Description
Study on Intra-organizational Influence Tactics: Explorations in Getting One's Way
Study on Intra organizational Influence Tactics: Explorations in Getting One's Way
The tactics used by people at work to influence their superiors, co-workers, and subordinates were investigated in two studies. In the first study, 165 lower-level managers wrote essays describing an incident in which they influenced either their bosses, co-workers, or subordinates. Through content analysis, a total of 370 influence tactics grouped into 14 categories were identified. The tactics ranged from the use of rational discussion through the use of exchange tactics to the use of clandestine tactics. In a second study, the 370 influence tactics were rewritten into a 58-item questionnaire. New respondents described the extent to which they used each item to influence their bosses (n = 225), co-workers (n = 285), or subordinates (n = 244). Based on afactor analysis of the questionnaire, eight dimensions of influence were found: assertiveness, ingratiation, rationality, sanctions, exchange, upward appeals, blocking, and coalitions. It was found that the frequency with which each influence dimension was used related to the relative power of the respondents and their targets of influence, the reasons for exercising influence, the resistance of the target person, the organizational status of the respondents, organizational size, and whether the organization was unionized. Sex of the respondents and sex of the respondents' bosses, however, were not related to the choice of influence tactics in the present study.
Organizational psychologists have not been particularly interested in studying the ways in which people at work influencetheir colleagues and superiors to obtain personal benefits or to satisfy organizational goals, For the most part, interest has centered on the ways subordinates can be influenced to improve subordinate productivity and morale. This latter use of influence is customarily called the study of leadership, whereas the former can be called the study of organizational politics. A consequence of this focus on leadership is that there is little systematic information available about how people use power to influence their colleagues or superiors. With
We wish to thank Sheldon Zalkind for gathering data attheBernard Baruch Graduate Center Requests for reprints should be sent to Stuart M.
but few exceptions (e.g., Izraeli, 1975; Schein, 1977), our thinking about this topic is guided by anecdotal evidence or armchair speculations that have been organized into rational classifications of power tactics (e.g., Etzioni, 1968, pp. 94- 1Q9; French & Raven, 1959). One problem with these classifications of power tactics, as Raven (1974) has pointed out, is that they overlap with each other, though each varies in the number of influence dimensions that are described. A further problem with existing classifications of power tactics is that when influence acts are actually studied, it is found that people do not exercise influence in ways predicted by rational classification schemes. This point was first explicitly made in a study by Goodchild, Quadrado, and Raven (Note
1) Jn which college Students wrote brief the t ; 1>H T t ' . .. , •a f „ Jt
Schmidt, Department of Industrial Relations and was found that many of the influence tactics Organizational Behavior, Temple University, Phila- described by these students could not be delphia, Pennsylvania 19122. classified into preexisting categories. Sev440
•
INFLUENCE TACTICS
441
eral tactics thought to be basic when classifying influence, such as the use of expert power, were not even mentioned by the students. There is a need, therefore, for empirical studies of the use of influence within organizations. The purpose of this article is to report two studies that sought to examine the tactics of influence used by people at work when attempting to change the behavior of their superiors, co-workers, and subordinates. In the first study, the range of tactics that people use at work was identified by applying content analysis to written descriptions by managers of their attempts to influence their bosses, co-workers, or subordinates. The second study sought to identify through factor analysis the dimensions of influence underlying the specific tactics that were uncovered in the first study. Study 1: Determining Intraorganizational Influence Tactics Method
The data for this study were collected from 165respondents, 25% women and 75% men. All respondents were taking graduate business courses part-time, in the evening. They were told to refer to their current employment experience when supplying the requested information. The respondents were employed mainly in managerial roles as engineers, technicians, and professionals. Since most respondents came from different organizations, there were almost as many organizations as respondents represented. Respondents were asked to describe an incident in which they actually succeeded in getting either their boss, a co-worker, or a subordinate to do something they wanted; 62 described how they got their way with their boss, 49 with a co-worker, and 54 with a sub-
ordinate. The reason for the unequal distribution of respondents over status levels was that fewer respondents had subordinates or co-workers than had bosses. In their description of an incident, each respondent wrote in essay form what they wanted from a target person, what they did, whether there was resistance from the target, and what further influence tactics were used in response to resistance from the target. Each respondent also completed a structured questionnaire containing demographic and specific organizational job-situation questions. The incidents were sorted first in terms of the goal sought from the target person. This sorting yielded fivegeneral categories of goals: assistance with own job —obtaining the assistance of the target in helping the respondent do his or herjob, when it was not part of the target's legitimate job duties;get others to do theirjob— getting the target to do his or her own work; obtain benefits—goals that personally benefited the respondent, such as salary increase, promotion, and improved work schedule; initiate change—initiating new organizational programs and systems or improving the coordination of organizational activities (e.g., changing a scheduling procedure); and improve performance— improving the target's on-the-job performance. These goals varied as a function of the target's job status, as shown in Table 1. Respondents sought mostly self-interest goals from their superiors. The primary goal sought from co-workers was to get assistance with the respondent's own job. The most prevalent reason for influencing subordinates was to get them to do their jobs. Finally, the goal of initiating change was sought both from bosses and from subordinates in almost equal proportion. However, the changes sought from superiors focused onjob-related organizational changes, such as launching a new accounting procedure or starting a special project. But with subordinates, the changes sought dealt with job performance, such as changes in the way ajob should be done or the manner of working in the organization. Next, the influence tactics reported by the respondents in attempting to achieve these goals were identified. A total of 370influence tactics were reported by the 165 respondents. The authors sorted these tactics into 14 categories (see Table 2). Consensus among the coders was used as the criterion for assigning a given tactic to a
Table 1 Reasons for Exercising Influence, by Target Status (Study 1) in Percentages
Target status Reasons Obtain assistance on own job Get others to do their jobs Obtain personal benefits Initiate change in work Improve target's job performance Note. Numbers in parentheses are Afs. Boss (62)
3 13 58 26 0
Co-worker (49)
48 23 10 15 4
Subordinate (54)
9 46 0 28 17
Total (165)
18 27 25 23 7
442
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON In addition to identifying influence tactics, the analyses of Study 1 focused on the correlates of the 14 categories of influence. To this end, a unit weight of 1 was assigned if the respondent reported using any of the items comprising a given influence tactic category; a score of 0 was assigned if none of the items comprising a category were mentioned. Thus, each respondent had 14 scores of 1 or 0. Next, multivariate analyses of variance, which if significant werefollowed by univariate analyses, were carried out to examine the relationship among the 14 categories of influence, the status of the target person, and the goals sought by respondents in exercising influence. The influence tactics used by the respondents varied with the goal sought from the target person. When the goal was self-interest, the most frequently reported tactics were self-presentationand personal negative actions; when the goal was to initiate change, the most
category. The 14 categories ranged from the use of administrative sanctions and personal threats through the use of logic and rational discussions to clandestine, dependency appeal, and ingratiating tactics. The individual tactics illustrating these 14 categories are also shown in Table 2. To determine the reliability of the assignment of items to these 14 categories, three coders who were not associated with the research independently sorted the 370 influence tactics into the 14 categories. Raters 1 and 2 agreed on the placement of items 61% of the time; Raters 1and 3 agreed 64% of the time; and Raters 2 and 3 agreed 65% of the time. Since there were 14 categories into which any item could be placed, the degree of agreement between the three raters suggests modestly high reliability for the classification scheme. The areas of disagreement between the raters involved primarily the three categories "weak ask," "explain," and "request."
Table 2 Classification of Influence Tactics by Category (Study 1)
Category/tactic Clandestine Challenged the ability of the target Lied to the target Acted in a pseudo-democratic manner Puffed up the importance of the job Manipulated information Made the target feel important Cajoled the target Showed understanding (pretended) of the target's problem Personal negative actions Fait accompli/went ahead on own Chastised the target Became a nuisance Slowed down on the job Held personal confrontation with target Threatened withdrawal of help Expressed anger Threatened to leave job Blocked target's actions Ignored target Administrative negative actions Filed a report with supervisor Sent target to superior for conference Gave unsatisfactory performance evaluations Gave no salary increase Threatened with unsatisfactory performance ratings Threatened job security Threatened loss of promotion Exchange Contributed in exchange for compliance Compromised Offered to make sacrifice Offered help to get the job done Invoked past favors Category/tactic Persistance Repeated reminders Argued Repeated previous actions Surveillance Training Explained how it was to be done Showed how to do it Reward Verbal reinforcement Salary raise Gave benefits Self presentation Demonstrated competence Performed well, then asked Waited until target was in the right mood Was humble Was friendly Direct request Weak ask Showed dependency Weak request Demand Invoked rules Ordered Convened formal conference Set time deadline Told target that it must be done as I said or better proposed Explained rationale for request Gathered supporting data Coalitions Obtained support from co-workers Obtained support informally from superiors Obtained support from subordinates Threatened to notify an outside agency Made formal appeals to higher levels
10 6
17 6 7
' Percentage of the 370 tabulated influence tactics reported by respondents.
INFLUENCE TACTICS frequently reported tactics were the use of logic and Michener and Schwertfeger (1972) to the rational discussions; when the goal was to improve a schemes described by French and Raven target's performance, respondents reported using ad- (1959) and by Cartwright (1965, pp. 1-47).
