Interpersonal and personality dimensions of behavior: FIRO-B and the big five.
The analysis of individual differences in human behavior has two distinct, yet compatible, conceptual frameworks. Personality draws heavily from the medical model of behavior, in that the sources of a person's actions are presumed to reflect inherent variations in motivation within the individual. The vernacular concept of "mental illness" and the traditional allopathic model of psychiatry both view individuals themselves as sources of behavior. Analysis of a person's behavior is assumed to yield meaningful dimensions of individual differences, which constitute the "personality." Some personal predisposition to pathology is assumed to appear as disruptive behavior in an individual's relationships. Presumably, the normal individual is one who is free of the potential dysfunctional aspects of behavior that would otherwise cause distress. This symptomatic model of the internal disturbance is the principle on which the diagnostic and statistical criteria of psychiatry are based. In its medical formulation, it appears as the DSM, and this model also pervades legal, ethical and cultural assumptions about deviance. By contrast, the scientific study of interpersonal relationships as entities independent of personality gained stature with small-groups research, classically reflected in the study of the dyadic relationship. The increasing focus on the situational determinants of behavior impelled the recognition of the dimensions of emergent interactions. Lewin's (1947) work on groups as models of democratic openness or authoritarian rigidity largely defined the nascence, and enhanced the evolution, of modern social psychology. The construct of personality to complex interactions was essentially negated as irrelevant (Mischel & Shoda, 1995), while recognition of the importance of the group dominated causal explanations of behavior.
Sullivan (1953) redefined personality itself as an interpersonal phenomenon. He viewed "personality" as a relatively stable pattern of interpersonal behaviors arising from interactions with others, especially during critical developmental periods. Sullivan argued that interpersonal skills and the sense of identity developed from human interactions. He asserted that interpersonal skills arise from individuals' reactions to an innate drive to reduce anxiety. This global, undifferentiated drive becomes transformed through maturity and experience to become more focused as differentiated needs for security and satisfaction. Empathy is a central component of socialization, as a skill for evaluating others' needs and abilities. An individual who fails to master the distinctions among social identity, power, and love is at substantial risk for the development of pathological relationships.
As fields of research, personality (with its intrapersonal emphasis), and interpersonal relations (emphasizing dimensions of interaction among individuals) are both maturing as alternative, yet complimentary, foci of individual differences. Personality, from the purely psychometric view of factor models, has tended to focus on a small number of predispositions, resulting in the increasingly prominent Five Factor Model (FFM; Costa & McRae, 1985; 1992; Wiggins, 1996). This perspective views the internal, more-or-less stable behavioral predispositions, as orthogonal dimensions of personality which have been labeled Extraversion, Neuroticism, Agreeableness, Conscientiousness, and Openness. The Big Five model was the basis for Basic Factors Inventory (John & Srivastva, 1999), a 44-item measure of extraversion, neuroticism, agreeableness, conscientiousness, and openness. The model has not only achieved a certain bedrock status by its scope and apparent robustness, but has the virtue of substantive convergent support from lexical, symbolic-interactionist and dyadic-interactional perspectives, each lending individual support to, and ultimately mutually validating, the factors (Wiggins, 1996).
The measurement of interpersonal constructs presents a more convoluted problem. The assessment of interpersonal behaviors is more complex than that of personality. For most purposes, the personality is conceived as static--a person is more or less the same as he or she was yesterday, and will presumably be very similar to his or her personality tomorrow. Measurement of interpersonal behavior requires that the person be viewed as interacting in a number of different situations. Insight into this perspective was provided by early analyses of group behavior (Lewin, 1947), in which it was argued that there are dimensions of interpersonal behavior that cannot be predicted by personality measures alone. This argument was further supported by Cattell (1948), who asserted that there was an interpersonal component of "syntality" that arose from interpersonal interaction. Syntality could not be predicted directly from measured personality traits. Instead, it was a distinct and relationship-specific phenomenon.
While researchers such as Leary (1957) and Kiesler (1996) have tended to focus on two-dimensional interpersonal models, there remains the issue of the minimum number of dimensions that actually exist. There is general agreement that the dimension of Dominance-Submissiveness is well established. However, there is considerable disagreement as to the components of positive and negative emotional aspects of interaction, since it is possible to interpret the dimension of positive and negative interactions as reflecting the existence of an additional component.
