fn‘I -'I1< :Iil’vi"§!;iiy§it;I_;II.I;I;I«.-'I i3: INITIATING STRUCTURE, CONSIDERATION; AND PARTICIPATIVE DECISION MAKING: DIMENSIONS OF LEADER BEHAVIOR . ‘ ' O‘Issertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY RAYMOND HOWARD JOHNSON 1973 IIIIIIII IIIIIIIIIIIIIIIII Y" , I . v-gm...wumuumwcwae< .mfinmuoomom mo hpoosu Houoom Mao“ m Sufi? unmam>wuuoww0 amcowuunwcwwuo wnwuuwvoum .uhosmmmm .N .m van mumkbm .0 .n “muuaom _ co m >uoa=o .4 1 maaaxu a a «o M U Acousnuoa museums m U gang-ago ucul do w 1 . u cowu, _ unansuaa Io>ua£ua.lllllllllllllt ouauushuu 1 1 M .oueuaaoau anon maaunauaau nasaaaumuw . oaou anon 1 1328p «535 Domaiuasn __ n no acquuuu _ m Isuumuuwn . 1 nsouuunbu madame sowuousouuo noabnaun . us0l0>oanuu o>wuouu unulufisuuuilllllllllllL _ I luwnfiah< souuosvoum «>«uuunao m n o nuwoa ” ooaoaalu w a unusuavo: w a no 1 occuauowuoa manage-u sausages 1 Ian“! use-cuauuu nouuusvoum Hoou W doom cu some mdwuauushum w uow>wnoa. , noduw>uoa=n amaze nomu sown m we avonuul Icowuumou huw>uuqmsom Iwuaauuau Iauwdwuou M nacho nacho m sowuuauousw scuuucuouan “ _ naouu 1 uncuuunSu «Hausa . M ousdnuunuua naouuaaou w asouu sols: aownu>uunsu 1 Imago sowu no 1 a Iaoaudaou Iuomaauma noonuooHu no uwuao no so as usuammanm . m abfluuoamsa woos flecho was no «mo Iuo Byline uncannm no uuouue H I ca 0 n u «Addendum uauvu>oum coauuusowuo canmnuuauu: oohqunfl Aommuv Aanmnv Aomaav Annaav Annmav Aeomav novaoN van ”wwafiw Auoaav Ammoww anus .Hn no soda: «@000 1nd oposnaom uauauiuulo x can: on non nun: «you vac swede: Hnwnunum was amazon .muouomwumo>su ucouowwwn mo ounooaoo awnuuovooq mo oocoosoamouu00uu.H manna ‘ w no.5? u "Av. l3 3-space solution and the strong correlations among the variables making up the dimensions. In addition, one may argue that a 2-space solution may have provided an adequate fit, judging from the coefficients of alientation presented by Taylor (1971, p. 256, 261). Other Approaches WOfford (1967) developed six theoretical dimensions of managerial behavior: 1. Security and maintenance. Behaviors concerned with maifitaining a neat environment, protecting and arranging the physical facilities, and assuring safety of the work situation. 2. Order and structure. Behaviors that are orderly, systematic, and well Structured. 3. Personal interaction. Behaviors involving social interaction, supportiveness, and acceptance of support. 4. Achievement. Behaviors relating to goal setting, seIféimprovement, independent advancement in work, and innovation. 5. Personal enhancement. Behaviors which seek to gain status, power, recognition. 6. Group achievement. Behaviors associated with team action, group'leadership, morale, group goals, and .group advancement. Factor analyzing 183 manager behavior items, written to reflect the six dimensional theory, Wofford (1970, 1971) found five orthogonal factors which explained a remarkable 14 100% of the variance. The five factor labels and their contributions to explainable variance follow: Order and group achievement (40.5%) Personal enhancement orientation (22.2%) Personal interaction (17.2%) Security and maintenance (11.3%) Ul-L‘LJDNH Dynamic and achievement oriented (8.8%) These labels are deceiving in that after inspecting the items loading on each of the factors, one puzzles over the origin of the labels, i.e., the Wofford theory (1967) or the structure of the data that was factor analyzed (Wofford, 1970). In fact, most of the items loading on a given factor have little face validity. Three reasons explain the absence of clearly inter- pretable factors. First, a factor analysis of 183 items based on a sample size of 136 is capitalizing to a great extent on chance. Second, tetrachoric correlations were the basis of the factor analysis, despite the fact it is a poor estimate (unreliable) of the Pearson product correla- tion and that no mathematical basis exists for employing tetrachoric correlation in multivariate analyses (Nunnally, pp. 123-124). This inappropriate correlation may explain three other peculiar results in the Wofford factor analysis: (1) unusually high factor loadings; (2) the sum of the Intrei . I u o" . "OI-'1'“: V"wlu n "4"“. .1."- :....ep q I-I p end-I is (“I 9... 15 squared factor loadings for a given item far exceeds its communality; (3) 100% of the variance was explained by five factors, unheard of in psychological research, since this event implies merfectly reliable items. Third, the theory (Wofford, 1967) and the items written to reflect it are conceptually murky . This last series of research by Wofford (1967, 1970, 1971), because of serious conceptual and methodological limitations, will not be given further attention. Fiedler (1967) presents a theory of leadership effectiveness that is indirectly related to the notion of leader behavior dimensions. Instead of discussing leader behavior, Fiedler's key concept is leadership style, defined as the underlying need-structure of the individual that motivates his behavior in various situations. The polar eXtremes of the leadership style dimension are: task oriented Versus relationship oriented. Fiedler operationalized his c-‘—<>l':1¢ept of leadership through the Least-Preferred Co-Worker instrument (LPC), the details and measurement problems of which will not be presented here. For a summary see Graen, 0:li‘tris, and Alvares (1971). The important points about the LPC are that: (1) it does not measure leader behavior directly; (2) it is not known exactly what is being measured; (3) presumably, it measures a single dimension of leader- Sflip, as opposed to the two relatively independent Ohio State I07}. ui-o‘) n.1, l'ozL‘ 16 leadership behavior dimensions. Attempts to find behavioral correlates of LPC scores have yielded inconsistent results (Sample and Wilson, 1965; Fiedler, 1967, p. 53; Nealey and Blood, 1968; Yukl, 1970; Gruenfeld, Rance, & Wéissenberg, 1969; Reilly, 1969). Thus, since Fiedler's model relates only indirectly to leadership behavior, this body of research will not be given further treatment here. Although none of the studies reviewed in this section )fiielded a dimension of leader behavior that could be clearly labeled participative decision making, some conceptual e‘flidence for its existence is available, and is presented in the following section. Participative Decision Makipg As a Leader Behavior Dimension Although the concept of participation in decision Inualcing has received considerable theoretical attention in 37€2<2ent years (Strauss, 1963; Lowin, 1968; Likert, 1961, 1967; Patchen, 1970), relatively little work has been done 2113. clearly conceptualizing participative decision making EIEB a dimension of leadership behavior. Newport (1962) noted the conceptual similarity of consideration and initiating Eitllructure to democratic and autocratic leadership, respec- t:i—Vely. Reviewing the literature on leadership styles, Sales (1966) found employee orientation (high consideration) 't‘3 be associated with democratic supervision. However, 17 others have argued for consideration and participative decision making (sometimes called Decision-Decentralization, Yukl, 1969) as separate, independent dimensions of leader behavior (Gomberg, 1966; McMurray, 1958; Schoenfeld, 1959; Stanton, 1962). The first direct theoretical attempt to deal with participative decision making as a third distinct leader behavior dimension was made by Yukl (1971) . His arguments follow. Lhe Theoretical Relation of brticipative DecisiontMaking to Confideration The Ohio State consideration scale contains par- ticipative decision making items. But these items are only considerate when subordinates desire participation in decision making, since there are individual differences in Preference for participation. Thus, when consideration is defined as "the degree to which the leader's behavior eJipresses a positive attitude towards subordinates" (Yukl, 1971, p. 417), participative decision making is conceptually d i stinct . g3 Theoretical Relation of QuicipativetDecision Making 20‘ Inititting‘S tructure If we define initiating structure as task oriented behavior: (1) concern for productivity, (2) making sure .n I. p O I l8 necessary task decisions are made, and (3) behavior insuring that decisions and higher level organizational directives (training and supervision) are carried out, then we have no mention of who will actually make the decisions. "The task orientation of the leader does not appear to be very closely related to the amount of influence he will allow subordinates in the making of task or maintenance decisions" (Yukl, 1971, p. 417). Thus participative decision making is made conceptually distinct from initiating structure. Lowin, Hrapchak, and Kavanaghl (1969) offer the idea that the presence of participation in decision making is what contributes to the intercorrelation among consideration and initiating structure scales. These researchers examined the factor analysis of LBDQ items by Fleishman, Harris, and Burtt (1955), and noted several participative decision making items that were found in both initiating structure and con- 8 ideration scales . Participative decision making items grouped with consideration it ems: l. 2 3. 4 He refuses to give in when people disagree with him. He insists that everything be done his way. He rejects suggestions for change. He changes the duties of people under him without first talking it over with them. He acts without consulting his foremen first. l9 6. He puts suggestions that are made by foremen under him into operation. 7. He gets the approval of his foremen on important matters before going ahead. Participative decision making items grouped with the initiat- ing structure items: 1. He.rules with an iron hand. 2. He waits for his foremen to push new ideas before he does. 3. He lets others do their work the way they think best. 4. He decides in detail what shall be done and how it shall be done. The notion of common participative decision making variance 11:1 both consideration and initiating structure scales is theoretically appealing in that allowing subordinates to IPétrticipate in decision making "implies a concern for their <>rxinions and for their desire to contribute (high considera- tion) and a relaxation of hierarchical control (low structure)" (Lowin, Hrapchak, 6: Kavanagh, 1969) . Although it is important to know that empirically al‘rldtheoretically, leader behaviors can be grouped into meaningful dimensions such as participative decision making, consideration, and initiating structure, it is also impor- tlétnt to knOW’hOW these groupings of leader behaviors are I:elated to one another. For example, if participative Sieecision making and initiating structure in leaders are 20 perceived by subordinates as being negatively related, then it would be a difficult task for a leader to be perceived as I high on pg h_participative decision making and initiating structure. The next section reviews studies of the empiri- cal relations mmong scales measuring consideration, par- ticipative decision making, and initiating structure in an attempt to better understand the nature of the dependencies among these leadership constructs. Empirical Relations Among Participative DeciSion Making, Consideration, and Initiating Structure The initiating structure and consideration dimen— Sions of leader behavior emerged from orthogonal factor analyses of the LBDQ (Hemphill and Coons, 1957; Halpin and Winer, 1957), and thus were by definition uncorrelated ciimensions. It should be noted, however, that scales (ieerived from factor analysis are usually correlated. That files, only the items that load most heavily on a factor are 17€etained for measurement of that factor. The typical scale score is simply the summation of responses to each of the Eiczale items. Given that a factor is a linearly weighted chambination of all_items being factor analyzed, and that <>1:thogonality refers to zero correlation between these w(eighted composites, it is easily seen why the actual scale scores tend to be correlated. The orthogonality is a III': 21 mathematically imposed constraint which may not be psychologically‘meaningful. Fleishman (1969) concluded that when averaged across a variety of situations, the median correlation between consideration and initiating structure scales is zero. The argument is that non-zero correlations between these scales in the literature represent a sampling dis- tribution whose true value is zero. However, several researchers have questioned the orthogonality of consideration and initiating structure Scales (Bales, 1958; Fiedler, 1964; Lowin, Hrapchak, and Kavanagh, 1969; Weissenberg and Kavanagh, 1972; Anderson, 1966; Bass, 1958; Fleishman and Harris, 1962; Fleishman and Peters, 1962; Halpin, 1954; Nealey and Blood, 1968; Oak- lander and Fleishman, 1964; Parker, 1963; and Rambo, 1958). Weissenberg and Kavanagh (1972) reviewed 72 studies Gennploying the LBDQ or the Leader Opinion Questionnaire (LOQ) 12:) assess the independence of the consideration and initiat- ing structure scales. Fifty-one percent (37) of these studies reported significant positive relationships between Qonsideration and initiating structure, 10% (7) significant negative relationships, and 39% (28) non-significant corre- lations. The mean correlationiwas .23 and the median was - 36, clearly not zero as predicted by Fleishman (1969) . Further analysis by type of instrument (LBDQ vs. LOQ) revealed . Ow J the I!‘ ~ 40-51! v M .ale A side: 22 that the correlation between consideration and initiating structure was more frequently positive for LBDQ measures, while the relationship was more frequently non-significant for LOQ measures. The LOQ is a self-report measure of how managers think they should behave. Also, for the LBDQ, 64% (ll) of studies involving first-line supervisors reported significant positive correlations between con- sideration and initiating structure, 11% (2) significant lmegative, and 23% (4) non-significant. As a summary state- nment, Weissenberg and Kavanagh (1972, p. 127) concluded: It appears that the independence of the dimensions is contingent mainly on the type of questionnaire used to measure the behavior. The high frequency of non- significant correlations for the LOQ indicates that managers think they should behave as if consideration and initiating structure are independent, however, descriptions by subordinates (LBDQ) indicate that their supervisors do not, in fact, behave in this manner, or at least, they are not perceived as behaving in this way. The literature relating participative decision making tho initiating structure and consideration is rather scarce. Measuring consideration and initiating structure by LBDQ éide participative decision making by the Decision Procedure Questionnaire (form C), Heller and Yukl (1969) found a cor- relation of .24 (p < .05) between participative decision IILaking and consideration in a sample of 67 second-line mana- Egems. Participative decision making was not significantly Irelated to initiating structure. ”a. , 1 :bv' ay- ‘1‘ 050 n-' uh- 23 Stogdill, Goode, and Day (1962, 1963, 1964) used a recent version of the LBDQ having consideration, initiating structure, and 10 other subscales to study a variety of samples. Participative decision making was measured by the "Tolerance for Member Freedom" subscale. The following correlations between participative decision making and con- sideration emerged: (a) .41 for subordinates of corporation (b) .42 for labor union subordinates; (c) .40 Beer (1966) Presidents; for community leaders; (d) .49 for ministers. found a correlation between participative decision making and consideration of .50 for a sample of office supervisors rated by female subordinates. In none of these five samples was participative decision making significantly related to initiating structure. In summary, empirical research supports the notion that initiating structure, participative decision making, and QQ‘tlsideration are oblique (correlated) dimensions of leader- ship behavior. Specifically, (l) participative decision making and consideration probably correlate moderate positive; ( 2 ) participative decision making and initiating structure jbt'l‘obably correlate low positive; (3) initiating structure Q11<1 consideration probably correlate zero to low positive. rhe exact relationships are difficult to predict because the measures of initiating structure and consideration in pre- J ions research are confounded with participative decision 'vv' . .5’ ,u .5. ‘1‘- I.