And shows the Berkeley guys have been smoking too much crack
The conclusion that President Bush was more likely to improve his vote in counties with e-voting is laughable on its face. Using the Excel spreadsheet provided by the authors, I totaled the votes for counties with and without e-voting, and came up with this:
Percentage Change for Bush in Counties WITH E-Voting: 2.25%
Percentage Change for Bush in Counties WITHOUT E-Voting: 2.54%
It looks like e-voting suppressed the President’s vote by about 0.29%_ — or 7,800 votes!
Taking each of these counties as data points, was the President “significantly more likely” to have increased his support in counties with e-voting? Again, no.
E-Voting Counties with Increased Bush Vote: 13/15 (86.7%)
Non-E-Voting Counties with Increased Bush Vote: 46/52 (88.5%)
Bad science loses to E-Voting again. Thanks to Pat for doing what I was too lazy to do.
Recent Comments