Schutz (1958) advanced the Fundamental Interpersonal Relations Orientation (FIRO) system. Schutz posited the existence of three basic dimensions of behavior. "Control" reflects the person's dominance in the interaction--an individual elevated in Control tends to direct, lead, or manipulate the relationship; those persons low in Control tend to emit patterns of behavior that facilitate others to initiate dominance. Schutz made a critical distinction between two components of interaction involving the definition of an individual's role in a relationship. "Inclusion" addresses the issue of personal significance in an interaction. A person elevated in inclusion is recognized as positively or negatively significant in an interaction. The third dimension is "Affection," a measure of the positive or negative emotional aspects of a relationship. Individuals elevated in Affection are emotionally bound to the relationship; those low in affection have little emotional investment in the relationship.
Schutz (1958) defined each of these three relational components as having two distinct tactical operations. Each aspect has an "Expressed" component and a "Wanted" component. Inclusion Expressed (IE) behaviors signify a desire to be a member of a relationship. Inclusion Wanted (IW) behaviors are internal desires to be included by another. If the person is socially competent, he or she will manifest appropriate matches in Expressed and Wanted aspects. Problems arise, however, for the individual who lacks the interpersonal skills to match Wanted and Expressed needs. Interpersonal incompetence arises from a disjunction in the level of expression versus wanting of a component.
The literature is mixed regarding evidence for the tripartite distinction posited by Schutz (1958). The three-dimensional model was developed by a careful analysis of self-reports. However, other researchers have failed to support a distinction between Inclusion and Affection (Gough & Bradley, 1996). This issue is further complicated by the subtleties inherent in assessment; indeed, a major paper argues that the failure to cross-validate personality or interpersonal measures founders on the actual structure of the language itself (Hofstee, DeRaad & Goldberg, 1992). Thus, the direct comparison of dimensions across linguistic communities presents problems.
The current study was conducted to explore the relationship between two established models of individual differences. The personality-within model was represented by the Big Five measure; the personality-between model was represented by the FIRO-B model. It was hypothesized that the two approaches would have common variance in some areas, but that unique dimensions of behavior would emerge for each measure.
A total of 192 students, 64 men and 128 women, participated in the study. All were recruited from Introductory Psychology classes at a large urban university in the southeastern part of the United States. Approximately 57% of the participants identified themselves within the category "Caucasian/white," 25% self-identified as "African-American," and the remaining 18% selected another category (Hispanic, Asian-American, Other). The median age of the sample was 19 years. The participants received partial course credit for research participation.
The Big Five Inventory consists of 44 items aggregately measuring five independent dimensions. Extraversion has 8 items, of which 3 are reverse-scored. Agreeableness and Conscientiousness are each represented by 9 items, including 4 reversals, for each scale. Neuroticism has 8 items, with 3 reversals; Openness has 10 items, 2 reversed.
The FIRO-B (Schultz, 1958) consists of 54 items, 9 for each of the 6 scales. For convenience, the scales are labeled IE for Inclusion Expressed, IW for Inclusion Wanted, CE for Control Expressed, CW for Control Wanted, AE for Affection Expressed and AW for Affection Wanted.
Participants were recruited via announcements and postings on the course webpage. Participants appeared at their leisure and completed a packet containing the FIRO and the BFI in functionally random sequence. Data were collected on computer answer sheets. Since the number of items for the BFI scales differed, a mean score was computed for each of the scales for each respondent. All items for both instruments were scored on a 5-point continuum where "1" indicated complete disagreement and "5" indicated complete agreement. Pearson correlation coefficients were computed on the combined matrix of the 6 FIRO-B and 5 BFI scales.
Table 1 presents the means and standard deviations for each of the FIRO and BFI scales. Table 1 also presents the reliability coefficients (alpha) for each of the 6 FIRO and 5 BFI scales. Each FIRO scale was found to be psychometrically stable, with alpha coefficients ranging from .81 to .94 (median = .87).
The BFI scales consisted of different numbers of items. Therefore, items on each scale were averaged to provide mean scale scores that could be compared directly to one another (each could range from 1 to 5). As shown in Table 1, each scale had adequate reliability, with alpha coefficients ranging from .81 to .88.
One goal of this project was to explore relations among the FIRO scales to examine the extent to which the three relationship dimensions (affection, control, and inclusion) were distinct from one another, as well as to determine whether expressed and wanted degrees of each dimension were correlated. The correlations among the FIRO measures appear in Table 2.
A clear pattern emerged in the relations among the FIRO scales. The correlations between the Affection and Inclusion measures were quite strong (r =.65 between Affection Expressed and Inclusion Expressed; r =.66 between Affection Wanted and Inclusion Wanted). In addition, the Wanted and Expressed levels of both Affection and Inclusion were highly correlated (r=.68 for affection and r=.68 for inclusion). Thus, participants did not seem to distinguish affection from inclusion, and seemed to desire and express similar amounts of these relationship dimensions.
