[Home] [Databases] [WorldLII] [Search] [Feedback] New Zealand Law Commission

[Database Search] [Name Search] [Previous] [Next] [Download] [Help]

Appendix C

Law Commission Right of Silence Surveys

IN THE COURSE of preparing Criminal Evidence: Police Questioning (NZLC PP21, 1992), the Commission decided that empirical data on the exercise of the right of silence before and during trial was needed before it could with any confidence advance reform proposals. The Commission wished to test some of the assumptions commonly made about the exercise and effect of the right of silence, before it publishes its final report on the right of silence and confessions.

The three surveys were designed to test the following assumptions commonly made about the right of silence:

• The exercise of the right of silence prior to or during trial increases the likelihood of an acquittal.
• The right of silence is invoked by guilty people to evade conviction.
• The right of silence leads to an undue number of wrongful acquittals.
• The right of silence prevents the detection of crime.

Jury trial survey

The Commission conducted a survey of all judges involved in jury trials in New Zealand for a three-month period in 1992. Survey forms were sent to every High Court Registry and District Court in the country for completion. There was a 92.4% response rate, and there were 398 defendants in the survey. The aim of the research was to consider how often defendants exercise the right of silence both before and during the trial, and the impact on conviction or acquittal of exercising the right. The research was designed to consider the impact of judicial comment on the exercise of the right of silence. The research also considered the relationship between the nature of the offence and the exercise of the right of silence, as well as the relationship between the nature of the offence and judicial comment. The survey responses were coded and inputted on a Topic search and retrieval full text database. In order to evaluate the significance of relationships in the survey, chi-square tests were undertaken. Chi-square is a statistical test which examines the relationship between two variables. A significant result is one where the two variables are found to be related.

Crown solicitor survey

With the assistance of the Crown Law Office, a survey was conducted in March and April 1992 for all trials in 11 District Courts and 7 High Courts. Crown solicitors recorded the information, the main aim of which was for the Commission to be in a position to compare the overall conviction rate with the conviction rate in those cases where the defendant:

• responded or did not respond to police questioning, or
• testified or did not testify, or
• responded to police questioning and testified, or
• did not respond and did not testify.

The Commission received a total of 376 responses. A total of 91 responses (24.2%) were invalid. Again, the survey responses were coded and inputted on a Topic search and retrieval full text database, and chi-square tests were conducted.

Police station survey

The third survey comprised interviews with suspects at the Christchurch and Henderson, Auckland police stations from June to December 1992. The survey was designed to discover how often the right of silence was exercised by police suspects in response to police questioning. Questions elicited the impact of demographic and other factors on statements made by suspects during interviews with police, the length of interviews and breaks, the timing of cautions by police, the requests for and provision of legal advice amongst different ethnic groups and the link between the age, gender and previous criminal record of the suspect. There were 312 usable questionnaires from the total of 322. The questionnaires were coded and then entered into an MS Access database. Validation checks were included in the data entry forms to ensure the accuracy of the transfer from the questionnaires to the database. Most of the data analysis was carried out using the Pivot Table tool in MS Excel 5. This enables rapid calculation of frequencies and cross-tabulations of nominal variables. The database sorting function in the spreadsheet was used to determine the relative time when specific events took place. Chi-square tests were conducted.