443
It
ministrative sanctions, training, and simplydemanding compliance. Finally, when the goal was to get others to is clear that the many influence tactics dedo the respondent's own work, the most frequently scribed here do not fit easily into any single reported tactic was the use of requests. All of the above classification scheme currently found in the findings were statistically reliable beyond the .05level. literature on power usage. Based on these The kinds of influence tactics used by the respondents varied with the power of the target person. Re- findings, we believe that new ways of classispondents significantly more often used the tactics of fying such tactics are needed that use the self-presentation, supporting data, and coalitions actual influence behaviors of organizational when attempting to influence their bosses. Different members as the starting point. An effort in tactics, however, were used to influence subordinates; this direction is reported in the next study. then respondents significantly more often used clandestine tactics, administrative sanctions, training, demanding, and explaining. Finally,the data showed that Study 2 only one tactic was significantly associated with influencing co-workers—the tactic of requesting help. Method At the .10 level, however, the tactics of exchange, reMany of the 14 categories of influence tactics in the quests, and rewards were also associated with influfirst study overlapped either conceptually or empiriencing co-workers. The use of influence tactics also varied with the cally or both (e.g., "weak ask" vs. "request"). The amount of resistance shown by target persons. When purpose of Study 2 was to determine the factor structhe respondents stated that the target at first refused to ture of the tactics found in the previous study. To this comply, the subsequent actions of respondents in- end, 58 items were developed from Study 1's tactics. cluded an increase in persistence and the use of per- These items were included in a questionnaireadminissonal negative actions. Additionally, when confronting tered to 754 employed respondents. Respondents were resisting bosses and co-workers, the respondents re- asked to describe on a 5-point scale how frequently ported an increase in the use of coalitions with fellow during the past 6 months they had used each item to inemployees. If the person resisting was a subordinate, fluence a target person at work. The 5-point scale had however, the respondents reported using more adminis- verbal anchors as follows: usually use this tactic to trative sanctions such as giving unsatisfactory per- influence him/her (5), frequently use this tactic to influence him/her (4), occasionally use this tactic to informance evaluations. fluence him/her (3), seldom use this tactic to influence him/her (2), and never use this tactic to influence him/ Results her (1).
The findings of the study suggest that in asked erespondentsetforms of thehoquestionnaire: One
Ther were thre
organizational settings the choice of influ- their bosses, another asked how they influenced their ence tactics is associated with what the re- co-workers, and the third asked how they influenced spondents are trying to get from the target their subordinates. The instructions read: person, the amount of resistance shown, and This questionnaire is a way of obtaining information the power of the target person. Combining about how you go about changing your boss's (or cothese findings suggests that administrative worker'witoryou. Below are)describesdo
variouhsewaysshef
o describe
w they influenced
s subordinate's mind that or agrees sanctions and personal negative actions are this.hPlease do not answer in terms of what yoou
more likely to be used when the target is a doingd like to do. woul subordinate who is actively resisting the re- Respondents were also told that if the tactics did not quest of the manager and when the reasons apply in their work to leave the space blank; blank refor exercising influence are based on the re- sponses were coded as "never used this tactic." spondent's role in the organization (e.g., Sample. The respondents were drawn from the improve target's performance). same managerial population as in Study 1 and were It is also important to note that many of taking graduate business courses part-time; 690ywere
the tactics reported by the respondents have enrollede inenrolledgacoursesBernard Baruch Graduatde received little mention in the organizational Center in New York City. Of the 754 respondents, 225 literature (e.g., the use of deceit, self-pre- described how they influenced their bosses, 285 desentation, and clandestine tactics). In fact, scribed how they influenced co-workers, .and 244 dethe tactics shown in Table 2 represent a be- scribed how they influenced subordinates wildering combination of several classifica- scribing hoforfrequently each influenceIntactic wanstused-, tion schemes from the exchange theories of a separate scale presented respondents with five
Reasons w exercising influence. additio o de 64 wer evenin t the at Temple Universit an
444
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Table 3
Mean Frequency of Reasons for Respondents Exercising Influence (Study 2)
Target status Reason for exercising influence Assistance on own job Assign work Obtain benefits Improve performance Initiate change Subordinate (244) 3.15. 4.08a 1.33a 3.93a 3.54a Co-worker (285) 2.75b 2.47b 1.65b 2.88b 3.27b Superior (225) 2.09C 1.86C 2.42C 2.39C 3.47a
F
64.36** 329.69** 90.41** 135.19** 6.66*
Note. Numbers in parentheses are Ns. Groups with different subscripts differ beyond the .05 level. High scores indicate that the goal was rated as a frequent reason for the respondent to influence the target person. * p < .05. **p < .01.
possible reasons for influencing the target person. These reasons were based on those found in Study 1 and read as follows: (a) have my boss (co-worker or subordinate) assist me on my job or do some of my work; (b) assign work to my boss (co-worker or subordinate) or tell him or her what to do; (c) have my boss (coworker or subordinate) give me benefits, such as raises, better hours of work, time off, better job assignments, and so on; (d) have my boss (co-worker or subordinate) do his or her own work better or do what they are supposed to do; (e) have my boss (co-worker or subordi- nate) accept my ideas for changes, for example, to accept a new way of doing the work more efficiently or a new program or project. For each of the five reasons, the respondents were asked to rate on a 5-point scale ranging from "very often" (5) to "never" (1) how frequently each reason had been the cause of their trying to influence the target person to do something. Paralleling the findings of Study 1, the reasons for exercising influence varied with whether the target person was a superior, co-worker, or subordinate. Table 3 shows that the major reasons for influencing subordinates were that respondents more frequently attempted to assign them work, to improve their task performance, and to have them assist the respondents in their own work (p < .01 compared to co-workers or superiors). Superiors were influenced most frequently to receive personal benefits (p < .01). Finally, re- spondents reported that they attempted to influence both their subordinates and their superiors with almost equal frequency to convince them to initiate change. This last result was also found in Study 1. Additional background data. The questionnaire also obtained information about the sex of the respondent, the sex of the respondent's boss, whether the organization was unionized, the number of persons employed in the respondent's work unit, and the job level of the respondent. Job level information was based on the respondent's description of his or her own work. This information was coded into four groups: clerical/ sales, professional (e.g., engineer, computer technician), first-line and middle managers (e.g., supervisor, manager of clerical unit, etc.), and top-level managers (e.g., vice president of marketing).
Results The 58 influence tactic items were factor analyzed using a principal component factor solution, with iterations for communality and varimax rotation. Forced two-factor through eight-factor solutions were carried out to aid in interpreting the findings. Factor analyses were carried out for the entire sample and separately for each of the three target status levels (superior, coworker, subordinate). This was done to examine the possibility that dimensions of influence would emerge at each target status level that did not emerge in the overall analysis. A second reason for carrying out separate factor analyses at each target status level was to ensure that any factors that emerged in the combined analysis also appeared in at least one of the separate factor analyses. It was assumed that a factor that did not appear in any of the separate analyses, yet appeared in the overall analysis, only reflected differences between target status levels in the exercise of influence. The factor analysis of the entire sample yielded six interpretable factors.1 These six factors accounted for 38% of the total item variance. Table 4 lists the 58 tactics presented in the questionnaire and their loadings on each of the six factors. The data in Table 4 are based on the overall factor analysis utilizing all 754 respondents.
1 One additional factor was dropped, since it did not emerge in any of the separate factor analyses by target status level. (text continued on page 447)
INFLUENCE TACTICS
445
Table 4 Rotated Factors and Item Loadings
Factors
?a
Dimension of influence/tactic
1
2
3
4
5
6
8"
Assertiveness
51. Kept checking up on him or her. 45. Simply ordered him or her to do what was 64 58 58 54 54 49 47 47 42 34 33 21 12 10 06 22 11 14 14 -13 -07 11 11 -10 -03 07 18 09 08 28 18 31 15 09 27 29 30 23 26 -11 -04 17 18 20 05 -05 -03 04 00 01 16 09 07 12 10 18 15 12 01 23 14 08 14 01 05 30 12 13 08 03 -09 03 33 28 36 24 27 -15 -14 -06 10 35 17 -10 -04 08 25 04
asked.
18. Demanded that he or she do what I requested. 39. Bawled him or her out. 11. Set a time deadline for him or her to do what
I asked.
19. Told him or her that the work must be done
as ordered or he or she should propose a better way. 53. Became a nuisance (kept bugginghim/her until he/she did what I wanted). 43. Repeatedly reminded him or her about what I wanted. 54. Expressed my anger verbally. 41. Had a showdown in which I confronted him or her face to face. 30. Pointed out that the rules required that he or she comply.
Ingratiation
46. Made him or her feel important ("only you
have the brains, talent to do this").
9. Acted very humbly to him or her while mak-
25 -01 -02 -02 27 32 25 -13 22 18 15
59 52 51 51 51 50 47 45 43 38 38
13 -02 11 19 14 09 17 21 12 15 15
15 -05 -01 10 20 22 05 -05 06 01 13
04 14 16 17 07 06 16 16 11 14 -04
09 07 04 03 06 12 15 -06 07 07 07
-17 07 -03 -05 -04 -09 05 04 -21 17 00
09 -02 -07 -04 05 19 12 21 11 02 23
ing my request.
17. Acted in a friendly manner prior to asking for
what I wanted.
28. Made him or her feel good about me before
making my request.
37. Inflated the importance of what I wanted him or her to do. 36. Praised him or her. 3. Sympathized with him/her about the added
problems that my request has caused.
44. Waited until he or she appeared in a receptive
mood before asking.
10. Showed my need for their help. 29. Asked in a polite way. 22. Pretended I was letting him or her decide
to do what I wanted (act in a pseudo-democratic fashion).
Rationality
40. Wrote a detailed plan that justified my ideas. -06 38. Presented him or her with information in support of my point of view. 06 31. Explained the reasons for my request. 17 13. Used logic to convince him or her. 10 24. Wrote a memo that described what I wanted. 15 42. Offered to compromise over the issue (I gave in a little). 14 16. Demonstrated my competence to him or her before making my request. -10 -04 17 20 13 03 23 23 65 64 57 55 42 40 35 06 04 -03 02 04 00 10 -02 05 08 03 00 23 04 21 -07 01 -02 26 00 02 -17 -15 07 -03 00 04 -11 61 37 49 32 31 47 17
446
Table 4 (continued)
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Factors Dimension of influence/tactic
1 2 3 4 5 6 7" 8"
Sanctions
49. Gave no salary increase or prevented the per-
son from getting a pay raise.