‘ {A '1 H (1" .1 n 24 Clearly more research on the relation making variance. between participative decision making and both initiating structure and consideration is needed before firm conclu- sions can be drawn. While reducing leader behaviors to a few dimensions Or groups and examining the relations among these groups of behaviors are helpful steps in dealing with the com- do these Plexity of leadership, the question remains: groups of leader behaviors exert any effects on subordinate The following section answers this attitudes and behavior? question by reviewing subordinate satisfaction and produc- t iVity correlates of participative decision making and supervisor initiating structure and consideration. As the e‘7:i.dence will show, the answer to the question of whether 1 ea.der behaviors make a difference is complex. Attitudinal and Behavioral Correlates of Leader’Behavior g\atis faction Considerable research has been addressed to the 1Iilestion of the relation between leader behavior dimensions Unless speci- Q-hd subordinate satisfaction with the leader. :Eied, all research cited in this section used self-report hmeasures of satisfaction. ,1 0V 7.: 1 25 Several studies (Halpin, 1957; Halpin and Winer, 1957; Anderson, 1966; Yukl, 1969; Nealey and Blood, 1968; Graen, Dansereau, and Minami, 1972) reported a positive relation between consideration'and satisfaction. Using indirect indices of satisfaction (turnover and grievance rate), Fleishman and Harris (1962) and Skinner (1969) found Significant curvilinear relationships with their measure 0f consideration. In experimental lab studies (Day and Hamblin, 1964; Misumi and Shirakashi, 1966), low considera- tion (defined as punitive leadership) was found associated with low satisfaction. However, Lowin, Hrapchak, and KaVanagh (1969) reported no significant relation between CV90 levels of consideration and satisfaction although the he ans were in the predicted direction. The relationship between satisfaction and initiating 3 tincture is much less clear than with consideration. Ralpin (1957) and Yukl (1969) reported significant positive 13% lations between initiating structure and satisfaction. HQWever, Vroom and Mann (1960) found a significant negative I: Qlation between pressure for production and satisfaction 3Q): delivery truck drivers; this relationship did not hold up for truck loaders. Similarly, Nealey and Blood (1968) Ecund a significant positive relationship between initiating a‘ttlz-ucture and satisfaction for first-level supervisors, but 110 significant relation for second-level supervisors. u" til .a 26 Numerous studies found no relationship at all between initiating structure and satisfaction (Baumgartel, 1956; Halpin and Winer, 1957; Argyle, Gardner, and Cioffi, l958--labor turnover and voluntary absenteeism; Misumi and Shirakashi, 1966; Lowin, Hrapchak, and Kavanagh, 1969 ; Anderson, 1966; Likert, 1961, pp. 16-18). A few non-linear relationships between initiating Structure and satisfaction have been found. Specifically, Fleishman and Harris (1962) and Skinner (1969) found a negative, curvilinear relationship between initiating Structure and two indices of satisfaction: turnover and grievances. Likert (1955) reported an inverted U relation between pressure for production and subordinate satisfaction. F inally, Fleishman and Harris (1962) found an interaction of initiating structure and consideration, such that high Q Qnsiderate leaders could increase initiating structure with l ittle accompanying increase in turnover and grievance. The relationship between participative decision making and subordinate satisfaction has generally been bositive (Baumgartel, 1956; Argyle, Gardner and Cioffi, 1958uvoluntary absenteeism and labor turnover; Vroom, 1959; Eachman, Smith, and Slesinger, 1966; Yukl, 196.9; and Tosi, 1970). Experimental studies also confirm this positive 1‘elationship (Coch and French, l948--turnover; Shaw, 1955; L'Iorse and Reimer, l956--turnover and self-report; Solem, 1958; Maier and Hoffman, 1962). Q l v «’Q "ca-o I a no. - 0 u ' "-on 53:9] n 0—. ‘D . .1 H (L? H II'OMI-z 0 V- - Iain ., \\ 6' do "an. Jud. I . I 4a.. u.5. I ~.,. .0“; :t’ : i151 Int 1 '_1 n *1. \.14) .‘In h-J r1- ~41 27 There is some evidence that the positive relationship between participative decision making and satisfaction is moderated by other variables. For example, Vroom (1959) found a positive relation between participative decision making and satisfaction for sub- ordinates with a high need for independence but not for those subordinates with a low need. However, T031 (1970) failed to replicate the moderated relationship found by Vroom, and found only consistent positive relationships. 3888 (1965, pp. 165-170) and French, Israel and As (1960) reported that subordinates' perceived participation in decision making did not result in more favorable attitudes t oVavard the leader unless they also perceived participation in decision making as a legitimate part of their role. Pflciuctivi ty The relationship between consideration and produc- ': iVity is far from clear. A positive relationship has been QB served by some researchers (Katz, Maccoby, Gurin, and Floor, 1951; Argyle, Gardner, and Cioffi, 1958; Besco and LG-Wshe, 1959; Schacter, Willerman, Festinger, and Hyman, 3‘ 961; Kay, Meyer, and French, 1965). However, instances of 110 significant linear relations have also been reported (Bass, 1957; Nealey and Blood, 1968; Scott, 1968). In an fixperimental laboratory study, Lowin, Hrapchak, and Kavanagh l IA-OI ’,fl .imte . ‘l a}. .‘y “DUE. I , t 5:.” 1‘. iii .1 m . 1,3131 I I." II 28 (1969) found a positive relation for objectively manipulated consideration but not for subordinate ratings of consideration. A negative relation occurred between consideration and performance for aircraft com- manders (Halpin and Winer, 1957), and for foremen of pro- duc tion departments, but not for non-production departments (Fleishman, Harris, and Burtt, 1955). It should be noted that in these last two studies, performance was measured by superiors' ratings of effectiveness, and not by objec- tiVe output of the leader's group. The relationship of initiating structure to pro- ductivity is a little clearer than that between considera- t ion and productivity. Generally, initiating structure is p Ositively related to productivity (Fleishman, Harris, and Burtt, 1955; Likert, 1955; Halpin and Winer, 1957; Maier th Maier, 1957; Besco and'Lawshe, 1959; Anderson, 1966; Nealey and Blood, 1968) . Some have found no relationship (Argyle, Gardner, and Cioffi, 1958; Bass, 1957; Halpin, 3‘ 957; Rambo, 1958; and Lowin, Hrapchak, and Kavanagh, 1969), But none have reported a negative relationship. Several instances of interactive effects of initi- ating structure and consideration on productivity have been sound. When Israeli foremen were described by superiors Q11 a Hebrew translation of the Supervisory Behavior Descrip- tion, foremen high on both consideration and initiating 29 Structure received the highest proficiency ratings from their superiors (Fleishman and Simmons, 1970). Similarly, superiors gave highest effectiveness ratings to aircraft commanders who were described by subordinates as high on both consideration and initiating structure. Hemphill (1957) found that high consideration-high initiating struc- ture department chairmen received highest faculty ratings of how well the department was administered. Personal Production norms (task motivation) was highest for workers whose leader encouraged proficiency and "went to bat" for Workers (Patchen, 1962). Finally, in an experimental lab 8 tudy, Misumi and Shirakashi (1966) found that their most F’ttoductive groups had leaders high on both consideration a~Imzi initiating structure. Dawson, Messé, and Phillips (1972) experimentally Inilanipulated teacher behavior in the college classroom situ- ation. Main effects for consideration and initiating Q tructure were found. Students under high consideration treatment handed in more annotated bibliographies than those in the low consideration groups; the same direction of foect appeared for initiating structure. Subj ects under high consideration also performed better on class exams, and participated in more psychological experiments to earn I‘esearch credits. The interaction effect to emerge involved research credits such that subjects in the low consideration- 30 high initiating structure condition earned a disproportionately low number of research credits. A considerable amount of evidence suggests that participative decision making tends to be positively related to group performance (Bachman, Smith, and Sle- singer, 1966; Coch and French, 1948; Fleishman, 1965; French, 1950; French, Kay, and Meyer, 1966; Lawrence and Smith, 1955; Likert, 1961; Mann and Dent, 1954; McCurdy It and Eber, 1953; Meltzer, 1956; and Vroom, 1959). 8I'lould be noted, however, that Argyle, Gardner, and Cioffi ( 1958) found a positive relation between participative decision making and performance only for those departments There have also been studies showing without piece rates. no relationship (Tosi, 1970; French, Israel and As, 1960; a‘~'-'lcl McCurdy and Lambert, 1952) or a negative relation (Shaw, 1 955; Morse and Reimer, 1956). As a general summary, (l)the relationship of c esearch findings suggest that: Q Qnsideration to productivity is unclear; (2) initiating Q tructure is positively related to productivity; (3) initi- at‘ting structure and consideration interact to yield greatest b7"?oductivity under the high initiating structure-high con- & ideration combination; (4) participative decision making is positively related to productivity. In talking about the attitudinal and behavioral Qffects of leader behavior, we are assuming that the leader . t .A-r ’ I 1v..- .u. l to“ v m L. U . n.. “c: I I.) T) a- (It ‘1 ‘5 '~c n- ( I) 14 31 behaviors are perceived by subordinates. This assumption raises the methodological issue of whether individual or group perceptions of leader behavior are most appropriate. The next section addresses this issue and finds the methodological and empirical evidence in favor of using iridiividual perceptions of leader behavior in the present study. Individual Versus Group Perceptions of Leader Behavior Most, if not all, research on leadership behavior has used group perceptions to describe the leader. Typi- Qally a leader's subordinates complete the LBDQ, and then the leader's level of initiating structure and consideration 1 8 defined as the average response of his subordinates on these scales. Although some will argue that average initiating structure and consideration scores are the most Q table indicators of the leader's behavior, these scores may accurately reflect the perceptions of 132 individual QI-‘lbordinate. Halpin and Winer (1957, p. 49) examined the batio of variance in initiating structure and consideration Q(tales between air crews to within air crews, finding reli- ably greater variance between leaders than within (29 air Ql‘ew commanders were described by 201 crew members). How- e\Ter, Dawson (1972, unpublished dissertation) studied the - __ 32 leadership of 88 auto industry foremen using the LBDQ, and found that variance in his measures of initiating structure and consideration within foremen work groups was actually larger than the variance between foremen work groups. The essential point is that much useful information is lost when individual perceptions of leadership behavior are collapsed into aggregate scores; differences in perception are treated as irritating error variance. Individual per- ceptions rule the individual's world; he acts and feels on the basis of his own perceptions, not those of a mythical average group member (with whom he has not even a passing acquaintance). Graen, Dansereau, and Minami (1972) point out another issue involved in individual versus group percep- tion of leader behavior. First, should leadership behavior be viewed as an average style or a set of particular leader- member relationships? If leadership behavior is viewed as an average style, then averaging subordinate perceptions is the most valid and reliable way to measure it. If, how- ever, leader behavior is viewed as a set of unique dyadic relationships, then averaging individual subordinate per- ceptions would not be valid, and would in fact result in loss of valuable information. In his recent paper on individualizing organizations, Lawler (1972) reaffirms the vital role of individual 33 differences in organizational psychology as a specific application of Cronbach's (1957) treatment of the two disciplines of psychology in which the differences between experimental and correlational psychology are discussed. Lawler makes the point that organizational theorists search- ing for universality "often ended up treating individual differences in how people react to organization practices as error variance." A methodological issue, related to the issue of using individual or group perceptions of leader behavior, is whether one should use multidimensional scaling or factor analysis in studying the perception of leader behaviors. The following section highlights the simi- larities and differences in these two approaches, and points out how much useful information about individual differences in perceptions of leader behaviors can be obtained through the INDSCAL method of multidimensional scaling. Multidimensional Scaling_Versus Factor Analysis In Studying Perception of Leader Behaviors The purpose of this section is to show how recent developments in multidimensional scaling (MDS) can con- tribute to the study of leadership in ways not possible with factor analysis. 34 As with factor analysis, the goal of MDS in general is to discover the underlying pattern or structure in a ‘matrix of empirical data. Unlike factor analysis, MDS attempts to represent this structure in a more visible manner, i.e., as a geometric model. "The objects under study (whether these be stimuli, persons, or nations) are represented by points in the spatial model in such a way that the significant features of the data about these objects are revealed in the geometric relations among the points (Shepard, 1972, p. 1)." Multidimensional scaling starts with a set of inter- related entities, in this study leadership behaviors. For each pair of entities, a measure of proximity (association, similarity) is the basic input datum for MDS. Commonly used measures of proximity are ratings of similarity; in the case of leadership behaviors, a rating of likelihood of co- occurrence is more appropriate. Basically, MDS provides a geometric configuration of the entities in which the inter- point distances correspond to the empirical input values of psychological relatedness. The success, or goodness-of-fit, of the geometric representation is defined as the degree of monotonic relationship between the measures of similarity and the interpoint distances in geometric space, described later as lack of "stress." Unlike unidimensional scaling, MDS permits points to position in two- or higher-space if necessary to capture 35 the complexity of the data. While factor analysis typically yields high-dimensional factor space, MDS usually yields solutions in two- or three-dimension space. One kind of MDS procedure is INDSCAL (individual differences scaling). The INDSCAL model has in common with other MDS procedures the assumption that different indi- viduals perceive the stimuli in terms of a common set of assumptions. However, these dimensions are differentially important or salient in the perception of different indi- viduals. In the extreme case of zero importance or salience, a given dimension does not affect the person‘s perception at all, which is the same as saying that he does not perceive that dimension. The model underlying INDSCAL is a modified Euclidean distance for each subject. For the ith subject, the distance equation is: (i)= r _ 2 1/2 dJ'k [E = 1 Wit (th xkt) ] where, d‘k) the distance between stimuli j and k 3 for subject i wit = weighting given the tth dimension (t = l to r dimensions) by subject i (w.t - wkt)2 = the squared difference be ween J stimuli j and k on the tt dimension t L ‘1 9 fish; als 36 According to Carroll (1972), the major purpose of the INDSCAL program is to determine both the stimulus coordinates and the subject weights that maximally explain the total variance in all subjects' data. Using an iterative least squares procedure, the program determines the orien- tation of the axes that maximizes the goodness-of-fit criterion. The resulting orientation usually results in dimensions which are directly interpretable without further rotation. Graphically, INDSCAL produces the information illustrated in Figure 1. Note that for subject 2, dimen— sion 1 is most important; he can discriminate among stimuli to a greater extent with dimension 1 than 2. The exact opposite dimension saliency occurs for subject 4. Carroll (1972) notes several useful properties in the subject map shown in Figure 1. Besides providing information about the relative saliency of each dimension for a given subject, the subject map gives an idea of the variance in a subject's judgments that is explained by the dimensions. Specifically, the dimensions are normalized so that the sum.of squares of projections for all subjects on each dimension is unity. Thus, by squaring the two projections for a given subject, we have an index of the communality of that subject with all other subjects in the analysis, i.e., how much of the particular subject's variance is explained by the group-based dimensions. 37 H SHQ .mu .mmocowom Hmuow>m£on map Ga mcofiumowaamm paw huoonu "mawamom $838333st .5va 33.82 .m .m 28 .398m .x .< .3323 .z .m SH .wcHHmom HmGOHmaoawpfiuana paw mooconommwp Hoppfi>wpaH .Haounmo .n .h "monsom .AuanH HoBoHv s paw Aumma HwBoav N mm £03m muoomndm How moommm Hmsummouom Hman>HvGH mospoum ou Aumma Hommsv momma mDHDBHum :msonw: mo mcoflmaoaflp ou poflammo ohm Aunwflu Hmmmnv momma uommnsm aoum muzwflmz .Hopoz A rIs,<: Hypothesis 4: Subordinate's perception of supervisor behaviors encouraging subordinate participation is positively related to:. (a) subordinate satisfaction with the supervisor. (b) subordinate motivation to work for the super- v1sor. (c) subordinate's attitude toward the task compe- tence of the supervisor. (d) subordinate's attitude toward the interpersonal style of the supervisor. Hypothesis 5: Subordinate's perception of initiating structure behaviors in his supervisor is: (a) negatively related to subordinate's satisfac- tion with supervisor. (b) positively related to subordinate's self- reported motivation to work under his super— visor. (c) negatively related to subordinate's attitude toward the interpersonal style of the super- visor. (d) negatively related to subordinate's attitude toward the task competence of the supervisor. 44 Hypothesis 6: Subordinate's perception of consideration Bahaviors in his supervisor is positively related to: (a) subordinate's satisfaction with his supervisor. (b) subordinate's motivation to work for his supervisor. - (c) subordinate's attitude toward the supervisor's interpersonal style. (d) subordinate's attitude toward the supervisor's task competence. A set of two studies was designed to test these hypotheses. In Study 1, the subjects' task was multidimen- sional scaling of leader behaviors to determine whether subordinates perceive three distinct dimensions in leader behaviors. The results of this study provided a test of hypothesis 1. In Study 2, subjects rated experimentally manipulated supervisor descriptions varying in the extent to Which they showed participative decision making, con- sideration, and initiating structure. The design of Study 2 permitted the testing of hypotheses 2-6. In addition to hypotheses 4, 5, and 6 above, the experimental design in Study 2 permitted an exploratory analysis of possible interactions among the three leadership behavior dimensions and subordinate personality character- istics (Authoritarianism and Need for Independence). CHAPTER III EXPERIMENT I: MULTIDIMENSIONAL SCALING OF LEADERSHIP BEHAVIORS This study was concerned with the testing of Hypothesis 1: Participative decision making, defined as the degree to which a leader allows or encourages his sub- ordinates to participate in making decisions related to their jobs, is a valid third dimension of leadership behavior,as perceived by subordinates. The general plan of the study was to use INDSCAL in the multidimensional scaling of 15 leader behavior statements, selected to represent leader participative decision making, considera- tion, and initiating structure dimensions. Method Subjects Fifty-two male undergraduate students in introductory psychology volunteered to participate in the experiment for extra credit toward their final class grades. 