Personality Structure Research Paper

1. Origins Of The Fi E Factor Model Of Personality Structure

The idea that variations in human characteristics have an evolutionary origin was suggested at the end of the eighteenth century by Erasmus Darwin (1731–1801) and it later bore fruit in the empirical researches of two of his grandsons, Charles Darwin (1809–82) and Francis Galton (1822–1911). That Galton was uniquely qualified to conduct research on ‘hereditary genius’ is suggested by the accomplishments and eminence of the Darwin family and by the fact that Galton was, by any criteria, a genius himself. Galton coined the term ‘eugenics’ (Greek eugenes ‘well-born’) for the field which studied the inheritance of desirable characteristics, especially intelligence. In 1904, Galton endowed a permanent chair in eugenics at University College, London and appointed the mathematician, Karl Pearson (1857–1936) as its first professor. Pearson developed the correlation coefficient to express the relation between the intelligence of parents and the intelligence of their offspring. His colleague, Charles Spearman (1863–1945) developed the method of factor analysis which, when applied to mental ability data, suggested a general factor of intelligence (‘g’) in addition to more specific factors. Raymond B. Cattell (1905–98), the founder of modern personality structure research, received his doctorate under Spearman and was the first to suggest extending the applications of factor analysis to the study of individual differences in personality characteristics.

Cattell (1967) had a master plan for the study of personality structure, which he and his associates implemented over a period of 65 years: (a) determine the totality of trait-descriptive terms in a given language that denote individual differences in thinking, feeling, and behaving (‘surface traits’); (b) reduce correlated clusters of these surface trait terms by factor analysis to their underlying determinants (‘source traits’); and (c) confirm, by factor analysis and experimentation, the extent to which these source traits, as measured by personality questionnaires, may also be found in different media of observation such as ratings by observers, self-ratings, and behavioral responses to controlled test stimuli. Cattell’s master plan set the stage for two distinct, but conceptually related, approaches to the study of personality structure that have only recently converged.

2. Two Traditions Of Fi E Factor Model Research

2.1 The Lexical Tradition

The totality of surface traits in a given language may be defined with reference to the words that are used to mark individual differences in thinking, feeling, and behaving found in an unabridged dictionary of that language. Within the lexical tradition, representative words (typically adjectives) that have been selected from dictionaries are administered in self-report and observer-report format to groups of respondents who indicate the extent to which each word is descriptive of themselves or of someone they know well. Such data are then analyzed by first calculating the extent of cooccurrences (correlations) among words, and then determining the underlying ‘structure’ of the entire set of words by factor-analytic techniques.

Lewis R. Goldberg (1981) maintained that: (a) the frequency with which an attribute is represented in a language corresponds to the general importance of that attribute in a given culture and (b) the most important dimensions of personality are those ‘universal dimensions’ that are common to all languages. Goldberg and his colleagues have studied personality trait-descriptors in English and 12 other languages (e.g., German, Italian, and Turkish) and have concluded that there are fi e factors (which they called the ‘Big Five’) that may be considered universal dimensions of personality surface trait structure. The Big Five dimensions of personality that have been identified in many different language groups are: (a) Extraversion (talkative, assertive, verbal); (b) Agreeableness (kind, cooperative, warm); (c) Conscientiousness (organized, systematic, practical); (d) Neuroticism (anxious, moody, emotional); and (e) Intellect/Openness (intellectual, imaginative, complex).

2.2 The Multivariate-Trait Tradition

The second part of Cattell’s master plan involved the development of personality questionnaires that would capture the basic dimensions of personality inherent in language. A personality questionnaire consists of items such as ‘I like to attend parties and other social gatherings where I can meet new people’ to which the test-taker is required to respond ‘true’ or ‘false’; or more typically, to rate the degree to which the statement is self-descriptive on a scale ranging from ‘not at all’ to ‘highly descriptive.’ Most questionnaires are designed to measure many dimensions or variables, and hence are called ‘multivariate.’ Because responses to questions (‘true vs. ‘false’; ‘descriptive’ vs. ‘nondescriptive’) can be scored with complete reliability, the tests are referred to as ‘objective.’