26. Threatened his or her job security (e.g., hint
17 26 06 24 11
03 05 11 09 00
06 03 07 01 02
63 59 52 52 46
01 02 11 -04 07
01 08 01 13 11
02 04 -05 02 04
08 11 15 15 -03
of firing or getting him or her fired).
15. Promised (or gave) a salary increase. 6. Threatened to give him or her an unsatisfac-
tory performance evaluation.
34. Threatened him or her withloss of promotion. 35. Offered an exchange (e.g. , if you do this for
Exchange me, I will do something for you).
27. Reminded him or her of past favors that I 03 07 21 19 02 01 03 04 70 53 08 03 22 17 06 -15
did for them.
50. Offered to make a personal sacrifice if he
or she would do what I wanted (e.g. , work late, work harder, do his/her share of the work, etc). 55. Did personal favors for him or her. 7. Offered to help if he/she would do what I wanted.
58. Made a formal appeal to higher levels to back
-07 04 23
28 37 33
10 03 07
10 07 10
52 48 40
05 03 20
10 15 -12
15 09 14
Upward appeal up my request.
20. Obtained the informal support of higher-ups. 20 25. Filed a report about the other person with 10 17 28 12 26 03 09 09 08 06 07 -01 06 31 13 11 05 04 05 59 47 42 40 29 -05 -02 -03 14 23 16 05
higher-ups (e.g., my superior).
33. Sent him or her to my superior.
Blocking
47. Threatened to notify an outside agency if he
or she did not give in to my request.
48. Threatened to stop working with him or her
17 10 02 05 06
-12 -11 03 03 19
04 02 -08 -14 04
19 16 03 08 06
12 20 07 06 13
19 07 16 01 10
80 63 59 37 33
10 -03 07 -08 -02
until he or she gave in.
4. Engaged in a work slowdown until he or she
did what I wanted.
5. Ignored him or her and/or stopped being
friendly.
14. Distorted or lied about reasons he or she
should do what I wanted.
12. Obtained the support of co-workers to back
Coalitions up my request.
56. Had him or her come to a formal conference 06 25 06 17 04 16 35 34 35 -13 15 11 02 02 -06 29 35 18 28 13 20 50 47 46
at which I made my request.
32. Obtained the support of my subordinates to
back up my request.
52. Kept kidding him or her until they did what
Unclassified items I wanted.
23. Ignored him or her and went ahead and did 33 10 34 11 13 00 -10 -08 25 11 -02 01 -01 16 -09 05
what I wanted.
INFLUENCE TACTICS
447
Table 4 (continued)
Factors Dimension of influence/tactic Unclassified items 8. Provided him or her with various benefits that they wanted. 1. Challenged his or her ability (' 'I bet you can't do that"). 21. Pretended not to understand what needed to be done so that he or she would volunteer
to do it for me.
07 17 02 03
22 13 24 22 6.5
17 -01 -04 13 5.2
25 13 05 07 4.2
28 07 10 -01 3.2
07 09 05 10 2.9
08 03 18 24 5.7
19 07 16 12 5.0
2. Concealed some of my reasons for trying to influence him/her. % of total variance
18.2
t>tote. N = 754. Decimals omitted. Item numbers denote the items' original position in the questionnaire. a Data based on factor analysis of tactics used to influence superiors. b Data based on factor analysis of tactics used to influence subordinates.
Factor 1 is identified by highest loadings on the influence tactics, including demanding, ordering, and setting deadlines. This factor is labeled Assertiveness. The factor emerged as a dimension of influence at all target status levels (superior, coworker, and subordinate). Factor 2 is described by the highest loadings on weak and nonobtrusive influence tactics. Included here were such tactics as "acting humble" and "making the other person feel important." This factor is labeled Ingratiation. The factor emerged as a dimension of influence at all levels of target status. Factor 3 is characterized by loadings on the use of rationality influence tactics and is labeled Rationality. It includes such tactics as "writing a detailed plan" and "explaining the reasons for my request." This factor emerged at each target status level. In the analysis of tactics directed toward subordinates, however, additional items that involved group pressure also loaded highly on this factor. These latter items will be discussed subsequently. Factor 4 involved the use of administrative sanctions to induce compliance. Tactics with high loadings included "prevented salary increases" and "threatened job security." This factor is labeled Sanctions. It emerged as a dimension of influence at all levels of target status. Factor 5 loaded on tactics involving the
exchange of positive benefits. Included here were such tactics as "offering an exchange" and "offering to make personal sacrifices." This factor is labeled Exchange of Benefits. This factor only emerged in the factor analysis of influence tactics directed toward superiors. Factor 6 is described by loadings on tactics that bring additional pressure for conformity on the target by invoking the influence of higher levels in the organization. Included here were such tactics as "making a formal appeal to higher levels" and "obtaining the informal support of higher-ups." This factor only emerged in the factor analysis of influence tactics directed toward superiors; it is labeled Upward Appeal. Two additional factors (Factors 7 and 8) did not emerge in the overall factor analysis but were found in the subanalyses. It was decided to retain these factors for heuristic purposes. Factor 7 emerged in the factor analysis of influence directed toward superiors. Items that loaded on this factor included "engaging in a work slowdown" and "threatening to stop working with the target person." Essentially, these tactics are attempts to stop the target person from carrying out some action by various kinds of blocking tactics. This factor is labeled Blocking. Factor 8 emerged from the factor analysis of tactics directed toward subordinates.
448
Table 5
D. KIPNIS, S. SCHMIDT, AND I.WILKINSON
Intercorrelation of Scale Scores Dimension of influence
23 8
1. Assertiveness .22 .32 .24 2. Ingratiation .14 .24 3. Sanctions .13 4. Rationality 5. Exchange 6. Upward appeal 7. Blocking 8. Coalitions Note. N = 754.
.26 .49 .24 .20
.35 .21 .28 .1.0 .32 .21 .22 -.03 .27 .22 .20
.28 .31 .17 .32 .29 .36 .16
The remaining factors show reasonably low levels of intercorrelation, with rs ranging from .03 to .36. It should be noted that the individual correlation matrices at each status level closely paralleled the averaged findings reported in Table 5. Table 6 presents the reliability of the eight scale scores (coefficient alpha) for each target status level. With the exception of low reliability for the use of Blockingtactics with subordinates and co-workers, the remaining factor scores have satisfactory reliabilities. Correlates of Influence Tactics Target status. Table 7 shows the means and standard deviations of each influence dimension as a function of the status of the target. Seven of the eight dimensions were significantly associated with the relative status of the target. Basically, the findings suggest that as the status of the target person increased, respondents placed more reliance on rationality tactics. An analysis using Duncan's multiple-range test found that Assertive tactics and Sanctions were used more often to influence subordinates than co-workers or superiors (ps < .01). The tactics of Ingratiation, Exchange of Benefits, and Upward Appeal were used with equal frequency among subordinates and coworkers but significantly less often when attempting to influence superiors (p < .01). Finally, respondents reported that they used Rationality tactics more frequently to convince superiors than co-workers or subordinates (p < .01). As will be shown later, these differences in choice of tactics in part reflect the fact that respondents have different reasons for influencing target persons at different status levels. Goals. It will be recalled that respondents in the present study rated the importance of five possible reasons for exercising influence. To determine whether different combinations of tactics were used as the reasons for influencing varied, a set of stepwise regression analyses was performed. These stepwise regression analyses examined the relation between the eight tactic scores and the rated importance of each reason for
2
Items in this factor were part of the previously described factor Rationality. However, this subset of items described the use of steady pressure for compliance by "obtaining the support of co-workers" and by "obtaining the support of subordinates." This is labeled Coalitions. Scale construction. To aid further analysis, scales were constructed whose items were selected to represent each of the eight dimensions of influence.2 Selection was made on the basis of two criteria. First, items were selected that loaded over .40 on agiven dimension and did not load above .25 on any of the remaining dimensions. Second, from the pool of items that were selected to represent each dimension, items were selected based on an examination of each item's correlation with other items representing the factor and their correlations with items in the remaining factors. High item intercorrelation within a factor and low item intercorrelation with the remaining items were used as the final criteria for selecting items. To determine the intercorrelation between the eight scales, correlation matrices were first computed at each of the three status levels. Next, all intercorrelations were converted to their z' equivalents, averaged, and then reconverted to productmoment correlations. This procedure provides an estimate of the correlation between the eight scales, with the variance attributable to target status partialed out. As shown in Table 5, there was a fair degree of independence among the factors, with the exception of the relationship between Ingratiation and Exchange tactics (r = .49).
Copiesof the scales are available from the authors.
INFLUENCE TACTICS
449
Table 6 Reliability of Tactic Scores (Alpha
Number of items 6 4 5 5 5 4 3 2
Coefficient)
Target status Subordinate .65 .61 .61 .69 .70 .65 .47 .71 Co-worker .73 .77 .68 .71 .74 .65 .42 .64 Superior .71 .70 .65 .54 .76 .68 .74 .81 Total .70 .71 .78 .73 .73 .67 .53 .75
of influence Ingratiation Rationality Assertiveness Sanction Exchange Upward appeal Blocking Coalitions
exercising influence. A restriction put on the analysis was that only tactics that correlated significantly (p < .05) with a given reason were entered into the regression analysis. Since we have already found that the type of tactic used and the reasons for exercising influence both varied with the status of the target person, separate regression analyses were carried out at each target status level.
