A paper based on reading the instructions in the attachements

spyyyz77

7_danserau_et_al.pdf

Home >Human Resource Management homework help >A paper based on reading the instructions in the attachements

ORGANIZATIONAL BEHAVIOR AND HUMAN PERFORMANCE 13, 46-78 (1975)

A Vertical Dyad Linkage Approach to Leadership within Formal Organizations

A Longitudinal Investigation of the'l~ole Making Process 1

FRED DANSEREAU~ JR. 2

State University o] New York at Buf]alo

GEORGE GRAEN

University of Illinois at Champaign-Urbana

AND

WILLIAM J . HAGA

Naval Postgraduate School, Monterey, California

The use of an alternative to the research strategies employed for the past 20 yrs and more to investigate leadership produced results which question the traditional models and open new avenues for empirical exploration. Ap- proaching leadership as an exchange relationship which develops w i t h i n the vertical dyad over time during role making activities, this longitudinal study found that the degree of latitude that a superior granted to a member to negotiate his role was predictive of subsequent behavior on the part of both superior and member. Contrary to traditional views of leadership, superiors typically employed both leadership and supervision techniques within their units. With a select subset of their members, superiors developed leadership exchanges (influence without authority), and with others, superiors devel- oped only supervision relationships (influence based primarily upon authority). Some of the many implications of these findings are discussod.

' T h e authors would like to express their appreciation to the administrators and participants in this study. Also we acknowledge the research assistance of Jim Cashman, Joan Graen, Carline Haga, Ginny Ingersoll, Chip Johnson and Warren Nielsen and the reading of earlier drafts by Jim Cashman, John Kimberly, .Bill Hoel, Walter Franke, and Joseph Alutto. This research was supported in part by the National Science Foundation, the National Institute of Education, the Institute of Labor and Industrial Relations at the University of Illinois, Champaign-Urbana, and the State University of New York at Buffalo.

2 Requests for reprints should be addressed to Fred Dansereau, Jr., Department of Organization and Human Resources, School of Management, State University of New York, Buffalo, NY 14214.

46 Copyright (~) 1975 by Academic Press, Inc. All rights of reproduction in any form reserved.

LONGITUDINAL INVESTIGATION OF ROLE MAKING 4 7

After more than 20 yrs of active research, contemporary models of leadership within organizations have failed to develop beyond rather primitive levels. Recent reviews of these models testify to the under- developed state of these formulations (Campbell, Dunnette, LaMer & Weick, 1971; Graen, A1vares, Orris & Martella, 1970; Guion & GoRier, 1966; Korman, 1966; Nash, 1966; McMahon, 1972; ShifteR, 1972). These contemporary models typically include two assumptions about the leader- ship setting (see Dansereau, Cashman and Graen (1973)). The first assumption is t h a t the members of an organizational unit who report to the same superior are sufficiently homogeneous on the relevant dimensions (e.g., perceptions, interpretations and reactions) that they can be con- sidered as a single entity: the "work group." The second assumption is t h a t a superior behaves in essentially the same prescribed manner toward each of his members. Following these assumptions, research attention is focused upon the average or usual behavior of a superior toward his "work group." If these assumptions are inappropriate, this could help to explain the slow progress in the leadership area during the past 20 yrs.

One alternative approach to leadership that does not include the re- strictions of these two assumptions focuses upon the vertical dyad and the relationship between a superior and a member contained in a dyad. This approach allows for the case where each of the vertical dyadic relationships contained within a unit are radically different, as well as for the traditional case where each are essentially the same. Only by focusing upon each of the dyadic relationships can the actual distribution of vertical relationships within any organizational unit be empirically discovered. Following this alternative approach, both members of the vertical d y a d become the foci of investigation into the leadership process. Tersely stated, the vertical dyad is the appropriate unit of analysis for examining leadership processes because the vertical dyad reflects the processes linking member and superior.

Considering the traditional approach, one method of testing the appro- priateness of measuring the behavior of the superior on the average, or toward his members in general (Average Leadership Style), is to investi- gate the extent of agreement between the superior and his members as a group in describing the behavior of the superior. This was done in two studies: Evans (1970) and Graham (1970) asked both superiors and their members to describe the behavior of the superior using the conventional measures of Consideration and Structure (Fleishman, 1955). In both studies, the agreement between the description obtained from the superior and that obtained by averaging his member's response was close to zero.

One method of testing the possible advantages of the vertical dyad linkage approach (VDL) is to compare the extent to which it and the

48 DANSEREAU, GRAEN AND HAGA

t r a d i t i o n a l a p p r o a c h order e m p i r i c a l relationships. T h i s was done in a s t u d y b y D a n s e r e a u , C a s h m a n a n d G r a e n (1973). I n this s t u d y the re- search m e t h o d s p r e s c r i b e d b y b o t h a p p r o a c h e s were a p p l i e d on d a t a collected f r o m 261 m a n a g e r s a n d t h e i r superiors. A f t e r i n v e s t i g a t i n g the r e l a t i o n s h i p between leadership a n d t u r n o v e r f r o m tlle t r a d i t i o n a l (ALS) a n d V e r t i c a l D y a d L i n k a g e p e r s p e c t i v e s tile a u t h o r s concluded:

. . . this VDL approach reveals orderliness in the data that the average leadership style approach would have assumed a priori to be mainly error variance. On the other hand, the orderliness revealed by the VDL approach could not have been extracted from the data using the ALS approach (Dansereau, Cashman & Graen, 1973, p. 197).

T h i s s t u d y suggested t h a t the v e r t i c a l d y a d a p p r o a c h m i g h t possess a good deal of p o t e n t i a l for u n c o v e r i n g the " m y s t e r i e s " of leadership.

Leadership and Supervision as Techniques

I n e m p l o y i n g t h e v e r t i c a l d y a d a p p r o a c h , it is n e c e s s a r y to focus upon each m e m b e r of t h e d y a d . F o c u s i n g first u p o n the b e h a v i o r of tile su- perior, J a c o b s (1971) points out t h a t m u c h of the difficulty with the l e a d e r s h i p l i t e r a t u r e p r o b a b l y s t e m s f r o m the f a i l u r e to distinguish be- t w e e n the c h a r a c t e r i s t i c s of a superior as a p e r s o n a n d " l e a d e r s h i p " a n d " s u p e r v i s i o n " as t e c h n i q u e s which are elicited b y the situation. I n illus- t r a t i n g this p o i n t J a c o b s s t a t e s :

Yet there is a continuing tendency in the literature for reference to be made to persons as leaders because some part of their role repertories con- sists of behaviors successful in influencing others without recourse to authority ("leadership"). This obscures the fact that almost certainly there are other parts of their repertories that are based u p o n . . , authority relationships ("supervision"), in order that the complicated machinery of formal organizations might operate as efficiently as possible (Jaeobs, 1971, p. 288)

Briefly s t a t e d , J a c o b s is suggesting t h a t a superior m a y engage in role b e h a v i o r s i n v o l v i n g b o t h leadership (influence w i t h o u t a u t h o r i t y ) a n d supervision (influence b a s e d u p o n o n l y a u t h o r i t y ) . Following this insight- ful o b s e r v a t i o n to its logical consequences raises s e v e r a l questions con- eerning the conditions u n d e r which a superior can be expected to b e h a v e as a leader a n d when he can be e x p e c t e d to p l a y the role of supervisor.

One m e t h o d of seeking answers to these questions is to i n v e s t i g a t e the d e v e l o p m e n t of new v e r t i c a l d y a d linkages w i t h i n an o r g a n i z a t i o n a l context. Beginning w i t h a s u p e r i o r a n d a m e m b e r located within an o r g a n i z a t i o n a l l y new relationship, a n d m o n i t o r i n g the d e v e l o p m e n t of this v e r t i c a l r e l a t i o n s h i p o v e r t i m e m a y r e v e a l v a r i o u s conditions which con- t r i b u t e to t h e emergence of leadership a n d supervision. B e f o r e we can

LONGITUDINAL INVESTIGATION OF ROLE !VIAKING 49

proceed with a longitudinal investigation of the development of vertical dyadic linkages, we must be prepared to empirically distinguish between leadership and supervision in the dyad and during the early stages of their development in the dyad. 3

Leadership and Supervision in the Dyad

A key distinction between supervision and leadership involves the nature of the "vertical exchange" t h a t takes place between a superior and a member. Employing the supervision technique, the nature of the verti- cal exchange is such t h a t a superior relies almost exclusively upon the formal employment contract in his exchanges with a member. As a con- dition of continued employment, the member agreed to fulfill all pre- scriptions and proscriptions of the formal organization regarding his position. These role prescriptions included among others t h a t he submit to the legitimate authority of his specified superiors regarding certain specified matters. In exchange for the member's fulfillment of these con- ditions, the organization (not the superior) agreed to compensate t h e member with coin and benefits of various kinds. At the extreme, the superior can fulfill the employment contract without entering into any but the most minimal social exchange. The superior need not negotiate with the member about unit-related matters or become dependent upon him as an individual, rather he has the means to treat the member much as he would a part in a complex machine.