Cattell’s suggestion that Spearman’s statistical method of factor analysis could be applied to personality data was indeed a fruitful one, but the new field of personality structure also inherited some of the problems that had been associated with the use of factor analysis to determine the number and nature of mental abilities. Foremost among these problems was the issue of determining the number of factors ‘underlying’ the intercorrelations among personality scales. For example, the two most prominent figures in the field of personality structure during the twentieth century, R. B. Cattell in the United States and H. J. Eysenck in England, strongly disagreed about the number of factors that were required for the comprehensive measurement of personality structure. Cattell argued that at least 16 factors were necessary and he included these factors in his Sixteen Personality Factor Questionnaire (16 PF). Eysenck held that there were, at most, the three factors included in the Eysenck Personality Questionnaire (EPQ). A 1953 survey of the literature by psychologists at the Educational Testing Service identified and categorized the approximately 450 factors that had been reported up to that date. By the 1960s, the field of personality structure was being severely criticized, not only for its failure to agree on a common set of factors, but for its failure to demonstrate any substantial ‘real world’ correlates of such factors. At the same time, however, it was argued that the ‘Big Two’ dimensions of personality, Extraversion and Neuroticism were large, ubiquitous, and almost unavoidable dimensions of personality for which substantial and important correlates had been reported (Wiggins 1968). These dimensions had figured prominently not only in the research programs of Cattell and Eysenck, but in many other research contexts as well.

2.3 Reconciliation Of The Two Traditions

1. T. Costa, Jr., and R. R. McCrae (1994) developed a personality questionnaire that included scales for the dimensions of Neuroticism and Extraversion suggested by Cattell and Eysenck, as well as a measure of Openness to Experience that was suggested by an analysis of Cattell’s questionnaire. Their instrument, the NEO-Personality Inventory (NEO-PI), was later expanded to include the additional dimensions of Agreeableness and Conscientiousness that had been identified in the lexical work of Goldberg and others. During the 1980s and 1990s, Costa and McCrae conducted an unprecedented number of empirical studies with the NEO-PI that established the utility of this instrument, as well as the comprehensiveness of what came to be known as the Five Factor Model (FFM) of personality structure. In addition to revitalizing the field of personality structure, the work of Costa and McCrae reaffirmed the basic logic of Cattell’s master plan which called for an integrative framework for research that included both the lexical and multivariate-trait traditions.

3. Research With The NEO Personality Inventory

The NEO-PI has been an especially useful measure in the ‘applied’ contexts of personnel selection (Barrick and Mount 1991) and psychiatric diagnosis (Costa and Widiger 1992). In addition, the validity of the NEO-PI as a conceptually-justified measure of personality traits has been demonstrated with reference to the following criteria that would be required of such a measure.

3.1 Heritability And Stability

All five of the NEO-PI dimensions are moderately heritable and remarkably stable over the adult life of an individual. After the age of 30, these traits, as William James put it, appear to be ‘set like plaster.’ Costa and McCrae (1994) reported retest correlations ranging from 0.6 to 0.8 even after an interval of 30 years.

3.2 Consensual Validation And Cross-Cultural Invariance

There are reasons to be skeptical of the ‘truthfulness’ of individuals’ self-reports to personality inventories and for that reason, it is desirable to obtain ratings by knowledgeable observers of the respondent’s personality characteristics on the same items. In many of the empirical studies conducted by Costa, McCrae, and associates, an Observer Form of the NEO-PI was completed by spouses, peers, or other knowledgeable informants. In general, substantial agreement has been obtained between self-and observer-ratings, indicating a satisfactory degree of ‘consensual validity.’ The NEO-PI has been translated into more than 30 languages, including non-Indo-European languages such as Thai and Shona. The success of these translations in producing the expected five-factor structure in other cultures attests to the cultural generalizability of the Five-Factor Model.

4. Origins Of The Interpersonal Circumplex Model

Circular representations of human nature can be traced from ancient astrology to the drawings of Leonardo Da Vinci to the modern interpersonal circumplex tradition, which began with Harry Stack Sullivan. Sullivan defined psychiatry as the study of interpersonal relations and considered ‘personality’ to reside in an individual’s patterns of relationships with significant others over time. Timothy Leary (1957) and his associates attempted to operationalize Sullivanian concepts in terms of concrete measurement procedures. Their analysis of clinicians’ observations of the things that patients did to each other (and to themselves) in group psychotherapy led to a taxonomy of interpersonal variables that was empirically well captured by a circular arrangement of 16 (later 8) variables organized around the two coordinates of dominance (vs. submission) and affiliation (vs. hostility).

4.1 The Structure Of Interpersonal Behavior

The Five Factor Model and Interpersonal Circumplex are based on different assumptions about the nature of personality and employ quite different structural models. The tradition that led to the current FFM focused on the Individual and attempted to provide a comprehensive ‘list’ of the factors necessary for describing the individual’s most important personality traits. The tradition that led to the current IPC focused on dyadic relationships and attempted a detailed description of a two-dimensional ‘interpersonal space’ within which interpersonal actions and reactions could be classified. The IPC tradition assumes that a person’s characteristic interpersonal behavior may be expressed as a combination of scores on the two bipolar coordinates of Dominance (vs. Submission) and Affiliation (vs. Disaffiliation).

