Suicidal ideation and the five-factor model of personality

Abstract

Trait predictors of suicidal ideation were examined within the taxonomic framework provided by the five-factor model of personality in a sample of young adults (N=185). Participants completed the revised NEO Personality Inventory (NEO-PI-R) and the Adult Suicidal Ideation Questionnaire (ASIQ), an inclusive measure of suicide-related thinking. ASIQ scores were positively associated with neuroticism and negatively associated with conscientiousness. Factor-level multiple regression analyses revealed significant gender differences; namely, suicidal ideation was positively predicted by neuroticism in females and negatively predicted by conscientiousness in males. More detailed analyses revealed distinctive patterns of association between facets and ASIQ scores within factor domains. Suicidal ideation was positively predicted by the neuroticism facets, angry hostility and depression, and negatively predicted by the conscientiousness facet, self-discipline. These findings are discussed in relation to previous research investigating personality and suicide-related constructs.

Introduction

Studies investigating the relationship between suicide ideation and putatively normal personality variables have most consistently indicated that neuroticism is strongly (positively) related to suicidal thinking in non-clinical samples (e.g. Mehryar, Hekmat, & Khajavi, 1977, Lester, 1987, Lester, & Francis, 1993). Indeed, neuroticism is likely the most salient of trait factors implicated in the development of a broad spectrum of suicide-related behaviors (e.g. Isherwood, Adam, & Hornblow, 1982; Ashton, Marshall, Hassanyeh, Marsh, & Wright-Honari, 1994, Nordstrom, Schalling, & Asberg, 1995) and major depression Hirschfeld et al., 1989, Maier, Lichtermann, Minges, & Heun, 1992, Kendler, Neale, Kessler, Heath, & Eaves, 1993, Clayton, Ernst, & Angst, 1994. Empirical support for the notion that suicidal ideation is negatively related to extraversion and positively related to psychoticism Irfani, 1978, Lester, 1987, Dyck, 1991, Lolas, Gomez, & Suarez, 1991 has been relatively less conclusive. Some of these relationships have been replicated cross-culturally (e.g. Irfani, 1978, Lynn, 1982, Lynn, & Martin, 1995).

With few exceptions, these research efforts have presumed a three dimensional trait model (P-E-N), based on principles of Eysenckian theory (Eysenck, 1991, Eysenck, 1992) and in the majority of cases, operationalized by the Eysenck Personality Questionnaire (EPQ; Eysenck, & Eysenck, 1975, Eysenck, Eysenck, & Barrett, 1985). Concurrently, an increasing number of trait theorists have recognized the potential utility of adopting alternative conceptual models, i.e. trait taxonomies comprised of five rather than three primary dimensions (Costa, & McCrae, 1985, Costa, & McCrae, 1992a, Digman, 1990, Digman, 1996, Goldberg, 1990, John, 1990, McCrae, & John, 1992, McCrae, & Costa, 1996, Wiggins, & Trapnell, 1996). These five factors (neuroticism, extraversion, openness to experience, agreeableness and conscientiousness) are operationally defined by Costa and McCrae's revised NEO Personality Inventory (NEO-PI-R; Costa, & McCrae, 1992b), a widely used measure of normal personality.

The present study was conducted to further investigate the relationship between suicidal ideation and potentially underlying trait constructs in a non-clinical (university) sample, by considering neuroticism, extraversion and the additional personality dimensions elaborated in the five-factor model (openness, agreeableness and conscientiousness). In at least one case, the NEO-PI has been employed to investigate suicidal behavior. Duberstein and associates Duberstein, Conwell, & Caine, 1994, Duberstein, 1995, reporting the results of a psychological autopsy study, found that NEO-PI ratings obtained from third-person informants discriminated completed suicides and case control subjects. However, no previous studies have directly examined suicidal thinking in relation to scores obtained on Costa and McCrae's measure. Insofar as fairly limited evaluations of suicidal ideation, i.e. a single item used to measure the construct, characterize many of the previously mentioned studies examining personality variables, a more inclusive measure, the Adult Suicidal Ideation Questionnaire (Reynolds, 1991a), was selected for use in the present investigation.

Results were expected to provide further support for the notion that Neuroticism, conceptualized as a trait vulnerability factor for depression, i.e. a propensity to experience negative affect, is positively related to suicidal ideation. Gender differences were also examined, based on evidence (e.g. Isherwood et al., 1982) suggesting that the relationship between neuroticism and suicidal behavior may be gender-specific, i.e. more strongly associated in women.

Moreover, the hierarchical structure of Costa and McCrae's trait model (i.e. factors comprised of facets) permits more detailed analyses of such relationships. The validity of the NEO-PI facet scales has been established (Costa, & McCrae, 1992b, Costa, & McCrae, 1995) and facet-level analyses have been shown to add considerable and meaningful specificity to the evaluation of various state-trait relationships (e.g. Velting, & Liebert, 1997, Wolfenstein, & Trull, 1997, Velting, 1999). Thus, distinctive patterns of association between suicidal ideation and facet scores were anticipated. Of the six facets of neuroticism, depression was expected a priori to be most predictive of suicidal ideation scores.