US Supreme Court Says: Need ID to Vote

[snoopy369]

Originally posted by Ramo
I'm not saying it's onerous. Not for me, anyways. But I'm not an old person living in a ghetto without access to a car. More importantly, regardless of the magnitude of the hassle involve, it reduces the number of people in the process for no good reason. I think that makes it a bad law.

do 75% of blacks vote during elections?(The answer is "No")

Black voter turnout is only slightly lower than white voter turnout. Controlled for income, actually, black voter turnout is higher IIRC.

1. There are generally DMV branches in appropriate areas (in Chicago, for example, near where I live, which while not a ghetto is a 'safe area' (university) surrounded by very poor neighborhoods on all sides, there is a huge DMV branch just outside the college area (63rd/Cottage Grove).

2. You SHOULD have an ID. Things that make IDs easier (such as requiring states to give them freely) are a GOOD thing for poor people (allowing them to get bank accounts, for example).

3. Proof of already-occurring fraud should not be necessary. This is simply something that should be done - you should not be able to vote without showing you're you. It's just common sense. Fraud could happen - and may well be - but regardless of whether it IS happening, this is just such an obvious thing - and in no way a meaningful deterrent. Focus on the other deterrents to voting if you are worried about voter turnout; this is a common sense plan that should be approved.

[b etor]

oh cool

[Ramo]

curious.

Looking at midterm data over time can be questionable. IN, for instance, was unusally politically active in '06; there were more Dem pickups in that state than any other (except a larger state - PA, with which it tied). And of course, '06 turnout was generally larger across the country than in '02.

The paper cited is using 2 data sets: Census and County Turnout and then uses the voting req's as variables.

It looks like they used a wide variety of variables (i.e. competitiveness in the Presidential race, having a competitive state-wide race, demographics, etc.). And then they did a linear regression...

Further, it uses data from one election, 2004. I don't find data from one election to be particularly persuasive, I would rather see data compared over time, before and after the voting requirements changed.

Putting in data from other years would make for a better model, but I don't see how that makes the current model inadequate.

Please correct me if I've missed something, I'm just a layman here.

Aren't we all?

[snoopy369]

Not having data from more than one election means you can't control for unique elements in that election...

I'd say it needs at least 20 years' of data to be particularly interesting, myself. Anything less would be quite subject to the whims of the political winds...

[Ramo]

Regarding unique elements in the election, that's why the authors of the study used independent variables to classify states that had competitive Presidential races and competitive state-wide races.

3. Proof of already-occurring fraud should not be necessary. This is simply something that should be done - you should not be able to vote without showing you're you. It's just common sense. Fraud could happen - and may well be - but regardless of whether it IS happening, this is just such an obvious thing - and in no way a meaningful deterrent. Focus on the other deterrents to voting if you are worried about voter turnout; this is a common sense plan that should be approved.

1. If, empirically, there is a significant reduction in turnout due to voter ID, would you still support it?
2. Obviously, there are other things that we can do to increase turnout. For instance, making election day into a holiday (or at least hold it on a Saturday) would substantially boost voter turnout. But the powers that be don't like that. They also, coincidentally I'm sure, like voter ID laws...

[asleepathewheel]

Originally posted by Ramo


Looking at midterm data over time can be questionable. IN, for instance, was unusally politically active in '06; there were more Dem pickups in that state than any other (except a larger state - PA, with which it tied). And of course, '06 turnout was generally larger across the country than in '02.

So that data is questionable, but the other that supports the theory isn't?

Originally posted by Ramo


It looks like they used a wide variety of variables (i.e. competitiveness in the Presidential race, having a competitive state-wide race, demographics, etc.). And then they did a linear regression...

Sure. But you can throw as many variables as you want at something, your base datasets are your base datasets.

Originally posted by Ramo
Putting in data from other years would make for a better model, but I don't see how that makes the current model inadequate.

Its inadequate to me for the same reasons you criticized looking at just the Indiana results.

Originally posted by Ramo
Aren't we all?

Sure.

[Sirotnikov]

Not requiring photo ID is stupid.

hundreds of dead people vote in israel
especially ultra orthodox that are all dressed in black and have beards... there is limit to how much you can tell those apart.

[Ramo]

So that data is questionable, but the other that supports the theory isn't?

That single piece of data is obviously not comparable in import to a thorough linear regression.

Sure. But you can throw as many variables as you want at something, your base datasets are your base datasets.

What's wrong with the base datasets? The only legitimate problem is that it might not be big enough, but I don't see how one can trivially arrive at that conclusion.

[chequita guevara]

Originally posted by Sirotnikov
Not requiring photo ID is stupid.

hundreds of dead people vote in israel

Hundreds? Out of millions? I guess you really, really need that id.

[Ming]

Originally posted by Sirotnikov
hundreds of dead people vote in israel

Only hundreds... what a bunch of amateurs... they should come to Chicago and see how it's properly done

[chequita guevara]

Chicago should come to Florida to see how it's done.

[Ming]

People in Florida only seem dead...

[chequita guevara]

In Chicago, a few thousand dead people vote. In Florida, a few thousand non-existent people vote, plus, a few thousand people who've moved vote more than once, and tens of thousands of people who should be able to vote aren't allowed.

[asleepathewheel]

Originally posted by Ramo

So that data is questionable, but the other that supports the theory isn't?

That single piece of data is obviously not comparable in import to a thorough linear regression.

Sure. But you can throw as many variables as you want at something, your base datasets are your base datasets.

What's wrong with the base datasets? The only legitimate problem is that it might not be big enough, but I don't see how one can trivially arrive at that conclusion.

They're taking census info and determining a causation between lower than expected turnout and voting requirements, without anything else to back up their claim. The census data (pg 15 for table) apparently didn't include a question for nonvoters on procedural voting booth hurdles. I find that problematic. Yes, the lower than expected turnout could be due to voting requirements, but it could be due to a myriad of other things.


I think they are making a leap here that may or may not be true, but isn't "proven" by their own data. I tend to be overly critical on things like this. while I am personally a layman on these things, my wife isn't, so I do get some exposure to these issues (she doesn't deal with elections so isn't overly familiar) and, as an odd coincidence, I was at the conference the paper was presented at, but didn't go to that panel. I wish I had, it sounds more interesting than what I was probably doing at the time.

Spoiler:

yes, I just played the Hillary "experience" card on you

[David Floyd]

If this stops poor liberals from voting, I'm all for it. The country doesn't need to get any more ****ed up, as it would if poor liberals turned out en masse.

What we should do is make voting largely irrelevant on the federal level, by actually enforcing a strict construction of the US Constitution. Then, no matter what ******* you vote into office, he can't actually do all that much anyway.