The Control dimension of the FIRO provided a very different pattern of correlation with the other FIRO scales. First of all, Control Expressed was not correlated significantly with Control Wanted (r=.05, ns.), indicating that respondents generally desired a different level of control in relationships than they expressed. Secondly, the Control dimension was not correlated highly with the other FIRO dimensions. For example, Affection Expressed and Affection Wanted were not significantly related to either Control Expressed or Control Wanted (strongest r =.10, all non-significant). Inclusion was correlated somewhat more strongly with Control, and Control Expressed was significantly correlated with both Inclusion Expressed (r=.20) and Inclusion Wanted (r=.25). These correlations were of modest magnitude. However, Control Wanted was not significantly related to Inclusion Wanted (r=.02) and Inclusion Expressed (r = -.06). The overall pattern of these correlations suggests that Control is distinct from both Affection and Inclusion, with Control Expressed moderately correlated with Inclusion, presumably because one must interact with another before issues of control become salient. In most circumstances, one must interact with others in order to express control over them.
The relationship among the dimensions of interpersonal skills and individual dimensions of personality was examined operationally by comparing the FIRO, an interpersonal measure, with the scales from the BFI trait measure. Correlations among the FIRO and BFI measures (see Table 2) disclosed a provocative pattern. First, the FIRO Affection measures (both Expressed and Wanted) were positively correlated with the Extraversion, Agreeableness, and Openness scales from the BFI (r's ranging from .26 to .37). This pattern of results suggests that both Affection Expressed and Affection Wanted are actually components of a single generally more positive and socially reactive pattern common to both personality and interpersonal dimensions.
Additionally, it was noted that both Inclusion Expressed and Inclusion Wanted were positively related to Extraversion (r = .49 and r =.39, respectively). For the remaining FIRO and BFI scales, Inclusion Expressed was correlated significantly only with Agreeableness (r = .23) while Inclusion Wanted correlated significantly only with Openness (r = .27). While the results for the FIRO Affection and Inclusion measures were not isomorphic, the dimensions were quite similar in that, generally, they were positively related to the Extraversion, Agreeableness and Openness scales from the BFI.
The Control Wanted and Control Expressed dimensions from the FIRO presented a somewhat nebulous pattern when contrasted with the Inclusion and Affection dimensions. Control Expressed had a modest correlation with Extraversion (r = .23), but did not correlate significantly with any other BFI scale. However, Control Wanted correlated negatively with Extraversion (r = -.21) and Conscientiousness (r = -.26) while being positively correlated with BFI Neuroticism (r = .35).
The findings of the present study provide partial support for a commonality between BFI personality measures and FIRO interpersonal components as correlates of interpersonal behavior. It is, however, also very clear that the BFI and FIRO measures are neither isomorphic nor strongly equivalent. It appears that the two systems of assessment reflect different domains for the explication of individual differences in relationships.
Some common components of the differing domains can be limned fairly clearly. For Extraversion, the effect is pronounced, where there was a pattern of consistently strong correlations with all six of the FIRO subscales. It would appear that this commonality arises from the likely reality that elevated scores on Extraversion merely indicate that the individual so elevated simply has a larger social sphere in which to acquire the social skills necessary to maintain a larger, more inclusive number of relationships. Persons for whom Extraversion is below average (labeled in the vernacular as "Introverts") presumably have fewer social interactions, which may be limited to a fairly low level of interpersonal demands. Thus, the social isolate only infrequently encounters opportunities to demonstrate any need for sophisticated interpersonal skills simply because he or she only rarely interacts with others.
As would be expected for the assumed tactic of Control in relationships, elevated scores on Extraversion correlated negatively with Control Wanted. Presumably, extraverts are not comfortable with their relationships being constrained by external demands from other people. Notably, Extraversion was correlated positively with several other interpersonal measures. Highly Extraverted individuals tended to express control, and to both express and want inclusion and affection from others.
Agreeableness, a social skill component of the BFI, was found to correlate positively with Affection Expressed, Inclusion Expressed and Affection Wanted--again, suggesting an affinity for social competence among persons desiring social activity. An individual deficient in social skills would not be expected to thrive in an arena of intense social interaction. It would be expected that the person with deficits in social skills would likely be actively shunned by those more extraverted persons, who would reasonably anticipate that others who were more extraverted would be able to reciprocate interactions effectively.