45 46 Stimulus Materials Leader behavior stimulus statements for the multi- dimensional scaling were selected to represent conceptual definitions of the three major leadership behavior dimen- sions of interest: 1. 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. ' 2. Initiating structure: Behaviors by which the supervisor organizes and coordinates the activities of subordinates. Behaviors by which the super- visor defines his relation to his subordinates, and defines the role he expects each subordinate to assume. 3. 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. With these dimension definitions in mind, the experimenter searched through leader behavior items reported by Fleish— man, Harris, and Burtt (1955), Dawson (1972), Lowin, Hrap- chak, and Craig (1968), and Stogdill (1963), selecting 13 items that seemed to fit the definition for participation, 12 for consideration, and 11 for initiating structure. Since the subject's task in multidimensional scaling is to judge the likelihood of co-occurrence of 105 pairs of 47 leader behavior statements using 15 stimuli, the initial set of 36 items had to be reduced. Toward this end, five graduate student judges were asked to sort independently the 36 leader behavior items into three groups: participa- tion, consideration, and initiating structure, using the dimension definitions given above. If a statement did not fit any of the three definitions, it was put in a fourth group--other. Five items related to each dimension of leadership behavior were selected under the criterion that 4 out of 5 of the judges agree in their sorting of each item in a dimension. Table 2 presents the final set of 15 leader behavior items selected for use in multidimensional scaling. These items were punched on computer cards so that computer generated printouts could be obtained. The printouts (see Appendix A for specimen) were designed to have the following characteristics: 1. Each contained E_S%;ll = 15 14 = 105 pairs of leader behavior stimuli. 2. A different random order of presentation of pairs was generated for each subject. 3. A 9-point likelihood of co-occurrence scale appeared to the right of each pair. 4. Half’the printouts had the same order of statements within each pair; half had the reverse order of statements within each pair. 48 Table 2.--Leader Behavior Items Selected for Multi- dimensional Scaling. Consideration 1. He is friendly and easily approached. 2. He expresses appreciation when I do a good job. 3. If I am a new member, he helps me make adjust- ments to the group. 4. He makes me feel at ease when talking with him. 5. He looks out for my personal welfare. Participative Decision Making 6. 7. 8. 9. 10. He allows me to modify the procedures required in my job. He has me share in decision making. He assigns a task, then lets me handle it. He allows me influence equal to his own in decisions which affect my job. He lets me do my work the way I think best. Initiating_Structure ll. 12. l3. 14. 15. He lets me know what is expected of me. He schedules the work I have to do. He makes sure that I and other group members under— stand his part in the group. He maintains definite standards of performance for me. He sees to it that I have the material I need to work‘with. 49 Procedure Because of the tediousness of the subjects' rating tasks, especially the multidimensional scaling, experiment I was conducted in two one-hour sessions. In the first ses- sion, subjects completed the multidimensional scaling of the leader behavior stimuli and two personality measures: Need for Independence Scale, and the F-Scale Measure of Authoritarianism (Vroom, 1960). Alpha estimates of inter- nal consistency reliability for the Independence and Authoritarianism scales, based on the subjects in experi- ment 1, were .43 and .80, respectively. These personality measures are presented in Appendix B. The session began when a sample pair of leader behaviors was placed on the chalkboard with its correspond- ing 9-point likelihood of co-occurrence scale. The experimenter then read these instructions: I want you to think of a hypothetical male supervisor in industry. Now look at your printout where you will see pairs of behaviors. If a hypothetical male super- visor does the first behavior, how likely is it that he will do the second? If you think it is very likely, you would circle the 9. If you think it is very unlikely, you would circle the 1. If you think they are unrelated, you would circle the 5. That is, if he 50 were just as likely as not to do the second behavior, the pair would be rated 5. Subjects were told that none of the pairs were identical, and so they were to rate each pair indepen- dently, starting with the first pair and going straight through to the end without looking back to see how they rated similar pairs. At session 2, using 5-point Likert scales, subjects rated each of the fifteen leader behaviors on (1) considera- tion, (2) participative decision making, (3) initiating structure, and (4) the degree to which they would be satis- fied with that behavior in their supervisor. This rating task is presented in Appendix C. It should be noted that each rating dimension was defined, and that a different, random order of leader behaviors was used for each rating dimension, although all subjects received the same random order within a dimension. For each subject, the order of presentation of rating dimensions was randomized. The major purpose of this Likert rating task was to gather data to help interpret the dimensions resulting from the multidimensional scaling. Toward this end, the subjects also rated each of the fifteen leader behaviors on seven semantic differential scales: 51 weak-strong good-bad active-passive don't prefer-prefer important-unimportant successful-unsuccessful cruel-kind \JO‘UI-l-‘UONH These semantic differential scales were selected because they were representative, though not exhaustive of, the possible dimensions which subjects might use in perceiving leadership behaviors. The order of semantic differential scales was randomly determined, and was the same for all leader behaviors. Four of the seven scales above are reversed to minimize response set. A different random order of pre- sentation of the 15 leader behaviors was used for each subject. The slightly modified semantic differential rating instructions (Osgood, Suci, & Tannenbaum, 1957) and leader behavior items and actual rating scales appear in Appendix D. Results and Discussion Before going further, an important assumption under- lying all multidimensional scaling techniques, including INDSCAL, is symmetry. This assumption requires that the distance from stimulus A to stimulus B be the same as the distance from B to A. ‘When symmetry is violated, the exact distances between stimuli in Euclidean space is difficult 52 to determine since Euclidean geometry is also based on the assumption of symmetry. . In a pilot study, it was determined that subjects had great difficulty understanding the more methodologically sound intersection approach to giving instructions, in which neither the first nor the second behavior in each pair is the specified anchor. Since the problem of asymmetry may arise using other than the intersection method, i.e., P(B1/Bz) # P(B2/B1) where B1 = behavior 1 and B2 = behavior 2, two male pilot subjects each performed the multidimensional scaling task twice, once rating P(Bl/B2) and once rating P(leBl), to check on the problem of asymmetry of ratings. For each subject, the correlation between his two sets of ratings was .67 and .62, indicating no substantial asym- metry problem with the non-intersection method of giving instructions. As an additional, intuitive check on the problem of asymmetry, the 105 pairs of leader behavior statements were examined in both the A/B and B/A orders. No glaring asym- metries were apparent. Table 3 presents the correlations between the com- puted scores in Euclidean space and the original proximity data for subjects. These correlations show the degree of fit of the three- and four-dimensional INDSCAL solutions for each subject, i.e., the degree to which the solutions 53 Table 3.--Correlations between Computed Scores and Original Data for Subjects. Three- Four- . Three- Four- Sléllrfijbegxf: Dimension Dimension Sfiffifg Dimension Dimension Solution. Solution Solution. Solution 1 .60 .62 27 .21 .23 2 .67 .67 28 .52 .54 3 .60 .63 29 .68 .74 4 .77 .79 30 .71 .76 5 .65 .66 31 .60 .63 6 .72 .75 32 .54 .57 7 .70 .72 33 .85 .84 8 51 56 34 67 71 9 71 70 35 58 64 10 53 .56 36 51 56 ll 57 .59 37 56 59 12 46 .45 38 74 75 13 37 .34 39 80 82 14 71 72 40 75 74 15 76 83 41 47 50 16 65 69 42 62 64 17 74 75 43 56 59 18 37 39 44 63 64 19 81 83 45 63 67 20 57 62 46 56 59 21 44 .46 47 81 83 22 68 71 48 74 76 23 81 82 49 39 41 24 52 .52 50 6O 63 25 62 .62 51 66 71 26 69 .72 52 73 77 Average subject correlation coefficient .62 .65 Mean squared correlation coefficient .40 .43 54 adequately describe each subject's perceptual world. The differences between correlations for both three- and four- dimensional solutions indicates that adding a fourth dimension does very little to improve the fit of the model. The average correlation coefficients are .62 for the three- dimensional solution and .65 for the four-dimensional solution. However, these coefficients are comparable to the degree of fit found in other applications of multidimen- sional scaling in industrial-organizationa1 psychology (Davison and Jones, 1973; Hamstra, 1973). These data-model fits are relatively poor compared with INDSCAL solutions using auditory and visual stimuli (Wessel and Null, 1973). Because the fit of the three-space INDSCAL solution to the data was already relatively poor compared with INDSCAL solutions using auditory and visual stimuli (Wessel and Null, 1973), no two-space solution was computed. Such a reduction in dimensions would have only reduced the accuracy of the data-model fit. Note in Table 3 that subjects 13, 18, 27, and 49 all have very low correlations (below .40) in the three- dimensional solution which do not tend to improve much when the fourth dimension is added. Two reasons may explain this relative lack of fit: (1) these subjects may have responded in a relatively random manner; (2) they may have been using dimensions other than those used by the majority 55 of the subjects. Fortunately, few of the 52 subjects showed such poor lack of fit between raw proximity data and dis- tances in Euclidean space. The locations of the 15 leader behavior stimuli in the three-space solution are presented in Figures 2, 3, and 4. From inspection of these figures, we can answer two major questions. Are the three a priori clusters of leader behaviors distinct enough from one another to justify their experimental manipulation in Experiment 11? What is the mean- ing of the perceptual dimensions that emerged from INDSCAL? In response to the first question regarding the rela- tive distinctiveness of the a priori clusters, Figure 2 shows the leader behavior stimuli plotted in the first two INDSCAL dimensions of the three-space solution. It is clear that there is both intracluster and intercluster variation, highlighted by the lines drawn around the three a priori groups of leader behaviors: consideration (numbers 1-5), participative decision making (numbers 6-10), and initiating structure (numbers 11-15). Note that with the possible exception of item 15, the three clusters are quite distinct from one another, even though there is some intracluster variation for the most part along dimension 2. In Figure 3, the leader behaviors are displayed along INDSCAL dimensions 1 and 3. Here the distinctiveness of the clusters is less clear, especially in regard to the 56 Participative F'6 Decision Making -.5 ‘;4 —.3 .—.2 I I I 1 1 l -.6 —.5 -.4 -.3 -.2 -.1 Initiating Structure Dimension 1 (horizontal) Consideration Dimension 2 (vertical) Figure 2.--Leader Behavior Stimuli Coordinates in Dimensions 1 and 2 from.Three-Space INDSCAL Solution. 57 —.6 Initiating —-5 Structure Consideration ’«4 ‘13 ”.3 —.2 _.l Dimension 1 (horizontal) Participative Decision Makingtr'- Dimension 3 (vertical) Figure 3.-- Leader Behavior Stimuli Coordinates in Dimensions 1 and 3 from Three-Space INDSCAL Solution. 58 Participative Decision Making Consideration Dimension 2 (horizontal) Initiating 3 Structure ' .4 .5 --.6 Dimension 3 (vertical) Figure 4.--Leader Behavior Stimuli Coordinates in Dimensions 2 and 3 from Three-Space INDSCAL Solution. 59 participative decision making and consideration clusters. In addition, the large amount of intracluster variation is attributable to dimension 3, and to a lesser extent to dimension 1. Figure 4 displays the leader behaviors along INDSCAL dimensions 2 and 3. Again, the distinctiveness of the a priori clusters is less apparent, and a large portion of the intracluster variation appears to be due to dimension 3. In summary, the distinctiveness of the a priori clusters in three-space is obscured by a large amount of intracluster variation along dimension 3, and to a lesser extent along dimension 2. The argument for the distinctive- ness of the a priori clusters is most strong in reference to INDSCAL dimensions 1 and 2 shown in Figure 2. 'Since the a priori clusters are most distinct in two rather than three dimensions, indicating some degree of cor- relation among the clusters, another analysis was performed to determine the feasibility of independently manipulating the consideration, participative decision making, and ini- tiating structure leader behavior clusters. In this analysis, only the consideration, participative decision making, and initiating structure ratings given to the a priori groups of leader behavior items were examined. Specifically, a 3 x 3 repeated measures design was used with type of item and type of rating as the respective factors. The dependent 60 variable was amount of rating given to each of the three groups of items. Table 4 presents the results of the item by rating analysis of variance, the critically important part of which is the expected, significant (p < .0005), and rela- tively powerful (n2 = .19) interaction. Intuitively, one would expect the amount of rating given to an a priori group of items to depend on the type of rating dimension being used. Table 5 presents the interaction cell means, and highlights some of the dependencies existing among the consideration, participative decision making, and initiating structure groups of leader behavior items. Our primary con- cerns were two: (1) whether items designated a priori as consideration items would be rated as showing more con- sideration than the a priori participative decision making items, i.e., whether consideration and participative decision making a priori groups were independently manipulable; and (2) whether participative decision making items would be rated as showing more participation than consideration items. Conceptually, concern (1) above was most important in that consideration and participative decision making a priori groups of leader behavior items might both be perceived as showing high consideration. However, a simple-effects analysis (Kirk, 1968, p. 267) indicated the consideration 61 Table 4.--Two-Way Analysis of Variance of Type of Item and Type of Rating on Amount of Rating. Source df MS F ‘p n2 Between 51 Within Item 2 116.64 5.64 .005 .024 Item x Subjects 102 20.68 Rating 2 94.44 9.45 <.0005 .019 Rating x Subjects 102 9.99 Rating x Item 4 471.82 40.13 <.0005 .194 Rating x Item x Subjects 204 11.76 SS = 9748.95 TOTAL 62 Table 5.--Mean Amount of Rating as a Joint Function of Type of Item and Type of Rating. Type of Ratinga Type of Item. Participative . . . Consideration Decision ngfizifiizg Making Consideration 21.36 17.64 16.73 a a Partipative ] ] Decision Making 19.94 21.36 16.35 Initiating Structure 16.65 15.35 20.52 8Significantly different, p < .05. bMean rating is based on summation across five Likert scales. Thus, instead of having values ranging between 1 and 5, the transformed means above can range 5 to 25. 63 ratings given to consideration and participation a priori groups of items to be significantly different (t = 2.12, df = 1/459, p < .05), as were the mean ratings of partici- pation given to consideration and participation items. How do the previous results and discussion bear on Hypothesis 1?: Participation, defined as the degree to which a leader allows or encourages his subordinates to participate in making decisions related to their jobs, is a valid third dimension of leadership behavior, as perceived A by subordinates. Based on the fairly good clusterings observed in the multidimensional scaling and the analysis of variance evidence that the three dimensions were approximately independently manipulable, hypothesis 1 is reasonably well supported. It is important to note that leader behavior dimensions do not have to be orthogonal to one another in order to be considered valid. Rather, the primary purpose of the multidimensional scaling analysis was to show that the leader behaviors would be perceived as belonging to three separate and distinct clusters, defined a priori as consid- eration, participative decision making, and initiating structure. Therefore, it was decided to attempt the experi- mental manipulation of each of the consideration, partici- pative decision making, and initiating structure a priori clusters in experiment II. 64 In response to our second question regarding the meaning of the three orthogonal dimensions produced by INDSCAL, analytical and intuitive approaches to interpreting the dimensions were used. The analytic interpretation was based on a tech- nique and computer program called PROFIT, developed by Chang and Carroll. Tables 6 and 7 represent criteria generated by PROFIT for trying to interpret or give meaning to the three-dimensional INDSCAL solution in which the three a priori clusters are located. As mentioned earlier in the introduction, INDSCAL is supposed to provide directly interpretable dimensions without rotation of the axes. Table 6, the first of the two criteria for finding suitable labels for the INDSCAL dimensions, shows the maximum corre- lation between the property (rating scale) and the projections on the fitted vector. As a brief explanation, consider the mean rating on the weak-strong property given to all 15 stimuli. The stimuli are rank ordered on the weak-strong dimension. This vector of rank orders is positioned in three- space to maximize the correlation between its rank values and the projections of the 15 stimuli in the three-space on it. A low correlation indicates a rather poor job of describing the stimulus points in three-space. Also, one may consider this a necessary but not sufficient criterion for deriving labels for the INDSCAL dimensions since it 65 Table 6.