LDCUS OF CONTROL, DOGMATISM, AND SATISFACTION WITH THE LEADER AS DETERMINANTS IN THE PERCEPTION 0F LEADERSHIP STYLES Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY THOMAS EDWARD HILL 1976 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII L PIE-ANN u ,9 I ABSTRACT LOCUS OF CONTROL, DOGMATISM, AND SATISFACTION WITH THE LEADER As DETERMINANTS IN THE PERCEPTION 0F LEADERSHIP STYLES By Thomas Edward Hill One of the experimental paradigms employed to investigate the superior-subordinate relationship involves the description of supervisors by their subordinates via standard instruments. These descriptions are then used to classify the superior along relevant leadership dimensions. In addition to the traditional consideration (CON) and initiating structure (IS) dimensions, a participative decision-making (PDM) dimension of leader- ship has recently been proposed. While sound theoretical arguments can be made for a PDM dimension, controversy still surrounds its operationalization. Arguments center around the lack of evidence for discriminant validity between CON and PDM dimensions. Another important issue concerning the study of leadership is the effect of subordinate related variables on the leader-follower relationship. Past research has suggested that the sub- ordinate's personality influences his or her perception of leadership style. Hence, contingency models of leadership have been advocated. The central hypothesis of these contingency models is that leadership style, subordinate variables, and situational variables are all determinants of the leader- follower relationship. Thomas Edward Hill The present study investigated the effects of subordinate related variables on perception of leadership style and the independence of the consideration (CON), participative-decision making (POM), and initiating structure (IS) dimensions of leadership. The subjects were 374 undergraduates who observed a videotape of a leader portraying either a participative or authoritarian role in a group problem solving exercise. At the conclusion of the tape, questionnaire packets designed to measure the predictors and criteria were distributed to the subjects. Perceived leadership styles, measured by the CON, PDM,C/”fl and IS scales of both the Leader Behavior Description Questionnaire and the Johnson 3-dimensional Leadership Scale, were the criteria and leader- ship style observed, locus of control, dogmatism and satisfaction with the leader were the predictors. The data were anlayzed by multivariate and univariate multiple regression. The results indicated that locus of control was significantly related to CON and dogmatismeas significantly related to CON and PDM. Neither was significantly related to IS. However, the sum of the variance accounted for that was attributable to linear regression for both personality variables was less than one percent on all dimensions. Therefore, it was concluded that the relationship between these personality variables and perception of leadership style is trivial. Satisfaction with the leader was found to be significantly related to all the leadership dimensions and accounted for 60 percent of the variance. Subjects who were satisfied with the leader described that leader as higher in CON and PDM and lower in I than subjects who observed the same leader but were dissatisfied with the leader. Thomas Edward Hill Construct validity of the leadership measures was assessed by means of the multitrait-multimethod matrix. While the CON, FUN, and IS scales marginally met the cirteria for construct validity, the high correlation between the CON and FUN scales of both measures demonstrated a lack of discrindnamt validity for these dimensions. The conclusions from this research have two important implications for field studies of leadership. When formulating contingency models of leadership subordinate attitudes specific to the job may be related to subordinate descriptions of the leader-follower relationship. On the other hand, personality characteristics of the subordinate appear not to account for a sizable portion of the linear variance in measures of leader-follower relationships. It was also concluded that for the sub- jects in the present study the CON and PDM dimensions of leadership were not meaningfully discriminable. Approved by Thesis Committee: Dr. Neal Schmitt, Chairperson: Dr. Carl Frost: Dr. Michael L. Moore: LOCUS OF CONTROL, DOGMATISM,'AND SATISFACTION WITH THE LEADER AS DETERMINANTS IN THE PERCEPTION OF LEADERSHIP STYLES By Thomas Edward Hill A THESIS submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1976 ACKNOWLEDGMENTS THANK YOU Bryan Coyle Carl Frost Jeff Coldblatt 'Alan Greentbal Greg Huszczo Eugene Jacobson Rod Lowman Mike MOore Stuart Ralsky Neal Schmitt Anna Toth Sue Weesner ii TABLE OF CONTENTS Page LIST OF TABLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . iV LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . v ./ REVIEW OF THE LITERATURE. . . . . . . . . . . . . . . . . . . . . . . 1 The Dimensional Approach to Leadership . . . . . . . . . . . . Locus of Control . . . . . . . . . . . . . . . . . . . . . . . . 23 Dogmatism. . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Satisfaction with the Supervisor . . . . . . . . . . . . . . . . 35 Concepts and Hypotheses. . . . . . . . . . . . . . . . . . . . . 42 monomGY O O I O O O O I O O I O O 0 O O O O O O O O O O O O I O O 5 2 Subjects . . . . . . . . . . . . . . . . . . Procedure. . . . . . . . . . . . . . . . . . Description of the Measures. . . . . . . . . Data Analysis. . . . . . . . . . . . . . . . 52 RESULTS 0 O O I O O O O O O O O O O O O O O O O O O 0 ~ 0 O O C 0 O O O 64 Manipulation Check . . . . . . . . . . . . . . . . . . . . . . . 64 M‘lltitrait-M‘llt imthOd mtrinx O O O O O O O O Q C O Q 0 O C O C O 65 HypotheSi-S Testing 0 O O O O O O O O 0 O O O O O O O O O O O O O 70 DISCUSSION 0 O l O O O O O O O O O O O O O O O O O O O O O O O O. O O O 80 APPENDICES A. The Desert Survival Situation. . . . . . . . . . . . . . . . 90 B. Leadership in Group Problem Solving Questionnaire. . . . . . 94 VII] '1 REFEENCES O O O O O O 0 O O O O O O O O O 0 O O O O O O O O O O O O O 104 iii LIST OF TABLES Table ' Page 1 A summary of the results from the Tannenbaum and All-port StUdy. O O O O O O O O 0 O O O O O O O O O O O O O 40 2 Results of the manipulation check. . . . . . . . . . . . . . . 66 3 Correlation matrix containing independent and dependent variables. . . . . . . . . . . . . ... . . . . . . . . . . . . 68 4 The Multitrait-Multimethod Matrix used to assess construct validity of the Johnson Three—Dimensional Leadership Questionnaire . . . . . . . . . . . . . . . . . . . 69 5 Summary of the regression analysis to test hypotheses 1 to 6. O O - O O I O O O O O O O O O O O O O O O O O 73 6 Final regression equations for the criterion variables . . . . 74 7 Mean criterion scores for three levels of the predictor variables. . . . . . . . . . . . . . . . . . . . . . 75 8 Summary of the regression analysis to test hypothesis 7. . . . 78 9 Mean criterion scores by experimental conditions . . . . . . . 79 iv LIST OF FIGURES Figure Page 1 Continuum of leadership behavior (Tannenbaum and Schmddt, 1958). . . . . . . . . . . . . . . . 4 2 Predicted interactions among subordinate related variables and the subordinate's perception of his leader's Participative Decision-making. . . . . . . . . . . . 46 3 Predicted interactions among subordinate related variables and the subordinate's perception of his leader's Initiating Structure . . . . . . . . . . . . . . . . 47 4 Predicted interactions among subordinate related variables and the subordinate's perception of his leader's Consideration. . . . . . . . . . . . . . . . . . . . 48 REVIEW OF LITERATURE In order to facilitate the efficient functioning of an organization few would question a basic assumption that some type of leadership is necessary. However, just what is leadership? Who are leaders? And how does one discriminate effective from ineffective leadership? Because of the ambiguity of the terms referring to leadership these-questions seem most difficult to answer at first glance. WOrds pertaining to leaders or leadership appear in most languages. In relatively simple cultures leader- ship terms merely differentiate rulers from other members of the culture. In more complex societies, particularly in cultures with an Anglo—Saxon heritage, leadership has become a sophisticated concept referring to a broad range of roles, situations and behaviors (Stogdill, 1974). Thus in order to study leadership, behavioral scientists have found it necessary to compose operational definitions of leadership. The first step in this task is to limit the scope of interest so that agreement on a frame of reference can be reached. In this thesis, leadership will be viewed only in terms of an industrial organizational context with formal rules and roles of leadership. Foremen, managers, administrators, executives and presidents are all common examples of this kind of a leader. Now leader- ship may be defined as, "behavior by one member of a group toward another member or members of the group, which advances some joint aim” (Bowers & Seashore, 1966). Within this definition the word behavior has considerable importance. Behaviors are definite events that can be reliably recorded and measured. This is necessary since past experience has demonstrated 1 that the study of leadership is greatly impeded by the absence of a measurement process. Historically the first attempts to study leadership were informal studies of the lives of great men. By examining the past accounts of powerful individuals, variables such as heredity, innate ability and specific abilities to enchant the masses were specified as casual mech- anisms for superior leadership. These erroneous conclusions resulted from the inadequacy of this method of study to deal with extraneous confounding variables. In addition, the great man theories have absolutely no measure- ment capabilities and therefore have yielded little useful understanding of the construct of leadership (Gibb, 1954). In light of these conclusions researchers have turned to classifica- tion systems. Using a classification system a researcher observes actual behavior that occurs when a person is leading and tries to identify and classify different leadership behaviors in terms of their structure and results. It is from.this system of study that the dimensional approach to leadership evolved. It is important at this point to note the progression of the study of leadership as a measurement system.. The goal of the "great man" studies is to examine historical data from a few exclusive persons and explain 2235 hog the cause of their success. Within this framework leader— ship is viewed, for the most part, as a unitary general factor with no real attempt to quantify or qualify it. The goal of the dimensional approach to leadership is to examine specific behaviors of anyone and predict his or her capabilities as a leader. Here leadership is viewed as a combination of many present behaviors which can be, and indeed must be, quantified and qualified. Operational definitions and classification systems solve only part of the problems associated with the study of leadership. They can answer the questions concerning what is leadership and who are leaders, but they cannot discriminate good from poor leadership without first establishing a set of criteria. In the past, objective indices such as turnover, grieve ances, and productivity have been utilized. In addition less tangible concepts such as subordinate satisfaction and paper and pencil ratings of the leadership style are often used. One thing all these measures have in common is that they all involve, in some way, the subordinate. Turnover, grievances, and productivity are actual outcomes of subordinate behaviors, and in some cases, the paper and pencil tests are administered to the followers in order to obtain their perception of the leader. Yet when these measures are used for evaluative purposes, they are interpreted only in terms of the leader himself. Under these circumstances, when a supervisor is able to elicit a relatively high standard of production and favorable employee ratings, he is ascribed to the status of a good leader. 'Many interpret this as a sole characteristic of that person which he will demonstrate in most leadership situations. The fallacy of this line of thought should be apparent. Since a leader cannot be assessed independently from the unit that he leads, the effectiveness or ineffectiveness should be considered a property of both the leader and the followers. In Victor Vroomis dissertation, following a comprehensive review of the literature, he states, The general conclusion is that leadership cannot be regarded as a unitary trait and must be evaluated in terms of a number of other variables including the attitudes, needs, and expectations of the followers. The most effective behaviors in dealing with indivi- duals with certain personality characteristics may be completely ineffective in dealing with persons with different personalities (Vroom, 1960, p.3). In a later study, Runyon reinforces this same conclusion when he states, ...The general finding that the personality of subordinates is an important variable in the supervisor-subordinate relationship has important implications. It suggests, for example, that management style alone is insufficient to account for differences in employee satisfaction, and that a broader, more comprehensive theoretical model is needed (Runyon, 1973, p.293). Such a theoretical model as alluded to by Nroom and Runyon has been reported by Robert Tannenbaum and warren Schmidt (1958). They analyzed leadership behavior in terms of the degree of authority used by a leader —*m——¢—1——m~v A. Han—... and the amount of freedom available to subordinates in making decisions. Presented in Figure l is a summary of this conceptualization. The leader— ship continuum demonstrates the wide variety of styles which a supervisor can utilize to lead a group. Boss ‘L_ Subordinate centered centered leadership ‘ leadership I‘— Use of authority by the manager Area of freedom for subordinates f Mangger Mangger Mangger Mangger Mangger Mangger Manager makes "sells" presents presents presents defines permits decision decision. ideas and tentative problem limits; subordinates and invites decision gets sug— asks to function announces questions. subject gestions, group to within limits it. to makes ‘make defined by change. decision. decision. superior. Reprinted from, R. Tannenbaum and W. Schmidt. How To Choose a Leadership Pattern. Harvard Business Review, 1958, 39, 95-101 Figure l. Continuum of Leadership Behavior. Inherent in this theory of leadership is the assumption that a leader occupies no one position on the continuum. In the short run a leader may have to change his style to meet present demands of the situation. Therefore, in each leadership situation and with each work unit a leader must consciously or unconsciously decide what leadership style will be most.effective to reach the goals of that unit. Or a leader may change his style over a long period of time as he matures and learns to better utilize his unit. Also inherent in this theory is the reality that a supervisor's choice of a leadership style is limited by a number of factors. Some of these limitations are imposed by his own ability to lead, some by the organization, some by his subordinates, and some by the specific problem at hand. All these factors work together to define the limits on the style of leadership that can be used in that situation. Actual antecedents which contribute to the chosen style of leader- ship are perceived by Tannenbaum and Schmidt as represented by three forces. These forces are trichomotised in terms of their origins. Thus we have forces in the manager, forces in the subordinates, and forces in the situation. The forces in the manager are shaped by his personality charac— teristics, his value system, his confidence in his subordinates, his own leadership style and his feeling of security. The manager's leader- ship behavior as well as the subordinates following behaviors, are determined by these variables which stem from the leader himself. Perhaps one way of looking at the ramifications of the force in the manager is exemplified in the dimensional approach to leadership. Here it is possible to examine three indexes of the way in which someone leads (consideration, initiating structure, and participative decision making). Although, admittedly, there are other leadership dimensions that could also be tapped to study the force of the leader, these seem to represent a broad sample covering the critical areas of leadership. Since different leaders typically exhibit diverse combina- tions and magnitudes of these three dimensions, one could theoretically determine the force in the manager which contributes to a pattern of leadership. The second force is the force in the subordinate. This force is defined by the subordinates' personality characteristics, his value system, his confidence in his superior, his own style of following, and his expectations about how his superior should behave. Once again the subordinates' following behaviors, as well as the superior's leadership behavior, are shaped by this force. Preferred style of following, previously mentioned as a component of the forces in the subordinate, needs to be emphasized since not much literature has accumulated on this topic. The concept is analogous to preferred style of leadership. Just as there are leaders who vary from democratic to autocratic in the way they lead their work group, subordinates also vary in a similar manner in the way they prefer to be led. This topic will be treated in greater detail in this thesis. The third and final force is that of the situation. Components of situational forces include type of organization, group effectiveness, size of working units, geographical distribution, the problem itself, and the time pressure to accomplish goals. In light of this discussion leadership can be conceived of as the results of the interaction from the three forces. Each force in its turn can influence the type of leadership that will be exhibited. For example, if the forces in the leaders are held constant, situational and subordinate forces could produce wide differences in the final leadership style. In one case a usually democratic leader may be re- butted by subordinates who do not wish to expend the effort to help make decisions and thus prefer more autocratic leadership. Eventually the forces operating may