Vous êtes sur la page 1sur 11

Personality and Individual Differences 30 (2001) 321331


Personality, perceptions of family and peer inuences, and

males' self-reported delinquency$
Patrick C.L. Heaven *, Michael Virgen
University of Wollongong, Department of Psychology, Northelds Avenue, Wollongong, NSW 2522, Australia
Received 24 September 1999; received in revised form 18 January 2000

The aim of this research was to assess the joint inuences of personality factors (extraversion (E), neuroticism (N), and psychoticism (P)), family control (parental inductiveness, punitiveness, and love withdrawal), and delinquent companionship on males' self-reported delinquency. Respondents were two
groups of 13-year-olds (n 110 and n 89). Structural equation modelling showed that personality and
delinquent companionship consistently had direct eects on self-reported delinquency. It is concluded that
this study provides important evidence on the interplay between personality and environmental factors on
delinquency. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Delinquency; Personality; Psychoticism; Companionship; Parental discipline style

1. Introduction
What are the most signicant predictors of self-reported delinquency, personality factors or
environmental inuences such as those involving family and peers? The psychology of delinquency continues to enjoy considerable research attention by behavioural scientists. This research
eort has most notably been dominated by the study of personality inuences (see, for example,
Binder, 1988; Eysenck, 1977; Eysenck & Gudjonsson, 1989; Feldman, 1977; Furnham &
Thompson, 1991; Gudjonsson, 1997; Heaven, 1993, 1996; Lane, 1987; Putnins, 1982; Rigby, Mak
& Slee, 1989; Weaver & Wootton, 1992), although there is growing interest in the role of major

An earlier version of this paper was presented to the 9th Biennial Meeting of the International Society For The
Study Of Individual Dierences, Vancouver, 59 July 1999
* Corresponding author. Tel.: +61-2-4221-3742; fax: +61-2-4221-4163.
E-mail address: patrick_heaven@uow.edu.au (P.C.L. Heaven).
0191-8869/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0191-8869(00)00049-0


P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331

environmental factors such as the eects of family processes (Baumrind, 1967; Conger, Ge, Elder,
Lorenz & Simons, 1994; Deater-Deckard, Dodge, Bates & Pettit, 1996; Lamborn, Mounts,
Steinberg & Dornbusch, 1991; McFadyen-Ketchum, Bates, Dodge & Pettit, 1996; Shaw & Scott,
1991; Peiser & Heaven, 1996) and delinquent peers (see Emler & Reicher, 1995 for a recent integrative approach).
The present research was designed to assess the joint eects of the major personality domains,
perceptions of parental discipline style, and association with delinquent companions on levels of
self-reported delinquency among male youth. As far as we can establish, these joint eects have
not yet been investigated.
1.1. The role of personality
This perspective posits that delinquency is the result of xed and biologically determined personality factors. Most (but certainly not all) research relating personality to delinquency has
adopted the approach of Eysenck (e.g., Eysenck, 1977; Eysenck & Gudjonsson, 1989), who proposed that the major personality domains psychoticism (P), neuroticism (N), and extraversion (E)
are crucial in predicting delinquency and criminality. Those high on E are said to be cortically
under-aroused and are therefore more likely to engage in thrill- and sensation-seeking behaviours. While it is suggested that N is linked to anxiety which acts as a drive ensuring that delinquent behaviours are amplied, high P scorers have been described as anti-social, aggressive,
cold, and unempathic (Claridge, 1981; Eysenck & Eysenck, 1975). Thus, perhaps not surprisingly,
P has been found to discriminate between criminals and non-criminals. Whereas early theorising
gave prominence to the role of E and N, several studies employing both cross-sectional and
longitudinal designs suggest that P is more inuential in predicting criminality (e.g., Gudjonsson,
1997; Heaven, 1993, 1996; Lane, 1987; Putnins, 1982). This may be due, in part, to the fact that
the impulsiveness component of E has been shifted to P, while N seems sensitive to the age of
respondents (Eysenck & Eysenck, 1985; Furnham & Thompson, 1991; Gudjonsson, 1997).
1.2. The role of family processes
It is well established that family functioning aects the emotional and behavioural outcomes of
children (Noller & Callan, 1991). Schaefer (1959) proposed his well known two-dimensional
model of parental behaviour patterns (autonomycontrol vs hostilitylove) suggesting that each
had dierential outcomes for youth. Baumrind (1967) dierentiated dierent parenting styles and
showed that an authoritative style was most likely to lead to behavioural and emotional competencies in children. In an early British longitudinal study (West & Farrington, 1973) it was shown
that parental psychopathology is linked to childrens' maladjustment and delinquency. Likewise,
Patterson and colleagues (Patterson, DeBarsyshe & Ramsey, 1989) reported that anti-social parents tend to raise anti-social teenagers.
In this research we were specically interested in the role of parental discipline style which has
been the focus of recent studies (e.g., Ge, Best, Conger & Simons, 1996; Peiser & Heaven, 1996;
Pettit, Bates & Dodge, 1997; Shaw & Scott, 1991). Grounded in earlier theorising regarding social
control (Hirschi, 1969) and moral development (Homan, 1963), researchers distinguish between
power assertive and psychological techniques of parenting. On this basis, Shaw and Scott (1991)