Table 8 shows these findings in terms of the particular combination of influence tactics that best predicted each reason for exercising influence. In addition, Table 8 presents the multiple correlations of these influence tactics with each reason for exercising influence. Table 8 shows that at all target status levels, the respondent's choice of influence
Table 7 Target Status and Average Frequency of Tactic Use
Target status Dimension of influence Ingratiation Subordinate 15.92b 3.97 14.04b 2.47 12.04a 3.00 6.39a 1.95 9.83b 3.10 7.05b 2.31 3.31a 0.82 4.37 2.04 Co-worker 16.08b 4.73 13.74b 3.20 8.06b 2.67 5.27b 0.91 10.08b 3.65 6.83b 2.29 3.50b 1.04 4.30 2.05 Superior 14.52a 4.48 14.72a 3.01 6.94C 2.25 5.16b 0.62 8.56a 3.53 5.43a 1.99 3.29a 0.91 4.53 2.14
F
16.69** 7.17** 243.91** 68.18** 13.52** 37.16** 4.05* .079
M
SD Rationality
M
SD Assertiveness M SD Sanctions M SD Exchange
M
SD Upward appeal M SD Blocking M SD Coalitions
M
SD
Note. Groups with different subscripts differ beyond the .05 level. * p < .05. **p < .01.
450
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Table 8 Multiple Correlations Between Tactics and Rated Importance of Reasons for Exercising Influence
Reason for exercising influence/target Receive assistance on own job boss co-worker subordinate Assign work to target boss co-worker subordinate Obtain benefits from target boss co-worker subordinate Improve target's performance boss co-worker Tactic category used Ingratiation Ingratiation Assertiveness Ingratiation Assertiveness Assertiveness Assertiveness Exchange Ingratiation Exchange Blocking Ingratiation Assertiveness Coalitions Assertiveness Blocking Rationality Assertiveness Exchange Coalitions Rationality Assertiveness Rationality Rationality Coalition Ingratiation Exchange Rationality Coalitions Exchange Assertiveness Rationality .20 .16 .29 .51 .28 .31
.38
.32 .21
.43
.46 .34
subordinate Initiate change boss
.57 .45 .42
co-worker subordinate
Note. Tactics are listed in order of their entry into the stepwise multiple regression equations. Allcorrelations are significant beyond the .05level.
tactics varied with the respondents' reasons for exercising influence. That is, respondents who frequently sought personal assistance from target persons used Ingratiation tactics; respondents who frequently assigned work to target persons used Assertiveness; and respondents who frequently tried to improve a target person's performance used Assertiveness and Rationality tactics. Fi-
nally, respondents who frequently tried to convince target persons to accept new ideas used Rationality tactics. Another finding of interest in Table 8 is that respondents showed the least variation in choice of tactics when attempting to influence their subordinates. No matter what the reason for influencing subordinates, the use of Assertiveness was associated with each of the five reasons and accounted for the most variance for each reason. In contrast, when influencing co-workers and superiors, the use of Assertiveness was associated only with two of the five reasons. Other situational factors. So far it has been shown that use of the eight dimensions of influence was associated with the relative power of the target person and the reasons why the respondents wanted to influence the target person. In this section we examine the relationship between the use of tactics and five personal or situational characteristics of the respondent: sex, level in the organization, whether the respondent's unit was unionized, number of people in respondent's work unit, and sex of respondent's boss. Thefindingsshowed that the respondent's own level in the organization was closely associated with use of influence tactics. Compared to those with low job status, respondents with higher job status reported greater use of Rationality and Assertiveness tactics when influencing both their subordinates and their superiors (ps < .01). In addition, respondents with higherjob status used Sanctions more frequently and sought aid from their superiors less frequently when influencing their subordinates (ps < .01). Thus, as the respondents' own job status rose, they were more likely to use more direct tactics of influence and be less dependent upon superiors. Size of work unit also related to the use of tactics on subordinates. In large work units, respondents more frequently used Assertiveness (p < .01), Sanctions (p < .01), and Upward Appeal (p < .05) when influencing subordinates. These findings are consistent with the general idea that as the number of persons in a work unit increases, a greater reliance is placed on strong and impersonal means of control.
INFLUENCE TACTICS
451
Finally, the presence of unions was as- influence tactics among peers and bosses. sociated with the use of certain tactics. If the Hence, the range of influence tactics used in organization was unionized, respondents that early research covered fewer areas than were more likely to use Ingratiating tactics those included here. to influencesubordinates (p < .01), to avoid Other factorial studies of influence tactics the use of Assertiveness when influencing have focused on dimensions of influence in co-workers (p < .05), and to use Ration- interpersonal settings (Falbo, 1977;Kipnis, ality tactics less frequently (p < .01) and Cohen, & Catalano, Note 2). These studies Blocking tactics more frequently (p < .01) have uncovered only two or three dimenwhen influencing bosses. sions of influence. For example, Kipnis et There were no significant relations as- al. reported that dating couples used three sociated with the sex of the respondent and dimensions of influence—strong, weak, and the sex of the respondent's boss in terms of rational—as ways of changing their dating the frequency of use of the eight dimensions partner's behavior. The items loading on of influence. Thus, in the present study men these three dimensions closely resemble and women chose similar tactics when at- the dimensions we have labeled Assertivetempting to get their way. ness, Rationality, and Ingratiation tactics. Thus, instead of acting "more loving" in order to influence, organizational members Discussion act "humble" and offer to "make sacriIn Study 1, influence tactics were sorted fices." The parallel is clearly there. The into 14 categories, but we suspected that a additional factors found in the present study more parsimonious classification system suggest that organizational resources exwas likely to exist. The results of Study 2 tend the range of tactics that organizational yielded eight dimensions of influence. members can use. Thus, organizational Four of these dimensions emerged at all members can invoke formal sanctions or the status levels. These dimensions were As- added authority of higher management as sertiveness, Sanctions, Ingratiation, and compliance-gaining strategies. Rationality. The remaining dimensions were It is recognized, of course, that findings uniquely associated with influencing either presented here are based on self-reportmeasuperiors or subordinates. The tactics of sures and require replication using different Exchange of Benefits, Blocking, and Up- methodologies. Nevertheless, the picture ward Appeal emerged when respondents de- portrayed by these data is of organizational scribed how they influenced their bosses. members actively seeking to influence peers, These dimensions did not appear in the anal- superiors, and subordinates for a variety of ysis of influence tactics used among sub- reasons, some personal and some based on ordinates. Finally, the use of Coalitions their management roles. This picture suptactics appeared only when respondents de- ports the view of those who argue that scribed how they influenced their sub- organizational leadership is more compliordinates. cated than is represented in organizational Unfortunately, there are few studies avail- behavior textbooks (Kochan, Schmidt, & able with which to compare the present DeCotiis, 1976). Such texts mainly focus on findings. Within organizational settings, the the ways in which higher levels in the organimajor empirical studies of influence tactics zation influence lower levels. This is called have been based on the Ohio State-Navy the leadership process. In fact, we would leadership studies (Fleishman, 1973). These suggest that everyone is influencing everystudies could identify only two dimensions one else in organizations, regardless of job of influence—consideration and initiating title. People seek benefits, information, satstructure. However, this early program of isfactory job performance, the chance to do research was only concerned with how better than others, to be left alone, coleaders exercised influence over subordi- operation, and many other outcomes too nates and was not concerned with the use of numerous to mention.
452
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Eastern Psychological Association, Philadelphia, In fact, there may be very little difference April 1979. in the frequency with which people try to influence their bosses, co-workers, and subReferences ordinates, given all these various reasons for trying to influence others. What shows re- Cartwright, D. Influence, leadership and control. In markable variation, however, are the kinds J. G. March (Ed.), Handbook of organizations. of tactics that are chosen when trying to Etzioni, A:. Rand McNally,dimensions and their inter-
obtain these various outcomes. The present article has shown that these tactics vary with the particular wants of the influencing agent
1965. Organizational relationship. In B. Indik & F. K. Berrien (Eds.), People, groups and organizations. New York: Teachers College Press, 1968.
Chicago
and his or her degree of control over the Falbo, T. .The multidimensional scalingSocialpower of strategies Journal of Personality and Psytarget of influence. The implications of chology, 1977, 35, 537-547. these findings for the understanding of or- Fleishman, E. A. Twenty years of consideration and structure. In E. A. Fleishman & J. G. Hunt (Eds.), ganizational politics remain to be explored. Current developments in the study of leadership. As a final point, the scales that have been Carbondale: Southern Illinois University Press, developed to measure the various dimensions of influence can be useful in further French., J. R. P., & Raven, B. H. The bases of social 1973 research. Even though the use of influence power. In D. Cartwright (Ed.), Studies in social within organizations is a topic of considerpower. Ann Arbor: University of Michigan Press, able interest, there have been few systematic Izraeli ., D. N. The middle manager and the tactics of
1959 power expansion. Sloan Management Review, 1975, attempts to develop a means of measuring 16, 57-70. such behavior. The present dimensions provide the potential for profiling the use of in- Kochan, T. A., Schmidt, S. M., & DeCotiis, T. A. fluence in a variety of organizational settings Superior-subordinate relations: Leadership and and at different levels as well as between Michener, .A.Human Relations,, 1976,Likin279-294. -
various groups. Reference Notes
1. Goodchild, J. D., Quadrado, C., & Raven, B. H. Getting one's way. Paper presented at the meeting of the Western Psychological Association, Sacramento, California, April 1975. 2. Kipnis, D., Cohen, E., & Catalano, R. Power and affection. Paper presented at the meeting of the
28, , & Schwertfeger M. g as a deter minant of power tactic preference. Sociometry, \972,35, 190-202. Raven, B. H. The comparative analysis of power and influence. In J. T. Tedeschi (Ed.), Perspective on social power. Chicago: Aldine, 1974. Schein, V. E. Individual power and political behavior. Academy of Management Review, 1977, 2, 64-72.
headship
Received July 2, 1979
doc_272223178.docx
Study on Intra-organizational Influence Tactics: Explorations in Getting One's Way
Study on Intra organizational Influence Tactics: Explorations in Getting One's Way
The tactics used by people at work to influence their superiors, co-workers, and subordinates were investigated in two studies. In the first study, 165 lower-level managers wrote essays describing an incident in which they influenced either their bosses, co-workers, or subordinates. Through content analysis, a total of 370 influence tactics grouped into 14 categories were identified. The tactics ranged from the use of rational discussion through the use of exchange tactics to the use of clandestine tactics. In a second study, the 370 influence tactics were rewritten into a 58-item questionnaire. New respondents described the extent to which they used each item to influence their bosses (n = 225), co-workers (n = 285), or subordinates (n = 244). Based on afactor analysis of the questionnaire, eight dimensions of influence were found: assertiveness, ingratiation, rationality, sanctions, exchange, upward appeals, blocking, and coalitions. It was found that the frequency with which each influence dimension was used related to the relative power of the respondents and their targets of influence, the reasons for exercising influence, the resistance of the target person, the organizational status of the respondents, organizational size, and whether the organization was unionized. Sex of the respondents and sex of the respondents' bosses, however, were not related to the choice of influence tactics in the present study.