move the leader to a style that is more acceptable to everyone. As can be seen here, leadership cannot be assessed by studying the leader alone, but must be conceptualized as a system in which the subordinates and the situation play important parts. The general interaction model of leadership proposed by Tannenbaum and Schmidt has been operationalized by several different investigators (Fiedler, 1967; House, 1971; and Vroom, 1973) in recent years. Perhaps, the most popular of these is Fiedler's contingency theory of leadership (Fiedler, 1967). This theory distinguishes two leader personalities based on differential motivational systems. Relationship-motivated leaders are most concerned with maintaining good relationships with their subordinates; while task-motivated leaders tend to concentrate on getting the job done. What Fiedler essentially postulates is that group per- formance is contingent upon matching a leadership style and the degree of favorableness of the situation, that is, the degree to which the situation provides the leader with influence over the subordinate. Therefore, the contingency theory of leadership takes the three forces and their integrations, defined by Tannenbaum and Schmidt, and explains systematic relationship among these variables in terms of group performance. For example, this contingency theory advocates that taskdmotivated leaders perform best in favorable situations (high control and influence or low uncertainty) or when the situation is unfavorable (low power, control and influence). Relationship-motivated leaders perform best in moderate power, control and influence situations. Fiedler's contingency theory of leadership is only one example of the contingency theories that are now being developed. Others, while examining the same concepts, predict different relationships among the variables and draw different conclusions about which leadership styles are most effective in different situations. In an attempt to examine a portion of the interaction model of leadership, four separate reviews of the literature will be presented in order to clarify essential issues. The first section will deal with the conventional dimensional approach to leadership. The next two sections will deal with two personality characteristics that have been chosen because past research on these two constructs suggest that they have a high probability of being critical components of the subordinate forces which influence the leadership style. These personality variables are locus of control and dogmatism. Finally, satisfaction with super- visors will be reviewed because of its obvious importance to the leader- follower relationship. As previously mentioned only part of the inter- action model of leadership will be examined in this thesis. In particular, only the forces in the subordinate will be examined. To obtain the complete picture of this model the forces in the manager and situation must also be examined. However, to accomplish this would demand a considerable escalation of the data collection effort beyond the means of this study. Although the inclusion of all the forces would be optimal, omitting the latter two creates no deficiency in this study. The forces in the manager and situation have in the past received a great deal of attention and are already well defined. Therefore, it is the intention of this study to add to the present body Of knowledge where evidence is most scarce. The Dimensional Approach To Leadership The mid 1940's the Bureau of Business Research at Ohio State University instituted extensive research to construct an instrument for' describing leadership. From a large pool of original items, the 150 that best represented ten a priori dimensions were administered to a large summer school sample. The subjects were asked to describe a leader of a group. Hemphill and Coons (1957) intercorrelated and factor analyzed mean scores of the ten dimensions and obtained three orthogonal factors. 1. Maintenance of membership character. Behaviors that are socially agreeable to group members. 2. Objective attainment behavior. Behaviors that result in the output of the group. 3. Group interaction facilitation. Behaviors that structure communication among group members. Halpin and Winer (1957) revised the above questionnaire and adminis- tered it to a large number of bomber crews who described their crew commanders. A factor analysis was performed on this data in order to reorgan- ize the items into fewer and more independent categories of behavior. This was accomplished by correlating eight g_priori dimensions with 130 10 items (mostly from the original Hemphill and Coons questionnaire). The above procedure yielded four orthogonal factors which are: 1. Consideration. Behaviors demonstrating friendship, trust and respect. 2. Initiating Structure. Behaviors that organize patterns of organization. 3. Production emphasis. Behaviors which stress the job to be completed. 4. Sensitivity. Awareness of social interrelationships. Halpin and Winer dropped the third and fourth dimensions because they accounted for too little common variance. These results and this scale (The Leader Behavior Description Question) have come to be well known and consideration and initiating structure are now identified as the Ohio State dimensions of leadership. Concurrent with the Ohio State studies the University of Midhigan Survey Research Center also developed a dimensional approach to leader—' ship. Katz, Maccoby, and Mores (1950) located clusters of characteristics . ..-—..- "4.. --.- ’—-.. Hun—“.... --. ‘— m w... which correlated positively with themselves and effectiveness of the leader. They found two dimensions of leadership defined as: 1. Employee orientation. Behaviors by the supervisor that denote his positive feelings toward his subordinates. 2. Production orientation. Behaviors by the supervisor that stress llmln_lvllm-r~umll‘ getting the work done and technical aspects of the job. The former dimension resembles the Ohio State Consideration dimension while the latter resembles the Initiating structure and Production emphasis dimension. Katz and Khan (1951) working from a large amount of accumulated data presented yet another four dimensions of leadership,these are: ll 1. Differentiation of supervisory role. Behaviors on the part of the supervisor that stress planning and performdng specialized skills. 2. Closeness of supervision. Behaviors on the part of the super- visor that limit or expand a subordinate's opportunity to control his own work. 3. Employee orientation. Behaviors on the part of the supervisor that denote his positive feelings toward his subordinates. I 4. Group relationships. Behaviors on the part of the supervisor that facilitate group cohesiveness. Bowers and Seashore (1966) have noted that Katz and Kahn's four dimensions are not directly congruent with the Ohio State dimensions. While differentiation of supervisory role and employee orientation correspond directly to initiating structure and the Ohio State employee orientation respectively, closeness of supervision appears to be repre- sented by a combination of consideration, initiating structure and production emphasis. Group relations, on the other hand, appears not to correspond to any of the four Ohio State dimensions. Likert (1961), working with data from the Life Insurance Agency Management Association, suggests five dimensions which he reports are necessary for effective supervisory behavior: 1. Principle of supportive relations. Building subordinate sense of personal worth and importance. 2. Group methods of supervision. Building a high degree of group loyalty and effective skills of interaction. 3. High performance goals. Supervisors must be both employee— centered and at the same time have high performance goals. 12 4. Technical knowledge. Supervisors must have adequate competence to handle technical problems. 5. Coordinating, scheduling, planning. The supervisor brings to the group the views, goals, and decisions of other groups to provide communication and influence decisions. Once again these dimensions are compatible with the Ohio State dimensions. Principle of supportive relations corresponds to consideration, group methods of supervision corresponds to sensitivity, high performance goals correspond to production emphasis, and technical knowledge and planning correspond to initiating structure. Reformulating the findings of the University of Michigan studies Mann (1965) proposed a typology of leadership based on essential skills. Mann's essential skills were: 1. Human relations skill. Ability in working with and through people. 2. Technical skills. Ability to use technical knowledge. 3. Administrative skills. Ability to understand and act according to the objectives of the total organization. When Mann conceptualized his theory of leadership he asserted that skills and not behaviors are the essential bases for effective leadership. However others, notably Bowers and Seashore (1966), have successfully argued that while there may not be a perfect parallel between the behaviors requiring particular skills and the skills themselves, a rough correspon- dence between the two is apparent. Hence this formulation can be con- sidered a dimensional approach similar to that of the Ohio State researchers. From a broader perspective Cartwright and Zander (1968) have defined leadership in terms of behaviors which facilitate the group to achieve their goals. They particularly define two sets of group functions. 13 Which are: 1. Group maintenance functions. Behaviors that maintain pleasant interpersonal relations. 2. Goal achievement functions. Behaviors that initiate action. These two group functions are clearly the same construct as consideration and initiating structure defined by the Ohio State researchers. Bowers and Seashore (1966) reviewed most of the studies cited here and concluded that a great deal of conceptual content is held in common. By integrating the studies on leadership they proposed four dimensions which they state "...seem to comprise the basic structure of what one may term leadership." They are: 1. Support. Behaviors that encourage a feeling of personal worth. 2. Interaction facilitation. Behaviors which encourage satisfying relationships. 3. Goal emphasis. Behaviors which encourage goal attainment. 4. Wbrk facilitation. Behaviors which help goal attainment by planning and providing resources. This formulation conforms to all the various research cited in this thesis with the exception of Katz and Kahn (1951). Therefore it is evident that in loose terms some agreement has been reached on what leadership is and how it can be measured. One issue that has been of considerable interest to researchers in this field is the independence of various dimensions of leadership. Fleishman (1953) and Halpin and Winer (1957), have claimed that the (iimensions of consideration and initiating structure are independent. They'reason that conceptually a supervisor's behavior on one dimension 14 has no affect on his behavior on the other. Fleishman (1969) supports this argument by stating that although there exist positive, negative and non-significant correlation between these two dimensions, the median correlation would approximate the true value of zero, and other significant correlations represent a sampling distribution around this true value. However, evidence from a variety of other studies have cast serious doubt on this explanation. Weissenberg and Kavanaugh (1972) have reviewed 72 studies employing the Leadership Behavior Description Questionnaire (LBDQ) to measure consideration and initiating structure. In summarizing the results of these studies they found 51% reporting significant positive correlations, 10% reporting significant negative correlations, and 39% reporting non- significant correlations. All in all they found a mean interdimension correlation of .23. This study effectively refutes the assumption that consideration and initiating structure are independent. Several other studies have come to a similar conclusion (Bales, 1958; Fiedler, 1964; Korman, 1966; Lowin, Hrapachak & Kavanagh, 1969; and Johnson, 1973). Although there has been extensive research concerning a dimensional approach to leadership there remains a great deal of confusion. This problem is typified by the seemingly endless parade of conflicting re- sults and conclusions reported in the research literature. In terms of productivity, Sales (1966) found that experimental studies showed no consistent results. He found one study reporting democratic supervision the most effective, one study showing authoritarian leadership most effective, and four studies demonstrating no differences between leadership styles. Anderson (1966) reviewed the literature on initia- ting structure and consideration and noted that differing results 15 occurred when sub-culture groups were analyzed separately. In summarizing the conflicting results that have been found in leadership studies, Korman (1966, p.354) states, "The great preponderance of con- current validity studies provide no evidence whatsoever, even in those cases of positive relationships, that variation on these scales is a predictor of different kinds of worker behavior..." A great many reasons can undoubtedly be cited for this problem. Perhaps the major reason is the lack of coherence in defining and labeling the leadership behaviors. This has led to over-generalization of the dimensions and comparisons of seemingly parallel dimensions, when in fact, the leadership behaviors that fall within the dimensions are not the same. To deal with this problem it has been suggested by many that leadership dimensions must be clearly defined and purified. In other words, initiating structure and consideration, the two most popular dimensions, are perhaps too broad and may contain other major dimensions within one or both of them. The word "major" is stressed because if too many dimensions are isolated their limited scope may be of little use for generalization or prediction. Within this context researchers have started to look for an additional dimension of parti- cipative decision making. The concept of participation in decision making is not a new one and has received much theoretical attention. However, the formulation of participative decision making into a formal dimension of leadership has received little attention and only of late has such a paradigm been brought forth (Yukl, 1971; and Johnson, 1973). In an innovative approach, Lewin (1944) classified leaders into a theoretical trilogy made up of democratic, autocratic and laissez—faire l6 leaders. Although he failed to give specific definitions of these categories he did describe them.in general terms. Democratic leaders tend to directly solicit ideas and views of their subordinates in order to base decisions. Autocratic leaders avoid such participation on the part of subordinates and make decisions themselves. Laissez-faire leaders can best be described as "no leaders," with a hands-off policy whenever possible. Lewin suggests that these types of leaders should Egg be thought of in terms of a continuum, but rather as a triangle. A democratic leader is not half-way between the others. It is a distinct type of leadership that may be closer to autocratic leadership than laissez-faire leadership. The importance in this is that Lewin divided leaders in terms of their participation in decision-making, thus clearing the way for more behaviorally based dimensions. Another similar typology of leadership that has received a great deal of attention is authoritarianism and democracy. Authoritarian leadership is characterized by the concentration of power and decision- making. The democratic leader is characterized by a sharing of power and participative decision-making (Sales, 1966). The major difference between this and the before mentioned leadership typology is that democratic and authoritarian leadership are generally thought of as a continuum along which a leader may fall. Theoretical work by Stogdill (1959) concerning the differentiation of leadership position in an organization has demonstrated that the con- ventional two dimensions of leadership are insufficient to account for the observable variation in leadership behaviors. To meet this criticism Stogdill, Goode, and Day (1962) constructed eight new sub- scales, to be used along with the original Consideration and Initiating 17 Structure. The subscale, Tolerance of Members Freedom, describes behaviors that allow subordinates scope for input into the decision- making process that concerns the activities of his unit in the organi- zation. Stogdill stated that this measure of participative decision— making is necessary to gain a complete picture of leadership behavior (Stogdill, Goode, & Day, 1962). Heller and Yukl (1969) have examined the participative decision- making capacity of subordinates in what they call Decision-Centralization. They have reasoned that the amount of subordinate influence in decision- making within the organization can be perceived as a continuum ranging from no subordinate influence to complete subordinate influence. The leader may make decisions on his own without explanation or on his own with explanation (low subordinate influence), with consultation with his subordinates (medium subordinate influence), or with joint decision~ making or delegation (high subordinate influence). The different decision—making procedures used by any leader will vary according to situational variables. However, the average degree of subordinate participation in decision-making can be computed on any set of typical decisions and this average is referred to as Decision—Centralization. A high Decision-Centralization score indicates a low amount of participative decision-making, and conversely a low decision-central- ized score indicates a large amount of participative decision-making (Heller S Yukl, 1964). Democratic leader behavior, Tolerance of Member Freedom, and Decision-Centralization all appear to be measuring participative decision-making. Although, as mentioned above, caution must be taken 18 when equating different measures of leadership behavior clusters, it seems well documented that participative decision-making is important in explaining leader behavior. Even if it is given that this is the case, the question that still remains concerns whether participative decision-making is a dimension of leadership. In other words, is participative decision making equivalent to consideration and/or initiating structure or is it a dimension in itself? While the empirical evidence is far from conclusive it suggests