Paper Presented at the 6Th Annual Conference of The

REFERENCE:

Peeters, G. (June, 1993). Mapping implicit theories of interpersonal conflict: The roles of dissatisfaction and disagreement.

Paper presented at the 6th Annual Conference of the

International Association for Conflict Management. Houthalen (Belgium).

https://lirias.kuleuven.be/handle/123456789/196711

https://lirias.kuleuven.be/retrieve/17621/1993REPORT_CONFLICT.doc

MAPPING IMPLICIT THEORIES OF INTERPERSONAL CONFLICT:

THE ROLES OF DISSATISFACTION AND DISAGREEMENT

Guido Peeters

N.F.W.O. and Catholic University of Leuven

Laboratory of Experimental Social Psychology

Tiensestraat 102

B-3000 Leuven, Belgium

1993

Abstract

A method and data are presented on the mapping of subjective

implicit theories underlying individual perceptions of interpersonal

conflict. Building on a previously established "relational information

processing paradigm" an experimental questionnaire was developed

requiring subjects to rate how conflictive they considered hypothetical

situations in which two individuals showed varying patterns of

(dis)satisfactions. Data obtained form economy freshmen revealed three

major implicit theories: (a) mere dissatisfaction associating conflict

with overall dissatisfaction, (b) unilateral dissatisfaction

associating conflict with specific dissatisfaction of a key-person, and

(c) disagreement associating conflict with dissatisfaction with an

other who is satisfied with him/her-self. The selective use of implicit

theories did not vary significantly between judgments concerning

conflict in formal labour-situations and informal socio-emotional

situations but unilateral dissatisfaction was confined to conditions

where the identification of a key-person was facilitated by having

conflicting parties clearly differentiated. The results are discussed

in the light of the relational information processing paradigm in

social cognition and possible applications to conflict management.

MAPPING IMPLICIT THEORIES OF INTERPERSONAL CONFLICT:

THE ROLES OF DISSATISFACTION AND DISAGREEMENT

Which factors make that some interpersonal situations are

"conflictive" while others are not? Common sense as well as experts

have various and even contradictory answers to this question. Consider,

for instance, the role of "self-esteem". Some may argue that high

self-esteem stimulates conflict in that it implies high self-confidence

making opponents unwilling to compromise. Others, however, may stress

that people with positive self-esteem are more prone to a conciliatory

"beau geste" than are people with low self-esteem who may dig in their

heels by fear of loosing their last grain of self-respect. Similar

rationales, which are often advanced in overt discussions, are referred

to as explicit theories. However, it is also possible to infer concepts

or rationales from the pattern of judgements made by a subject whereby

the subject may not be aware of them. Similar concepts or rationales

have been referred to as implicit theories (e.g.: Wegner & Vallacher,

1977). They are not always consistent with current explicit theories

and it may happen that only after a subject has made up his or her mind

for a judgment, the fitting explicit theory is added as a socially

acceptable justification.

It may be evident that adequate conflict management may require not

only accurate knowledge of explicit theories of conflict but also of

the implicit theories applied by the involved parties including the

conflict manager himself. However, while Wegner and Wallacher

inventoried implicit theories about motivation, personality, abnormal

psychology, social relations, and the self as early as 1977, implicit

theory of conflict has not been a main focus of research as yet. Hence

in the present paper, a research paradigm is presented on the issue and

first data are reported.

Conflict: A Relational Construct

Wegner and Vallacher (1977) classify implicit theories relative to

three models: (a) categorical models concerned with qualitative

either-or judgments, e.g.: group stereotypes, (b) dimensional models

concerned with quantities of qualities, e.g. spatial representations

such as the Semantic Differential (Osgood et al., 1957) designed to map

affective meanings, and (c) relational models concerned with relations

between people and objects rather than qualities, e.g.: reciprocity

rules such as "A likes B implies B likes A".

