It’s very refreshing to see someone in the news not jumping to strict comparisons and including a warranted disclaimer when nationwide statistics on a particular topic are released (e.g. the FBI crime stats). From a statistical point of view, what categories or mathematical combinations of data would, in your opinion, lead to a comparison between cities that minimizes the influence of exogenous factors not considered in the raw statistics themselves? As an example, violent crime per capita is one such measure that tries to eliminate a population based bias from the comparison, but there is host more that in some cases cannot be quantified (e.g. crime reporting practices of the citizenry). Assuming some factors span a geographic area or state more than just a city and can be considered ‘constant’, what are some important factors to consider when we DO directly compare cities like Norman, Tulsa, and Oklahoma City that are most correlated and account for the majority of differences in crime in these cities?

Thanks,

Don

http://docapp8.doc.state.ok.us/pls/sors

It’s also linked from our home Your Right to Know page here:

http://www.newsok.com/righttoknow

You’ll find it if you keep scrolling down in the “Data” section on the left-hand side.

Paul

Thanks