majority rule leader", and the "democratic cooperative leader" in terms of the leaders functions, his 18 attitudes towards followers, motivation within the group led, and educational philosophy implicit in the leader's relation- ship to his group. Gouldner (3h) in his studies in leader- ship continues the theme of types of leaders as described by Stone but presents a number of detailed analyses refer- ring to ”authoritarian” types of leaders as contrasted with the so-called "democratic" types of leaders. Gibb (30) at- tacks the analysis of leadership not from the viewpoint of the individual who does the leading but rather from the view- point of the group who permits his leadership. Gibb believes that leadership is an interactional process in which the syntality characteristics of the group itself are of con- spicuous importance in determining the nature of the leader selectee. Bell and French (6), however, do not adhere too closely to Gibb's line of thought, since in their studies they note that the leadership position is consistently held by the same leaders even though the groups are varied in memberships. The gglg 2; the leader. Cantor (1h) believes that the role of the professional leader is to focus the discussion. He encourages likeness and difference in others and opposes the group with his own difference. He does not strive for concensus or agreement, but helps to develop a spirit of compromise and willingness to work together. Carter and Haythorn (17) are essentially in agreement with Cantor but they divide the leader's role into two general areas of 19 responsibility: to analyze the situation; and to initiate the action required. Dulsky (23), however, finds five basic principles that the leader should observe. These are: to encourage sounding off; to promote discussion and active cooperation; encourage self-confidence, especially on the part of reticent members; to promote competition; and.to re- flect frequently both content and feeling in order to clarify and to draw out further discussion. Meier (51) suggests a single broad base of using "democratic" processes, though in elaboration upon the process it is readily seen that he is in agreement with most of the principles or factors else- where spelled out in the foregoing discussion. Factors influencing participation. Meier and Solem (50) believe that the quality of group thinking is stimulated when the discussion leader encourages particular expression of minority opinions. He found that this led to increased dis- cussion on the part of all, and with.more effectiveness in arrivhig at correct answers to the presented problems. Hem- phill (hl) found some relationship between the size of the group and the participation process. He concluded that as the group becomes larger, demands upon the leader's role become greater, and that at the same time the group increases its tolerance for leader-centered direction of its own activi- ties. Gullahorn (36) found two different types of factors that he believed influenced participation. These were the factors of distance and friendship. Individuals nearest each 20 other, geographically, tended to interact with each other, and when distance alone did not serve as an index of inter- action, it was found that friendship was likely to be the controlling factor. Leaderless gropps. After reviewing some of the litera- ture on both the nature of the leader as well as on the group that he leads, it might be well to examine some of the writ- ings concerned with the so-called "leaderless" groups. Es- sentially it will be noted that the characteristics of these groups are similar to those discussion groups which are directed by a democratically oriented leader. The main dif- ference is that leaderless groups begin with pg leaders and then select one or several from among their own group mem- bership to serve in the leadership roles. In the apparent progression from the traditional authoritarian group struc- ture to that of the recent emphasis on leader-led, group participation structures of the more democratic variety one might be tempted to assume that the "leaderless group dis- cussion" becomes the most recent step in the ongoing process of evolution. Strangely enough, however, the leaderless group discussion technique had its origins several decades back, and in a setting not ordinarily viewed as democratic in nature. Ansbacher (2) describes its use in the German Army back in 1925 when it was used as a leadership selection device by the German military psychology corps under the direction of J. B. Rieffert. It was exclusively a German 21 technique until the British Army became more directly ac- quainted with it in lQhO, and shortly thereafter it came to the attention of American military psychologists, who also employed it in personnel assessment tests. This technique made its first appearance in American literature in 19hl. French (29) discusses verbal output and leadership status in initially leaderless discussion groups. He found that group productivity increases with the urgency of the prob- lem and with the power of the group to deal with the prob- lem. Disagreement is high when there is a disagreement as to goals. Bass and Norton (h) drew some conclusions about the relationship between group size and leaderless discussion. In studying participants who were variously assembled in groups of two, four, six, eight, and twelve, he found a de- cline in participation as groups were increased in size. Maximum stratification occurred in discussion groups of six. Bass and Klubeck (5) also studied the effects of seat- ing arrangements on leaderless group discussions. They in- ferred that when effects of a person's outside status were eliminated, the particular seat a person occupied was of negligible importance in determining a participants's tend- encies to attain leadership status during the course of the discussion. Limitations pf previous studies. The literature that has been reviewed is of relevance to the current study in that it contributes an abundance of theory and research data i\> ‘\ N. ‘~ \ “5 22 relating to various aspects of the personality syndrome of authoritarianism as well as an abundance of information on group dynamics in general and certain group processes in particular. These resources have proved most important in designing the study, particularly in constructing the group. structure stereotypes and in building the test items. What these studies have omitted to do, however, is to relate the extremes of personality structures to the ex- tremes of group structures on a hypothesis of some kind of affinity between the two. Individual leaders have been studied, and collective processes have been experimentally examined, but there still remains a gap in the literature. This study is directed toward a building up of information in this area. Limitations of the Current Study The limitations to be found herein fall into two main categories: (a) limitations related to the sample; and (b) limitations related to the instruments used. The limitations inherent in the sample will be more ap- parent in the more detailed description of the selection of the sample in Chapter II. Essentially, however, this is a selective population, differing not only from general popu~ lation nonns in terms of education and maturity, but also differing from.many professional populations in terms of its working methodologies and emphases, and.more pertinently, in 23 its professional convention experiences and conferencing procedures. Since rehabilitation counseling is primarily an applied activity rather than a research-oriented or theoretical discipline, its conferencing procedures tend to be more of the demonstration type rather than of the re- search, paper-reading variety. Nevertheless, research in rehabilitation as conducted by other specialty groups is of necessary interest to workers in the field, and the paper-reading technique is by no means an unfamiliar con- ference experience for this population. However, it is a new profession and more or less unencumbered by tradition. Its program planning ordinarily includes some sampling of the newer group-conferencing procedures of the more un- structured type, as well as a substantial range of the more traditional conference methodologies. In addition to the limitations and Special sophistica- tions of the research population, it must be acknowledged that the instruments used to assess the population would possess certain limitations. Whenever validity and relia- bility are less than 100 per cent, some spuriousness must inhere in the assessing procedures. The "F" scale used to array the population in terms of "authoritarianism" appeared to the experimenter, after a review of the literature in the field, to be the best instrument that he could find, and this instrument, as would be true of any, has an efficiency that is necessarily less than perfect. A fuller description 2h of this scale, as well as the other questionnaire materials, will be found in the more detailed description appearing in Chapter II. With reference to these latter questionnaire materials, moreover, it must be likewise recognized that not only do limitations of the same character exist for each item in the additional questionnaires, but also there is an over-all, and somewhat measurable, effect of length of the total materials. It takes some time for an individual to work the total questionnaire through to completion. Since approximately h2 per cent of the six hundred original dis- tributions were never returned, it may be surmised that one of the factors in the failure to respond would be that of the very onerousness of filling out a long questionnaire. Thus, to a degree, the nature of the instrument itself pre- sents some impediment to an easy assessment of a sample population, and by that fact introduces an unknown factor to the sampling process which may or may not have been averaged out of the selected eXperimental samples. Thus, any interpretations arising out of this study must be tempered by an awareness of the limitations that are a part of its constitution. Organization of the Study This thesis is divided into ten chapters: Chapter I includes a statement of the problem, the importance and scope of the problem, a definition of terms, a review of the 25 literature, and some of the limitations of the study. Chapter II contains a description of the instruments used, a detailed account of the selection of the sample, and the general methodology and plan of procedure. Chapters III through VIII consist of comparative analyses of the mean attitudes of the two eXperimental groups toward each of the six kinds of group structures. Each chapter in this series consists of the analysis of a single group structure in terms of the contrasting attitudinal responses that it elicits from the two experimental groups. Chapter IX con- sists of an analysis of the data for the rank order of stated preferences for these group structures on the part of the two experimental groups; and Chapter X includes a summary of the study, the conclusions, and some of the im- plications for further research. 26 CHAPTER II METHODOLOGY AND PROCEDURE Introduction This is a causal comparative study of two groups of individuals who differed from each other in terms of the degree to which each possessed the personality variable known as "authoritarianism". As a result of a test selec- tion procedure, one group was characterized as being rather "high” in authoritarianism, and the other group was defined as being rather ”low”. This variable was believed to exert a predictable in- fluence over the types of attitudes which these individuals would display toward a variety of situations which are typi- cally encountered in a conference or convention setting. These special situations were comprised of six different types of group structures ranging from the completely spon- taneous or non-structured groupings that occur randomly in the convention corridors and lounges, through the semi- structured groupings of the workshop or group-discussion type, and into the extremely structured meetings of the 'paper-reading” variety. 27 Description of the "F" Scale In this study it was desired to select an otherwise homogeneous population, and to array this group from ”highest" to”lowest" in terms of the individual degree to which each possessed the personality variable of "authori- tarianism" as described in Chapter I of this paper. The \\ instrument selected was that developed by Else Frenkel- \ Brunswik and her associates (1:255-257), and designated as Form hO/hS. As described in the original work by these authors, "The Authoritarian Personality" (l), the authori- ‘ tarian syndrome was first identified by conducting clinical studies and case histories on those individuals whose public behavior was particularly believed to be authoritarian in essence. The more conspicuous of these were typically the persecutors of minority peoples; the less conspicuous were those who claimed partisanship in the same attitudes of deep prejudice or feelings of anti-democracy. Attitudinal statements reflecting these anti-democratic feelings were constructed so that individuals could place themselves at various levels of agreement or disagreement on a variety of points. The scale went through a number of revisions in an effort to increase the discrimination power of each item - until it reached the Form hO/hS which was employed in this study e 28 The authors cite data (1:258) which indicates that this scale was experimentally administered to a total of 1,518 individuals. The average discrimination power of each item.was found to be 2.85 which is at least at the 1 per cent level of confidence. The average of the relia- bility coefficients is .90 with a range of from .81 to .97. Each item is scored according to six levels of responses, these ranging from "strongly agree" to "strongly disagree", or from "plus three" at one end to "minus three" at the other end. In scoring the scale, the plus or minus values are converted into a weighted score as follows: Code Raw Score Conversion Key Strong agreement +3 7 Moderate agreement +2 6 Slight agreement +1 5 Omitted _O h Slight disagreement -l 3 Moderate disagreement -2 2 Strong disagreement -3 1 In framing the items, all statements were so worded that agreement was in the direction of agreeing with an authoritarian attitude. Thus the most authoritarian score is '7", the least authoritarian score is "1". When the scale is administered, the total score is divided by the number of items in order to establish an average score per item. In Frenkel-Brunswik's population, the average score for the 1,518 individuals proved to be 3.8h, with a standard deviation of 1.10. The complete ”F" Scale appears in the appendix. 29 The Experimental Population The population that was selected for this study con- sisted of the more experienced counselors who are employed by Vocational Rehabilitation Bureaus in each of the various states throughout the country. These vocational rehabilita- tion programs are operated under the administrative authority of the Department of Public Instruction in each state, but in operational conformity to a common Federal Act which spells out the program and provides much of the money. Basi- cally, this is a program in which the counselor meets with vocationally handicapped individuals who are of an employ- able age, offers them vocational counseling and diagnosis, provides for psychological testing, arranges for medical examination, plans a hospitalization or physical restora- tion program, plans a vocational training program according to the individual's best aptitudes, interest, and potential opportunities for employment, and then follows through to ultimate job placement. To the extent that the individual cannot participate financially in these costs, the vocational rehabilitation counselor arranges for the program to under- write these costs. Inasmuch as this program is a Federal-State operation, the federal participation requires that each state must recruit staff membership via a competitive merit system. This factor, plus the technical demands of the job itself, 30 tends to bring into each state program a more selective population than perhaps would be true if the applicant did not have to compete for the appointment on a merit system examination. Williams (75) recently made a sample study of the type of vocational rehabilitation field agent who is typically sent by his state program as a delegate to conferences of one kind or another. Inasmuch as this paper is concerned primarily with this type of counselor, his findings are likely to be fairly descriptive of the counselors who con- stitute the experimental population in question. This is especially true in that his sampling is drawn from the same general population at a relatively contiguous point in time. He finds in his study that vocational rehabilita- tion counselors are primarily a masculine population (about 9h per cent male); by age range, about 5 per cent are under twenty-five; 2k per cent are between twenty-five and thirty- four; 3h per cent are between the ages of thirty-five and forty-four; and 37 per cent are over forty-four. Of the entire group, about 80 per cent are married. With regard to length of service, the great bulk of the group entered the program within the last six to ten years. He also noted that practically all vocational rehabili- tation counselors had prior professional job experience before coming into the program. About 50 per cent of them were found to come out of the broad field of education, with 31 particular emphasis being noted in the field of special education, including speech correction, et cetera, some from school counseling assignments, and some from school administration. About hO per cent were found to possess a social agency background either in public welfare, private social agencies, or employment service work. About 10 per cent were found to have entered the prOgram directly from college, or from a job experience other than that described above. ‘lith regard to educational status, hl per cent were found to possess a master's degree, with the remainder hav- ing a bachelor's degree and typically some work beyond. Major fields of preparation were primarily in education, psychology, sociology, and social work. The rest were scattered in other fields of the arts and sciences. Thus, in Williams's summary, the more experienced vocational re- habilitation counselor is typically a person of maturity, who has exhibited job stability, and.Who possesses some professional training related to the job that he is perform- ing. Conference Experience. The National Rehabilitation Association is the official professional association to which counselors and other workers in the general field of rehabilitation typically belong. It has a paid professional membership drawn from rehabilitation and its allied fields of nearly six thousand individuals from whom it can expect 32 some degree of participation at its annual national, regional, state, and local conferences and conventions. These conferences are primarily professional in character, and are concerned with information-sharing activities on new research or practical applications of experiences in rehabilitation techniques, or with the advancement of the profession. Of the approximately one thousand individuals compris- ing the professional staff of the State-Federal programs for the rehabilitation of the vocationally handicapped, most of these are provided considerable opportunity to at- tend a rehabilitation conference or convention at least once during a working year. In addition, many of these profes- sional peOple have opportunity to attend other professional conferences in which they have memberships, such as those sponsored by the American Personnel and Guidance Association, the American Psychological Association, the American Associa- tion of Social WOrkers, and many other professional groups. Selection of the Specific Sample Since the design of this study was to select those counselors who had had the most experience in attending pro- fessional conferences or conventions, each program director in each state was asked to limit the distribution of the questionnaire materials to those counselors who possessed the most conference experience, or were most faithful, in 33 attending such conventions. As a criterion, they were asked to divide their counseling staff into two halves, the most experienced as opposed to the least experienced, and restrict the distribution of study materials to the more ex- perienced half. In order to call more specific attention to this process, each state program was given distribution materials sufficient for only the particular half of their staff as described above. This called for considerable cooperative assistance on the part of the various state directors and their supervi- sors in distributing these materials. However, by virtue of the mutuality of inter-state interests in furthering research studies--and with an established custom of con- stantly carrying on such national or regional studies on a state by state basis--this proved to be one of the most agreeable experiences in the whole survey. A total of six hundred questionnaires was placed in the hands of these selected counselors. Of that number, some 3&9 completed the materials and returned them to the experi- menter. Nine of the sets of materials were found to be in- completely or incorrectly filled out, and were rejected from the study. However, on the basis of known returns, it is apparent that a response of 58 per cent was secured, which, in view of the time and demands to complete each set of materials on the part of the individual respondent, may be 3h regarded as a very favorable individual cooperative effort. Rejecting the nine inadequate sets of material, the research population was narrowed down to a study of 3hO cases. The Selection of Contrasting Groups The total questionnaire submitted to each individual in the research population consists of a ten page document. The first section of this instrument consists of the "F" Scale Test elsewhere described in this paper. The purpose of this first section was to scale this population along a continuum of degrees to which they possessed or did not possess those traits representing the authoritarian con- stellation. The first and third quartile points were used as cutoffs to distinguish the least authoritarian from the most authoritarian ends of the range. Thus the lowest quarter was contrasted with the highest quarter in terms of their preferences for each level of group organization. In a sample of three hundred and forty individuals, this implies that the "lowest" 85 individuals are being contrasted to the "highest" 85 individuals throughout the rest of the experi- ment. These two groups represent the two actual experimental samples, each one serving as a control by which the other was measured in the usual causal-comparative manner. 35 Statistical Description of the Research Population An analysis of the ”F" Scale performance of the three hundred and forty individuals comprising the research popu- lations indicated a total mean score performance of 3.32 per item, with a standard deviation of .88. In comparison with the mean score obtained by Frenkel-Brunswik (1:258) and her associates on a population of 1,518 individuals, wherein the mean was 3.8h with a standard deviation of 1.10, it is apparent that the total selected population was some- what less authoritarian in nature and somewhat more homo- geneous with regard to this trait. Since the population tested was analagous in maturity and education to a number of the experimental samples re- ported on by Frenkel-Brunswik (1:258-260), and in which the reliability coefficients were found to range from a low of .81 to a high of .97 with a mean of .90, it could be reasona- bly inferred that the reliability of the conducted testing would approximate some point within that range. with regard to the two experimental groups which were chosen from each end of the research population in accordance with the definition set forth in Chapter I, these may be described as follows. The "non-authoritarian" group had a mean response on the "F" scale of 2.17, with a standard deviation of .h3. In contrast to this seemingly low 36 average response, the ”authoritarian" group had a mean ”F” scale response of h.39, with a standard deviation of .5. The Research Questionnaire It was the purpose of this study to assess the atti- tudes of preferences of individuals "high" in authoritari- anism as opposed to individuals ”low" in authoritarianism toward a range of group-structure stereotypes that are typically encountered at professional conferences. These group-structure stereotypes were to be arranged along a continuum, extending from the most rigid or formal type of group structure down through the most plastic or informal type of group structure. Six stereotypes were arbitrarily conceived to represent a suitable array of check points along the continuum. It was not assumed that these stereo- types would be constructed at equidistant points from each other, but it was intended that each successive step inter- val would progress in an orderly series, that is, each group structure would represent less rigidity than the one preced- ing it. These group structures are identified respectively as "A”, "B", "C", "D", "E", and "F". In addition to its letter designation, there follows a paragraph or two which portrays the particular group stereotype under consideration. 37 These six stereotypes might be more familiarly described with the following captions, though these captions do not appear on the questionnaire: Group Structure A Speaker-Centered Structure, Uninfluenced by Group Group Structure B Speaker-Centered Structure, Influenced by Group Group Structure C Synthetic Group Discussion Group Structure D Leader-Directed Group Discussion Group Structure E Leaderless Group Discussion Group Structure F Spontaneous Groupings In Group Structure A, it was intended to describe a group situation in which total authority resides in the speaker and in the subject matter that he is presenting. The role of the audience is to be essentially passive. The data, or subject matter, is more important than any changes in the audience's interests. In Group Structure B, it was intended to describe a situation in which authority resides fundamentally with the speaker, but in which the audience's moods and interests exert a definite influence over the speaker's choice and manipulation of subject matter. In Group Structure C, it was intended to describe a Chicago round-table type of discussion situation in which several individuals carry on a kind of group discussion in the presence of the larger, or audience, group. This kind of ”group discussion" is labeled "synthetic group discussion” 38 since its participants are preselected by the conference chairman for a preplanned activity, rather than being selected from the group membership by usual group processes. In addition, the actual audience group carries on no particu- lar role of discussion or group participation. In Group Structure D, it was intended to describe a basic group structure in which action and participation is carried on primarily by the group under the nominal direc- tion of the leader. The role of the leader is essentially de-emphasized in terms of authority, and his responsibility is to keep the center of gravity within the group and its natural leaders rather than upon himself. His skill as a group discussion leader is usually judged by the facility with which he is able to obtain a maximum of intra-group participation. In Group Structure E, it was intended to describe a group situation of the ”leaderless" variety. In this struc- ture there is no conference "leader" that has been assigned to the discussion group to help it formulate its own dis- cussion processes. Rather the group itself is simply as- sembled around a general interest and is given its own responsibility to carry on in any way that it sees fit. The only evidence of "structure" per se is that the convention chairman has recognized that a special interest, or discus- sion topic, exists; that a specially designated meeting room 39 and time has been set aside; and that an invitation has been issued to the interested to attend. In Group Structure F, it was intended to describe a group situation in which even these last vestiges of "struc- " have been largely removed. These situations are in- ture tended to represent the typical spontaneous groupings that occur in the corridors and lounges between and after meet- ings. They are largely transitory, forming, changing, and disappearing according to the immediate needs of its members. The actual descriptions of each group structure, of course, appear on the questionnaire itself, a copy of which is presented in the Appendix. Description of the Questionnaire The total questionnaire submitted to each individual in the research population consists of a ten-page document. The first section of this instrument consists of the "F" scale test; the second section consists of six sub-scales consisting of ten items each. Each sub-scale is headed up by a paragraph or two describing a specific stereotype of a group structure that is typically encountered at the aver- age convention or conference. This six-step series of group situations attempts to present structures that begin with a maximum of authority attached to the group leader and maximum constriction upon the group, then moves progressively toward a reduction of ho authority on the part of the leader and a reduction of con- striction upon the group, culminating finally in the more or less complete absence of either. Group Structure Attitudinal Statements Using each group structure stereotype as a reference point, a set of ten attitudinal statements were prepared about each one which provided opportunity for each respond- ent to agree or disagree with various aspects of the par- ticular group structure under consideration. Following the same scoring procedure as was established for the "F" scale portion of the total questionnaire, each individual was advised to indicate the level or strength of his agreement or disagreement in terms of the same code, i.e. "+3" for strong agreement and ”-3" for strong disagree- ment. In line with the basic hypothesis that authoritarian personalities would prefer the more rigid group structures and reject the more plastic group structures, the attitudi- nal statements were prepared in the direction that "strongly agree” responses would represent preferences for rigidity, and "strongly disagree" preferences would represent the converse. Similarly, in converting to uniform weighted scores, the same procedure was used as in scoring the "F" scale, namely "+3” would convert to "7", "+2" would convert to "6", and so on down to "-3" which would convert to "l". Lil In a few instances it was necessary to construct an item in an inverse direction in order not to overload its implied meaning. For these instances, in order to give the weighted conversion its correct value, the signs of the raw score must first be reversed before converting to the proper weighted score. Items which require the reversing of the plus and minus signs before conversion are as follows: Group Structure A - Item 9 Group Structure B - Items 14, 6, 7 Group Structure C - Item h Group Structure D - Items 1, 3, 5, 8, 9 Group Structure E - Items 1, 2, 9 Group Structure F - Items 3, Li, 5, 9 Group Structure Preference Statements The questionnaire employed in collecting the data is noted to contain an extra item.at the foot of each of the five pages which are concerned with group structures B, C, D, E, and F. In line with the special note that is set out in detail on page 3 of the questionnaire, the individual is to gradually build a progressive rank order of his general preferences for the six described group structures, that is, he is first to state his rank order of preferences between group structures A and B. Then, when he completes the next ten items under group C, he is to assign a rank order of preferences among the three groups. He then proceeds to rank the four groups from A through D, next the groups from A through E, and finally, the last six from A through F. h2 This procedure of having the respondent go through an experience of progressively building a final rank order of preferences for all six groups was felt to be more theoreti- cally sound than simply to present one final item at the end of the questionnaire which asked for a single listing of the entire rank order of preferences of all six groups. A progressive development of the final rank order provides greater opportunity for each group structure to be evaluated on an individual basis, and with only one new group structure at a time to be compared with group structures that have already had previous and independent evaluation. It was felt that for the respondent to make a single rank order of all the groups at but a single moment of consideration might lead to confusion and perhaps capriciousness in the assign- ment of values. For the purpose of analyzing and interpreting the data, only the sixth or final rank order of preferences for the entire six groups was considered. Method of Questionnaire Construction Using the Michigan Rehabilitation Association, with its professional membership of two hundred as a medium from which to select experienced program chairmen or veteran conference planners, six conference leaders were selected. Each of these individuals had either been president of the associa- tion and active in planning its annual professional convention, #3 or had served in the responsible post of being program chairman for the actual conference itself. These individu- als were selectively chosen because of their partisanship for particular conference techniques or processes. In order, these were as follows: 1. An bndividual whose particular interest was in research and who felt that the function of an annual pro- fessional conference was to provide an opportunity for re- search papers to be read. 2. An individual who believed that a conference should be built around outstanding speakers who were crowd drawers and attention holders. 