Reply #48: I'll play devil's advocate. [View All]

[Igel]

And claim that the correlation is derivative. On the one hand, there's an obvious correlation between voting machine type and spoilage rate. The 2000 Florida election noted that counties with greater per capita budgets had newer equipment that prevented overvotes and even prompted the voter when an undervote was detected; these trended repub. SES vs. machine type was the real correlation there.

It may be the case that less accurate technology's all recently been phased out in New Mexico; but that leads to another problem.

I also know from experience that voters new to a technology take longer and make more mistakes. I nearly did it the first time I voted in Oregon, in LA, and in Houston. Staffing precinct tables in NY I saw a lot of very young and very new voters screw up. New voters make more mistakes. Notice that a fair number of errors reported were of the sort "voter accidentally touched touchscreen and changed vote" or "circled optiscan bubble"; experienced voters would make the mistakes less often.

It would be also be interesting (as a devil's advocate not interested in crunching these particular numbers) to see two other correlations: the spoilage rate vs. acquisition date of the voting technology; and spoilage rate vs. percent of new voters, for precincts retaining older technology.