The photo above shows John McCain using the dreaded "Shaolin death grip" he learned in Vietnam from the late Kung Fu superstar Bruce Lee on President Bush, causing the president to scream in pain. It's too bad McCain ignored my advice to use the Shaolin death grip when he shook Barack Obama's hand before each debate. . . . . . Barack would collapse and writhe in pain, McCain could taunt him: So tell me whose bitch you are, Barack!, and the good senator from Illinois would be forced to squeal I is yo bitch, massuh John, I is yo bitch!
I believe such a moment could have indeed changed the course of this election, but that is not to be.
The Obama campaign has been very impressive in its organization, especially its use of the internet to mobilize supporters, while the McCain campaign has resembled a bit of concept-driven, avant garde performance art that has gone badly awry. Or maybe the "Red Zone Cuba" episode of Mystery Science Theater 3000 is a more apt comparison. At any rate, by this time 48 hours from now I firmly expect Obama to be our next president.
The real suspense is going to be in comparing the final polls to the actual election results. The final polls are all over the place, and it's clear that some major errors are being made in polling methodologies and analysis. For example, Gallup today gives Obama a lead of 8% while IDB/TIPP has Obama with a 2% lead. That's the sort of difference that can't be explained as normal variations in the data; something else is going on. And as someone fascinated by statistics-----three college courses in it-----I have some ideas.
One obvious problem is that the sample of a political population self-selects; no one can be forced to participate in a poll. I have read that about 20% of those contacted by political pollsters decline to take part. That has to introduce a huge error into the results, although the extent and direction of that error can't be determined. But it is definitely there.
Pollsters also contact people via landline telephone numbers. But an increasing number of people only have cell phones, and those people are omitted from any polling. I was in that situation when I lived in Las Vegas; I had a landline number, but I used it exclusively for my fax machine. If you wanted to make a voice call to me, you had to reach me at my cell phone. Contacting only those with landline phones is another source of sampling error.
But perhaps the biggest source of error in this election will be the Bradley effect. This is the dirty little secret we're not supposed to talk about; as a nation, we like to tell ourselves race will not be a factor in elections. But I don't believe it. I am confident there are quite a few white voters who will not vote for Obama because he is black but will not admit that to a pollster-----instead, they will say they are undecided or even say they are voting for Obama.
Versions of the Bradley effect are found in other areas. For example, it has been repeatedly shown that people are much less like to admit to certain beliefs and behaviors via personal interviews (either face-to-face or by telephone) than they will via anonymous written questionnaires. Just a couple of weeks ago I saw an item where a survey was conducted of married women from 25 to 40 on the subject of infidelity.Only 1% of women interviewed by telephone admitted to having had an affair; that number jumped to 8% on anonymous written questionnaires.
Because of the Bradley effect, I think Obama's margin of victory is going to be less than the polls indicate. I suspect he will win by 3% to 5% in the popular vote, but have a very healthy margin in electoral votes. It will be interesting to see if exit polling is any more accurate than it was in 2004, when exit polls had such laughably inaccurate results as John Kerry winning South Carolina. My advice would be to ignore any and all exit polls tomorrow; wait until actual vote totals start coming in before drawing any conclusions.
I cast my ballot in early voting last week, and I voted for Obama. I did so because I respect his intellect, because I feel we need a president uncontaminated by the "beltway mentality," because. . . . . . . ah crap, I'm not going to lie to you. I'm dying from cancer. I'm worried that maybe my college professors were wrong, that there really is a heaven and a hell, that I'm going to have a lot of explaining to do before much longer. . . . . . . . so maybe if I vote for a black guy for president, thereby demonstrating I am really A Good Person after all, maybe I can plea bargain down to probation and a couple of hundred hours of community service instead of eternity on the Rotisserie Of Divine Vengeance.
An act of desperation? Yes, but I am a desperate man.
I believe such a moment could have indeed changed the course of this election, but that is not to be.
The Obama campaign has been very impressive in its organization, especially its use of the internet to mobilize supporters, while the McCain campaign has resembled a bit of concept-driven, avant garde performance art that has gone badly awry. Or maybe the "Red Zone Cuba" episode of Mystery Science Theater 3000 is a more apt comparison. At any rate, by this time 48 hours from now I firmly expect Obama to be our next president.
The real suspense is going to be in comparing the final polls to the actual election results. The final polls are all over the place, and it's clear that some major errors are being made in polling methodologies and analysis. For example, Gallup today gives Obama a lead of 8% while IDB/TIPP has Obama with a 2% lead. That's the sort of difference that can't be explained as normal variations in the data; something else is going on. And as someone fascinated by statistics-----three college courses in it-----I have some ideas.
One obvious problem is that the sample of a political population self-selects; no one can be forced to participate in a poll. I have read that about 20% of those contacted by political pollsters decline to take part. That has to introduce a huge error into the results, although the extent and direction of that error can't be determined. But it is definitely there.
Pollsters also contact people via landline telephone numbers. But an increasing number of people only have cell phones, and those people are omitted from any polling. I was in that situation when I lived in Las Vegas; I had a landline number, but I used it exclusively for my fax machine. If you wanted to make a voice call to me, you had to reach me at my cell phone. Contacting only those with landline phones is another source of sampling error.
But perhaps the biggest source of error in this election will be the Bradley effect. This is the dirty little secret we're not supposed to talk about; as a nation, we like to tell ourselves race will not be a factor in elections. But I don't believe it. I am confident there are quite a few white voters who will not vote for Obama because he is black but will not admit that to a pollster-----instead, they will say they are undecided or even say they are voting for Obama.
Versions of the Bradley effect are found in other areas. For example, it has been repeatedly shown that people are much less like to admit to certain beliefs and behaviors via personal interviews (either face-to-face or by telephone) than they will via anonymous written questionnaires. Just a couple of weeks ago I saw an item where a survey was conducted of married women from 25 to 40 on the subject of infidelity.Only 1% of women interviewed by telephone admitted to having had an affair; that number jumped to 8% on anonymous written questionnaires.
Because of the Bradley effect, I think Obama's margin of victory is going to be less than the polls indicate. I suspect he will win by 3% to 5% in the popular vote, but have a very healthy margin in electoral votes. It will be interesting to see if exit polling is any more accurate than it was in 2004, when exit polls had such laughably inaccurate results as John Kerry winning South Carolina. My advice would be to ignore any and all exit polls tomorrow; wait until actual vote totals start coming in before drawing any conclusions.
I cast my ballot in early voting last week, and I voted for Obama. I did so because I respect his intellect, because I feel we need a president uncontaminated by the "beltway mentality," because. . . . . . . ah crap, I'm not going to lie to you. I'm dying from cancer. I'm worried that maybe my college professors were wrong, that there really is a heaven and a hell, that I'm going to have a lot of explaining to do before much longer. . . . . . . . so maybe if I vote for a black guy for president, thereby demonstrating I am really A Good Person after all, maybe I can plea bargain down to probation and a couple of hundred hours of community service instead of eternity on the Rotisserie Of Divine Vengeance.
An act of desperation? Yes, but I am a desperate man.
Posts