3. An individual whose favorite mode of arranging a meeting was to assemble a panel of experts representing a variety of viewpoints and to have them informally discuss the topic of the day. h. An individual who was highly interested in group participation processes under the guidance of a discussion leader. 5. An individual who was not only interested in demo- cratically led group discussion processes, but who also had had favorable experience with the "leaderless" group dis- cussion process. 6. An individual who placed particular value on the social or extra-curricular activities of conference hh activities, and had a primary interest in the informal activities and gatherings that occur before, between, and after the official meetings. These six individuals were used as judges or discrimi- nators in the process of constructing appropriate descriptions for each group structure stereotype. Each description was re-written or revised until a satisfactory variability of opinion was Obtained from each of the six. The basic criterion was that at least two individuals should indicate some level of negative preference for the stereotype under consideration, and at least two individuals should indicate some level of positive preference for that stereotype. After completing the descriptions of the six group stereotypes, a modification of the same procedure was em- ployed in preparing the ten attitudinal statements that are listed under each. The modification consisted of assigning a pair of discriminators to each group structure. This pair of individuals consisted of the person who represented great- est partisanship for the type of group process described, and the individual who was most negative to that type of group structure. In preparing attitudinal statements or items about the stereotype under consideration, the criterion for item acceptance was that it must yield a positive re- sponse from the individual representing the positive pole, and a negative response from the individual representing the negative pole. This was for the purpose, of course, of LLS eliminating "universals", items agreeable to all. Basic theoretical considerations underlying the construction of each item was the parallelism suggested by the nature of the authoritarian syndrome as conceivably applied to a group process. These would include items depicting such dynamisms as ethnocentrism, conventionalism, traditional- ism, preoccupation with the dominance-submission dimension, the uncritical acceptance of the leader, et cetera. Plan for Analysis of the Data The performance of the non-authoritarian group was compared with the performance of the authoritarian group, item by item on each of the six group structures compris- ing the remainder of the preference questionnaire. Signifi- cant differences in response were identified by the signifi- cance of the critical ratios between the two groups of scores. In addition, a study was made of the contrasting performances of the two experimental samples in terms of the rank order in which each arranged the six group stereotypes on a pref- erence basis of ”most liked" to ”least liked". These findings, together with their necessary conclu- sions, are presented in accordance with the plan of organi- zation as set forth in Chapter I of this study. h6 CHAPTER III ANALYSIS or ATTITUDES IN GROUP STRUCTURE A Introduction It is the purpose of this chapter to present an analysis of the data which was gathered under Group Struc- ture A of the questionnaire. The description of Group Structure A was intended to represent the most rigid of group structures. In essence, it presents one speaker who does all the talking. The role of the audience group is fundamentally a passive one. They are to sit, immobilized, and listen. Except for a desultory question and answer period at the end of the meeting, there is no audience participation whatsoever. Throughout the meeting the speaker addresses himself only to the factual data that he has immediately at hand. He is not interested in de- parting from his prepared facts to cater to audience inter- ests or topical disgressions that might lead elsewhere. All authority resides in the data and in the speaker who produced it. If general group interest conflicts with sub- ject matter, it is subject matter that still rules. Below the description of Group A are presented ten statements, each of which represents an attitude about dif- ferent aspects of this kind of a group structure. 1+? The response patterns of the two experimental group to these ten items will be presented in this chapter with the vieWpoint of comparing and interpreting the mean differences between the two. The hypothesis in each instance is direc- tional, that is, that the non-authoritarian group will always be "to the left of", or average a smaller weighted score on each item than will the authoritarians. The weighting scale runs from 1 to 7, with the higher scores theoretically repre- senting greater preference for the more rigid group struc- tures, and the lower scores keyed in the direction of pref- erence for the more plastic group structures. Presentation of Data on the Distribution of Responses on Group Structure A Tables I and II present the actual distribution of responses to Group Structure A on the part of each experi- mental sample. Before constructing the original frequency distribution table, all answers were converted to their re- spective weighted scores according to the following key: Raw Score Conversion Key +3 +2 41 Omitted -l -2 -3 H Nw-F'WO‘NJ to In the case of those attitudes which were stated in a reverse or negative direction, the signs were changed on the raw scores so that the correct weighted conversion could occur. DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE A (See page hi in Chapter II.) TABLE I BY'NON-AUTHORITARIANS m Item Weighted Score Responses Total 2 3 u 5 6 7 Number 1 8 26 10 26 11 h 85 2 8 22 19 23 10 3 85 3 16 1h 18 1h 15 8 85 h 3 18 25 17 15 7 85 5 3 8 10 31 26 7 85 6 S 12 27 19 18 85 7 19 26 23 6 6 5 85 8 11 18 18 12 16 10 85 9 6 3 8 27 2o 21 85 10 37 19 10 10 8 1 85 M9 TABLE II DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE A BY AUTHORITARIANS ._._._. —--—.—-.—— “-~~._.a——_ -—_. - —-——-. W. - -——.._4. *mr- _ ._.e _ _ — . ‘ Weighted Score Responses Total Item 1 2 3 h 5 6 7 Number 1 S S 15 19 2n 17 85 2 h 11 8 26 2o 16 85 3 2 5 7 9 18 uh 85 h o 3 13 17 21 31 85 5 o o 2 12 2h h? 85 6 1 6 10 21 23 24 85 7 11 15 29 12 9 9 85 a 6 9 16 8 21 25‘ as 9 h 8 7 3o 15 21 85 10 13 12 15 16 7 22 85 50 Analysis of the Differences Between the Mean Responses on Group Structure A In order to draw significance about the distribution of responses of the two experimental groups as they appear in Tables I and II, it was first necessary to calculate the mean response of the two experimental groups with regard to each item. The means presented in Table III are the arith- metic means calculated by the formula M s lgfi in which M represents the arithmetic mean, X represents each weighted score, and N represents the number of scores which in all cases will be 85. In addition, the standard deviations from the means were computed in the usual way such as is illustrated by Guilford (37:95-97) using the basic formula r= £25.. N Table III summarizes the means and standard deviations of the two experimental populations on these ten items. An inspection of the data indicates that the mean responses of the non-authoritarian group are in each case smaller than the mean responses of the authoritarian group to the same item with the exception of item 9 wherein the mean responses of the two groups are approximately equal. This substantially coincides with the directional aspects of the hypothesis in which it was predicted that these non- authoritarian means would occupy a point to the left of the authoritarian mean for an equivalent distribution. 51 TABLE III MEAN RESPONSES TO GROUP STRUCTURE A BY AUTHORITARIAN AND NON-AUTHORITARIAN SAMPLES m: J— Non-authoritarian Group Authoritarian Group Mean Standard Item Mean Standard Response Deviation Response Deviation 3.69 1.8 1 h.92 1.8 3.59 1.7 2 h.85 1.8 3.69 2.0 3 5.81 1.7 3.97 1.8 h 5.56 1.5 h.81 1.6 5 6.3h .9 5.00 1.7 6 5.3M 1.6 2.83 1.7 7 3.59 1.9 3.85 2.0 8 h.86 2.1 5.15 1.7 9 5.03 1.8 2.h7 1.8 10 h.21 2.2 When an experimental design is concerned with specifying direction, as is true of this one, then the experimenter is subjecting the data to a one-tailed test rather than to a two-tailed test on the distribution curve. Mood, in his book Introduction 32 the Theory 9f Statistics (56:263), discusses this one-tailed approach and in fact urges its use whenever it is justifiable because it is a more power- ful test, which is to say that it is more likely to detect a real difference when it exists. This, of course, leads 52 to the logical necessity of re-adjusting the indices for the various critical points of "t" since the degrees of freedom from chance are confined to the theoretical limits of but one tail of the distribution curve. Mood also presents a table of one-tailed and two-tailed .critical points for various group sizes (56:h25) of which Table IV comprises a useful summary. Item Analysis of Differences Between the Means It is apparent that for the ten items listed under Group Structure A in the questionnaire, there is a differ- ence in the response pattern of the two experimental groups. These differences may be more clearly recognized by comput- ing the critical ratios of the differences between their respective means. Using the method presented in Guilford's text (37:213-216) these critical ratios may be computed from the formula: ml ' m2 T: Fidirzdfi 1 + 1 n1 +. n2-2 n1 n2 in which "T" represents the critical ratio, "m" represents the mean, "n" represents the number of cases, and "d" repre- sents the difference. Table V presents a summation of the critical ratios of each of the ten items, plus an interpretation of their 53 om.m mo.m Hm.m mm.m mm.m .mm.m mm.m mo.m oo.H mo.H mo.a m~.H .pasuaH mm.m mH.m mm.m mr.m mo.m om.m om.m so.m mo.H mo.a mo.a o~.H owe os.m ma.m om.~ oo.m oo.m em.m om.~ oocm ooam soda ao.a om.a or Hess m Hess a Hess Niamey a ease m Hess a Hess m ease a mmeImummmmlwlaese m Hess a no H\ H mm} ma. mm mm RS 996 cocoouuooo no uaobon no mammana QMQHH 3mg“. Fm.” 5h level of significance. The interpretation as to whether the various levels of confidence are insignificant, sig- nificant, or very significant is also based on a table presented in Guilford (37:209) which may be summarized as follows: Level of Confidence Rough Conclusion Below 10% level Insignificant At the 10% level Insignificant At the 5% level Significant At the 2% level Significant At the 1% level Very significant At the .5% level Very significant Beyond .5% level Very significant An examination of Table V indicates rather strong statistical evidence that for most of the items in question there is a very significant difference in the performance of the two experimental groups, and that these performance differences are generally in line with the directional hypothesis of the study. Two items are excepted, items 6 and 9, but it may be noted that the first Of these is located at the 10 per cent level of confidence which, though statis- tically not of sufficient significance to suggest the re- jection of the null hypothesis, is at least inclining some- what in that direction. With regard to item 9, however, it is apparent that no statistical significance exists to dis- tinguish between the two groups. Over-all, the total per- formance of the two groups on the ten items is signified by a critical ratio of 8.58 which is indicative of a total per- formance that is statistically very significant. 55 TABLE v CRITICAL RATIOS FOR GROUP STRUCTURE A W Item Critical Level of Conclusion Ratio Confidence l h.39 0.1% Very significant 2 h.63 0.1% Very significant 3 7.31 0.1% Very significant h 6.2M 0.1% Very significant 5 7.73 0.1% Very significant 6 1.3h 10.0% Insignificant 7 2.71 0.5% Very significant 8 3.20 0.1% Very significant 9 -0.hh null No significance 10 5.59 0.1% Very significant Totals 8.58 0.1% _"Very significant 56 Analysis of Performance on Individual Items In addition to the interest that inheres in the statistical components of a study of this type, it may be assumed that considerable interest attaches to the content aspects of each item which elicited these statistical dif- ferences. Interpretation may be guided by statistical in- dices, but interpretation must be related to some axis of content or meaning. This study has been in a sense a study of attitudes, and the items that comprise these attitudes need be reviewed in order to present a basis from which such interpretations may be derived. To more properly set the atmosphere for inspecting these individual items, the group structure stereotype to which these items pertain is set out just as it appears in the questionnaire, as follows: Group Structure A The individual appearing on the rostrum at this meeting is an authority in the field of your gen- eral interest. By reputation you know that he is a very thorough and careful scientist. It has been reported to you that when he reads a paper at a convention, the subject is covered in abso- lute detail, and few, if any, questions are asked by members of the audience. This is the kind of a man who seeks to address himself to a serious minded audience. He is not so much concerned with trying to speak in an en- tertaining manner, but rather with an absolute devotion to the facts, regardless of what they might be. It has been his uncompromising pur- suit of this goal that has made him the outstand- ing authority that he is. You have not attended a meeting under his direction before. 57 The first of the ten items appearing below this stereotype was as follows: 1. This is the kind of a meeting I enjoy attending. This item had been constructed with the idea of pre- senting an over-all, unrefined attitudinal prOposition of accepting or rejecting, at any one of several levels, this general group structure. It will be noted that the mean weighted response of the non-authoritarian group was found to be 3.69 for this item, Which, when related to the scor- ing key, indicates a slight negative preference attitude fer this group structure. Conversely, for the authoritarian group, its mean weighted response of h.92 indicates a more definitely positive preference for this group structure. The critical ratio between these two meals is noted to be a very significant index of h.39. The second of the ten items was as follows: 2. As a general rule, audience members gain a great deal from listening to a speaker of this type. Students of the pragmatic educational philosophy of John Dewey emphasize that the learning process is an inter- actional process based on "doing" rather than just passively ”listening". This item was constructed with the idea of tapping some of the emotional flavor that surrounds levels of attitude toward a grOup process which is stereotyped fl primarily as a non-doing" activity. In Dewey's philosophy, 58 the traditional leader's role of authority is de-emphasized and its center of gravity becomes group-centered. Thus, an expression of agreement with this item would appear to lead in a direction away from Dewey's democratic philosophy, and toward a preference for more autocratic lines of authority. Of course, in the inspection of the item from a completely intellectual point of view, one could perhaps defend an agreement with the statement up to a certain pofilt on purely scientific grounds. However, points of view are not always purely intellectual, and it is interesting to note how the two experimental groups performed. The non- authoritarian group, with its mean response of 3.59, is again indicated by the key to be about as negative to this statement as it was for the first item in the series. Mean- while, the authoritarian group, with its mean response of h.85, is noted to be somewhat in favor of the proposition at about the same position as it took with respect to the first item. The critical ratio of h.63 between them is but further evidence of how significantly the two groups varied in their attitudes toward this item. The third item is noted to be: 3. A most sensible group arrangement is to have a speaker who can present the facts, and an audience that knows how to listen. Here the presentation of the power lines of authority, emphasizing the dominance-submission dimension, is presented 59 in rather frank nakedness. Frankel-Brunswik (1:22h-229) in her analysis of the authoritarian personality found consid- erable evidence that a key element in the authoritarian constellation was that of preoccupation with authoritarian submission. For such authoritarians the leader is ideally a strong power figure and the group is ideally a disciplined followership. The non-authoritarian group again manifests an ambigu- ous (slight increase in the standard deviation), but on the whole, somewhat negative response to this statement. Their mean response is noted to be 3.69. The authoritarian group, however, soars to a new high of agreement with an index of 5.81, or an unmistakable endorsement of the general proposi- tion. The critical ratios between these two levels of response is noted to be 7.31. The fourth item in the series is as follows: h. It is more professionally desirable for a speaker to stick to those facts which he can prove rather than to get off into speculations. Following the theoretical assumptions portrayed by Frenkel-Brunswik and her associates, "authoritarianism" is viewed essentially as a neurosis. As in any classical neurosis there is presumed to be elements of insecurity which, in ”authoritarianism", is manifested by clutching to rigid "truths", truths that have "stood the test of time", or have been enunciated by eminent and impeccable 60 authorities. Excursions into slippery speculations or "new- fangled" ideas are felt to be not only foolish in themselves but possibly to contain the seeds of dissolution for the belief system that currently governs the society. This item was written so that respondents could project their feelings as to the degree to which they require the leader to stick to a "safe” and rigid role. The non-authoritarian group, with its mean response of 3.97, manifests, according to the key, a generally indifferent or almost imperceptibly negative attitude toward the item in question. However, the authoritarian group indicates its presumed "authori- tarianism" with a definite response indexed at the 5.56 level. The critical ratio of 6.24 between the two groups confirms their differences. The fifth item in the series is: 5. It's rather nice to be able to sit back in the audience and listen to some- one talk who knows what he is talking about. This item was written with the view of obtaining 333 the attitude toward the role of the leader, but rather the attitude of the group toward the role of the followers. Implicit in the statement is the emphasis upon passive followership with a generalized uncritical attitude toward the idealized or acknowledged authority of the leader. These elements are inherent in the description of the authoritarian syndrome. In comparison to its previous mean responses, this 61 item was not wholly devoid of some attraction to the non- authoritarian group. Its mean response to this item of h.81 places it toward the upper end of the neutral zone and near to a level of "slight agreement". One could sur- mise that this increase of agreement occurred as the result of the presence of some theoretical neurotic influences within this group, or as a result of the attraction of other variables which it aroused. It must be conceded, of course, that the same hazard would be operative for all the items. In contrast with the non-authoritarian group, the authoritarian group made its highest mean score in the entire series of ten items with regard to this particular item. Its mean score of 6.3h placed it at a level between moderate agreement and strong agreement, thereby tending to confirm that this kind of group situation offered rather important satisfaction to this particular group of respond- ents. The critical ratio of 7.73 indicates that thoughthe level of agreement on the part of the non-authoritarian group did rise somewhat, the relative difference between the two groups was more than maintained. Item six appeared as follows: 6. It is to be hoped that the reading of scientific papers will continue to be en- couraged by the professional association. 62 The controversial aspects of whether a professional convention shall be devoted fundamentally to a series of paper-reading activities has been at least officially recognized to the point whereby the American Psychological Association has become engaged in an opinion poll on the matter. Item 6 and item 9 that follows later were prepared with the idea of focusing attention on this perhaps most common of group communication techniques in a professional conference setting. The paper-reading technique appears at face value to be the most precise and rigid of mediums to communicate professional information. Ordinarily the scientific paper contains little in the way of Speculative or untested ideas but is rather a fortress of data, buttressed by statistics, precedents, and references to supporting authorities. It was presumed that this item, and its related item - number 9- would be as discriminating in the different responses that it would elicit from each group operating as a whole as it was in discriminating between the several individuals who expressed the negative and positive agreements to the items at the time that they were prepared. Nonetheless, the differential responses of the two experimental groups to this item were not of statistical significance. The non-authoritarian group with its mean response of 5.00 occupied approximately the same position 63 of moderate endorsement toward this item as did the authori- tarian group with its index of 5.3h. The critical ratio of 1.3h is regarded as insignificant at the 10 per cent level of confidence. In interpreting this general acceptance on the part of both groups, one might surmise that the variable postulated was pure and that the resulting performances need be evalu- ated no further; or, one might wonder whether the item generally precipitated another variable of greater strength, Eggh‘gg the necessity £2 preserve the trappings 2; status for the association without acknowledging that this mode of communication is a personally satisfying one. At any rate this is a question for more proper investigation in a sepa- rate study. The seventh item is listed as follows: 7. Professional meetings of this type should generally be closed to non-members of the professional association. Levinson speaks of ethnocentrism (1:150) as one in which there is a rigid distinction maintained between the ingroup and the outgroup. The ingroup regards itself as rightfully dominant and in some way superior to the outgroup. It is an exclusive group and membership in it becomes a symbol of status that shall not be too casually given to others. Item 7 was written to give expression to the ethnocentric variable that may be characteristic of a professional association. The two eXperimental groups employed in this 6L1 study maintained a very significant difference of response to this item, but both occupied positions in the left half of the disagreement-agreement scale in contrast to the characteristic positions they had adopted in the first several items. The non-authoritarian group, with its mean response of 2.83, is placed at a level between "moderately disagree" and "slightly disagree" whereas the authoritarian group with its index of 3.59 is placed at a level half way between "slightly disagree" and "neutral". The critical ratio between the two of them was 2.71. Item eight is as follows: 8. Those who possess the least profes- sional training are usually least inter- ested in hearing someone read a scientific paper. Perhaps the basic texture of authoritarianism as described in the literature is the generalized hostility that is latent in the affected individual. The pnlcess of seeking status and recognition for himself is twin to the process of projecting fault or inferiority toward those who either cannot give him the status that he desires, or whom he feels may discredit the symbols of status which he himself so highly prizes. Item 8 was written in order to measure any group differences on a typical status en- hancing, fault projecting platitude of the more spurious and untested sort. The non-authoritarian group reacted to this particular platitude with a mean response of 3.85, or 65 at a point ranging from the very slight level of disagree- ment to close to neutral or indifferent. The authoritarian group, however, found some greater attraction in this item, as signified by their mean score of h.86, and located them- selves in the upper neutral ranges approaching very closely to the slight agreement range. The critical ratio between them of 3.20 indicates that the difference between the two is well founded. Item nine is listed as follows: 9. Program planners would do well to eliminate the reading of scientific papers. As referenced in the discussion of item 6, this item yielded no differences in performances between the two ex- perimental groups. The item was conceived originally to double check on the response patterns to item 6, and as such it was stated in a reverse direction and in even more sweeping scope. In other words, statements of agpeement to item 9 were to be expected from.the non-authoritarian group and statements of disagreement were to be expected from the authoritarian group. Instead, pgph experimental groups disagreed with the item, and at about the same level. The non-authoritarian group obtained a mean response of 5.15, whereas the authoritarian group obtained a score of 5.03. Bearing in mind that the ratings for this item 333 ig'phg reverse direction, both groups have an identical rating on the key of "slight disagreement". As was 66 suggested in the aforementioned discussion, the manifest content of the stated item must be accepted on its face value as non-discriminatory between the two groups, or else a variable other than the discrete activity of "paper-reading" as such was so elsewhere apparent in the item that both groups responded in unison to the latter. A possible statistical clue in support of this latter hypothesis lies in a compari- son of the paired critical ratios on each of the items. In item 6 when the statement is more moderately phrased, the critical ratio between the two is at the 10 per cent level of confidence; but when the item is more harshly and sweep- .yggly stated as it is in item 9, the critical ratio falls to a -0.hh. Therefore, refinement of these items into softer and less threatening statements for the purpose of re-testing becomes immediately suggestible as worthyof fur- ther research. Item ten appears as follows: 10. A real authority in the field should have the right to make his presentation without having to put up with constant questioning from his audience. The literature on authoritarianism includes considerable emphasis on the dimension of authoritarian submission. In manifestation of this symptom the group is characterized as to be rightfully submissive to the idealized authority figure, and not to be critical. The "father figure" merits this "right" by virtue of the superior traits through which 67 he achieved this position of leadership. Item 10 attempts to put in question the degree of acceptance to which each group would accept this principle. The non-authoritarian group exhibited a general disagreement with the item as in- dicated by a mean response of 2.h7; however, the authori- tarian group was not nearly so negative, exhibiting a mean response of h.21, or generally in the neutral zone With a barely perceptible trend in the direction of the slight agreement side of the scale. The critical ratio between the two groups was 5.59, which seems to be highly significant. Summary The stereotype of Group Structure A, with its emphasis on a subject matter centered leader, and a passive audience, had been constructed with the intent of portraying the most rigid end of a six-step scale of group structures vary- ing from most rigid to least rigid in construction. With the basic hypothesis in mind that non-authoritarians would tend to have a negative preference for this type of struc- ture, whereas authoritarians would have a positive prefer- ence for this type of structure, it would be expected that the differences between the two groups would be exaggerated when responding to an exaggerated or extreme stimulus. This chapter has presented a report on the evidence which pretty well confirmed that expectation. The critical ratios between 68 the two groups, item by item, has been substantially high. Only one pair of items, and these were the positive and negative statements of the same basic concept, failed to yield any significant differences in response between the two groups. Aggregating the performance of the two experi- mental groups on the total series of the ten items studied, the mean response for the non-authoritarian group was found to be 3.9, or at a key value which is located on the border- line between slight disagreement and neutrality. The authori- tarian group, on the other hand, exhibited an over-all mean response of 5.0, or a key value that coincides with slight agreement or preference for these items. The over-all critical ratio between the groups of 8.58 is statistically of great significance. 69 C HA PTER IV ANALYSIS OF ATTITUDES IN GROUP STRUCTURE B Introduction It is the purpose Of this chapter to present an analysis of the data which was gathered under Group Struc- ture B of the questionnaire. The description of Group Structure B was intended to represent a slightly less rigid structure than that of Group Structure A. It is true that in Group Structure B there is still considerable rigidity. It is speaker-centered and largely speaker- dominated. The influence of the audience is largely minimal, although it is present to a degree. The speaker addresses himself to the group rather than to more subject matter. He is sensitive to audience interests and, within limits, bends himself to suit what he believes is their will. He is what is more colloquially known as the "crowd pleaser” type of speaker. Nonetheless, though he caters to, and is influenced by, his audience, it is he that does all the talking and it is only he that has the focus of attention. Below the description of Group B are presented ten statements, each one of which represents an attitude about different aspects of this kind of group structure. 