Organizational psychologists have not been particularly interested in studying the ways in which people at work influencetheir colleagues and superiors to obtain personal benefits or to satisfy organizational goals, For the most part, interest has centered on the ways subordinates can be influenced to improve subordinate productivity and morale. This latter use of influence is customarily called the study of leadership, whereas the former can be called the study of organizational politics. A consequence of this focus on leadership is that there is little systematic information available about how people use power to influence their colleagues or superiors. With
We wish to thank Sheldon Zalkind for gathering data attheBernard Baruch Graduate Center Requests for reprints should be sent to Stuart M.
but few exceptions (e.g., Izraeli, 1975; Schein, 1977), our thinking about this topic is guided by anecdotal evidence or armchair speculations that have been organized into rational classifications of power tactics (e.g., Etzioni, 1968, pp. 94- 1Q9; French & Raven, 1959). One problem with these classifications of power tactics, as Raven (1974) has pointed out, is that they overlap with each other, though each varies in the number of influence dimensions that are described. A further problem with existing classifications of power tactics is that when influence acts are actually studied, it is found that people do not exercise influence in ways predicted by rational classification schemes. This point was first explicitly made in a study by Goodchild, Quadrado, and Raven (Note
1) Jn which college Students wrote brief the t ; 1>H T t ' . .. , •a f „ Jt
Schmidt, Department of Industrial Relations and was found that many of the influence tactics Organizational Behavior, Temple University, Phila- described by these students could not be delphia, Pennsylvania 19122. classified into preexisting categories. Sev440
•
INFLUENCE TACTICS
441
eral tactics thought to be basic when classifying influence, such as the use of expert power, were not even mentioned by the students. There is a need, therefore, for empirical studies of the use of influence within organizations. The purpose of this article is to report two studies that sought to examine the tactics of influence used by people at work when attempting to change the behavior of their superiors, co-workers, and subordinates. In the first study, the range of tactics that people use at work was identified by applying content analysis to written descriptions by managers of their attempts to influence their bosses, co-workers, or subordinates. The second study sought to identify through factor analysis the dimensions of influence underlying the specific tactics that were uncovered in the first study. Study 1: Determining Intraorganizational Influence Tactics Method
The data for this study were collected from 165respondents, 25% women and 75% men. All respondents were taking graduate business courses part-time, in the evening. They were told to refer to their current employment experience when supplying the requested information. The respondents were employed mainly in managerial roles as engineers, technicians, and professionals. Since most respondents came from different organizations, there were almost as many organizations as respondents represented. Respondents were asked to describe an incident in which they actually succeeded in getting either their boss, a co-worker, or a subordinate to do something they wanted; 62 described how they got their way with their boss, 49 with a co-worker, and 54 with a sub-
ordinate. The reason for the unequal distribution of respondents over status levels was that fewer respondents had subordinates or co-workers than had bosses. In their description of an incident, each respondent wrote in essay form what they wanted from a target person, what they did, whether there was resistance from the target, and what further influence tactics were used in response to resistance from the target. Each respondent also completed a structured questionnaire containing demographic and specific organizational job-situation questions. The incidents were sorted first in terms of the goal sought from the target person. This sorting yielded fivegeneral categories of goals: assistance with own job —obtaining the assistance of the target in helping the respondent do his or herjob, when it was not part of the target's legitimate job duties;get others to do theirjob— getting the target to do his or her own work; obtain benefits—goals that personally benefited the respondent, such as salary increase, promotion, and improved work schedule; initiate change—initiating new organizational programs and systems or improving the coordination of organizational activities (e.g., changing a scheduling procedure); and improve performance— improving the target's on-the-job performance. These goals varied as a function of the target's job status, as shown in Table 1. Respondents sought mostly self-interest goals from their superiors. The primary goal sought from co-workers was to get assistance with the respondent's own job. The most prevalent reason for influencing subordinates was to get them to do their jobs. Finally, the goal of initiating change was sought both from bosses and from subordinates in almost equal proportion. However, the changes sought from superiors focused onjob-related organizational changes, such as launching a new accounting procedure or starting a special project. But with subordinates, the changes sought dealt with job performance, such as changes in the way ajob should be done or the manner of working in the organization. Next, the influence tactics reported by the respondents in attempting to achieve these goals were identified. A total of 370influence tactics were reported by the 165 respondents. The authors sorted these tactics into 14 categories (see Table 2). Consensus among the coders was used as the criterion for assigning a given tactic to a
Table 1 Reasons for Exercising Influence, by Target Status (Study 1) in Percentages
Target status Reasons Obtain assistance on own job Get others to do their jobs Obtain personal benefits Initiate change in work Improve target's job performance Note. Numbers in parentheses are Afs. Boss (62)
3 13 58 26 0
Co-worker (49)
48 23 10 15 4
Subordinate (54)
9 46 0 28 17
Total (165)
18 27 25 23 7
442
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON In addition to identifying influence tactics, the analyses of Study 1 focused on the correlates of the 14 categories of influence. To this end, a unit weight of 1 was assigned if the respondent reported using any of the items comprising a given influence tactic category; a score of 0 was assigned if none of the items comprising a category were mentioned. Thus, each respondent had 14 scores of 1 or 0. Next, multivariate analyses of variance, which if significant werefollowed by univariate analyses, were carried out to examine the relationship among the 14 categories of influence, the status of the target person, and the goals sought by respondents in exercising influence. The influence tactics used by the respondents varied with the goal sought from the target person. When the goal was self-interest, the most frequently reported tactics were self-presentationand personal negative actions; when the goal was to initiate change, the most
category. The 14 categories ranged from the use of administrative sanctions and personal threats through the use of logic and rational discussions to clandestine, dependency appeal, and ingratiating tactics. The individual tactics illustrating these 14 categories are also shown in Table 2. To determine the reliability of the assignment of items to these 14 categories, three coders who were not associated with the research independently sorted the 370 influence tactics into the 14 categories. Raters 1 and 2 agreed on the placement of items 61% of the time; Raters 1and 3 agreed 64% of the time; and Raters 2 and 3 agreed 65% of the time. Since there were 14 categories into which any item could be placed, the degree of agreement between the three raters suggests modestly high reliability for the classification scheme. The areas of disagreement between the raters involved primarily the three categories "weak ask," "explain," and "request."
Table 2 Classification of Influence Tactics by Category (Study 1)
Category/tactic Clandestine Challenged the ability of the target Lied to the target Acted in a pseudo-democratic manner Puffed up the importance of the job Manipulated information Made the target feel important Cajoled the target Showed understanding (pretended) of the target's problem Personal negative actions Fait accompli/went ahead on own Chastised the target Became a nuisance Slowed down on the job Held personal confrontation with target Threatened withdrawal of help Expressed anger Threatened to leave job Blocked target's actions Ignored target Administrative negative actions Filed a report with supervisor Sent target to superior for conference Gave unsatisfactory performance evaluations Gave no salary increase Threatened with unsatisfactory performance ratings Threatened job security Threatened loss of promotion Exchange Contributed in exchange for compliance Compromised Offered to make sacrifice Offered help to get the job done Invoked past favors Category/tactic Persistance Repeated reminders Argued Repeated previous actions Surveillance Training Explained how it was to be done Showed how to do it Reward Verbal reinforcement Salary raise Gave benefits Self presentation Demonstrated competence Performed well, then asked Waited until target was in the right mood Was humble Was friendly Direct request Weak ask Showed dependency Weak request Demand Invoked rules Ordered Convened formal conference Set time deadline Told target that it must be done as I said or better proposed Explained rationale for request Gathered supporting data Coalitions Obtained support from co-workers Obtained support informally from superiors Obtained support from subordinates Threatened to notify an outside agency Made formal appeals to higher levels
10 6
17 6 7
' Percentage of the 370 tabulated influence tactics reported by respondents.
INFLUENCE TACTICS frequently reported tactics were the use of logic and Michener and Schwertfeger (1972) to the rational discussions; when the goal was to improve a schemes described by French and Raven target's performance, respondents reported using ad- (1959) and by Cartwright (1965, pp. 1-47).