that the latter may be most appropriate. Stogdill, Goode, and Day (1963, 1964) measured consideration and initiating structure by means of the LBDQ using the subscale Tolerance of Members Freedom to measure participative decision-making. From professional directories the researchers obtained a select group of corporation and union presidents with instructions to give the material to a subordinate that knew them well. Correlations between Participa— tive decisiondmaking and Consideration for corporate presidents and the union president were .41 and .42 respectively. There were no significant relations between Initiating Structure and Participative Decision-making for either sample. Stogdill gt 51, concluded that the leader behaviors of corporation presidents and union presidents cannot adequately be described only by the consideration and initiating structure, but can be described in terms of the additional eight subscales in which Tolerance of Members Freedom (participative decision-making) is included. In an earlier study Stogdill, GoOde, and Day (1962) compelled summer students to describe community leaders and another sample to 19 describe protestant ministers. Once again the LBDQ was used as the measuring device. They found correlations between Consideration and Participative Decision-making of .40 for community leaders and .49 for ministers. No significant relations were found for Initiating Structure and Participative Decision-making. From these results it was pointed out that if the LBDQ is used for comparison across popu- lations then the separate subscales are needed to adequately describe their leadership behaviors. Beer (1966) found a correlation between Participative Decision- making and Consideration of .50 for a sample of office supervisors in a large insurance company (using the LBDQ). No relation was found for Participative Decision-making and Initiating Structure. Argyle, Gardner, and Cioff (1957) measured democratic vs. authori- tarian leadership with a questionnaire constructed for their study. They administered the questionnaire to English foremen. Democratic foremen were defined as ones who allow discussion and participation in the running of their section, listened to ideas and suggestions and passed on information to their subordinates. The results of the study indicated that democratic leadership was correlated with employee- centered leadership (.42) and nonpunitive leadership (.41); both are significant at the .001 level. Employee-centered and non-punitive leadership were considered to be a part of the consideration dimension. Democratic leadership was not signficantly correlated with pressure for production, a component of initiating structure. In a study designed to assess the situational variables of participative decision-making, Heller and Yukl (1969) administered the Decision Procedure Questionnaire (form C) to 67 second-line supervisors 20 in three companies. A significant correlation (r = .24; p < .05) between consideration and participative decision—making was found. As usual no significant relationship was found for initiating structure and participative decision—making. To summarize the above studies, it appears that the empirical evidence demonstrates that participative decision—making, initiating structure and consideration are oblique dimensions of leadership behaviors. To elaborate on this, it can be said that participative decision-making and consideration have a moderate positive correlation, and participative decision-making and initiating structure have a low positive correlation. To account for the above mentioned correlations among these dimensions, Lowin, Hrapchak, and Kavanagh (1969) have hypothesized that the participative decision-making items are included in the Consideration and Initiating Structure scales and therefore contaminate them. As an example of this Lowin g£_§l,, examined the LBDQ and found the following participative decision-making items in the Consideration scale: 1. He refuses to give in when people disagree with him. 2. He insists that everything be done his way. 3. He rejects suggestions for changes. 4. He changes the duties of people under him without first talking it over with them. 5. He acts without consulting his foremen first. 6. He puts suggestions that are made by foremen under him into operation. 7. He gets the approval of his foremen on important matters. Participative decision-making items within the Initiating Structure scale were: He rules with an iron hand. He waits for his foremen to push new ideas before he does. He lets others do work the way they think best. He decides in detail what shall be done and how it shall be done. J-‘UNH O o o 21 The inclusion of these items in both the scales has led to an arti- ficially high correlation among the three dimensions of leadership. The empirical evidence for a leadership dimension of participative decision-making is at best suggestive and therefore the conclusions drawn from this evidence must be interpreted with caution. To justify the conclusion that participative decision—making is a separate dimen- sion of leadership a strong theoretical argument must also be made. Yukl (1971) has presented several such theoretical arguments. Con- sideration has been at times defined as including the sharing of decision-making with subordinates. However, as Yukl stated, this argument is only valid if the subordinates desire to participate in decision-making. Clearly if subordinates want no part in this activity the considerate leader would not force his subordinates to so engage. Inclusion of participative decision-making items in a consideration scale has resulted in scores which are not comparable across persons unless first adjusted for differences in preference in subordinates' desire to participate in that situation (Yukl, 1971, p.417). To further confound this issue a leader may behave in an extremely con- siderate manner but allow absolutely no participation in decision- making. Such a case is evident in MeMurray's (1958) benevolent autocrat. Here the supervisor makes all policy decisions but in such a way as not to offend his subordinates. To be more specific, he may confer with subordinates before his decision, explain in full, a course of action, hear all subordinates' complaints, and meet all arguments with information and concern. This type of leader is certainly a considerate one, but far from.being a participative leader. 22 Initiating structure, as it is usually defined, included three types of behaviors: concern for production, insuring that decisions from.bigher levels of the organization are carried out, and insuring that necessary task decisions are made. Note that this does not specify who will actually make the decisions (Yukl, 1971, p.417). Supervisors may initiate a great deal of structure in order to complete a task after a participative decision has already been made, or he may make the decision himself and initiate structure in the same manner. Because of this and in light of the strong empirical evidence, it seems reasonable to treat initiating structure and participative decision-making as distinct dimensions of leadership. Such an attempt has been made by Johnson (1973). By using multi- dimensional scaling he found three dimensions of leadership behaviors. In addition to the usual consideration and initiating structure, a participative decision-making dimension was also found. Johnson contended all three dimensions accounted for from 45.3 to 73.5 percent of the variance when subordinate motivation, subordinate atti— tude toward supervisor task competence, and subordinate attitude toward supervisor's interpersonal style were dependent measures. However, Johnson made no claim to the orthogonality of these dimensions. In fact, he stated that there is a low positive relation among the dimensions. However, the interaction effects caused by the inter- correlation of the scales account for only a minute amount of variation and therefore can be, for the most part, ignored. A.word here is needed to distinguish between psychological and objective measurements of leadership. Vroom (1960) defines psychological 23 participation as the influence an individual perceives he has in decision-making, while objective participation is the actual influence the subordinate has on decision-making. Vroom notes that if subordinates' perceptions are 100 percent correct the two indices will coincide, however, in most cases this would not be the case and the two would be discrepant. This concept can be generalized to other leadership dimensions or to leadership as a whole. In this study, unlike most,. both an objective and psychological measurement of leadership style were attempted, thereby allowing comparison of perception of leadership style and actual leadership behavior. Locus of Control Since Rotter's introduction of the construct of locus of control it has received a great amount of attention in the psychological literature. There are now 13 measures of locus of control, five literature reviews (Joe, 1971; Lefcourt, 1966; Hinton, 1967; and Rotter, 1966), and one extensive bibliography of 330 references (Throop & MacDonald, 1971). Perhaps the reason for this uncommon popularity is that it ties together several areas of research that were uninterpretable until locus of control was measured and applied to these problems. Rotter (1966) defines locus of control as the degree that a person perceives desired reinforcement as contingent upon his own behavior. He further states, When a reinforcement is perceived by the subject as following some action of his own but not being entirely contingent upon his actions, then in our culture, it is typically perceived as the result of luck, chance, fate, as under the control of powerful others, or as unpre— dictable because of the complexity of the forces surrounding him. When the event is interpreted in this way by an individual, we have labeled this a belief in external control. If the person perceived 24 that the event is contingent upon his own behaviors or his relatively permanent characteristics, we have termed this a belief in internal control (Rotter, p.l). The theoretical base for locus of control originates from social learning theory (Rotter, 1945, 1955, 1950). Generally, social learning theory advocates the proposition that when a behavior is reinforced an expectancy that that behavior will be reinforced_in the future is produced. Hence, a person will exhibit similar behaviors in new situations in order to receive the anticipated reinforcement. On the other hand, if a person does not believe the reinforcement is contingent on his behavior the expectancy will not increase. Therefore, it has been suggested that one's expectations or beliefs that reinforcement is dependent on his behaviors will depend on his previous reinforcement history, which in turn, will affect the way he behaves in any given situation (Rotter, 1966). Evidence for this contention has come from a variety of sources. Goodnow and Pettigrew (1955) have found that probabilistic learning theory is not applicable when subjects perceive that there is no systematic relation between reinforcement and behavior. In order to examine the growth and extinction of expectancies in perceived chance and skill conditions, Rotter, Liverant, and Crowne (1961) tested subjects in an ESP task (perceived chance) and a motor task (perceived skill). Both groups were subjected to 25%, 50%, 75% or 100% experi— mentally controlled varying ratio reinforcement schedule over eight trials followed by extinction trials. In three of the reinforcement conditions (all but 100%) the skilled group exhibited significantly greater increments or decrements following success or failure than the 25 chance group. In the other group on extinction trials it was found, contrary to usual Operant principles, that partial reinforcement was superior over continuous reinforcement only in the chance groups. Under the skill conditions, resistance to extinction was greater under 100% reinforcement than the 50% condition. The authors concluded that subjects in the chance groups perceived the extinction trials as a disappearance of their 'good luck.' The subjects in the skilled group took longer to accept the fact that they were no longer able to control the task successfully. Similar results have been reported by a number of researchers (James & Rotter, 1958; and Holdent S Rotter, 1962). In another study using the 'skilled-chance' group design, Phares (1962) looked at the perceptual thresholds for shock-associated stimuli. He found that recognition thresholds significantly decreased in the skill condition but not in the chance condition. Phares inferred that the subjects who perceived they could control the environment will exhibit behaviors that will enable them to cope more than subjects who perceive chance as the controlling factor. The implications of the preceding studies are explicit for the formulation of a theory of locus of control. In Rotter's words, In its simplest form, our basic hypothesis is that if a person perceives a reinforcement as contingent upon his own behavior, then the occurrence of either a positive or negative reinforce- ment will strengthen or weaken potential for that behavior to reoccur in the same or similar situations. If he sees the reinforcement as being outside his own control or not contingent, that is depending upon chance, fate, powerful others, or un— predictable, then the preceding behavior is less likely to be strengthened or weakened. Not only will there be a difference of degree but a difference in some instances, in the nature of the function or the result of a series of trials (Rotter, 1966, p.5). 26 At this point a theoretical social learning paradigm implemented with controlled lab studies provides a substantial base of evidence to support a locus of control construct. The next step was to demonstrate that this theory can be applied to people in general in their normal environment. Therefore, the problem switches from one of experimental design to one of measurement. The first attempt to measure locus of control as a personality variable was accomplished by Phares (1957). On an a_priori criterion he developed 26 five-point Likert-type items, half of which were purported to assess external attitudes and half internal attitudes. Using this measure in a study on skill versus chance contingencies for reinforcement, Phares found that subjects who scored high on external items demonstrated more unusual shifts, smaller magnitude of increment and decrements, and lower_frequencies of shifts of expectancy than subjects who scored low on these items. Although their results were not significant, they were encouraging in that their findings were congruent with studies where chance (external) and skill (internal) perceptions were manipulated. In his doctoral dissertation, James (cited in Rotter, 1966) revised Phares' test. He used only what he thought to be the most useful items from the Phares test plus some new items of his own. In a chance versus skill type study, James found results similar to Phares, but this time there were significant differences between dependent measures in the two conditions. In order to further refine the measurement of locus of control an attempt was made by Liverant, Rotter, and Seeman to construct a forced-choice questionnaire (reported in Rotter, 1966). From a primary pool of 100 items (written by Rotter's research team) 27 the instrument was subjected to both item analysis and factor analysis which reduced the number of items to 60. Additional items were eliminated due to their relatively high correlation with the Marlowe-Crown Social Desirability Scale (Crown S Marlowe, 1964). In the end, this procedure resulted in a 29 item instrument and is referred to as the I-E scale. Most of the studies reported later in this paper use this scale to measure locus of control (reliability and validity data is presented in the methods section of this thesis). Research that deals directly with leadership and locus of control have only recently been undertaken. Runyon (1973) hypothesized that persons who are identified as either internals or externals will react differently to styles of supervision, differentiated along a directive- participative continuum. The subject pool contained 110 hourly employees in a large chemical company. Each subject was administered four question- naires. They were: 1) style of management (seven questions developed for this study); 2) the I-E scale (locus of control); 3) Lodahl and Kejner job involvement scale; and 4) satisfaction with supervision (one question developed for this study). The independent variables were style of management and locus of control, while the dependent variables were job involvement and satisfaction with supervision. It is important to note here that the style of management measure was filled out by work-team members that were not subjects in this study. The author reported that this was done to assure that the personality character- istics of the subject (locus of control) would not confound management style ratings of their leaders. Using this procedure still yields a measurement of psychological participation, however, it is different 28 from the procedure that is normally used in studying leadership. The results of this study confirmed Runyon's hypothesis. Specifically he found: (1) that under a participative leader the satisfaction of internals is significantly greater than externals (p < .01); (2) that under conditions of directive leadership the satisfaction of externals is greater than internals (p < .01); and (3) internals exhibit signi- ficantly more job involvement than externals under both participatory and directive supervision. These findings are congruent with those that one might expect from the theoretical implication of locus of control. Externals, perceiving that reinforcement is determined by external sources, do not desire, nor are they satisfied with, a participative supervisor. Internals, on the other hand, prefer participative management styles. They believe they can positively influence their reinforcement and therefore desire a chance to do so. If not given this chance, dissatis- faction arises due to the perceived discrepancy between what they actually receive and what they feel they could receive if they were able to influence the outcomes. WOrking with a large (900) sample of public utility workers, Mitchell, Smyser, and Weed (1975) conducted a study similar to Runyon intended to appraise the relationship among locus of control, leadership style and satisfaction with the leader. They partitioned the sample into four groups by dichotomizing both locus of control (internals and externals) and type of leadership (participative and directive). A two-way analysis of variance design was used, yielding significant main effects and interactions. The results were interpreted to indicate 29 that while both internals and externals were more satisfied under participative than directive leadership, internals were more satisfied with participative leaders than externals, and externals were more satisfied with directive leaders than internals. However, the amount of variance attributable to locus of control was small. The authors concluded that, "...This personal characteristic, although theoretically meaningful, may not have a major practical impact" (Mitchell, Smyser, Weed, 1975, p.629). Dogmatism In 1950, T.W. Adorno and others introduced the authoritarian personality and it soon became a major topic for social researchers (Adorno, Frenkel-Brunswik, Levinson, S Sanford, 1950). They charac- terized authoritarians as anti-semitic, ethnocentric, politically and economically conservative, and rigid. Along with the theoretical structure they also introduced the F—Scale as a measure of the authori- tarian personality. It is a historical fact that the original research pertaining to the authoritarian personality was initiated when the problem of fascism accompanied by anti-semitism and ethnocentrism was of primary concern to social scientists. From this climate it is not at all surprising that the construct of authoritarianism became linked to fascist ideology (Rokeach, 1960). However, as pointed out by Rokeach, attributes which are indicative of the authoritarian personality are also manifest by leftist, righist, liberals, middle-of-the—roaders and conservatives. From the preceding consideration it is apparent that authoritariansim 30 is much broader than the facets studied by Adorno (Rokeach, 1954). As a result of this, Rokeach advanced the concept of dogmatism. Rokeach (1954) defined dogmatism as, (a) a relatively closed cognitive organization of belief and disbeliefs about reality, (b) organized around a central set of beliefs about absolute authority which, in turn, (c) provide a framework for patterns of intolerance toward others (p.195). From this framework Rokeach (1954, p.197) advanced a series of postulates concerning the belief and disbelief systems of dogmatic individuals. 