Interpersonal space is interpreted, quite literally, in terms of the geometry of the circle. The two underlying coordinates of this circle are Dominance (90 ) vs. Submission (270 ) and Affiliation (0 ) vs. Disaffiliation (180 ). A number of theoretical rationales have been given for the assumption that these coordinates are the ‘basic’ reference points for understanding interpersonal behavior. For example, it has been suggested that David Bakan’s (1966) distinction between agency (being a differentiated individual, striving for mastery and power) and communion (being part of a larger entity, striving for intimacy and union) provides a metatheoretical basis for interpreting Dominance and Affiliation as the basic axes of social life. Within this conceptual framework, Dominance vs. Submission and Affiliation vs. Disaffiliation are the nodal reference points on the IPC and all other behaviors may be interpreted as ‘blends’ of these two coordinates. For example, Extraversion (45 ) is an equal blend of Dominance and Affiliation; Introversion (225 ) is a combination of submissive and disaffiliative tendencies.

4.2 The Nature Of Interpersonal Space

Unlike the competing inventories of personality structure found within the multivariate-trait tradition (e.g., Cattell, Eysenck, Costa, and McCrae), different IPC measures share the same structure, but differ in their domains of application. For example, IPC models of personality and emotions have coexisted comfortably for many years and are likely to continue to do so (Plutchik and Conte 1997). Four contemporary IPC measures have quite different areas of application: The Interpersonal Adjective Scales provide a measure of relatively normal interpersonal traits or dispositions. The Inventory of Interpersonal Problems measures problems of living as perceived by respondents or others. The Impact Message Inventory measures the covert reactions of respondents to different target persons corresponding to different locations on the IPC. The Supportive Actions Scale measures tendencies to provide various kinds of social support to others in need of assistance. It is clear from the approximately 1,000 references that appeared in Donald Kiesler’s (1996) survey of the IPC literature that both the utility and generalizability of the IPC structure are well established.

5. Current Status

Examination of the relevant literature of the past decade reveals that the two major models of personality structure today are the Five-Factor Model or ‘Big Five’ and the Interpersonal Circumplex model. An early rapprochement between these two models occurred when proponents of each of the models agreed that the Extraversion and Agreeableness dimensions of the FFM were rotational variants of the interpersonal dimensions of Dominance and Affiliation and that a full description of personality structure cannot ignore the additional dimensions of Conscientiousness, Neuroticism, and Openness Intellect. It has also become increasingly apparent that the FFM, which had been characterized by some as an ‘atheoretical’ model, lends itself to interpretations from a variety of different theoretical perspectives (Wiggins 1996).

Bibliography:

1. Bakan D 1966 The Duality of Human Existence: An Essay on Psychology and Religion. Rand McNally, Chicago
2. Barrick M R, Mount M K 1991 The Big Five personality dimensions and job performance: A meta-analysis. Personnel Psychology 44: 1–26
3. Cattell R B 1967 The Scientific Analysis of Personality. Penguin, Harmondsworth, UK
4. Costa P T, Jr., McCrae R R 1994 ‘Set like plaster’? Evidence for the stability of adult personality. In: Heatherton T, Weinberger J L (eds.) Can Personality Change?. American Psychological Association, Washington, DC, pp. 21–40
5. Costa P T, Widiger T A Jr. (eds.) 1994 Personality Disorders and the Fi e-Factor Model of Personality. American Psychological Association, Washington, DC
6. Goldberg L R 1981 Language and individual differences: The search for universals in personality lexicons. In: Wheeler L (ed.) Review of Personality and Social Psychology. Sage, Beverly Hills, CA, Vol. 2, pp. 141–65
7. Kiesler D J 1996 Contemporary Interpersonal Theory and Research: Personality, Psychopathology, and Psychotherapy. Wiley, New York
8. Leary T F 1957 Interpersonal Diagnosis of Personality. Ronald Press Co., New York
9. Plutchik R, Conte H R (eds.) 1997 Circumplex Models of Personality and Emotions. American Psychological Association, Washington, DC
10. Wiggins J S 1968 Personality structure. Annual Review of Psychology 19: 293–350
11. Wiggins J S (ed.) 1996 The Fi e-factor Model of Personality: Theoretical Perspectives. Guilford Press, New York