Thursday, August 21, 2008

Follow Through Thursday

1) The allegations made against petition circulators for the Payday Lenders have faded to the back burner. Every entity collecting signatures for the referendum effort is required by law to file a 'Form 15' with the Secretary of State. Lisa Renee at Glass City Jungle tried to get the SOS to tell her who had filed the forms, and was given a pretty silly runaround. I can sympathize. In the meantime, although the consortium of Ohio Newspapers that is rating the truthfulness of political ads this season has seemed somewhat more subjective and capricious in their ratings than I would like, they have consistently rated the Payday Lenders as among the most dishonest advertisers on the air.

2) Despite my suspicions of polite dismissal, The Franklin County Board of Elections has, in fact, updated their front page with a modified version of the identification requirements. My main concern was with the Drivers' License Address issue, and most people will still find it unclear that a current and valid DL does not necessarily display the current address, and that the displayed address is irrelevant. It does remove the language regarding current addresses on Military IDs. The net effect, however, seems to more tightly link address information to the non-photo forms of ID. Which is an improvement, but still somewhat unfortunate, because most forms of photo Identification DO require a current address. There is a specific exception for Driver's Licenses and State ID's that does not apply to something like a passport that might otherwise meet the requirements. This is the new BOE webpage text:

Voters must bring identification to the polls in order to verify identity. Identification may include current and valid photo identification, a military identification, or a copy of a current utility bill, bank statement, government check, paycheck, or other government document, other than this reminder or a voter registration notification, that shows the voter's name and current address. Voters who do not provide one of these documents will still be able to vote by providing the last four digits of the voter's social security number and by casting a provisional ballot...
This is the section of law ORC 3501 (aa) that I feel is still not being communicated well:
(AA) “Photo identification” means a document that meets each of the following requirements:

(1) It shows the name of the individual to whom it was issued, which shall conform to the name in the poll list or signature pollbook.

(2) It shows the current address of the individual to whom it was issued, which shall conform to the address in the poll list or signature pollbook, except for a driver’s license or a state identification card issued under section 4507.50 of the Revised Code, which may show either the current or former address of the individual to whom it was issued, regardless of whether that address conforms to the address in the poll list or signature pollbook.

(3) It shows a photograph of the individual to whom it was issued.

(4) It includes an expiration date that has not passed.

(5) It was issued by the government of the United States or this state.
I know that I'm repeating myself, but I'm convinced that a significant number of provisional ballots are cast due to this particular misunderstanding of the law, and that a non-trivial number of people skip voting because they believe that they don't have a valid ID conveniently available, when in fact they do. I've emailed Mr. Stinziano to ask why the BOE prefers not to clarify this issue.

3) I hand entered the table of mock voting results that the consultants from Sagata and Lextant wouldn't give to me. So far, thankfully, no big red flags. I'm uncomfortable with their sampling logic (but that's an argument that goes far beyond this study, and they're certainly not out of the mainstream), but there is nothing in the data that gives a strong indication that their estimates are biased in a particular direction. There's not really enough data to reliably test whether or not income, education, voting experience, and age matter separately and/or together to influence voting time, so assuming that these things don't matter is actually better than simply guessing that they do. In fact, the few times where statistically significant differences appear, they aren't where you'd expect. The take-home: ratchet-back my alarm volume a notch. If and when I have time to play with the other tables, I'll update again.

No comments: