A meme that is currently picking up traction in the conservative blogosphere is that the list of dealerships to be shuttered as a result of Chrysler's bankruptcy contains a disproportionate number donors to Republican candidates. There have been furious efforts to prove this contention by looking up campaign contributor lists at the Huffington Post, Open Secrets, and other places.
There is just one problem with this theory. Nobody has bothered to look up data for the control group: the list of dealerships which aren't being closed. It turns out that all car dealers are, in fact, overwhelmingly more likely to donate to Republicans than to Democrats -- not just those who are having their doors closed.
Here, for instance, is what Huffington Post's Fundrace site turns up for those who list their occupation as "auto dealer":
Republican donations outstrip Democratic ones by about 8.6:1. Next, let's try "car dealer":
For some reason, those persons who describe themselves as "car dealers" are just slightly more likely to donate to Democrats than those who call themselves "auto dealers". Nevertheless, the list of contributions tilts Republican by better than a 3:1 margin.
Next up, "automobile dealer":
Roughly a 10:1 advantage for Republicans. Finally, we'll look at the slightly more obscure formation of "automotive dealer":
Big Republican edge here too.
Combining the data:
Overall, 88 percent of the contributions from car dealers went to Republican candidates and just 12 percent to Democratic candidates. By comparison, the list of dealers on Doug Ross's list (which I haven't vetted, but I assume is fine) gave 92 percent of their money to Republicans -- not really a significant difference.
There's no conspiracy here, folks -- just some bad math.
It shouldn't be any surprise, by the way, that car dealers tend to vote -- and donate -- Republican. They are usually male, they are usually older (you don't own an auto dealership in your 20s), and they have obvious reasons to be pro-business, pro-tax cut, anti-green energy and anti-labor. Car dealerships need quite a bit of space and will tend to be located in suburban or rural areas. I can't think of too many other occupations that are more natural fits for the Republican Party. Unfortunately, while we are still a nation of drivers, we are not a nation of dealers.
5.27.2009
News Flash: Car Dealers are Republicans (It's Called a Control Group, People)
@ 6:50 PM
Subscribe to:
Post Comments (Atom)
173 comments
Jeez, Nate, now you're providing context for numbers? Do you see CNN or Fox News doing that? I don't THINK so. This is yet another in a long string of examples of your obvious bias towards reality.
Great analysis, Nate!
Now the "dealer" and maybe the "sales manager" may be the main GOPers on the staff. I would guess that salespeople and mechanics, as well as accountants and other administrative staff, lean Democratic. But they probably don't give a lot of money to either party.
Great analysis. It's amazing how often people jump to conclusions like this without looking a little deeper. And that's what's wrong with this kind of crazy. These folks aren't looking for the truth, just for whatever they see that fits into their distorted world view.
Brilliant! Nate, I'm so glad you are on our side.
You make us white guys proud.
The one thing I'd love to see, though, is how many car dealership OWNERS can be accounted for. I mean, I'm sure the math is fine, and the difference would be even more stark, but I can't help but worry that those you examined could include a bunch of lot salesmen. Maybe "car dealer" implies ownership of the lot, but we might be throwing in a bunch of salesmen.
Those "car dearlers" USED to be Republicans. Odds are they are DEMOCRATS now!
Nothing like unemployment and bankruptcy to make you favor government subsidies!
Nate -
You forgot one thing.
Dealers of American cars (however we define "dealer") are probably more even likely to be Republicans than the general population of car dealers.
But in fact, those who reported this as a conspiracy are neither madmen nor idiots, despite how obvious the real explanation is.
No, they merely assume that their audience is made up of madmen and idiots, whom they can manipulate. As indeed it is, if it was fooled even for a second.
Don Beyer, who was listed in the first graph as a $4,600 Obama donor, is a former (1989-97) Lt. Governor of Virginia, and was the Democratic nominee for Governor in 1997.
Minor problem here: Simply saying car dealers are Republican donors does not explain the statistical result: Closed Chrysler dealerships did not give to the Obama campaign. Open dealerships tended to give to the Obama campaign.