70 The response patterns of the two experimental groups to these ten items will be presented in this chapter with the vieWpoint of comparing and interpreting the mean dif— ferences between the two. As in Chapter III, the hypothesis continues to be directional in that the non-authoritarian group is predicted to always produce a mean score to the left of each mean score produced by the authoritarian group on a given item. Presentation of Data on the Distribution of Responses on Group Structure B Tables VI and VII present the actual distribution of weighted score responses to Group Structure B on the part of each experimental sample. All raw score answers were first converted to weighted scores in accordance with the process and key as described in Chapter III. Analysis of the Differences Between the Mean Responses on Group Structure In order to draw significance about the distribution of responses of the two experimental groups as they appear in Tables VI and VII, it was first necessary to calculate the mean response of the two experimental groups with re- gard to each item. The means presented in Table VIII are the arithmetic means calculated in the same manner as described in Chapter III. In addition, the standard devia- tions are similarly included. TABLE VI DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE B BY NON-AUTHORITARIANS 71 W Item Weighted Score Responses Total 1 2 3 h 5 6 7 Number 1 6 3 31 30 12 85 2 13 21 20 20 6 85 3 32 25 17 8 2 1 85 h 5 22 19 13 21 5 85 5 10 29 23 1h 5 h 85 6 14 17 18 16 1h 6 85 7 21 17 In it 5 85 8 2 8 19 18 23 15 85 9 13 23 18 12 11 8 85 10 27 31 ll 8 h h 85 TABLE VII DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE B BY AUTHORITARIANS 72 Item Weighted Score Responses Total 1 2 3 h 5 6 7 Number 1 5 11 11 2h 22 12 85 2 15 28 1h 9 13 85 3 29 25 23 2 2 u 85 h In 10 18 1h 19 10 85 5 8 13 20 1 1h 18 11 85 6 11 2h 20 16 5 9 85 7 9 15 13 23 7 18 85 8 1 1 10 18 16 39 85 9 2 5 18 1h 23 23 85 10 8 19 12 18 1h 1h 85 73 Table VIII summarizes the means and standard deviation of the two experimental populations on these ten items. TABLE VIII MEAN RESPONSES TO GROUP STRUCTURE B BY AUTHORITARIAN AND NON-AUTHORITARIAN SAMPLES M '- ———-r~—v Non-authoritarian Group Authoritarian Group Mean Standard Item Mean Standard Response Deviation Response Deviation 5.21 1.5 l h.66 1.8 h.19 1.8 2 3.9a 1.9 2.26 1.h 3 2.33 1.5 3.90 1.9 h h.02 2.1 3.12 1.7 5 h.15 2.0 3.62 2.0 6 3.h3 1.9 3.72 1.7 7 h.25 2.1 h.80 1.7 8 5.79 1.5 3.h7 2.0 9 5.12 1.8 2.52 1.7 10 h.l6 2.1 An inspection of the foregoing data indicates that in seven out of the ten pairs of mean responses for the two experimental groups, the non-authoritarian group obtained a smaller mean score than the authoritarian group. This is supportive of the directional prediction of the under- lying hypothesis. Of the remaining three pairs of responses, 7A the mean responses to items 2 and 6 came within twenty- five hundredths and eleven hundredths respectively of obtaining the identical scores. Only with regard to item 1 was there a conspicuous conflict between the real location of the score and the predicted location of the score. In the main, the obtained data is fairly satisfactory in terms of confirming the prediction. However, it is apparent to some degree that as the group structure stereotype under- goes some reduction in the exaggeration of its rigidity features, the performance differences between the two ex- perimental groups tend to draw closer together. This trend will be noted more definitely in the differential perform- ances on other group structures that are to be described in subsequent chapters of this study. Item Analysis of Differences Between the Means It is apparent that for the ten items listed under Group Structure B in the questionnaire, there is a differ- ence in the response pattern of the two experimental groups. These differences may be more clearly recognized by com- puting the critical ratios of the differences between their respective means. Using the same processes as were employed in Chapter III with respect to Group Structure A, Table IX was constructed which presents a summation of these critical ratios plus an interpretation of their levels of significance. 75 The level of confidence index is based upon the same one- tailed distribution table as was used for treating the parallel data in Chapter III. TABLE IX CRITICAL RATIOS FOR GROUP STRUCTURE B W Item Critical Level of Conclusion Ratio Confidence 1 -2.lh null No significance 2 -0.88 null No significance 3 0.31 null No significance h 0.39 null No significance 5 3.68 0.1% Very significant 6 -0.6h null No significance 7 1.81 5.00% Significant 8 3.98 0.1% Very significant 9 5.73 0.1% Very significant 10 5.62 0.1% Very significant Totals 5.37 0.1% Very significant An examination of the foregoing table indicates that five of the indices are statistically significant in sug- gesting that the contrasting performances of the two ex- perimental groups were not due to pure chance factors. Four of these indices were significant beyond the one-tenth 76 of 1 per cent level of confidence. With regard to the other five items, no important differences in the performances of the two groups were observable. Over-all, the total per- formance of the two groups on the ten items is signified by a critical ratio of 5.37, so that the discriminatory power of the more efficient items lends an aggregate weight to the total performance that is statistically very significant. Analysis of Performance on Individual Items In addition to treating with the purely statistical aspects of the data, this study is also concerned with the content aspects of the items under examination. Following the same rationale as was presented in Chapter III regard- ing the ten attitudinal statements that pertained to the group structure in question, a similar presentation is ap- propriate at this point. As a point of reference, the stereotype of Group Structure B is presented just as it appears in the questionnaire, as follows: Group Structure B The principal Speaker at this meeting is known to be an engaging personality who tries to make his talks as interesting as possible for his audience. Though he works in your general field of interest, he has contributed no particular research to the professional journals. By reputation you know that he always speaks extemporaneously, Seldom refers to notes, and generally offers no documents- tion for his principal statements. Usually he deals with interesting and speculative ideas, rather than with factual data. 77 Ordinarily, several members of the audience engage him in a lively question and answer period at the conclusion of his remarks. The speaker seems to be principally motivated in stimulating his audience rather than in presenting an organization of highly dependable data. You have never attended a meeting of his before. The first of the ten items appearing below the stereo- type is as follows: 1. This is the kind of meeting I enjoy attending. This item is noted to be identical to the first item for Group Structure A. It was constructed with the same general purpose of presenting an over-all and unrefined proposition of accepting or rejecting this general group structure. Since the group structure is fundamentally a speaker-dominated structure with little or no participation on the part of the audience, the hypothesis would urge that the authoritarian personality would tend to be more drawn to this kind of setting than would be the non-authoritarian personality. Nonetheless, the reverse response actually occurred. The non-authoritarian group eXpressed some degree of preference for this structure with an average weighted score of 5.21. The authoritarian group, however, were on the whole rather indifferent about this type of group structure, obtaining a mean response of n.66 which located them about midway between neutrality and slight preference. 78 When the first draft of the stereotype of Group Structure B was constructed, it elicited such universal favor with the group of judges on whom it was tested that it had to be rewritten a number of times in order for it to stimulate some lggk of affection or preference on the part of a representative fraction of them, and thereby attain some discriminatory power. The direction of these rewrites was to place increasing emphasis on this individu- al's lack of obeisance to established subject matter or rigid truths, and to imply a willingness on his part to take occasional excursions into flights of fancy or unsub- stantiated speculations if such were the audience's inter- est. Theoretically, this element is repugnant to that part of the authoritarian personality which is most depend- ent on a stable and sober approach to the presentation of dependable subject matter. Thus Group Structure B presents somewhat of an ambivalent situation to the authoritarian personality, offering as it does the high level of authority organization that appears in Group Structure A, but not the same high level of authority in the delivery of the informa- tion which ostensibly is its reason for organization. It might be surmised then that broad items treating the whole of Group Structure B would not be as efficient for discrimination as would narrower items that would treat of but an element of this structure. Some of the remaining nine items to be inspected are of this latter type. 79 The second item in the series is as follows: 2. As a general rule, audience members find it most profitable to listen to a speaker of this type. In line with the immediately preceding discussion, this item is similar to the first item in that it treats the stereotype as a whole, rather than referring to a sub-portion of it. Since the authoritarian group had already demonstrated some apathy toward this less exact- ing and less precise type of information-sharing behavior (compared with the highly exact methodology described in Group Structure A), they remained consistent with them- selves in tending to almost disagree with the item in question. Their mean score of 3.9h places them at the border line between neutrality and slight disagreement. The non-authoritarian group is not so sure that it agrees aca- demically with the item as stated in spite of their general attraction for this type of group "entertainment". Their mean response of h.l9 indicates a response that is entirely neutral in attitude, not greatly different from that of their control group. The critical ratio between the two is observed to be insignificant. Item three appears as follows: 3. Professional meetings of this type should generally be closed to non-members of the professional association. Item 3 is noted to be a restatement of item 7 in the group A series, but of course, referenced this time to a 80 somewhat different kind of structure. As in the discussion for this item in the preceding chapter, this is an item devised to test the degree of ethnocentrism that is aroused on the part of each group when they relate themselves to this kind of a structure. The non-authoritarian group, with its mean response of 2.26, exhibited a rejecting attitude toward this item of moderate disagreement. It is interesting to note how the authoritarian group responded to this item. It will be recalled that in Group Structure A (which was possibly regarded by them as a more ”professional" type of meeting than is connoted in Group B), this sample of respondents obtained a mean score of 3.59 to this item in comparison to a non-authoritarian score of 2.83. To the same item in the present group struc- ture, their mean score dropped to 2.33, or just about iden- tical with that of their paired sample. Since ethnocentrism is theorized as being a property that is most manifest when the group has very "high" standards, or is exclusive, it might be inferred at this point that the authoritarian sample had defined this structure as one geared to the ad- mission and attention-attracting level of "the general public” rather than being an "exclusive" group at the level which they would prefer. Item four appears as follows: u. If the audience manifests other interests than in what the speaker is saying, he should 81 try to re-direct his discussion toward those topics which particularly attract their attention. Item h was constructed with the idea of focusing at- tention on what should be the role of the leader toward the question of following the center of group gravity, or toward pursuing his own ends and purposes. The result of the item testing on the two groups yielded no significant differences in attitude between the two. Both presented a completely neutral response pattern. ‘ This was an inverted item in which it was believed that the authoritarian personality would tend to reject any deviationism upon the part of the leader, or, in short, to disagree with the statement. Perhaps having built up a special kind of attitude toward this particular type of in- dividual, this group of respondents was not too receptive to the proposition that the stereotype fairly described their own ideal type of leader, and were perhaps reluctant that the described individual should assume a strong role at cross purposes to some of the interests of the group. Item five is set forth as follows: 5. When a speaker departs from his pre- pared assemblage of facts in order to attract audience interests, he is not properly discharging his professional obligation to report all the facts. This item was addressed to a Specific element in the stereotype wherein attention was focused upon the order of 82 ascendancy between absolute and relative rigidity in pre- senting subject matter. It was surmised that the non- authoritarian sample would not be as partial to a doggedly uncompromised offering as would be the more authoritarian population in keeping with the theoretical foundations that have been elsewhere described in this paper as con- comitant with that type of personality organization. Consistent with this premise, the non-authoritarian sample obtained a mean score of 3.12, or a rating definitely in the disagreement zone toward this item. The authoritarian sample, however, yielded a mean response of h.lS, which is to saytmmt they took no stand one way or another against the proposition. The critical ratio between the two groups of 3.68 is very significant at beyond the one-tenth of l per cent level of confidence. Item six appears in the questionnaire as follows: 6. It is more professionally desirable for a speaker to leave factual reporting to the journals, and to employ conference time in the general discussion of whatever items are of current interest to the dele- gate assembly. Item 6 was constructed with the intention of challeng- ing those processes of conference organization in which the specific professional information that is to be trans- mitted is highly preplanned. It is slanted in the direction of fostering group-centeredness and group authority in the choice of topics that are to be discussed and in the direction 83 that the discussion shall take place. It is not illustra- tive of actual group-centered processes in the fuller sense of the term, but is intended to represent some nascent stir- rings in that direction. The item also includes an element which rather depreciates the factual preplanning activity characteristic of many conference speakers when it suggests that such factual reporting should be more or less left to ' —- —_ --7172‘m_z" v- ' the journals! In examining the differential responses of the two experimental groups, it became apparent that the item did not bring out the total discrimination that it originally promised. Both group samples evinced a slightly positive, but almost neutral, reaction to the statement. The non-authoritarians obtained an average score of 3.62 which, for an inverted scale item, signifies a rating of slight agreement, as anticipated. The authoritarians, how- ever, obtained a score of only 3.h3, which was not in sig- nificant contrast to that of its control, and no particular critical ratio could be found between the performances of ..~.. ' it '75-“! each group on the item. Item seven appears in the questionnaire as follows: 7. If a program planner has to choose between one who is an excellent authority in the field but not too much of a speaker, as against one who is an excellent speaker but not too much of an authority, he should generally choose the better speaker. Like the preceding item, this item was constructed in an inverted direction with the eXpectation of eliciting 81L agreement from non-authoritarians and rejection from authoritarians. It was designed to focus attention on two aspects of a speaker's role: one in respect to the speak- er's delivery, personality, and rapport with his audience; the other with respect to the authority of subject matter. In a conflict of choice between the need for the "human touch" as opposed to the need for factual data, it was sur- mised that the triumph of data would occur for the authori- tarians but not for the non-authoritarians. The theoretical basis for this assumption may be noted in the discussion pertaining to item A in Chapter III. An examination of the evidence indicates some neutrality to this item on the part of the non-authoritarians with a slight trend in the direction of accepting the statement as evidenced by a score of 3.72 which, when inverted, would compare to an index of h.28. For the authoritarians, their index of h.25, when inverted to 3.75, indicates some level of disagreement with this item. The critical ratio between the two groups is noted to be 1.81 which is significant at the 5 per cent level of confidence. Item eight is noted to read as follows: 8. A truly professional organization should not use up the valuable time of its membership with speeches by those who are least qualified to make an au- thoritative contribution. This item was concerned with arousing attitudes of authoritarian aggression toward theoretical newcomers to 85 the group. It tends to venerate experience and the tradition-rooted statuses that are attached to the older and therefdre more familiar authorities, and to inhibit equalitarian participation on the part of all. Rather than imputing equal dignity and worth to each individual, this platitude imputes more dignity and worth to some and less to others. It was conceived with the idea that it would particularly attract the authoritarian personality, since it emphasizes a highly rigid arrangement of power figures, and implies that the proper role of the remainder of the group is that of subordination. The non-authoritarian group, with its mean score of h.80, placed itself essentially in the neutrality zone, but tending in the direction of an imperceptible degree of agreement. The authoritarian group, however, with its score of 5.79, placed itself very definitely in the agreement zone, thus giving strong support to the underlying hypothesis. The critical ratio between the two was noted to be 3.98, or significant beyond the one-tenth of l per cent level of confidence. Item nine is noted to read as follows: 9. Generally speaking, there is no particular place on a professional platform for speakers who hold audience attention by a presentation of ideas that are unorthodox, unproven, and perhaps sen- sational in nature. 86 Item 9 was intended to be conSpicuous by its rejection of the unconventional approach to conference processes. The authoritarian personality was theorized to possess a prefer- ence for conventional, sober, middle-of—the-road activities, and to distrust or be fearful of unusual or different points of view, particularly for those theories where the verifica- tion of research evidence or the benediction of hallowed leaders in the field had not yet been given. This is a categorical item which generally excludes all such speakers rather than a selective few, and therefore more properly portrays a statement of a generalized attitude. In accord— ance with the basic hypothesis, it was presumed that this attitude would be more characteristic of authoritarians than non-authoritarians. The non-authoritarian group, with its mean score of 3.h7, tended to disclaim this item as descriptive of their own set of attitudes. The authoritarian group, on the other hand, yielded a positive score of 5.12 which gave rather definite support to the predicted assumption. The critical ratio between them of 5.73 was but further concurrence of the true dissimilarity of the two groups with respect to this item. Item ten in this series is noted as follows: 10. For the most part, the role of a con- ference audience should be to give courteous attention to the speaker, rather than to raise 87 challenges to the soundness of his ideas, or to express contradictory viewpoints. This item states in another way an aspect of the leader- follower dimension in which the idealized power figure is to be accepted uncritically and submissively. It would be upsetting to the rigidity of organization and power balance if a group were to possess too strong an attitude of self- assertion rather than a readiness to identify with the lead- er. The item seems to give principal emphasis to audience discipline rather than to audience participation. As an attitude to be possessed by a non-authoritarian personality, such would be most unexpected, and in this the evidence is equally corroborative. The non-authoritarian sample obtained a mean score of 2.52 on the item, or a rating that rather definitely rejects it. Authoritarians, however, did not find the statement nearly so distasteful. They obtained a mean score of h.16, which in relation to the performances between the two groups was represented by a critical ratio of 5.62, or a discrimination beyond the one-tenth of l per cent level of confidence. Summary The stereotype of Group Structure B concerned a leader- centered process in which the group was allowed a minor role of influence as compared with its role in Group Structure A wherein such audience influence was almost totally absent. 88 Chapter IV has been concerned with attitudes toward a group structure in which the rigidity of organization has been slightly modified or loosened. As should be expected, a less extreme reference point should produce less extreme differences between two groups who are postulated to repre- sent the positive and negative poles toward this component of group rigidity. An overview of the reported data con- firms the expectation. In a few instances the critical ratios between the two experimental groups on the ten items yielded nothing of statistical significance. These dis- crimination failures were suggested to occur either as the result of some identity between the two groups, or as the result of unanticipated factors within these items which might have yielded to further refinement in the rewording of the attitudinal statement. A number of the items, however, were found to yield substantial differential performances between the two groups, and in the direction that had been predicted. Aggregating the performance of the two experimental groups on the total series of ten items, the mean response for the non-authori- tarian group was found to be 3.68, or at a key value which is near to the middle of the slight disagreement zone. The authoritarian group, on the other hand, exhibited an over- all mean response of h.18 which, in relation to the non- authoritarian group, indicates a greater acceptance of the 89 rigidity attitudes presented to them, but which on the key is represented by a rating that is in the middle or neutral zone. The critical ratio between the two groups on total performance was found to be 5.37 which is significant beyond the one-tenth of l per cent level of confidence. It was concluded, therefore, that the basic hypothesis was in ac- cordance with the data described in the chapter. 90 CHAPTER V ANALYSIS OF ATTITUDES IN GROUP STRUCTURE C Introduction It is the purpose of this chapter to present an analysis of the data gathered under Group Structure C of the questionnaire. The description of Group Structure C was intended to represent a further modification of the rigidity component which most characterized the stereotype of Group Structure A and was lessened only slightly in the stereotype of Group Structure B. Group Structure C was conceived with the idea of presenting what appeared to be a group participation process but yet not actually being a true group process of the interaction type. In the typical group-centered processes of the more democratic construction, leadership arises from the group itself out of its own intra-relationships. In the stereotype of Group Structure C, however, the group plays no particular role of self-organization or even of participation. It is essentially an audience group similar in role to that described for the group in Group Structure B, and not re- markably different from the role described for the group in the more extreme description of Structure A. 91 This structure is of the "Chicago Roundtable" type. It consists of a panel of experts who have been preselected by the conference chairman, and among which certain under- standings, procedures, and points of view are agreed upon as "shall come out in the discussion". It is essentially a synthetic group discussion, rather than pure group dis- cussion. The experts are not the natural leaders from the audience group, and in fact need have no membership with the group, or the group with them. In short, it is a syn- thetic group discussion that is conducted in the presence of, but not a g 2333.;2, the audience group that surrounds it. The structure is viewed as less rigid than that of the two previously treated structures in that, according to the specific stereotype, there is practically no rigidity in the organization of subject matter; each . speaker is an expert at speaking in an impromptu fashion, and with the motive of winning over some faction of the audience to his point of view, or to otherwise stimulate them in some direction or another. In addition, since each individual is to some extent an "authority", there is no orderly "one-way” presentation. In fact, conflict is not only anticipated but often encouraged. Varying with the skill or adroitness of each leader, it is entirely pos- sible for new power lines to develop in accordance with the weakness or strength of position that each is able to command. 92 Below the description of Group C are presented ten statements, each one of which represents an attitude about different aspects of this kind of group structure. The response patterns of the two experimental groups to these ten items are presented in this chapter with the viewpoint of comparing and interpreting the mean differences between the two. As in Chapters III and IV, the hypothesis continues to be directional in that the non-authoritarian group is predicted to always produce a mean score to the left of each mean score produced by the authoritarian group on the same item. Presentation of Data on the Distribution of Responses on Group Structure C Tables X and XI present the actual distribution of weighted score responses to Group Structure C on the part of each experimental sample. All raw score answers were first converted to weighted scores in accordance with the process and key as described in Chapter III. Analysis of the Differences Between the Mean Responses on Group Structure C In order to draw significance concerning the distribu- tion of responses of the two experimental groups as they appear in Tables X and XI, it was first necessary to cal- culate the mean responses of the two experimental groups TABLE X DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE C BY NON-AUTHORITARIANS Item Weighted Score Responses Total 1 2 3 h 5 6 7 Number 1 o o 1 7 u? 30 85 2 0 3 l 5 35 M1 85 3 O 11 20 26 2O 8 85 h 5 17 12 23 19 9 85 5 3 3 12 20 33 9 85 6 32 3O 9 5 3 6 85 7 13 9 18 1h 19 12 85 8 23 3O 17 6 9 O 85 9 1h 2h 13 19 10 5 85 10 O 8 IO 12 27 28 8S TABLE XI 9h DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE C BY AUTHORITARIANS Item Weighted Score Responses fifr 1 2 3 u 5 6 7 Number 1 5 2 5 18 25 3O 85 2 o 2 1 5 17 60 85 3 2 8 6 13 37 19 85 u 12 9 11 25 17 11 85 5 1 3 S 12 32 32 85 6 12 16 19 17 9 12 85 7 9 LL 21 23 17 11 85 8 9 19 20 1h 18 5 85 9 S 10 15 1 15 211 15 85 10 o 1 3 15 19 u? 85 95 with regard to each item. The means presented in Table XII are the arithmetic means calculated in the same manner as that described in the immediately preceding chapters. In addition, the standard deviations are similarly included. Table XII summarizes the means and standard deviations of the two experimental groups on these ten items. TABLE XII MEAN RESPONSES TO GROUP STRUCTURE C BY AUTHORITARIAN AND NON-AUTHORITARIAN SAMPLES Non-authoritarian Group Authoritarian Group Mean Standard Item Mean Standard Response Deviation Response Deviation 6.23 0.7 1 5.57 1.7 6.2h 1.1 2 6.52 0.98 h."6 1.6 3 5.36 1.7 h.32 1.9 h h.32 2.0 5.13 1.5 5 5.86 1.u 2.hO 1.8 6 3.81 2.0 h.l§ 2.1 7 h.h0 1.9 2.56 1.6 8 3.76 1.9 3.82 1.9 9 n.68 1.9 S.h6 1.7 10 6.22 l.h An inspection of the foregoing data indicates that in eight out of the ten pairs of mean scores for the two eXperi- 96 mental groups, the non—authoritarian group obtained a smaller mean score than the authoritarian group; in one instance the non-authoritarian group obtained the same score as the authoritarian group; and in the other case, the mean score of the non-authoritarian group exceeded that of the authoritarian group. Thus the obtained data confirmed rather strongly that the directional prediction on each pair of scores was largely sustained. Item Analysis of Differences Between the Means It is apparent that for the ten items listed under Group Structure C in the questionnaire, there is a differ- ence in the response pattern of the two experimental groups. These differences may be more clearly recognized by comput- ing the critical ratios of the differences between their respective means. Using the same processes as were employed in Chapter III with respect to Group Structure A, Table XIII was constructed. This table presents a summation of these critical ratios, plus an interpretation of their levels of significance. The level of confidence index is based upon the same one-tailed distribution table as was used for treating the parallel data in Chapter III. A An examination of Table XIII indicates that seven ofthe ten indices are statistically significant in indicating that the contrasting performances of the two eXperimental 97 TABLE XIII CRITICAL RATIOS FOR GROUP STRUCTURE C W Item Critical Level of Conclusion Ratio Confidence l -3.32 null No significance 2 1.75 5.0% Significant 3 3.19 0.1% Very significant h 0.00 null No significance 5 3.27 0.1% Very significant 6 h.78 0.1% Very significant 7 0.82 null No significance 8 h.h6 0.1% Very significant 9 h.26 0.1% Very significant 10 3.18 0.1% Very significant Totals 5.52 0.1% Very significant 98 groups were not due to pure chance factors. Six of these indices were significant at, or beyond, the one-tenth of l per cent level of confidence. Over-all, the total per- formance of the two groups on the ten items was signified by a critical ratio of 5.52 so that it was reasonably evi- dent that the majority of these items possessed a dis- criminatory power of substantial aggregate weight. Analysis of Performance on Individual Items As in Chapters III and IV, this chapter treats not only the statistical aspects of the derived data but also the content aspects of the various items under examination. These items specifically are the ten attitudinal statements that refer to the whole, or to some component part, of Group Structure C. As a point of reference, the stereotype of Group Structure C is presented as follows: Group Structure C This is a panel discussion meeting of the Chicago Round Table type in which a group of experts rep- resentative of the various interest segments of the audience will discuss a specified topic. These individuals are quite experienced in the panel dis- cussion technique. They seldom use notes or written materials but rather speak in impromptu fashion on the subject under consideration. Because of the contrasting viewpoints that they represent, the dis- cussion often becomes quite vigorous, and they raise pointed questions among themselves. Since the questions raised by the panel experts are assumed to be similar to many of those that would come from the audience, the majority of the period is given over to the experts on the panel. However, 99 there is usually some time allotted at the end of the discussion for a few pertinent questions to come from the floor which the experts may have overlooked. You have not yet attended a panel meeting composed of these particular in- dividuals. The first of the ten items appearing below the stereo- type is as follows: 1. This is the kind of a meeting I enjoy attending. This item is noted by being identical to the first items for Group Structures A and B. It was prepared with the same general purpose of presenting an over-all state- ment that would stimulate a generalized negative or posi- tive endorsement for the whole group structure in question. By construction, this stereotype was presumed to be a modification of Group Structures A and B. In it the reins of authority still remain in the hands of the speakers, and the audience group continues in check; however, the general atmosphere of the structure is one of decreased rigidity, with the first emergence of some kind of participation. Since the stereotype did not illustrate a true group- centered, participation process, the hypothesis was continued that the non-authoritarian group should generally manifest less preference for the operational aspects of this type of group process than would the authoritarian sample. If the evidence produced in response to this first item were all that were obtainable, the hypothesis would have to be lOO rejected at this point as untenable since both experimental samples reacted in concert and in favor toward this simple statement. Both agreed that they liked this type of group structure, the non-authoritarian group even slightly more so than the authoritarian. However, considerably more evidence was available for study. In addition to item 1 most of these latter items produced sufficient evidence that rejection of the hypothesis would be premature. Item two is noted to appear as follows: 2. Conference purposes are usually best served when the panel has been carefully pre-selected from established experts in the field rather than when it is made up at the time of the meeting from.untested members in the audience. This item addresses itself to a single aspect of the group process in question, the method of the selection of the speakers. The item was constructed to elicit atti— tudinal responses toward the process of the arbitrary selection of leaders by a remote authority as opposed to the democratic selection of leaders by the immediate group itself. Inspection of the item reveals that the attitude is stated in a positive tone as though this were a sensible process. Whether this positive flavor was a suggestible influence or not, both groups responded to the item with a reaction of agreement, but on somewhat different levels; the non-authoritarian group, with its mean score of 6.2h, 101 exhibited low moderate agreement, whereas the authoritarian group, with its mean score of 6.52, exhibited high moderate agreement to the statement. The critical ratio between the two of 1.75 is regarded as significant at the 5 per cent level of confidence. Though of some significance, the per- formance of the two groups on this item will not appear as meaningful as on most of the items that follow. Item three appears on the questionnaire as follows: 3. The average professional audience can gain more from listening to a discussion by a pre-selected panel of experts from the field than they can by discussing the sub- ject among themselves. In the nature of authoritarianism, considerable em- phasis is placed on confidence in the leader with a cor- responding lack of confidence in the group that provides him with followership. The leader's role is one of authority in which his position has arisen out of the proof of his right to eminence. Thus, by definition, the leaders are the primary repositors of authority and the acknowledged tutors for the group. For the less informed to attempt to teach each other is viewed as in the same logic as the concept of the blind leading the blind. Im- plicit in this rigid leader-follower learning process is a philosophical conflict with the learning processes described by Dewey in which the democratic interactions of group- sharing activities are regarded as accelerating rather than inhibiting influences to the learning process. 