A regression model fit to data from all 23 cities finds a 
statistically significant linear relationship between numbers 
of violent crimes in American cities and their populations.
For each increase of 1,000 in population, the number
of violent crimes increases by 0.1093 on average.
48.6% of the variation in number of violent crimes
can be explained by its relationship with population.
Because these data are observational, i.e. collected
without experimental intervention, it cannot be said
that larger populations cause larger numbers of crimes,
but only that such an association appears to exist.

However, this linear relationship is mostly determined
by New York City whose population and number of violent
crimes are much larger than any other city, and thus accounts
for a large fraction of the variation in the data.  When New York
is removed from the analysis there is no longer a statistically 
significant linear relationship and the linear relationship with 
population explains less than 9% of the variation in number of 
violent crimes.