Pre-Trial Detention Under the Young Offenders Act: A Study of Urban Courts

4. Judicial Interim Release: What Factors Affect Court Decision-Making? (cont'd)

4. Judicial Interim Release: What Factors Affect Court Decision-Making? (cont'd)

4.5 Factors Affecting the Number of Release Conditions and Type of Release: Race in Toronto

The research by Kellough and Wortley (2002) encouraged us to compare youth and adult bail processing in the Toronto courts. Their study examined the effects of demographic and legal variables on releases at bail hearings, the number of bail conditions and the type of release (i.e., whether the accused was released on a recognizance as opposed to a less serious mode of release). The main interest of Kellough and Wortley was the influence of race. This research reproduced both the dependent and independent variables used by Kellough and Wortley in order to find out if the same or similar findings applied to the youth court.

We first analyzed the factors affecting the number of bail conditions imposed by the two Toronto youth courts. Black youth [49] had an average of 4.2 bail conditions (median of 4), and those of other races had a mean of 3.5 (median of 3) - a difference in means that was statistically significant. [50] At the multivariate level, race was close to significance (p=.06) when legal factors were included in the regression model. Being female, living independently or with no fixed address, having more prior convictions and being charged with a violent offence were significantly associated with the number of bail conditions imposed on Toronto-area young persons.

The second analysis involved the issue of recognizance releases, which require a surety, usually a friend or a relative, and a monetary commitment. This form of release is in contrast to undertakings, which while often onerous in terms of conditions, lack the surety requirement. In the Toronto youth courts, race was related to being released on a recognizance: 71 percent of whites and other races but 95 percent of black youth were released on a recognizance (p<.01). When all variables were controlled, race and living arrangements (living independently or no fixed address) were the only factors significantly associated with the use of a recognizance. When Kellough and Wortley applied logistic regression to their adult dataset, black accused were three times as likely to be released on a recognizance as other races; sex, employment status, having a permanent address, number of current charges, serious violent charges and being charged with failure to appear also affected recognizance releases.

Thus, in one of the two instances examined in this section, the impact of race in the youth court was similar to its impact in adult courts in the same city: black youth had to locate a surety to be released more often than did others regardless of the characteristics of their current charges or prior record.


Non-legal (social) factors
  Number of bail conditions Recognizance vs. other forms of release
Being female *  
Age - being 15 or more years    
Being black   *

Socio-legal factors
  Number of bail conditions Recognizance vs. other forms of release
Living independently or no fixed address * *

Legal factors
  Number of bail conditions Recognizance vs. other forms of release
Number of prior convictions    
Number of current charges, including outstanding charges    
"Violent" charge at arrest *  
FTC/FTA charge at arrest    
Number of prior "violent" convictions    
Adjusted r2 and Nagelkerke r2 respectively 18 29


* p<.05. If the cell is blank, then the factor was not statistically associated with the dependent variable.

4.6 Summary

At the bivariate level, an array of factors were associated with youth court bail decisions. Compared to police detention, fewer socio-demographic and legal variables were related to pre-trial detention by the court. In particular, the presence of outstanding charges, the number of charges at arrest and the most serious current charge were associated with court detention. In contrast, most of the indicators of prior record were related. What we termed the bail-related charges such as reverse onus and number of mentions of primary and secondary grounds tended to be associated with the court decision.

The multivariate analysis, which controlled for all factors simultaneously, found the following significant relationships:

  • The longer and more serious the prior record, the more likely Halifax youth were detained by the court. A current charge of fail to appear or fail to attend was also statistically significant. Young persons not living in a family setting were more likely to be detained.
  • In the two Toronto courts combined, the longer and more serious the prior record and unconventional living situations increased the likelihood of court detention.
  • In Winnipeg, a history of "bad" behaviour on bail or a prior custody sentence, and some legal involvement at arrest, were the only factors associated with detention.
  • In Edmonton, the longer and more serious the prior record and a history of "bad" bail behaviour/prior custody sentence raised the probability of detention by the youth court.
  • In Vancouver and Surrey youth courts, only the length and seriousness of the prior record of the accused affected detention use.
  • In the sample overall, more factors were related to detention: mention of primary or secondary grounds or both, having an indictable current charge, having a long and serious prior record, a current charge of fail to attend court or failure to comply with bail conditions, the number of outstanding charges, age (older) and unconventional living situations.

Thus, the prior offence history of the accused was more influential in bail decision-making by the youth court than were the characteristics of the current charges when the courts are analyzed separately. However, indicators of prior and current offences were significant in the total sample.

The regression models for this decision point were less successful than were the models developed for police detention. Factors other than personal and case characteristics probably influence court bail decisions.

Multivariate analysis was employed to try to determine the personal and case characteristics associated with specific release conditions. Being black or Aboriginal increased the likelihood of having a "do not communicate with the victim" and "do not carry or possess weapons" conditions, even when other factors were controlled. No factors explained the imposition of curfews at bail, leading us to conclude that curfews are imposed for reasons other than the legal characteristics of the case. Since curfews so often precipitate bail charges, their use should be reconsidered.

An examination of the role of race in Toronto bail decisions found that race predicted the use of recognizances versus other means of court release even when other factors were controlled. Because of their requirement for a surety, recognizances require that the young person have a resource in the community, a more burdensome requirement than undertakings. Race was marginally related to the number of bail conditions in the Toronto youth courts.

  • [49] There were 36 black young persons in the Toronto-area sample.
  • [50] Anova F value = 10.42, p=.001
Date modified: