Greg J
04-28-2005, 12:37 PM
I made this post originally for a message board some friends of mine run, but like 5 people read that board. I thought I might as well post it here too. This is geared towards a more leftist audience, so those of you with a conservative bend, please forgive some of the unflattering ways I refer to Bush voters. This is meant to be an objective analysis, and I think it is.
***************************
I am one of those leftists that have a big problem with conspiracy theories. They are generally poorly thought out and pretty much always have an agenda. A lot of kooks and crazies on the left like to posit outrageous things like “we know where Osama Bin-Laden is, we just want to keep him uncaptured for propaganda purposes” or “the Bush Administration knew the 9/11 attacks were coming” and stupid stuff like that. I dismiss this kind of stuff for what it is: garbage.
So, you can imagine my standard response when someone tells me “the Republicans stole this election! Kerry actually won. There was MASSIVE voter fraud.” I generally give my standard stop-being-a-whiny-little-liberal-bitch response in my infamously untactful form (those of you that know me well know what I’m talking about here): “Yes but you forgetting about something very important.” “Oh yeah, what’s that Greg?” “The little green men on the grassy knoll – what role did they play in all this?” That leaves them feeling confused, then stupid, and my job is done – they feel dumb for having brought it up (as I thought they should), but more importantly, my message has gotten through: don’t believe this kind of crap.
However, a recent write up in the Washington Post has given me pause. This is done by a guy I respect: Terry Neal. He is a political reporter and columnist, and is, in my mind, pretty good at what he does. He is a political moderate and a registered independent. He pisses off people on both the left and the right, so, in my mind, he HAS to be doing something right. He recently brought up a report done by an academic trained as a public opinion analyst, which argues, based on exit polling data, that there WAS mass voter fraud in the 2004 Presidential election. This triggered my “this might need to be looked into” radar, so I have decided to do that. I am going to read the report, and post what I think of its findings here. I think I am qualified to do so – I have spend the last several years of my life becoming an expert in public opinion theory and analysis. It’s what I specialize in. As some of you know, I’m currently ABD. I’m not trying to toot my own horn here as to make the point that I know what I’m talking about when it comes to this kind of stuff. I say all this having not yet read the report – that comes next.
AFTER HAVING READ THE REPORT – my analysis:
First, who is conducting this study? Terry Neal notes the organization that published the report as “left leaning.” That, in my mind, is irrelevant when considering the actual researchers. There are twelve authors of this study, consisting of political scientists and statisticians. I have no doubt that this is not some disingenuous, partisan, or politically motivated report (crackpots on the right will believe otherwise). The people conducting this study are, in my opinion, trying to be fair and totally objective. Furthermore, the mathematical analyses are, as far as I can tell, impeccable. The theory, on the other hand, is not.
The authors discuss 3 explanations for the discrepancy between the exit poll and the election results:
1) Random error. This is the idea the there will always be some sampling error and that no poll will ever be 100% accurate. This is why when people on the news report polls you always here about a “margin of error.” This explanation is ruled out as virtually statistically impossible in the present case. The discrepancy between the exit poll and the election results is HUGE – much to big to be explained by random error.
2) Conventional wisdom. This can be summed up pretty simply as the idea that Bush voters were underrepresented in the exit polls. The writers of this report fault the authors of the initial report of the exit poll results for not being able to adequately answer why this is so. They criticize the main explanation of the difference between the exit poll and the election results: Bush voters systematically refused to answer exit pollsters questions.
3) Massive voter fraud. This report supports the notion that massive voter fraud took place in the most pro-Bush districts. However, they don’t really offer a whole lot of evidence to back up this assertion. Mainly they just criticize what the previous analysis of exit poll data for its flaws (and yes it does have some).
I won’t get into a detailed theoretical analysis, b/c no one wants to read that anyhow. I will say that this report basically blasts the initial report that came out by the group that conducted the exit polls for being wrong. They don’t offer that much to support their own argument. The most valid point I think they make is that response rates for the exit polls in more pro-Bush districts is higher, so how can it be that pro-Bush voters are refusing to answer the exit polls? Good question, and I think that both the initial report and the report being discussed here missed this point.
The point is that, from election to election, different people can vote. In this election, a lot of homophobic bigots, religious wackos, and pro-war psychopaths came out in droves to support ole W. Furthermore, Bush has largely brought the right wing militia, black helicopter crowd to the polls. These are people that are likely not to vote in most elections, but they turn out for a president they like. This is a group that is not, however, prone to want to talk to a lot of people. Getting past the pejorative language (sorry, could not resist), the Republicans did a great job mobilizing traditionally unmotivated voters by fostering issues those people cared about. This lead to the Bush vote being systematically underrepresented in the exit polls, especially in the most pro-Bush areas (since there are more of them to be underrepresented). If we are to conclude that the American electorate is stable from election to election, then there might be something to worry about here. But a lot of freaky, scary, and generally nasty people showed up to vote to show their support for bashing gays and killing Arabs. These also happen to be people that don’t like talk to exit pollsters.
It boils down to this: the problem with this report is that it does not take sociological points of view into consideration. It is done from a number crunching point of view. Like I said, the math in this report is legit. The logic and theory is not very good though. There are more plausible reasons than fraud to observe this kind of difference between the election and the results. I played a little at offering one here.
One legitimate issue this report brought up: this data needs to be made publicly available on the precinct level.
So there is my take. There was no massive voter fraud in my opinion. As much as I, and many of you, would like to think otherwise, Kerry did not really win the 2004 election. I wish I could say otherwise, but I can’t.