In contrast, employing the technique of leadership, the nature of the vertical exchange is such t h a t the superior cannot rely exclusively upon the employment contract. Instead, he must seek a different basis for influencing the behavior of a member. This alternative basis of influence is anchored in the interpersonal exchange relationship between a superior and a member. This source of influence, theoretically untapped by formal supervision, can involve highly valued outcomes not available under supervision for both the superior and the member. The superior for his

3 J a c o b s (1971) does n o t e m p l o y t h e v e r t i c a l d y a d i c a p p r o a c h . H e does suggest t h a t s u p e r v i s i o n a n d l e a d e r s h i p are n o t m u t u a l l y exclusive i n t e r m s of a superior's b e h a v i o r . B y n o t c o n s i d e r i n g the d y a d s inside of t h e u n i t h e fails t o d i r e c t l y de- scribe l e a d e r s h i p a n d s u p e r v i s i o n f r o m b o t h t h e m e m b e r ' s a n d superior's side of t h e exchange. Thus, h e did n o t logically consider w h e t h e r a m e m b e r can d e v e l o p b o t h s u p e r v i s i o n a n d l e a d e r s h i p r e l a t i o n s w i t h a superior. I n o r d e r to consider this p o i n t J a c o b s ' (1971) view was r e c o n e e p t u a l i z e d i n t e r m s of role t h e o r y . T h e view of s u p e r v i s i o n exchange d e r i v e d is s i m i l a r to K a t z a n d K a h n ' s (1966) role t a k i n g model. T h e view of l e a d e r s h i p exchange p r e s e n t e d is s i m i l a r to t h a t pre- s e n t e d b y G r a e n , D a n s e r e a u , M i n a m i , a n d C a s h m a n (1973). T h e difference b e - t w e e n this view a n d J a e o b s ' view of l e a d e r s h i p is e l a b o r a t e d u p o n in t h e Discussion section.

5 0 DANSEREAU~ GRAEN AND ttAGA

p a r t can offer the outcomes of job latitude, influence in decision making, open and honest communications, support of the member's actions, and confidence in and consideration for the member, among others. The mem- ber can reciprocate with greater than required expenditures of time and energy, the assumption of greater responsibility, and commitment to the success of the entire unit or organization, among others. The larger the extent of this vertical exchange, the more the superior must be ready to negotiate with his member on unit-related matters. The superior is inter- dependent with his member and must respect his dependence upon the particular member. The use of the supervision technique by the superior to solve problems with the member can only damage and eventually destroy the leadership exchange. Although the superior employing the leadership technique possesses the formal authority to command com- pliance from his member by virtue of his rights of office, this authority must not be used against the member. Its use rapidly transforms the basis of influence from leadership to supervision.

Leadership and Supervision in the New Dyad

When the members of the d y a d are in the initial process of organizing their roles (role making) the key outcropping of "emerging" vertical ex- changes should be the extent to which the superior allows the member to negotiate job-related matters. If a member is initially given the oppor- tunity to work through some of his job problems with his superior, the superior probably is attempting the leadership technique. However, if a member is initially allowed little or no chance to negotiate with the superior, the superior probably is employing the technique of supervision. In general, the greater the latitude initially given to the member to ne- gotiate job-related matters, the higher is the probability that the superior is attempting leadership and the lower the probability that he is using supervision with his member.

Using this distinction the purpose of the present study was to examine how early attempts to employ the techniques of leadership and super- vision contribute to the development of social exchanges in the dyad. Specifically this investigation focused upon the degree of negotiating latitude initially granted by the superior to his member, as a key variable in distinguishing between attempts at leadership and supervision and monitored the development of social exchange beginning at the stage of "near strangers" and following through to the stage of "established in- cumbents." The general hypothesis of this investigation was that nego- tiating latitude within the vertical dyad will produce leadership relations when given in generous amounts and will produce supervision relations when given in meager amounts.

LONGITUDINAL INVESTIGATION OF ROLE M A K I N G 5 1

M E T H O D

Participants and Setting

The participants in this investigation were 60 managers whose positions formed the administrative pyramid atop a housing division within a large public university. This organization had experienced a substantial reorganization of personnel at the beginning of the academic year. One- half of the management team was brought in from outside of the organi- zation to fill a vertical cross-section of positions. The reorganization was sufficiently extensive t h a t it produced approximately 90% "new" vertical dyads. A vertical dyad was considered "new" if at least one member was a newcomer to his position after the reorganization. Given the personnel policies of the university, this setting produced a stable cohort of 60 managers moving together through a 9-month "shake down" period.

Procedure

D a t a collection was divided into four interview waves: (a) The initial wave began immediately after the first month of the academic y e a r ; (b) the second wave began after the third month; (c) the third wave began after the sixth month; and (d) the final wave began after the eighth month. Each wave consisted of interviews with the 60 managers and with the superiors of these managers. Although the 60 managers in this study were superiors of other members, only the 17 managers who had members in this study were interviewed as superiors. Interviews of the 60 managers about their situation as members were conducted by one member of the research staff. Interviews with the 17 superiors about each of their members were conducted by two other staff members. For more detail on the setting, procedures, instruments and scaling methods employed, see Graen, Dansereau, Haga, and Cashman (in press) and Dansereau (1972).

Instruments

Four sets of instruments were employed to assess (a) negotiating lati- tude, (b) the superior's contribution to vertical exchange in terms of his activities toward the member, (c) the member's contribution to vertical exchange in terms of his role behavior, and (d) various outcomes of the exchange process.

Negotiating Latitude

Negotiating Latitude was defined as the extent to which a superior is willing to consider requests from a member concerning role development. At a conceptual level, at one end of the continuum a superior is unwilling

5 2 DANSEREAU, GRAEN AND HAGA

T A B L E 1 PRODUCT MOMENT CORRELATIONS OF THE NEGOTIATION LATITUDE

SCALE ACROSS TIME ( N = 6 0 )

Pearson product Time period moment correlation

Second month to fourth month Second month to seventh month Second month to ninth month Fourth month to seventh month Fourth month to ninth month Seventh month to ninth month

• 5 9 .59 .60 .73 .66 .75

to allow or help a m e m b e r to influence his role (low negotiating l a t i t u d e ) , while, a t the other end of the c o n t i n u u m a superior is willing not only to allow b u t also to help a m e m b e r to influence his role (high negotiating latitude).

T w o questions were e m p l o y e d to assess each m e m b e r ' s perception of two aspects of his o p p o r t u n i t y to negotiate on various aspects of his r01e. T h e first question asked the m e m b e r : " H o w flexible do y o u believe y o u r supervisor is a b o u t evolving changes in y o u r job a c t i v i t y struc- ture? ''* T h e second question asked the m e m b e r : "Regardless of how much formal a u t h o r i t y y o u r supervisor has built into his position, w h a t are the chances t h a t he would be personally inclined to use his power to help y o u solve problems in y o u r work? '''~ T h e correlations between these two c o m p o n e n t variables were .62, .71, .66, and .72 a t each time period, respectively. T h e responses to these two measures were s u m m e d to form a measure of the m e m b e r ' s negotiating latitude. T h e correlations across time for this scale are shown in T a b l e 1.

S u p e r i o r E x c h a n g e

I n order to measure the superior's contribution to the exchange process, some of the outcomes which could be exchanged b y the superior ( " L e a d e r - ship A t t e n t i o n " and " L e a d e r s h i p S u p p o r t " ) and s o m e of the conse- quences of these outcomes ( " D y a d i c P r o b l e m s " and "Superior Sensi- t i v i t y " ) were assessed.

*The response alternatives were (1) "He sees no need for change," (2) "He sees little need for change," (3) "He is lukewarm about change," and (4) "lie is en- thusiastic about change."

5 The response alternatives were (1) "No chance," (2) " l i e might o1' might not," (3) "He probably would," and (4) "He certainly would."

LONGITUDINAL INVESTIGATION OF ROLE MAKING 5 3

Measuring Leadership Attention

The first measure of leadership behavior, consisted of eight activities which included: (a) allowing a member to participate in decisions affect- ing him, (b) providing the member accurate information and feedback, (e) assuring the member of his confidence in him, and (d) paying atten- tion to a member's feelings and needs, among others. The members eval- uated these leadership activities in terms of (a) how much they were getting (a great deal to a little) and (b) how much they preferred to be getting (much more to much less). A superior employing the same eight activities reported how much attention each member needed to perform his role "adequately and without undue dissatisfaction" (a great deal to a little). The unit-weighted responses of the members and superiors were summed separately to form three leadership attention measures (received, preferred and needed).

