Why managers prefer some influence tactics to other tactics: a net utility explanation.
Many studies of influence tactics focus on descriptive research questions: identification and categorization of the most frequently used tactics, sometimes including the identification of the consequences of these tactics. Recently, researchers also developed an interest in the determinants of the use of influence tactics. It has been found, for example, that the use of influence tactics covaries with the direction of the influence attempt (Deluga & Perry, 1991; Kipnis et al., 1980; Yukl & Falbe, 1990; Yukl & Tracey, 1992). Personality factors (e.g. Machiavellianism) also seem to be important (Grams & Rogers, 1990). In addition, it has been found that the use of influence tactics is related to factors such as self-esteem (Raven, 1992), status (Stahelski & Paynton, 1995), leadership style (Deluga & Souza, 1991), one-to-one or group situations (Guerin, 1995), organizational culture (Steensma, Jansen, & Vonk, 2003), expectation of future interaction (Van Knippenberg & Steensma, 2003) and the various objectives of influence attempts (Yukl, Guinan, & Sottolano, 1995).
Several studies demonstrate that variables covary with influence tactics used by people. The point has been reached now at which it becomes important to explain the 'why' of the influencing behaviour. Which tactic will be used and why?
The current research was conducted to answer this explanatory question. The study is building on preliminary models proposed by Yukl and Tracey (1992) and on the general expected utility model (Edwards, 1955). Yukl and Tracey assumed that managers would prefer to use tactics that are not costly in relation to likely benefits. This cost-benefit explanation is the central tenet of the so-called expected utility model of preferences and behaviours. The expected utility model is widely used in social sciences (see, for example, Edwards, 1955; Feather, 1959). The core assumption of this model is that a person who has to make a choice, will prefer that alternative which has the highest subjective expected utility, i.e. the alternative which is likely to lead to the most favorable outcomes. Alternatives may refer to behavioural intentions, to actual behavior or to objects.
The subjective expected utility of a given alternative can be defined as:
SEU = [[summation].sup.n.sub.i=1][SP.sub.i][U.sub.i]
where SEU is the subjective expected utility associated with a given alternative; [SP.sub.i], the subjective probability that the choice of the alternative will lead to outcome i; [U.sub.i], subjective value (= utility) of outcome i; and n, the number of relevant outcomes (Edwards, 1955).
For SEU, the concept of 'net utility' will be used in the present study, i.e. expected 'costs' (negative outcomes) are subtracted from expected positive outcomes. Two predictions are tested in the present study:
(1) The higher the net utility of a given influence tactic is, the higher the 'real' (actual) frequency of using this tactic will be.
(2) the higher the net utility of a given tactic is, the higher the 'desired' (preferred) frequency of using this tactic.
A distinction is made between 'desired' (preferred) frequency and actual frequency since it is possible that situational constraints will hinder people from using their preferred influence tactic.
Context factors and personality variables may affect subjective probabilities and values of outcomes, and via this path the use of tactics. In this research note, attention will be paid only to the core constructs of the net utility model: outcomes, subjective probabilities and subjective values.
However, what outcomes should be studied? Based on Maslow's (1962) theory of needs, it was assumed that managers strive for outcomes such as control (to fulfill the need for security), good social contacts, (self-)esteem and self-actualization. Moreover, in a pilot study, in-depth interviews were held with 26 managers, to measure what their goals and expected outcomes were when using influence tactics. In all, this resulted in a list of 14 outcomes.
Questionnaires were sent to 300 managers. The response rate was 31.7% (N = 95). Seventy-three per cent of respondents were males; 53% were working in profit organizations, while 47% worked in non-profit organizations; 24.7% worked in a small-scale organization (N < 100), 38.7% were managers in a medium-sized organization (100 < N < 500) and 36.5% of respondents worked in large organizations. As for the management level, 46.2% were working as supervisors, 17.2% were working at the highest level and 36.5% were working at intermediate hierarchical levels. Most managers were in the age category 40-49 years (52%); 4% were younger than 30 years, 28% were 30-39 and 16% were older than 49 years.
The questionnaire is divided into sections: General information about respondent and organization, organizational structure, organizational culture, influence tactics and expected outcomes, evaluations of outcomes, leadership style, Machiavellianism, social desirability and special topics. The current study was part of a larger study. In this section, only that will be reported what is relevant to the current study.
