S. Bushway et al., Assessing stability and change in criminal offending: A comparison of random effects, semiparametric, and fixed effects modeling strategies, J QUANT CR, 15(1), 1999, pp. 23-61
An important theoretical problem for criminologists is an explanation for t
he robust positive correlation between prior and future criminal offending.
Nagin and Paternoster (1991) have suggested that the correlation could be
due to time-stable population differences in the underlying proneness to co
mmit crimes (population heterogeneity) and/or the criminogenic effect that
crime has on social bonds, conventional attachments, and the like (state de
pendence). Because of data and measurement limitations, the disentangling o
f population heterogeneity and state dependence requires that researchers c
ontrol for unmeasured persistent heterogeneity. Frequently, random effects
probit models have been employed, which, while user-friendly, make a strong
parametric assumption that the unobserved heterogeneity in the population
follows a normal distribution. Although semiparametric alternatives to the
random effects probit model have recently appeared in the literature to avo
id this problem, in this paper we return to reconsider the fully parametric
model. Via simulation evidence, we first show that the random effects prob
it model produces biased estimates as the departure of heterogeneity from n
ormality becomes more substantial. Using the 1958 Philadelphia cohort data,
we then compare the results from a random effects probit model with a semi
parametric probit model and a fixed effects legit model that makes no assum
ptions about the distribution of unobserved heterogeneity. We found that wi
th this data set all three models converged on the same substantive result-
even after controlling for unobserved persistent heterogeneity, with models
that treat the unobserved heterogeneity very differently, prior conduct ha
d a pronounced effect on subsequent offending. These results are inconsiste
nt with a model that attributes all of the positive correlation between pri
or and future offending to differences in criminal propensity. Since resear
chers will often be completely blind with respect to the tenability of the
normality assumption, we conclude that different estimation strategies shou
ld be brought to bear on the data.