P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331


constructed three measures of discipline style, namely, punitive, love withdrawal, and inductive.
The punitive style is a power assertive technique and employs physical punishment, whereas the
latter two are described as psychological techniques: love withdrawal incorporates shaming and
assumes that children become anxious over the possible loss of parental love, while inductiveness
emphasises the explanations by the parent about the eects of the child's behaviour on others.
Inductiveness (or calm discussion) has been shown to be negatively related to maladjustment
(including delinquency) among young people (Peiser & Heaven, 1996; Pettit et al., 1997), while a
punitive style predicts anti-social behaviour (Shaw & Scott, 1991). Poor discipline styles, but not
parental warmth, have been shown to be related to conduct disorders among high school youth
(Ge et al., 1996).
1.3. The role of peer inuence
Although criminologists and social psychologists have in the past made ample reference to
group norms and the inuence of `peer pressure' on adolescent behaviours, such an approach has
lacked a coherent theoretical framework. Social identity theory (Tajfel & Turner, 1986) provides
such a focused theoretical approach, and refers to ``. . . those aspects of an individual's self-image
that derive from the social categories to which he perceives himself belonging'' (p. 16). This
approach allows one to explore more deeply and reliably the nature of the groups within which
the individual functions. Similar social identities have `shared cognitive constructs' (Hogg &
Abrams, 1988), implying a sharing of norms, behaviours, attitudes, and values within groups.
According to social identity theory, delinquent behaviours are seen as a natural outcome of the
social categorisation process and ensuing group identity. Members of delinquent groups view
their group as a source of companionship, anonymity, security, reputation management and
behavioural norms (Emler & Reicher, 1995). These functions are closely linked to delinquent
behaviour, and facilitate the context for group members to engage in those sorts of behaviours, to
explore personal relationships, and to strengthen their sense of group belonging. This approach is
therefore diametrically opposed to the view that a particular personality conguration leads to
delinquency. Rather, this approach emphasises the collective character of delinquency.
Recently, Heaven and colleagues (Heaven, Caputi, Trivellion-Scott & Swinton, 2000) found
support for at least two dimensions, namely, delinquent norms and delinquent companionship
among two samples of Australian high school students. Reliable and valid scales were constructed
to tap these constructs. In both samples, psychoticism was found to have signicant direct eects
on delinquency, while the indirect eects through companionship were found to be weaker.
1.4. Aims and rationale of the present study
The present research sought to assess the extent to which three clusters of psychological variables predict males' self-reported delinquency. These include personality (P, E, and N), perceptions of parental discipline style (inductiveness, punitiveness, and love withdrawal), and
delinquent companionship. The following hypotheses guided the study:
1. Psychoticism will be signicantly related to males' self-reported delinquency (Gudjonsson,
1997; Heaven, 1996; Lane, 1987).