--The Maximum Correlation Between the Property and the Projections on the Fitted Vector. Rho Property 1 .37 weak-strong 2 .51 bad-good 3 .66 passive-active 4 .64 don't prefer-prefer 5 .41 unimportant-important 6 .41 unsuccessful-successful 7 .94 cruel-kind 8 .92 participation 9 .89 initiating structure 10 .78 satisfaction 11 .89 consideration Table 7.--Direction Cosines of Fitted Vectors in Normalized Space. Dimensions Vector l 2 3 l. weak-strong .70 -.67 .25 2. bad-good -.89 -.46 .07 3. passive-active .23 -.68 70 4. don't prefer-prefer -.98 -.10 16 5. unimportant-important -.99 -.05 01 6. unsuccessful-successful -.96 -.24 - 13 7. cruel-kind -.85 -.46 26 8. participation -.84 .54 - 01 9. initiating structure 98 .19 07 10. satisfaction - 95 .22 23 11. consideration 91 -.30 28 66 evaluates only the fit of the property vector to the stimuli in the three-space, independent of the relation of the property vector to the INDSCAL dimensions. It appears in Table 6 that properties 7, 8, 9, 10, and 11 are good possible candidates for labeling our three dimensions, according to the first PROFIT criterion. Table 7 shows the relation of the fitted property vectors to the INDSCAL dimensions. The ideal is to find values of the cosine equal to i 1.00 which indicates an angle of 0° between the property-vector and the INDSCAL dimension, a perfect geometrical alignment between the property vector and the dimension. Of the properties surviving the first criterion in Table 6, initiating structure seems to best explain the first INDSCAL dimension; however, the other four properties are also reasonably well aligned with dimension 1. The second and third dimensions are uninterpretable using the tremaining properties; none of the property vectors seem to align with either dimension 2 or 3. Although the analytic approach to interpretation Vmas.rather unsuccessful, an intuitive interpretation of the :IITDSCAL dimensions is possible, based on visual inspection (’15 the intra- and intercluster variation that appears in Figures 2, 3, and 4. 67 Looking back at Figure 2 and the content of the leader behavior items in Table 2, there is some justifica- tion for labeling INDSCAL dimension 1 initiating structure. The a priori initiating structure cluster differs markedly from the a priori consideration and participative decision making clusters on this dimension. These latter two clusters, as one would expect, are positioned at approxi- mately the same low end of the dimension. Examination of the intracluster variation in Figure 2 also supports the initiating structure interpretation of dimension 1. Within the initiating structure cluster, item 14 (He maintains definite standards of performance for me) clearly implies more initiating structure, as originally defined, than does item 15 (He sees to it that I have the material I need to work with). In the consideration a priori cluster, item 3 (If I am a new member, he helps me make adjustments to the group) implies more initiating structure than item 1 (He is friendly and easily approached). Inspec- tion of the participative decision making a priori cluster offers similar support. Item 8 (He assigns a task, then lets me handle it) implies more initiating structure than item 10 (He lets me do my work the way I think best). From inspection of the intra- and intercluster varia- tion along INDSCAL dimension 2 in Figure 2, a reasonable interpretation is that dimension 2 deals with the amount of 68 task concern by subordinates. To elaborate, note the positioning of the a priori clusters along this dimension. At the upper end of the dimension is the participative decision making cluster, all the items of which directly imply involvement by the subordinate in work related decision making. Understandably enough, there is little variability within this cluster along dimension 2. In the middle of dimension 2 is the a priori initiating structure cluster. In these items, there is a task concern component, but the emphasis is task concern by the leader, not the subordinate. Within this cluster, it seems reasonable that item 13 (He makes sure that I and other group members understand his part in the group) has a higher scale value on dimension 2 than does item 15 (He sees to it that I have the material I need to work with), since the subordinate sees this as a supervisory task con- cern. As expected, the a priori consideration cluster has the lowest values in INDSCAL dimension 2, since the content of this cluster has very little to do with thetask concern of either the supervisor or subordinate. As with the other clusters, the intracluster variability shows a pattern con- sistent with the task concern by subordinate interpretation of dimension 2. Item 2 (He expresses appreciation when I do a good job) expresses a contingency on subordinate task 69 concern, and accordingly, is given a higher scale value than item 3 (If I am.a new member, he helps me make adjustments to the group) which has absolutely no task concern content. Figure 3 illustrates the intra- and intercluster variability along INDSCAL dimension 3. It appears that the intracluster variability along dimension 3 is much greater than the intercluster variability. The patterning of the leader behaviors suggests that dimension 3 measures the degree of subordinate contact or involvement with the supervisor. Since each of the a priori clusters calls for interaction between subordinate and supervisor, it makes sense that the clusters differ very little from one another on dimension 3. The large within cluster variation along dimension 3 provides strong support for the subordinate-supervisor inter- action interpretation. Consider the initiating structure a priori cluster. Item 13 (He makes sure that I and other group members understand his part in the group) definitely implies more subordinate-supervisor interaction than item 15 (He sees to it that I have the material I need to work with), and appropriately, item 13 is scaled much higher than item 15 on dimension 3. Similarly, within the a priori consideration cluster, the interaction pattern holds. Item 1 (He is friendly and easily approached), the behavior with the highest scale 70 value in the cluster, seems to tap an interaction component moreso than item 2 (He expresses appreciation when I do a good job) which clearly specifies a contingency of when interaction will occur. In the participative decision making a priori clus- ter, additional support for the interaction interpretation is found. For example, item 7 (He has me share in decision making), the behavior with the highest scale value in the cluster, directly connotes a dynamic interaction pattern between subordinate and supervisor in the process of thrashing out a decision acceptable to both parties. Con- trast item 7 above with item 8 (He assigns a task, then lets me handle it), which has the lowest scale value in the cluster. Item 8 seems to be dealing with delegation rather than participative decision making, and thus does not require much personal interaction between subordinate and superior. In summary, although the analytic method of inter- preting the INDSCAL dimensions was little help, the intuitive approach yielded three rather consistent inter— pretations of the dimensions used by subordinates in per- ceiving leader behavior: (1) degree of supervisor initiating structure; (2) amount of task concern by sub- ordinate; (3) degree of personal interaction between sub- ordinate and supervisor. 71 Related to the meaning of the INDSCAL dimensions is the question of the relative saliencies or weights given the dimensions by subordinates as they perceive supervisory behavior. Figures 5, 6, and 7 present the saliencies of each of the three INDSCAL dimensions for each subject in the study. Figure 5 indicates that, on the average, subjects place greater perceptual weight on the initiating structure dimension 1 than on the task concern by subordinate dimen- sion 2. Similarly, in Figure 7, the average saliency for the initiating structure dimension 1 appears to be greater than that given the subordinate-supervisor interaction dimen- sion 3. Figure 8 shows that the subjects placed about equal weight on dimensions 2 and 3. The overwhelming importance placed on the initiating structure dimension by subordinates in their perception of leader behavior is reasonable, considering the context of the study. The subject's task was to scale the leader behaviors of a "hypothetical male supervisor in industry." Given the implied organizational context of those behaviors and the formal organizational role expectations associated with supervisors in industry, it is no surprise that initiating structure emerged as the strongest perceptual dimension used by subjects. Indeed, for many subjects, leadership may well be synonymous with initiating structure. 72 .9-— 0 33 .8— 039 019 .7— .40-15 .7 047 .6— . 29 017 O .16-1.45.51 6 050 03 . 34 . 4 "" . 5_ 02609 ‘. .14 g .44 030 .22 048 .,..| 9 ' .38 g .35 2 52 as, 011 .25 . .4— 020 Q o 36 o 10 ‘12 031 O 37 .3_ C 8 O 32 '46 ‘ 41 '28 .2“ ' 43 018 ‘ 049 . o 12 1_ .21 24 '27 0 I I I I I I .1 .2 3 .4 .5 .6 Dimension 2 Figure 5.--INDSCAL Subject Coordinates for Dimensions 1 and 2 from Three-Space INDSCAL Solution. 73 .9- ~33 C8— 039 ~19 .7- '40 ~15 . 7 .47 -6‘ .6 a? ~45 .1 051 :4 016 ~3 60 34 H .S-I 014 26’ '9 a ~ 48 -3 ~38 22'44 ~2 ’-52 ' 30 U) o 5 35 8 ‘11,25 ’3 ~4‘ 10..36 ‘30 ’42 .31 57 3— 08 .4; 32 . O 38 41 .2- .43 018 ~12 ~49 .24 .1— '21 '27 n ' I I I I I *‘I .1 .2 3 .4 .5 Dimension 3 Figure 6.--INDSCAL Subject Coordinates for Dimensions 1 and 3 from Three-Space INDSCAL Solution. 74 .6- .5- ~38 ~46 ~42 ~4 - 012 I32 0 '4 ' 028.14 :"£532 . . ~24 o, 6 . 31 ~37 22 a .47 O o 02 '3'; .3_ ~10 ~17 25 ~30 a I Q) I a 36 33.5“” ~41 ~21 ~9 ~43 c: 51-18 ~11~s .26 :3 ~29 ~49 .2_ .7 45 .15 27: 50 4O .1 O . .20 19 '33 d6 l—J ~39 0 1 I I l I 1 .1 .2 3 .4 .5 .6 Dimension 3 Figure 7.--INDSCAL Subject Coordinates for Dimensions 2 and 3 from Three-Space INDSCAL Solution. CHAPTER IV EXPERIMENT II: EXPERIMENTAL MANIPULATION 0F LEADER INITIATING STRUCTURE, CONSIDERATION, AND PARTICIPATIVE DECISION MAKING Experiment 11 was designed to test hypotheses 2-6. Given the approximate independence of leader behavior clusters found in experiment I, the general aim of this study was to experimentally manipulate in descriptions of supervisors the three aspects of supervisor behavior: participative decision making, initiating structure, and consideration. W Subjects Seventy-two male undergraduates enrolled in intro- ductory psychology volunteered to participate for extra credit points toward their final course grade. All subjects were recruited one term after experiment I, and none of experiment II subjects had participated in experiment I. 75 76 Design A 2 x 2 x 2 x 2 x 2 design was used to test the effects of high-low subject Need for Independence, high-low subject Authoritarianism, high-low supervisor participative decision making, high-low supervisor initiating structure, and high-low supervisor consideration factors on subject attitudes toward supervisor and subject work motivation. Specifically, Need for Independence and Authoritarianism were subject classification factors, while supervisor consideration, participative decision making, and initiating structure were repeated factors. Figure 8 illustrates the design. A 2 x 2 x 2 design, all repeated measures, was used to determine whether the intended leadership treatment manipulations were operating. This design was run three times, once for each rating dimension (discussed later under procedure section). Stimulus Materials Based on the findings of experiment I, eight descriptions of supervisors were constructed to vary from low to high along the consideration, participative decision making, and initiating structure dimensions. In order to construct these descriptions, the 15 original leader behavior items had to be reflected. Table 8 77 .nmflmon Houomhao>fim mo mmammwn ownmmuonugwouowwm .mHSuodnum wcflumHuHcH Homfi>uoanm u mH .GONummHofiuHmm Homw>uomnm n m .GOHuononaoo HomH>Homam n o .ooaooaommoaH mom omoz uoonnnm u Hz .amflamfiumuwuonu5¢ n :< .304 n A .nwwm u m "moowuma>onnn¢ 7. wow D,c: i r1>,Is > r13,0 The correlations among the dimensions do not support hypothesis 3. All three dimensions are about equally related in the present study. It is not immediately apparent why this hypothesis was not supported, nor in retrospection, why it should be supported. The rationale for the hypothe- sis was originally a hunch based on previous research that used measures of consideration, participative decision making and initiating structure that were conceptually sloppy and often confounded with variance excluded from the definitions of these dimensions used in the present study. Hypothesis 4: Subordinate's perception of super- visor behaviors encouraging subordinate participa- tion is positively related to: (a) subordinate satisfaction with the supervisor. (b) subordinate motivation to work for the super- visor. 114 (c) subordinate's attitude toward the task competence of the supervisor. (d) subordinate's attitude toward the inter- personal style of the supervisor. The consistently significant and strong main effects sup- port all parts of hypothesis 4. That is, having a super- visor who actively allows or encourages his subordinates to participate in decisiondmaking in matters related to their jobs tends to be related to increased subordinate general satisfaction with the supervisor as well as increased moti- vation to work for that supervisor. Specific subordinate attitudes toward the supervisor's task competence and inter- personal style tend to be more favorable when the supervisor tends to be high on the degree to which he encourages sub- ordinate participation than when he is low. This set of findings is consistent with the bulk of the research reported on the effects of participative decision making. Hypothesis 5: Subordinate's perception of initiat- ing structure behaviors in his supervisor is: (a) negatively related to subordinate's satisfac- tion with supervisor. (b) positively related to subordinate's self- reported motivation to work under his super- visor. (c) negatively related to subordinate's attitude toward the interpersonal style of the super- visor. (d) negatively related to subordinate's attitude toward the task competence of the supervisor. 115 The strong and consistently significant main effects for the initiating structure manipulation on all dependent variables support only hypothesis 15 (b), and suggest the opposite of hypothesis 15 (a), (c), and (d). When the hypotheses for initiating structure were originally conceived in the light of very inconsistent previous research findings, initiating structure was thought of as a negative attribute of supervisors because it seemed to represent a threat to the subordinate's perceived freedom (Brehm, 1966). Given the assumption that most people don't want to have their freedom restricted, it seemed plausible that supervisors high on initiating structure would have subordinates who were dissatisfied in general with their supervisor, and who would also express their dissatisfaction in more specific negative attitudes toward the supervisor's task competence and interpersonal style, if not in their motivation to work (which could be based on fear). One plausible reason Fleishman and Harris (1962) and Skinner (1969) found negative and curvilinear relation- ships between initiating structure and two indices of satisfaction (turnover and grievances) is that their measures of initiating structure appear to be confounded with or included a large portion of autocratic or "arbitrary” variance. For example, consider the following items on the initiating structure scales used by both Fleishman and Harris (1962) and Skinner (1969): 116 He rules with an iron hand. He asks for sacrifices from his foremen for the good of the entire department. He insists that his foremen follow standard ways of doing things in every detail. He "needles" foremen under him for greater effort. He decides in detail what shall be done and how it shall be done. U'I-L‘ (JO NH Psychologically, it is no great surprise that subordinates would be unhappy with this kind of supervisor because it largely represents something far beyond the concept of initiating structure as defined in the present study. It's one thing to organize, coordinate, and set mutual expecta- tions, and quite another thing to be brutal, arbitrary, and autocratic. The former kind of initiating structure can serve useful psychological functions for subordinates, while the latter may incur psychological costs subordinates are unwilling to pay. To carry the analysis of the positive functions of initiating structure a bit further, consider this set of leadership behaviors as communicative acts within the con- text of information theory. Gardner (1962) states that a communicative act yields information only if it reduces a condition of ignorance or uncertainty. Thus, prior uncer- tainty is a necessary prerequisite for information. Further- more, the amount of information provided by a communicative act is determined by the amount of ambiguity or uncertainty in the situation. 