102 This item was constructed with the anticipation that it would suggest a greater degree of willingness on the part of authoritarians to delegate professional discussion processes to a panel of experts rather than to the group at large. In observing the reaponses of the two experi- mental groups to the item in question, the authoritarian group behaved as expected with a definite agreement level at a point between slight agreement and moderate agreement. Their mean response score was 5.36. The non-authoritarian group, however, manifested a general attitude of neutrality toward the item, as evidenced by a score of h.56. That these differences in scores were significant is further borne out by the critical ratio between them of 3.19 which is significant at the one-tenth of l per cent level of con- fidence for a one-tailed distribution. Item four appears as follows: h. The general discussion should primarily be carried on by the audience group, using the panel experts only as resources for specialized information. This item is observed to be substantially the same as item 3, but stated in a reverse direction. In item con- struction, the alternatives of the positive or negative phrasing of an item are reputed to exercise variable de- grees of influence over the discrimination power of the item in question. In order to insure that the particular group dynmnic in question was to receive maximum focus from the 103 two experimental groups, item A was constructed as an in- version of item 3. The results of the testing of the two groups on item h failed to yield anything of statistical significance. Both groups responded with the identical mean score of h.32. An examination of the standard devia- tions for the paired scores on item h as compared with the standard deviations for item 3 reveals a definite increase in that index. One interpretation of this change which would seem to harmonize with the rest of the data is that item 8 may present a somewhat ambiguous concept. Item five is as follows: 5. It is a better panel discussion when the group obtains its information by questioning and discussing with the ex- perts on the panel, rather than by ques- tioning and discussing with each other. Item 5 attempts to describe a compromise or half-way position between a panel discussion process that ignores the group and a group discussion process that ignores the panel. Non-authoritarians would be presumed to welcome those aspects of the process which permit any increase in group participation, however structured the conditions of participation may be. Authoritarians would be presumed to welcome those aspects of the process which preserve the hierarchies of leader-follower lines of authorities, with the role of the experts being essentially that of arbiters of discussion points rather than of mutually equalitarian 10h discussants. Since the role of the panel expert to each group member is more enhanced than is the role of each group member to each other group member, the total attrac- tion of the item was believed to be weighted in the authori- tarian direction. In examining the performance of the two groups, both are noted to exhibit positive preference for the statement, though at different levels of agreement. The non-authori- tarian group, with its mean score of 5.13,placed itself at a scale point at the lower end of the "slight agreement" zone. The authoritarian group, however, obtained a mean score of 5.86 which located it at a point at the upper end of the "slight agreement" zone, very near to the level of moderate agreement. The critical ratio between the two of 3.27 is noted to be significant at the one-tenth of l per cent level of confidence. Item six is as follows: 6. In a good panel discussion, the ex- perts should attempt to resolve their differences and come up with a single point of view. As Frankel-Brunswik and her associates indicate in their theoretical discussions on the nature of the authori- tarian personality (1), such a personality dislikes ambiguity or complications in any operational standard. convention, or idea. They find no security in multi-basis explanations but rather seek the touchstone of a single causation. The 105 technique of the demagogue is illustrative of this prin- ciple when he attributes every aspect of social discontent to some single influence. The same general process is in- herent in those personalities who arbitrarily sweep aside all mitigating or conflicting evidences in favor of an "all black" or "all white" opinion or conclusion. Dogmatic points of view are disturbed by complicating or extenuating factors that undermine the security of a definite position, or becloud a clear and definite answer to a stated problem. Item 6 was prepared in order to touch upon some aspect of this manifestation such as would be found in the respec- tive viewpoints of each experimental group. This somewhat sweepingly stated item yielded negative levels of response from each of the samples in question, but at significantly different levels. The non-authoritarian sample, with its mean score of 2.hO, placed itself rather definitely in the disagreement zone of the scale. The authoritarian group, with its mean score of 3.81, located itself high up on the "slight disagreement" end of the scale, very close to the level of neutrality. The critical ratio between the two was found to be h.78, or well beyond the one-tenth of l per cent level of confidence. Item seven is as follows: 7. Top authorities in the field are professionally obligated to serve on panels of this type. 106 In the stereotype of the conventional personality, "duty" is idealized as a proper component. The authori- tarian personality professes rigid adherence to conventional values, and as such perceives conformity, duty, and obedi- ence as indispensable to the social discipline which must be preserved. Item 7 implies some limitation of free choice on the part of the individual and stresses the duty and obligation values that must govern one's personal behavior. To the degree that this principle is perceived equally by the two experimental samples, it would be expected that variability in their levels of acceptance would be a function of varia- bility in their personality organization. At any rate, the evidence indicates that the non-authoritarian group ex- hibited a rather neutral response to this item with a mean score of h.lS, and the authoritarian group also elicited a score in the neutral zone at a point a little further up the scale in the direction of "slight agreement", but not to a sufficient degree to be of statistical significance. The mean score of the authoritarian group was h.h0; the critical ratio was .82. Apparently this item was not clearly perceived in reference to the principle it was trying to illustrate, or that the two groups were similar in organization on this point. Some rise in the standard deviations suggests possible ambiguity in the statement. 107 Item eight is as follows: 8. A sound discussion will try to stick pretty much to the middle of the road and to avoid extremes. Discussions that are "left wing", extreme, or purely speculative, regardless of how provocative or stimulating they may be, are presumed to have less attraction for authoritarian personalities than are the safe, secure, conventional, middle-of—the-road discussions. This is predicated on that part of the theoretical basis of the authoritarian personality which ascribes to it an affinity for conventionalism with a rigid adherence to conventional, middle-class values. Item 8 was stated as an agreement with this authoritarian concept. In an examination of the evidence it was noted that the non-authoritarian sample rather frankly rejected the state- ment with a mean score of 2.56. The members of the authori- tarian group, however, were not so definite in their handling of the itan. Their score of 3.76 was sufficiently negative to place them just within the disagreement zone, but not greatly removed from the neutrality point. The critical ratio between the two groups of h.h6 was strong statistical evidence of some real differences between the two groups with respect to this item. 108 The ninth item in the series is: 9. Generally speaking, the panel should be composed of the respected leaders of the profession rather than of the youngest newcomers. In the traditional rules of a rugged-individualist society, leadership and recognition come about as the natural process and reward of toil, effort, struggle, and proof of fitness. Any disarrangement or suspension of the rules that proposes to capriciously give rec0gnition or roles of leadership to the untried, the untested, or the "tenderfoot", may be regarded by authoritarians as a threat to the rest of the group who have been playing the "rules of the game" and who now stand to lose out on their invest- ment. The item was phrased to elicit agreement from in- dividuals particularly high in the authoritarian complex. In confirmation of the expectation, the authoritarian sample performed at a significantly different level than did the non-authoritarians. The latter group exhibited a rejection toward the item with an average response of 3.h2, or about in the middle of the"slight disagreemenfi'zone. The authoritarian group, however, obtained a mean score of h.68, or well up the neutral zone in the direction of positive agreement. The critical ratio between the two of h.26 indicates a level of confidence that is significant beyond the one-tenth of l per cent. 109 The last item for Group Structure C is as follows: 10. A panel made up of experts who practice at their profession will generally produce a sounder discussion than if the panel is made up of theorists or academically minded profes- sionals. Frenkel-Brunswik discusses an aspect of authoritarianism that she calls anti-intraception (1:23h-235)- Defining intraception first as relating to the feelings and empathy for people, the quality of anti-intraception becomes one in which people are viewed merely as physical objects. The anti-intraceptive personality is primarily interested in practical pursuits, concrete facts, objective measurements, et esters, and is inclined to devalue human personalities who cannot produce these substances. He represents the personality who, in the athletic world, hires and fires coaches on the basis of whether games are won or lost, and not on whether the coach is teaching clever theory or get- ting the most out of material that is otherwise too inade- quate to meet the competition. In the professional world there necessarily exists an operational cleavage in that some pursue the work of building new theory and some pursue the work of prac- ticing that which they already have. Though the theorists and the practitioners are operationally indispensable to each other, group attitudes toward each are not necessarily synonomous. The anti-intraceptive quality that is implicit in the authoritarian personality should tend to enhance the 110 role or status of the practitioner whenever he is being compared with that of the theorist. Thus, for the item in question, authoritarians were anticipated to particularly prefer assembling a panel of experts from the field of con- crete practice rather than to select a group of theorists who would have a more academic orientation. Item 10 was phrased in the positive direction of agree- ing that practitioners are more desirable on a panel than theorists, and the item was responded to by two samples both of whom were professional practitioners. Thus, though some agreement was anticipated from each of the responding groups, it was surmised that the levels of agreement would be differentiated in accordance with the underlying prin- ciple on which they were selected. This prediction was confirmed by a mean response of 6.22 from the authoritarian sample which indicated fairly strong agreement with the presented attitude. The non-authoritarian sample also elicited some measure of agreement with a mean score of S.h6, which, on the scale, was substantially more moderate in tone. The critical ratio between the two was 3.18, which is statistically significant at the one-tenth of l per cent level of confidence. 111 Summary The stereotype of Group Structure C is concerned with a synthetic group-centered process which imitates group discussion by conducting a proxy activity in behalf of, but not with, the actual group at large. The stereotype was intended to portray a further reduction in the lines of authority and group rigidity than in the previous stereotypes. However, within the limits of operational construction, the stereotype was meant to illustrate a predominantly authoritarian structure in contrast to the predominantly non-authoritarian structures which are to be described in the chapters that follow. This chapter has contained a presentation of the statistical data derived from the contrasting performances of the two experimental samples to different aspects of the structure. In eight of the ten items that comprised the test series, the non-authoritarian sample presented mean response scores that were consistently more negative toward rigidity in group structure than did the authori- tarian sample, although in one instance the difference was not regarded as statistically significant. Aggregating the performance of the two experimental groups on the total series of ten items, the mean response for the non-authori- tarian group was found to be h.h5, or at a key value in the middle of the neutral zone. The authoritarian group, on the 112 other hand, exhibited an over-all mean response of 5.05, or a key value of slight agreement with the items in question. The critical ratio between the two groups on total performance was found to be 5.52 which is statisti- cally significant beyond the one-tenth of l per cent level of confidence. The evidence suggested, therefore, that the basic hypothesis was operationally sound with regard to the situation in which it was tested in this chapter. 113 CHAPTER VI ANALYSIS OF ATTITUDES IN GROUP STRUCTURE D Introduction It is the purpose of this chapter to present an analysis of the data gathered under Group Structure D of the questionnaire. The stereotype of Group Structure D was intended to represent a further modification in the rigidity organization of a group structure. Where in Group Structures A, B, and C lines of authority were rather tightly drawn and centered in the hands of the leader or leaders, Group Structure D marks a relinquish- ing of that authority to the group itself under the guid- ance, but not the command, of a more democratically oriented leader whose role is sharply circumscribed. Group Structure D is intended to represent essentially a group-centered discussion process in which the role of the leader is largely that of a guide to help the group organize itself. He attempts to stimulate the use of democratic processes during the participation activity as well as when the group begins to evolve its own natural leaders. He also endeavors to see that group purposes are met and that everyone has a chance to contribute. He does 11h not arbitrate discussions or make authoritative pronounce- ments; in fact, he need not make any contribution to the purely professional information-sharing activity at all. His eXpertness is defined as appropriate to the field of group processes rather than to the subject matter that com- prises the basis of discussion. The atmosphere of Group Structure D is essentially democratic and pliable. Lines of authority are not pre- planned, and no strong "power figures" are placed in a position of dominance over the group. The program planning committee may have outlined a basic agenda which it desires the group to consider, but it is necessarily unable to in- struct the group in the decisions or recommendations it must make, or even to possess foreknowledge of what they may choose to say in their discussions. Obviously a structure as plastic as this leads to a certain amount of confusion as individuals begin to explore and test themselves with each other, or to try to acquire concerted agreement or action on the part of other group members, each of whom may possess slightly different view- points on a given matter. Still the structure is not totally without order or structure since it does possess a preselected leader whose role it is to guide them into self-organization, and around this individual, whether he seeks it or not, the mantle of authority still reposes, if only in a negligible degree. 115 Below the description of Group Structure D are pre- sented ten statements, each one of which represents an attitude about different aspects of this kind of group structure. The response patterns of the two experimental groups to these ten items will be presented in this chapter with the viewpoint of comparing and interpreting the mean differences between the two. As in the previous chapters the hypothesis continues to be directional in that the non-authoritarian group is predicted to always produce a mean score to the left of each mean score produced by the authoritarian group on the same item. Presentation of Data on the Distribution of Responses on Group Structure D Tables XIV and XV present the actual distribution of weighted score responses to Group Structure D on the part of each experimental sample. All raw score answers were first converted to weighted scores in accordance with the process and key as described in Chapter III. Analysis of the Differences Between the Mean Responses on Group Structure D In order to draw significance concerning the distribu- tion of responses of the two experimental groups as they appear in Tables XIV and XV, it was first necessary to cal- culate the mean responses of the two experimental groups TABLE XIV DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE D BY NON-AUTHORITARIANS ll6 4 -; Item Weighted Score Responses Total 1 2 3 u 5 6 7 Number 1 31 37 18 1 85 2 20 39 20 h 2 O 85 3 10 2o 20 22 10 3 85 h l2 19 20 21 IO 3 . 85 5 l IO 19 26 20 9 85 6 8 25 3o 15 3 u 85 7 29 3O 20 5 1 O 85 8 13 18 18 19 10 85 9 5 16 21 1 18 19 9 85 10 20 2h 16 1h 6 5 85 DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE D BY AUTHORITARIANS TABLE XV 117 Weighted Score Responses 4‘ fl Item Total 1 2 3 h 5 6 7 Number 1 19 19 28 5 8 6 85 2 13 11; 3o 13 9 6 85 3 3 17 25 12 16 12 85 u 2 8 7 26 15 27 85 5 8 8 23 28 17 85 6 16 16 17 18 14 85 7 13 l7 30 1 13 8 3 85 8 7 1a 21 15 16 12 85 9 1 11 1 23 21 21 85 1o 3 u 16 12 2h 26 85 118 with regard to each item. The means presented in Table XVI are the arithmetic means calculated in the same manner as was described in the preceding chapters. In addition, the standard deviations are similarly included. Table XVI sum- marizes the means and standard deviations of the two experi- mental populations on these ten items. TABLE XVI MEAN RESPONSES TO GROUP STRUCTURE D BY AUTHORITARIAN AND NON-AUTHORITARIAN SAMPLES Non-authoritarian Group Authoritarian Group Mean Standard Item Mean Standard Response Deviation Response Deviation 1.92 0.96 l 3.01 1.8 2.23 1.1 2 3.h3 1.8 3.5M 1.8 3 h.lh 1.9 3.118 1.8 u 5.27 1.7 h.60 1.6 5 h.97 1.8 3.16 1.6 6 h.36 2.0 2.12 1.1 7 3.23 1.7 h.25 1.9 8 h.15 2.0 11.13 1.9 9 5.1.7 1.6 3.02 1.9 10 5.23 1.8 An inspection of the foregoing data indicates that in all but one of the ten pairs of mean scores for the two 119 experimental groups, the non-authoritarians obtained a smaller mean score than did the authoritarians. With refer- ence to the single exception, item eight, both groups ob- tained, within a tenth of a point, the same score. Thus the data confirmed rather strongly that the directional prediction on each pair of scores was uniformly sustained with the exception of one minor deviation. As a point of reference it will be recalled that the smaller weighted scores signify preferences in the direction of plasticity in group structure in contrast to the larger weighted scores which signify preferences in the direction of rigidity in group structure. Item Analysis of Differences Between the Means It is apparent that for the ten items listed under Group Structure D in the questionnaire, there is a differ- ence in the response pattern of the two experimental groups. These differences may be more clearly recognized by comput- ing the critical ratios of the differences between their respective means. Using the same processes as were employed in Chapter III with respect to Group Structure A, Table XVII was constructed which presents a summation of these critical ratios plus an interpretation of their levels of significance. The level of confidence index is based upon the same one- tailed distribution table as was used for treating the parallel data in Chapter III. 120 TABLE XVII CRITICAL RATIOS FOR GROUP STRUCTURE D Item Critical Level of Conclusion Ratio Confidence 1 n.8u 0.1% Very significant 2 5.17 0.1% Very significant 3 2.1a 2.0% Significant h 6.68 0.1% Very significant 5 1.u0 10.0% Insignificant 6 h.35 0.1% Very significant 7 5.02 0.1% Very significant 8 -O.33 null No significance 9 3.82 0.1% Very significant 10 7.8h 0.1% Very significant Totals 8.50 0.1% Very significant ========= :=_ An examination of the foregoing table indicates that eight of the ten indices are statistically significant in indicating that the contrasting performances of the two experimental groups were not due to pure chance factors. Seven of these indices were significant beyond the one- tenth of 1 per cent level of confidence. The other was significant at the 2 per cent level of confidence. Over-all, the total performance of the two groups on the ten items was signified by a critical ratio of 8.50, so that it was 121 fairly evident that the majority of these items possessed a discrimination power of substantial aggregate weight. Analysis of Performance on Individual Items As in the immediately preceding chapters it is the purpose of this chapter to treat not only the statistical aspects of the derived data but also the content aspects of the various items under examination. These items specifically are the ten attitudinal statements that refer to the whole, or to some component part, of Group Structure D. As a point of reference the stereotype of Group Structure D is set out below: Group Structure D The discussion leader for this group meeting has a good reputation for obtaining participation and discussion on the part of every member of the audience. The discussion leader has no important professional knowledge to contribute to your pro- fessional field of interest, but he does have professional knowledge of techniques by which he can draw opinions and contributions from other members of the group, as well as to draw out from you some statements of your own viewpoints on vari- ous professional t0pics. Frequently, the leader breaks up the audience into small discussion groups; at other times he employs round-table techniques or other devices. You are interested in the topical agenda of the meeting; you also know that if you at- tend you will be involved actively in group participa- tion. You have not attended a meeting with this dis- cussion leader before. The first of the ten items appearing below the stereo- type is as follows: 122 1. This is the kind of meeting I enjoy attending. This item is noted to be identical to the first items for the immediately preceding group structures. It was prepared with the same general purpose of presenting an over-all statement that would stimulate a generalized negative or positive endorsement for the whole group struc- ture in question. By construction this stereotype was pre- sumed to represent a decrease in the lines of rigidity which characterized the stereotypes of Structures A, B, and C. It is essentially a group-centered process rather than a leader-centered process though it does preserve the role of the leader, albeit a sharply circumscribed one. Since the stereotype marks a departure from preceding structures in that authority is transferred more or less to the group and rigidity lines are largely broken, the basic hypothesis would be inverted at this point in the direction that the non~authoritarian personalities would tend to be attracted to the structure in contrast to the authoritarian personalities who would be presumed to have a lesser relative preference for the structure. The evidence revealed that the non-authoritarian group obtained an anti-rigidity score of 1.92 on the item, which (when properly inverted) corresponded to a key value in the substantial agreement zone. The non-authoritarian group, however, indicated a reduced level of positive preference 123 for this type of structure with a mean score of 3.01 which inverts on the scale to correspond with the "slight agree-' ment” point. The critical ratio between the two was found to be h.8h, or statistically significant at a point beyond the one-tenth of l per cent level of confidence. Item two in the series is as follows: 2. Generally speaking, it is more profitable to sit back and learn something than to try to get into the discussion yourself. As previously noted in the discussion on the authori- tarian personality, one aspect is the emphasis on passivity, the willingness to accept a subordinate status in the pres- ence of a "higher authority", and to be told what to do, think, or learn rather than to talk through to a shared decision. This item attempted to portray this ego-reducing dynamic by suggesting endorsement of such behavior. By construction, the item is keyed in the direction that authoritarians were anticipated to exhibit higher agreement with the attitude than the non-authoritarians. A review of the results indicated that the non- authoritarian sample was substantially negative toward this statement as witnessed by their low mean score of 2.23. The authoritarian group, however, was not nearly so negative, as evidenced by a mean score of 3.h3 which located them on the scale at a point midway between slight disagreement and 12h neutrality. The critical ratio between the two was a very significant 5.17. The third item in the series is as follows: 3. Statements or opinions from the least experienced members of the group are just as worthwhile as statements from the most experienced. This item was constructed with the intent to challenge that tradition-rooted platitude which bases right to leader- ship and authority upon the "seniority" principle of years of experience. The authoritarian personality is theorized as one which seeks to preserve the rigid hierarchy of rules and taboos which attaches special dignity and value to the meticulous measurement of time itself, and in which, if the novitiate were to be accorded quick and equal recog- nition, would jeopardize the entire investment that each member of the group had already devoted to the system. Because it challenges the "rigidity" principle, the item was stated in an inverse direction with the expecta- tion that it would be more attractive to the non-authori- tarians than the authoritarians. The evidence indicated, characteristically enough, that the authoritarian sample was not too much in agreement with the statement. Their mean score of h.lh, when properly inverted on the scale, located them at a point of slight disagreement with the item, although inclining in the direction of being neutral. 