Measuring Leadership Support

The second set of leadership activities concerned the superior's sup- portive behavior. The members evaluated the superior's supportive be- havior in terms of 12 activities which included: (a) standing up for the member even if it made him unpopular with others, (b) backing up the member in dealing with the administration, and (c) following through on his or her promises and the like. For each activity the members indi- cated how often they preferred the superior to engage in the activity (more often, less often). A superior, in turn, indicated how often he supported each of his members by responding to the questions: "How often do you back up what (the member's name) suggests?" and "How often do you accept and implement changes which (the member's name) suggests?" (Never . . . Always). The unit-weighted responses of the su- periors and members were summed separately to form two leadership support measures (Preference for support and Frequency of support).

Measuring Dyadic Problems and Superior Sensitivity

Two measures were employed to index the consequences of the su- perior's activities for the member. The first measure, labeled "Dyadic Problems," was developed from information gathered in the first inter- view wave. During the second wave of interviews the members and the superiors (for each of their members) indicated the severity of the fol- lowing dyadic problems: Strains in the working relationship with my superior; unsure of what my superior wants me to do in this job; lack of evaluation of my job performance and back biting within the su-

54: DANSEREAU, GRAEN AND ttAGA

perior's u n i t 2 T h e second set of questions designed to m e a s u r e t h e con- sequences of t h e superior's b e h a v i o r was called " s u p e r i o r s e n s i t i v i t y . " T h i s scale included the m e m b e r ' s responses t o t h e following five items: the e x t e n t to which t h e superior (a) u n d e r s t o o d his p ro b l em s an d needs, (b) recognized his p o t e n t i a l , (c) helped him to k n o w w h a t he was sup- posed to be doing, (d) let him k n o w w h a t was expected, an d (e) let him k n o w where he stood.

M e m b e r E x c h a n g e

In order to m e a s u r e the m e m b e r ' s c o n t r i b u t i o n to the exchange process, some of the outcomes which could be exchanged b y the m e m b e r (Role B e h a v i o r ) a n d some of the consequences of these outcomes (Role Dis- c r e p a n c y and R o l e C o n g r u e n c y ) were assessed.

M e a s u r i n g R o l e B e h a v i o r

Role b e h a v i o r was defined as the a n m u n t of time an d en erg y t h e m e m - ber invested in each of the v a r i o u s activities included in his job. T o con- s t r u e t a list of t h e activities a c t u a l l y p e r f o r m e d b y the m em b ers an " a i d e d - r e c a l l " p r o c e d u r e (Graen, 1973) was used with the m e m b e r s and t h e i r superiors d u r i n g the first interview wave. Using a c o n t e n t a n a l y t i c p r o c e d u r e on the v e r b a t i m activities g e n e r a t e d d u ri n g this first wave, a m a s t e r a c t i v i t y list of 22 items was constructed. D u r i n g each s u b s e q u e n t interview wave, a m e m b e r indicated how m u c h t i m e an d energy he was spending on each of the 22 activities ( m e m b e r n o w ). T h e superiors in t u r n e s t i m a t e d for each of t h e i r m e m b e r s s e p a r a t e l y (a) how m u ch t i m e and e n e r g y t h e m e m b e r was expending on each of t h e 22 act i v i t i es (su- perior n o w ) , and (b) how m u c h t i m e and e n e r g y t h e superior p r e f e r r e d t h a t m e m b e r to expend on each a c t i v i t y (superior o u g h t ) . ~ T o av o i d analysis of r e d u n d a n t d a t a we r e d u c e d the 22 activities b y principal com- p o n e n t f a c t o r a n a l y s i s to four f a c t o r scales. T h e fo u r f a c t o r s were: (I) Counseling activities, ( I I ) P r o g r a m m i n g activities, ( I I I ) C o m m u n i c a t i n g activities, a n d (IV) A d m i n i s t r a t i n g activities. F o r b o t h perspectives (superior and m e m b e r ) t h e u n i t - w e i g h t e d tim e an d effort scores were s u m m e d according to t h e i r f a c t o r loadings.

W i t h i n the o r g a n i z a t i o n t h r e e m a j o r groupings of roles which co n t ai n ed r a t h e r different t a s k s t r u c t u r e s could be identified. T h e first group con- sisted of H e a d Residents who were responsible for t h e o p erat i o n s within a cluster of housing units. T h e second group consisted of S u p p o r t i v e Services who were responsible for the " B r i c k s an d M o r t a r " t h r o u g h o u t

~The response alternatives were "Not a problem," "A minor problem," "Some~ what of a problem," and "A major obstacle."

7The response alternatives for each activity were (1) "none," (2) "A little," (3) "A moderate amount," and (4) "A considerable amount."

LONGITUDINAL INVESTIGATION OF ROLE MAKING 55

the organization. The third group consisted of Student Affairs who were responsible for servicing the special programs for residents throughout the organization. These three major job types showed strong relationships to the activity scales. While these relationships supported the diserim- inant validity of these factor scales, the purpose of the current study was to examine the effect of negotiating latitude across differing types of roles. Thus the job type variance was partialled out. By subtracting each subordinate's time and energy allocations from the average of indi- viduals doing the same type of work, each individual's score was adjusted to reflect his time and energy allocation relative to others doing the same type of work. s

Employing this procedure the following two measures were used in the current analysis.

Member Now (MN). A member's estimate of the amount of time and energy he spends on a particular activity relative to others doing the same type of work. A positive value indicates t h a t the member's time and energy allocation on a particular activity is above the average of those performing the same type of work, and a negative value indicates it is below the average.

Role Discrepancy ( I S N - SOl). An estimate of the extent to which the superior perceives a particular member's activities (SN) as corre- sponding to his own preferences (SO) regardless of whether the member's activities are perceived as above or below the superior's expectations (the absolute value of the difference between the superior's relative per- eeption and his relative preferences for the subordinate). A higher score indicates a larger discrepancy between what a superior reports his mem- ber doing and what he preferred that member to be doing.

A third factor analytically derived measure, Role Congruency, was employed to tap the member's perception of his ability to change his role according to his own preferences. The three questions which were summed to form this scale were the extent to which the member had (a) the opportunity to contribute as much as he wanted to his role, (b) the ability to define his role, and (e) the ability to bring about changes he wanted in his role.

Role Outcomes

The final set of instruments were designed to measure the member's attitude toward and reaction to his situation. The members' attitudes toward their situation were assessed by the Role Orientation Index (ROI).

This procedure was e m p l o y e d for each a c t i v i t y factor (counseling, etc.), for each perspective (the superior and m e m b e r ) , and for each t i m e period separately. This last step e q u a t e d t i m e and energy allocation means between t i m e points thereby partialling o u t t h e effect of differential calendar demands (Boek, 1968).

5 6 DANSEREAU~ GRAEN AND HAGA

This instrument uses an adjective check list similar in format to the Job Description Index (Smith, 1967). (See Graen, Dansereau, and Minami (1972) and Dansereau (1970) for details of the development of the ROI.) Employing the ROI, members were asked to describe their situation in terms of the following outcomes: (a) the intrinsic value of their work, (b) the psychological value of their job performance rewards (pay, pro- motions, etc.), (e) the technical competence of their superior, and (d) the interpersonal relationship with their superior. In addition overall job satisfaction was measured by the Hoppoek Job Satisfaction Blank, (Hop- pock, 1935). A member's reaction to his situation also was indicated by whether he stayed with or left the organization during the 1-yr period after the study.

Because the interviews with the superiors and the members were rather lengthy, all instruments were not employed at all time periods. Table 2 presents a summary of all the instruments employed. For each instru- ment, the respondent, method and timing is shown.

Analysis

Since the purpose of the current study was to focus upon attempts to employ leadership and supervision early in the dyads development and trace the later consequences, the member's estimate of his negotiating latitude assessed one month after reorganization served as the inde- pendent variable. By holding a vertical dyad at its assessed degree of negotiating latitude early in the process, it was possible to analyze social exchanges as they developed over the remaining 7 months. One advantage of this design is t h a t all reliable relationships with the independent variable are predictive rather than concurrent.

Examination of the distribution of scores on the negotiating latitude scale at the second month after reorganization revealed a negatively skewed distribution. This distribution suggested that the sample could be divided into two nearly equal groups with those perceiving markedly high negotiating latitude numbering 29 and those perceiving lower ne- gotiating latitude numbering 31. Those reporting the opportunity to ne- gotiate with the superior on their role behavior (high negotiating latitude) will be called initially " I N " with their superior; while those reporting less of a chance to negotiate with tlle superior will be called initially "OUT" with their superior.

