The study of styles of handling interpersonal conflict has increased in importance in recent decades. Among different approaches carried out to distinguish and classify styles - see Blake & Mouton, 1964; Pruitt, 1983; Rahim & Bonoma, 1979; Thomas, 1976, 1992; Tjosvold, 1989; Van de Vliert & Prein, 1989 and Van de Vliert & Hordijk, 1989 -, empirical evidence shows an important support to GRID’s five style of conflict handling behavior approach (Ruble & Thomas, 1976; Rahim, 1983a; Van de Vliert & Kabanoff, 1990). This category, carried out by Thomas (1976, 1992) and Rahim (1992), is based on a two-dimensional approach - attempt to satisfay one’s own concerns and attempt to satisfy the concerns of the other person. As is shown in Figure 1, a combination of the dimensions results in grid’s five different conflict handling styles: integrating, compromising, obliging, dominating and avoiding.

The relations between these styles of conflict management have been contrasted using self-reporting questionnaires -with satisfactory levels of reliability- that were specifically designed to reflect the five styles of conflict management. Studies carried out in Spain on this subject (Munduate, Ganaza, Alcaide & Peiró, 1994), using the adaption of ROCCI-II (Rahim, 1983b), have shown that integrating is the style most widely used by Spanish managers, followed by compromising and avoiding. Obliging and dominating feature as the least used.

The findings concerning the influence of the hierarchical position of the other party, indicate that the style of conflict management adopted by the speaker varies depending on whether the conflict is with a superior, a subordinate or a peer. It has been shown that subjects adopt a style of domination in the resolution of differences with subordinates (Philips and Cheston, 1979; Lee, 1990); while they adopt a compromising style when both parties in a conflict situation hold a similar share of power -among peers, for example- (Lee, 1990; Philips & Cheston, 1979; Rahim, 1983a; 1986). Finally, subordinates tend to adopt an obliging style more in cases of confrontation with a superior than in those with peers or subordinates (Drake, Zammuto and Parasuraman, 1982; Munduate, Ganaza & Alcaide, 1993).

Research Objectives and Hypothesis

Lee proposes (1990) the need for empirical studies which analyze the actual styles of conflict management, instead of analyzing the perception that subjects have about their styles of conflict management through retrospective field studies, as has been the case in most of research to date. In fact, of the studies in print, only Lee’s (1990) establishes the styles used by managers in their relations with superiors, peers and subordinates by observing their behavior in the moment of confrontation. A different approach to that adopted through self-reported questionnaires -in spite of obtaining retrospective data-, is that adopted by Kabanoff & Van de Vliert (1993). They have used managers’ reports about their interventions in real conflict situations which are then evaluated and classified by coders. Cosier & Ruble’s (1981) study also moves away from self-reported questionnaires using an experimental game with a computer program.

The principal objective of the present study is to establish whether the findings about the styles of conflict management are confirmed using a different methodology. The first specific objective is to establish whether managers are more inclined to use the style of integration, followed by compromise and avoidance, in their conflict interactions in organizations.

Considering the interactive focus for the analysis of conflict situations put forward by Kabanoff & Van de Vliert (1993), and drawing on research into processes of social influence (Moscovici, 1976; Mugny & Doise, 1978), we have picked up on the design study used by Lee (1990). This approaches from the standpoint that in interactions of conflict with a superior, a peer or a subordinate, one of the factors that influences the style of conflict management adopted by the manager is the tactic of influence used by the other party. In other words, the subject tends to adopt a different tactic depending on whether she/he is dealing with a boss, a peer or a subordinate, and this in turn affects the style of conflict management adopted by the other party (Lee, 1990). Linked to the results of the effects of hierarchical positions on social influence (Lemaine, Lasch & Ricateau, 1971-1972; Sherif & Sherif, 1969), Kipnis, Schmidt & Wilkinson (1980) point out that subjects tend to use different tactics of influence depending on the relative status of the other party: superiors will tend to use assertive tactics, peers, blocking tactics, and subordinates, rational tactics.

The second objective of the present study is to analyze the style of conflict management used by managers, depending on the relative status and the tactics of influence of the other party, with the following predictions:

- Hypothesis 1: differences will be found in managers’ use of styles of conflict management depending on whether they are dealing with superiors, peers or subordinates, in the following direction: there will be a tendency to use an obliging style more with superiors than with peers or subordinates; that of compromise more with peers than with subordinates or superiors and, finally, dominating more with subordinates than with peers or superiors.

