variable is personal mitigating characteristics. It has a high coefficient, yet it is in the wrong direction. Again, this is likely caused by having relatively few cases in which this variable is cited by prosecutors in their decision to strike a strike. Results for Model 4 can be found in Figure 7.

^aFigures are unstandardized logit coefficients. The decision to use discretion is coded 1 if DA discretion is used to strike a strike, 0 if the DA chose not to strike a strike.
^bNumbers in bold are significant as follows: * Significant at p < .001. Significant at p < .005. �Significant at p < .01. Significant at p < .05. Observed probability levels based on a one-tailed test.
^cThe reduction in error (ROE) statistic is defined by the formula: (% Correctly Predicted - % Modal)/(l00 - % Modal)

Overall, the four models allow the discretionary actions of prosecutors to be compared against criminal history data and factors derived from the subjective evaluation data. The results of each support the hypothesis that prosecutors use their discretionary authority to strike prior strikes in accordance with crime control goals. The factors that increased the likelihood of discretion being used were all related to an assessment of potential crime control risk to the community. Offenders who were not seen to be typical recidivists, characteristics that were indicative of a reduced criminal character, and a reduced degree of offense seriousness (expressed mainly through the number of prison priors) all contributed to the decision to exercise discretion. Although the coefficients and/or odds ratio figures are artificially high for the variables derived from the subjective evaluation data (higher correlation is assumed because they are by design tied to the decision to strike a strike), their importance for testing this hypothesis is derived from the relative importance placed on each by prosecutors as they are compared to each other.

Conclusion

The California three-strikes law has received a great deal of media attention because of its relative harshness to similar measures. Its broad third strike encourages prosecutors and judges to exercise their limited discretionary authority in ways that reduces the culpability of offenders who commit minor third-strike felony offenses. Contrary to popular opinion, however, this analysis suggests that discretion is not being used haphazardly or irresponsibly. Rather, this study offers reassurance that prosecutors are acting in accordance to the standard �in the furtherance of justice� set forth by the legislature.

In particular, this study found that in a statewide survey of District Attorneys there was a large degree of consensus over the types of cases that would qualify for leniency. Almost three-fourths of prosecutors said that they would strike a prior strike if the current offense was trivial. Another two-thirds said that they would justify striking a strike if the strike offenses were remote in time or if the defendant had remained crime free in the interim. Over 90% of respondents also agreed that a strike would not be stricken if the defendant had been charged with a serious current offense and four-fifths agreed that the defendant would not qualify for lenient treatment if he was likely to re-offend or had a history of violence.

The analysis of prosecutorial discretion within San Diego County found that prosecutors were over three times more likely to strike a prior strike if the strike offense was remote in time and almost twelve times more likely to strike a prior strike if the defendant had no history of weapons possession. Similar to the survey findings, prosecutors in San Diego County were seven and a half times more likely to exercise discretion in those cases in which the current offense was trivial. They were also over 40% less likely to strike a prior strike for every additional prison prior accumulated by the defendant.

Both hypotheses tested with survey data and three-strikes case information from San Diego County indicate that prosecutors are not using their discretionary authority to strike a strike in a disparate manner nor are they striking prior strikes based on extra-legal factors. Although race data was not made available for this analysis, the strong performance of crime control variables in this study supports the conclusion that prosecutors are in fact using their allowable discretion in a manner that reinforces their theoretical crime control perspective.

or go to...
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11
12 | 13 | 14 | 15 | 16 | 17 | 18 | 19