Influence tactics and expected outcomes
The list of eight categories of tactics developed by Yukl and Falbe (1990) was used: pressure, upward appeals, exchange, coalition, ingratiation, rational persuasion, inspirational appeals and consultation. First, a tactic is described (for example: 'Pressure is the use of urgent demands and intimidation to motivate subordinates to accomplish their tasks'). Then it is asked how often the respondent uses this tactic (five-point scale, 1 = never, 5 = always). Also, the preferred or desired frequency is measured ('How often would you like to use [pressure]?' (five-point scale)). Next, 14 possible outcomes are described and for each outcome the subject is asked to indicate the subjective probability of the outcome. A sample item reads as follows: 'The use of pressure will bring to me:
* higher motivation of subordinates;
* resistance by subordinates;
* good personal contacts with my subordinates;
* (etc; in all, 14 outcomes).'
Subjective probabilities range from 'very low' (1) to 'very high probability' (5). In a separate part of the questionnaire, the subjective value of each outcome is measured on a five-point scale (very unimportant ... very important). Questions on the subjective value of outcomes were far apart from questions on preferred and actual use of influence tactics. For instance, preferred and actual use of pressure were measured by items 15 and 16, probabilities of outcomes of pressure by items 17-30, while subjective values of outcomes were measured by items 143-156.
Subjective expected (net) utility
The net utility of each separate influence tactic was measured using the SEU formula (Edwards, 1955).
Results and discussion
There were no meaningful relations between the background variables (age, etc.) of the managers and the variables under scrutiny.
Mean scores and standard deviations of the expected net utilities of the influence tactics are presented in Table 1. This also contains the means and standard deviations of the actual and the preferred frequency of using each influence tactic. Actual and preferred frequencies of use of rational persuasion and soft tactics (consultation, inspirational appeals) are higher than actual and preferred frequencies of exchange, and hard tactics such as upward appeals and pressure.
The eight means of expected utilities of influence tactics correlate as predicted with the means of the actual and the preferred use of tactics. In both cases, Spearman rank-order correlation coefficient [r.sub.s] = .81 (p < .05 two-tailed). So, at the level of mean scores the hypotheses are supported. However, there are large individual differences in expected utilities of tactics (and also in actual and preferred frequencies of use). Therefore, it is better to use all individual scores. Pearson correlations between net utilities of tactics and actual or desired frequencies of use of tactics are presented in Table 2.
The significance level was set at [alpha] = .01 for two reasons. Firstly, gathering data from the same source--the managers--could lead to the problem of common-method variance, resulting in artificially higher correlations. Secondly, there are many correlations to be tested and adoption of a strict criterion should reduce the probability of type I errors.
The pattern of correlations as presented in Table 2 is firmly supporting hypotheses 1 and 2. In almost all cases (15 of 16, i.e. 94%), net utility of an influence tactic correlates significantly (p < .01 or p < .001) with the (actual and preferred) frequency of use of that tactic. The only exception is the correlation between net utility of consultation and actual use of consultation: r = .26 (p < .02). In addition, there are only a few cases (12.5%, see Table 2) in which a net utility of a tactic correlates with the (actual or preferred) frequency of use of another tactic. Almost all exceptional cases can be explained, however, since there seems to be a connection between some tactics:
* Upward appeals are often used as a method of pressure. Moreover, upward appeals somehow seem to imply that the agent will form a coalition with high-ranking persons.
* Ingratiation may be considered to be a special case of exchange: one flatters a person in exchange for her cooperation.
* Rationality is valued in organizations; therefore, agents might use rationality to inspire their subordinates.
* Rational persuasion can benefit from consultation tactics, to be able to weigh one's arguments carefully.
* Inspirational tactics seem to work best where they can be connected with private wishes from persons--and such wishes can be brought out while consultation takes place.
Only the unexpected correlation between net utility of coalition tactics and preferred use of exchange tactics cannot be explained easily. It may be concluded that the subjective expected utility model seems to describe the process of selecting an influence tactic remarkably well.
However, survey studies have a methodological weakness. Common methods and self-reports to gather data enhance the risk that social desirability might distort the results. This problem does not seem to play a role in the present study. The scores of the managers on a social desirability scale had been computed. (1) Most correlations between social desirability and the variables under scrutiny in the present study were close to zero. Only one correlation was statistically significant: social desirability correlated with the reported use of inspiration (r = .35, p < .001).