In order to test the appropriateness of dividing the sample into an "IN" group and an "OUT" group, it may be assumed t h a t this I N - O U T di- chotomy capitalizes heavily on error variance. If this is the ease, accord- ing to Campbell and Stanley (1963), the high scorers (or IN group) should decline on subsequent measurements of negotiation latitude, while

LONGITUDINAL INVESTIGATION OF ROLE MAKING

T A B L E 2 DATA MATRIX

Timing months after reorganization

R e - Variable spCndent Method 2 4 7 9

Independent variable: Negotiation latitude Member I X X X X

Dependent variables: Superior exchange

Activities: Leadership attention

Received Member Q X X Preferred Member Q X X Needed Superior Q X X

Leadership support Preference for support Member Q X X X Frequency of support Superior Q X X

Consequences: Dyadic problems Member I D X Dyadic problems Superior I D X Superior's sensitivity Member I X X X

Member exchange Activities :

Member's role behavior Member I D X X X Consequences:

Role discrepancy Superior I D X X X Role congruency Member I X X X

Role outcomes Role orientation index Member Q X X (X) Overall satisfaction Member Q X X Turnover Records One year period after the study

Note: I indicates verbal interview questions; Q indicates self-administered question- naire; X indicates d a t a collected. D indicates if and when the measure was developed from within the sample.

t h e l o w s c o r e r s (or O U T g r o u p ) s h o u l d r i s e . O v e r t i m e t h e t w o g r o u p s s h o u l d r e g r e s s t o t h e s a m e p o p u l a t i o n m e a n . I n o r d e r t o e x a m i n e t h e a p p r o p r i a t e n e s s o f t h i s d i c h o t o m y , t h e g r o u p s ( I N a n d O U T ) w e r e c o n - s i d e r e d a s t h e i n d e p e n d e n t v a r i a b l e a n d t h e m e a s u r e s o f n e g o t i a t i n g l a t i t u d e a t t h e f o u r t h , s e v e n t h a n d n i n t h m o n t h w e r e c o n s i d e r e d a s t h e d e p e n d e n t v a r i a b l e in a r e p e a t e d - m e a s u r e s a n a l y s i s of v a r i a n c e d e s i g n ( W i n e r , 1972). A s s h o w n in T a b l e 3, b y e m p l o y i n g a o n e - w a y a n a l y s i s o f v a r i a n c e w i t h t w o g r o u p s a t e a c h t i m e p e r i o d ( H a y s , 1963), t h o s e w h o

T A

B L

3 M

E A

O V

E R

T IM

O F

N E

G O

T IA

T IO

N L

A T

IT U

D E

F O

R H

IG H

N D

L O

N E

G O

T IA

T IO

N L

A T

IT U

D E

G R

O U

P S

4 m

o n

th s

7 m

o n

th s

9 m

o n

th s

P ro

b a

b il

it y

O u

t In

u t

O u

t In

ro u

T im

e G

ro u

M ea

su re

ro u

g ro

u p

ro u

g ro

u p

ro u

g ro

u p

fe ct

, N

m e

S u

b o

rd in

at e'

s es

ti m

at e

6 .2

.4 *

7 .2

5 .4

.1 '

.0 0

0 4

0 1

h is

N eg

o ti

at io

n L

a ti

tu d

N o

te :

In (

H ig

h ,

2 9

) an

d O

u t

(L o

w ,

3 1

) g

ro u

p s

w er

e d

e te

rm in

e d

b y

t h

e s

u b

o rd

in a

te 's

e st

im a

te o

f h

is o

w n

n eg

o ti

at io

n l

at it

u d

e at

e se

co n

d m

o n

th a

ft er

r eo

rg an

iz at

io n

. T

h e

d e

p e

n d

e n

t v

ar ia

b le

i s

th e

su b

o rd

in a

te 's

e st

im at

io n

o f

h is

n eg

o ti

at io

n l

a ti

tu d

e a

t 4,

7 a

n d

m o

a ft

er r

eo rg

an iz

at io

n .

* p

.0 1

t t

es t

of t

h e

d if

fe re

n ce

b et

w ee

n g

ro u

p s

a t

o n

e ti

m e

p er

io d

LONGITUDINAL INVESTIGATION OF ROLE MAKING 59

scored high on n e g o t i a t i n g l a t i t u d e a t 2 m o n t h s a f t e r r e o r g a n i z a t i o n ( I N group) are significantly higher in n e g o t i a t i o n l a t i t u d e at 4, 7 an d 9 months. M o r e o v e r the significant r e p e a t e d - m e a s u r e s group effect indi- cates t h a t the I N group is significantly higher t h a n the O U T group on n e g o t i a t i o n l a t i t u d e across all time periods. I n ad d i t i o n to suggesting the a p p r o p r i a t e n e s s of dividing the sample into these two groups, this analysis d e m o n s t r a t e s the statistical design to be e m p l o y e d t h r o u g h the c u r r e n t study.

In sum, dichotomized n e g o t i a t i n g l a t i t u d e at t h e second m o n t h a f t e r the r e o r g a n i z a t i o n served as the i n d e p e n d e n t v a r i a b l e an d the r e m a i n i n g measures as tile d e p e n d e n t v a r i a b l e s for the puri)ose of analysis.

RESULTS

Unit Composition

B e f o r e considering the i m p a c t of n e g o t i a t i n g l a t i t u d e upon the develop- ment~ of d y a d i c relationships, it is helpful to examine t h e w a y in which n e g o t i a t i n g l a t i t u d e was initially perceived w i t h i n each unit. One possi- bility corresponds to the t r a d i t i o n a l assumed case t h a t n e g o t i a t i n g lati- t u d e m u s t be offered b y the superior to all m e m b e r s e q u a l l y and m u s t be perceived in the same m a n n e r b y all the mem b ers of t h e unit. I f this view is a c c u r a t e , t h e n all of the m e m b e r s within a u n i t should be either in or out with t h e superior. Only 15% of the units showed all m e m b e r s as either " i n " or " o u t " with the superior. T h e r e m a i n i n g 85% of t h e units showed a m i x t u r e of " i n " and " o u t " members. T h e significant difference between these p r o p o r t i o n s (.15 vs .85, p ~ .01) suggests t h e units in a g r e e m e n t with the more t r a d i t i o n a l view (15 %) were n o t r e p r e s e n t a t i v e of the distribution of negotiating l a t i t u d e within m o st units ( 8 5 % ) 2

T h u s the labeling of the negotiation l a t i t u d e groups as t h e "i n g r o u p " and the " o u t g r o u p " can be said to reflect n o t o n l y t h e level of n e g o t i a t - ing l a t i t u d e received b u t also the w a y in which n eg o t i at i n g l a t i t u d e was initially d i s t r i b u t e d in m o s t units. Specifically, within most units v e r y e a r l y in t h e i r development, we find an in group (those receiving high n e g o t i a t i n g latitude) and an out group (those receiving low n e g o t i a t i n g l a t i t u d e ) .

°An analysis using the average negotiation latitude in the unit as the inde- pendent variable and the same set of dependent variables as presented here, re- vealed results significantly weaker than presented here. This analysis suggested that the weaker results found using the traditional ~pproach may have been due to the research strategy of averaging the member's responses and thus losing valid variance.

6 0 DANSEREAU~ GRAEN AND ttAGA

High

i~ 27 i

~ 25

~ 24

-~ 23

22 Out group

LowT i I 7 9

Time (monlhs)

F r o . I. Leadership a t t e n t i o n received b y member.

Leadership Attention

The first aspect of the nature of vertical exchange within the dyad concerned the extent to which the members reported receiving leadership attention from the superior in terms of his allowing the member to par- tieipate in decisions affecting his work, giving the member complete and accurate information and the like. As shown in Fig. 1, the in-group mcm- bets reported receiving significantly higher leadership attention from the superior than that reported by the out-group members. The difference between the two groups on leadership attention at each time period and across all time periods was statistically reliable (p < .001). Focusing upon the member's preferences for this type of attention, as shown in Fig. 2, the out-group members reported preferring higher additional leadership attention than that reported by the in-group members. The difference between these two groups on their preferences for leadership

• ~ 26

~ 24

21 Some ] -

Oul group

J In group

I I 7 9

Time [months)

FIG. 2. Leadership a t t e n t i o n preferred b y member.

LONGITUDINAL INVESTIGATION OF ROLE I~IAKING 6 1

High

,~ 2s

2C g

, S ~' & T i m e ( mort U',S)

FIG. 3. L e a d e r s h i p a t t e n t i o n r e q u i r e d b y m e m b e r .

In group

Out group

attention was statistically reliable at each time period and across all time periods (p < .001). To summarize these two findings then, the out- group members reported receiving lower leadership attention but pre- ferred higher additional attention than the in-group members. These results suggest t h a t both groups valued leadership attention about the same; however, the in group reported receiving more of the attention than the out group.

