Academic journal article Canadian Journal of Behavioural Science

Inferential Structure of the NEO Five-Factor Inventory: Construct Validity of the Big Four Personality Clusters

Academic journal article Canadian Journal of Behavioural Science

Inferential Structure of the NEO Five-Factor Inventory: Construct Validity of the Big Four Personality Clusters

Article excerpt

Abstract

Four studies investigated the dimensionality of the NEO Five-Factor Inventory (NEO-FFI; P. T. Costa & R. R. McCrae, 1992). In Study 1, four inferential dimensions and four clusters represented the NEO-FFI when 114 undergraduates freely sorted items into categories. Construct validity for four item-clusters derived from the inferential space was obtained in Study 2 based on selfreport with 304 undergraduates. Study 3 validated these inferential clusters using self and peer reports for 420 undergraduates. Study 4 validated the cluster scales for predicting quality of life and significant social and cultural behaviours for 110 undergraduates. Implications for implicit personality theory, the number of dimensions issue in personality, and test construction are discussed.

Résumé

Quatre études ont examiné la dimensionnalité du NEO Five-Factor Inventory (NEO-FFI; P. T. Costa & R. R. , McCrae, 1992). Dans l'étude 1, quatre dimensions inférentielles et quatre groupes représentaient le NEO-FFI où 114 étudiants de premier cycle ont trié librement des articles en catégories. La validité du concept de quatre groupes d'articles dérivés de l'espace inférentiel a été obtenue dans l'étude 2 à partir d'une autoévaluation de 304 étudiants de premier cycle. L'étude 3 a validé ces groupes inférentiels à l'aide de l'autoévaluation et de l'évaluation par les pairs de 420 étudiants de premier cycle. L'étude 4 a validé les échelles de groupe pour prédire la qualité de vie et les comportements sociaux et culturels significatifs de 110 étudiants de premier cycle. Les conséquences sur la théorie de la personnalité implicite, le nombre de problèmes de dimension dans la personnalité et le concept du test font l'objet de la discussion.

Historically, researchers have relied on trait terms available in everyday language to define models of personality (e.g., Digman, 1996; Wiggins & Trapnell, 1997). Known as the lexical hypothesis, the assumption implicit in this approach is that the most important personality dimensions are likely to be represented in a culture and, therefore, in the language of that culture. In using this approach to study individual differences, one goal of many researchers is to identify the basic dimensions that define personality. Given the large body of research relevant to this goal, how many fundamental personality dimensions are there, and what are they?

The Number of Factors Debate in Personality

There is empirical evidence of three (e.g., Eysenck, 1991), five (e.g., McCrae & Costa, 1987), at least six (Jackson, Paunonen, Fraboni, & Coffin, 1996), seven (Almagor, Tellegen, & Waller, 1995), and even as many as sixteen (e.g., CatteII & Krug, 1986) fundamental dimensions to personality. Although subject to some dispute (e.g., Paunonen & Jackson, 2000), a growing consensus seems to favour what is referred to as the Big Five model of personality, typically characterized by dimensions of Neuroticism, Extraversion, Openness (Intellect/Imagination), Agreeableness, and Conscientiousness.

Despite the legitimacy of the model coming partly through its derivation from both lexical and questionnaire analysis, this cross-modal consensus may not necessarily reflect accuracy or completeness. For example, some researchers argue that the aforementioned discrepancies may exist as a function of different analytic procedures (e.g., Block, 1995a, b). Others take issue with the representativeness or content validity of the personality stimuli and are concerned that the reported consistencies in structure might partly reflect variable sampling. They argue that, sometimes, the choice of nonrepresentative descriptors might produce an underlying dimensionality that is neither comprehensive nor unbiased. Saucier (1997), for example, has demonstrated that the choice among solutions reflecting either three broad factors (Extraversion, Agreeableness, and Conscientiousness), or the Big Five, or seven dimensions (the Big Five plus Attractiveness and Low Base-Rate Attributes) depends, at least partly, on the breadth of sampling of dispositions and states. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.