Up to now, much influence tactics research has been rather exploratory. There have been explanatory studies, but often these studies focused on the content of determinants of influence tactics (e.g. culture, personality, situational determinants), and not on the process by which influence tactics are selected. The present study demonstrates the value of the general subjective expected utility model. This process theory can be applied to a broad range of choice situations, including situations in which influence tactics are applied. Subjective expected utility might play an important role in determining preferences and choices. It should be admitted, though, that even though the preferences are measured and explained satisfactorily in the present study, the actual use of tactics can more properly be ascertained by research methods which are less vulnerable to response errors. Self-reports of behaviour run the risks of rater bias.
Moreover, the important role of expected utility does not mean that other factors can be ignored. In particular, norms of fairness seem to be matters of the utmost weight to many people (Tyler & Lind, 1992). Future research should pay attention to the combined effects of utility and fairness. Future studies should also pay more attention to negative outcomes of influence tactics, to offer a more refined insight into the psychological process of determining preferences and decisions. Negative items could elicit different types of responses than positive items. Finally, future studies should also use alternative research methods. Experimental and longitudinal designs have a higher internal validity than cross-sectional surveys. Alternative explanations (e.g. an attributional explanation that managers believe that an influence tactic works best because they use that tactic) can be excluded better with such methods.
The author conveys his gratitude to Lisette Otto, Anneke de Rijk and Yvonne Spies, who participated as student members of the research team.
Received 21 February 2006; revised version received 27 April 2006
Blickle, G. (2000). Influence tactics used by subordinates: An empirical analysis of the Kipnis and Schmidt subscales. Psychological Reports, 86, 143-154.
Crowne, D., & Marlowe, D. (1964). The approval motive. New York, NY: Wiley.
Deluga, R. J., & Perry, J. T. (1991). The relationship of subordinate upward-influencing behavior, satisfaction, and perceived superior effectiveness with leader-member exchanges. Journal of Occupational Psychology, 64, 239-252.
Deluga, R.J., & Souza, J. (1991). The effect of transformational and transactional leadership styles on the influencing behavior of subordinate police officers. Journal of Occupational Psychology, 64, 49-55.
Edwards, W. (1955). The prediction of decisions among bets. Journal of Experimental Psychology, 50, 201-214.
Feather, N. T. (1959). Subjective probability and decision under uncertainty. Psychological Review, 66, 150-164.
Grams, W. C., & Rogers, R. W. (1990). Power and personality: Effects of Machiavellianism, need for approval, and motivation on use of influence tactics. Journal of General Psychology, 117(1), 71-82.
Guerin, B. (1995). Social influence in one-to-one and group situations: Predicting influence tactics from basic group processes. Journal of Social Psychology, 135, 371-385.
Kipnis, D., Schmidt, S. M., & Wilkinson, I. (1980). Intraorganizational influence tactics: Explorations in getting one's way. Journal of Applied Psychology, 65, 440-452.
Maslow, A. H. (1962). Toward the psychology of being. New York: Van Nostrand.
Raven, B. H. (1992). A power interaction model of interpersonal influence: French and Raven thirty years later. Journal of Behavior and Personality, 7, 217-244.
Schriesheim, C. A., & Hinkin, T. R. (1990). Influence tactics used by subordinates: A theoretical and empirical analysis and refinement of the Kipnis, Schmidt & Wilkinson subscales. Journal of Applied Psychology, 75, 246-257.
Stahelski, A. J., & Paynton, C. E (1995). The effects of status on choices of social power and influence strategies. Journal of Social Psychology, 135, 553-560.
Steensma, H., Jansen, S., & Vonk, C. (2003). Organizational culture and the use of influence tactics by managers. Journal of Collective Negotiations in the Public Sector, 30(1), 47-57.
Tyler, T. R., & Lind, E. A. (1992). A relational model of authority in groups. In M. Zanna (Ed.), Advances in experimental socialpsychology (Vol. 25, pp. 115-191). San Diego, CA: Academic Press.
Van Knippenberg, B., & Steensma, H. (2003). Future interaction expectation and the use of soft and hard influence tactics. Applied Psychology: An International Review, 52(1), 55-67.
Yukl, G., & Falbe, C. M. (1990). Influence tactics and objectives in upward, downward and lateral influence attempts. Journal of Applied Psychology, 75, 132-140.
Yukl, G., Guinan, P. J., & Sottolano, D. (1995). Influence tactics used for different objectives with subordinates, peers, and superiors. Group and Organization Management, 20, 272-296.
Yukl, G., & Tracey, J. B. (1992). Consequences of influence tactics used with subordinates, peers, and the boss. Journal of Applied Psychology, 77, 525-535.