Turning to the superior's view of leadership attention reveals the re- sults shown in Fig. 3. In contrast to the member's reports, the superior did not report that the out-group members needed as much leadership attention as the in group members. On the contrary, the superior reported t h a t his in-group members, in order to accomplish their jobs adequately and without undue dissatisfaction, required higher leadership attention than t h a t reported for the out-group members. This overall difference of the superior's report of leadership attention required by the in versus out group was marginally significant across all time periods (p < .08). This set of results suggests t h a t not only did the in group receive higher leader- ship attention but also that the superiors believed t h a t this differential attention within the unit was required for adequate unit functioning.

Leadership Support

The second set of measures of the superior's activities toward the member measured the extent to which the member preferred leadership support and the extent to which the superior reported offering support to each of his members. Analysis of the member's preferences for support produce the set of results shown in Table 4. As shown in this table, the out group reported higher preferences than t h a t reported by the in-group members. The difference between these two groups was statistically re- liable at each time period and across all time periods (p < .001). In

T A

B L

4 T

H E

M O

U N

T O

F L

E A

D E

R S

H IP

S U

P P

O R

T P

R E

F E

R R

E D

Y A

N D

IV E

N T

O I

G R

O U

A N

D O

U T

R O

U P

V a

ri a

b le

M e

a n

4 m

o n

th s

m o

n th

9 m

o n

th s

P ro

b a

b il

it ie

R e

p e

a te

d m

e a

su re

O u

t In

u t

O u

t In

ro u

T im

g ro

u p

ro u

g ro

u p

ro u

g ro

u p

ro u

e ff

e c

t e

ff e

c t

G ro

u p

m e

M e

m b

e r'

s v

ie w

: P

re fe

re n

c e

f o

r su

p p

o rt

S u

p e

ri o

r' s

v ie

w :

S u

p p

o rt

o ff

er ed

5 5

4 8

.9 *

5 .2

9 .6

5 5

4 9

.0 *

0 1

7 0

5 0

9 .0

8 .9

.7 *

6 0

8 0

* p

.0 5

t t

e st

o f

th e

d if

fe re

n ce

b e

tw e

e n

g ro

u p

s a

t o

n e

t im

e p

e ri

o d

LONGITUDINAL INVESTIGATION OF ROLE MAKING 6 3

contrast, as shown in Table 4, the superior reported giving his in-group members somewhat higher support than t h a t reportedly given to the out group. This difference between the two groups on the superior's report of the amount of support extended was marginally significant (p ~ .06). Taken together these results suggest t h a t in-group subordinates were receiving higher support from the superior and t h a t the out groups' un- fulfilled need for support was not reflected in the superior's supportive behavior.

Dyadic Problems and Supervisory Sensitivity

While the previous measures were concerned with a superior's behavior toward his members, the current set of measures are directed toward the consequences for the member of this behavior. A direct measure of these consequences is the extent to which the superior's behavior created prob- lems for the members. The results of the analysis of the superior's be- havior as a source of job problems for the members are shown in Table 5. As can be seen in this table, not only did the out-group members report the superior as more of a source of job problems, but also, the superior reported t h a t the out-group members experienced more problems with him in terms of knowing what he expects and so on. Also as shown in Table 5, compared to the out-group members, the in-group members per- ceived the superior's behavior as higher in responsiveness to their job needs. The difference between the out group and in group on their reports of the superior's sensitivity was statistically reliable at each time period and across all time periods (p < .001).

Summary of Superior Exchange Results

In terms of the nature of the superior's activities within the dyad, the in group received higher leadership attention and support than the out group. :Moreover, those initially receiving higher negotiating latitude (in group) reported less difficulty in dealing with the superior (for example, knowing what he expected) and perceived the superior's behavior as more responsive to their job needs {including the superior letting them know what he expected and where they stood with him, and helping them with their problems and role development, and the like). Thus those who initially received more latitude in negotiating their roles clearly were the favored members within the superior's unit.

Role Behavior

Turning to the member's behavior, Table 6 shows the average alloca- tion of time and energy by the in group and out group. Recall t h a t a posi- tive value indicates greater than average expenditure relative to others

T A

B L

5 P

E R

C E

IV E

D C

O N

S E

Q U

E N

C E

S O

F T

H E

U P

E R

IO R

/S B

E H

A V

IO R

O N

H E

E M

B E

R 'S

O R

K IN

G R

E L

A T

IO N

S H

V ar

ia b

M e

a n

2 m

o n

th s

m o

n th

9 m

o n

th s

P ro

b a

b il

it ie

R e

p e

a te

d m

e a

su re

O u

t In

u t

O u

t In

ro u

T im

G ro

u p

ro u

g ro

u p

ro u

g ro

u p

ro u

g ro

u p

ff e

c t

ef fe

T im

S u

p e

ri o

r' s

v ie

w :

D y

a d

ic p

ro b

le m

s 7

5 .1

M e

m b

e r'

s v

ie w

: D

y a

d ic

p ro

b le

m s

7 ,7

S u

p e

ri o

r' s

se n

si ti

v it

1 4

1 7

.3 *

3 .9

6 .8

" 1

2 .5

5 .8

• 0

0 1

" N

N A

.0 0

1 "

N A

. 0 01

0 1

. 5

9 0

* p

< :

.0 5

t te

st o

f th

e d

if fe

re n

ce b

e tw

e e

n g

ro u

p s

a t

o n

e t

im e

p e

ri o

d .

T A

B L

6 T

IM E

N D

N E

R G

A L

L O

C A

T E

A M

O N

V A

R IO

U S

C T

IV IT

IE S

A N

O U

G R

O U

O V

E R

IM E

M e

a n

4 m

o n

th s

m o

n th

9 m

o n

th s

P ro

b a

b il

it ie

R e

p e

a te

m e

a su

re s

O u

t In

u t

O u

t In

ro u

V a

ri a

b le

ro u

g ro

u p

ro u

g ro

u p

ro u

g ro

u p

ff e

c t

T im

e ff

e c

t X

ro u

C o

u n

se li

n g

.1 7

-- .

0 5

. 2

.2 2

.9 9

P la

n n

in g

.2 0

-- .

3 8

. 16

.9 7

.3 2

o m

m u

n ic

a ti

n g

1 .2

1 .4

8 *

. 5

.5 6

.6 1

" .0

.9 9

A d

m in

is te

ri n

9 1

1 "

-- .

8 7

7 *

. 8

.8 7

.0 1

.9 9

* p

.0 5

t t

e st

o f

th e

d if

fe re

n c

e b

e tw

e e

n g

ro u

p s

u t

o n

e t

im e

p e

ri o

d .

T A

B L

7 R

O L

D IS

C R

E P

A N

C Y

E T

W E

E N

H E

U P

E R

IO R

P E

R C

E P

T IO

N S

N D

R E

F E

R E

N C

E S

M e

a n

4 m

o n

th s

7 m

o n

th s

9 m

o n

th s

P ro

b ab

il it

ie s

R e

p e

a te

d m

ea su

re s

P er

fo rm

an ce

u t

O u

t In

u t

G ro

u p

is cr

ep an

g ro

u p

ro u

g ro

u p

ro u

g ro

u p

ro u

ef fe

T im

e G

ro u

ef fe

T im

C o

u n

se li

n g

.6 1

.0 7

1 .4

.9 7

.0 1

P la

n n

in g

.3 9

.5 9

.9 5

.2 3

.6 7

.1 4

.0 0

C o

m m

u n

ic at

in g

.3 5

.6 1

.1 6

.1 8

.6 4

.0 1

A d

m in

is te

ri n

2 .1

2 .0

l. 7

1 .5

1 .8

.3 5

.8 5

L O N G I T U D I N A L I N V E S T I G A T I O N O F R O L E M A K I N G 67

doing the same type of work. A negative value indicates a lower than average expenditure of effort relative to others doing the same type of work. For counseling and planning activities (Table 6) the difference between the in group and out group is not statistically reliable. By con- trast, the in-group members report expending significantly higher time and energy communicating and administering than t h a t reported by the out-group members (p < .01). Moreover, the reliable group-by-time interaction (p < .02) on communicating indicates t h a t the difference between the two groups on communicating becomes smaller over time. An additional analysis revealed t h a t the in group tended to show greater variance on these four activity factors than the out group.

Analyzing the superior's evaluation of member role behavior reveals the set of results presented in Table 7. Recall t h a t the higher the value, the larger the discrepancy between the superior's report of a member's behavior and his preferences for t h a t member's behavior. As shown in Table 7, a superior generally tended to evaluate his in-group member's role behavior as corresponding more to his preferences than the out-group member's role behavior. A conservative estimate of overall discrepancy was calculated by summing the discrepancies for all 22 activities at each time period. The difference between the in and out groups on the overall discrepancy measure is shown in Fig. 4. This difference is statistically reliable across all time periods (p < .04). These results suggest t h a t over time the superior tended to evaluate the in-group member's behavior more favorably than t h a t of the out group (despite the apparent diversity of the in-group's role behavior).