125 The non-authoritarian group, however, with its mean score of 3.5h, presented a somewhat more positive reference to the item. When appropriately inverted on the scale, they are found to be at a point midway between neutrality and agreement, or a significant relative difference in the predicted direction. The critical ratio between the two was 2.1h, which is significant at the 2 per cent level of confidence. Item four appears as follows: h. One important drawback to a general group discussion is that confusion is in- creased, time tables become upset, and not taamuch is accomplished. This item was constructed with the intent of presenting in the form of a statement some of the more frequent criti- cisms which are addressed to the group-process technique. The basic criticism.seems to devolve around disturbances that impede the orderly processing of the agenda, subject matter or business of the day. So much emphasis on group interaction and participation is viewed as an unfortunate sacrifice in efficiency. The item was written in the direction of encouraging agreement that this kind of an apparent result in a group-centered activity is a bad thing. The authoritarian group demonstrated considerable will- ingness to accept the item.as witnessed by their rather h1g3 mean score of 5.27. The non-authoritarian group, however, manifested the contrary point of view. Their mean score of 126 3.h8, which placed them on the disagreement side of the scale, marked a rather clear cut difference in performance. The high critical ratio of 6.68 indicated a level of con- fidence beyond that of the one-tenth of l per cent. Item five is as follows: 5. It is better for a discussion leader to see that everyone participates rather than to concern himself with what is being said. This item was constructed with the view of focusing attention on a special aspect of the discussion leader's role. As elsewhere mentioned in this chapter, the quali— fications of good discussion leadership were a function of the ability to elicit participation rather than to impart authoritative information on the subject matter at hand. In the construction of this item it was anticipated that the authoritarian personalities would tend to reject this definition of the discussion leader's role and incline toward furthering his concern in the actual subject matter being discussed. An examination of the results indicated that the authori- tarian group did indeed reject this statement, as evidenced by their mean score of h.97, which, for an invertedly phrased item, located them on the scale at the slight dis- agreement point. The non-authoritarian group, however, made a somewhat similar score though their position was found to be at a midpoint between slight disagreement and 127 neutral. Their mean score, before inversion, was h.60. Thus, though the authoritarian group manifested slightly more negative attitude toward the discussion leader's emphasis on participation rather than content, the differ- ences in performances on this item cannot be regarded as significant. The critical ratio between the two is noted to be only l.h0, or at the 10 per cent level of confidence. Item six is noted to be: 6. Individuals who are not professionally qualified to speak on the subject under discussion should generally remain quiet. Item 6 was designed to tap the latent hostility that has been described as characteristic of the authoritarian personality, particularly in his attitudes towards those of a lesser "caste" than the elite. The release of hos- tility is presumed to be ethnocentrically directed by authoritarian personalities in a spiral fashion, with each such individual seeking identification with the "higher status" members of the group and a minimum of identification with those of "lower status". In the competition for status, individuals possessing the least status may engender hos- tility by intruding on the statusrights of superiors. The item in question presents an attitudinal statement that lends encouragement to the preservation of rigid status lines, and, as a natural effect, would inhibit the demo- cratic participation of all members of the group on an equal 128 basis. As such, it was anticipated that the authoritarian sample would find the statement less objectionable than the non-authoritarian sample. In reviewing the data it was found that the non- authoritarian group rejected the item, as demonstrated by a mean score of 3.16 which, on the scale, located them well down in the slight disagreement zone. The authori- tarian sample failed to reject the statement but instead exhibited a neutral attitude toward the item, trending somewhat in the direction of slight positive agreement. Their mean score was h.36. The critical ratio between the two was h.35, or very significant at the one-tenth of l per cent level of confidence. Item seven is noted to be: 7. One drawback to a group discussion meeting of this type is that you know the leader is going to call upon you to do or say something. The case histories discussed by Sanford (1:787-816) focus attention on that aspect of authoritarianism which may be more aptly described as the "lone wolf" behavior. These are individuals who surround themselves with a psychological shell, who cannot be reached easily by others, nor, in turn, can they reach out and enter into the main stream of other peoples lives. Their reference to people is primarily one of dominance or of submission. As a protective mechanism they may occasionally prefer those 129 situations which provide them with anonymity and a surcease from the struggle of competitive comparisons with their fellow man. In the construction of the item, it was implied that the inability to be a non-participant is an unfortunate aspect of group-centered processes. Agreement with the item was conceived to represent a preference for a return to a more rigid structuring of the group process; disagreement with the item was conceived to signify a preference for the more plastic and free-forming group processes which expose each member to participation. An inspection of the performance of the two experimental samples revealed that the non-authoritarians rather defi- nitely rejected the item_with a mean score index of 2.12. The authoritarian group, however, was relatively much less rejective toward the item in question. Their mean score was 3.23, and the critical ratio of the differences in per- formances between the two groups was found to be 5.02, or very significant at the one-tenth of 1 per cent level of confidence. Item eight is noted to be: 8. It is professionally sound for the group leader to absorb a significant portion of the conference time in seeing that each member becomes socially ac- quainted with the other members present. In Sanford's discussion of the genetic aspects of the authoritarian personality (1:787-816), further illustrations of the anti-intraceptive component are brought out. Basically 130 this is an attitude in which other individuals are regarded in a depersonalized light. Social processes which cater to human needs for recognition, response, et cetera, are re- garded by authoritarians as of less "real" importance than the strictly businesslike procedures which deal directly and "efficiently" with the concrete matters at hand. In a sense the socializing processes are viewed more nearly as delaying actions which simply hold up the real purposes of the meeting. Item 8 was constructed to reflect negative endorsement toward this type of anti-intraceptiveness, and as such, was expected to evoke statements of agreement from the non-authoritarians. This, the item failed to do. Both groups yielded an entirely neutral response to the statement with mean scores of h.25 and h.15 being obtained from the non-authoritarian and authoritarian groups respectively. A restudy of the item suggests that a possible weakness or ambiguity in the statement might reside in that part of it which reads "a significant portion ... in seeing that each group member becomes socially acquainted ...", wherein the interpretation of "significant portion" may have been per- ceived at various quantitative levels by the various re- spondents. Item nine is noted to be: 9. How the group discusses and produces factual data is perhaps more important than what the data is that they produce. 131 Item 9 concerns itself with the controversy between end results versus the means used to achieve them. In a sense the item has attempted to focus a choice between the value placed on the extent to which factual truths were developed, and the value placed upon the manner in which in- dividual personalities were reSpected during the fact devel- opment process. In line With concepts that have elsewhere been related to authoritarianism, the authoritarian per- sonality was believed to be more prone to approve the collection or establishment of hard and solid facts by direct means than to obscure the establishment of truth by muddling through a maze of democratic processes. Since the item was written in an inverted direction, the mean scores had to be inverted in order to obtain the correct scale values of the responses. The authoritarian group with its obtained score of 5.17 (inverted score, 2.83) located itself definitely in the disagreement zone, as predicted. The non- authoritarian group with a mean score of h.l3 (inverted score, 3.87) located itself just outside the zone of neu- trality in the area of very slight disagreement, thus main- taining a relative position to the authoritarian group in accordance with the prediction. The critical ratio between the two was 3.82, or very significant beyond the one-tenth of l per cent level of confidence. 132 The last item in the series is as follows: 10. If the leader for this discussion group had considerable professional knowledge of his own to contribute to the group thinking, it would save much useless discussion. Item 10 was intended to place in question one of the essential bases on which equalitarian discussion rests. The item implies that discussion as such is not particularly useful, and that it would be more constructive to have a group leader who is thought to be capable of supplying the group with many of its answers. The content of the item seems to suggest a movement away from a general group- participation process toward more of an orientation meeting in which facts are released by a hierarchy of the most pro- fessionally qualified, with the group leader occupying the most ascendant position. The manner in which the two ex- perimental groups reacted to the item proved to be most significant. The non-authoritarian group demonstrated a rejecting attitude toward the item with a disagreement score of 3.02. The authoritarian group, however, demon- strated their priority of interest in the direction of the more leader-centered type of structure with a positive in- dex of agreement at 5.23. The critical ratio of the differ- ences in performances was exceedingly significant as wit- nessed by an index of 7.84. 133 Summary The stereotype of Group Structure D marked a departure from leader-dominated group organizations to a structure that was essentially group-centered. The stereotype in- cluded the continuation of the role of a predetermined group leader, but sharply limited that role to a consultant figure rather than an authority figure. The chapter was concerned with a study of the differential attitudes mani- fested by the two experimental groups toward a series of ten statements that pertained to different aspects of this type of a group-centered process. The design of the stereo- type suggested that non-authoritarians should respond to the series with rejection attitudes toward items which en- couraged a return to the rigidity features of previous stereotypes, and the authoritarians with permissiveness toward these types of items. The data, when analyzed, con- firmed the expectation. In seven of the ten items the dif- ferential performances of the two groups were found to pos- sess critical ratios beyond the one-tenth of l per cent level of confidence. Of the remaining three, one was found to be significant at the 2 per cent level of con- fidence, one at the 10 per cent level, and only one in which no differential significance could be attached. Aggregating the performance of the two experimental groups on the total series of ten items, the mean response 131+ for the non-authoritarian group was found to be 3.2h, which signified negative preferences for rigidity features pre- sented in the series at a level definitely in the disagree- ment zone. The authoritarian group demonstrated a lack of rejection toward these rigidity features with a completely neutral mean-response of h.28. The critical ratio between the two groups on total performance was found to be 8.50, or very significant beyond the one-tenth of l per cent level of confidence. Thus, the data set forth in the chapter sup- ported the basic hypothesis that authoritarians would respond toward a group-centered process with attitudes consistently more negative than those demonstrated by the non-authori- tarians. 135 CHAPTER VII ANALYSIS OF ATTITUDES IN GROUP STRUCTURE E Introduction It is the purpose of this chapter to present an analysis of the data gathered under Group Structure E of the questionnaire. This group structure was constructed with the intent to illustrate a further reduction in lines of group organization, authority, and rigidity. As in the stereotype of Group Structure D, the presentation in this chapter is concerned primarily with a group-centered proc- ess, but one in which the responsibility and independence of the group has been increased to the point where it has no pre-appointed group leader of any kind to give it direc- tion. Essentially the group is a "leaderless" group, and the process that it illustrates is the "leaderless dis- cussion technique". In the stereotype that describes the group situation under consideration, almost all evidences of group "structure", as such, have been removed. A group is formed by the self-initiated process of each member to assemble at a given room at a given time, and to discuss among themselves a given special interest topic. Who comes to the meeting, how they organize themselves, what they 136 discuss, and how they discuss it is left entirely up to the individuals themselves. No conference chairman or official makes an appearance to see that "they get organized", et cetera. The only traces of organization or structure that remain are that the conference chairman has observed that certain special interests are still in need of discussion; a room and a time has been set aside for those who are interested; and an invitation to carry on informal group discussions has been issued. Below the description of Group E are presented ten statements, each one of which represents an attitude about different aspects of this kind of group structure. The response patterns of the two experimental groups to these ten items will be presented in this chapter with the vieWpoint of comparing and interpreting the mean dif- ferences between the two. As in the preceding chapters, the hypothesis continues to be directional in that the non-authoritarian group is predicted to always produce a mean score to the left of each mean score produced by the authoritarian group on a given item. Presentation of Data on the Distribution of Responses on Group Structure E Tables XVIII and XIX present the actual distribution of weighted score responses to Group Structure E on the part of each experimental sample. All raw score answers TABLE XVIII DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE E BY NON-AUTHORITARIANS '1 137 Item Weighted Score Responses Total 1 2 3 h 5 6 7 Number 1 15 18 18 ll 16 7 85 2 29 32 15 3 5 l 85 3 u 17 19 27 9 9 85 u 26 33 16 6 1 3 85 5 3 1h 13 23 23 9 85 6 1h 29 22 12 7 l 85 7 17 28 17 1 1h 7 1 85 8 16 31 23 11 h 0 85 9 6 7 21 18 23 IO 85 10 9 27 13 18 ll 7 85 DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE E BY AUTHORITARIANS r TABLE XIX 138 Item Weighted Score Responses Total 1 2 3 11 5 6 7 Number 1 8 11 12 16 18 19 85 2 2o 29 17 1 8 8 2 85 3 2 u 13 1 11 27 27 85 u 11 27 17 1 8 11 10 85 S 3 1 2 1 19 26 33 85 6 7 10 15 18 18 17 85 7 u 7 18 1 2h 15 16 85 8 1 10 26 2 13 17 16 85 9 6 8 8 l 25 25 12 85 10 S 7 10 1 11 23 25 a; 139 were first converted to weighted scores in accordance with the process and key as described in Chapter III. Analysis of the Differences Between the Mean Responses on Group Structure E In order to draw significance about the distribution of responses of the two experimental groups as they appear in Tables XVIII and XIX, it was first necessary to calculate the mean responses of the two experimental groups with re- gard to each item. The means presented in Table XX are the arithmetic means calculated in the same manner as was de- scribed in the preceding chapters. In addition, the stand- ard deviations are similarly included. Table XX summarizes the means and standard deviations of the two experimental populations on these ten items. An inspection of the data in Table XX indicates that in all ten pairs of the mean scores for the two experimental groups, the non-authoritarians obtained a smaller mean score than did the authoritarians. Thus the data rather conclu- sively confirmed that the directional prediction on each pair of items was sustained throughout the series. As a reference point for interpretation, it will be recalled that the smaller weighted scores signify anti-rigidity preferences in group structure in contrast to higher weighted scores which signify pro-rigidity preferences in group structure. 110 TABLE XX MEAN RESPONSES TO GROUP STRUCTURE E BY AUTHORITARIAN AND NON-AUTHORITARIAN SAMPLES M— ‘ Non-authoritarian Group Authoritarian Group Mean Standard Item Mean Standard Response Deviation Response Deviation 3.59 2.0 1 4.59 2.1 2.23 l.h 2 2.76 1.7 h.08 1.8 3 5.h0 1.7 2.32 1.3 h 3.h8 2.0 1.58 1.8 S 5.88 1.8 2.90 1.6 6 h.57 2.0 2.90 1.7 7 h.68 1.8 2.66 l.h 8 h.5h 1.8 h.h8 1.8 9 h.81 1.8 H O 3.61 1.9 5.13 1.9 111 Item Analysis of Differences Between the Means It is apparent that for the ten items listed under Group Structure E in the questionnaire, there is a consist- ent difference in the response pattern of the two experi- mental groups. These differences may be more clearly recognized by computing the critical ratios of the differ- ences between their respective means. Using the same processes as were employed in Chapter III with respect to Group Structure A, Table XXI was constructed which presents a summation of these critical ratios plus an interpretation of their levels of significance. The level of confidence index is based upon the same one-tailed distribution table as was used for treating the parallel data in Chapter III. An examination of Table XXI indicates that nine of the ten indices are statistically significant in indicating that the contrasting performances of the two experimental groups were not due to pure chance factors. Eight of these nine ratios were significant at or beyond the one-tenth of l per cent level of confidence. The other statistically significant index was found to be at the 2 per cent level of confidence. Over-all, the total performance of the two groups on the ten items was signified by a critical ratio of 9.51 so that it was rather evident that the aggregate of these items possessed a discrimination power of substantial weight. 1&2 TABLE XXI CRITICAL RATIOS FOR GROUP STRUCTURE E Item Critical Level of Conclusion Ratio Confidence 1 3.18 0.1% Very significant 2 2.20 2.0% Significant 3 h-9U 0.1% Very significant h h.hl 0.1% Very significant 5 5.26 0.1% Very significant 6 6.03 0.1% Very significant 7 6.69 0.1% Very significant 8 7.58 0.1% Very significant 9 1.18 null No significance 10 6.27 0.1% Very significant Totals 9.51 0.1% Very significant W 1&3 Analysis of Performance on Individual Items As in the immediately preceding chapters it is the purpose of this chapter to treat not only the statistical aspects of the derived data but also the content aspects of the various items under examination. These items specifi- cally are the ten attitudinal statements that refer to the whole, or to some component part, of Group Structure E. As a point of reference the stereotype of Group Structure E is set out as follows: V Group Structure E In a,survey of the convention membership. the con- vention chairman has discovered various areas of professional interest which may not have been com- pletely covered on the official agenda. He informs the membership that the smaller conference rooms will be immediately available to any interested in- dividuals who desire to meet among themselves and continue further discussions in line with their own professional interests. There is to be no assignment of conference officials to help organize these discussion groups. Rather, each group will have to determine for itself its own kind of organization and what it wants to do. Attendance at any of these free discussion groups is entirely optional. You may assume that one of the discussion groups will concern itself with an area of your general interest. The first of the ten items appearing below the stereo- type is as follows: 1. This is the kind of meeting I enjoy attending. This item is noted to be identical to the first items for the immediately preceding structures. It was prepared mm with the same general purpose of eliciting a generalized negative or position reaction to the group structure as a whole. Since, by construction. the group structure in question is intended to represent an extremely plastic or- ganization rather than a rigid one. it would be expected that non-authoritarian personalities would be more attracted to the stereotype than authoritarians according to the de- sign of the study. The evidence revealed that the non-authoritarian group obtained an anti-rigidity score of 3.59 on the item, which (when properly inverted) corresponded to a key value of h.hl, or about half-way between neutrality and agreement with the item. The authoritarian group obtained a mean score of h.59 which (when properly inverted) corresponded to a key value of 3.h1, or in the zone of disagreeing with or rejecting the statement. Thus, on a relative basis, it was apparent that authoritarians demonstrated a somewhat more negative attitude to the stereotype as a whole than did the non-authoritarians. The critical ratio of the differences between the two performances was found to be 3.18, or exactly at the one-tenth of l per cent level of confidence for a one-tailed distribution organization. Item two in the series is noted to be: 2. ‘lhan professional information is ex- changed through participation, it is ordinarily more clearly comprehended and better remembered. 1&5 Although traditional textbooks on the psychology of learning, particularly Thorndyke's theory of the "stamping- in" process, rather give basic foundation to the import of this item, it was still hypothesized that the level of ac- ceptance toward this statement would vary between the two experimental groups. On the assumption that authoritarian personalities would possess a bias against the participa- tion process, it was presumed that they would resist as complete an endorsement of this type of learning situation as would the authoritarians who were assumed to be biased in the opposite direction. The results of the testing seemed to support the conjecture. Both groups were in agreement that the item was substantially true. but varied somewhat in their levels of endorsement, with the non-authoritarians manifesting greater ”enthusiasm" for the stated principle than the authoritarians. Both groups obtained scores that were weighted according to the key at the small or "anti-rigid” end of the seven step key. These scores were 2.23 and 2.76 for the non-authoritarians and authoritarians respectively. Since it was an inverted item in that both agreement re- sponses were keyed to anti-rigidity weightings. the level of agreement for the two groups would have to be found by inverting the positions of the two scores to the correspond- ing points at the opposite end of the scale. The non-authori- tarian score of 2.23 would correspond to a 5.77 scale point lh6 of definite agreement. The authoritarian anti-rigidity score of 2.76 would correspond to a 5.2h scale point which, though in the agreement zone of the scale, represents a more modified level. The critical ratio between the two was found to be at the 2 per cent level of confidence. The third item in the series is: 3. In a free discussion group it is unfortunate that those least equipped to make a professional contribution consume equal amounts of the available discussion time as do those most equipped. This item was intended to represent an anti-democratic concept. It implies that individual rights in a free dis- cussion situation should be limited in accordance with one's position in the professional status hierarchy of the group. Those who know the most should have not only the most to say but the most time to say it, et cetera. The item in- clines toward endorsing a return to rigidity in structure and the establishment of power lines of authority over various members of the group in various degrees. As such it was presumed to elicit a greater level of endorsement from the authoritarians than from the non-authoritarians. The results confirmed the prediction. The authoritarian group obtained a mean score of 5.h0 on the item, or a rather definite manifestation of agree- ment. The non-authoritarians with a mean score of h.08, however, exhibited a pronouncedly neutral attitude to the 1&7 item in question. The critical ratio between the two of n.9h was very significant at beymnd the one-tenth of l per cent level of confidence. The next item in the series is: h. A group that has been gathered around an interest rather than around a leader is not likely to accomplish too much. As elsewhere more extensively discussed in this study, the authoritarian personality is most concerned with the leader-follower power dimension. As Frankel-Brunswick repeatedly brings out in her discussions (1), the authori- tarian personality is one that, in a political reference, is psychologically predisposed to fascism. In contrast to this, the non-authoritarian personalities are assumed to possess an equalitarian outlook which predisposes them toward more democratic processes. The construction of item h was intended to highlight the contrasting philoso- phies of authoritarian and non-authoritarian personalities in a group process. The authoritarian personalities were anticipated to emphasize that the leader would be more important than group interests, and the non-authoritarians that group in- terests are more important than the leader. The variable performances of the two groups did reflect a tendency in the direction predicted. The non-authoritarians, with a ‘mean response of 2.32, manifested a rather definite rejection lh8 of the attitudinal statement. The authoritarians, however, were not nearly so rejective though they too did not endorse the item. Their mean score of 3.h8 placed them half-way between the scale point of slight agreement and neutrality. The critical ratio between the two of h.hl was significant at beyond the one-tenth of 1 per cent level of confidence. Item five in the series is as follows: 5. Groups like this should adOpt some form of parliamentary procedure rather than to risk letting the discussion get completely out of hand. Within the limits of individuals' ability to communi- cate with one another through a common language, the in- timate and free-flowing exchanges that characterize the ”bull-session" are as a polar opposite to the august, exacting, and, it must be conceded, rather stilted phraseologies that comprise the language of the parlia- mentarians. Rooted in its historical growth to the an- ciently evolved English governing system, the purpose of parliamentarian procedure has been accepted as a code of procedures that, when rigidly observed, assures to each the right to be heard. As such it governs the behavior of in- dividuals on the power principle: communication is not free but restrictive. Authority lines are rigidly set, and, when wielded by an expert parliamentarian, can be so manipulated as to keep inarticulate those who might disturb the group with needless disruptions. lh9 Since by its very nature parliamentary procedure is restrictive of natural group processes, the inhibition of participation would be assumed to evoke a negative response from the non-authoritarians; because it emphasizes a return to structure and authority, it would be presumed to arouse a positive response from the authoritarians. The differ- ential performances of the two groups indicated that, on a relative basis, they were trending in the direction pre- dicted. The authoritarian group responded to the items with a positive endorsement at the 5.8h level. 'The non- authoritarian group exhibited a much more neutral attitude with a mean score of n.5h. The critical ratio between the two was 5.26, or significantly beyond the one-tenth of l per cent level of confidence. Item six in the series is as follows: 6. Meetings of this sort are generally no more than informal "bull sessions" which accomplish little as compared with the more formally planned meetings. Implicit in this item is a general lack of confidence in the ability of the ”average" group member to make a con- tribution that would be comparable to those produced by the more recognized authorities. A formally planned meeting connotes structuring a meeting with pre-selected experts, which, according to syllogistic reasoning, might connote an ability to accomplish more than would be possible for a group of unselected (and therefore not so expert!) 150 "average" members of the profession. At any rate the item was conceived to test the attitudes of the two experimental groups toward their confidence (or lack of confidence) in the ability of a discussion group to be really effective. The non-authoritarian group was substantially negative toward.a statement so sharply critical of informal, group centered, processes. Their mean score was 2.90. The authori- tarians, however, found little to criticize about the state- ment, as manifest in their score of h.57, and in fact adopted a position about midpoint between neutrality and slight agree- ment. The critical ratio between the two was found to be 6.03, or well beyond the one-tenth of l per cent level of confidence. Itmn seven in the series is: 7. New scientific ideas are more apt to be developed by individuals inspired by the challenge of competition, rather than to be the product of cooperative group discussions between socially minded colleagues. Implicit in the society in which we find ourselves are those enduring cultural concepts that are believed to have stood the test of time and are, therefore, worthy of venera- tion. Our setting is one that has included a belief in the principle of ”rugged individualism", or progress stimulated on the spirit of competition. In opposition to this tradi- tional concept, there emerged in the last great depression a counter-set of beliefs in the direction of group coopera- 151 tives. In.this,rugged individualism was to be largely abandoned in favor of group action, as exemplified in the growth of unionism.and a variety of social legislation. Thus the society was confronted with two antagonistic principles of action, the traditional one of competition, with emphasis upon the individual keeping his identity, and the new one of cooperation with emphasis upon the individual sharing his identity with that of the group. Because the authoritarian personality is believed to be one in which strict adherence to traditional, middle- class values is a component, this item was constructed with the expectation that parallel attitudes would be manifested by them in choices between parallel conference processes: that is to say, they would be expected to possess the attitude that a rugged individualist under the stimulus of competition can accomplish more by doing a research study and then reporting on it than can a group of individuals cooperatively sharing a common problem and trying to talk it through, all other things being’equal. The evidence on the contrasting performance of the two groups was strongly indicative that this was so. The non- authoritarians definitely rejected the concept that new ideas are best developed by individual competition as was evident in their low mean score of 2.90. The authoritarians, on the other hand, were not so resistive to the statement. 152 Their mean score of h.68 located them.two-thirds of the way up the neutrality section of the scale, trending toward slight agreement. The critical ratio between them of 6.69 was well beyond the one-tenth of l per cent level of con- fidence. Item eight reads as follows: 8. A formally planned meeting is better than an unplanned meeting of this type in that the speaking assignments have already been definitely assigned, and you have no. worries about having to "do something”. This item may be noted to be quite similar in content to item 7 which is discussed under Group Structure D in the previous chapter. However, the item is stated with the in- tention of soliciting a greater appeal for agreement from those individuals who are uncomfortable in completely democratic settings, and of course, it is stated with reference to a different group structure. The dynamics of the authoritarian personality in seek- ing to have all events pro-planned and ordered, and not havhng to be ”worried” about suddenly bebng thrust into an unexpected situation, are adequately described in the previous chapter, as noted, and need not be repeated here. In examining the date it was apparent that the appeal to agree with this statement urging preplanning was sub- stantially rejected by the nan-authoritarians who produced the rather low score of 2.66. However, the appeal elicited much better reception from the authoritarians. They 153 responded with a mean score of n.5h which indicated a level half-way between neutrality and agreement. The critical ratio between the two was 7.58, or well beyond the one- tenth of l per cent level of confidence. Item nine is as follows: 9. A discussion group should be as much concerned with the individual's social attitudes toward the other members of the group as with the professional information that is being exchanged. Item 9 was written with the intent of contrasting those emphases in group processes which touch alternately on social values and subject matter values. With the theoretical as- sumption that the authoritarian personality frequently places more value on the concrete realities of subject mat- ter in preference to the tenuous abstractions of feelings, it was thought that this item would offer greater attrac- tion to the non-authoritarians, and less for the authori- tarians. In reviewing the data, no important differences in performance between the two groups was noted, however. Both made scores that were not too different in rigidity values. The non-authoritarians obtained a mean score of h.h8 or only slightly less rigid than the h.81 score of the authoritarians. Since the item is an inverted statement, these two scores would invert to 3.52 and 3.29 respectively which signified that both groups were somewhat in disagree- ment with the idea that a discussion group should be as 15h equally concerned with social attitudes as with subject matter. However, the authoritarians were very slightly more in disagreement with the statement than the non- authoritarians. The critical ratio of 1.18, however, is noted to be less than at the 10 per cent level of confidence, and therefore, of no real statistical significance. A pos- sible weakness of the item might be contained in that part of the item which implies that the two aspects of the proc- ess should be given Eggs; consideration by all members of the group. However, that is a possibility for further in- vestigation. The last item in the series is as follows: 10. A spontaneously formed discussion group organized around a newly discovered interest is not as likely to be as profitable as a program that has been carefully thought out in advance and staffed with suitable speakers. Item 10 is noted to be somewhat similar to item h in this series but with a slightly different emphasis. Rather than directing attention to the possibility of rallying a group around a leader or power-figure who would be able to I'organize" them as in item h, the emphasis here is with re- gard to the general principle of pre-staffing with qualified speakers so that the conference chairman can be sure that the subject matter is there and has a chance to come out. .Again there is evident a general lack of confidence in the average group member to relate with other group members, and through collective efforts, produce anything too profitable. 