1. "The greater the dogmatism the more will the belief system be perceived as different in content or aim from the disbelief system (e.g. Catholicism and Protestantism ...)" 2. "The greater the dogmatism the more will ideological arguments pointing to similarities between belief and disbelief systems be perceived as irrelevant." 3. "The greater the dogmatism the greater the denial of events contradicting or threatening one's belief system." Rokeach went on to state that when a person exhibits a closed belief system he finds it difficult to discriminate between the accuracy of a statement and the authority of the source. The exact difference between the construct of authoritarianism, as measured by the F-Scale, and dogmatism, as measured by the Dogmatism Scale, has been an issue of concern for some time. Rokeach contends that dogmatism is a generalized theory of authoritariansim, while the F-Scale yields a specific dimension or "right" authoritariansim, (Fruchten, Rokeach, S Novak, 1958). In support of his theory Rokeach (1952) reported a significant correlation of .60 between dogmatism and authoritariansim when political- economic conservatism was held constant. When dogmatism and authorit- arianism were correlated with political-economic conservatism, on four 31 geographically diverse samples, the consequent coefficients ranged from .13 to .28 for dogmatism and .22 to .43 for authoritariansim. The discrepancy between these sets of correlations supports the con- clusion that dogmatism is actually measuring general authoritariansim (Rokeach, 1960). To further test this conclusion Hanson (1968) administered both the Dogmatism Scale and the F-Scale to college students. He examined the relation between authoritariansim and dogmatism by analyzing dogmatism scores and responses to categorical and probabilistic authoritariansim items. Hanson's results supported Rokeach's argument that the Dogmatism Scale is a measure of general authoritariansim of both the left and right. Plant (1960) also found with a sample of 2350 college students that the Dogmatism Scale tapped general authoritarianism. Kerlinger and Rokeach (1966) administered the F-Scale and the Dogmatism Scale to 1239 subjects located in three diverse areas of the United States. They then factor analyzed the intercorrelations by using the principle axes which were labeled dogmatism, fascistic authoritariansim, and authoritarian aggressions. The authors concluded that the F-Scale and Dogmatism Scale are factorially discriminable even though they both measure the same construct. The evidence presented here supports Rokeach's contention that, although dogmatism and authoritarianism are related, they are discernable in that dogmatism is not uniquely connected with rightist ideologies. But rather is a general measure of the openness or closedness of beliefs and disbeliefs about reality. This implies that those who score high on the F-Scale will also score high on the Dogmatism Scale. In addition, 32 persons who are not ofzirightist ideology (thereby scoring low on the F-Scale) but do in fact have a closed belief system will score high on the more general Dogmatism Scale. As predicted by Rokeach, dogmatism can be used as an index to study interpersonal behaviors. In a classroom situation Zagona and Zurcher (1964) found that students with high dogmatism scores were greatly concerned with the selection of leaders and group structure, but when challenged by authority withdrew and became insecure. The authors also noted that high dogmatic persons are not problem solvers and have a great deal of trouble relating to others. Druckman (1967) discovered that high dogmatism subjects tend to be more resistant to change than do low dogmatism subjects. Several studies have supported Rokeach's contention that dogmatism interferes with the processing of information, thus acting as a defense mechanism. For instance, one study's results suggest that subjects with high dogmatism exhibit behavior patterns that reduce sensory activity with the result of reduced self-awareness (Kaplan S Singer, 1963). Along this line of research several studies have found that high dogmatism is related to personality maladjustment and a poor self-image (Vacchiano ££_§l,, 1968; Kemp, 1966; and Norman, 1966). Authoritarianism and dogmatism have been of interest in the study of leadership. Typical results have shown that work groups with leaders low in authoritarian characteristics exhibit greater satisfaction, more participation, less pressure for performance, less task-orientation, and less negative sanctions, than groups with high authoritarian leaders (Halpin S Winer, 1952; Kahn S Katz, 1953; Morse, 1953; Vroom, 1959; 33 Vroom S Mann, 1960; Singh S Arya, 1965; and Dustin S Davis, 1967). A final conclusion that has been drawn by many is that in general high authoritarians tend to be rejected as leaders and in sociometric choices (Stogdill, 1974). The next logical extension of this research is to behavior of high and low authoritarian followers. Once again this area has been well researched. Stanford (1950) administered an authoritarian-equalitarian measure to a sample of 963 adults. He found that those who scored high on authoritarianism prefer status-laden leaders, accept strongly directive leaders, and regard authoritarians as more effective than democratic leaders. Those scoring low preferred leaders who were guided by the people and friendly. He also found that the leaders who tell subordi- nates exactly what to do and how to do it were accepted by authoritarians but rejected by nonauthoritarians unless some special circumstances demanded it. Haythorn 35 31., (1956) investigated the effect that different levels of authoritarianism, of both the leader and the follower, had on the performance of the group. College students in a group problem solving exercise were assigned different combinations of authoritarian and equalitarian leaders and followers. Generally, it was reported that authoritarian leaders behaved in a more autocratic manner, were less concerned with group approval, and less sensitive than non- authoritarians. Authoritarian followers were also less equalitarian, less sensitive to others, and less goal oriented. However, the signi- ficant interaction indicated that differences between high and low authoritarian leaders were contingent on whether the followers are high or low authoritarians. For example, it was found that leaders, regardless 34 of their authoritarian orientation, with high authoritarian followers were rated higher on autocratic behavior than leaders with low authori- tarian followers. These results were interpreted as supporting an interaction theory of leadership. In part, these results were confirmed by Newcomb (1961) who found that low-authoritarians were better able to name which of their subordinates agreed with them and hence let attraction preferences be determined by this more than high authoritarians. Wilkins and DeCharms (1962) reported an experimental investigation concerning the relationship between authoritarianism and a person's reaction to internal or external power or status. Internal power cues were manipulated by reported socio-economic status. The results supported the theoretical position that high authoritarians are not more sensitive to external power cues, but are more influenced by them in making evaluations about others. On the other hand, low authori— tarians are less influenced by external power cues and use more internal cues in evaluating others. Vroom (1959, 1960) working with 108 first, second, and third line supervisors in a large company whose function was the delivery of parcels to retail stores, assessed their participative decision~making, job attitude, need for independence, and authoritariansim. He divided the sample into three approximately equal subgroups on the basis of their scores on each of the personality variables. Significant positive correlations were found between participative decision-making and job attitude in the high and moderate need for independence, and moderate and low authoritarianism subgroups, but no significant correlations were found for the low need for independence and high authoritariansim 35 subgroup. In a similar manner, significant positive correlations were found between participative decision~making and job performance for high and moderate need for independence, and moderate and low authoritarianism but not for the low need for independence and high authoritariansim subgroups. This study demonstrates that personality variables of a subordinate have an effect on how that person reacts in a participative decision-making situation. Authoritarians and persons with weak independence needs are relatively unaffected by the opportunity to participate in decision-making, while equalitarians and those who have strong independence needs develop more positive attitudes toward their job and greater motivation for effective performance through participa- tion. Satisfaction with the Supervisor Throughout the previous sections of this thesis the concept of job satisfaction and satisfaction with the supervisor has been mentioned. This topic, an already popular one, is commanding more and more attention due to the rising public awareness of employee's discontent at the work place. Despite several decades of investigation, sound theoretical formulations have yet to be discovered. Lawler (1973) suggests that the major reasons for this state of the science are that most research has been both atheoretical and unorganized. This strategy, or lack of strategy, has resulted in a substantial mass of contradicting and un- interpretable facts. With this in mind it is not the purpose of this review to pull together the vast literature or to offer any new theore- tical contributions to the study of job satisfaction. Its purpose is, 36 however, to introduce a basic model of satisfaction with the supervisor and cite only the studies that deal directly with this study. Most theories of job satisfaction treat it as a complex set of variables. An employee may be very satisfied with some components of the job while being quite dissatisfied with others. The thesis of most theories of job satisfaction is directed at combining these parts into a reportable generalized attitude (Vroom, 1964). In these terms Beer (1964) defines job satisfaction as, "The attitude of workers toward the company, their job, their fellow workers and other psychological objects in the work environment." Before the generalized attitude can be defined, the parts that contribute to it must also have a theoretical basis. Such components as pay, working conditions, company policy, participation, advancement, and supervision have been labeled facets of job satisfaction. A model of facet satisfaction comprised of the strengths of both equity and discrepancy theory has been presented by Lawler (1973). This model, while not being definitive, is sufficient to define the construct as it will be used in this study. The model proposes that a worker's satisfaction with any facet of the job is determined by a discrepancy between what the worker perceives he should receive, in terms of rewards, and what the worker perceives he actually receives. In this model, if a person perceives his outcome level as lower than his input level, he will experience dissatisfaction. When, on the other hand, inputs and outputs are perceived as equal that person will be satisfied. Perception of inputs and outputs are influenced by past experience and with reference to what others receive. 37 The nature of inputs and outputs are clear for facet satisfiers such as pay and working conditions, but they are less clear in the case of satisfaction with supervisor. Perhaps a useful way to define inputs in this case is to examine the behaviors a worker must exhibit in order to initiate and sustain a functional relationship with his supervisor. Such behaviors would include communication, 'stroking behaviors,' minimum standard of performance, attendance, and ability to comply to both general company policies and specific demands of the supervisor. By functional relationship it is meant one in which both parties are able to interact in order to efficiently carry out their job respon- sibilities with a minimum of animosity or of friction between the two. The outputs here are the usual extrinsic (pay and promotion) and intrinsic (autonomy and feedback) outcomes defined by Lawler and others (Lawler, 1973; Vroom, 1964; and Adams, 1965). Thus, in terms of the discrepancy model, if a worker perceives that in order to maintain a functional relationship with his supervisor his inputs are greater than the outputs he receives, he will experience dissatisfaction. But, if the relationship is maintained with inputs equal to outputs, the worker will be satisfied with the supervisor. This is very much a situational model, in that, for each subordinate-supervisor pair the level of inputs, type of relationship that is optional, and the desire outcome will be unique. In some cases where the worker and the leader have compatible personalities and working styles, few inputs will be needed to maintain a functional relationship, thereby receiving the necessary outcomes. However, when the two are not compatible in either personality or work style a great deal of effort must be exerted in the form of inputs in order to maintain the functional relationship. 38 A common procedure for measuring the subordinate-supervisor re— lationship has been to reconnoiter the subordinate's perception of the level of consideration he feels his supervisor displays. Several studies have found significant positive relations between consideration and satisfaction with supervisor (Fleishman S Harris, 1962; Halpin S Winer, 1957; Likert, 1961; and Wager, 1965). However, not all studies have found this relation. Pelz (1952) has found that when a supervisor is low in upward influence, consideration does not correlate with sub- ordinate satisfaction, while the same behavior exhibited by influential leaders is related to subordinate satisfaction. Wager (1965), in a similar study, found that the relation between consideration and satisfaction was greatest in groups who shared a common "organizational marginal" status. This relationship may have been facilitated by work group members compenséting for their precarious status by becoming dependent upon one another to obtain their goals. House, Filley, and Gujarati (1971) found support for wager's concept of marginal status in a study on research units. Further confusion has been contributed to this area of study by a recent investigation by David Herold (1974). He measured consideration (LBDQ), leader hierarchical influence, need for independence, and three facets of satisfaction (advancement, pay, security) as perceived by 91 nonsupervisory employees of a manufacturing company. The moderating effect of upward influence found in the above mentioned study was not supported here. Relationships between consideration and satisfaction were the same for both the high and low influence groups. When high or low need for independence was introduced along with high or low influence 39 (yielding four groups) a significant relationship between consideration and satisfaction was found for the low influence - low independence and high influence - high independence group, but no significant relations were found for the other two groups. Herold interpreted these results to indicate that past findings of moderate relationship may be the result of a failure to identify population subgroups which exhibit widely different relationships. He goes on to suggest that the contradictory findings in the area will not be resolved until subordinates' charac- teristic interactions with leader behaviors are better understood. Tannenbaum and Allport (1956) have completed an experimental field study to test whether patterns of personality constitute a basis to predict satisfaction in a work-program. Within the confines of Allports' Event Structure Theory (Allport, 1954) they determined whether a person's personality was suited or unsuited to work in a democratic or autocratic form of organization. The subjects were 400 female clerical workers of a large