There is a growing body of international literature that report high levels of language impairment in young offenders which is disproportionate when compared to the typical population. This data also indicates that language impairments have previously gone undetected, suggesting young offenders are not considered to be particularly vulnerable to language and communication difficulties by organisations they come into contact with.
Studies report favourable outcomes regarding rates of recidivism and victim satisfaction following a Restorative Justice Conference. However no studies have been carried out which aim to investigate the role of offenders' language abilities in the outcomes of the Restorative Justice Conference. The project will provide key information regarding the role of language impairment and other factors in the Restorative Justice Conference ensuring that a modern creative strategy has its most positive potential impact in society.
The aims of the study are:
1. To examine the linguistic, psychological and environmental profiles of youth offenders.
2. To ascertain the role of offenders’ oral language abilities in the Restorative Justice Conference process. Specifically,
3. To determine the key factors associated with higher rates of re-offending.
The project will be undertaken within a mixed methods framework involving direct assessments, observational schedules, questionnaire and interview protocols as well as examination of records.
Standardised assessments and tasks will be utilised and an observational schedule will be developed. Victim Satisfaction will be measured by means of a short questionnaire which the victim will be asked to complete.
It is anticipated that basic descriptive statistics will be used (e.g. frequencies, percentages, means, standard deviations, medians) followed by formal independent samples significance tests (contingency table analyses, analyses of variance or, depending on the nature of the distributions Mann-Whitney U-tests). The results will be presented as percentages, mean or median differences (with p values, confidence intervals and effect sizes). Regression analyses will be used to examine group differences where appropriate and in particular when co-variance of particular variables in desirable (for example, co-varying nonverbal skills when examining language).
Statistical power: Power calculations assuming 75 participants per group and an alpha of 0.5 suggests the study has over 80% power to detect small effect size for simple group comparisons and preserves 80% power for regression analyses for detection of medium effect sizes.