443
It
ministrative sanctions, training, and simplydemanding compliance. Finally, when the goal was to get others to is clear that the many influence tactics dedo the respondent's own work, the most frequently scribed here do not fit easily into any single reported tactic was the use of requests. All of the above classification scheme currently found in the findings were statistically reliable beyond the .05level. literature on power usage. Based on these The kinds of influence tactics used by the respondents varied with the power of the target person. Re- findings, we believe that new ways of classispondents significantly more often used the tactics of fying such tactics are needed that use the self-presentation, supporting data, and coalitions actual influence behaviors of organizational when attempting to influence their bosses. Different members as the starting point. An effort in tactics, however, were used to influence subordinates; this direction is reported in the next study. then respondents significantly more often used clandestine tactics, administrative sanctions, training, demanding, and explaining. Finally,the data showed that Study 2 only one tactic was significantly associated with influencing co-workers—the tactic of requesting help. Method At the .10 level, however, the tactics of exchange, reMany of the 14 categories of influence tactics in the quests, and rewards were also associated with influfirst study overlapped either conceptually or empiriencing co-workers. The use of influence tactics also varied with the cally or both (e.g., "weak ask" vs. "request"). The amount of resistance shown by target persons. When purpose of Study 2 was to determine the factor structhe respondents stated that the target at first refused to ture of the tactics found in the previous study. To this comply, the subsequent actions of respondents in- end, 58 items were developed from Study 1's tactics. cluded an increase in persistence and the use of per- These items were included in a questionnaireadminissonal negative actions. Additionally, when confronting tered to 754 employed respondents. Respondents were resisting bosses and co-workers, the respondents re- asked to describe on a 5-point scale how frequently ported an increase in the use of coalitions with fellow during the past 6 months they had used each item to inemployees. If the person resisting was a subordinate, fluence a target person at work. The 5-point scale had however, the respondents reported using more adminis- verbal anchors as follows: usually use this tactic to trative sanctions such as giving unsatisfactory per- influence him/her (5), frequently use this tactic to influence him/her (4), occasionally use this tactic to informance evaluations. fluence him/her (3), seldom use this tactic to influence him/her (2), and never use this tactic to influence him/ Results her (1).
The findings of the study suggest that in asked erespondentsetforms of thehoquestionnaire: One
Ther were thre
organizational settings the choice of influ- their bosses, another asked how they influenced their ence tactics is associated with what the re- co-workers, and the third asked how they influenced spondents are trying to get from the target their subordinates. The instructions read: person, the amount of resistance shown, and This questionnaire is a way of obtaining information the power of the target person. Combining about how you go about changing your boss's (or cothese findings suggests that administrative worker'witoryou. Below are)describesdo
variouhsewaysshef
o describe
w they influenced
s subordinate's mind that or agrees sanctions and personal negative actions are this.hPlease do not answer in terms of what yoou
more likely to be used when the target is a doingd like to do. woul subordinate who is actively resisting the re- Respondents were also told that if the tactics did not quest of the manager and when the reasons apply in their work to leave the space blank; blank refor exercising influence are based on the re- sponses were coded as "never used this tactic." spondent's role in the organization (e.g., Sample. The respondents were drawn from the improve target's performance). same managerial population as in Study 1 and were It is also important to note that many of taking graduate business courses part-time; 690ywere
the tactics reported by the respondents have enrollede inenrolledgacoursesBernard Baruch Graduatde received little mention in the organizational Center in New York City. Of the 754 respondents, 225 literature (e.g., the use of deceit, self-pre- described how they influenced their bosses, 285 desentation, and clandestine tactics). In fact, scribed how they influenced co-workers, .and 244 dethe tactics shown in Table 2 represent a be- scribed how they influenced subordinates wildering combination of several classifica- scribing hoforfrequently each influenceIntactic wanstused-, tion schemes from the exchange theories of a separate scale presented respondents with five
Reasons w exercising influence. additio o de 64 wer evenin t the at Temple Universit an
444
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Table 3
Mean Frequency of Reasons for Respondents Exercising Influence (Study 2)
Target status Reason for exercising influence Assistance on own job Assign work Obtain benefits Improve performance Initiate change Subordinate (244) 3.15. 4.08a 1.33a 3.93a 3.54a Co-worker (285) 2.75b 2.47b 1.65b 2.88b 3.27b Superior (225) 2.09C 1.86C 2.42C 2.39C 3.47a
F
64.36** 329.69** 90.41** 135.19** 6.66*
Note. Numbers in parentheses are Ns. Groups with different subscripts differ beyond the .05 level. High scores indicate that the goal was rated as a frequent reason for the respondent to influence the target person. * p < .05. **p < .01.
possible reasons for influencing the target person. These reasons were based on those found in Study 1 and read as follows: (a) have my boss (co-worker or subordinate) assist me on my job or do some of my work; (b) assign work to my boss (co-worker or subordinate) or tell him or her what to do; (c) have my boss (coworker or subordinate) give me benefits, such as raises, better hours of work, time off, better job assignments, and so on; (d) have my boss (co-worker or subordinate) do his or her own work better or do what they are supposed to do; (e) have my boss (co-worker or subordi- nate) accept my ideas for changes, for example, to accept a new way of doing the work more efficiently or a new program or project. For each of the five reasons, the respondents were asked to rate on a 5-point scale ranging from "very often" (5) to "never" (1) how frequently each reason had been the cause of their trying to influence the target person to do something. Paralleling the findings of Study 1, the reasons for exercising influence varied with whether the target person was a superior, co-worker, or subordinate. Table 3 shows that the major reasons for influencing subordinates were that respondents more frequently attempted to assign them work, to improve their task performance, and to have them assist the respondents in their own work (p < .01 compared to co-workers or superiors). Superiors were influenced most frequently to receive personal benefits (p < .01). Finally, re- spondents reported that they attempted to influence both their subordinates and their superiors with almost equal frequency to convince them to initiate change. This last result was also found in Study 1. Additional background data. The questionnaire also obtained information about the sex of the respondent, the sex of the respondent's boss, whether the organization was unionized, the number of persons employed in the respondent's work unit, and the job level of the respondent. Job level information was based on the respondent's description of his or her own work. This information was coded into four groups: clerical/ sales, professional (e.g., engineer, computer technician), first-line and middle managers (e.g., supervisor, manager of clerical unit, etc.), and top-level managers (e.g., vice president of marketing).
Results The 58 influence tactic items were factor analyzed using a principal component factor solution, with iterations for communality and varimax rotation. Forced two-factor through eight-factor solutions were carried out to aid in interpreting the findings. Factor analyses were carried out for the entire sample and separately for each of the three target status levels (superior, coworker, subordinate). This was done to examine the possibility that dimensions of influence would emerge at each target status level that did not emerge in the overall analysis. A second reason for carrying out separate factor analyses at each target status level was to ensure that any factors that emerged in the combined analysis also appeared in at least one of the separate factor analyses. It was assumed that a factor that did not appear in any of the separate analyses, yet appeared in the overall analysis, only reflected differences between target status levels in the exercise of influence. The factor analysis of the entire sample yielded six interpretable factors.1 These six factors accounted for 38% of the total item variance. Table 4 lists the 58 tactics presented in the questionnaire and their loadings on each of the six factors. The data in Table 4 are based on the overall factor analysis utilizing all 754 respondents.
1 One additional factor was dropped, since it did not emerge in any of the separate factor analyses by target status level. (text continued on page 447)
INFLUENCE TACTICS
445
Table 4 Rotated Factors and Item Loadings
Factors
?a
Dimension of influence/tactic
1
2
3
4
5
6
8"
Assertiveness
51. Kept checking up on him or her. 45. Simply ordered him or her to do what was 64 58 58 54 54 49 47 47 42 34 33 21 12 10 06 22 11 14 14 -13 -07 11 11 -10 -03 07 18 09 08 28 18 31 15 09 27 29 30 23 26 -11 -04 17 18 20 05 -05 -03 04 00 01 16 09 07 12 10 18 15 12 01 23 14 08 14 01 05 30 12 13 08 03 -09 03 33 28 36 24 27 -15 -14 -06 10 35 17 -10 -04 08 25 04
asked.
18. Demanded that he or she do what I requested. 39. Bawled him or her out. 11. Set a time deadline for him or her to do what
I asked.
19. Told him or her that the work must be done
as ordered or he or she should propose a better way. 53. Became a nuisance (kept bugginghim/her until he/she did what I wanted). 43. Repeatedly reminded him or her about what I wanted. 54. Expressed my anger verbally. 41. Had a showdown in which I confronted him or her face to face. 30. Pointed out that the rules required that he or she comply.
Ingratiation
46. Made him or her feel important ("only you
have the brains, talent to do this").
9. Acted very humbly to him or her while mak-
25 -01 -02 -02 27 32 25 -13 22 18 15
59 52 51 51 51 50 47 45 43 38 38
13 -02 11 19 14 09 17 21 12 15 15
15 -05 -01 10 20 22 05 -05 06 01 13
04 14 16 17 07 06 16 16 11 14 -04
09 07 04 03 06 12 15 -06 07 07 07
-17 07 -03 -05 -04 -09 05 04 -21 17 00
09 -02 -07 -04 05 19 12 21 11 02 23
ing my request.
17. Acted in a friendly manner prior to asking for
what I wanted.
28. Made him or her feel good about me before
making my request.
37. Inflated the importance of what I wanted him or her to do. 36. Praised him or her. 3. Sympathized with him/her about the added
problems that my request has caused.
44. Waited until he or she appeared in a receptive
mood before asking.
10. Showed my need for their help. 29. Asked in a polite way. 22. Pretended I was letting him or her decide
to do what I wanted (act in a pseudo-democratic fashion).
Rationality
40. Wrote a detailed plan that justified my ideas. -06 38. Presented him or her with information in support of my point of view. 06 31. Explained the reasons for my request. 17 13. Used logic to convince him or her. 10 24. Wrote a memo that described what I wanted. 15 42. Offered to compromise over the issue (I gave in a little). 14 16. Demonstrated my competence to him or her before making my request. -10 -04 17 20 13 03 23 23 65 64 57 55 42 40 35 06 04 -03 02 04 00 10 -02 05 08 03 00 23 04 21 -07 01 -02 26 00 02 -17 -15 07 -03 00 04 -11 61 37 49 32 31 47 17
446
Table 4 (continued)
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Factors Dimension of influence/tactic
1 2 3 4 5 6 7" 8"
Sanctions
49. Gave no salary increase or prevented the per-
son from getting a pay raise.