117 Weick (1969) relates the twin notions of information and uncertainty to the social psychology of organizing. He states that the major purpose of organizing is the removal of equivocality from information and structuring processes to make this removal possible. The important point behind these references to information theory concepts is that when initiating structure is conceived of as a means of reducing uncertainty for sub- ordinates by letting them know what is expected of them and the ways in which the work is organized and coordinated, then the initiating structure activities of the supervisor are perceived in a favorable light. Of course, this entire analysis is predicated on the reasonable assumption that people in general, and people in organizations in particular, find uncertainty an unpleasant psychological state. The information theory explanation of the positive effects of leader initiating structure is compatible with the instrumentality theory explanation offered by House (1971). His major point is that in certain situations where there is role ambiguity and/or the task is very complex, leader initiating structure exerts positive attitudinal and motivational effects because such leader behavior reduces subordinate role ambiguity which is usually associated with low path goal instrumentality. In other words, by initiating structure in an ambiguous situation, the leader helps to 118 clarify the path to the subordinate's desired outcomes or pay-offs. Hypothesis 6: Subordinate's perception of con- sideration behaviors in his supervisor is: (a) positively related to subordinate's satisfac- tion with his supervisor. (b) positively related to subordinate's motivation to work for his supervisor. (c) positively related to subordinate's attitude toward the supervisor's interpersonal style. (d) positively related to subordinate's attitude toward the supervisor's task competence. The strongly significant main effects of supervisor con- sideration on all four of the dependent variables support the above hypotheses. When subordinates perceive supervisors as being considerate in their behaviors, subordinates tend to be generally satisfied with the supervisor, motivated to work for the supervisor, and have positive attitudes toward the supervisor's interpersonal style and task competence. In the contexts of hypotheses 4, 5, and 6, it is appropriate to comment on the patterning of main effects on the dependent variables. Consistent with previous litera- ture reviewed earlier, consideration and participative decision making exert a greater influence on subordinate satisfaction and motivation than initiating structure, although the effects of initiating structure were in a direction opposite that predicted by previous research. 119 It is also gratifying to see (Table 16, page 99) some experimental evidence for the construct validity for the task competence and interpersonal style dimensions of interpersonal evaluation. For example, the consideration factor is the supervisor dimension conceptually closest to interpersonal style (Johnson, Siegel, and Wakeley, 1972) and explained twice as much interpersonal style variance as task competence variance. Similarly, the initiating structure factor is closest conceptually to task competence, and not surprisingly explains about four times more task competence variance than interpersonal style variance. Participative decision making which can be viewed as con- ceptually similar to both interpersonal style and task competence, produced effects of about equal magnitude on these two dependent variables. Beyond the testing of hypotheses, a number of other interesting and unexpected findings have emerged and should be further discussed. First, in terms of the magnitude of the main effects of the manipulated consideration, partici- pative decision making, and initiating structure dimensions, we seem to have an additive, main class model of leadership. That is, at least in terms of the attitudinal and motiva- tional dependent variables studied, the main effects dominated while the observed interactions were always quite trivial by comparison. In general, the explanatory power 120 of the significant interactions was disappointing. Even the consistently significant interaction of initiating structure and participative decision making on all the dependent variables was interesting but trivial. In relation to this finding of trivial inter- active effects, it should be noted that in previous research findings interactions between consideration and initiating structure (Fleishman and Harris, 1962) reported no index of the strength of the effect (or even the F values for that matter). The studies of the interactive effects of participation and personality on subordinate attitudes by Vroom (1960) and Tosi (1970)'were expressed as differences between correlations, and are not readily comparable to an n2 for a significant interaction in an analysis of variance design. One suspects thata re-analysis of the Vroom (1960) study would be enlightening and disappointing. Of course, a number of reasons may explain why more and stronger interactive effects of the supervisor dimen- sions and subordinate personality characteristics did not emerge. One reason that the previously observed interaction of initiating structure and consideration on subordinate satisfaction (Fleishman and Harris, 1962; Skinner, 1969) did not emerge is that initiating structure was defined quite differently in their study. Specifically, initiating structure was defined to include autocratic, arbitrary 121 variance which is so obviously distasteful and psychologically costly to subordinates. Hence, with our new definition of initiating structure, one would not expect level of con- sideration to help compensate for the undesirable effects of autocratic initiating structure since this undesirable variance in initiating structure was not present. Another reason, this one for the lack of meaningful interactions between personality characteristics (Need for Independence and Authoritarianism) and the supervisor behavior dimensions, especially participative decision making, may be that there was relatively little variability in the personality characteristics for the college popula- tion studied. Vroom (1960) did not report cut—off scores for the Authoritarianism and Need for Independence measures he used. Thus it is not possible to directly compare results, a problem noted by Tosi (1970) in his attempted replication of the Vroom study. It is likely, however, that the college subjects used in the present study were much.more homogeneous in their level of Need for Indepen- dence and Authoritarianism than Vroom's sample, so that median splits on each of these personality characteristics may not have produced meaningful high and low groups. All of the subjects in the present study may possibly be com- parable to high Need for Independence and low Authoritarian- ism respondents in Vroom's (1960) sample. CHAPTER V SUMMARY, CONCLUSIONS, PRACTICAL IMPLICATIONS, AND AREAS FOR FUTURE RESEARCH Summary Subjects perceived supervisor behaviors belonging to three relatively distinct groups: participative decision making, consideration, and initiating struc- ture . These three groups of supervisor behaviors showed low positive correlations with one another. The three groups of supervisor behaviors can be manipulated approximately independently of one another. The INDSCAL dimensions used by subjects in perceiving supervisory behaviors, and which explained variation within and between the a priori clusters in three- space, were: (a) initiating structure, (b) task concern by subordinate, and (c) subordinate-supervisor interaction. 122 123 5. Supervisor participative decision making, consideration, and initiating structure each exerted powerful main effects on subordinate attitudes and motivation. 6. Unexpectedly, a high level of initiating structure in the supervisor was associated with positive subordinate attitudes. 7. Interactive effects among supervisor behavior dimensions and subordinate personality variables were trivial in magnitude, though sometimes statistically significant. Conclusions On the basis of the findings of the present study, it is concluded that subordinates like and are most moti- vated to work for supervisors who (1) allow or encourage subordinates to participate in decisions related to their jobs; (2) show a warm, positive attitude toward and concern for subordinates; and (3) organize, coordinate, and set mutual expectations. Going back to the troubled young vice-president mentioned in the introduction, the answer to his question about how leadership fits in with a participative management system is clear: Don't think about participative decision making, initiating structure, and consideration as opposing dimensions. Rather, think of them as mutually reinforcing 124 aspects of leadership behavior. From this perspective, it is easy to see how supervisors can organize, coordinate, and set mutual expectations £23 participative decision making by subordinates while showing a positive, warm attitude toward them. In making the above conclusions, a few statements of caution are in order. First, the subjects in the present study did not actually experience the supervisor's behaviors; they only read descriptions, and responded a§_i£ he were their supervisor. While some have praised such role-taking methodology in psycholdgical research (Schultz, 1969; Kel- man, 1967) as an alternative to pure deception, others have condemned it (Miller, 1969; .Aronson auui Carlsmith, 1968). The controversy is far from being resolved; however, it is fair to say that the role-taking methodology used in the present study may limit generalization of its findings beyond statements of abstract preferences of subjects. Second, and implicit in the above tentative con- clusions, is the assumption that other important employee behaviors such as quantity and quality of productivity, turnover, and absenteeism are related to employee satisfac- tion and motivation. As Brayfield and Crockett (1955) point out in their classic review of the literature, the assumed link between employee attitudes and employee performance is at best a very tenuous one. Locke (1969, 1970) and Lawler 125 (1970) have both commented at length on the theoretically complex relations existing among satisfaction-motivation concepts, and job performance. Therefore, tests of the effects of supervisor consideration, participative decision making and initiating structure on other employee per- formance and satisfaction variables are needed, preferably in a field situation. Practical Implications Given the limitations of the present study, and the need for extension of the study to include other dependent variables in a field setting, the implications for training of supervisors are clear. Because the leader- ship dimensions are expressed in terms of usually concrete behaviors, training programs could be devised in which supervisors are first assessed as to where they fall in each of the three dimensions, then efforts directed toward moving them toward more optimal regions of the leadership behavior cube, shown in Figure 9. One might conceive of the training program as an extension of the Blake and Mouton (1969) managerial grid to three dimensions. The Scanlon Plan (Lesieur, 1958) is a company wide system of participative management, complete with organiza- tional structures to encourage and maintain participative 126 Consideration ) Participative Decision Making Figure 9.--Johnson Leadership Behavior Cube. 127 decision making throughout the organization. However, when encouraging participative decision making is viewed as a supervisory function, participative decision making in the Scanlon Plan becomes much more dynamic, and complements the organizational structures for participative decision making. Therefore, a managerial supervisory training program using the three-dimensional leadership cube shown above may be a helpful tool for preparing companies which are planning to adopt the Scanlon Plan. For companies with the Scanlon Plan already installed, the three-dimensional leadership cube training could serve as a useful refresher course. Areas for Future Research While the practical implications of a three- dimensional model of leadership behavior are fascinating, future research is needed to verify and extend the findings of the present study in other settings. For example, though it is clear in the present study that being high on consideration, participative decision making, and initiating structure is the optimal set of leadership behaviors for maximizing subordinate satisfaction and motivation, other combinations may be more optimal in certain organizational contexts. Thus, we need to examine further the situations in which the supervisor functions. Is the optimal 128 combination of leader behavior dimensions the same for female supervisors? Do sex of supervisor and sex of subordinate interact to affect optimal leadership behavior combinations? How do consideration, initiating structure, and participative decision making depend on the degree of group favorableness, a la Fiedler's (1967) method of measuring the situation, in affecting subordinate attitudes and behavior? When the subordinate's job is routine, repetitive, and/or unpleasant, what then are the effects of supervisor participative decision making, initiating structure, and consideration behaviors on subordinate's attitudes, motivation, and performance? How effective are the three leadership dimensions in organizations having dif- ferent technologies? Are different leadership styles needed in different levels of a formal organization? Perhaps the optimal combination of leader behavior dimensions is different for different indices of supervisor effectiveness such as quality and quantity of productivity, turnover, absenteeism, etc. Another research problem is whether the considera- tion, initiating structure, and participative decision making leader behaviors are relatively independent in actual super- visors in industry, and not simply in the cognitive maps of perceivers. To what degree can these classes of behaviors be changed? Are consideration, initiating structure, and 129 participative decision making behaviors manifestations of a relatively permanent and unchanging, underlying motivational structure as Fiedler (1967) suggests? Viewing leadership behavior as the result of a dynamic interaction between supervisor and subordinate, it is possible that subordinate attitudes and behavior may in fact cause supervisor behaviors (Lowin and Craig, 1968). How do participative decision making, initiating structure, and consideration interact and change over time in the development of a dyadic relationship between super- visor and subordinate? Thus, to further examine the causal relations among supervisor consideration, participative decision making, and initiating structure, and subordinate attitudes and behavior, future cross-lagged panel analysis in the field situation may provide us with useful answers to these questions. Future research should be directed toward further investigating the degree to which supervisor consideration, participative decision making, and initiating structure is related to subordinate's overall job satisfaction, of which satisfaction with supervision is only a part. REFERENCES REFERENCES Anderson, L. R. Leader behavior, member attitudes, and task performance of intercultural discussion groups. Journal of Social Psychology, 1966, 62, 305-319. Argyle, M., Gardner, G., and Cioffi, F. Supervisory ‘methods related to productivity, absenteeism, and labor turnover. Human Relations, 1958, 11, 23-40. Aronson, E., and Carlsmith, J. M. Experimentation in social psychology. In G. Lindsey and E. Aronson (Eds.), Handbook of social psychology. Vol. 2. Reading, Mass.: Addison4wesley,il968. Bachman, J. G., Smith, C. G., and Slesinger, J. A. Con— trol, performance, and satisfaction: an analysis of structural and individual effects. Journal of Personality and Social Psychology, 1966, 4, 127- 136. . Bales, R. F. Task roles and social roles in problem- solving groups. In E. E. Maccoby, T. M. Newcomb, and E. L. Hartley (Eds.), Readings in social 8 cholo (3rd ed.). New YOER: Holt, Rinehart, E Winston, 1958, pp. 437-447. Bass, B. M. Leadership opinions and related characteris- tics of salesmen and sales managers. In R. M. Stogdill and A. E. Coons (Eds.), Leader behavior: its description and measurement. 7Cqumbus: *Bureau Of Business Research, Ohio State University, 1957. Bass, B. M. Organizationalypsychology. Boston: Allyn and Bacon, 1965. Baumgartel, H. Leadership, motivations, and attitudes in research laboratories. Journal of Social.Issues, 1956, 12(2),_24-31. 