P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331

2. Delinquent companionship will be signicantly related to males' self-reported delinquency

(Emler & Reicher, 1995).
3. Perceptions of inductiveness will be signicantly negatively related to males' self-reported
delinquency (Peiser & Heaven, 1996; Pettit et al., 1997).
4. Perceptions of punitiveness will be signicantly positively related to males' self-reported
delinquency (Shaw & Scott, 1991).
It is also reasonable to argue that personality factors and perceived family inuences may be
exacerbated by a youth's interactions with a particular group. Thus, delinquent friends may play
a mediating role on delinquent behaviour patterns in those cases where youth score high on the P
dimension or have experienced a punitive discipline style. These relationships, which will be tested
below, are illustrated in Fig. 1.
2. Method
2.1. Participants
Respondents were groups of adolescent males attending two independent Catholic schools in
dierent cities of New South Wales, Australia. Sample 1 comprised 110 boys, while sample 2
comprised 89 boys. Both groups of boys were volunteers from Year 8 classes and their ages ranged from 12 to 15 years (M age=13.49, SD=0.60). It is not suggested that these samples are
representative of Australian adolescent males.

Fig. 1. Model showing hypothesised links between predictor variables and delinquency.

P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331


2.2. Materials
Students were asked to complete a test booklet containing the following measures:
1. Eysenck personality questionnaire (Eysenck & Eysenck, 1975): Corulla's (1990) revision of
this well known instrument suitable for children in their young teens was used. It measures
extraversionintroversion (E), neuroticismemotional stability (N), and psychoticismsuperego (P). In addition, it contains a lie scale to measure dissimulation. Alpha coefcients for samples 1 and 2, respectively, were E=0.84 and 0.77; P=0.76 and 0.62; N=0.81
and 0.72; lie=0.73 and 0.52.
2. Parental discipline style (Shaw & Scott, 1991): This instrument assesses perceptions of
inductiveness, punitiveness, and love withdrawal. Past studies have demonstrated excellent
internal consistencies (e.g., Shaw & Scott, 1991; Peiser & Heaven, 1996). Alpha coecients
for samples 1 and 2, respectively, were punitiveness=0.87 and 0.77; inductiveness=0.84 and
0.80; love withdrawal=0.81 and 0.74.
3. Delinquent companionship (Heaven et al., 2000): This 9-item scale, based on the qualitative
data of Emler and Reicher (1995), is highly internally consistent and has demonstrated
concurrent validity (sample item: ``Doing bad things with my friends is really exciting''). On
the present occasion, alpha coecients were 0.89 (sample 1) and 0.92 (sample 2).
4. Self-reported delinquency (Gold, 1970): This 14-item scale measures two forms of delinquency, namely, interpersonal violence and vandalism/theft and has been demonstrated to
have acceptable psychometric properties among Australian samples (Shaw & Scott, 1991).
Two additional items were added to broaden the scope of the instrument, namely, ``Have
you ever smoked a cigarette?'' and ``Have you ever drunk alcohol?'' All items were answered
on a 4-point Likert scale from never (scored 1) to often (scored 4). Cronbach's coecient
alpha was 0.89 (sample 1) and 0.92 (sample 2).
2.3. Procedure
After obtaining parental as well as student consent, questionnaires were administered by teachers during regular class times. In order to standardise the testing conditions, teachers were
provided with written instructions which were read to the students at the commencement of
testing. Students were asked to answer each question as honestly as possible and reminded that
there are no correct or incorrect answers. They were informed that all information would be
treated in the strictest condence and used by the researchers and not their school. They were
asked not to write their name on the booklet. Teachers were also provided with a standardised
debrieng sheet to be used once all questionnaires had been completed.
3. Results
Participants who scored more than two standard deviations above the mean on the lie scale
were eliminated from their respective samples. Ten students were eliminated for sample 1, while
four were eliminated from sample 2.


