Three general periods of development can be identified: the segregation period of the early 19th Century, which focused on inmate age, sex, mental function, and offense severity; the early 20th Century approach to classification for diagnosis and treatment planning; and the classification for treatment phase that began in the 1950's and emphasized differential treatment and program evaluation. Current trends suggest an increasing emphasis on classification for security and custodial purposes. Subjective or traditional classification systems typically rely on the expertise and clinical judgment of individuals or teams in making security, custodial, and programmatic decisions about inmates. By contrast, objective systems usually assign points to inmates based on characteristics shown to be correlated with institutional adjustment. Objective systems generally use validated test and classification instruments, use the same approach for all offenders, and arrive at application factors shown to be related to placement decisions. Such systems promote uniformity of decisionmaking, are easily understood by staff and inmates, assign inmates to placements consistent with their backgrounds, and promote systematic and efficient monitoring. 18 references. (Author abstract modified)