(1) Social desirability was measured by combining the answers to 10 Yes-No items. Five of the items were adapted versions of items taken from the 33 items of the Social Desirability Scale developed by Crowne and Marlowe (1964). Five items were constructed specially for the present study. Mean score was 4.20, standard deviation = 1.897 (with 0 = low, 10 = high social desirability).
Herman Steensma *
Leiden University, The Netherlands
* Correspondence should be addressed to Herman Steensma, Department of Social and Organizational Psychology, Leiden University, PO Box 9555, 2300 RB Leiden, The Netherlands (e-mail: email@example.com).
Table 1. Influence tactics: Mean scores (M) and standard deviations (SD) of net utilities, frequencies of actual use and preferred frequencies; also included are the ranks of M Net utility Influence tactic M SD Rank of M Consultation 208.94 31.39 1 Inspirational appeals 207.23 33.27 2 Rational persuasion 197.36 37.77 3 Ingratiation 136.03 48.58 4 Coalition 132.71 44.33 5 Exchange 117.54 51.47 6 Upward appeals 98.79 34.78 7 Pressure 95.26 38.83 8 Frequency of use Influence tactic M SD Rank of M Consultation 3.76 .614 2 Inspirational appeals 3.70 .620 3 Rational persuasion 4.02 .483 1 Ingratiation 2.41 .940 5 Coalition 2.60 .868 4 Exchange 1.86 .911 8 Upward appeals 1.87 .662 7 Pressure 2.07 .672 6 Preferred frequency Influence tactic M SD Rank of M Consultation 3.99 .583 3 Inspirational appeals 4.10 .630 2 Rational persuasion 4.18 .645 1 Ingratiation 2.35 .907 5 Coalition 2.45 .893 4 Exchange 1.81 .893 7 Upward appeals 1.60 .612 8 Pressure 2.07 .836 6 Table 2. Correlations between net utilities of tactics and (actual and preferred) frequencies of use of tactics (N varies from 87 to 95) Net utilities of. Actual (A) and preferred (P) frequencies Pressure Upward appeals Pressure A .40 ** .31 * P .30 * .22 Upward appeals A .16 .36 ** P .10 .45 ** Exchange A .02 .22 P .15 .26 Coalition A .05 .29 * P .05 .28 * Ingratiation A .23 .22 P .27 .24 Rational persuasion A .05 .04 P .26 .11 Inspirational appeals A -.05 -.09 P .00 -.11 Consultation A -.15 -.25 P -.13 -.23 Net utilities of. Actual (A) and preferred (P) frequencies Exchange Coalition Pressure A .10 .19 P .07 .26 Upward appeals A .13 .20 P .22 .17 Exchange A .78 ** .18 P .78 ** .29 * Coalition A .18 .62 ** P .22 .62 ** Ingratiation A .16 .24 P .19 .23 Rational persuasion A .05 .17 P .09 .15 Inspirational appeals A -.04 -.08 P .04 -.02 Consultation A -.06 -.07 P .07 -.10 Net utilities of. Actual (A) and preferred (P) Rational frequencies Ingratiation persuas. Pressure A .19 .13 P -.01 -.02 Upward appeals A .13 -.11 P .05 -.22 Exchange A .34 ** -.04 P .36 ** .01 Coalition A .21 .09 P .16 -.08 Ingratiation A .70 ** -.04 P .66 ** -.07 Rational persuasion A -.02 .41 ** P .17 .58 ** Inspirational appeals A -.19 .32 * P -.07 .27 Consultation A -.11 .04 P -.16 .22 Net utilities of. Actual (A) and preferred (P) frequencies Inspir. appeals Consultation Pressure A .17 .04 P -.08 -.16 Upward appeals A -.13 -.19 P -.18 -.27 Exchange A .02 -.09 P .04 -.01 Coalition A .08 -.01 P .00 -.08 Ingratiation A .03 -.08 P -.09 -.19 Rational persuasion A .29 * .35 ** P .37 ** .35 ** Inspirational appeals A .50 ** .38 ** P .53 ** .30 * Consultation A .17 .26 P .32 * .47 *** * p < 0.01 two-tailed; ** p < .001, two tailed.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Short research note|
|Publication:||Journal of Occupational and Organizational Psychology|
|Article Type:||Author abstract|
|Date:||Jun 1, 2007|
|Previous Article:||Perceived similarity and complementarity as predictors of subjective person-organization fit.|