The results for the member's perception of the extent to which he per-

~ T g:

group

1~ ~'oup

4 Time (monlhs)

FIG. 4. Deviation of member's behavior from that preferred by his superior.

6 8 DANSEREAU~ GRAE:N AND HAGA

~5 In gtou p

9.5

Time [monlhsl

FIG. 5. Congruence of role with member's preferences.

ceives his role b e h a v i o r as c o n g r u e n t with his own preferences are pre- s ented in Fig. 5. As shown in this figure, the in-group m em b ers r e p o r t e d t h a t t h e i r role was more highly c o n g r u e n t with t h e i r own preferences t h a n t h a t r e p o r t e d b y the o u t - g r o u p members. T h i s difference between these two groups on role c o n g r u e n c y was s t a t i s t i c a l l y reliable across all t i m e periods (p < .01).

Summary o] Member Exchange Results

T h e s e results show t h a t the in-group memb ers spent more t i m e and e n e r g y in c o m m u n i c a t i n g and a d m i n i s t e r i n g activities t h a n the o u t - g r o u p members. M o r e o v e r , the b e h a v i o r of the in group was e v a l u a t e d as cor- responding more closely to both the m e m b e r ' s an d t h e superior's p r e f e r - ences t h a n was the b e h a v i o r of the out group. T h e s e findings suggest t h a t the w a y in which a superior initially m e d i a t e d n eg o t i at i n g l a t i t u d e h a d consequences for the m e m b e r ' s o p p o r t u n i t y t o " p e r s o n a l i z e " his job. E q u a l l y i m p o r t a n t , these findings suggest t h a t t h e superior m a y h a v e been seeking and l a t e r e v a l u a t i n g a c o n t r i b u t i o n f r o m his in-group m e m - bers t h a t is f a r more v a l u a b l e to his unit t h a n si m p l y a g reat er expendi- t u r e of effort.

Role Outcomes

T h e final a n a l y s e s were designed to assess t h e m e m b e r ' s e v a l u a t i o n of and r e a c t i o n to his job situation. In t e r m s of a m o re general s a t i s f a c t i o n m e a s u r e ( H o p p o c k J o b S a t i s f a c t i o n B l a n k ) , the in-group m e m b e r s across t h e second and seventh time periods expressed higher sat i sfact i o n with t h e i r jobs t h a n t h a t expressed by the o u t - g r o u p m em b ers (p < .03). In t e r m s of more specific outcomes, as shown in T a b l e 8, t h e i n -g ro u p m e m - bers expressed more positive a t t i t u d e s t h a n those expressed b y t h e o u t group t o w a r d the intrinsic outcomes of t h e i r wo rk (p < .01), t h e i r inter-

T A

B L

E V

A L

U A

T IO

O F

O L

O U

T C

O M

E S

T H

G R

O U

A N

O U

G R

O U

M e

a n

2 m o n t h s

7 m o n t h s

9 m o n t h s

P ro

b a

b il

it ie

R e

p e

a te

d m

e a

su re

O u

t In

u t

O u

t In

ro u

T im

V a

ri a

b le

ro u

g ro

u p

ro u

g ro

u p

ro u

g ro

u p

ff e

c t

e ff

e c

t G

ro u

T im

In tr

in si

c o

u tc

o m

e s

1 6

1 7

.3 *

5 .4

7 .0

1 5

1 6

.6 *

f w

o rk

S u

p e

rv is

o r:

I n

te r-

9 .3

2 .9

1 8

2 2

.4 *

7 .6

1 .5

p e

rs o

n a

l re

la ti

o n

S u

p e

rv is

o r:

T e

c h

n i-

1 0

.3 *

9 .3

8 .0

.4 *

l c

o m

p e

te n

c e

R e

w a

rd s:

P sy

c h

o -

2 0

2 1

1 7

2 1

.8 *

g ic

al v

a lu

e s

•0 1

• 0

0 1

.0 1

• 0

0 1

• 0

0 1

• 0

0 1

• 0

0 1

• 0 3

• 3 7

• 9 2

• 6 6

• 0 3

* p

.0 5

t t

e st

o f

th e

d if

fe re

n c

e b

e tw

e e

n g

ro u

p s

a t

o n

e t

im e

p e

ri o

d •

70 DANSEREAU, GRAEN AND HAGA

personal relations with the.supervisor (p < .001), their supervisor's tech- nical competence (p < .01), and the value of their job performance re- wards (p < .001). As shown in this table, the differences between the two groups on these attitudinal variables are significant at each time period except for the assessment of the psychological value of job performance rewards at the second month. The significant group × time interaction effect (p < .03) on the psychological value of rewards suggests t h a t while the in group evaluated their rewards more favorably across time than the out grdup, this difference became more pronounced over time. For the in group the mean of the rewards received remained stable (21.7 to 21.8), but for the out-group members the mean of the job performance rewards decreased over time from 20.2 to 17.8. I t may be of interest to note t h a t one reason this role outcome was not measured at the ninth month was the strong objection by some to the relevance of these rewards in their situation. I t seems plausible t h a t these objections m a y have come from the out-group members. Indeed, 55% of the out-group members left the organization during the 1-yr period after the study, and only 34% of the in-group members left. This set of results suggests that these two groups evaluated and reacted to the outcomes of their respective situa- tions very differently.

DISCUSSION

The results of this longitudinal investigation of the managerial role making process, taken as a whole, produce a rather elaborate scenario of the differentiation of the organizational unit. Contrary to the views of most organizational scientists, organizational units were shown to be differentiated based upon the relationships between the superior and each of his members. Quite soon after the reorganized unit began to function, the differences in the developing vertical dyad linkages were identified in terms of the members initial ability to negotiate with their superior. These initial differences foreshadowed the differentiation of the units into the cadre and the hired hands.

Superior Exchange Over time, the cadre became the objects of the superior's concern and

attention as compared to the hired hands. Compared to the hired hands, the cadre reported receiving higher amounts of information, influence, confidence, and concern from the superior (leadership attention). T h a t these were valued outcomes for even the hired hands was demonstrated by the higher preferences for these outcomes reported by them compared to the cadre. The superiors for their part showed t h a t this differential treatment was not merely a fantasy of the members by reporting t h a t the

L O N G I T U D I N A L I N V E S T I G A T I O N O F R O L E M A K I N G 71

cadre actually required higher levels of these supervisory mediated out- comes than the hired hands to perform their jobs adequately and without undue dissatisfaction. Based upon these reports from the superiors, these outcomes appear to function at least partly as inputs to adequate per- formance in the eyes of the superiors. This suggests t h a t the differential treatment of members by superiors may be instrmnental to adequate unit functioning. The question t h a t these results raise is how can differ- ential treatment aid unit functioning? The answer quickly suggested by differential psychology is based upon merit: Reward the more dependable with a more attractive package of outcomes than that given to the less dependable. But how is dependability defined by superiors?

Member Exchange

The evidence clearly indicates that the cadre members were seen by their superiors as more dependable than the hired hands. The superiors reported that the cadre were expending time and energy more closely to that preferred by them than t h a t reported for the hired hands. On the members' part, the cadre reported higher involvement in administering and communicating activities than did the hired hands, but the cadre was also seen as doing more what the superior wanted them to do. Even though the cadre reported higher levels of latitude in performing their functions (job enrichment?) apparently this latitude was not abused but exercised dependably.

Another aspect of these differential relationships was the support the superior gave to his members. The exercise of latitude in performing one's job is closely dependent upon the support extended by the superior. With- out a good deal of confidence t h a t one's superior will support innovative approaches to problems and tasks on the job, the assumption of such risk as an exercise of latitude is not likely to occur. In this study, the su- periors reported giving higher support for the actions of the cadre than that reported for the actions of the hired hands, even though the hired hands reported a higher preference for support than t h a t reported by the cadre.

Consequences of Vertical Exchanges

Some of the costs of unit differentiation appear in the reported severity of job problems and in the reported job attitudes of the members. Both the members and the superiors reported that the hired hands experienced higher difficulty with the superior's behavior than that reported for the cadre. Judging from the sensitivity of the superior reported by the mem- bers, this set of problems, rather than being resolved somewhat over time, remained more severe for the hired hands as opposed to the cadre.

72 DANSEREAU, GRAEN AND HAGA

Finally, the attitudes and reactions (turnover) of the members clearly reflected the reward value of the differential t r e a t m e n t over time.

Superior-Member Exchange

Based upon these results, the fact of unit differentiation is indeed diffi- cult to deny. However, once we have accepted the inescapable fact of unit differentiation, we must consider its implications for our understanding of how superiors react to their situation (Fig. 6).