26. Threatened his or her job security (e.g., hint
17 26 06 24 11
03 05 11 09 00
06 03 07 01 02
63 59 52 52 46
01 02 11 -04 07
01 08 01 13 11
02 04 -05 02 04
08 11 15 15 -03
of firing or getting him or her fired).
15. Promised (or gave) a salary increase. 6. Threatened to give him or her an unsatisfac-
tory performance evaluation.
34. Threatened him or her withloss of promotion. 35. Offered an exchange (e.g. , if you do this for
Exchange me, I will do something for you).
27. Reminded him or her of past favors that I 03 07 21 19 02 01 03 04 70 53 08 03 22 17 06 -15
did for them.
50. Offered to make a personal sacrifice if he
or she would do what I wanted (e.g. , work late, work harder, do his/her share of the work, etc). 55. Did personal favors for him or her. 7. Offered to help if he/she would do what I wanted.
58. Made a formal appeal to higher levels to back
-07 04 23
28 37 33
10 03 07
10 07 10
52 48 40
05 03 20
10 15 -12
15 09 14
Upward appeal up my request.
20. Obtained the informal support of higher-ups. 20 25. Filed a report about the other person with 10 17 28 12 26 03 09 09 08 06 07 -01 06 31 13 11 05 04 05 59 47 42 40 29 -05 -02 -03 14 23 16 05
higher-ups (e.g., my superior).
33. Sent him or her to my superior.
Blocking
47. Threatened to notify an outside agency if he
or she did not give in to my request.
48. Threatened to stop working with him or her
17 10 02 05 06
-12 -11 03 03 19
04 02 -08 -14 04
19 16 03 08 06
12 20 07 06 13
19 07 16 01 10
80 63 59 37 33
10 -03 07 -08 -02
until he or she gave in.
4. Engaged in a work slowdown until he or she
did what I wanted.
5. Ignored him or her and/or stopped being
friendly.
14. Distorted or lied about reasons he or she
should do what I wanted.
12. Obtained the support of co-workers to back
Coalitions up my request.
56. Had him or her come to a formal conference 06 25 06 17 04 16 35 34 35 -13 15 11 02 02 -06 29 35 18 28 13 20 50 47 46
at which I made my request.
32. Obtained the support of my subordinates to
back up my request.
52. Kept kidding him or her until they did what
Unclassified items I wanted.
23. Ignored him or her and went ahead and did 33 10 34 11 13 00 -10 -08 25 11 -02 01 -01 16 -09 05
what I wanted.
INFLUENCE TACTICS
447
Table 4 (continued)
Factors Dimension of influence/tactic Unclassified items 8. Provided him or her with various benefits that they wanted. 1. Challenged his or her ability (' 'I bet you can't do that"). 21. Pretended not to understand what needed to be done so that he or she would volunteer
to do it for me.
07 17 02 03
22 13 24 22 6.5
17 -01 -04 13 5.2
25 13 05 07 4.2
28 07 10 -01 3.2
07 09 05 10 2.9
08 03 18 24 5.7
19 07 16 12 5.0
2. Concealed some of my reasons for trying to influence him/her. % of total variance
18.2
t>tote. N = 754. Decimals omitted. Item numbers denote the items' original position in the questionnaire. a Data based on factor analysis of tactics used to influence superiors. b Data based on factor analysis of tactics used to influence subordinates.
Factor 1 is identified by highest loadings on the influence tactics, including demanding, ordering, and setting deadlines. This factor is labeled Assertiveness. The factor emerged as a dimension of influence at all target status levels (superior, coworker, and subordinate). Factor 2 is described by the highest loadings on weak and nonobtrusive influence tactics. Included here were such tactics as "acting humble" and "making the other person feel important." This factor is labeled Ingratiation. The factor emerged as a dimension of influence at all levels of target status. Factor 3 is characterized by loadings on the use of rationality influence tactics and is labeled Rationality. It includes such tactics as "writing a detailed plan" and "explaining the reasons for my request." This factor emerged at each target status level. In the analysis of tactics directed toward subordinates, however, additional items that involved group pressure also loaded highly on this factor. These latter items will be discussed subsequently. Factor 4 involved the use of administrative sanctions to induce compliance. Tactics with high loadings included "prevented salary increases" and "threatened job security." This factor is labeled Sanctions. It emerged as a dimension of influence at all levels of target status. Factor 5 loaded on tactics involving the
exchange of positive benefits. Included here were such tactics as "offering an exchange" and "offering to make personal sacrifices." This factor is labeled Exchange of Benefits. This factor only emerged in the factor analysis of influence tactics directed toward superiors. Factor 6 is described by loadings on tactics that bring additional pressure for conformity on the target by invoking the influence of higher levels in the organization. Included here were such tactics as "making a formal appeal to higher levels" and "obtaining the informal support of higher-ups." This factor only emerged in the factor analysis of influence tactics directed toward superiors; it is labeled Upward Appeal. Two additional factors (Factors 7 and 8) did not emerge in the overall factor analysis but were found in the subanalyses. It was decided to retain these factors for heuristic purposes. Factor 7 emerged in the factor analysis of influence directed toward superiors. Items that loaded on this factor included "engaging in a work slowdown" and "threatening to stop working with the target person." Essentially, these tactics are attempts to stop the target person from carrying out some action by various kinds of blocking tactics. This factor is labeled Blocking. Factor 8 emerged from the factor analysis of tactics directed toward subordinates.
448
Table 5
D. KIPNIS, S. SCHMIDT, AND I.WILKINSON
Intercorrelation of Scale Scores Dimension of influence
23 8
1. Assertiveness .22 .32 .24 2. Ingratiation .14 .24 3. Sanctions .13 4. Rationality 5. Exchange 6. Upward appeal 7. Blocking 8. Coalitions Note. N = 754.
.26 .49 .24 .20
.35 .21 .28 .1.0 .32 .21 .22 -.03 .27 .22 .20
.28 .31 .17 .32 .29 .36 .16
The remaining factors show reasonably low levels of intercorrelation, with rs ranging from .03 to .36. It should be noted that the individual correlation matrices at each status level closely paralleled the averaged findings reported in Table 5. Table 6 presents the reliability of the eight scale scores (coefficient alpha) for each target status level. With the exception of low reliability for the use of Blockingtactics with subordinates and co-workers, the remaining factor scores have satisfactory reliabilities. Correlates of Influence Tactics Target status. Table 7 shows the means and standard deviations of each influence dimension as a function of the status of the target. Seven of the eight dimensions were significantly associated with the relative status of the target. Basically, the findings suggest that as the status of the target person increased, respondents placed more reliance on rationality tactics. An analysis using Duncan's multiple-range test found that Assertive tactics and Sanctions were used more often to influence subordinates than co-workers or superiors (ps < .01). The tactics of Ingratiation, Exchange of Benefits, and Upward Appeal were used with equal frequency among subordinates and coworkers but significantly less often when attempting to influence superiors (p < .01). Finally, respondents reported that they used Rationality tactics more frequently to convince superiors than co-workers or subordinates (p < .01). As will be shown later, these differences in choice of tactics in part reflect the fact that respondents have different reasons for influencing target persons at different status levels. Goals. It will be recalled that respondents in the present study rated the importance of five possible reasons for exercising influence. To determine whether different combinations of tactics were used as the reasons for influencing varied, a set of stepwise regression analyses was performed. These stepwise regression analyses examined the relation between the eight tactic scores and the rated importance of each reason for
2
Items in this factor were part of the previously described factor Rationality. However, this subset of items described the use of steady pressure for compliance by "obtaining the support of co-workers" and by "obtaining the support of subordinates." This is labeled Coalitions. Scale construction. To aid further analysis, scales were constructed whose items were selected to represent each of the eight dimensions of influence.2 Selection was made on the basis of two criteria. First, items were selected that loaded over .40 on agiven dimension and did not load above .25 on any of the remaining dimensions. Second, from the pool of items that were selected to represent each dimension, items were selected based on an examination of each item's correlation with other items representing the factor and their correlations with items in the remaining factors. High item intercorrelation within a factor and low item intercorrelation with the remaining items were used as the final criteria for selecting items. To determine the intercorrelation between the eight scales, correlation matrices were first computed at each of the three status levels. Next, all intercorrelations were converted to their z' equivalents, averaged, and then reconverted to productmoment correlations. This procedure provides an estimate of the correlation between the eight scales, with the variance attributable to target status partialed out. As shown in Table 5, there was a fair degree of independence among the factors, with the exception of the relationship between Ingratiation and Exchange tactics (r = .49).
Copiesof the scales are available from the authors.
INFLUENCE TACTICS
449
Table 6 Reliability of Tactic Scores (Alpha
Number of items 6 4 5 5 5 4 3 2
Coefficient)
Target status Subordinate .65 .61 .61 .69 .70 .65 .47 .71 Co-worker .73 .77 .68 .71 .74 .65 .42 .64 Superior .71 .70 .65 .54 .76 .68 .74 .81 Total .70 .71 .78 .73 .73 .67 .53 .75
of influence Ingratiation Rationality Assertiveness Sanction Exchange Upward appeal Blocking Coalitions
exercising influence. A restriction put on the analysis was that only tactics that correlated significantly (p < .05) with a given reason were entered into the regression analysis. Since we have already found that the type of tactic used and the reasons for exercising influence both varied with the status of the target person, separate regression analyses were carried out at each target status level.