- Hypothesis 2: differences will be found in managers’ use of styles of conflict management, depending on the tactics of influence adopted by the other party in the following direction: there will be a tendency towards using the obliging style more when the other party uses assertive tactics; compromise when the other uses blocking; and dominating when the other turns to rational arguments.

- Hypothesis 3: Differences will be found in managers’ use of styles of conflict management depending on who s/he is dealing with -a superior, peer or subordinate- and the tactics of influence used by that person -assertive, blocking or rational-, in the following direction: they will tend to use the obliging style more when the superior uses assertive tactics; the compromising style more when the peer uses tactics of blocking, and dominating more when the subordinate turns to rational arguments.

Method

Sample. From a class of subjects attending a post-graduate managerial course, 45 subjects were selected (27 men and 18 women, with an average age of 31). The subjects’ jobs involved them in tasks of a managerial nature, or their curriculum was one which would lead to them assuming responsibility, as this was one of the criteria for them being accepted on the courses.

Experimental design. A 3x3 factorial design was used with the following intergroup variables: Relative status of the manager and Tactics of influence used by the managers. The first of the variables has three dimensions: Relations with superiors vs. Relations with peers vs. Relations with subordinates. The second independent variable has three dimensions: Assertive tactics vs. Blocking tactics vs. Rational tactics. The dependent variable is the style of conflict management adopted in each experimental condition, according to the following five dimensions: integrating, compromising, obliging, dominating, and avioding.

Procedure. The 45 subjects are randomly assigned to one of the groups of six subjects, from the nine experimental conditions. Each group is made up of five experimental subjects and an accomplice. The group is asked to carry out an exercise, and each subject is given a role which specifies the department they are going to represent as well as the position they hold within it. General information is also given about the organization and the aims of the meeting which is the same for all the participants. They are given twenty minutes to prepare the exercise during which time they are not allowed to exchange information. The accomplice subject is treated in the same way as the rest of the members of the group so as not to arouse suspicion, but his position at the table has been previously agreed so as to give him the planned role.

As the exercise unfolds a conflict situation develops. The exercise consists of the allocation of a half-yearly budget in which each member assumes the managerial role of a certain department and has to negotiate with the other subjects -acting as managers of other departments- the quantity corresponding to each one. They must reach an agreement within the 40 minutes allowed for the meeting. The accomplice provokes the conflict situation by requesting an excessive proportion of the overall budget. In each of the nine experimental situations the accomplice is assigned a different managerial role - high-ranking, low-ranking, or same hierarchical level-, and adopts different tactics of influence -assertive, blocking, and rational. The meetings of the nine experimental situations are filmed from the control room for subsequent evaluation of the variables.

Evaluation of the dependent variable. Two coders carried out structured observations of the videos. The categorial system used to codify the interactions corresponds to the five styles of conflict management, and also includes a sixth style labelled miscellaneous which caters for those behaviors that cannot be included in the previous categories and that will be left out of later analyses. The units of observation selected for the codifying of the interactions are the intervention turns. These turns are the utterances that the subjects make during the sessions. Each intervention turn must be codified into one of the categories mentioned. The aspects of the intervention turns that are recorded have a mainly semantic content while not ruling out the possibility of recording non-verbal aspects when these clearly substitute utterances.

The intercoder reliability of 0.91, calculated using Cohen’s Kappa coefficient (1968) is very satisfactory. The disagreements in codification of the categories of analysis are resolved by going back over the videos, discussing them and agreeing on a codification. To settle differences in the event of a disagreement persisting, the codification is carried out by a third observer. Every four intervention turns the observers compare to see whether they have been codifying the same intervention turn -by comparing the first and last phrase of the turn codified. The disagreements over intervention turns are sorted out at that point by the observers.

The success of the experimental manipulation. To test whether the accomplices have correctly fulfilled their roles as superiors, peers and subordinates, observations are made of these behaviors by the judges with coefficients being obtained of 0.93, 0.93, and 0.94 respectively. The same procedure is repeated for the tactics of influence used by the accomplices, with coefficients being obtained of 0.95, 0.85 and 0.87 for rational, blocking and assertive tactics, respectively. In short, the success of the experimental manipulation for both variables is confirmed.