business organization. Two departments of the organization were manipulated to exhibit the democratic or autocratic management styles. In the autocratic, or hierarchical, program the workers were rigidly regulated. Decisions were centralized in top management, lectures were given on work standards, and scientific management principles were utilized. In the democratic, or autonomous, program the workers were given increased control over working conditions and authority was delegated throughout the organization. The people in this department participated in meetings in which they made decisions concerning work distribution, rest periods, working hours, and personal freedoms. The analysis of data took the form of a 2 x 2 factorial design with democratic- 40 autocratic departments and suited-unsuited employees as the independent variables and four questions pertaining to satisfaction as the dependent measures. The results of this study are summarized in Table 1. By examination of this table it can be seen that in every case but one the satisfaction scores are higher for suited workers than the unsuited. When dichotomized by democratic and autocratic departments only one satisfaction measure is significantly different; however, after collapsing over departments, three of the measures show significant differences between suited and unsuited employees and the fourth is in the predicted direction. Table l A Summary of the Results from the Tannenbaum and Allport Study Percentage of subjects respon- Mean Percentagg ding in the Satisfied direc- on all 4 Depen~ tion on the 4 dependent dent measures measures Departments Subject 1 2 3 4 Autonomous unsuited 36.7 21.8 37.5 22.6 32.2 suited 76.0 42.2 60.0 31.8 v 52.5 Hierarchical unsuited 23.1 53.9 40.0 20.0 43.2 suited 33.3 63.3 36.4 45.5 44.6 Combined unsuited 26.2 48.6 23.1 65.2 40.8 suited 60.0 59.8 47.2 80.0 61.8 From this Tannenbaum and Allport concluded that employee satisfaction or dissatisfaction is a function of the interaction between the personality structure of the individual and the management style. Or, more simply, a person is more likely to be satisfied if the working environment is compatible with his personality than if it is not. 41 . Here it should be noted that none of the dependent measures reported above dealt directly with satisfaction with the supervisor. They all pertain to attitudes toward the general managerial environment of the two departments. However, since attitudes toward general satisfaction are contingent on the personality of the employee it can, in a similar manner, he justifiably hypothesized that satisfaction with the leader is also contingent on the personality of the employee. In the context of this study it can be said that employees who are suited to a particular leadership style will be more satisfied than those who are unsuited to that leadership style. In order to place the aforementioned studies in perspective a few observations must be made. One reason for the wide range of discrepant results can be accounted for by the measurement of satisfaction itself. Many of these studies purport to measure different kinds of satisfaction i.e., general job satisfaction, satisfaction with supervisor, and pay. This situation has arisen because each study deals only with a limited segment of the field, and each researcher has defined his measures in terms of concepts which are convenient to his specific situation. The diverse semantic approaches prevent the direct comparison of the various results across studies (Carroll, 1973). Another issue that must be dealt with is that of perceived versus actual behavior changes. This subject was discussed in the leadership section of this thesis, and, once again, applies here. For example, whether or not satisfied workers' leaders are really more considerate or the worker just perceives them as so has been untested. Since no study has directly measured actual consideration, all that can be said is that in some cases satisfied workers perceive their leader as more considerate. 42 Summarizing the research on satisfaction with the supervisor for interpretive purposes is difficult in that strong conclusions cannot be made. The substantial number of conflicting findings suggest that pre- sent models are inadequate to act as a framework for an inexhaustive investigation of this subject. The solution to this problem seems to be the evolution of models that are more encompassing while still main— taining the ability to account for, on an individual level, variables associated with the subordinate, the leader, and the situation. The need for this is demonstrated by the conclusion that personality charac- teristics of both the leader and the subordinate influence the satisfaction with the leader. Lawler's (1973) synthesis of equity and discrepancy theory partially meets the issues raised above and is a step in the evolutionary process. Concepts and Hypotheses In the previous sections of this paper, the dimensional approach to leadership, an interaction model of leadership, the personality charac- teristics of locus of control and dogmatism, and satisfaction with the supervisor have been reviewed. The task that remains is to pull this rather extensive amount of information together in order to empirically test the proposed interaction model of leadership. Basically this model suggests that leadership cannot be assessed solely in terms of the leader himself, but rather, may be conceptually thought of as an inter- action between both the leader and his followers. Therefore, it is proposed that followers who possess different personality characteristics will perceive their leader as exhibiting diverse leadership styles. 43 In this study, leadership style will be described by the combination of the leadership dimensions of participative decision~making, initiating structure, and consideration, although operationally these dimensions are defined in terms of the instrument used to measure them, conceptually they are defined as (from Johnson, 1973): Participative Decision-making: Behaviors by which the supervisor allows or encourages subordinates to have a say in matters related to their jobs. Behaviors by which the supervisor allows or en- courages subordinates to influence decisions related to their jobs and the way in which they do their jobs. Initiating Structure: Behaviors by which the supervisor organizes and coordinates the activities of subordinates. Behaviors by which the supervisor defines his relation to his subordinates, and defines the role he expects each subordinate to assume. Consideration: Behaviors by which the supervisor'shows warmth and rapport toward subordinates. Behaviors by which the supervisor expresses a warm, positive attitude toward and personal concern for subordinates. Viewing these subordinates as either external or internal on locus of control, high or low on dogmatism, and satisfied or dissatisfied with their supervisor, yields eight groups of subjects containing all possible combinations of these independent variables. It is a major underlying assumption of this study that persons within each of these cells will systematically differ in the way they believe their leaders behave on the three dimensions of leadership. In order to formulate hypotheses it is necessary to examine the personalities and typical behaviors of the person placed within each of the cells. It seems reasonable to suggest that if a subordinate is satisfied with his leader, or in terms of the model presented earlier has a functional relationship with the leader, he will perceive the behavior of that leader in a different manner than one who is dissatisfied. 44 For example, if one believes that participative decision-making is an important function in obtaining his work goals and he is satisfied with his leader, he will perceive his leader as affording opportunities for Participative Decision-making. On the other hand, if a subordinate believes that participative decision-making is a waste of time and not goal oriented, but he is satisfied with his leader, he will perceive his leader as low in participative decision-making. Therefore, it is possible for a leader to receive scores on opposite ends of the Participative Decision—making Scale depending on who is rating him. This type of result is not uncommon in the literature and has led some researchers to be somewhat critical of such instruments as the LBDQ (Keeler S Andrews, 1963; Hood, 1963; and Stogdill, 1974). A person who scores as internal on the I—E Scale can be typified as a person who believes his abilities and skills determine his outcomes, is willing to take moderate chances, is a high achiever, is not easily, nor does he prefer to be, externally manipulated, strives for self improvement, exhibits problem solving behavior, and would rather rely on himself than others. An external may be described as one who believes outside sources determine his outcomes, is conservative in risk taking, is intimidated by authority, is easily manipulated by external sources, expresses difficulty in perceiving and solving problems, and relies on others rather than on himself. Because of these personality precursors typical internals and externals can be expected to desire and perceive different leadership styles. Internals will desire leaders who are relatively higher in participative decision-making and relatively lower in initiating structure than externals. There is no evidence that 45 internals and externals should differ in the extent in which they perceive their leader as considerate. Now, when these observations are integrated with the ones discussed on satisfaction it can be hypothesized that internals who are satisfied with their leaders will perceive them as high in participative decision-making and high on initiating structure. In a similar manner, it can be hypothesized that externals who are satisfied with their leaders will perceive them as low in participative decision-making and high in initiating structure. Once again, as with internals, the converse of this statement should also be true. Graphic representations of these predictions are displayed in Figure 2 for participative decision-making, Figure 3 for initiating structure and Figure 4 for consideration. To typify the highly dogmatic personality it may be said that he has a great respect for authority, has a closed belief system, is not a problem solver, is not respected as a sociometric choice, is not equalitarian or goal oriented, is rigid, and is greatly influenced by external influences that are perceived as powerful. Persons scoring low on dogmatism can be described as exhibiting an open belief system, are less influenced but more aware of external power sources, are problem solvers, are able to cope with change, and are concerned with equality and participation. As with locus of control, persons differing in dogmatism should desire and perceive leadership style differently. Low dogmatic persons who are satisfied with their leader will perceive their leaders as high in participative decision-making and low in initiating structure; while low dogmatic persons who are dissatisfied with their leader will perceive their leader as low in participative 46 .9:on 5.230 22092:?“ a ...oooo. m2 .0 cozooocoo m .3233 .53 of can 332.9, “.229. 335233 o:.oEo 3989.25 aflofioi .N 230E mmmmSmdS owkdjwm w._. ucoocoooo mom unavooemocH moanemucou xenon: cowumaouuou m Nanak 69 Table 4 The Multitrait Multimethod Matrix Used to Assess Construct Validity of the Johnson Three-Dimensional Leadership Questionnaire LBDQ Johnson 3-D Traits CON PDM IS CON PDM IS CON (849) LBDQ } PDM 84 (903) IS _ -32 -33 (681) CON h g t: "'77 T T T :23} (847) U \\ \ \ | uohnson PDM : o‘\\ 86 ‘ . \ 32', 81 (844) 3-D ; \ \ ‘ ~I IS I-1.2-_-..1l 1». g 27 17 (725) Notes. Traits CON, PDM and IS refer to Consideration, Participative Decision-making and Initiating Structure respectively. The values in parentheses form the reliability diagonal and the underlined values form the validity diagonal. Each heterotrait-monomethod triangle is enclosed by a solid line. Each heterotrait-heteromethod triangle is enclosed by a broken line. 70 this study generally meet this criterion. Although there is not a one to one correspondence among the correlations within the heterotrait triangles, they do fit the pattern of high correlation between CON and PDM, and much lower and similar correlations between these two dimensions and IS. This analysis has demonstrated strong, although not unequivocal, evidence for the construct validity of the Johnson measure. However, since the Johnson measure has been recently developed, further research is needed to confirm these conclusions. No notable differences were found between the Johnson measure and the LBDQ in terms of their psycho- metric properties. These measures appear to be assessing the same constructs with merely minor discrepancies in internal consistency and correlations among the dimensions. In as much as the LBDQ has been extensively used in leadership and the Johnson measure is relatively new, for the remainder of this study only the LBDQ will be used as the measure of perceived leadership behavior. Hypothesis Testing The hypotheses were tested utilizing the multivariate multiple regression anlaysis presented earlier in this thesis. In order to test the significance of variance accounted for by each of the predictor variables they were placed in sets in a manner suggested by Cohen (1968). The sets are as follows: A = {IE, DOG, SAT} B = {IE x DOG, IE x SAT, DOG x SAT} 0 = {IE x DOG x SAT} 71 where: IE = locus of control score DOG = dogmatism score SAT = satisfaction with the leader score A multivariate multiple regression equation was computed on Set A using the three leadership dimensions as the criteria. The multivariate F was significant, {(18, 1032.86) = 29.10, p_< .0001. Since the multivariate F was significant, a univariate multiple regression was performed for each criterion variable. All univariate F's were signi- ficant at the .001 level. When Set A was found to account for significant variance for all criterion variables, the individual predictor variables were "broken out" and tested for significance. The sets are tested for significance first, followed by the variables in an attempt to control the alpha levels over the complete analysis (Cohen, 1968). Three multivariate multiple regression equations were then computed on Set A with one predictor variable missing (RT°IE,DOG' RT'IE SAT; and RT'DOG SAT). The R2 was sub- , 9 tracted from the original R2 for the complete Set A to determine the increment of variance accounted for that can be attributed to each predictor 2 2 Z Z ..2. 2. ,. A variable (RY-Set A -RY'IE,DOG’ RY-Set A "RY-IE,SAT' RY-Set A -RY'DOG,SAT)- the increment attributed to each predictor was tested for significance by a F ratio computed from the formula described by McNemar (1962, p. 284). The variable locus of control accounted for significance variance for the CON criterion, F(l,370) = 3.87, E.< .05, but not for the other two criteria. Dogmatism accounted for significant variance for CON, E(l,370) - 100.20, p_< .001, and PDM, [(1,370) - 4.07, 2_< .05, p but was not significant for IS (F<:l). Satisfaction with the leader 72 accounted for significant variance for CON, F(l,370) = 595.84 p_< .OOl, PDM,F(1,370) = 497.52, p < .001, and IS, [(1,370) = 43.61, p_< .001. Since leadership style was viewed in terms of a multidimensional concept it was decided that if a predictor variable accounted for significant variance on at least one dimension it would be left in the regression analysis. Therefore, all predictor variables in Set A were maintained. Set B was then included in the analysis by computing a multivariate multiple regression equation for Set A and Set B combined and for Set B itself. The procedure outlined above was repeated and it was found that Set B did not account for a significant increment in the variance accounted for on any of the criterion variables. It was therefore dropped from the analysis. Set C, likewise was found not to contribute a signi- ficant increment in variance accounted for and was dropped. Presented in Table 5 is a summary of the final regression model, and Table 6 displays the final regression equations for each criterion variable. It can be noted from Table 5 that, in the regression equations for PDM and 18, both locus of control and dogmatism have negative regression weights while satisfaction with the leader has a positive regression weight. Therefore, pursuant to Darlington's (1968) definition, these two variables can be labeled as suppressors. Here locus of control and dogmatism improves the prediction of the criterion by subtracting out of the satisfaction with the leader variable variance which is not common with the criterion. The first hypothesis predicted that subordinate related variables would affect the subject's perception of CON. The regression analysis revealed that there was a significant relationship between the three subordinate related variables and the subject's perception of CON. 73 .oouumovo ecu Eoum omboamn Houowooum umnu sues coaumHouuoo OHeHuama osu mum m one x N Hammad emu sues mqoa. -oo. coco. Nwom. mama. osma. mace. emmo. mend. aowuowmmaumm Nmoo. mmoH. mHNM. nqoo. «can. mans. omoo. wHNo. cams. amwumawon Honuaoo mace. emoa. moan. Hooo. oawm. . much. «Hoe. memo. Noon. mo mauoa I need. comm. I Hamm. mmwu. I smuo. cams. HH< < 4 < who u on NM NM M um NM M NM NM m u HQ m ououopuum wsaumauwnH :owmwoonIo>Humm«uwuuom mowumuoofimdoo moumaom oumawvuonam moann«um> Homoeooum an» mo mHo>OA mouse pom mouoom :oHNOuwuo one: u wanna 76 between the predictors and criteria, the sample was independently divided into equal thirds on each predictor variable. Presented in Table 7 are the mean criterion score for each prediction variable. Inspection of this table reveals that subjects who were satisfied with the leader described that leader as higher in CON and PDM, and lower in IS than those who were less satisfied with the leader. Subjects who score low on the locus of control scale (internal) perceived the leader as higher in CON than did externals. Subjects who score lower on the Dogmatism scale described the leader as higher in CON and PDM than did those who were highly dogmatic. Hypothesis 6 predicted that subjects who were satisfied with the leader would describe him as higher in CON than those who were less satisfied with the leader. From Table 7 it can be Observed that this hypothesis was supported. Hypothesis 7 predicted that subjects who observed the authoritarian leader would describe him higher in IS and lower in CON and PDM than subjects who observed the participative leader. In order to test this hypothesis and to explore the possibility of significant interactions among the experimental conditions (type of leader that the subject observed) and subordinate related variables a multivariate multiple regression analysis was performed. Utilizing the same notation as in the first regression with the addition of condition (CO) the variables were placed in the following sets: A = {C0, IE, DOG, SAT} B - {CO x IE, co x DOG, CO x SAT, IE x DOG, IE x SAT, DOG x SAT} C . {CO x IE x DOG, CO x IE x SAT, CO x DOG, x SAT, IE x DOG x SAT} D = {C0 x IE x DOG x SAT} 77 The regression analysis was performed using exactly the same method as the previously reported analysis. Main effects for condition, dogmatism, and satisfaction with the leader were found to contribute significant unique variance. Sets B, C, and D (the interactive terms) were all nonsignificant and therefore dropped from the analysis. Table 8 presents a summary of this analysis. Main effects for condition, satisfaction with the leader, and dogmatism were the only variables to obtain a significant increment in the variance accounted for. The results of this analysis are similar with the first model with the major differences being the omission of locus of control as a predictor in the second analysis. When the experimental condition was added to the pre- dictors the increment in variance accounted for attributed to locus of control failed to reach statistical significance. This was due to the relatively small increment in R2 attributed to locus of control in both analyses and suggests that its relationship with the criterion is trivial. The mean score on each of the criterion variables within conditions is presented in Table 9. The subjects in the authoritarian condition described the leader significantly higher in IS and significantly lower in CON and PDM, therefore hypothesis 7 was confirmed. 