130 131 Beer, M. Leadership, employee needs, and motivation. Columbus: *Bureau of Business Researoh, Ohio State University, Monograph No. 129, 1966. Besco, R. 0., and Lawshe, C. H. Foremen leadership as per- ceived by supervisor and subordinate. Personnel Psychology, 1959, 12, 573-582. Blake, R. R., and Mouton, J. S. Building a dynamic corpora- tion through grid organization development. Reading, Mass.: Addison:Wesley, 1969. Bowers, D. G., and Seashore, S. E. Predicting organiza- tional effectiveness with a four factor theory of leadership. Administrative Science Quarterly, 1966, 11, 238-263. Brayfield, A. H. , and Crockett, W. H. Employee attitudes and employee performance. Psychological Bulletin, 1955, 52, 396- 424. Brehm, J. W. A theory of psychological reactance. New York: Academic Press,51966. Carroll, J. D. Individual differences and multidimensional scaling. In R. N. Shepard, A. K. Romney, and S. B. Nerlove (Eds.), Multidimensional scaling: theory and applications in the behavioral sciences. volume 1nyheory. New York: SeminarPPress, 1972. Cartwright, D., and Zander, A. (Eds.). Group dynamics: research and theory. Harper & Row, 1968. Chang, J. J., and Carroll, J. D. How to use PROFIT, a com- puter program for property fitting by optimizing nonlinear or linear correlation. Murray Hill, N.J.: Bell Laboratories Technical Report, undated. Coch, L., and French, J. R. P. Overcoming resistance to change. Human Relations, 1948, 1, 512-532. Cohen, J. Statistical power analysis in the behavioral sciences. *Now‘York: Academic Press, 1969. Cronbach, L. J. .The two disciplines of psychology. Ameri- can Psychologist, 1958, 12, 671-684. Davison, M. L., and Jones, L. E. Differences in multi- dimensional perceptions of the organization as a function of formal and informal status. Paper 132 presented at the Midwestern Psychological Association Convention, 1973. Dawson, J. E., Messé, L. A., and Phillips, J. L. Considera- tion and initiating structure: instructor-leader behavior influencing student performance. Journal of Applied Psychology, 1972, 56(5), 369-376. Day, R. C., and Hamblin, R. L. Some effects of close and punitive styles of supervision. American Journal of Sociology, 1964, 69, 499-510. Fiedler, F. E. A contingency model of leadership effective- ness. In L. Berkowitz (Ed.), Advances in egperi- mental socialypsychology. Volume I. New York: Academic—Press, 1964, 149-190. Fiedler, F. E. A theory of leadership effectiveness. New York: McGraw-Hill, 1967. Finn, J. D. Univariate and multivariate analysis of variance and covariance: a Fortran IV program. Occasional Paper No. 9. Office of Research Con- sultation, School for Advanced Studies, College of Education, Michigan State University, March, 1970. Finn, J. D. Multivariate analysis of repeated measures data. Multivariate Behavioral Research, 1969, 4, 391-413. Fleishman, E. A. Manual for leadership opinion question- naire (1969 Revision). Science Research Associates, Inc., 1969. Fleishman, E. A., and Harris, E. F. Patterns of leadership behavior related to employee grievances and turn- over. Personnel Psychology, 1962, 15, 43-56. Fleishman, E. A., Harris, E. F., and Burtt, H. E. Leader- ship and supervision in industry. Columbus: Bureau oflEducational Research, Ohio State Uni- versity, Research Monograph No. 33, 1955. Fleishman, E. A., and Peters, D. A. Interpersonal values, leadership attitudes, and managerial "success." Personnel Psychology, 1962, 15, 127-143. Fleishman, E. A., and Simmons, J. Relationship between leadership patterns and effectiveness ratings among Israeli foremen. Personnel Psychology, 1970, 23, 169-172. 133 French, J. R. P. Field experiments: changing group productivity. In J. G. Miller (Ed.), Egperiments in social process: a symposium on social psychology. New York: ‘McGraw-Hill, 1950. French, J. R. P., Israel, J., and As, D. An experiment on participation in a Norwegian factory. Human Relations, 1960, 13, 3-19. French, J. R. P., Kay, E., and Meyer, H. Participation and the appraisal system. Human Relations, 1966, 19, 3-20. Gardner, W. R. Uncertaintyyand structure as psychological concepts. New’York: Wiley, 1962. Gomberg, W. The trouble with democratic management. Trans- action, l966, 3(5), 30-35. Graen, G., Dansereau, F., Jr., and Minami, T. Dysfunctional leadership styles. Organizational Behavior and Human Performance, 1972, 1(2), 2164236. ‘ Gruenfield, L. W., Rance, D. E., and Weissenberg, P. The behavior of task-oriented (low LPC) and socially- oriented (high LPC) leaders under several conditions of social support. Journal of Social Psychology, 1969, 12, 99-107. Halpin, A. W. The leader behavior and effectiveness of air commanders. In R. M. Stogdill and A. E. Coons (Eds.), Leader behavior: its description and measurement. COlumbus: Bureau of Business Researdh, Ohio State University, 1957. Halpin, A. W. The leadership behavior and combat performance of airplane commanders. Journal of Social Psychology, 1954, 42, 19-22. Halpin, A. W., and Winer, B. J. A factorial study of the leader behavior descriptions. In R. M. Stogdill and A. E. Coons, (Eds.), Leader behavior: its descrip- tion and measurement. Columbus: Bureau of Business Research, Ohio State University, 1957. Hamstra, B. W. The perception of role positions: a multi- dimensional scaling analysis. Paper presented at Midwestern Psychological Association Convention, 1973. 134 Hays, W. L. Statistics for the social sciences. Second- edition. Neinork: Holt, Rinehart, and Winston, 1972. Heller, F., and Yukl, G. Participation, managerial decision making, and situational variables. Or anizational Behavior and Human Performance, 1969, 2, 227-241. Hemphill, J. K. Leader behavior associated with the admin- istrative reputations of college departments. In R. M. Stogdill and A. E. Coons (Eds.), Leader behavior: its descriptions and measurement. Colum- bus: Bureau ofiBusiness Research, Ohio State University, 1957. Hemphill, J. K., and Coons, A. E. Development of the leader behavior description questionnaire. In R. M. Stogdill and A. E. Coons (Eds.), Leader behavior: its description and measurement. Columbus: Bureau of Business ResearCh, Ohio State University, 1957. House, R. J. A path goal theory of leader effectiveness. Administrative Science Quarterly, 1971, 16(3), 321- 338. Hummel, T. J., and Sligo, J. R. Empirical comparison of univariate and multivariate analysis of variance procedures. Psychological Bulletin, 1971, 16(1), 49-57. Johnson, R. H., Siegel, A. L., and Wakeley, J. H. The effects of personality needs on interpersonal evaluation. Unpublished manuscript, 1973. Jones, E. E. Authoritarianism as a determinant of first impression formation. Journal of Personality, 1954, 16, 107-127. Kahn, R. L. Human relations on the shop floor. In E. M. Hugh-Jones (Ed.), Human relations and modern . mana ement. Amstefdam, The Netherlahds: North- Holland Publishing Company, 1957. Katz, D., Maccoby, N., Gurin, G., and Floor, L. Productivity, supervision, and morale amongrailroad workers. Ann Arbor: SurveyResearch Center, University of Michigan, 1951. 135 Katz, D., Maccoby, N., and Morse, N. C. Productivity, supervision, and morale in an office situation. Ann Arbor: University of Michigan, Institute for Social Research, 1950. Kay, E., Meyer, H. H., and French, J. R. P. Effects of threat in a performance appraisal interview. Journal of Applied Psychology, 1965, 62, 311-317. Kelman, H. C. Human use of human subjects: the problem of deception in social psychological experiments. Psychological Bulletin, 1967, 61, 1-11. Kirk, R. E. Experimental design: yprocedures for the ' behavibralisciences. Belmont, Galifornia:’iWads- worEh, 1968. Lawler, E. E., III. Individualizing the organization. Paper presented at the Midwestern Psychological Association Convention, 1972. Lawler, E. E., III. Job attitudes and employee motivation: theory, research, and practice. Personnel Psychology, 1970, 11, 223-237. Lawrence, L. C., and Smith, P. C. Group decision and employee participation. Journal of Applied Psy- chology, 1955, 12, 334-337. Likert, R. Developing patterns in management. Strengthening management for the new technology. New Ybrk: AmericanManagement Associatibn, 1955. Likert, R. New patterns of management. New York: McGraw- Hill, 1961. Locke, E. A. Job satisfaction and job performance: a theoretical analysis. Opganizational Behavior and Human Performance, 1970, 6, 4849500. Locke, E. A. What is job satisfaction? Or anizational Behavior and Human Performance, 1965, 5, 309-336. Lowin, A. Participative decision-making: a model, litera- ture critique, and prescriptions for research. Organizational Behavior and Human Performance, 1968, 1, 68-106. 136 Lowin, A., and Craig, J. R. The influence of level of performance on managerial style: an experimental object lesson in the ambiguity of correlational data. Organizational Behavior and Human Performance, 1968, 1, 440-458. Lowin, A., Hrapchak, W. J., and Kavanagh, M. J. Considera- tion and initiating structure: an experimental investigation of leadership traits. Administrative Science Qparter1y, 1969, 16, 238-253. Maier, N. R. F., and Hoffman, L. R. Group decision in England and the United States. Personnel Psychology, 1962, 11, 75-87. , Maier, N. R. F., and Maier, R. A. An experimental test of the effects of "developmental" vs. "free" discus- sions on the quality of group decisions. Journal of Applied Psychology, 1957, 61, 320-323. Mann, F. C. Toward an understanding of the leadership role in formal organizations. In R. Dubin, G. C. Homans, F. C. Mann, and D. C. Miller, Leadership and productivity. San Francisco: Chandler Publishing Cempany, 1965. Mann, F. G., and Dent, J. The supervisor: member of two organizational families. Harvard Business Review, 1954, 11(6), 103-112. McCurdy, H. G., and Eber, H. W. Democratic vs. authori- tarian: a further investigation of group problem- solving. Journal of Personality, 1953, 11, 258- 269. McCurdy, H. G., and Lambert, W. E. The efficiency of small groups in the solution of problems requiring genuine cooperation. Journal of PerSonality, 1952, 16, 478-494. 'McMurray, R. N. The case for benevolent autocracy. Harvard Business Review, 1958, 16(1), 82-90. Meltzer, L. Scientific productivity in organizational settings. Journal of Social Issues, 1956, 11(2), 32-40. Miller, A. G. Role playing: an alternative to deception? A review of the evidence. American Psychologist, 1972, 11(7), 623-636. 137 Misumi, J., and Shirakashi, S. An experimental study of the effects of supervisory behavior on productivity and morale in a hierarchical organization. Human Relations, 1966, 11, 297-307. Morse, N. C., and Reimer, E.' The experimental change of a major organizational variable. Journal of Abnormal and Social Psychology, 1956, 11, 120-129. Nealey, S. M., and Blood, M. R. Leadership performance of nursing supervisors at two organizational levels. Journal of Applied Psychology, 1968, 11, 414-421. Newport, G. A. A study of attitudes and leadership behavior. Personnel Administration, 1962, 11(2), 42-46. Nunnally, J. C. Psychometric theopy, New York: McGraw- Hill, 1967. Oaklander, H., and Fleishman, E. A.. Patterns of leadership related to organizational stress in hospital settings. Administrative Science Quarterly, 1964, 1, 520-532. Parker, T. C. Relationships among measures of superviSory behavior, group behavior, and situational character- istics. Personnel Psychology, 1963, 16, 319-334. Patchen, M. Participation, achievement, and involvement on the job. Englewood Cliffs, N.J.: Prentice-Hall, Patchen, M. Supervisory methods and group performance norms. Administrative Science_Quarter1y, 1962, 1, 275-294. Rambo, W. W. The construction and analysis of a leadership behavior rating form. Journal of Applied Psychology, 1958, 61, 409-415. . Reilley, A. J. The effects of different leadership styles on group performance: a field experiment. Paper presented at the American Psychological Association Convention, Washington, D.C., 1969. Sales, S. M. Supervisory style and productivity: review and theory. Personnel Psychology, 1966, 11, 275- 286. Sample, J. A., and Wilson, T. R. Leader behavior, group productivity, and rating of least preferred co- worker. Journal of Personality and Social Psy: chology, 1965, 1, 266-270. 138 Schacter, S., Willerman, B., Festinger, L., and Hyman, R. Emotional disruption and industrial productivity. Journal of Applied Psychology, 1961, 61, 201-213. Schoenfeld, E. Authoritarian management: a reviving con- cept. Personnel, 1959, 16, 21-24. Schmidt, W. Personal communication, 1973. Schneider, D. J. Implicit personality theory: a review. Psychological Bulletin, 1973, 11(5), 294-309. Shaw, M. E. A comparison of two types of leadership in various communication nets. Journal of Abnormal and Social Psychology, 1955, 11,127-134. Schultz, D. P. The human subject in psychological research. Psychological Bulletin, 1969, 11(3), 214-228. Shepard, R. Introduction to Volume 1. In R. N. Shepard, A. K. Romney, and S. B. Nerlove (Eds.), Multi- dimensional scaling; theory and applications in the behavioral sciences. .Volume 1, theory. New York: Seminar Press, I972. Skinner, E. W. Relationships between leadership behavior patterns and organizational-situational variables. Personnel Psychology, 1969, 11, 489-494. Solem, A. R. An evaluation of two attitudinal approaches to delegation. Journal of Applied Psychology, 1958, 61, 36-39. Stanton, E. S. Which approach to management--democratic, authoritarian, or . . .? Personnel Administration, 1962, 11(2), 44-47. Stogdill, R. M. ‘Manual for the leader behavior description questionnaire. Form XII. .Columbus: Ohio State University,iBureau of Business Research, 1963. Stogdill, R. M., and Coons, A. E. Leader behavior: Its description and measurement. Ohio State University, Bureau of Business Researbh, Business Research ‘Monograph No. 88, 1957. Stogdill, R. M., Goode, O. S., and Day, D. R. New leader behavior description subscales. Journal of Psychology, 1962, 16, 259-269. 139 Stogdill, R. M., Goode, O. S., and Day, D. R. The leader behavior of corporation presidents. Personnel Psychology, 1963,.16, 127-132. Stogdill, R. M., Goode, O. S., and Day, D. R. The leader behavior of presidents of labor unions. Personnel Psychology, 1964, 11, 49-57. Strauss, G. Some notes on power equalization. In H. J. Leavitt (Ed.), The social science of organizations: four ers ectives. Englewood Cliffs, New Jersey: Prentice-Hall, 1963. Taylor, J. C. An empirical examination of a four-factor theory of leadership using smallest space analysis. Organizational Behavior and Human Performance, 1971, 6(3),249-266. Tosi, H. A. A.re-examination of personality as a determinant of the effects of participation. Personnel Psychology, 1970, 11, 91-99. Vaughan, G. M., and Corballis, M. C. Beyond tests of. significance: estimating strength of effects in selected ANOVA designs. Psychological Bulletin, 1969, 11, 204-213. ii" Vroom, V. H. Some personality determinants of the effects of artici ation. Englewood Cliffs, New Jersey: rent ce- a , 960. Vroom, V. H. Some personality determinants of the effects of participation. Journal of Abnormal and Social Psychology, 1959, 11, 322-327. Vroom, V. H., and Mann, F. C. Leader authoritarianism.and employee attitudes. Personnel Psychology, 1960, 11, 125-140. Wéick, K. E. Laboratory experimentation with organizations. In J. G. March (Ed.), Handbook of organizations. Chicago: Rand McNally, 1965. Wéick, K. E. The social_psychology of organizing. Reading, Massachusetts: *Addison-Wesley, 1969. weissenberg, P., and Kavanagh, J. The independence of initiating structure and consideration: a review of the evidence. Personnel Psychology, 1972, 11(1), 119-130. 140 Wéssel, D., and Null, C. Personal communication, 1973. Winer, B. J. Statistical principles in experimental design. Second edition. New Ybrk: ’McGraw-Hill, 1971. WOfford, J. C. Behavioral styles and performance effective- ness. Personnel Psychology, 1967, 16, 461-496. WOfford, J. C. Factor analysis of managerial behavior variables. Journal of Applied Psychology, 1970, 54, 168-173. Wofford, J. C. Managerial behavior, situational factors, and productivity and morale.. Administrative Science Quarterly, 1971, 16(1), 10-18. Yukl, G. A. Conceptions and consequences of leader behavior. Paper presented at the annual convention of the California State Psychological Association, Newport Beach, 1969. Yukl, G. A. Leader LPC scores: attitude dimensions and behavioral correlates. Journal of Social Psychology, 1970, 16, 207-212. Yukl, G. A. Toward a behavioral theory of leadership. Organizational Behavior and Human Performance, 197l,‘6,_4lE-440. APPENDIX A SPECIMEN SUBJECT PRINTOUT SHOWING MULTIDIMENSIONAL SCALING TASK GIVEN SUBJECTS IN EXPERIMENT I 141 panama rum) 0". -l r ‘41 a .»H UJQZa: u: when 20:» .ywa» a w7chwmq my .cmruqoaaca >4~wqd oz: 1: up sorruzau mu m: a .00 o» w>¢1 H xwoz wt» Mmo5CmIow w: .xhux 2103 Oh cum: H Jauowwar or» w>cx p but» »H c» vumm mr « .pwmc xzur» H >¢z w1» yac: >x cc 1r mpuo w- .monchuc ozfiyar z» uo¢.v 1: wax u: a .waamquz Jazovamu >r «cu ~30 myoca ur .u: uo nuhowexu me but: gory tr mew; c: « .mmaanu: 442cmoua >: new 130 myccg m: .ooa coco q es H zmzx zeppcmcu~aaa mumvugAxu w: a . .co 0» w>c1 w ywcz wt» muqfituruv wr .wZOHMHUmo 02h¥u~co¢ o» u: mxeuna my .m: «on mozcraouamn ac mcaaoz¢»m ucuzeame vrhnpzuq: m: « .moa coco c on a rut: zo~ca~um~aaa mmummmaxu w: .rnz that cszua» 7u13 draw ha menu m1 mwxct w: u out «on wOZatmoumwQ no mcycozdhr “Fuzhuuo m2H¢~2HdI » 200: u:» MtdromrLU m: « .mcn oooo a on H zmrz reupcuuwwoaa wwwmuadxm ur .owxuqcadaa >4Hm¢m c2¢ m: a» yorrueou v» w: p u ohm uJOZaI wt mhma zmxb .ym«» « wrumwmc w: .m7OHMHCmo c2H¥t howuud IUHII MZOnmuouo an 230 war 0» Jaacm u02m34u2~ wt m1044¢ u: .th3 2UO:.O» cum: u Acmumhcr wz» w>az H he!» by c» mwuw w: >4wxuoz: >¢w> . :aowptsz hzmonbv -umrqz MVOu>Jw2ud hfius >4w¥~423 .co 0» m>aI H yvo: at» Utbrcdzum .m05 >t buuuud IUHII monmhcuo In 236 mm: o» JcDCM LQZuDJuZn ur #3044: .005 >3 bomuuc Ivar: mzo~mHomo 2H :36 My: ch JaDCm uLszJuzw ur mchqd .mOfi oooc c cm H zwzx zouhdwuwonaq vumvuooxw omen pt Zn Guanacma mwdacmocma m1» >uHccr or w: mzcaac .oo o» w>q1 H ymc: u!» mwaacurcm .mOauou: 442cmmud rr «cu bro mycca .ocs >2 huuuuc roux: wZONmHowo zn zxo mu: o» dance muzmaouzu mt mzoonc out «on mutttmoumua uc madcozchm MHHZwuuc vzncpzudt .GOfi >t wouuud rant: mzo~m~uwo an 230 mm: o» JcDCm muzwDJuZH ur vxoaqa .0: oh u>r 2H omwnacua muoncwcowa mt» >u~=ox 0» wt wxoqqc .mr you uoz4quu 024 wt Oh racruwou Va out mo nubomoxw Mu p.23 302! at aka; .IhH: vac: Ob cwuz u Jaudmbct wt» w>41» kn c» muum our «on mutttdnudua mo wcdtozf mcu »rc mxooa .mec 22H!» n has wt» xvcx >: CO at mpuJ .th! rum: ulnchh 2w!) umcu b. Juuu u: muxcx .12 «on uczaxaouaua Lo moaaozacw mhczauuc mzzqtzaqx .rnx xbux quch» zwxz umcu p¢ auwu u: wags: rrub u: w! w: m: w: u: w: NI KI w: LI m: w: wt u» u: m: u: m: m: m: w: u: w: w: 143 paws»; ram) pauvudza >dw> ormr thwz ozHyacb ZmIZ umaw he dawn mi muvct ocuzo4hwam 62¢ wt 0» >chuumu mH annexe Nth 2H hadn Mu: archwauozn mawcru: macaw mwtho 02¢ H pct» uwnv mwxat obmwt 22H!» m >¢3 mrh xvo: rt 00 Lt whwa .maouu U!» o» mbzuthmaficc u¥1 2H owaaaawd mmmaowucmn m1» >u~cor c» wt mzoaat .uu waozaz m: memo zwxh .2ma» « mzcnmma .r»: run: ozuyda» 2mzs umau he oumu u: mega: oaaoxa wt» 0» muzwthmafiaq mxct wt modux w: .ywmrwx 1m: a :4 u .monchwc ozuyqx Zu woarv wr Md: .0: oh u>¢x u rye: dz» muenourom coauto mt» :u want an: 02¢»deaz: mcwcxwt cacao amrho 024 n her» u~3m wqut ooOfi >2 bcuuuc rear: monMHowo Zn 230 mm: c» acacw muzwDJuZN wt m3044< our to Cwbcmoxw mm kcrx 302v NY waA .»