In criminal justice and criminology research (as in most other social sciences), regression models usually produce fairly modestR²values. Even explaining a relatively small fraction of the variance in a dependent variable is frequently considered good, even though anR² value considered robust within social sciences (such as .40, for instance) leaves the majority of the variance unaccounted for (in this case, 60% of the variance is unaccounted for). Why do you think this is? That is, why do regression models commonly explain less than half of the variance? (Hint: Think about the dependent variables that criminology and criminal justice researchers examine, and our ability to explain their variation.)