78 ~00. v o «as no. v o « .oowunouo Ono scum oppose» oanmuun> use» sues sowumaauuoo oaofiuaas_onu one NM one m mw.H coo. oom. nom.fmm.c moo. amo. mam. mo.¢ moo. «on. amo. Emfiumawon needed .2: SE. HH.H moo. Hon. men. “wthH ooH. wNm. «mm. mmwaoN hmH. omm. Hen. defluummmwumm ou.hoa w.oH ooa. own. m¢.aonowuuuunm sowumuuowmaoo m mfimonuoehm mama Ou mamhamo< downpouwum moo mo humaaam m manna 79 Table 9 Mean Criterion Scores by Experimental Conditions Condition CON PDM IS Authoritarian 20.168 14.799 29.054 Participative 31.321 27.795 22.921 Overall 25.83 21.40 25.93 Subjects DISCUSSION Prior to the discussion of the findings a few observations must be made concerning the experimental procedures used in this study. The generalizability of lab experiments of leadership may be low for several reasons. It may be pointed out that the limited contact between the leader and subordinate prevents the variables from operating as they do in actual leadership situations where the relationship is shaped over an extended period of time. It can also be suggested that motivational structures operating in the work place cannot be duplicated in the lab. Security, pay, work involvement, ambition, and social factors are often cited as variables which do not lend themselves to duplication in artificial environments. These constituents, along with many others, have the effect of limiting the generalizability of lab obtained results in the study of leadership. These criticisms are all relevant to this study. Here, the subject's interaction with the leader was limited to observing that leader by means of a 25 minute videotape. It could be argued that this contact was too short and superficial for the subjects to form good impressions of the leader's behaviors. However, this would affect the variables of interest in this study in a conservative manner. It is inherently less likely to obtain significant results attributable to subordinate related variables in this type of lab study than in a field situation where the subordinates have had extensive contact with their superior. In addition, the relatively high reliability of the leadership measures suggest that subjects were able to form concrete 80 81 images of the leader's behavior. Nevertheless, the differential effects of subordinate related variables, in particular the personality character- istics, between simulated and authentic leadership conditions are here untested and therefore must be a limiting factor in the interpretation and generalizability of the results of this study. This study was undertaken in order to examine some major issues in the study of leadership. The first such issue concerns the number of dimensions that are needed to adequately describe a leader's behavior. Another question is the effect that subordinate related variables have on the perception of leadership style. The stimuli for this study were videotapes presenting two diverse styles of leaderships. The two styles of leadership can be described as bipolar since, in their construction, an attempt was made to depict two opposing styles. From both the data obtained in the manipulation check and careful observation of the tapes by persons familiar with the formal study of leadership, it can be confidently concluded that this objective was met. Since the method used in this study permitted the fabrication of definite leadership styles, which were corroborated by external sources, and were measured in the usual manner by the subjects responding to the LBDQ and the Johnson 3-D, it may be presumed that both objective and psychological measures of leadership were obtained. With measures in both these categories the construct validity of the leadership measures can be directly assessed within the limitations of the lack of scaling pre- cision referent to the objective measure of leadership. In as much as no attempt was made to construct a more sophisticated objective scale other than a high-low dichotomy, here, the discussion of validity must be limited to these restraints. 82 Within the participative condition the objective leadership style was evaluated to be: (1) high on CON, (2) high on PDM, and (3) low on IS; within the authoritarian condition objective leadership style was: (1) low on CON, (2) low on PDM, and (3) high on IS. The mean scores within conditions for the LBDQ scales reported in Table 9 demonstrate that the relative magnitude of dimensional scores are congruent with the objective appraisal. However, the scores are not as extreme as they were expected to be suggesting that, perhaps, central tendency and/or other factors are reducing discriminability of this instrument. Overall this examination confirms the construct validity of this measure. It should be noted that this analysis has dealt with mean scores of subjects observing the same leader and has completely ignored individual differences among the subjects' ratings. This issue is important and will be discussed later in this thesis. Construct validity of the measures employed in this study was also evaluated by means of the multimethod~multitrait matrix. As reported in the results section general support was found for both discriminant and convergent validity. Consequently, the construct validity of the measures was empirically verified. The results with respect to two of the dimension were, however, somewhat ambiguous. This was demonstrated by the high correlation between the dimension of CON and PDM in both measures. This connotes that, in this study, the subjects were unable to discriminate behaviors which are considered on an §_priori basis to belong in different dimensions. In most instances, leader behavior segments were, on a content basis, perceived as either both participative and considerate or not participative and not considerate. Ergo, these dimensions did not demonstrate evidence of discriminant validity. 83 Lowin, Hrapchak, and Kavanagh (1969), as previously noted, have hypothesized that the correlation between CON and PDM may be artificially high attributable to overlapping items. This hypothesis was not sup- ported by the present study. The Johnson 3-D measure was constructed in response to Lowin gt g1, contentions and the scales were purged of contaminant items. The similar high correlation between these dimensions (.84-LBDQ and .8l-Johnson-3D) on both instruments refuted this assertion as an explanation of the intercorrelation between the CON and PDM dimension. Stogdill (1969) conducted an experiment somewhat similar to the present study. Subjects observed movies of actors portraying different leadership styles. The movies were made availing prepared scenarios that depicted the leader acting out the pattern of behaviors described by the items in the scales of the LBDQ. 'Apparently no attempt was made to manipulate the level of a leadership dimension, as in this study, but rather comparisons were made among scores on the dimension the actor was portraying versus scores on the dimensions the actor was not portraying. This resulted in significant differences in mean scores between the CON and PDM dimensions. Stogdill failed to report the correlations between these dimensions and did not discuss the issue of discriminant validity. Nonetheless Stogdill concluded that the LBDQ can be regarded as valid under his experimental conditions. Schriesheim and Kerr (1974) have reviewed the aVailable studies that have used the LBDQ. Summarization of their data have shown constant correlations between the CON and PDM dimensions of .40 to .50. Because of these relatively high intercorrelations, they have concluded that these dimensions lack discriminant validity. 84 In the review of the literature segment of this thesis, a great deal of time was spent on reviewing and obtaining both theoretical and empirical evidence for three dimensions of leadership. It was fully expected by this author that three leadership dimensions with reasonably low intercorrelations would emerge. The results of the present research have not found this to be true. The present study along with the two studies summarized above (Schriesheim S Kerr, 1974; and Stogdill, 1968) exemplify the three most common methods of examining dimensions of leadership. In studies similar to Stogdill (1969) and Johnson (1973) attempts have been made to orthogonalize the dimensions by creating scripts, descriptions, or scenarios that are congruent with the measuring instrument. For example, leaders may be directed to behave high in CON and low in PDM and IS, which, in turn, yields the lowest correlations among the dimensions. This technique is subject to the criticism that the resultant combinations of behaviors are unrealistic and are unlikely to be actually exhibited by a genuine leader. Therefore results obtained from these studies have questionable external validity. Studies reviewed by Schriesheim and Kerr (1974) used actual sub- ordinates describing their superior. This method produced moderate correlations between these dimensions and is perhaps the most accurate. A criticism of the method of the studies reviewed by Schriesheim and Kerr is the absence of an objective measure of leadership. Also there is the considerable likelihood that confounding variables, exterior to the superior-subordinate relation, limit interpretation. In the present study the leader portrayed two opposing leaders. In as much as two idealized general leaderships were desired, no scripts were used and no attempt to conform to any §_priori dimensions were 85 undertaken. This method yielded an extremely high correlation between the CON and PDM dimensions. A pertinent criticism of this method is that actual leaders exhibit several different types of leadership behaviors while the leader in this study tended to be quite internally consistent in his leadership style. The differential results with respect to the intercorrelation of PDM and CON which are procured by use of these methods are of importance in determining the appropriateness of a third dimension of leadership. To date, no empirical research, excluding studies that orthogonalize the leadership style by manipulation, have obtained results that concur with the standard Campbell and Fiske criteria for discriminant validity. On the other hand, on the conceptual or theoretical level there seems to be sound arguments that it may be advantageous to evaluate leader- ship in terms of three dimensions (Yukl, 1971). However, the results of the present study as well as others concerned with the discriminant .validity of PDM and CON suggest that if one is dealing with leadership on an operational level, and if the assessment of leadership is for evaluative or predictive intent, then parsimony would dictate that the dimensions of CON and PDM be conceived as one and the same. The results of this study have generally supported past findings which suggest that subordinate related variables effect the perception of leadership style. In terms of the personality Variables, locus of control and dogmatism, it was found that both were significantly related to perception of leadership style. Externsls perceived the leaders as more considerate than did internals who observed the same leaders, but no differences were obtained for perception of PDM or IS. Highly dogmatic subjects perceived the leaders as lower in CON and PDM than did subjects 86 who were characterized as low in dogmatism. Once again no differences were found between level of dogmatism and perception of IS. However, the two personality variables together accounted for less than one percent of the variance attributable to linearity. This implies that while these variables were found to be statistically significant, their inclusion in the predictor set adds little to the accuracy of the predic- tion. It must be remembered, as stated previously, that the limited contact between the leaders and the subjects in the study may not have permitted the variables to operate in the same manner in which they do in real leadership situations. Within the limits of the above statement, it can be concluded that, although these personality variables are associated with perception of leadership style, the infinitesimal magni- tude of this relationship suggests that it is of little practical use. It was also ascertained that satisfaction with the leader was significantly related to perception of leadership style and accounted for approximately 60 percent of the variance. Table 7 demonstrates the greater the subject's satisfaction with that leader the higher he described that leader on CON and PDM and lower in IS. This could be interpreted as supporting past findings that subordinates tend to be more satisfied with high CON-PDM and low IS leaders, and less satisfied with low CON-PDM and high IS leaders. The inclusion of satisfaction with the leader as a subordinate related variable, thereby placing it within the predictor set rather than the criterion set, differentiates this study from most others in the field of leadership. Satisfaction with the leader is most commonly employed as a dependent variable and is envisioned as the result of the leader-subordinate relationship. However, the results of this study 87 indicated that the level of satisfaction with the leader may influence the subject's description of that leader. For example, subjects within the authoritative condition who were somewhat satisfied with the leader described him as higher in CON-PDM and lower in IS than did subjects in the same condition who were less satisfied with the leader. In a similar manner, subjects within the participative condition who were somewhat dissatisfied with the leader described him as lower in CON-PDM and higher in IS than did subjects in the same condition who were more satisfied with the leader. The net result of this is that two leaders who exhibited greatly diverse and opposing leadership styles received similar scores on the leadership dimensions. This effect is not apparent when the scores on several subordinates are averaged, since persons tend to be more satisfied with leaders high on PDM and CON. However, on an individual level, it can be stated that when assessment of leadership style is undertaken by obtaining subordinate descriptions of leader behavior, this measurement is confounded by the level of subordinate satisfaction with the leader. This also suggests that when this type of instrument is used, in lieu of measuring explicit leader behavior, as intended, what is really being assessed is the subordinates' global attitude toward the leader. In common leadership paradigms subordinates describe their super- visor on standardized instruments and then the supervisors are classified on the relevant leadership dimensions. This research demonstrates that the classification may be erroneous if the subordinates are a hetero- geneous group in terms of subordinate related variables. To obtain accurate measures of leader behavior, here defined as a psychological measurement which is congruent with a hypothetical objective measure of 88 leader behavior, relevant subordinate related variables must be utilized as covariates. This study has advocated satisfaction with the leaders as one such variable, but more may be defined through further research. Conclusions concerning the effects of leadership style on subordinate behaviors and attitudes (performance or satisfaction) that have been inferred from past research are suspect. Plausible alternative explana- tions of these findings can be made in terms of general subordinate attitudes toward the supervisor, rather than leadership style, since a case can be made that this is what was actually measured. Resolution of these questions may be obtained most efficiently by viewing leadership in terms of a contingency model. Here, leader behaviors, subordinate related variables, and situational variables are all considered important in defining the leader-subordinate relation and in predicting resultant outcomes. However, the results of this study suggest that personality characteristics of the subordinate are not productive variables for use in a contingency model. What seems to be most useful are subordinate attitudes that are specific to his or her job. Here, satisfaction with the leader was found to be an important variable in the perception of leadership style. Similar variables such as subordinate attitude toward the task, peer relations, and individual goals may prove to be meaningful subordinate related variables in a contingency model of leadership. Additional research in the form of field studies is essential in order to fully test‘the hypotheses supported in the present study. Particularly, the finding that the personality variables have an incon— sequent effect on the perception of leadership style needs to be corrobor- ated by subordinates and leaders in a work situation. Further investigation 89 must also be implemented to determine the feasibility of constructing an instrument that will exhibit discriminant validity when assessing CON, IS and PDM. Current evidence indicates that, despite sound theoretical arguments, attempts to operationalize independent CON and PDM scales have been, by and large, unsuccessful. Further investigation with other scales and in other situations should be undertaken before accepting CON and PDM as essentially unidimensional. Perhaps, the major issue that was left unanswered by this inquiry concerns the causal relationships among the subordinate related variables and perception of leaderShip styles. This study was able to determine that under laboratory controlled conditions, where style of leadership is held constant, that subjects who differed in satisfaction with the leader systematically differed in their perception of leadership style. However, whether the perceived style of leadership caused the level of satisfaction or whether the level of satisfaction caused the perceived leadership style is unknown. Longitudinal field studies are the most appropriate methodological procedure to resolve this query. The practical usefulness of this research lies in the confirmation of the formerly published findings that individual differences are important in the perception of leadership style. It will benefit a supervisor to be aware that not all of his/her subordinates perceive him or her identically. Supervisors must be cognizant of how each subordinate perceives them and must be able to interact uniquely with each individual to maximize their leadership potential. APPENDICES APPENDIX A The Desert Survival Situation 90 THE . DESERT SURVIVAL _ SITUATION . - A GROUP DECISION MAKING EXPERIENCE FOR EXAMINING 4i. - p AND INCREASING INDIVIDUAL AND TEAM EFFECTIVENESS f’lu .5! . 