In a few days when the numbers have finished being calculated, from open source data that can be freely examined, will you be retracting your post if it shows an obvious disproportionate bias on the grounds of political donations. After all, if the Republicans had used political donations to determine which companies where destroyed, you would be all over it.
Georgfelis: Erm...He didn't simply say car dealers are Republicans. He provided a link and search terms that show that 88% of car dealers' campaign contributions go to Republicans. If you have any numbers that indicate that figure is incorrect, provide them. Otherwise, saying you disagree with them because...umm...it doesn't fit your world view(?)... is not going to convince anyone.
One thing to be careful about when counting is owners vs. dealerships. Many owners own several dealerships, as the list you link to shows. If you count each occurrence of a name, you might get different proportions than if you count each name once no matter how many dealerships each owns.
Another thing to note, which is impossible to ascertain from donor lists, is the amount of "successful" dealerships (by whatever metric that can be measured) that donate Democratically vs. Republican. Also, dealerships that are less likely to be closed should be generally nearer urban centers, hence have a higher incidence of Democratic loyalty.
All in all, saying that the dealerships that are getting closed are being closed because they donated to the Republicans is ludicrous. If Chrysler and any other car manufacturer doesn't base their decision to close dealerships on anything other than good business sense (keeping open highly successful or large target audience dealerships only,) I would be surprised. Of course, lack of business sense is what got them into this in the first place...
Oh, that brings me back to the day when the owner of the dealership I worked at called a meeting and told us to vote for the Republican candidate- that we should "vote for our paychecks." Unfortunately for conspiracy theorists, that dealership didn't get cut.
The sort of analysis the conservative blogosphere does NOT go into is hilarious. And then they dispute anything that anyone might say to dispute the non-analysis.
All you neo-con troglodytes - Will you agree with me that not all equilateral triangular polygons (a polygon which has all sides of the same length) have corners that form a 60 degree angle, but in fact you can form an equilateral triangular polygon with corners that form 90 degree angles?
Read and think VERY carefully before you respond. And if you don't respond, we'll just presume that either:
1. You're too intellectually lazy to think about it; or
2. You agree, but are too lazy to state how it's possible.
Mike in Maryland
My Blogger ID is http://www.blogger.com/profile/02848893412251095965
Mike in Maryland.
I'm not a conservative blogger, but...
You can draw an equilateral triangle with three 90° angles...
... you just need to do so on a sphere and not on a flat piece of paper.
I think I missed the point. I mean, I could imagine that a blogger could be ignorant of spherical geometry and still be able to put together a coherent political opinion. Not that this was a particularly coherent example of conservative thought.
~ Rebecca,
(Does the tilde prior to your screen name mean 'somewhat', so your screen name is 'somewhat Rebecca'?)
So you are not a conservative blogger? But you DO admit to being a 'neo-con troglodyte'.
You'll notice that I didn't address my challenge to conservative bloggers.
Quote:
All you neo-con troglodytes. . . .
Thank you for identifying yourself as such. Now we won't have to read your neo-con rubbish to determine your political orientation. We'll be able to skip it, as you've already proven that your powers of observation are seriously lacking.
I will state that you seem to be a somewhat more intelligent neo-con troglodyte than most who post here.
Mike in Maryland
My Blogger ID is http://www.blogger.com/profile/02848893412251095965
Mike in Maryland = Math troll? (and not even particularly difficult math?)
Now I've seen everything.
It's not that difficult people. Once all the open dealers have been evaluated you can combine the two stay pools and find the general affiliations according to donations of ALL Chrysler dealers, that becomes the control. You then evaluate closed vs open vs control. Save other variables (I.e. Profit, location, etc) all should be similar. Major discrepencies between the donation leanings would be highly suspect.
Mike in MD:
One (maybe the only one?) equilateral triangular polygon with corners that form a 90 degree angle is called a square.
WV: charicat: Republicans might enjoy red rice and kidney beans if they treat Sonia Sotomayor charicatively.
Andrew:
You may be getting a little ahead of the ball here. At least two of the sites that have jumped on the issue have, in fact, started working their way down the list of dealerships that have not been axed. If it turns out that Democratic donors preponderate on the surviving side, the very aggregated data you're relying on to make your case for dismissal, as it were, will actually give considerable heft to the truther version.