Table 8 shows these findings in terms of the particular combination of influence tactics that best predicted each reason for exercising influence. In addition, Table 8 presents the multiple correlations of these influence tactics with each reason for exercising influence. Table 8 shows that at all target status levels, the respondent's choice of influence
Table 7 Target Status and Average Frequency of Tactic Use
Target status Dimension of influence Ingratiation Subordinate 15.92b 3.97 14.04b 2.47 12.04a 3.00 6.39a 1.95 9.83b 3.10 7.05b 2.31 3.31a 0.82 4.37 2.04 Co-worker 16.08b 4.73 13.74b 3.20 8.06b 2.67 5.27b 0.91 10.08b 3.65 6.83b 2.29 3.50b 1.04 4.30 2.05 Superior 14.52a 4.48 14.72a 3.01 6.94C 2.25 5.16b 0.62 8.56a 3.53 5.43a 1.99 3.29a 0.91 4.53 2.14
F
16.69** 7.17** 243.91** 68.18** 13.52** 37.16** 4.05* .079
M
SD Rationality
M
SD Assertiveness M SD Sanctions M SD Exchange
M
SD Upward appeal M SD Blocking M SD Coalitions
M
SD
Note. Groups with different subscripts differ beyond the .05 level. * p < .05. **p < .01.
450
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Table 8 Multiple Correlations Between Tactics and Rated Importance of Reasons for Exercising Influence
Reason for exercising influence/target Receive assistance on own job boss co-worker subordinate Assign work to target boss co-worker subordinate Obtain benefits from target boss co-worker subordinate Improve target's performance boss co-worker Tactic category used Ingratiation Ingratiation Assertiveness Ingratiation Assertiveness Assertiveness Assertiveness Exchange Ingratiation Exchange Blocking Ingratiation Assertiveness Coalitions Assertiveness Blocking Rationality Assertiveness Exchange Coalitions Rationality Assertiveness Rationality Rationality Coalition Ingratiation Exchange Rationality Coalitions Exchange Assertiveness Rationality .20 .16 .29 .51 .28 .31
.38
.32 .21
.43
.46 .34
subordinate Initiate change boss
.57 .45 .42
co-worker subordinate
Note. Tactics are listed in order of their entry into the stepwise multiple regression equations. Allcorrelations are significant beyond the .05level.
tactics varied with the respondents' reasons for exercising influence. That is, respondents who frequently sought personal assistance from target persons used Ingratiation tactics; respondents who frequently assigned work to target persons used Assertiveness; and respondents who frequently tried to improve a target person's performance used Assertiveness and Rationality tactics. Fi-
nally, respondents who frequently tried to convince target persons to accept new ideas used Rationality tactics. Another finding of interest in Table 8 is that respondents showed the least variation in choice of tactics when attempting to influence their subordinates. No matter what the reason for influencing subordinates, the use of Assertiveness was associated with each of the five reasons and accounted for the most variance for each reason. In contrast, when influencing co-workers and superiors, the use of Assertiveness was associated only with two of the five reasons. Other situational factors. So far it has been shown that use of the eight dimensions of influence was associated with the relative power of the target person and the reasons why the respondents wanted to influence the target person. In this section we examine the relationship between the use of tactics and five personal or situational characteristics of the respondent: sex, level in the organization, whether the respondent's unit was unionized, number of people in respondent's work unit, and sex of respondent's boss. Thefindingsshowed that the respondent's own level in the organization was closely associated with use of influence tactics. Compared to those with low job status, respondents with higher job status reported greater use of Rationality and Assertiveness tactics when influencing both their subordinates and their superiors (ps < .01). In addition, respondents with higherjob status used Sanctions more frequently and sought aid from their superiors less frequently when influencing their subordinates (ps < .01). Thus, as the respondents' own job status rose, they were more likely to use more direct tactics of influence and be less dependent upon superiors. Size of work unit also related to the use of tactics on subordinates. In large work units, respondents more frequently used Assertiveness (p < .01), Sanctions (p < .01), and Upward Appeal (p < .05) when influencing subordinates. These findings are consistent with the general idea that as the number of persons in a work unit increases, a greater reliance is placed on strong and impersonal means of control.
INFLUENCE TACTICS
451
Finally, the presence of unions was as- influence tactics among peers and bosses. sociated with the use of certain tactics. If the Hence, the range of influence tactics used in organization was unionized, respondents that early research covered fewer areas than were more likely to use Ingratiating tactics those included here. to influencesubordinates (p < .01), to avoid Other factorial studies of influence tactics the use of Assertiveness when influencing have focused on dimensions of influence in co-workers (p < .05), and to use Ration- interpersonal settings (Falbo, 1977;Kipnis, ality tactics less frequently (p < .01) and Cohen, & Catalano, Note 2). These studies Blocking tactics more frequently (p < .01) have uncovered only two or three dimenwhen influencing bosses. sions of influence. For example, Kipnis et There were no significant relations as- al. reported that dating couples used three sociated with the sex of the respondent and dimensions of influence—strong, weak, and the sex of the respondent's boss in terms of rational—as ways of changing their dating the frequency of use of the eight dimensions partner's behavior. The items loading on of influence. Thus, in the present study men these three dimensions closely resemble and women chose similar tactics when at- the dimensions we have labeled Assertivetempting to get their way. ness, Rationality, and Ingratiation tactics. Thus, instead of acting "more loving" in order to influence, organizational members Discussion act "humble" and offer to "make sacriIn Study 1, influence tactics were sorted fices." The parallel is clearly there. The into 14 categories, but we suspected that a additional factors found in the present study more parsimonious classification system suggest that organizational resources exwas likely to exist. The results of Study 2 tend the range of tactics that organizational yielded eight dimensions of influence. members can use. Thus, organizational Four of these dimensions emerged at all members can invoke formal sanctions or the status levels. These dimensions were As- added authority of higher management as sertiveness, Sanctions, Ingratiation, and compliance-gaining strategies. Rationality. The remaining dimensions were It is recognized, of course, that findings uniquely associated with influencing either presented here are based on self-reportmeasuperiors or subordinates. The tactics of sures and require replication using different Exchange of Benefits, Blocking, and Up- methodologies. Nevertheless, the picture ward Appeal emerged when respondents de- portrayed by these data is of organizational scribed how they influenced their bosses. members actively seeking to influence peers, These dimensions did not appear in the anal- superiors, and subordinates for a variety of ysis of influence tactics used among sub- reasons, some personal and some based on ordinates. Finally, the use of Coalitions their management roles. This picture suptactics appeared only when respondents de- ports the view of those who argue that scribed how they influenced their sub- organizational leadership is more compliordinates. cated than is represented in organizational Unfortunately, there are few studies avail- behavior textbooks (Kochan, Schmidt, & able with which to compare the present DeCotiis, 1976). Such texts mainly focus on findings. Within organizational settings, the the ways in which higher levels in the organimajor empirical studies of influence tactics zation influence lower levels. This is called have been based on the Ohio State-Navy the leadership process. In fact, we would leadership studies (Fleishman, 1973). These suggest that everyone is influencing everystudies could identify only two dimensions one else in organizations, regardless of job of influence—consideration and initiating title. People seek benefits, information, satstructure. However, this early program of isfactory job performance, the chance to do research was only concerned with how better than others, to be left alone, coleaders exercised influence over subordi- operation, and many other outcomes too nates and was not concerned with the use of numerous to mention.
452
D. KIPNIS, S. SCHMIDT, AND I. WILKINSON
Eastern Psychological Association, Philadelphia, In fact, there may be very little difference April 1979. in the frequency with which people try to influence their bosses, co-workers, and subReferences ordinates, given all these various reasons for trying to influence others. What shows re- Cartwright, D. Influence, leadership and control. In markable variation, however, are the kinds J. G. March (Ed.), Handbook of organizations. of tactics that are chosen when trying to Etzioni, A:. Rand McNally,dimensions and their inter-
obtain these various outcomes. The present article has shown that these tactics vary with the particular wants of the influencing agent
1965. Organizational relationship. In B. Indik & F. K. Berrien (Eds.), People, groups and organizations. New York: Teachers College Press, 1968.
Chicago
and his or her degree of control over the Falbo, T. .The multidimensional scalingSocialpower of strategies Journal of Personality and Psytarget of influence. The implications of chology, 1977, 35, 537-547. these findings for the understanding of or- Fleishman, E. A. Twenty years of consideration and structure. In E. A. Fleishman & J. G. Hunt (Eds.), ganizational politics remain to be explored. Current developments in the study of leadership. As a final point, the scales that have been Carbondale: Southern Illinois University Press, developed to measure the various dimensions of influence can be useful in further French., J. R. P., & Raven, B. H. The bases of social 1973 research. Even though the use of influence power. In D. Cartwright (Ed.), Studies in social within organizations is a topic of considerpower. Ann Arbor: University of Michigan Press, able interest, there have been few systematic Izraeli ., D. N. The middle manager and the tactics of
1959 power expansion. Sloan Management Review, 1975, attempts to develop a means of measuring 16, 57-70. such behavior. The present dimensions provide the potential for profiling the use of in- Kochan, T. A., Schmidt, S. M., & DeCotiis, T. A. fluence in a variety of organizational settings Superior-subordinate relations: Leadership and and at different levels as well as between Michener, .A.Human Relations,, 1976,Likin279-294. -
various groups. Reference Notes
1. Goodchild, J. D., Quadrado, C., & Raven, B. H. Getting one's way. Paper presented at the meeting of the Western Psychological Association, Sacramento, California, April 1975. 2. Kipnis, D., Cohen, E., & Catalano, R. Power and affection. Paper presented at the meeting of the
28, , & Schwertfeger M. g as a deter minant of power tactic preference. Sociometry, \972,35, 190-202. Raven, B. H. The comparative analysis of power and influence. In J. T. Tedeschi (Ed.), Perspective on social power. Chicago: Aldine, 1974. Schein, V. E. Individual power and political behavior. Academy of Management Review, 1977, 2, 64-72.
headship
Received July 2, 1979
doc_272223178.docx