01.12 f. r .‘IE. ‘ ' ...“; , k. .-. Developed by ' i' '- I. ll. ' l . 3'“ ‘ ,,' «32‘ *2}, r ,7; w ‘W‘ 31,3137. _: (5%,, ., gt“ -1 -. 3 , f ”(it '.';_',':1. J. CLAYTON LAFFERTY, PHD. , f ‘3‘ ‘ j 7.? ‘ " , t a}? Consulting Psychologist , I . " , . -_ ., r ,,3; PATRICK M. EADY, men. ~ n . ' ‘ ' .1 ‘3; ‘ 39-. v." 1"» .K *I ' "‘ "H, :1} ' ' ‘N .. ‘-"\' I : ...”..g: ' '21? I?“ i. f ' ’ 3 " ":3",- - ’ ..‘ig‘ ‘ I‘- If . .. - . . . . ..‘Pj ’ 4m: , 31,. ‘ -. in consultation Wlth ‘--~~ 'u .‘ 3 5' 3 .5- ‘3: '- -_- if}; s z ALONZO W. POND, MA. I ..., - o -'.:‘.n"‘.".' ' " _ f ' _ I; T; .’ i‘ "5;: ., ; ,,’ a; Former Chief of Desert Branch \ " l‘" ‘3’" Arctic, Desert. Tr0pic Information Center .-. 7 Air University, Maxwell Air Force Base . '-.' .41.; .‘ . ,. 1&1:- n'. " The situation described in this exercise is based on over 2,000 actual cases in 3'6, 4 - ' ., 9 upon the survival decusions “ \, n , , ' . . p, -, lyour [cup ‘7 | .‘#.I_.‘s ‘ L ;' “13‘? Var-E4391. , _I.,‘_ ‘t 5,3}, . .'— .~ a. ‘\ .‘vv' . 91 THE SITUATION It is approximately 10:00 AM. in mid August and you have just crash landed in the Sonora Desert in southwestern United States. The light twin engine plane, containing the bodies of the pilot and the co-pilot, has completely burned. Only the air frame remains. None of the rest of you have been injured. The pilot was unable to notify anyone of your position before the crash. However, he had indicated before impact that you were 70 miles south-southwest from a mining camp which is the nearest known habitation, and that you were approximately 65 miles off the course that was filed in your VFR Flight Plan. ' The immediate area is quite flat and except for occasional barrel and saguaro cacti appears to be rather barren. The last weather report indicated the temperature would reach 110° that day, which means that the temperature at ground IeVel will be 130° . You are dressed in light weight clothing — short sleeved shirts, pants, socks and street shoes. Everyone has a handkerchief. Collectively, your pockets contain $2.83 in change, $85.00 in bills, a pack of cigarettes, and a ballpoint pen. YOUR TASK Before the plane caught fire your group was able to salvage the 15 items listed on the next page. Your task is to rank these items according to their importance to your survival, start- ing with “1" the most important, to "15" the least important. You may assume - the number of survivors is the same as the number on your team; you are the actual people in the situation; the team has agreed to stick together; all items are in good condition. .59.”? Step 1: Each member of the team is to individually rank each item. Do not discuss the situation or problem until each member has finished the individual ranking. Step 2: After everyone has finished the individual ranking, rank order the 15 items as a team. Once discussion begins do not change your individual ranking. Your team will have until—o'clock to complete this step. Step 1 Your ITEMS individual Ranking flashlight (4 battery size) 92 Step 2 The Team's Ranking Step 4 Difference Between Step 1 8i Step 3 iack knife sectional air map of the area plastic raincoat (large size). magnetic compass compress kit with gauze .45 caliber pistol (loaded) parachute (red and white) bottle of salt tablets (1000 tablets) 1 quart of water per person a book entitled, Edible Animals of the Desert a pair of sunglasses per person 2 quarts of 180 proof Vodka 1 top coat per person a cosmetic mirror Please complete the following steps and insert the scores under your team's number. TOTALS (the lower the score the better) Your Score Step 4 TEAM NUMBER 1 Z 3 4 Step 8 AVERAGE INDIVIDUAL SCORE Add up all the individual scores (Step 4) on the team and divide by the number on the team. Step 7 TEAM scone Step 8 GAIN SCORE The difference between the team score and the Average Individual Score. if the team score is lower than Avg. Ind. Score then gain is ”+". l f team score is higher than Avg. Ind. Score then gain is "—". Step 9 LOWEST INDIVIDUAL scone on the team ~ Sup‘“ NUMBER OF lNDlVlDUAL SCORES lower than the team score. 93 Synergistic Decision Making Process Resources Materials Man Math Natural Kno Facts and principles relating to the subject Plants, animals,the elements, terrain, raw materials, etc. Tools, facilities, products, etc. Interpersonal Skills Task Skills Rational Skills The Skills nacusary to perform a specific job; specific survival skills (i.e. fire building, hunting, etc.) The ability to deal with the situation rationally The ability to work with peeple The Interpersonal Skills . The Rational Skills Active Listening/Clarifying by: Building on others' ideas Decldng on a Rational Proeau Developing Alternative Coums Paying attention and responding Responding in an open, What issues need to be dealt of Action to others’ feelings and ideas spontaneous way with and in what sequence in What actions could possibly be Not interrupting Encouraging divergent points of order to arrive at a rational taken to achieve: Making open-ended inquiries view decision? 'tha minimum outcomes? Not judging othe's Freely offaing new ideas at “..sz the Situation 'the best outcomes? Summarizing and reflecting appropriate times Survivors' mental kiwi “Ck GEMS, Ith.‘ and Tall", mewmm w; manic" Supportinglauildng by: Continually focusing attention Materials on hand and their What would stand in the way of Accepting what other; have on the problem solving proces utilization taking each couru of action? to say Questioning own and others' Location What would be the adverse Not debating, persuading, assimptions in a non- Weather conditions consequences of each controlling or manipulating threatening way Surrounding environment alternative? others Dealing directly and specifically What are the teams' concerns? How likgly are they to occur? speaking in friendly, warm with apparent discrepancies How serious is each? How serious would it be if terms Reflecting on how the team is Setting Objectives they did? Creating opportunities for 90509 WI“) res-rd ‘03 What are the minimum bedding others to make their thoughts .9'09'0' _ outcom“ hoped for? Which alternative 5, most likely end feeling! known .oerml "hm“ What are the best outcomes that to achieve: Assuming other; m1 time ‘ can be reasonably hoped for? 'tha minimum outcomes? Wm" M. '"lommm- 0"» What are the probable 'the best outcomes? outcomes? 'the lust adverse consequences? Other matarids publishd by ELM: Subaru-c Survival Situation BIBLIOGRAPHY 9'0".“ 9"”an swam" LIN L. W. Listening Behavior. Englewood Cliffs, New Jersey, Prentice Hall 1971 M'mdq'mvm SM“ Cal E. Gregory, The Management of Intelligence: Scientific Problem Sowing, New York, l . _ Lear «ship and MotivatiOn McGraw-Hill, 1967 Learning Styles Inventory m, by Alonrow pm MA. Charles H. Kepnar and Benjamin B. Tregoa, The Rational Manager. New York, The Survrval Book, Funk and Wagnalls, N. Y. MCGraleII, 7965 . , . Sulvrval m sun and sand (3mm, Md Dual“). N, y Norman R. Meier, Prmclples of Human Relations. New York, John Wiley 8 Sons. 1952 Deserts. Grosser and Dunlap, N. v. George M. Prince. The Prect ice of Creativity, New York; Harper 3. Row, 1970 rususneo BY ELM 39819 Plymouth Rd., Plymouth, Michigan 48170 Experiential Learning Methods TEICPOOOC (313) 4594040 APPENDIX B Leadership in Group Problem Solving Questionnaire LEADERSHIP IN GROUP PROBLEM SOLVING QUESTIONNAIRE This questionnaire consists of 104 questions divided into 4 parts. Each part contains its own directions. Please read the directions carefully and record your answers on the computerized scoring sheets by filling in the box which cor- responds to your answer. For Parts I and II use the purple scoring sheet and for Parts III and IV use the blue scoring sheet. PLEASE BE SURE TO ANSWER ALL QUESTIONS THANK YOU. 94 95 PART I 1 This is a questionnaire to find out the way in which certain important events in our society affect different people. Each item consists of a pair of alternatives, numbers 1 or 2. Please select the one statement of each pair (and only one) which you more strongly believe to be the case as far as you're concerned. Be sure to select the one you actually believe to be more true rather than the one you think you should choose or the one you would like to be true. This is a measure of personal belief: obviously there are no right or wrong answers. Please answer these items carefully, but do not spend too much time on any one item. Be sure to find an answer for every choice. In some instances, you may discover that you believe both statements or neither one. In such cases, be sure to select the one you more strongly believe to be the case as far as you're concerned. Also try to respond to each item independently when making your choice; do not be influenced by your previous choices. 1. Your sex is 1. male 2. female 2. 1. Children get into trouble because their parents punish them too much. 2. The trouble with most children nowadays is that their parents are too easy with them- a 3. 1. Many of the unhappy things in people's lives are partly due to bad luck. 2. People's misfortunes result from the mistakes they make. 4. 1. One of the major reasons why we have wars is because people don't take enough interest in politics. 2. There will always be wars, no matter how hard people try to prevent them. 5. 1. In the long run, people get the respect they deserve in this world. 2. Unfortunately, an individual's worth often passes unrecognized no matter how hard he tries. 6. 1. The idea that teachers are unfair to students is nonsense. Most students don't realize the extent to which their grades are influenced by accidental happenings. 7. 1. Without the right breaks one cannot be an effective leader. 2. Capable people who fail to become leaders have not taken advantage of their opportunities. 8. 1. No matter how hard you try some people just don't like you. 2. People who can't get others to like them don't understand how to get along with others. ’ 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 96 2 Heredity plays the major role in determining one's personality. It is one's experiences in life which determine what one is like. I have often found that what is going to happen will happen. Trusting to fate has never turned out as well for me as making a decision to take a definite course of action. In the case of the well prepared student, there is rarely if ever such a thing as an unfair test. Many times exam questions tend to be so unrelated to course work that studying is really useless. Becoming a success is a matter of hard work, luck has little or nothing to do with it. Getting a good job depends mainly on being in the right place at the right time. The average citizen can have an influence in government decisions. This world is run by the few people in power, and there is not much the little guy can do about it. When I make plans, I am almost certain that I can make them work. It is not always wise to plan too far ahead because many things turn out to be a matter of good or bad fortune anyhow. There are certain people who are just no good. There is some good in everybody. In my case, getting what I want has little or nothing to do with luck. Many times we might just as well decide what to do by flipping a coin. Who gets to be the boss often depends on who was lucky enough to be in the right place first. Getting people to do the right thing depends upon ability, luck has little or nothing to do with it. As far as world affairs are concerned, most of us are the victims of forces we can neither understand, nor control. By taking an active part in political and social affairs, the people can control world events. Most people don't realize the extent to which their lives are controlled by accidental happenings. There really is no such thing as "luck." One should always be willing to admit mistakes. It is usually best to cover up one's mistakes. It is hard to know whether or not a person really likes you. How many friends you have depends on how nice a person you are. 22. 1. 2. 23. 1. 2. 24. 1. 2. 25. 1. 2. 26. 1. 2. 27. 1. 2. 28. 1. 2. 29. 1. 2. 30. 1. 2. PART II 97 3 In the long run, the bad things that happen to us are balanced by the good ones. Mbst misfortunes are the result of lack of ability, ignorance, laziness, or all three. With enough effort, we can wipe out political corruption. It is difficult for people to have much control over the things politicians do in office. Sometimes I can't understand how teachers arrive at the grades they give. There is a direct connection between how hard I study and the grades I get. A good leader expects people to decide for themselves what they should do. A good leader makes it clear to everybody what their jobs are. Many times I feel that I have little influence over the things that happen to me. It is impossible for me to believe that chance or luck plays an important role in my life. People are lonely because they don't try to be friendly. There's not much use in trying too hard to please people, if they like you, they like you. There is too much emphasis on athletics in high school. Team sports are an excellent way to build character. What happens to me is my own doing. Sometimes I feel that I don't have enough control over the direction my life is taking. Most of the time I can't understand why politicians behave the way they do. In the long run, the people are responsible for bad government on a national as well as on a local level. On the next page there are twenty statements people have made as their opinion on several topics. You may find yourself agreeing strongly with some of the statement, disagreeing just as strongly with others, and perhaps uncertain about others. Whether you agree or disagree with any statement, you can be sure that many other people feel the same as you do. 98 We want your personal opinion on each statement. When you read each statement, mark in the appropriate box whether, in general, you: . Agree very much Agree on the whole . Agree a little Disagree a little . Disagree on the whole . Disagree very much O‘U‘J-‘UJNH 31. In this complicated world of ours, the only way we can know what's going on is to rely on leaders or experts who can be trusted. 1 2 3 32. My blood boils whenever a person stubbornly refuses to admit he's wrong! 1 2 3 33. There are two kinds of people in this world: those who are for the truth and those who are against the truth. 1 2 3 34. Mbst people just don't know what's good for them. 1 2 3 35. Of all the different philosophies which exist in this world there is probably only one which is correct. 1 2 3 36. The highest form of government is a democracy and the highest form of democracy is a government run by those who are most intelligent. l 2 3 37. The main thing in life is for a person to want to do something important. 1 2 3 38. I'd like it if I could find someone who would tell me how to solve my personal problems. 1 2 3 39. Most of the ideas which get printed nowadays aren't worth the paper they are printed on. 1 2 3 40. Man on his own is a helpless and miserable creature. 1 2 3 41. It is only when a person devotes himself to an ideal or cause that life becomes meaningful. l 2 3 42. Most people just don't give a "damn" for others. 1 2 3 43. To compromise with our political opponents is dangerous because it usually leads to the betrayal of our own side. 1 2 3 44. It is often desirable to reserve judgment about what's going on until one has had a chance to hear the opinions of those one respects. 1 2 3 45. 46. 47. 48. 49. 50. 99 . Agree very much Agree on the whole Agree a little Disagree a little Disagree on the whole . Disagree very much O‘U14§UJNl-' The present is all too often full of unhappiness. it is only the future that counts. The United States and Russia have just about nothing in common. In a discussion I often find it necessary to repeat myself several times to make sure I am being understood. While I don't like to admit this even to myself, my secret ambition is to become a great man, like Einstein, or Beethoven, or Shakespeare. Even though freedom of speech for all groups is a worthwhile goal, it is unfortunately necessary to restrict the freedom of certain political groups. It is better to be a dead hero than to be a live coward. 3 4 5 3 4 5 3 4 5 3 4 5 3 4 5 3 4 5 100 PART III Instructions: Imagine yourself as the sixth member of the group that you just watched on the television. Keeping this in mind, please answer the following questions by marking the response which best fits your own reaction. Strongly agree Agree somewhat Neither agree nor disagree Disagree somewhat . Strongly disagree M§WNH l. I felt inhibited from expressing my feelings during the group discussion. 