BIRTH COHORT SIZE, CRIME, AND ALCOHOLISM
Steffensmeier, Streifel and Shihadeh (1992) have noted that:
Involvement in crime generally diminishes after adolescence, implying that fluctuations in the age composition of a population may have a significant impact on overall crime rate. (p. 306)
The authors provide an example in the dramatic increases in the rates of serious crime in American observed during the 1960s being linked to the rise in babies associated with the post-World-War II "baby boom." Similarly, they note that the fact that these rates leveled off in the 1980s was associated with the post-1960s "baby bust."
Given that there appears to be a systematic association between crime and birth cohort size, Easterlin (1978, 1987) contended that the general volume of crime may also be related to birth cohort size. However, tests of Easterlin's views (e.g., Easterlin & Shapiro, 1979; Maxim, 1985; O'Brien, 1989; Steffensmeier, Streifel & Harer, 1987) have yielded mixed results.
According to Steffensmeier, Streifel and Shihadeh (1992), the mixed findings are primarily due to methodologically problematic research designs. Design flaws are said to include:
1. Failure to control for the linear dependence between age, period, and cohort, so that cohort effects could not be isolated.
2. Use of total crime rates rather than specific crime rates as dependent variables--a procedure that confounds age, composition, period and cohort size.
3. Reliance on inadequate age groups (limiting observations to only certain ages or combing age groups thereby losing small single-year birth cohorts).
4. Failure to distinguish between crimes of assault and property crimes; and
5. Failure to partition data so as to determine whether the size effect varies over the life course of the cohort.
One implication of the methodological problems with existing research is that Easterlin's (1978, 1987) theory of birth cohort...