P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331

3.1. Mean scores on dependent variable

The mean scores (and standard deviations) on the delinquency measure for samples 1 and 2
were 24.61 (6.67) and 28.92 (10.04), respectively. Given that this dierence is signicant
[t(185)=3.36, p < 0:001], further analyses were done on the two groups separately.
3.2. Correlations
Table 1 presents the correlations between the variables for both samples. Delinquency was
signicantly related to psychoticism for both groups as well as to companionship and love withdrawal (all ps < 0:01). Delinquency was signicantly related to punitiveness (r 0:61, p < 0:01)
and neuroticism (r 0:28, p < 0:01) in sample 1 only. Self-reported delinquency was not signicantly related to inductiveness for both samples. None of the other personality or discipline
style measures were signicantly related to delinquency.
3.3. Structural equation modelling
In order to test the model proposed in Fig. 1, each data set was subjected to a covariance
structure analysis (SPSS, 1999 Amos 3.6). Table 2 presents the results for the proposed
model for both samples and lists the goodness of t index (GFI), adjusted goodness of t index
(AGFI), comparative t index (CFI) and chi-square. A good t of the data to the model is indicated by the GFI, AGFI and CFI being as close to 1 as possible (usually >0.90) and chi-square
not being signicant (Hair, Anderson, Tatham & Black, 1992). It is clear that, in both cases,
model t was weak with both chi-squares highly signicant (both ps < 0:0001) and the GFIs
being too low. The inclusion of either neuroticism, extraversion or love-withdrawal as single
additional variables in both samples did little to improve overall t. Indeed, in each case model t
was weakened. Table 2 shows the changes in indices when adding these variables to the proposed model. The indices of the best-tting models (indicated as nal model) are also shown in
Table 2 and illustrated in Figs. 2 and 3. In both samples P was found to have a signicant direct
eect on self-reported delinquency as well as an indirect eect through companionship. However,
Table 1
Partial correlations among variables for samples 1 and 2a

1 Companionship
2 Delinquency
3 Extraversion
4 Inductiveness
5 Love withdrawal
6 Neuroticism
7 Psychoticism
8 Punitiveness















Lie scores have been partialled. Sample 1 is given below the diagonal.  p < 0:05,  p < 0:01.

P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331


Table 2
Results of structural equation modelling for sample 1 and sample 2a

Sample 1
Proposed model
Add: N
Add: E
Add: Love withdrawal
Final model
Sample 2
Proposed model
Add: N
Add: E
Add: Love withdrawal
Final model


















the indirect eects of P were observed to be weaker than the direct eects (0.14 in sample 1 and
0.16 in sample 2). As expected, companionship had direct signicant eects on delinquency
in both samples. In sample 1 perceptions of punitiveness had direct eects on delinquency
(0.31) while in sample 2 punitiveness had a weak indirect eect (0.08) on delinquency through

Fig. 2. Model showing links between predictor variables and self-reported delinquency in sample 1 (n 110).


P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331

Fig. 3. Model showing links between predictor variables and self-reported delinquency in sample 2 (n 89).

4. Discussion
The analyses of both data sets conrm the importance of P as a predictor of males' selfreported delinquency with signicant direct eects being observed in both samples. This nding
appears quite robust even in the presence of several other factors known to be associated with
self-reported delinquency. Neither N nor E were found to have signicant direct eects on
delinquency and their inclusion only led to a deterioration of model t. These results therefore
conrm several previous reports which have indicated the crucial role that the P factor plays in
delinquent and criminal behaviour (e.g., Eysenck & Eysenck, 1985; Eysenck & Eysenck, 1976;
Eysenck & Gudjonsson, 1989; Gudjonsson, 1997). Thus, for example, male and female prisoners
have been found to have higher mean P scores than matched groups of controls (Eysenck &
Eysenck, 1976), while high P-scoring adolescents found it dicult to control their temper (Furnham & Gunter, 1983). Likewise, it has also been noted that P discriminates after ve years those
young people who do not re-oend and those who do (Sigurdsson, Gudjonsson & Peerson, in
Perceptions of a punitive discipline style were here found to be related to delinquency in both
samples, although only in sample 1 did punitiveness have a direct eect on the dependent measure. This supports the general thesis (Baumrind, 1967; Noller & Callan, 1991; Schaefer, 1959)
that family climate plays a critical role in shaping the emotional and behavioural adjustment of
adolescents. This also supports the research ndings of others who have shown that a punitive
discipline style is directly related to high levels of self-reported delinquency (Shaw & Scott, 1991),
and is in line with those authors who have argued that poor discipline styles in parents are related
to poor outcomes in children (Ge et al., 1996).

