So I read this story on Reddit and I wondered how true is it?
Quick Summary
What I found from doing my analysis was that on an average both Red and Blue states have similar rates of casualties. However in states with most casualties/100,000 Red states dominate (75%). In states with least number of casualties Blue states dominate (69%). The contention that there are more casualties in Democratic states is Bullshit!
Analysis
The contention seems to be that while Red states (Republican) voted for war the people that seem to be killed are from mostly for Blue states (Democrat). So I ran some numbers to check the validity of the statement.
Median = 1.3
Mode = 13.
What the above statistics mean is that the average casualties per state per 100,000 was 1.3. (Mean)
The most common rate for casualties/100,000 was 1.3. (Mode)
The middle value for casualties was 1.3 (Median)
What this seems to suggest is that most of the states seem to have similar rates of casualties. There were 9 states with 1.3 casualties/100,000 and 5 of them were Blue states. A pretty even split. Looking at the data from this angle seem to say that voting patters here seem to have no effect at all.
Lets now take a look at how the data shapes up by each quartile. I sorted the data in Ascending order (least casualties to most casualties per 100,000).
1st Quartile -
Number of states = 13
Number of Red States = 4
Number of Blue States = 9
Casualties Range = 0.1 to 1.0
Damn! This quartile breaks down almost exactly opposite to the Reddit story. As it would seem the Blue states dominate in the states with the least amount of casualties per 100,000. 69% of the states with the least casualties are Democrat.
2nd and 3rd Quartiles -
Number of states = 25
Number of Red States = 15
Number of Blue States = 10
Casualties Range = 1.0 to 1.5
I grouped these two quartiles together because their data had so little differentiation. In the middle 50% we see very little differentiation. Raw data would seem to indicate here toot here are more Reds than Blues. But we must remember that there are inherently more reds than Blues. 39% of all US states voted Democrat in 2004, 40% of the states in Q2&Q3 are Democrat. Thus we see that voting patterns seem to have no effect on the medial data.
4th Quartiles -
Number of states = 12
Number of Red States = 9
Number of Blue States = 3
Casualties Range = 1.5 to 2.9
Whoops! In the data in the 4th Quartile once again seems to directly refute the contention. In the states with the most casualties per 100,000 Red states seem to dominate. 75% of the states with the most number of casualties per 100,000 are Republican. Interestingly the state with the highest casualty rate was Vermont a Democratic state.
Monday, May 14, 2007
Subscribe to:
Comments (Atom)