بررسی رابطه بین مدل پنج عاملی شخصیت، عوامل اجتماعی و بزهکاری خود گزارش شده

Abstract The ‘Big Five’ is one of the predominant models of personality structure but relatively little research has focussed on how the five factors might be related to self reported offending separately for males and females. This is problematic for understanding the relationship between personality and offending as females and males typically have different personality profiles and differ considerably in self-reported offending. In this research 720 adolescents (376 males and 344 females) completed a Big Five personality measure along with measures of self-reported offending, socioeconomic status and family structure. The results suggested that low agreeableness and low conscientiousness were independently related to the prevalence of self-reported offending for males and that low agreeableness was independently related to frequent male offending. Low conscientiousness was independently related to female offending, but so too were interactions between disrupted family and extraversion and disrupted family and openness. The interaction between extraversion and a disrupted family was also independently related to frequent female offending. Limitations and directions for future research are discussed.

Introduction The most influential theory of personality and crime was that of Eysenck (1996), in which he persuasively argued that the relative balance of three superordinate personality factors that one possessed predicted their involvement in crime. Specifically, those who have high extraversion (E; sensation seeking, venturesomeness), high neuroticism (N; anxious, depressed) and high psychoticism (P; aggressive, impulsive, unempathic) were more likely to be offenders ( Eysenck, 1996). Eysenck also suggested that female offenders have the same personality profile as male offenders, and controversially, that the factors which commonly comprise sociological theories of crime (e.g. low socioeconomic status) have their impact on offending through these personality structures ( Eysenck, 1996 and Eysenck and Eysenck, 1973). Currently, the Five Factor Model (FFM) of personality is one of the primary models for organising personality traits. The FFM was based on the lexical hypothesis which suggests that all personality traits have been encoded in language (e.g. Costa & McCrae, 1995), and includes the dimensions of E and N (defined similarly to those of Eysenck), as well as Agreeableness (A; altruism, modesty), Openness (O; imaginative, aesthetic sensitivity) and Conscientiousness (C; self-discipline, competence). Research has suggested that Eysenck’s P is inversely related to both A and C ( Costa, McCrae, & Dye, 1991). A limited number of studies have compared offending to the FFM. For example, John, Caspi, Robins, Moffitt, and Stouthamer-Loeber (1994) assessed the relationship between the five factors of personality and delinquency for a group of 350 12–13 year old boys. The results suggested that the most delinquent boys scored lower on agreeableness, conscientiousness and openness and higher on extraversion. Other studies have generally found low agreeableness and low conscientiousness to be associated with offending, while a lesser degree of support has been found for an association with low openness, high neuroticism and high extraversion and offending (e.g. Bartlett and Anderson, 2012, Jones et al., 2011 and Miller and Lynam, 2001). The variation in the support for the link between specific facets of the FFM and offending might be attributable to the differential operationalisation of ‘offending’ across studies. For example, a number of studies use prisoners to represent offending behaviour, but it is well known that not all those who commit offences are convicted, let alone imprisoned for their offences (e.g. Farrington et al., 2003). There is also evidence that personality might be influenced by the experience of incarceration (Newton, 1998). Generally, self-reported offending provides a more accurate picture of more typical, but still serious offending behaviour (Thornberry & Krohn, 2003). Self-reports also have the benefit of giving a more valid picture of the number of offences that individuals commit compared to official measures of offending (Farrington, 1998). This distinction between the prevalence and the frequency of offending was an important contribution of the ‘criminal career’ approach to offending (Piquero, Farrington, & Blumstein, 2007), which has yet to be well integrated with the study of personality. This body of literature suggests that offenders are versatile in their offending (as opposed to specialising in certain types of offences), and that a small proportion of offenders in any cohort commit a disproportionate number of offences. These frequent offenders differ from more typical offenders in important individual and social background characteristics (Farrington & West, 1993). It is not known whether different aspects of the FFM might be associated with frequent as opposed to more typical offending. An additional potential source of variation in studies linking the FFM and offending is the gender composition of the samples used in past studies. While many studies do not include females, some combine males and females into a single category of ‘offenders’. This is problematic as research has consistently shown that males are much more likely to be offenders than females, males commit many more offences than females (Becker & McCorkel, 2011), and male and female offending has different correlates. Females appear more influenced by social factors (e.g. relationships, low socioeconomic status; Blanchette & Brown, 2006), and males appear more influenced by individual factors (e.g. impulsivity; Farrington, 1998). In addition, females consistently demonstrate different personality profiles than males on the FFM, with females generally higher on all facets except Openness (Marsh, Nagengast, & Morin, in press). Only one study has previously compared the FFM of personality and self-reported offending separately for males and females. Heaven (1996) administered the FFM to a group of 214 high school students (108 females and 106 males) along with a delinquency scale. Females scored significantly higher than males on the five personality domains, but significantly lower on delinquency. A significant negative correlation between low agreeableness and violence for males and females was found, as was a significant positive relationship between neuroticism and vandalism for boys and violence for girls. Conscientiousness was significantly and negatively correlated to vandalism for both boys and girls. Recently researchers have examined the potential moderating role of social environments when examining the relationship between personality and offending (Zimmerman, 2010). For example, in a large study of adolescents Meier, Slutske, Arndt, and Cadoret (2008) found that the relationship between impulsivity and delinquency was stronger in more deprived neighbourhoods. Similar findings have been noted in other studies (e.g. Lynam et al., 2000), but these have been limited to the examination of single personality traits as opposed to superordinate personality structures. It might be expected, however, that the association between personality structures (especially low agreeableness and low conscientiousness) and offending might be stronger in more negative social contexts. This study examines the independent relationship between the FFM of personality and self-reported offending for a group of male and female secondary school students. In addition the potential moderating role of low SES and disrupted families are explored. These measures of social environment were selected as both have been associated with an increased likelihood of offending (e.g. Becker and McCorkel, 2011 and Juby and Farrington, 2001) and might influence personality development. It was hypothesized that low agreeableness and low conscientiousness would be related to both male and female offending with the association stronger for more frequent offending. A negative social environment might moderate these relationships, but the limited research made this hypothesis more tentative.

Conclusion Like much previous research, the current study highlighted the importance of low conscientiousness and low agreeableness for understanding offending. However, this research also demonstrated that the relative importance of these varied by gender and degree of criminal involvement. Future research should aim to replicate and expand on these results by continuing to examine the relationship between gender, personality, additional measures of social background and knowledge about criminal careers.