It may be evident that interpersonal conflict is a matter of

interpersonal relationship and hence the corresponding implicit

theories are of the relational type. This has consequences in that a

relational configuration has a multiple informational structure which

sets the stage for multiple implicit theories (Peeters, 1989). For

instance, consider the following relational configuration: "A is

satisfied with B and B is satisfied with himself". One possible reading

of this configuration may point to the agreement between A and B who

share satisfaction with B. However, this agreement may be detracted

from by an alternative reading arguing that the object of B's

satisfaction is the self while that of A's is just not the self but an

other. Consistently extensive research on relational information

processing (Peeters, 1983, 1987, 1989, 1991) has revealed two modes of

thinking allowing for two alternative ways to interpret the same

reality. The one way has a personalized character stressing, for

instance, socio-emotional categories, while the other has a rather

depersonalized character dealing with persons as mere "roles" or

"functions" that, in principle, could be performed as well by

impersonal devices such as robots. Thus in the abovementioned example,

a personalized interpretation might view A's satisfaction with the

other as a manifestation of altruism or sociability, while B's

satisfaction with the self as a manifestation of conceit or

egocentrism. A depersonalized interpretation then might may focus on

the agreement between A and B and explain it by A and B evaluating B

using the same standards which may have an impersonal character such as

a certain level of performance B is expected to achieve. If B succeeds,

then he or she is object of satisfaction for A as well as for B whereby

it does not matter that A is an "other" while B the "self".

Given this duality one could wonder if there are also personalized

and depersonalized implicit theories of interpersonal conflict. When

consulting dictionaries we find interpersonal conflict defined as a

"disagreement" which, in the light of the above example, might be

associated with the depersonalized discourse. At the same time,

however, the notion of "conflict" is often associated with more

"personalized" socio-emotional states such as fights and quarrels

involving anger and hate. Hence it seems reasonable to expect both

personalized and depersonalized facets in implicit theories of

conflict. However it is not clear at al whether these two facets are

linked together as parts of one unified whole, or whether they form

distinct full-fledged implicit theories that are independent of each

other and perhaps even mutually exclusive. Data presented in the

highermentioned studies on relational information processing suggest

that, although they are formally compatible, personalized and

depersonalized discourses seem psychologically incompatible subjects

avoiding to combine them in their interpretations of reality. In

addition subjects seem biased to use the personalized rather than the

depersonalized discourse. The latter may only be prominent in settings

that include appropriate cues for it. Generalizing to the issue of

interpersonal conflict, we may advance the following working

hypotheses:

Hypothesis 1. There are at least two independent implicit theories

of interpersonal conflict, one focussing on socio-emotional aspects of

the interpersonal situation, an other focussing on the formal amount of

interpersonal agreement and disagreement implied by the situation.1

Hypothesis 2. Implicit theories underlying perceptions of

interpersonal conflict are more frequently of the socio-emotional type

than of the formal type.

Hypothesis 3. Implicit theories of the formal type are more

prominent in perceptions of conflict in formal settings such as work

situations than in rather informal settings such as love and friendship

relations.

Introduction to the Experiment

The experiment was designed in order to map implicit theories

underlying judgments about "conflictivity" of hypothetical

interpersonal situations. A standard procedure was used which was an

adaptation of the one used in the previous studies on relational

information processing mentioned higher. It involved a fixed set of 16

descriptions of situations. Each situation was the realisation of a

relational pattern involving two individuals (A and B) and each

entity's relations with the self (relations AA and BB) and with the

other (relations AB and BA). Each relation could be either positive or

negative allowing for the 16 relational patterns 'a,b,..p' represented

by the +/- signs in the upper part of table 1 (row 1-4). Positive and

negative relations were formulated respectively "is satisfied with" and

"is dissatisfied with". For instance, the situation corresponding to

pattern 'g' was formulated as follows (the experiments being run in

Dutch, "oneself" is the translation of "zichzelf" which means also

"himself" and "herself"):

"A is satisfied with B; B is not satisfied with A

A is not satisfied with oneself; B is satisfied with oneself"

Using this implementation, a standard questionnaire of 16 items was

constructed, each item consisting of a situation and the following

10-point rating scale:

'0 1 2 3 4 5 6 7 8 9'

Subjects were asked to indicate how conflictive they considered each

situation, the higher the number they marked, the more conflict they

expected between A and B. In the processing of the data, however, the

numerical values of the scales were reversed. Thus '0' indicated

maximal conflict also referred to as minimal harmony, and '9' indicated

minimal conflict also referred to as maximal harmony.