~ mJozaz 1: memo zmxp .yma. a m2c~mm¢ .xpux xaox o» owe: a aqumubqt mt» u>t 2H owanaamd Mw&30muoma m1» >uwoor 0» mt v3044¢ .maqumI szcvcuu >1 mou 5:0 vxooa omos >t 2H owdnacum Mmaaowcooa wt» >ch0t o» m: mxoaa¢ otHr Ira: cruyaab ZuII wmcm >< amen m: mux«: .wacuduz Adzomomd >! new ~30 mXOOJ .mzo~vHowc wszdt 2H macrm mt Md: s .pmmc yzaxc H >¢x ex. gee: s: on tr when owdduJux JdZCVIuQ >t don 2.0 mxoon- w: w: u: ur a“ u: m: u: w: m: um wr u: w: w: m1 m: w: m: w: u: w: u: m1 m: m: 144 >4U8HJ >¢u> >4uxH423 .ow10408Qaa >JHMcm czd wt 0» >4HZLH1u mH oQDOUQ mt» 2H bade mHI czahm¢wcz3 maumtm: aaowo «mxpo c2¢ H wax» mnrm muyqx .IHHI xxoz OH auuz H AcHacht NIH w>4Hch c2: wt OH >chuHmu mH .QOfi rt bowuud IUHII monwHowo 2H :30 MH: Oh Jcbcw mozwndqu mt wreaaq .tHI :hH: ozHXJab rut: mmcw H4 meu u: vwxcr onaouu UZH 2H hddm me czabmxmoza mcwerw: aanmu mwrbc :2: H Help 003w ward: .0: OH w><1 H x101 LI» mugncwrum .er IHH: oszaah rwI: umcu ha Juwu m: mwydt . .va9 sz1H H >43 wt» rye: >r on wt mbwq .003 pt Hammad :onz wonmHomo 2H 23c MHz 0» Jancw moruaqqu m: wxcaau .muunc: Jazomama >2 cos »:c axooo .aaouu wt» 2H page mH: ozqhmmmoz: mauotmx abovn mmx»c 02: H were myom vcyqr out no nmhcmdxw mH pct: xczy m7 whwa .QDOEQ uxH 2H bran mHI 02¢Hmdmoz: maumtwt «some curve 02¢ H H41» myrw vaqt .xHx run: oznyoab zuz: umam he 4mmu m: muxar .mr no cupcmaxm mu par: :ozy a: mhmo .IHHI 2&0: OH owuz H Jdewpqt m1» u>dr H Harp HH OH mwwm .mOfi >t 2H omeacud mwabowcowo U!» >uHoor CH wt mzcaqd .uzoxe m!» 7H buds mHI archwauoz: mawmtw! maowo meHc cza H Harp worm mwxcx .thx xao: OH owwz H acHauhct uIH w>¢r H Her» HH 0» Mmum obwmo 32H!» H >¢3 mt» xro: >: C: at Mpua .z»~x xao: o» emu: a manqueax or» use: c haze .H o» muum .mos coon a on _ zmzx zoneaHUcaaaa mummmauxw .96 o» u>¢t H 260: mIH quacuxom .aao¢o wt» OH whzwxbm3501 wxct m7 moqu u: .mwmtuz 3w: « t‘ H >¢m> my u: or w: m: m: w: w: wr w: w: wr w: wr ur m7 w? w: u: m: u: u? ur w: w: w: m: uH 145 >4uxHJ >¢w> >4w¥HJ23 .HH unozar Hr mHuo zmxH .quH a «7ovaq .aaocu uxH 2H paqe me crchauoz: maumxw: aaoau umzHo 024 H HazH wyau mcya: .moq >:.2H.oH«H=cma «waaomcoua HxH HcHoor OH or m2033. .HVmc ngxH H Ha: mzH xmc: Hr co tr me4 .monchwc 02H¥qt 2H uwdzm wt max .005 Doom 4 00 H ZmI: ZOHHQchandq wuwmmmnxu .Hmmm szzH H >a: wzH ywo: >: on H: mHuH .oc o» w><1 H rec: m1» vwmzcuruw oQOfi 6006 4 00 H zwIJ onbcHuwnooq mmmmumoxw .hwum yzHIH H >¢3 wry xao: >t on at mhma ombo¢w w!» o» mszthDficc uydz mt roam: w: .Uumrm: :w: « ta H .IHHI umo: o» owuz H Jchuch wt» w>cz H Hal» HH 0» mwum .QOS >2 Houuud IQHI: monmHuwo 2H zzo er OH JdDCH wczmaqqu u! m1044« .uDo¢u wt» 2H h¢¢n me orchmauoza wnuerut canon nwxho oz: H barb mysm wHY¢x .ourodoxadc >4HmJC7UHmu mH .wacuam: aczovown >1 mom H30 myoca .Hme yzHIH H >43 NIH xwo: >: on H! mbma our com murardoudma no MCO¢QZ¢HM uszHumo vzHaHquz .mos H: 2H owcHach mmanomcoya HIH HuHoor cH or mzonna .HH moczax u: meo zuz» .ymaH « mchmva .mos Hz 2H cuaHacmm mumaoucoya qu HuHoor OH or mxcdoc .aaoau uxH 0H mquthasca uyqx m: menu: m: .«czrcz am: a :4 H «meadow: Jazovauo >r non H30 myooo .IHH: recs 0» owuz H JanmHaz m1» w>¢r H wax» HH 0» muum ouch GOOD 1 on H tux: onthowaad< mwwmmmoxm .;. .oo o» u>cx H xuox w!» mwaacmzov .HH mnazm: u: «Hm; zwzH .ymaH a monmm< .mr do ouHouaxu mH H¢w> u: m: w: w: m: mr m: w: uH w: m: w: u: a: w: w: w: mr uH w: u: w: w: m: m: 146 >JuxHa >¢u> "B >Jw¥Haz= .mr you uczarwnuana no moquchm HHHzHcHH qucHzHaz .HH woozqz Hr mHmH zHIH .ymcH a mchmua .m: «an mozqraouumd do mcyaczaHm LHHzHumc mquHzHar .wonmHumc 07HVc: 7H awarv at war .CmIqumdad >4HM¢U 62¢ w: or >4rrmHmu VH .xHHz xuox OH cwwz H Aquchr m1» w>cx H HdIH HH CH owum .HH moczcr H: va4 Zqu .YMcH c vchmmc .mOfi >t Huwuud ronx monmHuwc 7H 23c mHI o» 4<:Cu uu7urgqu Hr vxouac .wz «on wczqtmnuuua do wc~¢cr¢Hm HHHzHuuo mzHaHzHcr .Hr do cchwcxu wH quw :ozy tr mHHJ casedo m!» 2H Hana mHI D7dbmmwa23 mdwctut nacwu aerC 62¢ H Hex» m02m muyqr omen ccoo « CO H zwr: rcHHaHuuwdda chvuanxm .aaouo wa oH mHzcmenscc ~24: my maJHr m: .nmcym: :ur a :4 H .mon >r Howuuc ronx monmvao 2H :30 er oH Dazed mczmnoqu ur uzoqoq .a:o¢o m1» OH mhzuthm3fi6¢ wxcr wt wdgmr m1 Hmmcrmx 3w? a 1d H .QCfi ocoo 4 o: H rut: ZCHHcHouwdca wumuWnoxw .QCfi >t Howuua IUHII monmHowD 2H 230 MHz OH Jcscm woruzgqu w! U3CJJ< .GOH >t 2H cwaHDcmx «moaomccyd Ur» ruHCoa CH “v m1044¢ otHz IHH: ozHyaaH fur: mmcb Ha 40mm qr Vuvct .003 >2 Humumq IUHII monmHuwc 2H 220 mHI CH octcu mL7wDJu2H my VIOJHd orHI IHHI QzHy4aH ZuII uwam Ha Juuu a: rmyqx .wroHUHouc czHyaT 2H uwcrv “I ma: .Hvuo yzHrH H >4: wzH yro: H: cc tr mHHJ .dnoau urH o» mszthzfica mxar mt went: c1 .~ucrdz 3“: a in H .00 3H m>¢z H soc: tzH wtnzcurcm .Omrucorddc >HHv«m €21 wt a» >Hr71Hou VH s .monmHuuo ozHycx 2H macro a: ma: .cos Hz 2H ouuHacmm mHOHOHCoua uxH HHHoor oH ur uxcnna >uu> u? “1 m: m: m: H? v.7. Hr u: w: my uH U: uH w: H: uI w: 1H m: ml u: u: 147 ngxHJ Haw) HawaJZD .aDomo .mow >t Homuua Ionx monmHowo 2H 210 .nnomu >mw> .HH moozax t2 vHHo ztxH .ywHH a mrm _ .moauowz oarrmrtc >r ace H:; anCH .aDOdO wa OH MHZmerDfiCd urcz mr many? mi .wmmrmr x”: a r« H NIH 2H Hana mHI oquwmeZD muwdrml QDOWQ ourHo czq H HdIH worm Unvar .vonmHtoc orqur 2H mmarm n: ma: .mr do Cuchnxu rH H423 :czy m: tho .Hmmm yzHrH H cc: mrH ywo: >: o- my uHma .CMIUQCUQOG >JHUdu r2: u: CH >Jdrdeu mH .on 0» m>¢r H yyox tiH mmnnomzom au.ux nczowama Hr wcu Ho» nyooo V" (K .mroHqumo c7Hqu zH toaww w: m4: er rH JanrH acrmnoqu hr mwoonu u: OH Harrlwmu mH .ouxodoyddd >4qu rq mH HqTI Iorv tr VHHJ m n .mr uc nUchnxm .tHI IHHI OIHVJnH rwI: “mam Ha Jmmm m1 vaqr .moauntz chcvnmc >r non Hrc mynoa c . mH Her: xnyy u a: .rH ywr: Mr or nr mHuo ll} X |v.” 1 uu . U .Hmwu vthH H > .m:.Hva.: mszar rH conrv -1 Me: n y x ' th 7H Head er azaHmamnza mummruz nhpnm a :Ho n.H H HHIH “04p o.y vqx 2H uNnYm . _ H m .1HH1 vac: oH owmz H Jauwu .H at. u>qz H HaxH HM cH wntv .mr .un .(LHCanxu mu Halt. 107v. m2 MHu... .mCfi >I 2H CmmHDCmm wmobbutbwa mIH >CHCCT CH Hr MICJHc .rHI IHH1.ozHy4aH rmtr mmqu Hd Humu hi Mmycx .IHHI two: oH Cmmz HchHmuHa: wIH mqu H HqIH HH OH muuv .cow >r 2H emaHnoma mwajomocod HIH HuHcoz 0H m: mxoooq .ouroqmaanc HHqum 024 m1 0H >nr2cuqu mH a: m3 1;; lg I I II U) 'A.’ ll, [1' II w: my m1 m1 ur mt 148 >4w¥HJ >¢w> >JwvH42: >du> orH: IHHI quyacH Zurz wwam H< Huwu mt muvq: .aaoao qu OH Mquthaqca urn: m: meow: m: .vmarcr :w: a rd H .m: «on uoz¢r~0uawn do mcycozan HHHzHuuc mzHcHzHc: caseda NIH 2H Hutu mH: o7¢Hdeczn mauetwt daoyn amIHc cz< H HHIH term vuycz .oos ooou a as H :12: onHcHuuanuc ruvaaaxH .HH uqoqu c: mHuo zuxH .yV¢H c wrunva .oo oH use: H ywc: uxH umsacHrem .HH uoozcx or vao zuxH .ymaH a mrsHmuq .HH maozcr w: mHmJ zmzH HymcH < mrcHum< cabomu wIH OH Mszeraficc mxct wt mcquz m: Howotw: :uz c ta H .Omxu4Hmaw D?d w: OH >chwau vH .v7chHuwo qu¥¢r 2H wmdtv Lt mar .omxoqoaaaa HJHmau ozc m: OH >4rzHku mH .aaoao uxH oH mqumeaaca uya: m: meow: H: .wumtmx :w: H :c H oaaodu utH OH mquthafiac mitt mt maawt u: .«umrwt 3w: 4 :1 H .wachdwx aczomaua >t «on H30 M¥OCJ .Hmuc yszH H ».3 HxH xaoa H: on c: mHms .HH usozH: u: vHHn zmzH .ymaH « mzeHmmq w: uH w: u: w: v: w: u: mt uH Ur wt u: uH uH m1 m: u: APPENDIX B NEED FOR INDEPENDENCE (OPINION QUESTIONNAIRE I) AND THE F-SCALE MEASURE 0F AUTHORITARIANISM (OPINION QUESTIONNAIRE II) USED IN EXPERIMENT I AND II Instructions: 149 OPINION QUESTIONNAIRE I Answer each of the following questions by checking one of the alternatives to the right of the question that best describes your true beliefs or feelings. It is important to the conduct of this experiment that you respond to all the questions in a frank and honest manner. Your responses will be kept confidential. Your name and student number are required only for the purposes of (l) ensuring that you receive full experimental credit; and (2) matching up of data for statistical analyses. Your Name: Student Number: Sex: Male Female 150 How important is it for you to feel that you can run your life without depending upon people who are older and more experienced than you? How often do you find that you can carry out other people's suggestions without changing them any? How much humility do you think you should show to those whom you respect and admire? How much respect do you think should be shown to a judge even outside his court- room? How much do you usually want the person who is in charge of a group you are in to tell you what to do? When you have a problem how much do you like to think itlthrough yourself without help from others? How much respect do you think people should show to policemen? How hard do you find it to disagree with others even in your own thinking? Not at all Slightly Somewhat Very Extremely Rarely Sometimes Often Very often Almost always None at all A little Some Quite a bit Very much None at all Some Quite a bit Very much Extremely much Not at all A little Somewhat Quite a bit Very much Not at all Somewhat Quite a bit Very much Extremely much None at all Some Quite a bit Very much Extremely much Not at all Slightly Somewhat Quite Very 9. 10. ll. 12. 13. 14. 15. 151 How much do you think that the leaders of organizations to which you belong have the right to expect certain things from you to which you should conform? Not at all A little Somewhat Quite a bit Very much How much do you feel that you are not as Not at all good in most things as people who are A little older and more experienced than you? Somewhat Quite a bit Very much In school how much do you dislike teachers Not at all who have forceful and dominant personali- A little ties? Somewhat Quite a bit Very much If you have thought about something and Not at all come to a conclusion, how hard is it for A little someone else to change your mind? Somewhat Very Extremely How much do you feel that officers of the Not at all law should tell people what to do rather A little than ask them? Somewhat Quite a bit Very much In school, under which of If I were left completely alone these conditions would you to seek out whatever I wanted feel most comfortable? If I were given suggestions from teachers as to what might be the best to study If I were given some suggestions and some assignments to complete If I were instructed, given assignments, and tested occasion- ally If I were given daily instruc- tions, daily assignments, and frequent tests How much do you dislike being told to do something by a superior that is contrary to your wishes? Almost always Very often Often Sometimes Rarely Instructions: 152 OPINION QUESTIONNAIRE 2 Answer each of the following questions by checking one of the alternatives to the right of the question that best describes your true feelings or beliefs. It is important to the conduct of this experiment that you respond to all the questions in a frank and honest manner. Your responses will be kept confidential. Your name and student number are required only for the purposes of (l) ensuring that you receive full experimental credit; and (2) matching up of data for statistical analyses. Your Name: Student Number: Sex: Male Female 153 Obedience and respect for authority are the most important virtues children should learn. A person who has bad manners, habits, and breeding can hardly expect to get along with decent people. If people would talk less and work more, everybody would be better off. Science has its place, but there are many important things that can never possibly be understood by the human mind. Every person should have complete faith in some supernatural power whose decisions he obeys without question. Young people sometimes get rebellious ideas, but as they grow up they ought to get over them and settle down. What this country needs most, more than laws and political programs, is a few courageous, tireless, devoted leaders in whom the people can put their faith. Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree 10. ll. 12. 13. 14. 154. No sane, normal, decent person could ever think of hurting a close friend or relative. An insult to our honor should always be punished. There is hardly anything lower than a person who does not feel a great love, gratitude, and respect for his parents. Most of our social problems would be solved if we could somehow get rid of the immoral, crooked, and feeble-minded people. When a person has a problem or worry, it is best for him not to think about it, but to keep busy with more cheerful things. Nowadays more and more people are prying into matters that should remain personal and private. Some peeple are born with an urge to jump from high places. Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree ll llll I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree 15. 16. 17. 18. 19. 20. 21. 155 People can be divided into two distinct classes: the weak and the strong. Someday it will probably be shown that astrology can explain a lot of things. No weakness or difficulty can hold us back if we have enough will power. Mbst people don't realize how much our lives are controlled by plots batched in high places. Human nature being what it is, there will always be war and conflict. Familiarity breeds contempt. Nowadays when so many different kinds of people move around and mix together so much, a person has to protect himself especially carefully against catching an infection or disease from them. Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree Strongly agree Agree I can't make up my mind Disagree Strongly disagree 156 22. The wild sex life of the old Greeks and Strongly agree Romans was tame compared to some of the Agree goings-on in this country, even in I can't make up places where people might least expect my mind it. Disagree Strongly disagree APPENDIX C INSTRUCTIONS AND RATING DIMENSIONS FOR LIKERT SCALES IN EXPERIMENT I 157 Instructions: You are asked to rate each of the following leadership behaviors on the extent to which they show participation. By participation we mean: 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 encourages subordinates to influence decisions related to their jobs and the way in which they do their jobs. Circle one of the alternatives to the right of the state- ment that best represents how much participation is shown in the statement. m o H4 H4 >Hz .c m u >.u no 0 E u up w s a o -H Q-H > s a m v4 >u4 He assigns a task, then lets me handle it. A B C D E He emphasizes that I meet deadlines. A B C D E He lets me know what is expected of me. A B C D E If I am a new member, he helps me make adjustments to the group. A B C D E He schedules the work I have to do. A B C D E He maintains definite standards of per- formance for me. A B C D E He knows what work I am responsible for. A B C D E He allows me influence equal to his own in decisions which affect my job. A B C D E He sees to it that my work is coordinated with that of other members. A B C D E He encourages me to make decisions on my own. A B C D E He makes sure that I and other group members understand his part in the group. A B C D E He helps me with my personal problems. A B C D E He is friendly to me and easily approached. A B C D E He allows me,a high degree of initiative. A B C D E 158 He encourages me to set my own performance goals. He allows me to modify the procedures required in my job. He criticizes a specific act rather than me as a person. ‘He puts suggestions made by me into operation. He criticizes poor work I do. He expresses appreciation when I do a good job. He does personal favors for me. He does little things that make it pleasant for me to be a member of the group. He has me share in making decisions. He looks out for my personal welfare. He sees to it that I have the material I need to work with. He asks for my opinion before making decisions which affect my job. He lets me do my work the.way I think best He treats me as his equal. He makes me feel at ease when talking with him. He makes his attitudes clear to.me. very much much u 8 01118 0 little U very little I?! 159 Instructions: You are asked to rate each of the following leadership behaviors on the extent to which you would be satisfied with that leadership behavior in your male supervisor. Circle one of the alternatives to the right of the state- ment that best represents how satisfied you would be with that leadership behavior in your supervisor. H >. H: .4 u >. m ‘” .2 3 u 8 s .r: m H >» m on u H a H u x m o .4 o (D > 03 (D I: He is friendly to me and easily approached. A B C D E He treats me as his equal. A B C D E He sees to it that my work is coordinated with that of other members. A B C D E He schedules the work I have to do. A B C D E He maintains definite standards of per- formance for me. A B C D E He sees to it that I have the material I need to work with. A B C D E He knows what work I am responsible for. A B C D E He allows me influence equal to his own in decisions which affect my job. A B C D E He has me share in making decisions. A B C D E He assigns a.task, then lets me handle it. A. B C D E He makes me.fee1 at ease when talking with , him.. A. B C D E He makes his attitudes clear to me. A B C D E He looks out for my personal welfare. A B. C D E He does little things that make it pleasant for me to be a member of the group. A B C D E He makes sure that I and other group members understand his part in.the group. A B C D E He helps me with my personal problems- A B C D E 160 He expresses appreciation when I do a good job. He does personal favors for me. He allows me to modify the procedures required in my job. He criticizes a specific act rather than me as a person. He puts suggestions made by me into operation. He asks for my opinion before making decisions which affect my job. He lets me do my work the way I think best. He allows me a high degree of initiative. He encourages me to make decisions on my own. He encourages me to set my own performance . goals. He lets me know what is expected of me. If Iran a new member, he helps me make adjustments.to the group. He emphasizes that I meet deadlines. He criticizes poor work I do. extremely a, > very U! somewhat O O slightly U U not at all [-13 [11 161 Instructions: You are asked to rate each of the following leadership behaviors on the extent to which they show initiating structure. By initiating structure we mean: 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. Circle one of the alternatives to the right of the state- ment that best represents how much initiating structure is shown in the statement. 