I amended my lawsuit and filed it again on April 1 in the Sedgwick County Courthouse. A reporter talked to me about it and gave me a small write-up for the AP. The Winfield Courier has the longest and most informative. It’s pretty cool and generating a surprising amount of interest.
I took another look at the Ohio 2012 Presidential race data. It proves to be another illustration of the pattern I find concerning. In this case, the pattern holds across all the voting units, not just the ones with more than 500 votes cast.
It also displays an interesting pattern of high variability for both the high count precincts as well as the low count precincts, which is somewhat odd. A linear regression of precinct size on % Rep. votes shows a straightforward positive trend with a p-value of less than one in a million. The slope is small, and the R2 value is low, which means that this trend does not explain much of the data variation. But the trend is consistently positive and statistically significant across many different elections