Apparently, there is a group of dealers who are challenging the closures, and from what I've read so far, the metrics being used don't systematically include what would strike me as a rather obvious profitable vs. unprofitable assessment. There are certainly other potentially salient factors, but it is as yet unclear what they are.
This may well be a tempest in a teapot, but I think the jury is still out.
I'd like to see a geographic analysis to determine if the closed dealerships tend to be in republican areas. I found local news sources from AL, OH, and TX, and their maps included dealers from metro areas as well as more outlying areas. So with a cursory search, I don't see any biases based on location. But then I would need to see a map of all Chrysler dealers. It does look like the smaller the dealership, the more likely it was to close.
But the contributions of the dealership owners should not be the only variable here. What about checking the contributions of the employees? I can't see the political benefit in punishing one person's contribution by eliminating the jobs of his/her employees who may have voted democrat. The premise doesn't make sense. Seems to me like a conspiracy theory.
Surprised to see just how much disparity there is in donations, especially in a year that Obama pulled in far more money than McCain.
Hey! I bought my Mazda from Jackie Cooper in OK! Good to know that the Coopers are fine, upstanding Democratic contributors.
Georgefellis said
Minor problem here: Simply saying car dealers are Republican donors does not explain the statistical result: Closed Chrysler dealerships did not give to the Obama campaign. Open dealerships tended to give to the Obama campaign.
-------------------------
Its customary to read a post before commenting on it. Nate's whole post was about the fact that car dealerships as a whole donate disproportionately to GOP candidates. Nowhere have I seen any statistics that suggest that 'Open chrysler dealerships tended to give to Obama.' Nate gave a statistical analysis of car dealerships to show that as a group car dealers donate to Republicans. He even used numbers to do so. Would you like to link to something that shows open Chrysler dealerships having donated to Obama?
"Unfortunately, while we are still a nation of drivers, we are not a nation of dealers. There's More... " - Nate
Why would we want to be a nation of wheeler dealers, especially when they disproportionately support Republicans? Surely the decline of the Auto industry is good news both for the environment and for Democrats? It would be interesting to see if Dealers of more economical car makes are proportionately any less Republican!
Pointing out the obvious here, but I'm guessing that the fact that the very first name on that list is a Republican Congressman from Florida with statewide ambitions may have had something to do with this meme emerging.
Thanks for the data analysis. It provides better context than anything else I have seen but it still is near worthless (no offense meant). You are correct that we need to compare the closed dealerships to the open dealerships and using the political affiliation of all Chrysler dealerships as a control group would be an improvement. However, what is really needed is a look at the profitability of the dealerships, whether they were closed, and the donation history of the owner of the dealership. Multivariate regression is great for this type of problem. The R-squared value would let us know the percent of variance that each variable explains. You would hope that the R-squared for profitability would be very high (approaching 1) and the value for donations would be low. That is the type of analysis that should be conducted before making accusations but unfortunately the right has been too lazy or innumerate to do so.
Thanks for taking the conversation in the right direction!
If they were REAL CONSERVATIVES, of course, they'd be delighted at the chance to GO GALT!!!
Good statistics. Thanks for making my day a little brighter (despite the fact it seems it will never stop freaking raining.....aaaagggghhhh!!!)
There have been two claims in the reports that I've read.
One is that the axe is falling on dealerships whose owners are contributors to Republican or non-Obama Democratic candidates. That will lend itself to quantitative analysis and sort itself out.
The other is that sources at the auto company say they didn't want to cut any dealers but the administration made them do it -- and that the dealers being cut are making money and not costing the auto company any money. I'd like to see some fact-checking on that.
Although your statistics are compelling, my understanding is that the complaint is not simply that there are a disproportionate number of republican-owned dealerships being shuttered, but that (1) specific owners, who are democratic supporters are being left open in areas where it is not clear they are objectively superior to their discarded competitors and (2) the process of selecting dealers to be closed is not transparent. number 2 seems to be the real issue..
(Reposted to fix mistake)
OK, people who list their occupation as some variation of "car dealer" tend Republican.
Fine. But not conclusive proof of anything either way.
First, your stats are for "car dealerships"; that muddies the issue, when you should be focused on Chrysler dealers.
