Legal Aid Research Series Court Side Study of Adult Unrepresented Accused in the Provincial Criminal Courts Part 1: Overview Report

2. Methodology

2. Methodology

2.1 Site Selection

A key initial task of this project was to select nine court sites that would, taken together, provide an understanding of the range of issues - and the range of data collection opportunities and challenges - that one would confront in researching the current situation with respect to legal representation in Canada. The rationale followed in selecting sites for this project was to provide a wide spectrum of communities where criminal cases are adjudicated, from large urban centres, to major areas within large urban centres, to medium-sized cities, to towns. Rural and isolated areas, and areas served by circuit courts were not part of this study, being the subject of a separate inquiry.

The first step was to develop a set of criteria to guide the selection process. These criteria are shown in Figure 2.1 below.

Figure 2.1 Site Selection Criteria

Application of these criteria led Department of Justice officials, in concert with members of the Federal-Provincial-Territorial Permanent Working Group on Legal Aid, to select the following nine court locations, one in each of nine of the ten provinces of Canada:[7]

  • Kelowna, British Columbia
  • Edmonton, Alberta
  • Regina, Saskatchewan
  • Brandon, Manitoba
  • Scarborough (in eastern Greater Toronto), Ontario
  • Sherbrooke, Quebec
  • Bathurst, New Brunswick 
  • Halifax, Nova Scotia
  • St. John’s, Newfoundland

Figure 2.2 illustrates the variation captured within the selected sites. Other important dimensions on which sites exhibit further variations – and similarities – are discussed in the chapters that follow (e.g., with respect to the organizational and operational structures through which the duty counsel function is provided).

2.2 Data Collection Methodology

The Methodology of the study consisted of three major independent quantitative and qualitative data collection techniques and sources of information – which in turn allowed us to triangulate and cross-check findings on specific issues in each court. Specifically, in each site we collected and analyzed data from:

A disposed cases sample

Extraction of empirical data from court records on a sample of cases recently disposed in the court.

A direct court observation sample

Systematic observation of case appearances in up to 10 full court dockets during the study period.

Structured interviews

with key informants.

2.2.1 Disposed Cases Sample

The initial research design called for data to be collected on a sample of roughly 500 recently disposed cases in each site. Implementation of the research design varied according to local circumstances. The actual approach followed was developed with the advice and considerable co-operation and assistance of justice officials in each province.

For each of the disposed cases in the samples, data was extracted into an electronic file for separate analysis by the researchers.  The databases consisted of separate data for each appearance of each case in the sample – with that data covering key characteristics of the accused and the case, the key events that occurred at each appearance (e.g., show cause hearings, pleas, elections), the status of the case at each appearance (including legal representation, custody status, and plea) and key outcomes at each appearance (e.g., verdicts and sentences).

A common core set of general protocols was used to select the samples of cases and the types of data from each case.  However, the specific rules used were necessarily adapted to fit the unique characteristics of the data capture and storage systems of each province and site.  In two provinces, there was an automated data system which was suited for the purpose, while in two other provinces the automated data system had to be supplemented by manually captured data on key factors (such as representation at each appearance), and in the remaining five provinces an entirely manual data capture method proved to be the most cost-effective and timely approach.

The quality of the data obtained was significantly enhanced by ensuring that all manual data collection and all extraction of data from automated systems was undertaken by persons with extensive knowledge of local court processes and local and provincial manual and court automated information systems.

Figure 2.2 Key Characteristics of Sites Selected for Inclusion in Study

Figure 2.3 Disposed Cases Sample: Description of Court Case Data Used for Each Site

Finally, readers should note that, throughout this report, references to ‘private counsel’ encompass both privately retained counsel and private counsel paid through a legal aid referral or certificate. Since it is inappropriate to make this distinction known during the court process, the distinction is typically not captured in court manual or automated systems in judicare jurisdictions.  We were, therefore, not able to make this distinction in our collection or analysis of either the disposed cases data or the court observation data.

2.2.2 Court Observation

Local persons with experience working in or around the court were also hired and trained to observe and record data on individual case appearances during a sample of (usually 10) court days.  Again the observations were based on a core template that was modified slightly to reflect the unique characteristics of each site.  The court observation was focused on first appearance/set date/arraignment/bail (i.e., non-trial) courts. (The time required for each trial would reduce the sample captured in court observation, and the disposed case sample would accurately capture factors at trials.)

Data was recorded separately for each appearance of each case on the docket.  As with the disposed cases sample, data was collected on the characteristics of the case, the events that occurred at the appearance, and the outcomes of those events.  However, a special focus of the court observations was to capture the comments made with respect to the accused’s representation status – by the accused, the judge, duty counsel, or the Crown.

2.2.3 Key Informant Interviews

Study team members visited each site for about a week, talking to judges, Crown attorneys, legal aid line and management staff, private defence counsel, court administration personnel and court clerks, sheriffs and other court security staff, and local service agencies who may have contact with accused adults in the criminal court system.  The interviews, which lasted up to one hour, followed a common structured interview guide and provided extremely valuable information regarding perceptions related to a wide range of issues, including:  the numbers of unrepresented persons, the reasons for lack of representation, issues affecting representation – and impacts of lack of representation.  Some 10 to 30 persons were interviewed at each site – the majority jointly conducted by two interviewers. Interviewees were assured of anonymity, so the names of the individuals interviewed are not reported herein.

2.3 Lessons Learned Regarding Methodology

A number of important methodological lessons were learned in the conduct of this research:

  • For the extraction of data from local court automated information systems it was essential to rely on local computer programmers with extensive experience with those systems.
  • Use of local persons with extensive expertise in court processes – in the specific courts studied – was a major advantage in terms of the efficiency and reliability of the data collection (all of our coders and court observers had previous experience in their courts, with a number being current or former managers in those courts).
  • It was important to give serious consideration to the use of manual data collection methods – even where partial information was available from automated systems.  In a number of situations the automated systems were not designed to support the type of analysis we required, and we found it more cost-effective and timely to rely on manual data collection methods.
  • Consultation with local court administrators on the design of the project greatly enhanced the quality and reliability of the data.  Those with local expertise often suggested methods that were superior to those originally envisaged.
  • It was important to follow common core protocols across the nine sites, but essential that the project consider local resources and capabilities – and change the research design to accommodate local circumstances.
  • Timing of data collection tasks had to accommodate local protocols and circumstances.  Courts operate in a very time-sensitive and event-sensitive environment.  The research had to be extremely flexible with respect to method and timing in order to be minimally intrusive on day-to-day operations.

The researchers were fortunate that both local court officials and the federal and provincial clients for the research understood the importance of each of the above principles.


[7] Prince Edward Island was the only province without a court in the study.

Date modified: