Use this URL to cite or link to this record in EThOS:
Title: Behaviour and personality in delinquent children
Author: Shapland, Joanna
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 1976
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Please try the link below.
Access through Institution:
The thesis is concerned with aspects of personality that may distinguish those children from a high delinquency area who become delinquent from those who do not. In Chapter 1 the limitations this approach places on the selection of the area and the sample to be studied are discussed. The two main methods of measuring criminality, convictions and self-report delinquency, are compared as to their usefulness for the study. The theoretical background for the choice of the personality variables to be considered is given. Two of these, susceptibility to reward and susceptibility to punishment, derive from operant conditioning tasks in which the subject presses a lever to obtain reward. Tests possibly measuring susceptibility to reward were stimulus generalization (from an intradimensional discrimination) and performance under differential reinforcement of low rates of response. In children, tests of susceptibility to punishment must be tests of susceptibility to frustrative nonreward (considered equivalent to punishment in its behavioural effects). In the present study the ones used were behavioural contrast on both extradimensional and intradimensional discriminations and a measure of peak shift. Questionnaire measures of extraversion, neuroticism and psychoticism were also used. The main hypothesis of the present study is that delinquents should be more susceptible to reward and less susceptible to punishment than nondelinquents. This may be tested in three ways: The first is to administer tests of susceptibility to reward and to purishment (the behavioural measures), and tests of delinquency (self-report delinquency) and to correlate the results of the two. The second is to correlate the scores of the children on tests of susceptibility to reward and to punishment and on personality questionnaires. The prediction (from Gray, 1973) is that susceptibility to reward should be correlated with extraversion and neuroticism and susceptibility to punishment with introversion and neuroticism. The third is to correlate delinquency scores with personality questionnaire scores, with the prediction that delinquents should be more extravert but not more neurotic than nondelinquents. All three ways are used in the present study. The total sample of the present study consisted of 54 school boys aged 11.1 to 12.9 years at the time of first testing, from a working-class housing estate. They were first given the personality questionnaires and delinquency tests (test administration). 31 boys (the behavioural sample) were selected from the total sample so that a full range of delinquency scores was obtained and given the behavioural tests of susceptibility to reward and to punishment. After this, the total sample was then given the personality questionnaires and delinquency tests again (retest administration). A study performed by Dr. J.P. Eushton on younger children aged 7 to 11 years was designed to run parallel with the present study. Comparisons are made with this study and with the previous work of Nicholson (1972) and de Wit (unpublished: described in Appendix C) throughout the thesis. Chapter 2 is concerned with the description of the personality questionnaires used and their reliability and intercorrelations. The questionnaires used were the Junior Eysenck Personality Inventory (JEPI), the High School Personality Questionnaire (HSPQ), the Junior Personality Questionnaire (JPQ) and a Teacher's Rating Scale (TRS: Nicholson, 1972), in which the form teachers of the boys rated the boys in their year. These give measures of extraversion, neuroticism, psychoticism and tendency to fake. Chapter 3 is concerned with the measurement of delinquency by the self-report method. A review of the rationale "behind the method and the various methodological problems encountered is given. It is also shown that delinquency is heterogeneous so that it is necessary to measure participation in various aspects of crime, for example, group versus solitary crime. The self-report interview used in the present study is described. This produces a measure of number of different types of crime committed and measures of involvement in group crime, solitary crime, self-suggested crime and other-suggested crime. The interview was shown to have high test-retest reliability and internal consistency. Principle components analysis produced a large first component, loading on all the delinquency items. Inconsistent responders between the test and retest administration of the interview were identified and inconsistent responding shown to relate positively to the JPQ, lie scale and negatively to number of crimes admitted. The third way of testing the main hypothesis, the correlations between personality and crime, is examined in Chapter 4. A review of the studies used to test Eysenck's theory of the relationship between personality and crime is given. Significant positive correlations of number of delinquent acts admitted with both extraversion and psychoticism were found. No such correlations were found with neuroticism. The hypotheses of Eysenck (1974) and Burgess (1972a) that the extravert neurotic quadrant should contain more criminals than other quadrants in the Eysenckian two-dimensional space was not supported. The behavioural measures and their correlations with the personality questionnaire scores and crime scores (the first and second ways of testing the main hypothesis) are discussed in Chapters 5 to 8. Chapter 5 deals with the phenomenon of behavioural contrast and its use to measure susceptibility to frustrative nonreward. A review of the animal and human literature is given, in which it is concluded that the frustration theory of behavioural contrast is the best currently available. The extra-dimensional and intradimensional discriminations of the present study are then described, together with possible controls for the behavioural contrast shown. Measures of the time taken to reach the baseline criterion, the time taken to be able to describe the discrimination conditions and the time then required to perform the discrimination to criterion were also used. The correlations with introversion and neuroticism found by Nicholson (1972) were not replicated with all the personality measures used in the present study. Chapter 6 discusses the phenomena occumng in the generalization test given after the intradimensional discrimination; peak shift, mean shift and stimulus generalization. A review of the animal and human literature on these phenomena is given. Mean shift, although shown by 18 out of 24 boys tested, was not found to be correlated with either introversion or neuroticism as found by Nicholson (1972). Stimulus generalization, however, did show some correlations with extraversion and neuroticism. Chapter 7 is concerned with the differential reinforcement of low rates of response tasks. A review of these tasks in am'mis and humans with regard to the use of a series of progressively more difficult tasks in measuring susceptibility to reward is given. The results of the present studies tend to indicate that the first task in the series correlates with developmental measures such as age, whereas the second and third tasks correlate more with the personality measures, notably extraversion and neuroticism. The third task (and, in the present study, the second task) may hence be used as a measure of susceptibility to reward. Chapter 8 discusses the correlations between all the measures taken. The two measures of susceptibility to reward show some moderate positive correlations.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Juvenile delinquency ; Juvenile delinquents ; Identification