|
Figure 7 shows the predicted profile based on
the augmented equation
for low- and high- density counties—represented
by population densities of 100 people per square
mile and 1500 people per square mile—for
aggravated assault and rape. Crime is not
statistically
different from zero in the years before casino
introduction and immediately thereafter, but
begins
29
Figure 7: Population Density Interaction
to rise three or four years after introduction.
By the fifth year after casino introduction, the
figure
shows a statistically significantly elevated
crime rate for both low- and high-density
counties. The
other crimes (omitting murder) display a similar
pattern. Overall 10 of the 12 estimated effects
are statistically significant after five years,
and all 12 are positive. Introducing a density
effect
does not change the prediction of the model.
These results give us confidence that the effect
of
casinos on crime is similar in large and small
counties.
Figure 8 displays the estimated effect of
casinos on large and small counties for auto
theft.
Neither figure displays a profile similar to
that of the other crimes, where crime begins to
rise
three years after casino introduction. Large
counties appear to have stable or declining auto
theft rates, consistent with the Lojack
explanation, while smaller counties that would
employ
Lojack less extensively or not at all appear to
have a rising auto theft rate independent of
casino
introduction.
Back to casino
news
|