In order to use the model presented in table 1 as a data processing

tool, + and - signs are handled as +1 and -1. Further, the four

original rows are extended with rows 5-15 obtained by multiplicative

combinations of the values in rows 1-4 as indicated in the extreme

right column (between brackets). Thus row 5 is obtained by multiplying

the corresponding +1 and -1 values (+ and - signs) in rows 1 and 2. In

this way a grid of 15 rows is obtained against which we can plot a

subject's scale scores for the 16 items by correlating scores with rows

resulting in a "profile" of 15 correlations. Correlations between rows

being zero, the correlations that form the profile do not share common

variance. It follows that using product-moment correlations r, the

squared profile r's indicate unequivocally the proportions of the

variance of the subject's scale scores that is accounted for by the

corresponding rows of the model.

A socio-emotionally oriented subject who associates conflict with

dissatisfaction with the other may produce ratings that are positively

correlated with rows 1 and 2. A formally oriented subject who

associates conflict with mere disagreement would produce ratings

positively correlated with rows 9 (disagreement about A) and 10

(disagreement about B). Thus hypothesis 1 predicts subjects producing

high positive correlations either with rows 1 and 2, or with rows 9 and

10. Hypothesis 2 predicts a dominance of the positive correlations for

rows 1-2 as compared with those for rows 9-10. Hypothesis 3 predicts a

dominance of positive correlations for rows 9-10 in a formal setting as

compared with an informal setting.

Readers acquainted with statistics may observe the similarity

between the model in table 1 and a 4-factorial ANOVA model. However,

for the analysis of the data the present correlational method was

preferred to ANOVA because it allowed to diagnose implicit theories on

the level of the single subject. Aggregating scores over subjects, as

required by ANOVA, might result in novel data configurations that are

not at all diagnostic for the implicit theories used by the subjects.

------

table 1 about here

------

Method and Design

Subjects. 37 male and 27 female Dutch-speaking freshmen from the

Faculty of Economical Sciences of the University of Leuven were

individually asked to complete a questionnaire in the course of a more

encompassing experimental-psychological session in which they

volunteered as subjects.

Questionnaire. The questionnaire consisted of two sheets. First

there was a small instruction sheet which was was read aloud by the

experimenter and was exposed all the time to the subject. It contained

first of all the manipulation of conditions (see below). Further it was

explained that the situations that would be presented were mutually

unrelated A and B not representing the same pair of individuals in the

various situations. Then followed a brief explanation of the use of the

rating scales as described higher. Finally it was mentioned that the

subjects could handle the items in the order they wanted and that it

was allowed to correct previous answers. The second sheet was a large

item sheet containing the 16 items described higher. Two equivalent

versions were used, the main difference concerning the presentation

order of the items.

Experimental Conditions. The experiment involved two replications or

parallel experiments. In the first replication A and B were explicitly

presented in differential complementary roles while A and B were not

differentiated in the second replication. Within each replication a

formal condition was contrasted with an informal one. In this way the

four following conditions were obtained:

1.- Formal condition of replication 1 or "secretary/boss condition" (17

subjects). In the instruction sheet belonging to this condition A and B

were specified respectively as a secretary named "Ann" and her "boss".

2.- Formal condition of replication 2 or "technicians condition"(15

subjects). In the instruction sheet A and B were presented as two

technicians put to work on the same task and thus dependent on each

other for their job.

3.- Informal condition of replication 1 or "wife/husband" condition (18

subjects). In the instruction sheet A and B were specified respectively

as "Ann" and "Bert" presented as a "married couple".

4.- Informal condition of replication 2 or "friends condition"(14

subjects). In the instruction sheet A and B were presented just as "two

friends".

Summary of design. Subjects rated how conflictive 16 situations

were. The situations belonged to one of four between-subjects

conditions: two formal ones ("secretary/boss" and "technicians") and

two informal ones ("wife/husband" and "friends"). The gender of the

subjects and two conceptually equivalent versions of the item sheet

(with the same items in different orders) were counterbalanced across

the conditions.

Results

A Q-type factor analysis (principal components with varimax

rotation) was applied to the product-moment correlations computed

between subjects over items. Eight factors had eigenvalues higher than