A major finding was that the probability of a felony conviction given a felony arrest (net felony conviction rate) varied substantially across the 10 counties. Two factors, by definition, determined net felony conviction rate, the rate at which counties selected felony arrests for upper court processing (screening rate) and the rate at which prosecutors obtained felony convictions for cases processed in upper court. Only the screening rate, however, was strongly associated with the net felony conviction rate. The rate at which counties obtained felony convictions in upper court was more uniform across counties and was minimally associated with variation in the net felony conviction rate. Four variables strongly related to the screening rate were caseload, prior criminal history, crime type, and average crime seriousness. Counties with lower average caseloads and higher proportions of defendants with one or more felony convictions tended to have relatively higher screening rates. In addition, counties that handled more serious crimes overall tended to have relatively higher screening rates. Further research is recommended to investigate fully the potential effect of study variables on the net felony conviction rate. 4 tables, 18 figures, 1 reference.