Using Representation Pathways to Explore Court Data

Court data is an important source of information that can increase our understanding of justice system issues. Research is currently under way at Osgoode Hall Law School and the Canadian Forum on Civil Justice using Ontario court data examining two issues: the impact of unrepresented accused on the efficiency of the courts and the disadvantages that may be experienced by unrepresented accused in terms of outcomes. The research is being carried out using an appearance-based data set from the Ontario Cout of Justice. The data set includes 17,622,670 appearances nested within 2,002,306 disposed cases from 2011 to 2022.

Court data are based on appearances, recording appearances and attributes of appearances as separate events. However, the experience or the journey of accused persons in the courts do not occur one disconnected appearance at time. An important part of that reality is whether or not the accused is represented by counsel. This brief article reports on research in progress using a representation pathways approach to analyzing court data that is examining the impact of appearances at which there is no representation on court efficiency and the disadvantage experienced by unrepresented accused in terms of case outcomes.

A Closer Look at the Data

Some people facing criminal charges are not represented at any appearances, some are represented at only some appearances, some are represented at all appearances. At appearances where there is representation, people may be represented by retained counsel or by duty counsel. The patterns are more varied than all, some, or none. In order to examine the data in more detail the original court data was recoded to create representation pathways rather than treating representation (not represented, represented by duty counsel, and represented by retained counsel) as attributes of individual appearances. The representation pathways are:

1. Not represented at any appearances
2. Not represented at some appearances + represented by duty counsel at some appearances
3. Not represented at some appearances + represented by retained counsel at some appearances
4. Not represented at some appearances + represented by retained counsel at some appearances + represented by duty counsel at some appearances
5. Represented at all appearances by duty counsel
6. Represented at all appearances by retained counsel
7. Represented at all appearances by mixed duty counsel and retained counsel

The research is examining court efficiency using two variables: the number of appearances and the number of days (minus bench warrant days) to disposition. The data are reported using medians and means because the data on appearances and days are highly skewed.[1]

Table I shows the mean and median number of appearances to disposition for representation pathways.[2]

In terms of appearances to disposition the lowest median and mean numbers are for disposed cases in which the accused is not represented at any appearances. This is followed by the duty counsel pathway, in which all appearances have representation by duty counsel. These two pathways probably represent relatively simple cases or ones in which charges are dismissed early in the process. It is very interesting that the third lowest representation pathway is disposed cases in which representation at all appearances is by retained counsel, pathway 6.

The representation pathways that have the highest median and mean numbers of appearances to disposition are the ones in which representation at appearances is mixed. The pathway involving the highest mean and median number of appearances to disposition is number 4, not represented at some appearances, represented at some appearances by duty counsel and represented at some by retained counsel. The mean number of appearances is 16.9 and the median is 14. This is followed by the pathway defined by a mix of duty counsel and retained counsel, pathway number 7, with a mean number of appearances of 11.5 and a median of 9 appearances to disposition. The pathway defined by not represented at some appearances and represented by retained counsel combination, number 3, with a mean number of 10.4 appearances to disposition and a median of 8 is the next highest ranked at number 5.

An interesting comparison is pathway 6, represented by retained counsel at all appearances and pathway 3, not represented at some appearances and represented at some appearances by retained counsel. The mean and median numbers are much higher for pathway number 3, at 10.4 for the mean and 8 for the median number of appearances compared with pathway 6, representation by retained counsel at all appearances in which the mean number of appearances is 6.8 and the median is 5. When there is duty counsel representation at some appearances, the median and mean numbers of appearances and days to disposition increase.

Overall, the data suggest mixed representation pathways are not efficient for the courts. An approach to representation, or a legal aid system in which duty counsel is concentrated at the early stages and there is a quick transition to retained counsel, privately retained or paid by a legal aid certificate, might be the most advantageous approach to representation in terms of court efficiency and efficiency for the legal aid system.

Analysis of duration in days to disposition is consistent with the analysis of appearances. Table 2 shows the mean and median duration in number of days to disposition for the 7 representation categories, representing different representation pathways.

Representation at all appearances by duty counsel has the lowest mean and median number of days to disposition, 95 and 59, respectively, followed by the not represented at any appearances pathway, 100.4 and 65. The ranks are reversed compared with appearances to disposition, with not represented at any appearances ranking second. Similar to the analysis based on appearances, it appears that mixed pathways are the least efficient in terms of days to disposition.

