I'm not a statistician. I don't even play one on TV. But I like to think I have some basic logic skills that I can apply when looking at statistical studies.
So I read a study earlier that took a ratio of A to B from a group and compared it to ratios of A to B from other groups. Then when the first ratio was shown to be out of line, the authors concluded that it was due to A. My concern was that it seemed to me, that the problem could equally be with B.
Simply stated, if A/B = a number that you don't expect, then either A or B could be the problem.
This diary from earlier today does just that. They are looking at Florida and Michigan vote totals to determine if the voter turnout was suppressed and how many Dem voters should have turned out.
They didn't include the data so I have found some myself. I went to each state site and used their data. Also, if you look at their pdf you will see that there are around 36 dots on their graph. Since Florida is a closed primary state I looked only at other closed primary states (I used the CNN site to see which states had closed primaries).
Democratic Turnout
Connecticut 51.09%
Maryland 50.71%
Arizona 50.47%
District of Columbia 44.48%
Florida 42.30%
Oklahoma 41.20%
Deleware 37.59%
New York 35.44%
New Mexico 28.42%
Louisiana 25.66%
----
Republican Turnout
Arizona 51.98%
Florida 50.96%
Oklahoma 42.37%
Connecticut 36.74%
Maryland 36.08%
Deleware 28.01%
Louisiana 22.96%
New York 22.35%
District of Columbia 22.08%
New Mexico June 3rd
Looking closer at those numbers:
Average Dem Turnout 40.74%
Median Dem Turnout 41.75%
Florida Dem Turnout 42.30%
Average Repub Turnout 34.84%
Median Repub Turnout 36.08%
Florida Repub Turnout 50.96%
So while the Florida Dem turnout was slightly better than average, the Florida Repub turnout was much better than average. In fact, the only state that had a higher Repub turnout was Arizona, McCain's state.
Most people have been looking at the ratio of the Dem turnout to the Repub turnout. So here that is.
New Mexico No Repub Yet
District of Columbia 2.01
New York 1.59
Maryland 1.41
Connecticut 1.39
Deleware 1.34
Louisiana 1.12
Oklahoma 0.97
Arizona 0.97
Florida 0.83
No doubt that Florida number is the lowest. But is it the lowest due to Dem voter turnout or Repub voter turnout? The easiest explanation, because it is what most people expect, is that it is due to suppressed Dem voter turnout. The authors of the study thought that. In fact they went further.
They compared the primary voter turnouts to 2004 Presidential votes in oder to calculate the missing voters from Florida and Michigan. And they came up with:
Democratic voters should have numbered 2.85 million and 1.305 million in Florida and Michigan, respectively. Given actual numbers of 1.7 million and 590,000, an estimated 1.15 million Floridians and 715,000 Michiganders stayed home on primary day, who otherwise would have voted had they behaved like voters in other states.
They believe 2.85 million Florida Democratic voters should have turned out to vote. That's a lot of voters. In fact that is a 68.89% voter turnout.
Let's look at that with the other closed primary states.
Average Dem Turnout 40.74%
Median Dem Turnout 41.75%
Actual Florida Dem Turnout 42.30%
Projected Florida Dem Turnout based on Repub Turnout 68.89%
That would put Florida way ahead of every other closed primary state. The next closest would be Connecticut with 51.09%. Does this seem reasonable? For a primary?
Now does that prove there was no voter suppression? Of course not. But it doesn't prove there was either. At least to my non-statistical mind.
|
|
|
Permalink :: 5 Comments :: Post a Comment
|
In order to post a comment, you must be logged in. If you have a member account, please log in to comment.
If not, you can make an account right here. It's quick and free.