155 In line with the general concept of the authoritarian personality as previously described, this item was believed to be cued in the direction of obtaining agreement from the authoritarians and disagreement from the non-authoritarians. The results indicated that this was a sound conception. The authoritarian group exhibited agreement to the item with a mean score of 5.13, or definitely in the slight agreement zone of the scale. The non-authoritarian group exhibited a rejection to the statement with a score of 3.61, or near to the midpoint of the slight disagreement zone of the scale. The critical ratio was found to be 6.27, or very significant at beyond the one-tenth of l per cent level of confidence. Summary The stereotype of Group Structure E was in contrast to the stereotype of Group Structure D in that it described a leaderlggg group discussion process in contrast to a leaderfilgg group discussion process. Both group structures were alike in that they exhibited group-centeredness with a decrease of organizational structure. The leaderless group process treated in this chapter represented the ex- treme stereotype of an officially sponsored professional meeting conducted in an atmosphere of the least possible amount of "structure" as such. 156 The chapter was concerned with a study of the differ- ential attitudes manifested by two experimental groups toward a series of ten statements that pertained to differ- ent aspects of this type of group-centered process. The design of the stereotype suggested that non-authoritarians should respond to the series with rejection attitudes toward items which encouraged a return to the rigidity features of previous stereotypes, and the authoritarians with permissive- ness toward these types of items. The data, when analyzed, confirmed the expectation. In all ten pairs of scores, the non-authoritarian group produced a score signifying prefer- ences for the less rigid components of the group structure in question than did the authoritarians. In eight of these cases the differential performances were so significantly different as to elicit critical ratios at or beyond the one-tenth of l per cent level of confidence. In the other two instances one critical ratio was found to be significant at the 2 per cent level of confidence and the other was viewed as statistically insignificant at less than the 10 per cent level of confidence. Aggregating the performance of the two experimental groups on the total series of ten items, the mean response for the non-authoritarians was found to be 3.37 which signified negative preference for rigidity features pre- sented in the series at a level definitely in the disagree- ment zone. The authoritarian group demonstrated a lack of 157 rejection for these rigidity features with a mean response of h.59 which located them still in the neutral zone but trending in the direction of slight agreement or acceptance. The critical ratio between the two groups on total performance was found to be 9.51, or very significant beyond the one- tenth of l per cent level of confidence. Thus the data presented in this chapter supported the basic hypothesis that authoritarians would respond toward a group-centered process of the leaderless type with attitudes consistently more negative than those demonstrated by non-authoritarians. 158 CHAPTER VIII ANALYSIS OF ATTITUDES IN GROUP STRUCTURE F Introduction It is the purpose of this chapter to present an analysis of the data which was gathered under Group Structure F of the questionnaire. This group structure was constructed with the intent to illustrate the final removal of any evidence of official structure in a group organization. In the pre- vious chapter, the description of the "leaderless group‘I very nearly approximated this elimination of structure as such. However, it had been recognized that for that situa- tion the convention chairman had given official recognition that certain needs to discuss special interest topics were found to exist: that a designated time and.meeting place had been set aside for interested individuals to pursue these discussions: and an official invitation had been tendered that they do so. In Group Structure F, however, it was intended that these last vestiges of structure would be absent.' This stereotype refers to the spontaneous types of groupings that occur during intermissions, before and after meetings, in and about the corridors, lounges, and hotel 159 rooms. The groupings are often in a state of flux, with the membership changing almost moment by moment. In other situations the grouping maintains its identity for longer periods of time. The purposes of each grouping may vary in relation to the needs of its current membership. For some it may be an opportunity to serve as a sounding board, with a strong power figure doing most of the talking, and the other members in various postures of respectful approval. In other instances the grouping is democratically oriented and all members are participating equally and at about the same level. Since ”structure” connotes an organized procedure as to how one may enter a group meeting as well as the manner in which one should conduct himself after he has so entered, the total lack of structure for these spontaneous groupings offers no comfortable, secure, or "authorized” procedure by which many of the delegates feel free to enter, and as a result some of them may be isolated. Those who are left out may consist of those individuals who would like to be "in", if possible, as well as those other ”lone wolf” type of personalities who consciously or unconsciously seem to avoid entry into such groupings. Thus Group Structure F has been hypothesized to repre- sent an ambiguous structure. For those types of authori- tarian personalities which the literature describes as pessessing the "lone wolf" type of attitude, the social and 160 informal groupings described in the stereotype represent situations which they seek to avoid. For those other types of authoritarian personalities in which the need to dominate or be.a power figure is most important, many, but not all, of these informal groupings present ideal opportunities for. this type of release. The non-authoritarians, in line with the basic hypothe- sis of this study, would be presumed to welcome opportunities for participation in a democratic setting. Some individuals, basically non-authoritarian in organization, might desire entry into these types of groupings, but for other reasons of a personal or professional nature cannot do so, or at least to the extent desired. In predicting anticipated preferences toward spontane- ous groupings on the part of the two experimental samples, it was hypothesized that the authoritarian sample would be more ambivalent in their attitude toward this type of struc- ture than the non-authoritarians, and that, in general, the non-authoritarians would be expected to exhibit a generally positive attitude toward this type of structure as compared with the authoritarians. This was formulated on the basis that Group Structure F is essentially a participation process and as such would appeal to the participation-minded non- authoritarians: whereas to the authoritarians it is somewhat disturbing as a free participation process except for those 161 particular authoritarians who are dominating such a struc- ture, or who are being dominated by a leader whom they respect and in whom they place considerable confidence. In addition, for those individuals who find it difficult to enter such informal groupings, it was felt that the desire to do so would be more characteristic of the less authori- tarian but perhaps "shy" personality than it would be of those authoritarian types who are presumed to possess the "lone wolf" attitude. Thus it was anticipated that non- authoritarians would be more generally pro-disposed in favor of this type of informal group situation whereas the authoritarians were anticipated to be more divided among themselves. Below the description of Group F are presented ten statements, each of which represents an attitude about dif- ferent aspects of this kind of group structure. The response patterns of the two experimental groups to these ten items will be presented in this chapter with the viewpoint of comparing and interpreting the mean differ- ences between the two. As in the preceding chapters the hypothesis continues to be directional in that the non- authoritarian group is predicted to always produce a mean score to the left of each mean score produced by the authori- tarian group on a given item. 162 Presentation of Data on the Distribution of Responses on Group Structure F Tables XXII and XXIII present the actual distribution of weighted score responses to Group Structure F on the part of each experimental sample. All raw score answers were first converted to weighted scores in accordance with the process and key as described in Chapter III. Analysis of the Differences Between the Mean Responses on Group Structure F In order to draw significance relative to the distri- bution of responses of the two experimental groups as they appear in Tables XXII and XXIII, it was first necessary to calculate the mean responses of the two experimental groups with regard to each item. The means presented in Table XXIV are the arithmetic means calculated in the same manner as that described in the preceding chapters. In addition, the standard deviations are similarly included. Table XXIV summarizes the means and standard deviations of the two experimental populations on these ten items. An inspection of Table XXIV indicates that in nine of the ten pairs of mean scores for the two experimental groups, the non-authoritarians obtained a smaller mean score than did the authoritarians. In the single exception the non- authoritarian score was larger by only one-tenth of a point. TABLE XXII DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE F BY NON-AUTHORITARIANS 163 Item weighted Score Responses Total 1 2 3 h 5 6 7 Number 1 7 19 22 26 9 2 85 2 7 25 35 1h h 0 85 3 6 18 26 20 l2 3 85 u o 2 10 30 3o 13 85 S 2 9 it 29 23 8 85 6 18 25 20 16 h 2 85 7 22 36 16 lO 1 O 85 8 13 ll 1 22 19 ll 85 9 10 13 25 23 10 85 10 7 ll 20 27 12 8 85 TABLE XXIII DISTRIBUTION OF RESPONSES ON GROUP STRUCTURE F BY AUTHORITARIANS 16h Item Weighted Score Responses Total 1 2 3 Li 5 6 7 Number 1 u 13 16 22 13 17 85 2 5 19 25 12 1n 10 85 3 11 1o 22 17 1h 11 85 u o u 7 3o 23 21 85 5 1 8 1o 17 20 29 85 6 12 21 19 1h 12 7 85 7 13 3o 17 10 12 3 85 8 o 6 5 19 17 38 85 9 3 16 12 2n 13 17 85 1o 5 9 17 17 33 85 165 TABLE XXIV MEAN RESPONSES TO GROUP STRUCTURE F BY AUTHORITARIAN AND NON-AUTHORITARIAN SAMPLES W AuthoritaW Mean Standard Item Mean Standard Response Deviation Response Deviation 3.63 1.7 l h.53 1.9 3.01 1.3 2 3.90 1.9 3.68 1.7 3 h.03 2.0 5.35 1.2 u 5.h6 1.3 n.72 1.6 5 5.35 1.7 2.89 1.6 6 3.55 1.9 2-33 1.3 7 3.1n 1.8 h.37 1.9 8 5.76 1.5 n.66 1.7 9 8.56 1.9 ...I O n.1h 1.8 S.h0 1.8 166 Thus the data rather strongly confirmed the directional prediction that the non-authoritarians would score to the left of (or smaller) than the score of the authoritarians in this series. As a reference point for interpretation, it will be recalled that the smaller weighted scores signify anti-rigidity preferences in group structure in contrast to higher weighted scores which signify pro- rigidity preferences in group structure. Item Analysis of Differences Between the Means It is apparent that for the ten items listed.under Group Structure F in the questionnaire, there is a con- sistent difference in the response pattern of the two ex- perimental groups with but one exception. These differences may be more clearly recognized by computing the critical ratios of the differences between their respective means. Using the same processes as were employed in Chapter III with respect to Group Structure A, Table XXV was constructed to present a summation of these critical ratios plus an interpretation of their levels of significance. The level of confidence index is based upon the same one-tailed dis- tribution table as used for treating the parallel data in Chapter III. An examination of Table XXV indicates that seven out of the ten indices are statistically significant in indicat- ing that the contrasting performances of the two experimental 167 TABLE XXV CRITICAL RATIOS FOR GROUP STRUCTURE F Item Critical Level of Conclusion Ratio Confidence 1 3.27 0.1% Very significant 2 3.55 0.1% Very significant 3 1.23 null No significance h 0.55 null No significance 5 2.h5 1.0% Very significant 6 2.38 1.5% Significant 7 3.38 0.1% Very significant 8 5.17 0.1% Very significant 9 -0.36 null No significance 10 u.52 0.1% Very significant Totals 5.06 0.1% chy significant W 168 groups were not due to pure chance factors. Five of these seven ratios were found to be significant at a point beyond the one-tenth of l per cent level of confidence. 0f the other two, one was found to be beyond the l per cent level of confidence, and the other to be at approximately the l per cent level of confidence, but stated herein as at the 1% per cent level of confidence. Over-all, the total per- formance of the two groups on the ten items was signified by a critical ratio of 5.06 so that it was rather evident that the aggregate of these various items possessed a dis- crimination power of substantial weight. Analysis of Performance on Individual Items As in the immediately preceding chapters it is the purpose of this chapter to treat not only the statistical aspects of the derived data but also the content aspects of the various items under examination. These items specifically are the ten attitudinal statements that refer to the whole, or to some component part, of Group Structure F. As a point of reference the stereotype of Group Structure F is set out as follows: Group Structure F Between meetings, or during intermissions, it is quite typical for many spontaneous groupings to occur. These clusters of small groups usually take place in the corridors and lounges adjacent to the formal meeting rooms. Each of these is usually made up of several individuals who 169 gravitate together for varying lengths of time. The membership of these circles is often quite temporary, with new members dropping out to join other circles, or to pursue other ends. Some of these groups seem to be centered around some one or two particular leaders or speakers; others seem to exhibit equal participation by all. Often a particular circle dissolves, and the member- ship becomes separately absorbed in other groupings or other interests. Occasionally these circles continue throughout the period of some regularly scheduled meetings, and their membership is necessarily comprised of absentees from.such official meetings. Ordinarily, a certain portion of the convention membership does not participate in these spon- taneous groupings except on infrequent occasions. Some of these individuals are occasionally absent from the regularly scheduled meetings, and either engage themselves in activities removed from the scene of the convention, or sit about the lounges in personal relaxation for various lengths of time. A number of these individuals have personal needs or satisfactions which they feel will not particu- larly be met by joining such informal groupings: other individuals would like to participate in such memberships but due to factors such as shy- ness or other personal reasons, they hesitate to actively pursue entrance into these spontaneously formed circles. The first of the ten items appearing below the stereo- type is as follows: 1. There is a great deal of time wasted in corridor gatherings which could be put to better planned use. This item was constructed with the intent to attract the presumed interest of authoritarians to place subject matter and organization well ahead of social purposes. It suggests that much of the process of informal group inter- action is in a sense ”wasted time", and that if replaced by 170 a planned and therefore "structured" activity, it would be better for everybody. This conception would imply a contrary attitude on the part of non-authoritarians who were presumed to be oriented toward a more democratic outlook on such processes, and to possess a lesser need for formalism. In line with such an expectation the data revealed that the non-authori- tarian group responded to the item with a mean score of 3.63, or a point on the scale located in the slight dis- agreement zone. The authoritarians, however, were unable to reject the statement, as witnessed by their mean score of h.53, a point on the scale corresponding to the neutral- ity zone. The critical ratio between the differences in the two performances was found to be 3.27, or beyond the one- tenth of l per cent level of confidence. The second item in the series is found to be: 2. Too much attention is being given to the "social" aspect of the average pro- fessional conference. Item 2 was constructed with the view of defining the attitudes of the two experimental samples toward both the spontaneous as well as the pre-planned ”social" activities of the professional conference. Though it is somewhat broader in scope than item 1, it was presumed that it would state a principle that would continue to be more unacceptable to non-authoritarians than to authoritarians. The evidence 171 indicated that concurrently with the slight variation in the item, a slight variation in performance was elicited from both groups although their relative differences be- tween each other was maintained at about the same degree. To this item the non-authoritarians were somewhat more rejecting in attitude than they were to item 1. Their mean score was found to be 3.01, or at the "more definite" and of the "slight disagreement” zone of the scale. The authoritarians obtained a mean score of 3.90 which, though within the slight disagreement zone, was at the opposite end, trending very strongly in the direction of neutrality. The critical ratio between the two was noted to be 3.55, or beyond the one-tenth of l per cent level of confidence. Item three is noted to be: 3. Actually the individual obtains about as much worth-while professional information in the informal meetings as he does in the official meetings. Item 3 was constructed with the idea of highlighting one of the controversial points between the two general types of convention attenders who are presumed to represent the positive and.negative attitudes toward this area of con- ference activity. In line with the basic premise that non- authoritarians are philosophically pro-disposed toward learning processes that maximate group-centeredness and participation, it was anticipated that this sample would tend to agree with the item. Therefore, by construction, 172 this item.was written in an inverted direction which is to say the higher the level of agreement, the more non- rigid becomes the interpretation of the score. The non-authoritarians yielded an anti-rigidity score of 3.68 which, when inverted in order to ascertain the level of agreement, would correspond to a neutrality level at the h.32 point. The authoritarians were not quite so anti-rigid. Their "anti-rigidity" score proved to be h.03, which, when inverted (3.97), located them barely within the zone of very slight disagreement with the stated item, but border- ing very closely to a frankly neutral response. However, the differences in performances between the two groups did not appear to be of statistical significance. The critical ratio between the two of 1.23 was noted to be at less than the 10 per cent level of confidence. Perhaps a partial reason for the failure of the two groups to exhibit more conspicuous differential performances might reside in the fact that an underlying premise of the study concedes that a representative portion of the authoritarians are believed to be adherents to this type of group process (though for different reasons than the non-authoritarians). Item four is as follows: h. The major emphasis of a conference should be on good fellowship and getting acquainted rather than on the formal ex- change of professional information. 173 This item was conceived as of the "testing the limits" variety. It was constructed as a deliberate overstatement of the value of social processes in contrast to the more professional purposes of the conference. It was presumed that in rejecting the item, the authoritarian group would react with significantly more negative behavior than the non-authoritarians. The obtained data indicated that the extremeness of the statement found both experimental samples reacthng at about the same level of negativity. The non-authoritarians with their mean score of 5.35 exhibited a rather definite disagreement response corresponding to the 2.65 point on the inverted scale. The authoritarians with a mean score of 5.h6, converted to a 2.5h on the inverted scale, signified an equal level of disagreement, if not even slightly more so. However, the differential performances was not sig- nificant to any degree. The critical ratio was found to be 0.55. Item five is noted to be: 5. At the average conference I am more interested in what takes place between meetings rather than in what takes place during meetings. This item was constructed with the intent of focusing upon those individuals who primarily attend conferences for the purposes of becoming socially involved with people as opposed to individuals who primarily attend in order to 17h acquire professional information. The object of the item was to ascertain whether individuals responding most nega- tively to the statement would not indeed be the authori- tarians, even though many such might otherwise profess much satisfaction with certain kinds of informal group situations. The evidence tended to support the expectation. The authoritarian group obtained a mean score of 5.35 which, for an inverted item, signified a level of disagree- ment at the 2.65 point. The non-authoritarians, however, obtained a score of h.72 which corresponded to a much milder level of disagreement at the 3.28 point. The critical ratio between the two was found to be 2.h5, or very signifi- cant at the l per cent level of confidence. Taken as a single item only, this bit of evidence would tend to sug- gest that the "corridor prone" population at the average professional conference is comprised more extensively of non-authoritarians than of authoritarians. Item six is as follows: 6. Rather than to pursue these corridor relationships, I usually prefer to use my free time for such relaxations as reading the paper, going for a walk, shopping, visiting elsewhere in the city, or taking a short rest in my room. In the discussion that introduces this chapter, mention was made of the concept that the various individuals who do not enter into the spontaneous groupings described in the stereotype might be of two general types: those who would 175 like to be ”in” but, for reasons of shyness, et cetera, are not able to make such entries: and those who seem to actually avoid getting ”in". It was hypothesized that a number of non-authoritarians might be of this first category but that, to a predominant degree, individuals of the second category would tend to give larger representation to the authori- tarian sample. This item was designed to take a measurement of the degree to which individuals of the 'lone wolf” type would be found in each of the two experimental samples. The non-authoritarians exhibited a majority opinion rejecting the item as signified by a mean score of 2.89. The authoritarians, however, apparently were represented by many more individuals who possessed attitudes in the speci- fied direction since the mean score of their total group was found to be at the relatively higher point of 3.55. The critical ratio between the two was at the 1% per cent level of confidence. Again as a single piece of evidence, it would appear that individuals who prefer not to participate in purely social group processes are more likely to be found among the authoritarian non-participators than among the non-authoritarians. Item seven is noted to be: 7. I usually find it rather difficult to enter into these spontaneous groupings. 176 Item 7 was conceived with the idea of ascertaining not which sample possessed the most number of ”lone wolf" personalities, but rather which sample possessed the most number of individuals who tend to be left out of the in- formal groupings even though interested in joining such groups. It would be expected that a portion of the ”lone wolf" personalities would consist of personalities who, upon exercising self-choice, could enter into these spon- taneous groupings as readily as many other practiced in- dividuals. With this assumption as a premise, it would be expected that some individuals of this particular category might properly have responded in agreement to item 6, but in disagreement to item 7. In an examination of the data the evidence indicated that the non-authoritarians obtained the rather low score of 2.33. Their rejection of the item would suggest that the proportion of individuals in their sample who represent the "left-outs" is apparently rather small. In comparison, the authoritarians obtained the relatively higher score of 3.1h, thereby inferring a greater incidence of this type of personality in their population. The critical ratio between the two was 3.38, or significant at beyond the one- tenth of l per cent level of confidence. 177 Item eight is noted to be: 8. Generally, I prefer to be in a group which possesses a speaker of sufficient authority in his field that the group can really profit by what he has to say. In the discussion pertaining to the stereotype of Group F, it was suggested that informal corridor groupings might provide different kinds of opportunities to different kinds of personalities, and that for authoritarians such a setting might provide an opportunity to be either a power figure in dominance over such a group, or to be a respectful follower of such a power figure and, in such capacity, to become a member of his small surrounding group. Item 8 at- tempts to state just such a setting. The evidence indicated that the authoritarian sample performed in accordance with the dynamic processes antici- pated for them. Their mean score of 5.76 indicated quite definite agreement with the stated item. The non-authoritarian group failed to accept the item as stated, and in fact exhibited an entirely neutral re- sponse at the h.37 level. The critical ratio between the two was found to be 5.17, or very significant at beyond the one-tenth of l per cent level of confidence. Item nine is as follows: 9. Generally, I prefer to be in a group. composed of colleagues of about the same professional status as myself, wherein nothing too important is being said, but everyone is saying something. 178 In the process of item construction, it has often been noted that an item stated in a positive direction may prove to have no particular discrimination power, but when re-worded in a reverse or negative direction, it proves to have a very significant discrimination power. In order to insure that the concept discussed in item 8 would have a more maximal opportunity for discrete consideration by the two experimental samples, items 8 and 9 were prepared as a pair in order to state the concept in two different directions, item 8 being weighted in the authoritarian direction and item 9 being weighted in the non-authoritarian direction. An inspection of the data indicated that item 9 con- tributed no information that could be said to distinguish the performance of the two groups. Both made neutral scores of h.66 and h.56 respectively, and the critical ratio was found to be insignificant. Item ten is noted as follows: 10. Individuals who are prone to spend extra time in corridor gatherings instead of attending the official meetings are exhibiting a real discourtesy to the scheduled speakers and the other delegates. Item.10 was intended to portray the hostility attitudes of the true authoritarian personalities toward any trans- gressors of the idealized conventional codes. This com- ponent of the authoritarian syndrome seemed to be manifested in the data by a definite agreement score of 5.h0 on the 179 part of the authoritarian sample. In contrast, the non- authoritarians demonstrated a completely neutral attitude toward.the statement with a mean response of h.lh. The critical ratio between the two of h.52 was found to be significant at beyond the one-tenth of 1 per cent level of confidence. Summary The stereotype of Group Structure F was in contrast to that of Group Structure E in that it carried the modi- fication of ”structure" to an ultimate degree, removing the last vestiges of pre-planning, and placing each.member on his own. Because of the extreme lack of structure in this type of group situation, the hypothesis was varied at this point. For the non-authoritarians it was presumed that their preferences for participation processes in a democratic setting would continue to be endorsed by the group as a whole: for the authoritarians, however, it was suggested that the ambiguities of entirely free groupings would evoke a variety of responses, with some authoritarians perceiving certain kinds of informal group settings as pro- vidhig opportunities for certain kinds of authoritarian satisfactions, and others in the population generally per- ceiving all such group operations as distractions to the real business of the convention. .T-m. .14-”- 1. Jv' ’." 180 In order to investigate the hypothesis more particularly, the chapter concerned itself with a study of the differential attitudes manifested by two experimental groups toward a series of ten statements that pertained to different aspects of this type of purely group-centered process. The data, when analyzed, seemed to bear out the prem- ises upon which this part of the study were conditioned. 