The negative correlation between Conscientiousness and Control Wanted suggests that Conscientiousness may reflect a sophistication in selecting relationships in such a manner as to avoid control by others. Notably, Neuroticism is strongly and positively correlated with Control Wanted, suggesting that individuals who have problems with emotional expression may welcome being controlled by others. Finally, Openness is correlated with Affection Expressed, Affection Wanted and Inclusion Wanted, suggesting that people who are more responsive to novelty are willing to experiment with novel aspects of relationships.
While there were certainly clear relations among the personality and interpersonal measures that appear to show the models to be conjunctive along some dimensions, there were also areas in which the two models appear to be assessing two quite distinct domains of individual differences. First, while Conscientiousness was negatively correlated with Control Wanted, it was not related to Control Expressed nor to any of the other FIRO measures. In particular, one would expect a highly conscientious person to be quite low in Control Expressed toward others of similar social power. It could be that persons low in Conscientiousness feel little, if any, obligation to interact with others simply because people who are relatively low on Conscientiousness may be indifferent to social rituals. Similarly, Neuroticism was not correlated significantly with either of the behavioral measures of Affection or Inclusion in the FIRO, something that would have been expected if the personality and interpersonal models were essentially homologous.
From an overall analysis of the FIRO, it is apparent that Affection and Inclusion are particularly closely related, at least in the current sample of young, diverse urban college students. Both Affection and Inclusion correlated with Extraversion, Agreeableness and Openness, suggesting that the constructs in this domain may be more closely related, perhaps to the point where the distinction is purely academic--a distinction without a difference.
There is an important methodological point to be considered here. It may be the case that only the more socially and interpersonally sophisticated individuals may be aware of the distinction between Inclusion and Affection, and this distinction in Schutz's (1958) original three-dimensional model may be an artifact of the construction of the FIRO system. In general, the distinction between Inclusion and Affection may be conceptually sound, but its operationalization in the separation of Inclusion and Affection scales may be an artifact of measurement, in that it only appears among certain populations. Schutz (1958) developed his original analysis based on data collected from a highly homogeneous and socially exclusive population. All were males, matriculating at Harvard, and all data for the initial model were collected in the 1950s.
Harvard in the 1950s was a bastion of wealth, as well as one of political, social, and economic conservatism. There was a strong element of presumed patrician social class and a norm of ethnic limitation, or even exclusion. Students were admitted with criteria that made positive exceptions for sons of graduates, and for applicants who could bring substantial economic resources. The Ivy League then was far more "clubbish" than would be the case early in the 21st century. In the socially rarified cultural atmosphere of the Ivy League, it is reasonable to assume that most participants would have had many opportunities to ruminate and introspect on their privileged situations. They would more likely possess the luxury of time to contemplate subtle differences in relationships. The Harvard men could experience a greater opportunity for more sophisticated relationships. They would have had the opportunity to become acutely aware of the differences between deficiency-driven opportunistic relationships of the non-elite and their own privileged opportunity to experience highly contrived and opportunistic interactions available then only to a relatively small and socially isolated, yet powerful, elite. Thus, the opportunity to explore subtle differences in the distinctions between Inclusion and Affection existed as clear distinctions apparent to an elite attuned to nuances of interaction often lacking in a less homogeneous sample. Schutz's sample was elite, and that may have accounted for the isolation of a third component only when sophisticated and highly socially skilled individuals evaluated similar others. In a more heterogeneous general population, however, there would be relatively fewer opportunities to experience subtleties of interaction.
It is notable that there were important distinctions evident in the FIRO control dimensions. For four of the five BFI factors, the directions of the correlations are opposed, even when the correlations did not achieve significance. The sole exception is in the intensity of the association. Control Wanted correlated strongly with Neuroticism, but the correlation between Neuroticism and Inclusion Wanted was essentially zero. This variation in the assessment of the Control dimension suggests that a more complex interaction may be operative. A reconsideration of the measurement of personal control, possibly of variations in the bases of power, appears to warrant rigorous exploration.
Based on our findings, it is difficult to support the three dimensions in Schutz's FIRO system for use with the general population, or at least within the young group of urban college students examined here. Reflecting the preponderance of findings, it would appear that a two-dimensional model with Dominance (Schutz's Control) and Socio-emotional Affect (Inclusion plus Affection) as the basic dimensions, would retain adequate rigor for interpersonal analysis. Relatively few people may be able to negotiate the distinction between the two subcomponents of Inclusion and Affection, and the appearance of a third dimension may signal the operation of a nuanced perception, rather than the far more common tendency to equate the two components maintained as conceptually distinct in the FIRO system.