P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331


The punitive discipline technique operates on the parent's physical power over the child. Many
adolescent delinquents may nd it easier or more convenient to be in harmony with the delinquent group than with their parents. For them their group is a source of comfort as well as
companionship, security, and anonymity. If they believe their reputation to be irredeemable at
home (as manifest by punitive disciplining), they believe it can be salvaged within the delinquent
group (Emler & Reicher, 1995).
As predicted, endorsement of delinquent companionship was signicantly and directly related
to delinquency in both groups. This supports the argument that the extent to which youth identify with delinquent groups is predictive of their levels of delinquency (Emler & Reicher, 1995).
These results accord with the view that delinquency is not just the function of internal processes,
but is also a group process which requires the co-operation or collusion of like-minded others,
and a behaviour that reects one's identity with particular groups. The data support the view that
identifying with delinquent companions is a powerful shaper of the values and behaviours of
The important contribution of the present research is the clear demonstration of the extent to
which personality, family, and peer group factors interact in inuencing delinquency. Although
the importance of each has previously been demonstrated, these data are the rst to show their
relative strengths and interconnections. Quite clearly, each set of variables is an important contributor to delinquency, yet their joint eects are quite complex. Both models suggest that the role
of P is quite powerful and that companionship also plays an important mediating role.
The present research is not without its limitations. First, the samples are small and selective.
Thus, future research will need to validate the models generated here on larger and more representative samples. Moreover, as these models have been generated on male samples an important
question that arises is the extent to which they apply to samples of young females. Finally, as
these data are cross-sectional in nature, future research will also need to address the extent to
which these models are valid using longitudinal research designs.
In conclusion, delinquency is multifaceted. These data have clearly shown the extent to which
behaviours of this sort result from the interplay of individual and environmental factors.
Baumrind, D. (1967). Child care practices anteceding three patterns of preschool behavior. Genetic Psychology Monographs, 75, 4388.
Binder, A. (1988). Juvenile delinquency. Annual Review of Psychology, 39, 252282.
Claridge, G. (1981). Psychoticism. In R. Lynn, Dimensions of personality: Papers in honour of H. J. Eysenck (pp. 79
109). Oxford: Pergamon.
Conger, R. D., Ge, X., Elder, G. H., Lorenz, F. O., & Simons, R. L. (1994). Economic stress, coercive family process,
and developmental problems of adolescents. Child Development, 65, 541561.
Corulla, W. J. (1990). A revised version of the Psychoticism scale for children. Personality & Individual Dierences, 11,
Deater-Deckard, K., Dodge, K. A., Bates, J. E., & Pettit, G. S. (1996). Physical discipline among African American
and European American mothers: Links to children's externalizing behaviors. Developmental Psychology, 32, 1065
Emler, N., & Reicher, S. (1995). Adolescence and delinquency. Oxford: Blackwell.
Eysenck, H. J. (1977). Crime and personality. London: Routledge & Kegan Paul (revised edition).