0f the ten pairs of scores obtained, nine Were found to be in the direction predicted, although from a statistical point of view only seven of these nine were importantly con- spicuous. In terms of the over-all performances of the two groups, the critical ratio of 5.06 signified very real opera- tional differences between the two populations. The consistently uniform performance of the non-authori- tarian group in the direction of responding with scores of a less rigid value than that of their control group provided considerable support for the hypothesis which had been pre- dicted to govern their performance. In reference to the authoritarian group, the hypothesis that their performance would tend to be governed by ambiva- lent forces is best corroborated by an inspecticn of their performance on those particular items in the series that have an apparent hearing. Items 6, 7, and 8 are some of the principal ones that bring out important highlights. The evidence for item 6 seemed to substantiate the concept that though many authoritarians might find satis- 181 factions for their authoritarian needs in informal group situations, the pepulation of their sample would include more individualscf the "lone wolf” type than would be found in the non-authoritarian sample. The evidence for item 7 gave support to the somewhat parallel concept that more individuals who are typically “left-out” of informal groupings, even though frankly desirous of being ”in”, are also more represented in the authoritarian sample. The evidence for item 8 gives equally sound support to the concept that some of these informal group situations offer specific opportunities for achievement of authoritarian satisfactions. Thus, item 6 suggests that a more significant propor- tion of the authoritarian sample is composed of individuals who seek to avoid entry into these types of group situations: item 7 suggests that a more significant proportion of the authoritarian sample is composed of individuals who would like to be "in", but who are frustrated in their desires for one reason or another: and item 8 confirms the sug- gested concept that certain-kinds of these group situations are participated in by other authoritarian individuals since they afford a kind of a satisfaction not sought to the same degree by the non-authoritarians. These three conflicting preferential differences are indicative of the ambivalence that had been hypothesized for this group . 182 CHAPTER IX , ANALYSIS OF TEE DATA FOR RANK ORDER OF GROUP PREFERENCES Introduction The questionnaire employed in collecting the data is noted to contain an extra item at the foot of each of the five pages which are concerned with group structures B, C, D, E, and F. In line with the special note that is set out in detail on page 3 of the questionnaire, the individual was to gradually build a progressive rank order of his general preferences for the six described group structures, that is, he first stated his rank order of preferences be- tween Group Structures A and B. Then, when he had completed the next ten items under Group C, he assigned a rank order of preferences among the three groups. He then proceeded to rank the four groups from A through D, next the groups from A through E, and finally, the last six from A through F. This procedure of having the respondent go through an experience of progressively building a final rank order of preferences for all six groups was felt to be more theoreti- cally sound than simply to present one final item at the end of the questionnaire which asked for a single listing of the. 183 entire rank order of preferences fbr all six groups. A progressive development of the final rank order provides greater opportunity for each group structure to be evaluated on an individual basis, and With only one new group struc- ture at a time to be compared with group structures that have already had previous independent evaluation. It was felt that for the respondent to make a single rank order of all the groups at but a single moment of consideration might lead to confusion and perhaps capriciousness in the assignment of values. For the purpose of analyzing and interpreting the data, only the sixth or final rank order of preferences for the entire six groups was considered. Distribution of Stated Preferences for the Various Group Structures Tables XXVI and XXVII present the actual distribution of the stated preferences for the various group structures on the part of each experimental sample. A first logical refinement of the data presented in Tables XXVI and XXVII was to compute the mean rank that each group structure received from the two experimental populations and then compare them with each other. An ideal statistical treatment of the differences between the means would consist in establishing critical ratios between each pair. However, inasmuch as the distribution of rank choices TABLE XXVI DISTRIBUTION OF PREFERENCES FOR VARIOUS GROUP STRUCTURES BY NON-AUTHORITARIAN GROUP -_ -: 18h Group Rank Order of Preference Structure First Second Third Fourth Fifth Sixth Total A 3 3 9 21 22 27 85 B 7 8 l7 17 20 16 85 C 35 3o 12 7 1 85 D 27 30 11; 9 2 3 85 E 11 12 22 10 18 12 85 F 2 2 ll 21 22 27 85 TABLE XXVII DISTRIBUTION OF PREFERENCES FOR VARIOUS GROUP STRUCTURES BY AUTHORITARIAN GROUP Group T Rank Order of Preference I To; Structure First Second Third Fourth Fifth Sixth A 21 22 19 12 7 )1 85 B 10 18 27 111 10 6 85 o 36 26 1h 5 3 1 85 D 12 114. 16 30 9 b, 85 E 6 17 110 15 85 F 2 2 3 7 16 55 85 Ag 185 were governed somewhat within the restrictive framework of a limited or closed-end range of choices rather than the open-ended range of choices which is normally implicit in computing "t' scores, the interpretation of the derived critical ratios would have to be somewhat broadened. In other words, rather than setting a point at 2.66 for re- jecting the null hypothesis at the l per cent level of confidence for samples of this size, one should probably demand an index higher than 2.66 in order to compensate for this factor. Since there exists no presently available formulae to establish the exact amount of compensation that should be added to the critical point of 2.66, if any, one is restricted to attaching significance to only those critical ratios which are substantially in excess of that point. Table XXVIII summarizes the mean rankings, the dif- ference between the means, and the critical ratios between the two. An inspection of the critical ratios which are listed in Table XXVIII indicates that for five out of the six 't" scores, the critical ratios are of such magnitude beyond the point 2.66 that they cannot be dismissed without attach- ing some significance to them. On the basis that these ”t” scores do have relative significance, there is indication that the non-authoritarians have considerably less stated preference than the authoritarians for Group Structures A 186 mb.o+ ec.ma m.a mn.: m.H mm.m a me.:s mn.oe c.H em.m m.H :m.: m nmc.:+ an.o+ m.H ~m.m :.H em.m o m:.o+. mo.o* an. ma.H m.a Ho.m o :e.m- No.0- m.H we.m :.H oa.m m co.a- No.a- m.H He.: :.H ce.m < Ad ooz : 4v room noes Noam one: oapmm ence: ca ooaeea>on sedan» nofipcfipon nodes» oasuosaum Haoapaau ooooeomman peepoohm aaaonpSsanoz peeomspm nuaonpsd noose wmmDBoDmem mbomo mDOHm¢> mom mmozmmmmmmm zMN mamen sodas» ooapefibon means» onopooapm Hoonpaao cocoaomuan whoomopm Iaeonpooinoz oeoocepm nanonpfid macaw mmmbaobmam mbomc QHmH¢m mom mMozmmmmmmm zams zahmaeHmomena-zoz-oz< enemaeHmomeoa no zomomeasOo KHNN mqm¢e 189 possessed individual significance resulted in an even more enhanced significance when viewed in the aggregate. It seems reasonable to interpret these differences in per- formance in the light of some real variable that is in- fluencing the preferences of the two experimental groups. These preference differences may be portrayed.more graphically by constructing two frequency polygons based on the mean ranks assigned to the various group structures by the two contrasting populations. Figure 1 presents this data. The vertical axis represents the six levels of rank- ings: the horizontal axis represents the six different group structures that are being ranked. An inspection of the two frequency polygons indicates that in relation to each other, there is a skewing of the authoritarian curve toward the left in the direction of the leader-centered Group Structures A and B, whereas the non-authoritarian curve is skewed to the right in the direction of the group-centered Structures D, E, and F. Both curves, however, regress to an apparent common mean in the middle at point C. Group Structure C, it will be recalled, had been devised with the intent to describe a structure that imitates "group centeredness", but was not to be actually a group-centered structure. A Both Table XXVIII as well as Figure 1 seem to indicate that there are significant differences in the stated prefer- ences of the two experimental populations for the highly 190 Rank Order 1st _1_ Straight line - Authoritarian 1.5 Dotted line - Non-authoritarian 2d 2.5 3d 3.5 hth 11.5 5th S-S 6th A B C D E F Group Structures Figure 1. Graphic Comparison of Mean Preferences for Various Group Structures structured group situations represented by StrUctures A and B, and for the loosely structured group situations represented in Structures D, E, and F. There seems to be a strong prefer- ential factor on the part of the authoritarians for the more rigidly structured groupings, and a converse preferential factor on the part of the non-authoritarians for the less rigidly structured groupings. 191 From a statistical viewpoint, one must investigate whether these mean rankings were produced by chance or under the influence of a definite variable. One index would be to ascertain the extent of homogeneity within each group concerning the extent of agreement which each individual had for the rank order of preferences stated by the other mem- bers of his selected sample. The null hypothesis would be that in the absence of a homogeneous factor, there should be no particular consistency of agreement between one another; but conversely, if the group were absolutely homogeneous, then there should be perfect consistency. Kendall, in his book ”Rank Correlation Methods" (h7:81) presents a statistical method for investigating this factor of group homogeneity by computing an index knows as the co- efficient of concordance. The formula for this coefficient is: W I 12 S m (n3 - n) in which "W" represents the coefficient, "m” represents the number of observers, and ”n" represents the number of rankings. The value of ”S”, however, must first be derived from the following formula: 3 12614—1“ 2‘92 192 in which "S" represents the sum of the squared deviations from the expected value, "t1" represents the sum of all ranks assigned to a particular or "i"th object, ”m" repre- sents the number of observers, and "n" represents the number of rankings. If a texture of homogeneity is an influence that assists a group in making concordant choices, then three groupings should be compared on this index. One might computsthe coefficient of concordance on each of the experimental groups, and then compute the coefficient on both groups added together. In investigating the non-authoritarian group, one first derives "S" from the stated formula, and then makes appropriate substitution in the formula: V' = 12 S m2 (n3 - n) which appears as: w = 1g (11838110-952301'1" = .386 852 (63 - 6) The authoritarian group is then computed by first deriving its "S” and.then substituting as follows: W - 12 £53.228e502; or w = ch00 852 (63 - 6) Adding both groups together, "S" is derived as above, and substitution is made as follows: w - 12 (156,820) . : 2 3 . or W .310 170 (6 - 6) 193 By comparing each sub-group with the total group, it is immediately apparent that the sub-groups are more homo- geneous in their choices than the total combined groups. we might investigate whether these several "W's" are sig- nificant by testing them against the standard Fisher Z dis- tribution as follows. For the non-authoritarian group, employing the standard formula, we obtain: N I 1/2 1ch (In-12 w - 1.98 1-w Vl = [1-1 "’ g = n.98 For the authoritarian group, the computation will appear as: Z . 2e01; V1 . (1.98; VZ - “.23e3 and for both groups together, the computation will appear as: Z = 2.16; V1 = h.993 V2 = 758.8 In the Fisher Z distribution, the null hypothesis is rejected at the l per cent level of confidence when Z is greater than .6h72 for a group of 85, or greater than .5522 for a group of 120. Therefore, since the three groups under investigation have considerably higher coefficients than either of these two points, then it may be concluded that the 2's in question are quite significant. Accordingly, it may be concluded that the coefficients of concordance for . 19h the three groups may be tested by the Z distribution with favorably significant results. Summary An analysis of the differences in stated preferences for various group structures on the part of two experimental samples has been presented in this chapter. Mean stated preferences for a group of 85 "non-authoritarians” seem to have directional inclination for group structures which are more plastic, less rigidly organized, and in which there is a maximum opportunity on the part of each.member for direct participation in intra-group relationships. Conversely, mean stated preferences for a group of 85 individuals who were relatively high in ”authoritarianism" were directional in favor of group structures which possess more rigid and definite organization, and in which the audience group plays a passive role, with little opportunity for participating in intra-group relationships. These mean preferences were statistically treated, first, by computing critical ratios between the two experimental samples on each structure. The largest variation between the two populations was with regard to their preference, or lack of preference, for Group Structure A, which, by opera- tional definition, was described to be the most rigid group structure in the series. The non-authoritarians had a negative preference for this grouping, compared to the 195 authoritarian sample as expressed by a critical ratio of -9.06. To a lesser degree they had a negative preference toward Group Structure B, when compared with the authori- tarian sample, by the evidence of a critical ratio of -3.6h. For Group Structure 0, which was meant to be a pseudo type of group discussion, there was no particular critical differ- ence between them. For the more plastic group structures D, E, and F, the critical ratios are positively oriented in the direction of the non-authoritarian sample with critical ratios of +h.65, +h.h3, and +3.68 respectively. By pairing the more rigidly organized groupings, A and B, the combined critical ratios of these two structures yields a negative index of ~8.66 which rather conclusively argues that there is a strong differential factor between the preferential attitudes of the two populations that did not arise from pure chance. Likewise, in combining the audience-centered groupings identified as D and E, the large critical ratio of +5.9h urges a parallel and supporting conclusion. Since the 't' scores so compared in the foregoing may have been partially modified by the fact that the distribution of choices or rankings was under the restriction of a closed- end distribution, it was thought prudent to test the data , with other statistical analysis. Accordingly, each group was analyzed to the extent with which it possessed internal 196 homogeneity in its choice making performance. An index known as the coefficient of concordance was obtained on each group and then compared with the same index of the two groups added together. The coefficient for the non- authoritarian group was found to be .386: for the authori- tarian group it was .hOO; and for both groups together it was .310. This yielded evidence that the sub-groups were more homogeneous than the aggregate group when these indices are related to each other. Each of these indices was then tested for its significance on the Fisher Z distribution with excellent statistical results. It may, therefore, be concluded that each population sample expressed with some validity its pattern of prefer- ences for group structures, that these preferences were different, and that their direction was predictable. 197 CHAPTER X SUMMARY, CONCLUSIONS, AND IMPLICATIONS FOR RESEARCH Summary This is a study that has been essentially concerned with conference processes and convention "engineering”. It was designed to contribute an aspect of information that might particularly guide conference planners and group leaders in the appropriateness of their choice and utiliza- tion of various group processes and structural arrangements. In essence, it has attempted to demonstrate a causal re- lationship between a personality syndrome known as ”authori- tarianism’ and the type of attitudes that individuals will manifest toward certain group structures that are typically encountered in modern conference settings. In this study it was predicted that individuals who particularly possessed the authoritarian type of personality would be individuals who would particularly prefer group structure and conference arrangements of the highly rigid, leader-dominated type. In contrast to these individuals it was predicted that individuals who would be particularly "low" in authoritarianism would be those who would typically prefer group structure and confer- ence processes of the more group-centered type in which a 198 maximum of participation and democratic responsibility would be assumed by each member of the group. The approach to this problem was to conduct a causal comparative study of two groups who were presumed to be selectively similar in all aspects except that of the de- gree to which the one was imbued with authoritarian char- acteristics and the other was relatively free of such. These two groups were compared with regard to their characteristic preferences and lack of preferences for various kinds of group organizations that they had experienced in conference settings, ranging from the completely spontaneous or non- structured groupings that occur randomly in the convention corridors and lounges, through the semi-structured groupings under democratically planned processes, and into the most authoritarian or extremely structured official meeting. In the absence of agreement upon some of the constructs upon which this study was conditioned, certain operational definitions were proposed in order to develop a common frame of reference. These were particularly in the area of the nature of the authoritarian personality and the method- ology by which it was to be assessed. The research population was composed of 3h0 vocational rehabilitation counselors. These individuals represented a sample of the most experienced conference participants who were on the staffs of vocational rehabilitation programs in 199 each state throughout the country. The design of the study was to array these individuals along a continuum ranging from most authoritarian to least authoritarian. Then, using the appropriate quartile points as out offs, the top and bottom groups were to constitute the contrast- ing experimental samples. These were to be studied in re- lation to the differential attitudes that they might demon- strate toward the remainder of the study materials. Quan- titatively, these represented the eighty-five individuals who were found to be highest in authoritarianism as opposed to the eighty-five individuals who were found to be lowest in authoritarianism. The instruments employed in the study were of two kinds. First in order was the instrument selected to assess the personality variable of authoritarianism. This was the "F" scale produced by the research of Else Frankel-Brunswik and her collaborators. This scale is more extensively described in Chapter II of this study. Second was the development of six sub-scales which comprised the remainder of the research questionnaire. These consisted of six stereotypes of group structures, ranging from the most authoritarian and rigid type of leader-centered structure down through the most plastic and group-centered type of structure. Below each stereotype was a scale of ten attitudinal statements relating to some aspect or dynamic of the group process in qiestion. 200 These aggregated to sixty items for the six sub-parts of the scale. In addition, an extra item of a different character was added to each sub-scale. This consisted of a spiral system of progressively ranking the various stereo- types until a final rank order of group structure preferences could be evolved. lith the cooperation of the directors of the various state programs for vocational rehabilitation, all of these questionnaire materials were placed in the hands of the selected research population. Ordinarily this activity occurred in conjunction with a staff conference conducted by a district supervisor. Following the initial orientation, the supervisors at the local level encouraged the completion of the questionnaires and then mailed them in. This data gathering activity was spread over the period from July 1, 1953. to December 31, 1953. This data was then analyzed into the findings of this study. Findings The findings resulting from the analysis of the data were concerned.with the differential performances of the two experimental groups in terms of their attitudes toward various aspects of the six different types of group struc- tures. In addition, a comparison was made of the rank order 201 of preferences in which each sample arranged these six stereotypes. These findings are presented as follows: 1. Grogp Structure A, This stereotype was constructed to represent the most rigid of the six structures. It repre- sents a speaker-dominated structure in which the audience group plays no particular role of influence; rather, the audience group's main function is to remain passively silent, and to absorb as much information as it can. On the ten item.aub-scale relating to this group structure, the non- authoritarian sample responded to nine of the attitudinal statements with smaller mean scores than those obtained from the authoritarian sample. As in each of the analyses for the six stereotypes, the smaller scores signify atti- tudes that reject rigidity in group processes, and favor the more plastic and democratic arrangements. The relatively higher scores signify the reverse: positive attitudes toward the rigidity aspects of group structure, and negative atti- tudes toward the less-structured arrangements. That these differential performances characteristic of the two samples were significantly different was found in an inspection of the critical ratios that were obtained on each performance, item by item. Eight of these critical ratios were found to be beyond the one-tenth of l per cent level of confidence. Over-all, the total performance of the two groups on the ten items was found to be represented by a 202 critical ratio of 8.58, or again a ratio beyond the one- tenth of 1 per cent level of confidence. 2. ggggp Structure E. This stereotype was constructed to represent a first modification in the rigidity axis of the group structure series. Though as in Structure A it still represents primarily a speaker-centered structure, with a role of considerable passivity assigned to the audience group, nonetheless the group is allowed to exert some in- fluence over the speaker. The arrangement is such that the speaker is motivated to attempt to capture and hold the in- terest of the group and, within limits, he is prepared to modify his presentation of material in order to do so. On the subscale relating to this stereotype, the non- authoritarian sample responded with smaller mean scores than the authoritarians on seven out of the ten items. In two of the remaining three items, both groups exhibited the same level of responses. In analysing the data to determine how significant these differential performances were, it was found that five of the critical ratios were of important significance, with four of these being of significance at beyond the one-tenth of l per cent level of confidence. Over-all, the total dif- ferential performances of the two groups on the ten items were found to have a critical ratio of 5.37, so that the dis- crimination power of the more efficient items were observed to possess an aggregate weight of considerable significance. "203 3. 93333 Structure 9. This stereotype was constructed as a further modification in group-structure rigidity. Its purpose was to illustrate a synthetic group-discussion proc— ess without actually presenting the true group-centered ar- rangement of modern design. In this structure a number of preselected experts appear before the audience group and conduct a kind of group-participation process among them- selves. The audience group, however, still plays essentially the same passive role as in the two previous arrangements. On the subscele relating to this stereotype the non- authoritarian group produced eight out of ten mean responses that were smaller in weight than those produced by the authoritarians. Of the other two performances, one score was found to be identical for both groups. The data was further refined into critical ratios in order to determine the significance of these characteristic differences. It was found that seven of the ten indices were statistically significant in indicating that the con- trasting performances did not occur as the result of chance. In six of these the ratio was found to be statistically significant at beyond the one-tenth of 1 per cent level of confidence. The total differential performance of the two groups were found to possess a critical ratio of 5.52. h. ‘ggggp Structure 2. This stereotype was constructed to represent a reduction of the rigidity and speaker-centered lines of authority to the point wherein the center of gravity 20h became group-centered. In this process the group employs a leader only as a guide to help itself in its own organiza- tion. This activity is ordinarily conducted along democratic lines with every member sharing in the responsibility and the participation. On the subscale relating to this stereotype the non- authoritarian group obtained smaller mean scores than the authoritarians in nine out of the ten cases. With regard to the single exception, both groups produced the same general level of response. Upon refinhng the data into critical ratios pertaining to the differential performances of the two groups, eight of these were found to have important statistical signifi- cance, with seven of these being at or beyond the one-tenth of l per cent level of confidence. The total differential performance of the two groups to the ten items was found to be signified by a critical ratio of 8.50, or significant at well beyond the one-tenth of 1 per cent level of confidence. 5. {£3232 Structure g. This is a stereotype that was designed to illustrate an extreme process of group-centered- ness in which the reduction in rigidity and organization would be pushed close to the maximum.limit. The group ar- rangement in question is more popularly known as the ”leader- less” discussion-group process. By construction, the group arrangement is one that is devoid of any power figure or 205 group leader to whom the group can immediately turn for direction. Rather they are required to assume these re- sponsibilities themselves. On the ten-item scale relating to this stereotype, the non-authoritarians obtained smaller mean scores than the authoritarians in every instance. When the data was examined in terms of the critical ratios that obtained between these differential performances, it was found that nine of these indices were of statistical significance. Eight of these nine were found to be signifi- cant at a point beyond the one-tenth of l per cent level of confidence. The total differential performance of the two experi- mental samples to these ten items was found to be represented in a critical ratio of 9.51, or statistically very signifi- cant at a point beyond the one-tenth of l per cent level of confidence. 6. £3922 Structure 2. This stereotype was intended to portray informal group arrangements of a purely spontane- ous nature in which no prior "structure" as such could be said to be characteristic of the manner of group organization. These were described as the informal groupings that spring up in the corridors and lounges between meetings; they may change, disappear, and sometimes re-appear again. In essence they were to represent the maximum in group fluidity rather than group rigidity. 206 For the ten-item scale relating to this general stereo- type, the non-authoritarians obtained smaller mean scores on nine of the items; with reference to the single exception the two experimental samples demonstrated practically identical performances. Upon refining these differences into critical ratios, it was found that seven of these indices were quite signifi- cant; five were found to be beyond the one-tenth of l per cent level of confidence, and two at about the l per cent level of confidence. Over—all, the total differential per- formance of the two groups to the ten items was found to be signified by a critical ratio of 5.06, or well beyond the one-tenth of l per cent level of confidence. 7. Eggg‘ggggg.g£ Preferences. In addition to the six subscales that were appended to each group-structure stereo- type, an added item of a different character was included as a terminal question. This consisted of a problem in progressively assigning a rank order of preferences to the various stereotypes under consideration until, at the con- clusion of the questionnaire, a final rank order of prefer- ences for the series of six stereotypes was Obtained. These rankings were arithmetically averaged for each group. In the over-all rankings the two groups were found to each possess a contrasting pattern of preferences. Group Structures A and B were substantially more preferred in the ranking system by the authoritarians than by the non- 297 authoritarians. Group Structures D, E, and F were sub- stantially more preferred by the non-authoritarians than by the authoritarians. With the progressive reductions in rigidity of organization that characterized each change in the six group stereotypes, the compromise or regression point was found to occur at Group Structure C wherein the authoritarian and non-authoritarian groups were found to be most congruent in their choices; and as such, this stereo- type was found to possess the top order of ranking for each experimental sample. The statistical data which more clearly delineates the differential preferences of the two samples may be best observed by referring to the more elaborate treatment that it is given in Chapter IX. Conclusions This was a study that had been predicated upon the following'hypotheses= 1. £212; Hypothesig. Preferences for structured and unstructured groups are a function of the degree of authori; tarianism. 2. Eiggg‘fiypotheses. Highly structured groupings are more preferred by the authoritarian personality. a. Organized groupings with a minimum.of structure are more preferred by the non-authoritarian personality. 208 b. A greater unity of attitude in favor of spontaneous groupings is more characteristic of the non-authoritarian personality. c. Unorgmnized or spontaneous groupings engender the greatest ambivalence of attitude among the authoritarian personalities. The findings immediately reported in the foregohng dis- cussions seem to rather conclusively indicate that the two experimental samples exhibited a consistently opposed atti- tudinal response pattern toward the various group structures described in the study. Inasmuch as the two experimental samples were drawn from the same research population, but were selected on the basis of a single difference in the personality variable of authori- tarianism, it would seem plausible therefore to relate this consistent difference to the influence of this particularized variable. If such be the case, the major hypothesis of this study seems to be well founded. In order to substantiate the minor hypotheses referenced above under a, b, and c, the study specified direction in predicted behavior for the two groups. In a weighting scale in which the smallest numerical answers were to represent responses keyed in the direction of preferences for non- rigidity in structure, and the higher numbers to represent preferences for rigidity in group structure, these hypotheses 209 would be confirmed in accordance with the degree to which the non-authoritarian sample would consistently score with mean responses of a smaller quantitative value than would the authoritarians to the same item. The findings reported in the preceding demonstrated that such did indeed occur as a consistent pattern. For the sixty items in the six subscales, the non-authoritarians obtained a smaller mean score than the authoritarians in fifty-two instances; Of the remaining eight exceptions, five of these were noted to consist of relatively equal responses by both groups, and in only three instances did the authoritarians exhibit a definitely smaller mean response to a given item. In addition, the rank order of preferences characteristic of each experimental sample (described in point 7 of the find- ings presented in this chapter) also gave rather clear in- dication that on a comparative basis the authoritarian sample had much more positive attitudes toward the highly rigid Group Structures A and B than did the non-authoritarians, and much.more negative attitudes toward the highly unstructured stereotypes of D, E, and F than did the non-authoritarians. Therefore, the substantiation of these hypotheses becomes a further conclusion of this study. The establishment of minor hypothesis "c” (that spon- taneous groupings would elicit the greatest amount of divided opinion or ambivalence in attitude among the authoritarians) was examined in this study in relation to the content items 210 of some of the key items in the subscale pertaining to Group Structure F. As may be observed in examining Chapter VIII of this study, the authoritarians demonstrated three characteristic differences in their sample as compared with the non-authoritarians which.were anticipated to bear upon the assumption in question. These were: that the authoritarians contained more individuals in their sample than the non-authoritarians who were prone to use these spontaneous group situations as opportunities for the gratification of special authoritarian needs; that they contained more individuals in their sample of the "lone- wolf" type who seek to avoid entry into these spontaneous groupings; and that they contained more individuals who were desirous of entering such groupings but who, for vari- ous reasons, represent the population who are typically left out of such groupings. The significantly greater possession of these rather contrasting pools of opinion by the authoritarians become the basis for the final conclusion that authoritarians are more ambivalent in attitude toward the spontaneous groupings than are the non-authoritarians. Recommendations This study seems to have practical import in guiding conference planners and group leaders in their choice and utilization of various group processes as a communication procedure. 