One view of the differentiation process is t h a t the superior of a new or reorganized unit finds himself in a problematic situation. His most valu- able resource, his own time and energy, is often inadequate to perform all of the functions for which he is held primarily responsible in the eyes of his superiors. He simply cannot devote the required time and energy to each and all of his members to ensure their optimum performance. For- tunately, a subset of his members usually can perform the majority of the critical functions of the unit. Therefore, the superior invests a dispro- portionate amount of his time and energy in developing a select subset of his members. Once these members shoulder their share of the burden of the unit, the superior maintains his disproportionate attention to these members and their critical tasks (leadership relationship).

But what about the unselected subset of members who receive only the residual amount of time and energy of the superior? What happens to these almost forgotten members in the rush to develop and maintain a cadre of trusted assistants? How does the superior influence these im- poverished ordinary members? The one concept t h a t describes the rela- tionship of the superiors to these hired hands is "supervision," influence by direct appeal to the formal authority of his office.

Commitment and Leadership

This view of the organizational unit as becoming differentiated based upon the superior's behavior into leadership relations with some mem- bers and supervision relations with other members is a plausible alter- native to the traditional view of leadership. In addition, the findings of the present study directly contradict the traditional view t h a t while superiors as individuals may v a r y in their ability to do so, they must

FIG. 6. Traditional unit becomes differentiated during role making.

L O N G I T U D I N A L I N V E S T I G A T I O N OF ROLE M A K I N G 7 3

r e l y solely u p o n t h e i r own f o r m a l a u t h o r i t y to ensure a d e q u a t e u n i t func- tioning. *° T h i s t r a d i t i o n a l v i e w a p p a r e n t l y is b a s e d u p o n the a s s m n p t i o n t h a t t h e m e m b e r ' s roles are such t h a t p r o p e r u n i t f u n c t i o n i n g will flow d i r e c t l y f r o m a m e m b e r ' s c o m p l i a n c e w i t h some l e g i t i m a t e l y p r e s c r i b e d role. H o w e v e r , our results suggest t h a t once these roles a r e filled w i t h real people, c o m p l i a n c e with a p r e s c r i b e d role m a y n o t be enough t o ensure the a d e q u a t e f u l f i l l m e n t of all of the u n i t ' s functions. I n addition, t h e a d e q u a t e fulfillment of all of t h e u n i t ' s functions m a y r e q u i r e t h e p e r s o n a l c o m m i t m e n t of a few k e y m e m b e r s to the success of t h e u n i t ' s mission. I t seems l i k e l y t h a t faced w i t h such a s i t u a t i o n a s u p e r i o r will s e e k to d e v e l o p such c o m m i t m e n t in some of his m e m b e r s .

T o d e v e l o p this c o m m i t m e n t , t h e superior c a n n o t e m p l o y his f o r m a l a u t h o r i t y since c o m m i t m e n t c a n o n l y be v o l u n t e e r e d b y t h e m e m b e r s . T h u s in t h e hope t h a t t h e m e m b e r will develop " o w n e r s h i p " feelings t o - w a r d the unit a n d w o r k for the success of the unit, t h e s u p e r i o r m a y offer t h e m e m b e r wider l a t i t u d e in his role, m o r e l e a d e r s h i p a t t e n t i o n , g r e a t e r s u p p o r t , a n d l a r g e r influence on t h e u n i t ' s o p e r a t i o n . I n sum, t h e s u p e r i o r m a y e n c o u r a g e t h e m e m b e r to a c t u a l l y n e g o t i a t e his duties a n d r e l a t i o n s h i p s w i t h i n the u n i t in t h e hope t h a t the m e m b e r will i n v e s t some of h i m s e l f in t h e role a n d a s s u m e some r e s p o n s i b i l i t y for t h e u n i t ' s success. As a result, e v e n t s affecting the u n i t m a y be p e r c e i v e d b y the m e m b e r as iiot o n l y h a v i n g consequences for the s u p e r i o r b u t also as h a v i n g consequences f o r t h e c o m m i t t e d m e m b e r s . T h e results of this s t u d y suggest t h a t b y allowing a m e m b e r to n e g o t i a t e his role t h e su- p e r i o r m a y g a i n critical a s s i s t a n c e f r o m m e m b e r s c o m m i t t e d to t h e u n i t ' s success. I n this m a n n e r the superior c a n i m p r o v e the s u r v i v a l p o t e n t i a l of his a d m i n i s t r a t i o n .

C o n s i d e r i n g this c o m m i t m e n t p h e n o m e n o n suggests t h a t b e c o m i n g in- v o l v e d in this process restricts the b e h a v i o r of the superior. H i s b e h a v i o r t o w a r d his p o t e n t i a l t r u s t e d a s s i s t a n t s cannot, even in crisis situations, d e g e n e r a t e into influence b a s e d solely u p o n t h e l e g i t i m a t e a u t h o r i t y of the office. T h i s q u i c k a n d effective c o n f o r m i t y m e c h a n i s m is a p t to de- s t r o y t h e m e m b e r ' s c o m m i t m e n t to the u n i t ' s success. M o r e o v e r , t h e s u p e r i o r m u s t c o n s i s t e n t l y s u p p o r t the exercise of job l a t i t u d e t h a t results f r o m the n e g o t i a t i o n process as well as e n c o u r a g i n g the selected m e r e -

lOThis is basically the supervision technique. This is not to deny people may vary in their ability to use supervision. Indeed Jacobs (1971, p. 316) hypothesizes several methods or styles for reducing the cost of supervision, for example, avoiding punitive methods in supervising. Nor is this to deny that a superior may be in a position where he must use supervision and only requires compliance from the members. However, as Jacobs points out, an individual's ability to use the super- vision technique is not leadership.

74 DANSEREAUt GRAEN AND ttAGA

bers to take action. This is especially the case when such agreements between the superior and member are unlikely to be formally sanctioned. If the superior fails to support the member's exercise of negotiated lati- tude, the negotiated promise is empty and commitment would probably not occur. This suggests t h a t the superior would be likely to offer ne- gotiating latitude to members he believes can be trusted to accept some responsibility for the success of the unit and do the right thing or, more simply, members he can depend upon. Within most units it seems un- likely t h a t a superior would typically perceive all of his members as " r e a d y " for the substantial amount of latitude and support of their actions which could be involved in this negotiation process.

Supervision and Leadership as Processes

This process of allowing some members the opportunity to negotiate their roles goes beyond Jacobs' (1971) view of leadership as involving "an interaction between persons in which one presents information of a sort and in such a manner t h a t the other becomes convinced t h a t his outcomes (benefits/costs ratio) will be improved if he behaves in the manner suggested." The results of the current study suggest t h a t leader- ship within formal organizations involves the active exchange of leader- ship inputs and outcomes. During the negotiation process, the trusted assistants of the leader share in both the leader's resources (superior exchange) and the leader's responsibilities (member exchange). This points to an exchange process in which the superior assists the member in developing a "vested interest" in desirable role behaviors. In a real sense these members' own outcomes are improved or maintained by en- gaging in behaviors which are desirable from the supervisor's point of view. For example in the current study the role behavior of the trusted assistants was in line not only with superior's preferences but also with the members' own role preferences.

Over time as the trusted members become trusted assistants, the su- perior may become more dependent upon them not only to assure ade- quate unit functioning but also to deal with problems that may arise within the unit. On the other side the trusted assistants may become more dependent upon the superior for the rewards of their contributions (e.g., visibility in the organization, and eventual promotions). As this inter- dependence (exchange) develops, a relatively tight clique may form between the superior and his trusted assistants and preclude free move- ment into or out of the clique over time. Additionally, the vast amounts of time invested by the superior in the trusted assistants during the negotiation process may also discourage the superior from replacing them.

LONGITUDINAL INVESTIGATION OF ROLE MAKING 7D

I n sum, these results suggest t h a t units become d i fferen t i at ed into t r u s t e d assistants and o r d i n a r y members. T h e t r u s t e d assistants accept a d d i t i o n a l responsibili~ties and duties and are c o m p e n s a t e d b y g r e a t e r l a t i t u d e in doing t h e i r job and by special t r e a t m e n t fro m t h e superior. In this w a y the superior m a y invest his most v a l u e d resource, his own t i m e and energy, most efficiently to assure a d e q u a t e u n i t functioning. As a consequence, the r e l a t i o n s h i p between t r u s t e d assistants an d t h e superior becomes c h a r a c t e r i z e d b y leadership relations, while t h e rel at i o n sh i p be- tween the o r d i n a r y m e m b e r s and the superior r e m a i n c h a r a c t e r i z e d b y supervision relations.