P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331

Eysenck, H. J., & Eysenck, M. (1985). Personality and individual dierences: a natural science approach. New York:
Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck Personality Questionnaire. London: Hodder and
Eysenck, H. J., & Eysenck, S. B. G. (1976). Psychoticism as a dimension of personality. London: Hodder & Stoughton.
Eysenck, H. J., & Gudjonsson, G. (1989). The causes and cures of criminality. New York: Plenum.
Feldman, M. (1977). Criminal behaviour: A psychological analysis. Chichester: John Wiley.
Furnham, A., & Gunter, B. (1983). Sex and personality dierences in self-reported social skills among British adolescents. Journal of Adolescence, 6, 5769.
Furnham, A., & Thompson, J. (1991). Personality and self-reported delinquency. Personality & Individual Dierences,
12, 585593.
Ge, X., Best, K. M., Conger, R. D., & Simons, R. L. (1996). Parenting behaviors and the occurrence and co-occurrence
of adolescent depressive symptoms and conduct problems. Developmental Psychology, 32, 717731.
Gold, M. (1970). Delinquent behavior in an American city. Belmont, CA: Brooks/Cole.
Gudjonsson, G. (1997). Crime and personality. In H. Nyborg, The scientic study of human nature: tribute to Hans J.
Eysenck at eighty (pp. 142164). Oxford and New York: Pergamon.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1992). Multivariate data analysis with readings (3rd ed.).
New York: Macmillan Publishing Company.
Heaven, P. C. L. (1993). Personality predictors of self-reported delinquency. Personality & Individual Dierences, 14,
Heaven, P. C. L. (1996). Personality and self-reported delinquency: a longitudinal analysis. Journal of Child Psychology
& Psychiatry, 37, 747751.
Heaven, P. C. L., Caputi, P., Trivellion-Scott, D., & Swinton, T. (2000). Personality and group inuences on selfreported delinquent behaviour. Personality & Individual Dierences, 28, 11431158
Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press.
Homan, L. (1963). Childrearing practices and moral development: Generalisations from empirical research. Child
Development, 34, 295318.
Hogg, M. A. & Abrams, D. (1988). Social identications: a social psychology of intergroup relations and group
processes]. London: Routledge.
Lamborn, S., Mounts, N., Steinberg, L., & Dornbusch, S. (1991). Patterns of competence and adjustment among
adolescents from authoritative, authoritarian, indulgent, and neglectful families. Child Development, 62, 10491065.
Lane, D. (1987). Personality and anti-social behaviour: a long-term study. Personality & Individual Dierences, 8, 799
McFadyen-Ketchum, S. A., Bates, J. E., Dodge, K. A., & Pettit, G. S. (1996). Patterns of change in early childhood
aggressive-disruptive behavior: Gender dierences in predictions from early coercive and aectionate mother-child
interactions. Child Development, 67, 24172433.
Noller, P., & Callan, V. (1991). The adolescent in the family. London: Routledge.
Patterson, G., DeBarsyshe, B., & Ramsey, E. (1989). A developmental perspective on antisocial behavior. American
Psychologist, 44, 329335.
Peiser, N., & Heaven, P. C. L. (1996). Family inuences on self-reported delinquency among high school students.
Journal of Adolescence, 19, 557568.
Pettit, G. S., Bates, J. E., & Dodge, K. A. (1997). Supportive parenting, ecological context, and children's adjustment:
A seven-year longitudinal study. Child Development, 68, 908923.
Putnins, A. (1982). The Eysenck Personality Questionnaires and delinquency predictions. Personality & Individual
Dierences, 3, 339340.
Rigby, K., Mak, A., & Slee, P. (1989). Impulsiveness, orientation to institutional authority and gender as factors in
self-reported delinquency among Australian adolescents. Personality & Individual Dierences, 10, 689692.
Schaefer, E. (1959). A circumplex model for maternal behavior. Journal of Abnormal & Social Psychology, 59, 226235.
Shaw, J., & Scott, W. A. (1991). Inuence of parent discipline style on delinquent behaviour: The mediating role of
control orientation. Australian Journal of Psychology, 43, 6167.

P.C.L. Heaven, M. Virgen / Personality and Individual Dierences 30 (2001) 321331


Sigurdsson, J. F., Gudjonsson, G. H., & Peerson, M. (2000). Dierences in the cognitive ability and personality of
desisters and re-oenders: A prospective study among juveniles. Psychology, Crime & Law (in press)
SPSS (1999). SPSS/PC, version 9.0. Chicago: SPSS, Inc
Tajfel, H., & Turner, J. (1986). The social identity theory of intergroup behaviour. In S. Worchel, & W. G. Austin,
Psychology of intergroup relations (pp. 724). Chicago: Nelson.
Weaver, G., & Wootton, R. (1992). The use of the MMPI special scales in the assessment of delinquent personality.
Adolescence, 27, 545554.
West, D., & Farrington, D. (1973). Who becomes delinquent? Second report of the Cambridge study in delinquent development. London: Heinemann.