Reply #23: Thanks. I appreciate your work as well as eomer's. You are not afraid to [View All]

[TruthIsAll]

step in and crunch numbers. We need more of that and less of the petty, personal attacks which serve no useful purpose.
I have never claimed to know all the answers. Cooperation, not blind naysaying is the key.

You:
..vary tremendously within a single county. Now what about the outliers in terms of ballots cast ... should they be given any special treatment?

Me:
The outliers are very significant.
How do you get more ballots than registrations.

You:
We know also that the entire range (all three groups combined) is neither uniform nor normal. Now consider each quartile group separarately -- do they have a uniform distribution or is there still some clustering that may effect the outcome of our tests?

Me:
I doubt they are normal.
Do a histogram.

You:
Moving on to the big picture, given that the distribution varies from county to county (in terms of strongholds) and hence their respective quartiles, should we estimate each county separately or do you think it is reasonable to throw them all into one big melting pot?

Me:
I say do both.
1) Do the differences in the counties offer clues? They might.
2) Does an exhaustive data set of all th counties show us the big picture and effectively eliminate individual the outlier concern? I believe it does.



PS: I appreciate your work very much TIA ... even if we disagree at times