The Use of Leadership and Supervision

In terms of the leadership process, these results suggest t h a t a superior m a y allow at least some of his members to n e g o t i a t e t h ei r roles with him, in order to produce new u n d e r s t a n d i n g s a b o u t t h ei r respective role be- haviors. 11 In a f u n c t i o n a l l y i n t e r d e p e n d e n t system, t h e superior is de- p e n d e n t upon the b e h a v i o r of his m e m b e r s or a t least some of his m e m - bers, and the m e m b e r s are d e p e n d e n t upon the b e h a v i o r of the superior: m u t u a l dependence. M o r e o v e r , the superior can be expected to m a k e de- m a n d s upon his members a n d t h e y upon him. One significant aspect a b o u t the o r g a n i z a t i o n a l s i t u a t i o n is t h a t the superior al m o st a l w a y s has f o r m a l a u t h o r i t y o v e r his members. T h e k e y distinction between leader- ship and supervision in the o r g a n i z a t i o n is the w a y in which a superior uses his f o r m a l a u t h o r i t y . In supervision, a superior uses his a u t h o r i t y to force the m e m b e r to comply with some prescribed role. T h e cost of this to the superior p r o b a b l y involves the t i m e n e c e s s a r y to initiate con- trol systems and to m o n i t o r r e w a r d and punish a p p r o p r i a t e b e h a v i o r . I n leadership, a superior m a y share his a u t h o r i t y to gain c o m m i t m e n t . I n the c u r r e n t s t u d y the leader b y using w h a t e v e r power he h a d to enrich the roles of his t r u s t e d assistants, m a y h a v e won the c o m m i t m e n t of these members, t h e r e b y e n h a n c i n g his control o v e r t h e unit. One cosl~ to the superior p r o b a b l y involves the time neces sary to allow, p r o m o t e and s u p p o r t the m e m b e r in developing his role g r a d u a l l y over time.

U n d e r c e r t a i n conditions the use of the leadership t e c h n i q u e can be more costly t h a n the use of supervision. F o r example a n u m b e r of " c o s t " f a c t o r s m a y be involved in a superior's e s t i m a t i o n of a m e m b e r ' s de- p e n d a b i l i t y or " r e a d i n e s s " for the negotiatio n (leadership) process. A p a r t i c u l a r superior m a y be personally inclined to select members who are

J1 This suggests that superiors as individuals can also vary in terms of their ability to use the leadership technique. For example Jaeobs (197l) suggests that individuals may vaD" in their ability to effectively use and learn social exchange skills.

76 DANSEREAU, GRAEN AND HAGA

c o m p a t i b l e with him in t e r m s of technical competence an d i n t erp erso n al skills. I n some cases the selection m a y be based upon t h e preiudices (style) of the superior or prior experience with t h e member. I n addition, if a superior feels sufficiently t h r e a t e n e d b y the technical or i n t erp erso n al competence or incompetence of all of his members, the superior m a y n o t use the leadership technique. Likewise, if a superior is g ran t ed little or no l a t i t u d e b y his superior or if the role system is a complete program, the superior's a b i l i t y to negotiate with his memb ers m a y be r e s t r i c t e d - a n d the leadership process m a y be unlikely to unfold. M u c h more research is needed to d e t e r m i n e the b o u n d a r y conditions u n d e r which a superior can be expected to use the leadership technique during the role m a k i n g process.

CONCLUSION

I n conclusion, the results of this investigatio n i m p l y t h a t c o n t r a r y t o most c o n t e m p o r a r y models of leadership a given superior can establish " l e a d e r s h i p " relationships with some of his m em b ers an d at the same t i m e establish " s u p e r v i s i o n " relationships with o t h er of his members. F u r t h e r m o r e , this differentiation of o r g a n i z a t i o n a l units o v er t i m e based upon the a p p a r e n t n e g o t i a t i o n of the m e m b e r s ' roles opens new an d u n e x p l o r e d avenues for research of leadership in fo rm al organizations. No longer need researchers examining leadership b e h a v i o r be c o n t e n t to search only for the b e h a v i o r of the leader t o w a r d his m e m b e r s in general on the average, b u t instead t h e y might prob e t h e exchange processes between t h e leader and each of his members. R a t h e r t h a n simply s e e k i n g the o p t i m u m balance between concern for people and concern for p ro - duction, the researcher can search for the effective m ean s of i d e n t i f y i n g and developing the a p p r o p r i a t e mix of m e m b e r s req u i red for effective u n i t p e r f o r m a n c e on a n u m b e r of dimensions, while a t t h e same t i m e minimizing the negative consequences of this process. Although t h e m a n - a g e m e n t of an o r g a n i z a t i o n a l unit clearly involves dealing with the entire set of members, leadership can only occur in t h e v e r t i c a l d y a d . Lo n g enough we h a v e been fixated on the entire set of members. N o w it is t i m e to examine the various v e r t i c a l d y a d s an d t h ei r d e v e l o p m e n t over time.

R E F E R E N C E S

BOOK, R. D., & HAGGARD, E. A. The use of multivariate anal3~is of variance in be- havioral research. I n D. K . Whitla (Ed.), Handbook o] measurement ~nd assess- m e n t in behavioral sciences. Reading, MA: Addison-Wesley, 1968.

CAMPBELL, D., • STANLEY, J. E x p e r i m e n t a l and quasiexperimentaI design ]or re- search. Chicago: Rand McNaUy, 1966.

LONGITUDINAL INVESTIGATION OF ROLE MAKING 77

CAMPBELL, J. P., DU~ETTE, M. D., LAWLI~I~, E. E., & WEIC~, K. E., JR. Managerial behavior, performance and effectiveness. New York: McGraw-Hill, 1970.

DANSEREAU, F. The invisible organization. Unpublished Ph.D. dissertation, Univer- sity of Illinois, 1972.

DANSEREAU, F., JR. Some correlates of supervisory behavior. Unpublished M.A. Thesis, University of Illinois, 1970.

DANSEREAU, F., CASHMAN', J., & GRAEN, G. Instrumentality theory and equity theory as complementary approaches in predicting the relationship of leader- ship and turnover among managers. Organizational behavior and human pe~'- formance, 1973, 10, 184-200.

EvANs, M. G. The effects of supervisory behavior on the path-goal relationship. Organizational behavior and human perjormance, 1970, 5, 277-298.

~q~LEISHMAN, E. A., HARRIS, E. F., & BURTT, H. E. Leadership and supervision in industry. Columbus: Bureau of Educational Research, Ohio State University, 1955.

GRAHAM, W. A. Perceptions of leader behavior and evaluation of leaders across or- ganizational levels. Experimental publication system (APA), 1970, 4, 144A.

GRAEN, G. Role making processes within complex organizations. I n M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology, in press.

GRAEN, G., ALVARES, K., ORRIS, J. B., & MARTELLA, J. Contingency model of leader- ship effectiveness: Antecedent and evidential results. Psychological b~dIetin, 1970. 74, 285--296.

GRABS, G., DANSEREAU, F , HAGA, W., & CASHMAN, J. The invisible organization. Boston, MA: Schenkman, in press.

GRAEN, G., DANSEREAU, F., JR., & MINAMI, T. Dysfunctional leadership styles. Or- ganizational behavior and human performance, 1972, 7, 216-236.

GRAEN, G., DANSEREAU, F., MINAMI, T., & CASHMAN, J. Leadership behaviors as cues to performance evaluation. Academy of management journal, 1973, 16, 611-623.

GuIoN, R., & GOTTIER, R. Validity of personality measures in personnel selection. Personnel psychology, 1965, 18, 135-164.

HARMON, H. H. Modern factor analysis. Chicago: University of Chicago Press, 1967. HAYS, W. L. Statistics ]or psychologists. New York: Holt, Rinehart & Winston,

1963. HoPPOCK, R. Job satisfaction. New York: Harper and Brothers, 1935. JAcoss, T. Leadership and exchange in formal organizations. Alexandria, VA: Human

Resources Research Organization, 1971. /(,~wz, D., ~z KAHN, R. The social psychology of organizations. New York: Wiley,

1966. :KORMAN, A. K. "Consideration," "initiating structure" and "organizational criteria:':

A review. Personnel psychology, 1966, 19, 349-361. McMAHoN, T. The contingency theory: Logic and method revisited. Personnel psy-

chology, 1972, 25, 697-709. NAsH, A. Development of an SIVB key for selecting managers. Journal of applied

psychology, 1966, 50, 250-254. SHIFLETT, S. The contingency model: Some implications of its statistical and meth-

odological properties. Report No. 996, US Army Medical Research Laboratory, F o r t Knox, KY, 1972.

SMITH, P. C. The development of a method of measuring job satisfaction: The

78 DANSEREAU~ GRAEN AND HAGA

Cornell studies. I n E. H. Fleishman (Ed.), Studies in personnel and industrial psychology. Homewood, I L : Dorsey, 1967.

STOGDILL, R. M., & CooNs, A. E. Leader behavior: Its description and measurement. Columbus: Ohio State University, Bureau of Business Research, 1957.

VROOM, V., & MANN, F. Leader authoritarianism and employee attitudes. Personnel psychology, 1960, 13, 125-140.

WINEa, B. J. Statistical principles in experimental design. New York: McGraw- Hill, 1972.

RECEIVED: O c t o b e r 3, 1973