0 m 5.: 4: cm: 7'4 >33 :3 g s a an: > s s m .4 >r4 He has me share in making decisions. A B C D E He assigns a task, then lets me handle it. A B C D E He encourages me to make decisions on my own. A, B C D E 7 He encourages me to set my own performance goals. A B C D E He does little things that make it.pleasant for me to be a member of the group. A B C D E He makes sure that I and other group members understand his part in the group. He helps me with my personal problems. He criticizes poor work I do. >’ >» >» >- n: c: c: a: He lets me know what is expected of me. If I am.a new member, he helps me.make adjustments to the group. A B C D E He allows me to modify the procedures required in my job. A B C D E He criticizes a specific act rather than me as a person. A B C D E He puts suggestions made by me into operation. A B C D E He asks for my opinion before making decisions which affect my job. A B C D E 162 He lets me do my work the way I think best. He allows me a high degree of initiative. He knows what work I am responsible for. He allows me influence equal to his own in decisions which affect my job. He is friendly to me and easily approached. He treats me as his equal. He looks out for my personal welfare. He sees to it that I have the material I need to work with. He makes me feel at ease when talking with him. He makes his attitudes clear to me. He expresses appreciation when I do a good job. He does personal favors for me. He schedules the work I have to do. He maintains definite standards.of per— formance for me. He emphasizes that I meet deadlines. He sees to it that my work is coordinated with that of other members. very much on much U! some 0 U little U U very little 163 Instructions: You are asked to rate each of the following leadership behaviors on the extent to which they show consideration. By consideration we mean: behaviors by which the super- visor shows warmth and rapport toward subordinates. Behaviors by which the supervisor expresses a warm, positive attitude toward and personal concern for sub- ordinates. Circle one of the alternatives to the right of the state- ment that best represents how much consideration is shown in the statement. o o >nfl .c o 73 >J3 $3 3 8 :3. 51$ >8 8 CD ---I >v-l He criticizes poor work I do. A B C D E He knows what work I am responsible for. A B C D E He allows me influence equal to his own on decisions which affect my job. A B C D E He is friendly to me and easily approached. A B C D E He allows me a high degree of initiative. A B C D E He encourages.ma to set.my own performance goals. A B C D E He does little things that make it pleasant for me to be a member of the group. A B C D E He has me share in making decisions. A B C D E He puts suggestions made by me into Operation. A. B C D E He asks for my opinions before making. decisions which affect my.job. A B C D E He lets me do my work the way I think best. A B C D E He allows me to modify the procedures required in my job. A B C D E He criticizes a specific act rather than me as a person. A B C D E He sees to it that my work is coordinated with that of other members. A B C D E 164 He lets me know what is expected of me. If I am a new member, he helps me make adjustments to the group. He schedules the work I have to do. He maintains definite standards of performance for me. He expresses appreciation when I do a good job. He does personal favors for me. He treats me as his equal. He looks out for my personal welfare. He sees to it that I have the material I need to work with. He encourages me to make decisions on my own. He makes sure that I and other group members understand his part in the group. He helps me with my personal problems. He assigns a.task, then lets me handle it. He emphasizes that I meet deadlines. He makes me feel at ease when talking to him. He makes his attitudes clear to me. very much > :> p :> > as much 0 some U little U very little APPENDIX D EXAMPLE AND INSTRUCTIONS TO SUBJECTS FOR USING SEMANTIC DIFFERENTIAL SCALES FOR RATING LEADER BEHAVIOR ITEMS IN EXPERIMENT I 165 INSTRUCTIONS The purpose of this study is to measure the meanings of certain leader- ship behaviors to various people by having them judge them against a series of descriptive scales. In taking this test, please make your judgments on the basis of what these leadership behaviors mean to you. On each page, you will find a different leadership behavior to be judged, and beneath it a set of scales. You are to rate the leader- ship behavior on each of these scales in order. Here is how you are to use the rating scales: If you feel that the leadership behavior is very closely related to one end of the scale, you should place your check-mark as follows: fair X : : : : : : : unfair or fair : : : : : : X : unfair If you feel that the leadership behavior is quite closely related to one or the other end of the scale, you should place your check~mark as follows: : weak strong : X : or strong : : : : : X : : weak If the leadership behavior seems only slightlyirelated to one side as opposed to the other side (but is not really neutral), then you should check as follows: active : : X : : : : : passive or active : : : : X :- : : passive The direction toward which you check, of course, depends upon which of the two ends of the scale seem most characteristic of the leadership behavior you are judging. If you consider the leadership behavior to be neutral on the scale, both sides of the scale equallygassociated with the leadership behavior, or 166 if the scale is completely irrelevant, unrelated to the leadership behavior, then you should place your checkdmark in the middle space: safe : : : X : : : : dangerous IMPORTANT: (1) Place your check-marks in the middle of spaces, not on the boundaries: this not this X : : : : : bX (2) Be sure you check every scale for every leadership behavior-—do not omit any. (3) Never put more than one checkdmark on a single scale. Sometimes you may feel as though you've had the same item before on the test. This will not be the case, so do not look back and forth through the leadership behaviors. Do not try to remember how you checked similar items earlier in the test. Make each item a separate and independent judgment. work at a fairly high speed through this test. Do not worry or puzzle over individual items. It is your first impres- sions, the immediate "feelings" about the leadership behaviors, that we want. 0n the other hand, please do not be careless, because we want your true impressions. If I am a new member, he weak good active don't prefer important successful cruel 167 helps me make adjustments to the group. ° strong : : : : bad : : . : : passive : : . . : : prefer : : : : : unimportant : : : : : : : unsuccessful : : : : : kind APPENDIX E EIGHT SUPERVISOR DESCRIPTIONS USED IN EXPERIMENT II, MANIPULATED TO SHOW LOW-HIGH COMBINATIONS OF CONSIDERATION, PARTICIPATIVE DECISION MAKING, AND INITIATING STRUCTURE 168 Example of Supervisor Description Showing High Consideration, He He If He He He He He He He He He He He He Low Participation, and Low Initiating Structure SUPERVISOR DESCRIPTION I is friendly and easily approached. expresses appreciation when I do a good job. I am.a new member, he helps me make adjustments to the group. makes me feel at ease when talking with him. looks out for my personal welfare. does not allow me to modify the procedures required in my job. does not have me share in decision making. assigns a task, but does not allow me to handle it. does not allow me influence equal to his own in decisions which affect my job. does not let me do my work the way I think best. does not let me know what is expected of me. does not schedule the work I have to do. does not make sure that I and other group members under- stand his part in the group. maintains no definite standards of performance for me. does not see to it that I have the material I need to work with. 169 Example of Supervisor Description Showing Low Consideration, He He He He He If He He He He He He He He He High Participation, and High Initiating Structure SUPERVISOR DESCRIPTION 2 sees to it that I have the material I need to work with. lets me know what is expected of me. schedules the work I have to do. makes sure that I and other group members understand his part in the group. maintains definite standards of performance for me. I am.a new'member, he does little to help me make adjustments to the group. is not friendly or easily approached. is not concerned with my personal welfare. expresses no appreciation when I do a good job. makes me feel uneasy when talking with him. assigns a task, then lets me handle it. allows me to modify the procedures required in my job. allows me influence equal to his own in decisions which affect my job. lets me do my work the way I think best. has me share in making decisions. 170 Example of Supervisor Description Showing Low Consideration, He If He He He He He He He He He He He He He High Participation, and Low Initiating Structure SUPERVISOR DESCRIPTION 3 expresses no appreciation when I do a good job. I am.a new member, he does little to help me make adjustments to the group. is not friendly or easily approached. is not concerned with my personal welfare. 'makes me feel uneasy when talking with him. ‘maintains no definite standards of performance for me. does not see to it that I have the material I need to work with. , does not schedule the work I have to do. does not make sure that I and other group members under- stand his part in the group. does not let me know what is expected of me. allows me to modify the procedures required on my job. lets me do my work the way I think best. allows me influence equal to his own in decisions which affect my job. has me share in making decisions. assigns a task, then lets me handle it. 171 Example of Supervisor Description Showing High Consideration, He He He He He He He He He He If He He He He High Participation, and High Initiating Structure SUPERVISOR DESCRIPTION 4 schedules the work I have to do. sees to it that I have the material I need to work with. ‘makes sure that I and other group members understand his part in the group. maintains definite standards of performance for me. lets me know what is expected of me. lets me do my work the way I think best. allows me influence equal to his own in decisions which affect my job. allows me to modify the procedures required in my job. assigns a task, then lets me handle it. has me share in making decisions. I am a new member, he helps me make adjustments to the group. 'makes me feel at ease when talking with him. looks out for my personal welfare. is friendly and eaSily approached. expresses appreciation when I do a good job. 172 Example of Supervisor Description Showing High Consideration, He He He He He He He He He He He He He If He Low Participation, and High Initiating Structure SUPERVISOR DESCRIPTION 5 does not allow me to modify the procedures required in my job. assigns a task, but does not allow me to handle it. does not allow me influence equal to his own in decisions which affect my job. does not have me share in decision making. does not let me do my work the way I think best. schedules the work I have to do. lets me know what is expected of me. makes sure that I and other group members understand his part in the group. sees to it that I have the materials I need to work with. maintains definite standards of performance for me. makes me feel at ease when talking with him. is friendly and easily approached. looks out for my personal welfare. I am.a new member, he helps me make adjustments to the group. expresses appreciation when I do a good job. 173 Example of Supervisor Description Showing Low Consideration, He He He He He He He He He He He He He He Low Participation, and Low Initiating Structure SUPERVISOR DESCRIPTION 6 maintains no definite standards of performance for me. does not schedule the work I have to do. does not see to it that I have the material I need to work with. does not make sure that I and other group members under- stand his part in the group. does not let me knOW'what is expected of me. does not let me do my work the way I think best. does not allow me to modify the procedures required in my job. assigns a task, but does not allow me to handle it. does not allow me influence equal to his own in decisions which affect my job. does not have me share in decision making. expresses no appreciation when I do a good job. is not friendly or easily approached. is not concerned with.my personal welfare. I am.a neW'member, he does little to help me make adjust- ments to the group. makes me feel uneasy when talking with him. 174 Example of Supervisor Description Showing Low Consideration, He He He He He He He He If He He He He He He Low Participation, and High Initiating Structure SUPERVISOR DESCRIPTION 7 does not let me do my work the way I think best. does not have me share in decision.making. assigns a task, but does not allow me to handle it. does not allow me to modify the procedures required in my job. does not allow me influence equal to his own in decisions which affect my job. is not concerned with my personal welfare. makes me feel uneasy when talking with him. is not friendly or easily approached. I am.a neW'member, he does little to help me make adjustments to the group. expresses no appreciation when I do a good job. lets me know what is expected of me. makes sure that I and other group members understand his part in the group. schedules the work I have to do. ‘maintains definite standards of performance for me. sees to it that I have the material I need to work with. 175 Example of Supervisor Description Showing High Consideration, He He He He He He If He He He He He He He He High Participation, and Low Initiating Structure SUPERVISOR DESCRIPTION 8 does not let me know what is expected of me. does not schedule the work I have to do. does not make sure that I and other group members under- stand his part in the group. does not see to it that I have the material I need to work with. maintains no definite standards of performance for me. is friendly and easily approached. I am a neW’member, he helps me make adjustments to the group. makes me feel at ease when talking with him. expresses appreciation when I do a good job. looks out for my personal welfare. allows me to modify the procedures required in my job. allows me influence equal to his own in decisions which affect my job. assigns a task, then lets me handle it. lets me do my work the way I think best. has me share in making decisions. APPENDIX F INSTRUCTIONS TO SUBJECTS FOR SEMANTIC DIFFERENTIAL RATINGS OF SUPERVISOR DESCRIPTIONS IN EXPERIMENT II 176 On each page, you will find a different supervisor description to be judged, and beneath it a set of scales. You are to rate the supervisor on each of these scales in order. Here is how you are to rate the scales: If you feel that the supervisor is very closely related to one end of the scale, you should place your check-mark as follows: fair X : : : : : : : unfair or fair : : : : : : X : unfair If you feel that the supervisor is quite closely related to one or the other end of the scale, you should place your check-mark as follows: strong : X : : : : : : weak or strong : : : : : X : : weak If the supervisor seems only slightly related to one side as opposed to the other side (but is not really neutral), then you should check as follows: active : : X : : : : : passive or active : : : : X : : : passive The direction toward which you check, of course, depends upon which of the two ends of the scale seem most character- istic of the supervisor you are judging. 177 If you consider the supervisor to be neutral on the scale, both sides of the scale equally associated with the super- visor, or if the scale is completely irrelevant, unrelated to the supervisor, then you should place your check-mark in the middle space: safe : : : X : : : : dangerous IMPORTANT: (1) Place your check-marks in the middle of spaces, not on the boundaries: this not this X : : : : : X (2) Be sure you check every scale for every supervisor description--do not omit any. (3) Never put more than one check-mark on a single scale. Sometimes you may feel as though you've rated the same supervisor description. 'This will not be the case, so do not look back and forth through the supervisor descriptions. Do not try to remember how you checked similar supervisors earlier in the test. Make each rating of a supervisor a separate and independent judgment. ‘Work at a fairly high speed through this test. Do not worry or puzzle over indi- vidual scales. It is your first impressions, the immediate "feelings" about the supervisor, that we want. On the other hand, please do not be careless, because we want your true impressions. APPENDIX G INSTRUCTIONS AND SCALES FOR POST-EXPERIMENTAL LIKERT RATING TASK TO TEST EXPERIMENTAL MANIPULATIONS IN EXPERIMENT II 178 You are asked to rate each of the eight supervisors described in your packet on three dimensions. Specifically, we want you to rate each supervisor on consideration, initiating structure, and participation. By consideration we mean: 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. By initiating structure we mean: 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. By participation we mean: 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 encourages subordinates to influence decisions related to their jobs and the way in which they do their jobs. Proceed through the rating task on the following pages. Feel free to refer to your supervisor description packet. 179 SUPERVISOR DESCRIPTION 1 Initiating_structure defined: behaviors by which the superv130r 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. To what extent does SUPERVISOR 1 show initiating structure? (Circle one). very little some much very little much SUPERVISOR DESCRIPTION 1 Participation defined: behaviors by which the supervisor allbws or encourages subordinates to have a say in matters related to their jobs. Behaviors by which the supervisor allows or encourages subordinates to influence decisions related to their jobs and the way in which they do their jobs. To what extent does SUPERVISOR 1 show participation? (Circle one). very little some much very little much 180 SUPERVISOR DESCRIPTION 1 Consideration defined: behaviors by which the supervisor OEhOWS‘warm H and rapport toward subordinates. Behaviors by which the supervisor expresses a warm, positive attitude toward and personal concern for subordinates. To what extent does SUPERVISOR 1 show consideration? (Circle one). very little some much very little much l'IICHIGAN STATE UNIV. LIBRRRIES 1|HIIWIHHIIW “1| H“ “1111111111111Ml HHI 31293105605251