December 5, 2007

The database underlying CNE's First Bulletin, anyone?

Quico says: As soon as CNE published this website breaking down the Referendum Results by location, my more computer savvy readers realized it would be possible to harvest the data that went into it.

That allows you to reconstruct the database underlying CNE's election night results announcement. With that information in hand, flushing out any hanky panky with the numbers becomes easy, regardless of whether CNE gives out any more information or not.

Longtime reader amieres got there first, for which we all owe him a big thank you:
Click here to download the data underlying CNE's First Bulletin. [excel spreadsheet, 1.3 MB, zipped.]
The data show the first bulletin was based on a count of 86.42% of the machine tallies (29,072 out of 33,639), though it's nuts what we had to go through to get a precise number on that. The data is still incomplete (there is no field for null votes or for abstention, for instance) because, obviously, you can't harvest data CNE hasn't published. But at least now it's in easy-to-read, easy-to-compare and, crucially, easy-to-fill-in-the-blanks form.

The student movement made a big thing about having "all the tallysheets in hand." If they have them, it's straightforward (if time consuming) to fill in the gaps in this spreadsheet with the data from the 4567 actas that CNE did not include in its first bulletin.

That means we can have complete, official, table-by-table, audited results whether or not CNE deigns to announce them.
Enough speculation. Enough rumors. Enough quick counts and projections and guesstimates. Enough bullshit:
Acta por acta.
The rest is noise.