A descriptive analysis of 519 felony cases also was conducted. While State law outlines the factors judges must consider in determining bond types and amounts (e.g., the offense, past record, ability to pay), there is actually much variability in the information judges receive. Most bail decisions are reached quickly: the average hearing lasts only a minute or two. Although judges are often given information about defendants' community ties and other circumstances, such information is often incomplete, unavailable, or unverified. Of the felony defendants sampled for descriptive analysis, 87 percent were not rearrested before trial. This can be attributed to the fact that over 50 percent were not released before trial. However, 4 percent of the sample was arrested multiple times after release from custody. There was little relationship between the offense at initial arrest and rearrest. Serious, repeat offenders were more likely to be arrested for trial than less serious offenders, suggesting that pretrial detention of this group (72 percent) is in the public interest. Supplemental research data are appended. 9 tables, 15 figures, and 41 references.