1 2 3 4 5 2. The group needed a strong leader to keep on the track. 1 2 3 4 5 3. I felt tense and uncomfortable in this group. 1 2 3 4 5 4. Some people in the group talked too much. 1 2 3 4 5 5. I felt that I could trust the leader of the group. 1 2 3 4 5 6. The leader of the group was annoying. 1 2 3 4 5 7. The leader of the group contributed to its success. 1 2 3 4 5 8. The leader of the group was willing to make changes. 1 2 3 4 5 9. I was generally satisfied with the leadership of my group. 1 2 3 4 5 PART IV On the following pages is a list of items that may be used to describe the actual behavior of the group leader you just watched on television. Each item describes a specific kind of behavior, but does not ask you to judge whether the behavior is desirable or undesirable. Although some items may appear similar, they express differences that are important in the description of leadership. Each item should be considered as a separate description. This is not a test of ability or consistency in making answers. It's only purpose is to make it possible for you to describe, as accurately as you can, the behavior of the group leader. Directions: A. READ each item carefully. B. THINK about how frequently the leader actually engaged in the behavior described by the item. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 101 C. DECIDE whether he (1) always (2) often (3) occassionally (4) seldom, or (5) never acts as described by the item. D. MARK one of the five numbers to show the answer you have 1 = Always 2 = Often 3 = Occasionally 4 = Seldom 5 = Never He lets group members know what is expected of them. 1 He allows the members complete freedom in their work. 1 He is friendly and approachable. 1 He encourages the use of uniform procedures. 1 He permits the members to use their own judgement in solving problems. 1 He does little things to make it pleasant to be a member of the group. 1 He tries out his ideas in the group. 1 He encourages initiative in the group members. 1 He puts suggestions made by the group into operation. 1 He makes his attitude clear to the group. 1 He lets the members do their work the way they think best. 1 He treats all group members as his equals. 1 He decides what shall be done and how it shall be done. 1 He assigns a task, then lets the members handle it. 1 He gives advanced notice of changes. 1 He assigns group members to particular tasks. 1 He turns the members loose on a job, and lets them go to it. 1 He keeps to himself. 1 He makes sure that his part in the group is understood by the group members. 1 selected. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. He He He He He He He He He 102 = Always Often Occasionally Seldom Never U14>WNH is reluctant to allow the members any freedom of action. 1 looks out for the personal welfare of group members. schedules the work to be done. allows the group a high degree of initiative. is willing to make changes. maintains definite standards of performance. trusts the members to exercise good judgement. refuses to explain his actions. asks that group members follow standard rules and regulations. He He He He He He He He If permits the group to set its own pace. acts without consulting the group. is friendly and easily approached. allows me to modify the procedures required in my job. lets me know what is expected of me. expresses appreciation when I do a good job. has me share in decision making. lets me know what is expected of me. I am a new member he helps me make adjustments to the group. He assigns a task, then lets me handle it. He makes sure that I and other group members understand his part in the group. He He makes me feel at ease when talking with him. allows me influence equal to his own in decisions which affect my job. 1 l 51. 52. 53. 54. 103 l = Always 2 = Often 3 = Occasionally 4 = Seldom 5 = Never He maintains definite standards of performance for me. He looks out for my personal welfare. He lets me do my work the way I think best. He sees to it that I have the material I need to work with. 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 REFERENCES References Adams, J.S. Injustice in social exchange. In 1. Berkowitz (Ed.), Awareness ip_gxperimental social psychology. Vol. 2 New York: Academic Press, 1965. Adorno, T., Frenkel-Brunswik. E., Levinson, D., & Stanford, R.M. The authoritarian personality. New York: Narper and Brothers, 1950. Allport, F.H. The structuring of events: Outline of a general theory with applications to psychology. Psychological Review, 1954, 61, 281-303. Anderson, L.R. Leader behavior, member attitudes, and task performances of intercultural discussion groups. Journal pf_Social Psychology, 1966, 62, 305-319. Argyle, M., Gardner, 6., & Cioffi, F. The measurement of supervisory methods. Human Relations, 1957,.19, 295-313. Bales, B.F. Task roles and social roles in problem-solving groups. In E.E. Maccoby, T.M. Newcomb, and E.L. Hartley (Eds.), Readings ip_ social psychology (3rd ed.), New York: Holt, Rinehart, and Winston, 1958, pp. 437-447. Beer, M. Leadership, employee needs, and motivation. Columbus: Bureau of Business Research, Ohio State University, Monograph, No. 129, 1966. Beer, M. Organizational size and job satisfaction. Academy pf_Management Journal, 1964, 1, 34-44. Bowers, D. and Seashore, S. Predicting organizational effectiveness with a four-factor theory of leadership. Administrative Science Quarterly, Campbell, D.T., & Fiske, D.W. Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 1959, g9, 81-105. Carroll, B. Job satisfaction. Ithaca, New York: New York State School of Industrial and Labor Relations, A Statutory College of the State University, Cornell University, 1973. Cartwright, D., & Zander, A. (Eds.) Grogp dynamics: Research and theory. Harper and Row, 1968. Cohen, J. Multiple regression as a general data-analysis system. Psychological Bulletin, 1968, ZQ_(6), 426-443. 104 105 Crowne, D.P., & Marlowe, D. The ppproval motive. New York: Wiley, 1964. Darlington, R.B. Multiple regression in psychological research and practice. Psychological Bulletin, 1968, 92, 161-182. Drackman, D. Dogmatism prenegotiation experience, and simulated group representation as determinants of dyadic behavior in a bargaining situation. Journal pf_Personality and Social Psychology, 1967, 6, 279-290. Dustin, D.S., & Davis, H.P. Authoritarianism and sanctioning behavior. Journal of_Personalipy and Social Psychology, 1967, p, 222-224. Fiedler, F.E. A contingency model of leadership effectiveness. In L. Berkowitz (Ed.), Advance ip_experimental social psychology, Vol. 1, New York: Academic Press, 1964, pp. 149-190. Fiedler, F.E. .A.theo£y pf_leadership effectiveness. New York: McGraw- Hill, 1967. Fleishman, E.A. Manual for leadership opinion questionnaire, (1969 re- vision), Science Research Associates, Inc., 1969. Fleishman, E.A. The description of supervisory behavior. Journal of (Applied Psychology, 1963, 36, 1-6. Fleishman, E.A., & Harris, E.F. Patterns of leadership behavior related to employee grievances and turnover. Personnel Psychology, 1962, 12, 43-56. Fruchter, F., Rokeach, M., & Novak, E. A factorial study of dogmatism, opinionation, and related scales. Psychological Reports, 1958, 4, 19-22. Gibb, C.A. Leadership. In G. Lindzey, Handbook p£_social psychology. Cambridge, Mass.: Addison-Wesley, 1954. Goodnow, J.J. & Pettigrew, T.F. Effects of prior patterns of experience upon strategies and learning sets. Journal p£_Experimental Psychology, 1955, 42, 381-389. Halpin, A.W. & Winer, B.J. A factorial study of the leader behavior descriptions. In R.M. Stogdill and A.E. Coons (Eds.) Leader behavior: Its description and measurement. Columbus: Bureau of Business Research, Ohio State University, 1957. Halpin, A.W. & Winer, B.J. The leadership behavior of the airplane commander. Columbus: Ohio State University Research Foundation, 1952. Hanson, D.J. Dogmatism and authoritarianism. Journal pf_Socia1 Psychology, 1968, 26, 89-95. 106 Haythorn, W.W., Couch, A., Haefner, D., Langham, P., and Carter, L.F. The effects of varying combinations of authoritarian and equalitarian leaders and followers. Journal p£_Abnormal and Social Psychology, 1956, 55, 210-219. Heller, F., and Yukl, G. Participation, managerial decision-making, and situational variables. Organizational Behavior and Human Performance, 1969, A, 227-241. Hemphill, J.K. & Coons, A.E. Development of the leader behavior description questionnaire. In R.M. Stogdill and A.E. Coons (Eds.), Leader behavior: Its descripsion and measurement. Columbus: Bureau of Business Research, Ohio State University, 1957. Herold, D.M. Interaction of subordinate and leader characteristics in moderating the consideration-satisfaction relationship. Journal 2: Applied Psychology, 1974,.52, 649-651. Holden, K.B. Interaction of subordinate and leader characteristics in moderating the consideration-satisfaction relationship. Journal 25 pApplied Psychology, 1974, 55, 649-651. Hood, P.D. Leadership climate for trainee leaders: The army AIT platoon. washington, D.C.: George Washington University, Human Resources Research Office, 1968. House, R.J. A path goal theory of leader effectiveness. Administrative Science Quarterly, 1971, 16(3), 321-338. House, R.J., Filley, A.C., & Gujatati, D.M. Leadership style, hierarchical influence, and the satisfaction of subordinate role expectations: A test of Likert's influence proposition. Journal p£_Applied Psychology, 1971, 55, 422-432. Hummel, T.J. & Sligo, J.P. Empirical comparison of univariate and multi- variate analysis of variance procedures. Psychological Bulletin, 1971, Z_6_, 49-57. James, W.H., & Rotter, J.B. Partial and 100% reinforcement under chance and skilled conditions. Journal p£_Experimental Psychology, 1958, 55, 397-403. Joe, V.C. Review of the internal-external control constructed as a personality variable. Psychological Repprts, 1971, Monograph Supplement 3-V28, 619-640. Johnson, R.B. Initiating structure, consideration, spd psrticipation 1p_ decision making: Dimension 2: leader behavior. Unpublished Ph.D. Thesis, Michigan State University, 1973. Kahn, R.L., & Katz, D. Leadership practices in relation to productivity and morale. In D. Cartwright and A. Zander (Eds.) Group dynamics. New York: Harper and Row, 1953. 107 Kaplan, M.F., & Singer, E. Dogmatism.and sensory alienation: An empirical investigation. Journal'pf_Consultipg_Psychology, 1963, 21, 486-491. Katz, D., & Kahn, R.L. Human organization and worker motivation. In L.R. Tripp (Ed.) Industrial productivity, Madison, Wisc.: Industrial Relations Research Association, 1951, pp. 146-171. Katz, D., Maccoby, N., & Morse, M.C. Productivit , supervision and morale is sp_office situation. Ann Arbor, University of Michigan, Institute for Social Research, 1950. Keeler, B.T. & Andrews, J.H.M. Leader behavior of principals, staff morale and productivity. Alberta Journal 9; Educational Resources, 1963, 2, 179-191. Kemp, C.G. Influence of dogmatism on counseling. Personnel and Guidance Journal, 1961, 22, 662-665. Kerlinger, F., & Rokeach, M. The factorial nature of the F and D scales, Journal s§_Personality and Social Psycholsgy, 1966, A, 391-399. Korman, A.K. Consideration, initiating structure, and organizational criteria a review. Personnel Psychology, 1966, 52, 349-361. Lawler, E.E. Motivation gp_work organizations, Monterey, California: Brooks/Cole Publishing Company, 1973. Lefcourt, H.M. Internal versus external control of reinforcement: a review. Psyshological Bulletin, 1966, 55, 206-220. Lewin, K. The dynamics of group action. Educational Leadership, 1944, A, 195-200. Likert, R. New patterns s£_management. New York: McCraw-Hill, 1961. Lowin, A., Hrapchak, W.J., & Kavanagh, M.J. Consideration and initiating structure: An experimental investigation of leadership traits. Administrative Science_Quarterly, 1969, 55, 238-253. Lowman, R.L. Secretarial job enrichment in a Scanlon Plan organizational climate. unpublished Masters Thesis, Michigan State University, 1974. Mann, F.C. Toward an understanding of the leadership role in formal organizations. In R. Dubin, G.C. Homans, F.C. Mann, D.C. Miller (Eds.), Leadership and productivipy. San Francisco: Chandler, 1965. MbMurray, R.M. The case for benevolent autocracy. Harvard Business Review, 1958, 25, (1), 82-90. McNemar, W. Psychological statistics, (3rd. ed.), New York: Wiley, 1962. Minton, H.L. Power as personality construct. In E.A. Maher (Ed.), Progress 5p_experimental personality research, Vol. 4, New York: Academic Press, 1967. 108 Mitchell, T.R., Smyser, C.M. & Weed, S.E. Locus of control: Supervision and work satisfaction. Acadspy’pffManagement‘Journal, 1975, 5, 624-631. Morse, W.C. Satisfaction ip_the white collar job. Ann Arbor, Michigan: Institute for Social Research, 1953. Newcomb, T.M. The acguaintance pgpcess. New York: Holt, Rinehart, and Winston, 1961. Norman, R.P. Dogmatism and psychoneurosis in college women. Journal 92_ Consulting Psychology, 1966, 59, 278. Overall, J.E., 8 Kleet, C. Applied multivariate analysis. New York: McGraw-Hill, 1972. Overall, J.E. & Spiegel, D.K. Concerning least squares analysis of experimental data. Psychological Bulletin, 1969, 12, 311-322. Pelz, D.C. Influence: A key to effective leadership in the first-line supervisor. Personnel, 1952, 22, 209-217. Phares, E.J.a Expectancy changes in skill and chance situations. Journal s£_Abnorma1 and Social Psychology, 1957, 55, 399-342. Phares, E.J. Perceptual threshold decrements as a function of skill and chance expectancies. Journal p£_Psychology, 1962, 52, 399-407. Plant, W.T. Rokeach's dogmatism scale as a measure of general authori- tarianism, Psychological Reports, 1960, 5, 164. Ralsky, S.L. The effects of leadership style on some aspects of group growth. Unpublished Ph. D. Thesis, Michigan State University, 1975. Robinson, J.P., & Shaver, P.R. Measure 92 social psyphological attitudes. Ann Arbor, Michigan: Institute for Social Research, the University of Michigan, 1973. Rokeach, M; Dogmatism and opinionation on the left and on the right. American Psychologist, 1952, Z, 310. (abstract) Rokeach, M. The nature and meaning of dogmatism. Psychological Review, 1954, 1;, 194-204. Rokeach, M. The ppen and closed mind. New York: Basic Books, 1960. Rotter, J.B. Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 1966, 29, 1-27. (1 whole No. 609) Rotter, J.B. Social learning and clinical psychology. New York: Prentice-Hall, 1954. 109 Rotter, J.B. Some implications of a social learning theory for the pre- diction of goal directed behavior from testing procedures. Psychological Review, 1960,.51, 301-316. Rotter, J.B. Some problems and misconceptions related to the construct of internal versus external control of reinforcement. Journal 2: Consulting and Clinical Psychology, 1975, 32, 56-67. Rotter, J.B. The role of the psychological situation in determining the direction of human behavior. In M.R. Jones (Ed.), Nebraska sypposium.sp_motivation, 1955, Lincoln; University of Nebraska Press, 1955, 245-269. Rotter, J.B., Liverant, 8., & Crowne, D.P. The growth and extinction of expectancies in chance controlled and skilled tests. Journal of_ Psychology, 1961, 52, 161-177. Runyon, K.B. Some interaction between personality variables and management styles. Journal stApplied Psychology, 1973, 52, 288-294. Sales, S.M. Supervisory style and productivity: Review and theory. Personnel Psycholsgy, 1966, 22, 275-286. Schriesheim, C. & Karr, S. Psychometric properties of the Ohio State leadership scales. Psychological Bulletin, 1974.125, 756-765. Singh, S.W. & Arya, H.P. Value-orientations of local village leaders. Manas, 1965, 22_(2), 145-156. Stanford, F.H. Authoritarianism and leadership s study s£_the follower's orientation pp authority. Philadelphia: Institute for Research in Human Relations, 1950. Stogdill, Ralph,.M. Handbook s§_leadership: .A survey p£_theory and research. New York: The Free Press, 1974. Stogdill, R.M. Individual behavior and grasp achievements. New York: Oxford University Press, 1959. Stogdill, R.M. Manual for the leader behavior description spestionnaire- Form XII. Bureau of Business Research, College of Commerce and Administration, Ohio State University, 1963. Stogdill, R.M. Validity of leader behavior description. Personnel Psychology, 1969, 22, 153-158. Stogdill, R.M., Goode, O.S., & Day, D.R. New leader behavior description sub scales. Journal p£_Psychology, 1962, 53, 259-269. Stogdill, R.M., Goode, O.S., & Day, D.R. The leader behavior of presidents of labor unions. Personal Psychology, 1964, 21, 49-57. 110 Stogdill, R.M., Goode, 0.8., & Day, D.R. The leader of corporation presidents. Personal Psychology, 1963,.19- 127-132. Tannenbaum, A.S., & Allport, F.H. Personality structure and group structure: An interpretive study of their relationship through an event structure hypothesis. Journal of Abnormal Social Psychology, 1956, 55, 272-280. Tannenbaum, R., & Schmidt, W.H. How to choose a leadership pattern. Harvard Business Review, 1958, 55, 95-101. Throop, W.E., & MacDonald, A.P., Jr., Internal-external locus of control; “a bibliography. Psychological Reports, Mbnograph Supplement, 1-528, 1971, pp. 175-190. Troldahl, V.C. & Powell, E.A. A short-form dogmatism scale for use in field studies. Social Forces, 1965, AA, 211-214. Vroom, V.H. Some personality determinants o; the effects o£_psrticipation. Englewood Cliffs, New Jersey: Prentice-Hall, 1960. Vroom, V.H. Some personality determinants of the effects of participation. Journal o£_Abnormal and Social Psychology, 1959, 52, 322-327. Vroom, V.H. WOrk and motivation. New York: Wiley, 1964. Vroom, V.H. & Mann, F.C. Leader authoritarianism and employee attitudes. Personnel Psychology, 1960, 55_(2), 135-140. Vroom, V.H. and Yetton, P.W. Leadership and decision-making, University of Pittsburgh Press, 1973. vacchiano, R.B., Strauss, P.S., & Schiffman, D.C. Personality correlates of dogmatism. Journal o£_Consulting and Clinical Psychology, 1968, 83-85. Wagner, L.W. Leadership style, hierarchical influence, and supervisory ‘ role obligations. Administrative goiencegguarteriy, 1965, 2, 391-420. Weissenberg, P., & Kavanagh, J. The independence of initiating structure and consideration: Review of the evidence. Personal Psychology, 1972, 25 (1), 119-130. Wilkins, B.J., & DeCharms, R. Authoritarianism and response to power cues. Journal o£_Personality, 1962, 55, 439-457. Yukl, C.A. Toward a behavioral theory of leadership. Organizational Behavior and Human Performance, 1971, 5, 414-440. Zagona, S.V., and Zurcher, L.A. Participation, interaction, and role behavior in groups selected from the extremes of the open-closed cognitive continuum. Journal g£.Psychology, 1964, 55, 255-264. HICHIGRN STATE UNIV. LIBRQRIES 31293103533745