Thursday, September 4, 2014


It is quite an education to look closely into FOIA data from the Pentagon.

I have been busy (sorry I missed posting yesterday) writing computer programs to analyze the FOIA data about the 1033 program as a explained in the last post I did.

When data entry software is not tightly written the data entry folks will input data differently and it makes the data difficult to use.

So. obviously my first exercise was to conform the data so that it can be useful to analytical programs I am developing (I really do like letting the computer do the work on large databases, because it would take years to do it by hand).

For example there were a few "unknown" or "blank" entries for "county" but they did manage to enter data for the state column or field.

The descriptions of the military equipment varied on almost everything, nevertheless one could manage to, for example, discern whether it was a rifle or a pistol.

Anyway, the bottom line is now that I have cleaned up the data in the ~250,000 records returned by the Pentagon in response to the FOIA request, now that I have built spreadsheets and SQL databases for all of it, and now that I have verified the data in them, I can begin to look closer.

I still plan to integrate some Census Bureau information into the basic databases.

At first blush it was obvious that some counties are more violent, brutal, and military oriented in their policing than other counties are.

I plan to secure some demographic data about the counties to compare the militarism of the police with percentage of minorities and other factors.

In general, I see the social science associated with this data as supporting the general theme expressed in the series Symbolic Racism: A Look At The Science - 3.

On that note, have you noticed that the DOJ has opened a second Ferguson investigation as to the general habitual racism at work in police departments there.

Perhaps they should read some of these posts that will be coming out as I now work to analyze the data after having cleaned it up.

Keep in touch.

"I'll be back."

I have downloaded data from the Census Bureau for integration into the "will the military become the police" databases.

The difference between the Census Bureau data and the Pentagon data is like the difference between a first grader and a Phd.

The Census Bureau data will need no cleaning whatsoever, and is a professional example of how to do databases and spreadsheets.

Some of the example data they have compiled for all states and counties:
Accommodation and Food Services
Building Permits
Business, Private Non-farm
Civilian Labor Force

Federal Government
Hispanic or Latino Population
Income and Poverty
Internal Revenue Service
Labor Force
Land Area
Local Area Government

Non-employer Statistics
Personal Income
Race Data
Retail Trade
Survey of Business Owners
Sex Data
Social Programs
Vital Statistics
Water Usage
Wholesale Trade
I will be able to present relevant information on some of those topics for comparison.

I expect that the police and citizenry, to a relevant degree, in counties which hog military hardware, will have demographics (democrapics?) that are of a certain cultural amygdala.

We shall see.

No comments:

Post a Comment