Files
Abstract
This study seeks to identify factors that are predictive of youth involvement in adverse incidents while placed in detention. The subjects included 13,557 youth involved in 21,179 detention placements in Georgia between July 1, 2001 and June 30, 2002. A Generalized Linear Mixed Model with fixed and random effects was used to identify variables that are predictive of youth involvement in three types of adverse incident including whether a youth was involved in any incident, in a use of force incident or a self harm incident. Independent variables included demographic, facility and public safety risk factors. The studys use of placement level data produced a rich picture of facility and demographic factors that would not likely have been identified with facility level data. The study found that placement length of stay was the strongest predictor of youth involvement in adverse incidents. Length of stay was significant in all three regressions and had the largest F statistic in two models. The difference between average and maximum population during placement (UTILDIFF) proved to be more predictive than either average or maximum population. That the likelihood of adverse incidents increases as UTILDIFF increases is important because it suggests how overcrowding is related to poor conditions of confinement. Younger youth, youth with more prior detentions and males were at higher risk of involvement in all three types of adverse incidents. Minority youth were more likely to be involved in use of force incidents than white youth. White males were more likely to be involved in self-harm incidents than minority males, both of who were far more likely to be involved in self-harm incidents than females. The study produced inconclusive evidence that youth in the moderate category of the Detention Assessment Instrument (DAI) were at higher risk than youth scoring low or high. DAI scores suggest that no overcrowding would have occurred during the study period if DAI placement recommendations had been implemented. The study considers that policy and constitutional implications of the overcrowding that results from the secure detention of youth who might be better served in less restrictive settings.