Background and aim

1. Heaven (1996) is the study that put Eysenck's Criminal Personality Theory to the test.
2. Eysenck's claim: extraversion, neuroticism and psychoticism are associated with criminal behaviour.
3. Heaven wanted to test that claim by measuring the traits before the offending rather than at the same time, so he followed the same teenagers for two years.
4. Prediction: psychoticism, extraversion and self-esteem would all be significant predictors of self-reported delinquency.
5. Self-esteem was included because it is associated with neuroticism, which gave Heaven a way of testing that trait indirectly.

Method

1. Design: a longitudinal correlational study run over two years.
   1. Traits were measured first and then correlated with self-reported delinquency at the end.
2. Sample: 282 thirteen to fifteen year olds from two Catholic independent schools in Australia.
   1. About half the sample was male, so male and female delinquency scores could be compared.
   2. 80% of the original participants were surveyed again two years later using the same questionnaires.

Materials and procedure

1. Materials: anonymous questionnaires using closed questions, including rating scales, measured psychoticism, extraversion and self-esteem.
2. Delinquency was measured the same way, so participants scored their own offending.
3. Procedure: participants completed the questionnaires at the start, then completed the same measures two years later.

Results

1. Psychoticism: a positive correlation with delinquency at both the start and the end of the two years.
2. Extraversion: only a weak positive correlation with delinquency, and only at the end.
3. Self-esteem: no significant correlation with delinquency at either point.
4. Psychoticism was the best predictor of delinquency out of the three measures.
5. Males had higher delinquency scores than females at both time points.

Conclusions

1. Personality traits are associated with delinquency, and psychoticism is associated with it most strongly.
2. The relationships were modest and self-esteem predicted nothing, so Eysenck's theory receives only limited support.
3. Because the design is correlational, the study cannot show that personality causes delinquency.
4. The study backs the psychoticism strand of Eysenck's account, and because self-esteem stood in for neuroticism, that strand came out unsupported.

Criticisms

1. Strengths: three things the study does well.
   1. The two-year longitudinal design with 80% retention placed the personality measure before the delinquency measure, which a one-off study cannot do.
   2. Standardised, anonymous, closed-question questionnaires are easy to repeat and quantify, so reliability is high.
   3. A sample of 282 with a near even sex split allows a meaningful comparison between males and females.
2. Weaknesses: four things that limit what the study can tell you.
   1. The sample was narrow, drawn from one country, one religion and two independent schools, and other cultures treat delinquency differently.
   2. Scoring your own offending invites social desirability and exaggeration, with no objective check to correct it.
   3. The correlational design leaves the direction open, so psychoticism could be a consequence of a delinquent lifestyle rather than its cause.
   4. Only 13 to 15 year olds were studied, so the findings cannot be stretched to adult offending.

Turning this into marks

1. When two findings are asked for, keep them genuinely separate, meaning psychoticism correlating at both time points and self-esteem correlating at neither, rather than one finding rephrased.
2. Name the design as a two year longitudinal correlational study, because writing that Heaven used questionnaires does not secure the knowledge mark.
3. The mark most often lost is writing that Heaven showed personality causes delinquency, when a correlational design leaves the direction of the relationship open.