2ll Rather than basing conference arrangements upon the employment of a basic group-process technique it suggests that a variety of techniques should be employed if the various personality needs are to be apprOpriately satisfied. Attention is directed to the influence of the personality syndrome known as "authoritarianism". To the degree that this personality variable is highly characteristic of a given pro- fessional grouping, the data suggests that it would not be unsound to organize the conference procedures around the more rigid type of group structures. Conversely if the given professional grouping be characterized as quite non-authoritarian in personality make-up, the most rigid type of group structuring procedures might be expected to possess a minimum of appeal to them. Since most large groupings would likely possess a range of distribution in a given personality variable, the widespread employment of either extreme of group-process technique would be most effective only for those individuals who possessed similar extremes in personality organization of the type described in this study. Since both experimental samples exhibited a common satisfaction with the group structure stereotype designated in the study as Group Structure C, to the degree that this was the most popular choice for both, it would appear that this compromise process (representing semi-rigidity in group 212 structuring) might more logically represent a basic group- processing standard from which appropriate departures might be made. With respect to the continuing development and utiliza- tion of the more democratic procedures in group structuring the following might be said. Since the theoretical founda- tions of the authoritarian syndrome includes a factor in which ”traditionalism’ has an important valence for attrac- ting the authoritarian element, then it might logically follow that if a given professional community pursues the use of "unstructured" group processes over a long enough time, and under the endorsement of its respected leaders, the authoritarian personality as such would eventually ac- cept the employment of such techniques as the ”proper" kind of communication vehicle, and perhaps would direct its resistance to such other "unconventional" techniques as would be then in.experimental try-out. Implications for Research The following approaches to further research seem to be implied by the findings of this study in order to cor- roborate the implications of the data and to extend the scope of the investigation. 1. This study was concerned with the evidence produced by a particular professional population at a particular point 213 in time. A follow-up study of the same population at a later point in time might be appropriate as a check on the stability of the evidence produced. 2. Comparative studies involving other populations 'would appear to test further the extent to which univer- sality might be applied to the conclusions attached to the instant study. 3. The "F" scale employed in this study to array the research population in terms of the degree to which each is possessed of ”authoritarianism" is noted to divide into nine subscales, each one of which is keyed to a particular com- ponent of the authoritarian syndrome. The possibility is immediately apparent that the collected data could be re- treated in nine separate substudies with the view of as- certaining the particular significances that each component in the authoritarian syndrome might possess in relation to the rest of the obtained data. h. In analyzing the performances of the two experi- mental groups to the various items on each of the subscales, it was noted that the discrimination power of each item was not always of important statistical significance. Further research is suggested in the area of refining some of these items as well as in experimenting with new ones. Similarly, further research in the area of conducting studies on other 21h types of group stereotypes and processes provides broad opportunity for additional areas of research. 2. 9. 10. 11. 12. 215 BIBLIOGRAPHY Adorno, T. W., E. Frankel-Brunswik, D. J. Levinson, and R. N. Sanford. The Authoritarian Personality, New York: Harper & Brothers, 1950. 990 pp. Ansbacher, H. L. "The History of the Leaderless Group Discussion Technique," Psychological Bulletin, h8 , No. 5, September, 1951: 383-391. Ash, Philip, and Edward Abramson. "The Effect of Anonymity of Attitude-Questionnaire Response,“ Journal ‘pg'Apnormal and Social Psychology, h7:722-723, 1952. Bass, Bernard M., and Fay-Tyler M. Norton. “Group Size and Leaderless Discussions, Journal pg Applied Psy- cholo , 35:397-h00, 1951. Bass, Bernard M., and Stanley Klubeck. "Effects of Seating Arrangement on Leaderless Group Discussions," Journal 2; Abnormal and Socig; P_ychology, h7: 72h-727, 1952. Bell, Graham B., and Robert L. French. "Consistency of Individual Leadership Position in Small Groups of Vary- ing Membership,” Journal of Abnormal and Social Psy- chology, h5: 76h-737, 19 50. Berkowitz, Leonard. "Some Effects on Leadership Sharing in Small, Decision-Making Conference Groups," (Abstract of Ph. D. Thesis, University of Michigan, 1951. 153 pp. ). Bernstein, S. "There are Groups and Groups,” Group 13 Bettelheim, B., and M. Janowitz. amics pf Prejudice, New Ybrk: Harper and Brothers, 193%. 227 pp. Bion, W. R. "Experiences in Groups: V.," Human Relations, 3:3‘111-9 1950- Bion, W. R. "Experiences in Groups: VII.," Human Relations, h:221-227, 1951. Bovard, Everett W., Jr. "The Experimental Production of Inter-Personal Affect," Journal of Abnormal and Social Psy_hology, h6. 521- 523, 1951. 13. 11,. 15. 16. 17. 18. 19. 20. 21. 22. 23. /25. 216 Bradford, Leland P. ”Leading The Large Meeting," Adult Education Bulletin, December, 19h9. pp. 38-50. Cantor, Nathaniel. "Focus and Function in Group Dis- cussion," Teachers' College Record, 53:375-382, 1952. Cantor, Nathaniel. "Skills in Conference Leadership," Personnel Journal, 31:135-137, 1952. Carter, Launor F. "Some Research on Leadership in Small Groups," in Guetzkow, H., Groups, Leadership and Men, 1h6-157. (See 35) Carter, Launor, William Haythorn, Beatrice Shriver, and John Lanzetta. "The Behavior of Leaders and Other Group Members," Journal of Abnormal and Social Psychology, ‘ h6=589-S9S. 19§f7"' , Chowdhry, Kamla, and Theodore M. Newcomb. "The Relative Abilities of Leaders and Non-Leaders to Estimate Opinions of Their Own Groups," Journal 9; Abnormal and Social Psychology, h7:51-57, 1952. Courtney, Douglas, F. Loyal Greer, and Joseph M. Masling. Leadership Identification ang Acceptance, Philadelphia: Institute for Research in Human Relations, 1951 (Rep. No. l. 39 pp. Courtney, Douglas, F. Loyal Greer, and Joseph.Masling. geadership Ideptification and Acceptance, Philadelphia: Institute for Research in Human Relations, 1952 (Rep. No. 1.) lhO pp. Darley, John G. "An Overview of the Conference and Its Controversies," in Guetzkow, H., Grou s, Leadership, and Men, 257-266. (See 35) Deutsch, M. ”Task Structure and Group Process," American Psychologist, 6:32h-325, 1951 (Abstract). Dulsky, Stanley G. ”Improving Conference Leadership," Personnel, 27:189-193. 1950. Festinger, Leon, and John Thibaut. "Interpersonal Com- munication in Small Groups," Journal 2; Abnormal and Social Psychology, h6:92-99. 1951. Festinger, L., A. Pepitone, and T. Newcomb. "Some Consequences of De-Individuation in a Group," Journal 2; Abnormal and Social Psychology, h7:382-389, 1952. 26. 27. 28. 29. 30. 31o 32. 33. 3h. 36. 37. 38. 39. 217 Festinger, Leon. ”Informal Communications in Small Groups," in Guetzkow, H., Groupg, Leadership and Men, 28-h3. (See 35) Fouriezos, Nicholas T., Max L. Hutt, and Harold Guetzkow. nMeasurement of Self-Oriented Needs in Discussion Groups," Journal 2: Abnormal ppd Social Psychology, h5:682-690, 1950. French, John R. P., Jr. "Group Productivity," in Guetzkow, H., Groups, Leadership and Men, hh-Sh. (See 35) French, R. L. "Verbal Output and Leadership Status in Initially Leaderless Discussion Groups," American Psychologist, 5:310-311, 1950 (Abstract). Gibb, Cecil A. "The Research Background of an Inter- actional Theory of Leadership," Australian Journal pf Psychology, 2:19-h2, 1950. Gilchrist, J. D. "The Formation of Social Groups Under Conditions of Success and Failure," Journal pf Abnormal and Socig; Psychology, h7:17h-187, 1952. Goode, Cecil B. "Significant Research on Leadership," Personnel, 27:3h2-350, 1951. Gordon, Leonard V. "Personal Factors in Leadership," Journal pf Social ngchology, 36:2h5-2h8, 1952. Gouldner, Alvin W., (Ed.) Studies lp Leadership, New York: Harper, 1950. 736 pp. Guetzkow, Harold, (Ed.) Groups, Leadership, and Men; Research in Human Relations, Pittsburgh: Carnegie Press, 195I,—_293 pp. Gullahorn, John T. "Distance and Friendship as Factors insthe Gross Interaction Matrix,” Sociometry, 15:123-l3h, 19 2. Guilford, J. P. Fundamentg; Statistipg in Psychology and Education, New York: McGraw-Hill, 1950. 633 pp. Haiman, Franklyn S. Group Leadegship and Democratic Action, Boston: Houghton Mifflin, 1951. 309 pp. Hartley, Eugene. ”Psychological Problems of Multiple Group Membership," in Rohrer, J. H., and.M. Sherif, Social Ps cholo pp ppg Crossroads, 371-387. New York: Harper, 1951. E37 pp. ”£6. #7. m3. #9. 50. 51. 52. 218 Hemphill, John K. "Situational Factors in Leadership," (unpublished Ph. D. Thesis, University of Maryland, 19h8). Hemphill, John K. "Relations Between the Size of the Group and the Behavior of 'Superior' Leaders,” Journal 2; Social Psychology, 32:11-22, 1950. Hollander, Edwin Paul. "An Investigation of the Re- lationship Between Authoritarian Attitudes and Leader- ship Selection in a Military Setting," (unpublished Ph. D. Thesis, Columbia University, 1952). Holtzman, Wayne H. "Adjustment and Leadership: a Study of the Rorschach Test,” Journal 2; Social Psychology, 36:179-189. 1952. " " ' Hyman, H. H. ”The Psychology of Status," Archives pf Psychology, No. 269, l9h2. Jenkins, W. O. "A Review of Leadership Studies with Particular Reference to Military Problems,” Psychological Bulletin, on: Sh-79. 19h?- Jones, Edward. ”The Role of Authoritarianism in the Perception and Evaluation of a Prospective Leader," American Psychologisp, Vol. 8, No. 8, p. 375, 1953 (Abstract). Kendall, Maurice G. Rank Correlation Methods, London: Charles Griffin and Company, Ltd., 19h8. Leslie, Robert C. "Growth Through Grou Interaction,” Journal 2; Pastoral Care, 5 (Spring): 3 -h5, 1951. Livingston, Donald Gene. "The Effects of Varying Group Organization Upon Perception of Power and Benefit,” (ugpublished Ph. D. Dissertation, University of Kansas, 19 1 . Maier, Norman R. F., and Allen R. Solem. "The Con- tribution of a Discussion Leader to the Quality of Group Thinking: the Effective Use of Minority Opinions," Human Relations, 5:277-288, 1952. Meier, Norman R. F. "The Quality of Group Decisions as Ingluenced by the Leader,” Human Relations, 3:155-l7h, l9 0. Marquis, D. G., Harold Guetzkow, and R. W. Reyna. "A Social Psychological Study of the Decision-Making Conference," in Guetzkow, H., Grou s, Leadership, spg ‘Mpp, 55-67. (See 35) 53- Sh- 55. 56. 57e 58. 59. 60. 61. 62) 63. 6h. 65. 219 Martin, William E., Neal Gross, and John G. Darley. ”Studies of Group Behavior: Leaders, Followers, and Isolates in Small Organized Groups," Journal of 5p- normal and Social Psychology, h7:838-8K2, 1952. Messing, P. W. Rehearsal For Destruction, New York: Harper and Brothers, l9K-. 351 pp. McCurdy, Harold G., and Wallace E. Lambert. "The Efficiency of Small Human Groups in the Solution of Problems Requiring Genuine Co-operation, Personnel, 20:h78-h9h, 1952. " Journal 2; Mood, Alexander M. Introduction to the Theory of Statistics, New York: McGraw-HilI: 19§o. Moreno, J. L. "A Note on Sociometry and Group Dynamics," Sociometry, 15:36h-366, 1952. Newcomb, Theodore M. "Social Psychology and Group Processes," Annual Review 2; Psychology, h:183-21h, 1953. Ratcliffe, T. A. ”The Function of Discussion Groups," Bulletin of World Federation p£,Menta1 Health, A (2): 59‘6- E‘. 7935. Richardson, Helen M., and Nelson G. Hanawalt. ”Leader- ship as Related to the Bernreuter Personality Measures: V. Leadership Among Adult WOmen in Social Activities," Journal 2; Social Psychology, 36:1hl-153, 1952. Rokeach, Milton. "Generalized Mental Rigidity As a Factor in Ethnocentrism," Journal‘pp Abnormal and Social Psychology, h3:259-278, 19KB. Sanford, Fillmore H. Authoritarianism snd Leadership: ‘5 Study p§_the Follower's Orientation pp Authority, Philadelphia: Institute for Research in Human Relations, 1950. 189 PP. Sanford, Fillmore H. ”Leadership Identification and ’ f) Acceptance,” in Guetzkow, H., Grou s, Lsadership, and.téglet. ” Mpg, 158-176. (See 35) Scheidlinger, Saul. Psychoanalysis and Group Behavior: 2 Stud in Freudian Group Psychology, New York: Norton, 1952. 2&3 pp- Semeonoff, Boris. "On the Reliability of the Leaderless Group Discussion Technique, h9:5h0-5hl. 1952. Psychological Bulletin, 66. 67. 68. 69. 70. 71. 72. 73- 7h. 75. 76. 77. 78. 220 Sewell, Walter E. "The Structure of Leadership," Military Review, Ft. Leavenworth, 32:62, 1952. Sherif, Muzafer. "A Preliminary Study of Inter-Group Relations,” in Rohrer, J. H., and M. Sherif, Social Psychology at the Crossroads, 388-h2h; New York: Harper. 195T? E37 pp- Sherif, Muzafer. and O. J. Harvey. "A Study in Ego Functioning: Elimination of Stable Anchorages in Individual and Group Situations," Sociomstpy, 15: 272-305, 1952. Smith, M. Brewster. "Social Psychology and Group Prgcesses,” Annual Review 2; Psychology, 3:175-20h, 19 2. Sterling, Theodor D., and Bernard G. Rosenthal. ”The Relationship of Changing Leadership and Followership in a Group to the Changing Phases of Group Activity," American Psycholo ist, 5:311, 1950. (Abstract) Stogdill, R. M. "Personal Factors Associated With Leadership,” Journal p; Psychology, 25:35-71, 19h8. Stone, W. L. "Youth Leaders Are of Three Types," Group, 15 (2):23-2u, 1952. Taylor, Lawrence J. "Conference Dividends in Dollars agd Sense,” College and University, April, 1950, h56- h 2. Titus, C. H. The Processes 23 Leadership, Dubuque, Iowa: William C. Brown Company, 1950. Williams, Richard H. Vocational Rehabilitation Personnel, National Institute of Mental Health, U. S. Department of Health, Education, and Welfare, October, 1953. (Staff Memorandum) Wittenberg, Rudolph M. The Art pf Group Discipline: 2 Mental Hygiene A roach pp Leadership, New York: Association Press, 1951. 12h pp. Whyte, William Foote. "Small Groups and Large Organiza- tions," in Rohrer, J. H., and M. Sherif, Social Psy- cholo sp,pps Crossroads, 297-312. New York: Harper, 19:1, E37 pp. Zilboorg, Gregory. "Authority and Leadership" Bulletin 2; World Federation.p£_Menta1 Health, 2 (5):l3-17, 1950. 221 APPENDIX A COPY OF ORIGINAL DATA-COLLECTING LETTER Q 9 2. I DEPARTMENT OF PUBLIC INSTRUCTION CLAIR L. TAYLOR. IUPERINTENDENT OFFICE OF VOCATIONAL REHABILITATION P. 0. BOX 928. LANSING. MICHIGAN 222 H. EARLS CORREVONT ASSISTANT I UPIRINTINDENT AND DIRECTOR August 9, 1953 Mr. Marlow B. Perrin, Director Bureau of Vocational Rehabilitation 83 S. High Street. Room 15 Columbus 15. Ohio Dear Marlow: I am soliciting the assistance of each state director of vocational re- habilitation for help in carrying out a research project. This study is to be concerned with.certain aspects of how we plan and conduct our state. regional, and national professional conferences of the National Rehabilitation Association, as well as with those other professional associations with which we so frequently participate from time to time. The gathering of the necessary research depends for its success upon the extent to which a suitable number of selected counselors can partici- pate. In order to collect the soundrst type of data it would be most preferable to have as reapondents those of your counselors who are most experienced in the particular area under investigation. Since this is a study having to do with conference processes, it would seem likely that those on your staff who participate the most in attending conference or- ganizations would be in the best position to reflect a considered opinion on the various items constituting the study materials. §s_g criterion, I would suggest that you consider your counselors as being divided into two halves, and that you would distribute the enclosed stgdy materials only among the more exuerienoad half. Using the national directory as a guide, I am enclosing study materials for approximately one half of your total staff. Each set of materials contains complete directions. I am sure your supervisors should have no trouble in interpreting to the counselors the general nature of what is being asked of each respondent. You can indeed be certain, Marlow, how very appreciative I would be for whatever assistance you and your fine staff in Ohio can give me. Sincerely, Roi-1 Q. PM Half A. Peckham Supervisor, Guidance Services RAP:mms 9.2.131. STATE or CALIFORNIA D E P A R T M E N T o F E D U c A T I o N 223 Division of Special Schools and Services BUREAU OF VOCATIDNAL REHABILITATION September 22, 1953 Mr. Ralf A. Peckham, Supervisor Guidance Services Office of Vocational Rehabilitation P. O. Box 928 Lansing, Michigan Dear Mr. Peckham: Shortly after I received your letter of September 9 enclosing questionnaires we had a staff training meeting at which I distributed a questionnaire to each of our selected professional staff members. I am sure that within a short while you will be re- ceiving them back. Best of luck in your research work. Yours very truly, Donald T. Blyth District Supervisor DWB:CG APPENDIX B "F" TEST OF AUTHORITARIANISM 22h 225 RESEARCH WORKSHEET An Attitudinal Questionnaire ********** Foreword Fellow Colleagues in Counseling: This questionnaire is being submitted to you as one phase of a research project concerned with levels of attitudes. TSince the nature of the pOpulation that you represent is a soPhisticated one in terms of familiarity with testing scales, the statistical clarity of the data that you submit for analysis will depend very frankly upon the degree to which you score each item in terms of your first emotional or attitudinal impulse, rather than to mark the item according to your intelligent estimation of what is "proper" or idealj 'Your utmost c00peration in responding to each statement below in accordance with your feelings rather than with your reason should result in an accumulation of data that could guide all of us to plan more realistically and wisely in the area of group relation- ships. In completing this questionnaire you are not required to reveal your identity. Simply return this form in the attached envelope. DIRECTIONS FOR SCORING All items are to be scored with numerical weighting in accordance with the strength or level of your agreement or disagreement with each statement. Be sure to place the plus signs in front of numbers indicating positive agreement, and negative signs before those num- bers indicating disagreement. TABLE FOR SCORING +.3 (Strong Agreement) +.2 (Moderate Agreement) ri-l (Slight Agreement) --1 (Slight Disagreement) -2 (Moderate Disagreement) -3 (Strong Disagreement) A Record each numerical score within the parentheses at the left of each item. Don't forget to use the plus and minus signs. 226 AAAAAAAA ‘ SCORE A A AA V V VV VVVVVVVV ll. 12. 13. 1h. 15. 16., 17. 18. 19. 20. 21. '22. 23. 2h. 25. -Nowadays more and more people are prying into matters that should remain .wars and social troubles may someday be ended by an earthquake or flood QEIW Awnings SCALE 1 Obedience and respect for authority are the most important virtues l children should learn. ~No weakness or difficulty can hold us back if we have enough will pcnnnq Science has its place, but there are many important things that can neVQ possibly be understood by the human mind. , Human nature being what it is, there will always be warand conflict. . Every person should have complete faith in some supernatural power whose; decisions he obeys without question. When a person has a problem or worry, it is best for him not to think about it, but to keep busy with more cheerful things. A person who has bad manners, habits, and breeding can hardly expect to get along with decent peOple. What the youth needs most is strict discipline, rugged determination, and the will to work and fight for family and country. A Some people are born with an urge to Jump from high places. - Nowadays when so many different kinds of people move around and mix togeflm so much, a person has to protect himself especially carefully against canfi ing an infection or disease from them. An insult to our honor should always be punished. »Young people sometimes get rebellious ideas, but as they grow up they ouyfi~ to get over them and settle down. , i It is best to use some prewar authorities in Germany to keep order and u prevent chaos. what this country needs most, more than laws and political programs, is a few courageous, tireless, devoted leaders in whom the people can put them? faith. a 1 Sex crimes, such as rape and attacks on children, deserve more than mere 5 imprisonment; such criminals ought to be publicly whipped, or worse. PeOple can be divided into two distinct classes; the weak and the strong. There is hardly anything lower than a person who does not feel a great love, gratitude, and respect for his parents. Some day it will probably be shown that astrology can explain a lot of things. personal and private. that will destroy the whole world. Most of our social problems would be solved if we could somehow get rid of the immoral, crooked, and feebleminded people. The wild sex life of the old Greeks and Romans was tame compared to some of the goings-on in this country, even in places where people might least expect it. If people would talk less and work more, everybody would be better off. Most people don't realize how much our lives are controlled by plots hatched in secret places. Homosexuals are hardly better than criminals and ought to be severely punished. 227 APPENDIX C ff GROUP STRUCTURE QUESTIONNAIRE .-( $26. '2 27. )9- DRE ) 26. ) 27. ) 28. ”a ) 29. GENERAL ATTITUDE SCALE 228 The businessman and the manufacturer are much more important to society than the artist and the professor. _ No sane, normal, decent person .could ever think of hurting a close friendy or relative. Familiarity breeds contempt. Nobody ever learned anything really important except through suffering. Foreword to Part II The next section is composed of six different kinds of group. situations such as may be typically encountered at the average professional conference or convention. Each group situation is described in a paragraph or two, after which there is a series of 10 statements about each. You are to respond to these statements in the same manner as you have done in Part I, using +8,.&2, or +1 to signify agreement; and 61, ‘2, or -3 to signify disagreement. NOTE As you respond to each of these six group situations you should be considering which of these you generally prefer the most, and which you generally prefer the least. Under each group situation--with exception of the first one, you are to progressively rank your preferences for one group situation over another.v Thus you will first choose between "A" and "B"; then between "A", "B", and "C"; and so on. Group Structure A The individual appearing on the rostrum at this meeting is an authority in the field of your general interest. By reputation you know that he is a very thorough and careful scientist. It has been reported to you that when he reads a paper at a convention, the subject is covered in absolute detail, and few, if any, questions are asked by members of the audience. : This is the kind of a man who seeks to address himself to a serious minded audience. He is not so much concerned with trying to speak in an entertaining manner, but rather with an absolute devotion to the facts, regardless of what they might'qu It has been his uncompromising pursuit of this goal that has made him the outstanding authority that he is. You have not attended a meeting under his direction before. ( ) 1. This is the kind of a meeting I enjoy attending. ( ) 2. As a general rule, audience members gain a great deal from listening to a speaker of this type. ( ) 3. A most sensible group arrangement is to have a speaker who can present the facts, and an audience that knows how to listen. ( ) h. It is more professionally desirable for a Speaker to stick to those facts which he can prove rather than to get off into speculations. ( ) 5. It's rather nice to be able to sit back in the audience and listen to someone talk who knows what he is talking about. ‘( ) 6. It is to be hOped that the reading of scientific papers will continue to be encouraged by the professional association. ( ) 7. Professional meetings of this type should generally be closed to non- members of the professional association. ( ) 8. -Those who possess the least professional training are usually least interested in hearing someone read a scientific paper. ( ” ) '9. Program planners would do well to eliminate the reading of scientific papers. ( ) 10. A real authority in the field should have the right to make his presentation without having to put up with constant questioning from his audience. ‘1- 1w r—v W _: .v .I: "" _n- -..-“A . , , oup Structure B e principal speaker at this meeting is known to be an engaging personality who tries to ke his talks as interesting as possible for his audience. Though he works in your general eld of interest, he has contributed no particular research to the professional journals. ‘s'reputation you know that he always speaks extemporaneously, seldom refers to notes, and yxrnerally offers no documentation for his principal statements.- Usually he deals with.in- gwresting and speculative ideas, rather than with factual data. ‘°dinarily, several members of the audience engage him in a lively question and answer period :the conclusion of his remarks. The speaker seems to be principally motivated in stimua i¢ating his audience rather than in presenting an organization of highly dependable data. You .|1ve never attended a meeting of his before. 21 ) l. vThis is the kind of meeting I enjoy attending. ) 2. As a general rule, audience members find it most profitable to listen to a 'speaker of this type. ) 3. Professional meetings of this type should generally be closed to non-members of the professional association. ) h. If the audience manifests other interests than in what the speaker is saying, he should try to re-direct his discussion toward those tOpics which particularly attract their attention. ) 5. When a speaker departs from his prepared assemblage of facts in order to attract audience interests, he is not prOperly discharging his professional obligation to report all the facts. ) 6. It is more professionally desirable for a speaker to leave factual reporting to the journals, and to employ conference time in the general discussion of whatever items are of current interest to the delegate assembly. ) 7- If a prOgram planner has to choose between one who is an excellent authority in the field but not too much of a speaker, as against one who is an excellent speaker but not too much of an authority, he should generally choose the better speaker. ) 8. A truly professional organization should not use up the valuable time of its mem- . bership with speeches by those who are least qualified to make an authoritative contribution. ) 9- Generally speaking, there is no particular place on a professional platform for speakers who hold audience attention by a presentation of ideas that are unortho- dox, unproven, and perhaps sensational in nature. ( ) 10. ,For the most part, the role of a conference audience should be to give courteous attention to the speaker, rather than to raise challenges to the soundness of his ideas, or to express contradictory viewpoints. Inserting the letters "A" or "B" in the boxes below, indicate the order of your preference between Group Structure A and Group Structure B. First Choice Second Choice 0 Group Structure C This is a panel'discussion meeting of the Chicago Round Table type in which a, group of ex- perts representative of the various interest segments of the audience will discuss a speci- ' fied topic. --These individuals are quite experienced in the panel discussion technique. They; seldom use notes or written materials but rather Speak in impromptu fashion on the subject under consideration. ~ Because of the contrasting viewpoints that they represent, the dis- cussion often becomes quite vigorous, and they raise pointed questions among themselves. - Since the questions raised by the panel experts are assumed to be similar to many of those that would cane from the audience, the majority of the period is given over to the experts on the panel. ~ However, there is usually some time allotted at the end of the discussion fort a few pertinent questions to come from the floor which the experts may have overlooked. You have not yet attended a panel meeting composed of these particular individuals. ( ( ) 1. ) 2. )‘ 3- h. ) 5- ) 6. ) 7. ) s. ) 9- ) 10. This is the kind of a meeting I enjoy attending. Conference purposes are usually best served when the panel has been carefully preselected from established experts in the field rather than when it is nude up at the time of the meeting from untested members in the audience. The average professional audience can gain more from listening to ,a discussion by a pre-selected panel of experts from the field than they can by discussing the subject among themselves. ' -'1'he general discussiOn should primarily be carried an by the audience group: using the panel experts only as resources for specialized information. It is a better panel discussion when the group obtains its information by ques- tioning and discussing with the experts on the panel, rather than by questioning and discussing with each other. In a. good panel discussion, the experts should attempt to resolve their differ- ences and come up with a single point of view. ~Top authorities in the field are professionally obligated to serve on panels of this type. ' . A sound discussion will try to stick pretty. much to the middle of the road and to avoid extremes. . Generally speaking, the panel should be composed of the respected leaders of the profession rather than of the youngest newcomers. A panel made-up of experts who practice at their profession will generally prod” 8 flounder discussion than if the panel is made up of theorists or academically minded professionals. . - a - — - a a c: - a a o 9 :3 O I, O a - I: o Inserting the letters "A", "B", or "c" in the boxes below, indicate the order of yourlflffi’ ence between Group Structure A, Group" Structure B, and Group Structure C. First Choice Second Choice "- Mr Third Choice 1 a. J Group Structure D {The discussion leader for this group meeting has a good reputation for obtaining participa- ftion.and discussion on the part of every member of the audience. The discussion leader has In: important professional knowledge to contribute to your professional field of interest, lnxt he does have professional knowledge of techniques by which he can draw opinions and con- ‘tributions from other members of the group, as well as to draw out from you some statements