The representation pathway with the highest number of days to disposition is pathway number 4, not represented at some appearances, represented at some appearances by duty counsel and represented at some by retained counsel, mean and median number of days to disposition 289.8 and 239, respectively. Pathway number 3, not represented at some appearances and represented at some by retained counsel ranks just below in sixth position with the median number of days to disposition of 241.3 days and a mean of 194 days.

An interesting comparison is between representation pathway number 6, representation at all appearances by retained counsel, and pathway number 3, not represented at some appearances + represented at some appearances by retained counsel. The mean and median values for the representation pathway defined by representation at all appearances by retained counsel are 171.3 days and 117 days to disposition. The mean and median days to disposition are higher for representation pathway number 3, not represented at some appearances and represented at some appearances by retained counsel; the mean is 241.3 days and the median is 194 days. Similar to the analysis of appearances to disposition, the presence of duty counsel in the mix of these pathways produces higher mean and median days to disposition.

Mixed pathways tend to require longer durations to disposition. This supports the tentative conclusion that early representation by duty counsel and then a quick transition to full representation by counsel of record might produce the greatest efficiency in terms of days to disposition.

The representation pathways data can be summarized by averaging the mean and median numbers for the 3 pathways in which there is representation at all appearances with the 4 pathways in which there is no representation at all or some appearances. Table 3 shows that the mean and median values are higher for representation pathways in which there is an absence of representation in some or all appearances.

A second research question is the extent to which accused persons experience disadvantage by not being represented by legal counsel in terms of outcome. For this part of the analysis odds ratios[4] are used to express how many times greater or less the odds are of an outcome, for example an acquittal or a withdrawal of charges, if represented or not represented. The representation pathways are used for this part of the analysis as well. . Table 4 show odds ratios for acquittals comparing not represented at any appearances vs three different representation pathways.[5]

Even odds equals 1 or 1:1. The odds ratios, the odds of an acquittal if not represented compared with the odds of an acquittal if represented under different representation pathway scenarios, show the number of times less the odds of an acquittal are when not represented at any appearances compared with being represented. In all but two instances the odds do not favour an acquittal if not represented at any appearances on the path to a disposition. The data show that the odds of an acquittal are slightly greater than even if not represented compared with represented some of the time. The odds of an acquittal are more than 2.5 times greater than even if not represented compared with when represented some of the time. At this point the analysis has not progressed sufficiently to explore anomalies in greater depth.

Conclusion

Representation pathways is a promising approach for analyzing appearance-based court data. The quantitative data in a court database cannot explore all the issues raised by the analysis. Arguments for the need for legal representation are often expressed in terms of fairness or in terms of due process that assures the protection of rights. The court data cannot address the absence of these aspects of justice. The anomalies might occur because of other factors at play in the justice system. For example, police charging practices might result in cases coming into the system that result in outcomes in favour of the accused such as dismissals in favour of the accused. The odds ratios reveal disadvantage in statistical terms. Analysis of court data using statistical methods may provide valuable indicators raising questions about patterns or anomalies that should be examined by other research using other methods.

_

Ab Currie, PhD
Senior Research Fellow
Canadian Forum on Civil Justice

Trevor CW Farrow, PhD
Dean & Professor
Osgoode Hall Law School

Lisa Moore
Director
Canadian Forum on Civil Justice

__________________________

[1] That means that there is a very large number of disposed cases for which the number of appearances and days to disposition is very high. That is because it takes a long time to resolve many cases. In statistical terms the data do not represent a normal distribution or bell curve. There is a sharp rise at the beginning of the curve representing the data and a very long right-hand tail.

[2] In the main report currently being prepared, the data are analyzed for outcome and most serious offence at disposition.

[3] The mean is the average of all observations. The median is the midpoint of the distribution. The mean is more highly effected by skewed data than the median. Both measures are reported.

[4] Perhaps the most common sense meaning of odds expressed in ordinary language is the degree to which one outcome will happen as opposed to another. Terms such as odds and probability are commonly used, often interchangeably, in everyday discourse. The two terms have different meanings statistically and are calculated differently. Terms such as chance, likelihood or risk that technically are associated with probability are often used in relation to odds.

[5] The analysis related to the disadvantage of not being represented is not as far advanced as the work on representation and court efficiency. There are 20 possible combinations or pairs of representation pathways and 6 outcomes. The number of permutations and combinations is large.