Evaluation strategies to assess the effectiveness of legal knowledge based systems enable strengths and limitations of systems to be accurately articulated. This facilitates efforts in the research community to develop systems and also promotes the adoption of research prototypes in the commercial world. However, evaluation strategies for systems that operate in a domain as complex as law are difficult to specify. In this paper, we present an evaluation framework put forward by Reich and describe how this motivated the evaluation of our systems in Australian family law. Strategies surveyed include a comparison of linear regression with neural networks, user acceptance surveys, a comparison of system predictions with those from past cases, and a comparison of system outputs with those proposed by a panel of lawyers. Specific criteria for the evaluation of explanation facilities are also described.

The evaluation of legal knowledge based systems

La recherche en sentencing : un survol contextualisé

Drawing on theoretical insights from basic psychological research, influences sourced in unanticipated or extra-legal aspects of criminal cases that can influence sentencing decisions are reviewed. Findings from empirical studies of observed disparities in sentencing decisions are summarized, including archival data, observational studies, field research and experimental investigations of simulated sentencing decisions. A criterion for inclusion was that the factor had to exert an influence outside the conscious awareness of the judge. The paper examines the effects on sentencing of some traditional extra-legal factors, such as judge and offender gender, as well as less traditional elements, such as the judge's attitudes and sentencing philosophy, the physical appearance of the offender, unforeseeable consequences of the crime, and reactions to acts of terrorism. Four sources of potential bias are distinguished: (a) features associated with the judge; (b) features associated with the offender, (c) unantic...

Unconscious influences in sentencing decisions: a research review of psychological sources of disparity

Defences to Homicide Final Report

This thesis explores the use and utility of suspended sentences in Tasmania.
Suspended sentences are currently available as a sentencing disposition in all
Australian jurisdictions, but there is conflict between public and legal perceptions
about the severity of the sanction, which may contribute to a lack of public
confidence in the criminal justice system as a whole. The following issues are
explored: How are suspended sentences used by judges and magistrates? What is the
process for imposing a suspended sentence? How effective are suspended sentence as
a specific deterrent or rehabilitative measure? How are breaches dealt with?
The study examines the history of suspended sentences and the arguments for and
against their use, as well as considering the principles and practice governing their
use in Australia and overseas. In-depth interviews with Tasmanian judges and
magistrates provide an invaluable source of information on judicial views on a range
of issues pertaining to the use of suspended sentences. A quantitative analysis
presents empirical information on the use of suspended sentences in the Supreme and
Magistrates’ Courts, while a qualitative analysis of sentencing remarks in the
Supreme Court examines the relevance of a range of sentencing factors to the
decision to suspend. The findings of a reconviction analysis are presented, indicating
that suspended sentences may be an effective deterrent. Finally, a breach analysis
examines offending conduct in breach of a suspended sentence, explores prosecution
practices in respect of breaches and analyses judicial sentencing remarks in breach
cases. The conclusion reviews the key findings and discusses how they support the
main arguments for and against suspended sentences. The practical and policy
implications of my findings are also considered. My research is not only of
significance for Tasmania, but also has relevance and resonance for the Australian
and international use of suspended sentences and will inform broader discussions
about the utility of this sentencing option.

Sword or Feather?The use and utility of suspended sentences in Tasmania

List of figures List of tables Acknowledgments 1. Judicial decision making and sentencing policy: continuation of a study 2. A sentencing decision model: single and multiple similar counts 3. A sentencing decision model: multiple disparate counts 4. Testing the decision model for multiple disparate counts 5. The techniques of data collection 6. Judges' thoughts on sentencing the multiple offender 7. An alternative sentencing decision model for the multiple offender 8. Validity and development of the alternative decision model: the data collection 9. Towards a requisite decision model for sentencing the multiple offender 10. The armature of judicial sentencing Appendix: Case 37 from Sentencing Research Exercise - Part 3B References Index.

The Framework of Judicial Sentencing: A Study in Legal Decision Making

Systems providing statistical information on past sentencing practices for cases in respect of certain offence and offender characteristics are creating worldwide interest and yet have not been subject to rigorous analysis and evaluation. This article attempts this, using the systems in Scotland and New South Wales as an illustration. The proper use of SISs is explained, followed by an assessment of their usefulness as an aid to the sentencing discretion where there is individualized justice. In this respect they are found wanting. Following this, there is an analysis of the factors, such as the intercorrelatedness of the underlying data bases, acting to limit the validity of the information provided by SISs. Corrective measures for some of these deficiencies are proposed. Finally, it is concluded that these systems do not inform the sentencing process with greater rationality, the principal reason being that current sentencing practice is not the product of a policy for consistency of approach. Nevertheless, if SISs were part of a sentencing system in which judges applied a well-developed policy, then these systems might have the potential to promote justice. But it must be policy reflecting decision making in sentencing as an analytic, not a holistic process.

Austin Lovegrove

Papers

The Framework of Judicial Sentencing: A Study in Legal Decision Making

Statistical information systems as a means to consistency and rationality in sentencing

The techniques of data collection

A sentencing decision model: multiple disparate counts