The sociological/criminological literature for many years has emphasized the existence of an association between prevalence and/or incidence of criminal behavior and certain sociological variables. (e.g., Hirschi 1969). It is undeniable that there are positive correlations between criminal convictions and race (Hawkins 1985; Matsueda and Heine 1987), gender (Hindelang 1979; Jensen and Eve 1976), age (Farrington 1986; Hirschi and Gottfredson 1983), place of residence (Messner and Tardiff 1986; Reiss and Tonry 1986) family structure (McCord 1982; Kolvin, Miller, Fleeting, and Kolvin 1988), socio-economic status (Miethe, Stafford, and Long 1987; Cohen, Klugel, and Land 1981) and ethnic group (Hagedorn 1988; Rieder 1985). However, the correlations at best are in the moderate range, and the underlying causes of these correlations remain uncertain. A number of these researchers have gone beyond the simple correlations to show that the relations of social context with criminality are often a product of third variables or are indicative of complex interactions with other behaviors also related to social context (e.g., Farrington 1986; McCord 1982; Reiss and Tonry 1986).