Cattell, R. B. (1948). Concepts and methods in the measurement of group syntality. Psychological Review, 55, 48-63.
Costa, P. T., Jr., & McRae R. R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psychological Assessment Resources.
Costa, P. T., Jr., & McRae R. R. (1992). Four ways five factors are basic. Personality and Individual Differences, 12, 887-898.
Gough, H. G. & Bradley, P. (1996). CPI manual. (3rd ed). Palo Alto, CA: Consulting Psychologists Press.
Hofstee, W.B.K., DeRaad, B., and Goldberg, L. R. (1992). Integration of the Big Five and the circumplex approaches to trait structure. Journal of Personality and Social Psychology, 63, 146-163.
John, O. P., & Srivastva, S. (1999). The Big Five trait taxonomy: History, measurement and theoretical perspectives. In L. A. Pervin & O.P. John, (eds.). Handbook of personality: Theory and research (2nd ed.., pp. 102-138). New York: Guilford.
Kiesler, D.J. (1996). Contemporary interpersonal theory and research: Personality, psychopathology and psychotherapy. New York: Wiley.
Leary, T. (1957). Interpersonal diagnosis of personality. New York: Ronald.
Lewin, K. (1945). Frontiers in group dynamics: Concept, method and reality in social science. Human Relations, 1, 5-41.
Mischel, W. & Shoda, Y. (1995). A cognitive-affective system of personality: Reconceptualizing situations, dispositions, dynamics and invariances in personality structure. Psychological Review, 102(2), 246-268.
Schutz, W.C. (1958). FIRO: A three-dimensional theory of interpersonal relations. NY: Holt, Rinehart & Winston.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Wiggins, J. S. (ed.). (1996). The Five-Factor model of personality: Theoretical perspectives. New York: Guilford.
The authors gratefully acknowledge the assistance of an internal grant from the College of Humanities and Sciences in the completion of this research.
Author info: Correspondence should be sent to: John M. Mahoney, Department of Psychology, Virginia Commonwealth University, Richmond, VA 23284-2018; e-mail: email@example.com.
John M. Mahoney
Virginia Commonwealth University
Mark F. Stasson
Metropolitan State University
TABLE 1 Descriptive Statistics and Internal Consistency Measures for Fundamental Interpersonal Relationship Orientation and Basic Factors Inventory Scales. Mean S.D. Alpha Affection Expressed (AE) 32.9 6.4 0.9 Control Expressed (CE) 24.7 6.5 0.9 Inclusion Expressed (IE) 30.3 6.3 0.9 Affection Wanted (AW) 35.7 6.0 0.8 Control Wanted (CW) 20.7 6.4 0.9 Inclusion Wanted (IW) 34.7 7.4 0.9 Extraversion (E) 3.4 0.8 0.8 Agreeableness (A) 3.7 0.6 0.8 Conscientiousness (C) 3.4 0.6 0.9 Neuroticism (N) 3.0 0.8 0.9 Openness (O) 3.5 0.5 0.8 TABLE 2 Correlations Among Fundamental Interpersonal Relationship Orientation and Basic Factors Inventory Scales. AE CE IE AW CW IW AE -- CE .10 -- IE .65 .20 -- AW .68 .04 .49 -- CW .03 .05 -.06 -.02 -- IW .64 .25 .68 .66 .02 -- E .37 .23 .49 .29 -.21 .39 A .32 -.13 .23 .35 -.09 .17 C .13 -.07 .11 .07 -.26 .04 N -.05 -.06 -.08 -.02 .35 .02 O .26 .08 .17 .36 -.13 .27 E A C N O AE CE IE AW CW IW E -- A .17 -- C .16 .47 -- N .27 -.32 .35 -- O .32 .14 .15 -.11 -- Notes: Coefficients in bold are significant at p<.01 (2-tail). Key to variable abbreviations: AE = Affection Expressed; CE = Control Expressed; IE = Inclusion Expressed; AW = Affection Wanted; CW = Control Wanted; IW = Inclusion Wanted; E = Extraversion; A = Agreeableness; C = Conscientiousness; N = Neuroticism; O = Openness
|Printer friendly Cite/link Email Feedback|
|Author:||Mahoney, John M.; Stasson, Mark F.|
|Publication:||North American Journal of Psychology|
|Date:||Jun 1, 2005|
|Previous Article:||A new approach to the study of deception in organizations.|
|Next Article:||The responses of male and female managers to workplace stress and downsizing.|