Links: http://electionarchive.org/ucvAnalysis/US/Exit_Polls_2004_Mitofsky-Edison.pdf
This is a link to the study I just talked about.
http://www.washingtonpost.com/wp-dyn/content/article/2005/04/24/AR2005042401545.html
This is a link to Terry Neal’s column.
***************************
I am one of those leftists that have a big problem with conspiracy theories. They are generally poorly thought out and pretty much always have an agenda. A lot of kooks and crazies on the left like to posit outrageous things like “we know where Osama Bin-Laden is, we just want to keep him uncaptured for propaganda purposes” or “the Bush Administration knew the 9/11 attacks were coming” and stupid stuff like that. I dismiss this kind of stuff for what it is: garbage.
So, you can imagine my standard response when someone tells me “the Republicans stole this election! Kerry actually won. There was MASSIVE voter fraud.” I generally give my standard stop-being-a-whiny-little-liberal-bitch response in my infamously untactful form (those of you that know me well know what I’m talking about here): “Yes but you forgetting about something very important.” “Oh yeah, what’s that Greg?” “The little green men on the grassy knoll – what role did they play in all this?” That leaves them feeling confused, then stupid, and my job is done – they feel dumb for having brought it up (as I thought they should), but more importantly, my message has gotten through: don’t believe this kind of crap.
However, a recent write up in the Washington Post has given me pause. This is done by a guy I respect: Terry Neal. He is a political reporter and columnist, and is, in my mind, pretty good at what he does. He is a political moderate and a registered independent. He pisses off people on both the left and the right, so, in my mind, he HAS to be doing something right. He recently brought up a report done by an academic trained as a public opinion analyst, which argues, based on exit polling data, that there WAS mass voter fraud in the 2004 Presidential election. This triggered my “this might need to be looked into” radar, so I have decided to do that. I am going to read the report, and post what I think of its findings here. I think I am qualified to do so – I have spend the last several years of my life becoming an expert in public opinion theory and analysis. It’s what I specialize in. As some of you know, I’m currently ABD. I’m not trying to toot my own horn here as to make the point that I know what I’m talking about when it comes to this kind of stuff. I say all this having not yet read the report – that comes next.
AFTER HAVING READ THE REPORT – my analysis:
First, who is conducting this study? Terry Neal notes the organization that published the report as “left leaning.” That, in my mind, is irrelevant when considering the actual researchers. There are twelve authors of this study, consisting of political scientists and statisticians. I have no doubt that this is not some disingenuous, partisan, or politically motivated report (crackpots on the right will believe otherwise). The people conducting this study are, in my opinion, trying to be fair and totally objective. Furthermore, the mathematical analyses are, as far as I can tell, impeccable. The theory, on the other hand, is not.
The authors discuss 3 explanations for the discrepancy between the exit poll and the election results:
1) Random error. This is the idea the there will always be some sampling error and that no poll will ever be 100% accurate. This is why when people on the news report polls you always here about a “margin of error.” This explanation is ruled out as virtually statistically impossible in the present case. The discrepancy between the exit poll and the election results is HUGE – much to big to be explained by random error.
2) Conventional wisdom. This can be summed up pretty simply as the idea that Bush voters were underrepresented in the exit polls. The writers of this report fault the authors of the initial report of the exit poll results for not being able to adequately answer why this is so. They criticize the main explanation of the difference between the exit poll and the election results: Bush voters systematically refused to answer exit pollsters questions.
3) Massive voter fraud. This report supports the notion that massive voter fraud took place in the most pro-Bush districts. However, they don’t really offer a whole lot of evidence to back up this assertion. Mainly they just criticize what the previous analysis of exit poll data for its flaws (and yes it does have some).
I won’t get into a detailed theoretical analysis, b/c no one wants to read that anyhow. I will say that this report basically blasts the initial report that came out by the group that conducted the exit polls for being wrong. They don’t offer that much to support their own argument. The most valid point I think they make is that response rates for the exit polls in more pro-Bush districts is higher, so how can it be that pro-Bush voters are refusing to answer the exit polls? Good question, and I think that both the initial report and the report being discussed here missed this point.
The point is that, from election to election, different people can vote. In this election, a lot of homophobic bigots, religious wackos, and pro-war psychopaths came out in droves to support ole W. Furthermore, Bush has largely brought the right wing militia, black helicopter crowd to the polls. These are people that are likely not to vote in most elections, but they turn out for a president they like. This is a group that is not, however, prone to want to talk to a lot of people. Getting past the pejorative language (sorry, could not resist), the Republicans did a great job mobilizing traditionally unmotivated voters by fostering issues those people cared about. This lead to the Bush vote being systematically underrepresented in the exit polls, especially in the most pro-Bush areas (since there are more of them to be underrepresented). If we are to conclude that the American electorate is stable from election to election, then there might be something to worry about here. But a lot of freaky, scary, and generally nasty people showed up to vote to show their support for bashing gays and killing Arabs. These also happen to be people that don’t like talk to exit pollsters.
It boils down to this: the problem with this report is that it does not take sociological points of view into consideration. It is done from a number crunching point of view. Like I said, the math in this report is legit. The logic and theory is not very good though. There are more plausible reasons than fraud to observe this kind of difference between the election and the results. I played a little at offering one here.
One legitimate issue this report brought up: this data needs to be made publicly available on the precinct level.
So there is my take. There was no massive voter fraud in my opinion. As much as I, and many of you, would like to think otherwise, Kerry did not really win the 2004 election. I wish I could say otherwise, but I can’t.
Links: http://electionarchive.org/ucvAnalysis/US/Exit_Polls_2004_Mitofsky-Edison.pdf
This is a link to the study I just talked about.
http://www.washingtonpost.com/wp-dyn/content/article/2005/04/24/AR2005042401545.html
This is a link to Terry Neal’s column.