In conducting the survey, the staff made over 350 telephone contacts with classification experts, research organizaitons, and justice system agencies. These contacts and the literature review indicated a recent trend toward formalizing offender classification, establishing criteria for screening decisions, and increasing reliance on standardized instruments for decisionmaking input. This sourcebook includes three main sections: a state-of-the-art summary, site reports, and telephone interview summaries. The first section describes the current classification instruments and practices that are employed at the level-of-supervision decision stage. The site visit reports provide an indepth look at currently used instruments and how they operate in four specific agencies. Telephone interview summaries contain succinct descriptions of 19 agencies and their use of classification tools. The survey revealed that the use of classification instruments at this stage of the criminal justice process is relatively uncommon, but interest in this area is growing. Agencies now using instruments report generally satisfactory experiences. Instrument use tends to divert more cases to lower levels of supervision, a trend that obviously could reduce costs. However, a number of potential problems are associated with instrument use, particularly in the area of legal complications. These problems stem from lack of instrument validation, low predictive accuracy of many instruments, and failure to assess the impact of instrument use on caseload distributions or departmental operations. Most of these problems could be resolved, and the judicious use of instruments does have demonstrable advantages. Footnotes, scoring forms, client analysis scales, case management/screening forms, tables, and organizational charts are provided. For related documents, see NCJ 76058 and 76060-62.