Re: statistical study after next Tues election

From: Douglas W. Jones <jones_at_cs_dot_uiowa_dot_edu>
Date: Sun Oct 31 2004 - 19:36:30 CST

On Oct 31, 2004, at 12:23 PM, Kathy Dopp wrote:

> Douglas W. Jones said:
>> Such a study should include, on a county by county basis, the
>> percent undervote on the presidential race and the percent overvote
>> (where overvotes are allowed).
> I'm not sure how to fairly consider over-votes in counties where they
> can
> occur with counties where e-voting machines prevent over-votes.

DRE machines and lever voting machines do not permit overvotes to be
recorded, so you get overvote counts of zero. The statistics I've seen
show that DRE machines seem to be recording undervote rates of between
1 and 2 percent, while precinct-count optical mark-sense readers seem
to be able to get rates down under 1 percent.

This implies that around 1 percent of the voters who encounter DRE
seem to have trouble expressing a preference that they would be able to
express on mark-sense ballots, a source of major concern.

If we could find differences in undervote rates between counties that
correlated to the voting systems being used, after correction for
demographics (for example, pairing counties by median income or some
other census data), we'd begin to learn what it is that is causing this.
My guess is that the central problem is one of bad GUI design, but this
is only intuition.

> The postal absentee plus paper-based early voting could be the "control
> group" with which to compare the election results by county to look for
> patterns.

Yes, but you don't know that the early-voting rates correlate with the
election day rates. I expect that undecided voters don't generally get
sucked into early voting, while strongly committed voters don't mind
voting early.

> Are these three breakdowns of "in precinct, postal absentee, and early
> voting" available on

... on what? Anyway, depending on the state rules and county
the early vote may not be separated from the absentee vote. In Iowa,
example, early voting is just on-the-spot absentee voting administered
county employees in what they call "satellite polling places". You sign
the request for absentee ballot, they look you up in the voter rolls,
you vote, all in a space of minutes, and then you stuff your ballot in
an absentee ballot privacy folder, stuff that in an envelope, and drop
whole thing, paperclipped to your absentee ballot request form, into the
absentee ballot box. So, no distinction.

In some areas, early votes cast at satellite polling places are entered
on DRE machines, while postal votes are on paper. These totals must be
done separately, but whether they lump them for public release or keep
them separate will depend on state rules.

> I have chosen five swing states to study: Florida, New Jersey, Ohio,
> Arkansas, and Iowa. Arkansas and Iowa are control states, having no
> touchscreen machines.

But we have a double handful of counties that use DRE voting machines
here, mostly Microvote MV464 push-button machines. Not touch screen,
but definitely not VVPT. On the other hand, we've got 99 counties, so
these make up under 10% of the counties, and they aren't the big urban

>> Visit county web sites. You can grab sample ballots there, and these
>> samples are important!
> Is that because residual errors may be due to ballot design? Darnit, I
> need an assistant to do this study. It is a huge project just to
> gather
> all the data and info.

Indeed. My preliminary analysis of the data from election 2000 in
Florida shows lots of dependency on typography for optical mark-sense

>> Do get the figures before the recount season begins! Don't let
>> lawyers
>> meddle with the numbers you use in this kind of statistical analysis.
> Youch! What are you saying? The numbers posted on the county web
> sites
> keep changing? Yikes.

Depends on the state and the county. Some will put up unofficial
early returns, first official canvass and second canvass as separate
items. Some just keep one public portal that holds the current best
available numbers and keeps changing until the lawyers are agreed.

                Doug Jones
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Mon Nov 1 15:28:58 2004

This archive was generated by hypermail 2.1.8 : Mon Nov 01 2004 - 15:28:59 CST