A prisoner subgroup (n = 132) defined by extent of SIB was matched by conviction prefix and security level to a group of prisoners who had never engaged in SIB (n = 132) and compared on selected variables. Relative to prisoners who either engaged in less extensive SIB or none at all, prisoners who engaged in three or more forms of SIB while in prison were found to be twice as likely to be housed in long-term administrative segregation, less educated, began their criminal careers at an earlier age, engaged in more violence in the community and in prison, involved in property destruction, and experienced a much greater degree of housing instability in prison. (Published Abstract)