Results

Although Table 1 confirms the greater overall use of the integration style, the highlight of the findings in this paper is the use of domination in second place. There is then a large distance to the style of compromise which, according to findings in previous studies, should have been in second place.

As for Hypothesis 1, concerning the incidence of relative hierarchical status of the other party on the style of conflict management, there is confirmation (see Table 2) of the following: the use of integration more often with superiors than peers; compromise more with peers than superiors; and domination more with subordinates and peers than superiors. There is nothing to show that an obliging style is used more with superiors.

In the case of Hypothesis 2 about the incidence of the tactics of influence used by the other party on the style of conflict management (see Table 3), there is only confirmation of the greater use of domination when the other turns to rational arguments.

The predictions in Hypothesis 3 about the interactive effect of both variables are not confirmed (see Table 4).

Discussion

The data about the greater use of the style of integration confirms the tendency observed in managers in recent years of addressing the interests of other parties as well as their own, seeking new and improved alternatives for both sides (Munduate, Ganaza, Alcaide and Peiró, 1994; Rahim, 1992). However, what stands out is that domination is the style least used by Spanish managers according to the self-reporting questionnaires (García-Echevarría, 1991; Munduate, Ganaza & Alcaide, 1993; Osorio, 1992; Serrano & Remeseiro, 1987; Vidal Abascal, 1991), and the second to last style used by American managers (Rahim, 1992), while in this paper it appears as the second most widely used style. This seems to indicate a difference between the findings previously obtained through managers’ reports about their styles of conflict management, and those obtained through the observation of their behavior in conflict interactions.

The differences found concerning the use of different styles depending on the relative status of the other party, confirm data from Kabanoff & Van de Vliert (1993) and Mannix, Thompson & Bazerman (1989) about the influence of relative power. They state that as the imbalance in the power relations grows, the likelihood of mutual cooperation in the resolution of the conflict decreases. In fact, we have observed that although managers are seen to be more integrating with superiors than with peers, and have a greater tendency towards compromise with peers than with superiors, they do, however, tend to use a dominating style with subordinates and peers.

As a conclusion of the present paper, we should highlight the significant differences between managers’ perception of the way they act in conflict interactions, and their actual behavior. One explanation for this could lie in the behavior-attitude inconsistency described by LaPiere back in 1934, and which has emerged as the difference between the cognitive, affective, and behavioral components of attitudes. The cultural and social changes which Spanish society has undergone in recent years may explain the development of styles of conflict management on a cognitive level, but not on a behavioral level. A study carried out by Porat (1970), to analyce the influence of cultural differences in the way executives of various countries handle conflict, indicated that when Spanish executives of the beginning of the seventies reached agreements, they did so in shorter times than Danes, Swiss or British, but that Spanish executives tended to lead negotiation to a zero-sum situation in which both parties lost. They adopted hard and inflexible tactics, higly coercive, when they could not arrive at an agreement in the first part of the process. Although the data in the present paper about the presence of integration and domination as the most widely used styles tie in with the tendency described by Porat (1970), the temporal phases of the process have not been analyzed. As Haire, Ghiselli & Porter (1963) pointed out in their study about managerial styles in different countries, we see the absence of a link between the ideas about managerial practices -that tend to be based around persuasion and participation- and basic beliefs about human nature -that tend to remain traditional and unchanging. The authors point out that this paradox between a basic lack of confidence in other people and at the same time a tendency towards group-based participative styles may be showing the effect of a partial acceptance of modern managerial concepts (Munduate, Ganaza, Alcaide & Peiró, 1994). It may be possible to stretch this explanation to cover the discrepancies between the descriptions that the managers gave about their styles of conflict management, and the actual styles of conflict management they were seen to adopt. In fact, although the use of bargaining styles similar to integration and compromise have been present in studies on the subject in recent years (García-Echevarría, 1991; Munduate, Ganaza & Alcaide, 1993; Osorio, 1992; Serrano & Remeseiro, 1987; Vidal Abascal, 1991), the data obtained in the present study by using an observational methodology show that managers continue to fall back on pressure styles on more occasions than they are willing to admit.

Agradecimientos

This research was supported by a grant from the Spanish CICYT (Comisión Interministerial de Ciencia y Tecnología)(PB91-0955).
Paper presented at the 8th annual conference af the International Association of Conflict Mamagemente, Elsinore, Denmark, June 11-14, 1995.