Posting this to support for now -- we'll upgrade this to an RCR if it comes to that. This is posted on behalf of Jane Barth.
Jane, we GEMS does not include blank votes in the under-vote statistic as long as the blank voted race stat is shown on the report. In short, we already report exactly as they require. In the example they cite, if there were no undervotes (instead of 2595) and 35 blank votes, then the number of undervotes would have been reported as -70 following their logic. The 2595 total does not include the 70 votes due to blank races. They are just the undervotes – that is, number of times people voted for only one candidate out of two in this race.
----- Original Message -----
From: "Jane Barth" <email@example.com>
To: "Dmitry Papushin" <firstname.lastname@example.org>
Subject: RCR for N.E.A. - Reporting change
> They are running
> 220.127.116.11 and would like this request by May of this year. This is their
> request via email sent to me:
> "The National Education Association runs its elections in a non-standard
> way. In a "multi-candidate / vote for many" race, the NEA needs a final
> report that does NOT count a blank race as fully undervoted. GEMS counts
> blank race as all undervotes, while we need a report that would report a
> blank race as a "blank voted race".
> -----Original Message-----
> From: Dunne, Sebastian [NEA] [mailto:SDunne@nea.org]
> To: Jane Barth
> Subject: RE: Word document for review
> How about use the attached as a starting point. It's pretty close to what
> you've got already and it should describe exactly what we want. Let me
> if it doesn't make sense.
Description: MS-Word document