Next, your list fails to distinguish between owners and employees; the former are the most salient factor in this discussion.
Third, you haven't compiled numbers comparing the donations of dealers who are being closed versus the dealers who are remaining open. There should be three lists - donations to GOP candidates (both towards primary campaigns and towards the McCain campaign), donations to Clinton and other Democratic primary candidates, and donations to Obama, and then compare the lists of closures to these lists.
But what about Bud Selig!!!
Another way to look at it here:
http://theautoprophet.blogspot.com/2009/05/no-political-bias-in-chrysler.html
This is one of the worst examples of faux-analysis I have ever seen. Let me make this very simple for you, the author, and your readers. If the ratio of GOP and non-Obama-Democrat donors compared to Obama donors is the same on both lists, no foul. If the ratio of GOP and non-Obama-Democrat donors compared to Obama donors is higher on the "closed" list than it is on the "open" list, foul.
To put it another way, if there are just as many (proportionally) GOP and non-Obama-Democrat donors on the "open" list as there are on the "closed" list, no foul. If there is a higher number (proportionally) of GOP and non-Obama-Democrat donors on the "closed" list than on the "open" list, foul.
Citing that 90% of dealerships are GOP donors does not address the argument in any way. Allow me to illustrate. Agreeing with the premise that 90% of dealerships overall are GOP donors - if 98% of the dealers on the "closed" list are GOP donors, and only 82% of the dealers on the "open" list are GOP donors, the numbers are showing a clear disparity based on political affiliation. If 90% of the dealers on both lists are GOP donors, the numbers are not showing a disparity.
Get it?
The overall 90% is only a baseline. It does not address, and the author does not even attempt to address, any disparity between the two lists. I don't believe the author is stupid, so I can only assume this falls under the category of intentionally deceiving those who are.
Re: the comment left by "The Auto Prophet"
The article you linked to only considers the political affiliations of 25 (out of several hundred) dealerships on the "open" list. This is akin to doing the first step of a lengthy algebra problem and claiming to have completed the equation.
Nate, use your immensely capable statistical brain and not your biased political one. If (and it is only a hypothesis at the moment) it turns out that the partisan distribution of car dealers is 88/12 while the distribution of closed dealers is, say, 98/2 that is absolutely a statistically significant difference and one of the biggest political stories of the last decade.
Amidst all of the smugness here is a whole lot of self-distraction from the facts of this story.
The original story reported on Red State initially counted only $200-300 given to Obama by closed dealers VS. MILLIONS given to Republicans by closed dealers.
Hundreds VS. Millions cannot simply be explained away by the fact that productive business owners and job creators naturally tend to be Republicans. Obama thuggery would have to be involved if such a disparity were in fact true.
Now that was just a blogger happening upon a coincidence and he stressed that he was only going by numbers available through fundrace (although most if not all should be available there).
There is also now emerging the suggestion that whole Republican congressional districts were targeted while Democrats' districts were spared.
This blogger either didn't read the full accusations against the administration or is simply attempting to sweep it under the rug.
It seems to me that the GOP was defeated so soundly in '08 and '06 that they find themselves now looking for a Golden Ticket back to prominence. Or maybe, to snatch Obama's golden ticket. They are so hungry for a scandal, so desperate to believe that his secret evil is real that they are mining every non-story for sparks of rage to fuel the kindling of their political resentment.
Honest to God. Non-story here. You guys should go find a new way to imply Sotomayor is stupid or something.
The point that all you right wing hacks seem to be missing is that most car dealers (as Nate has shown here) tend to be Republican. There is nothing in anything that I have seen on this story that actually suggests seriously that Chrysler dealerships were closed because someone had donated to Republican candidates, save for the squeeling of a paranoid right wing desperate to find something to smear the Obama administration with. Its actually quite embarrasing.
Let me sum up Nate's article again- yes the Chrysler dealerships that closed were largely Republican, but actually most car dealers are in fact Republican.
Even if there is a difference in campaign donations between closed and open auto dealers, this does not imply causality.
The democratic contributors might just be better dealers with better community networks, hence doing better in crisis...
Tomas
"Nate, use your immensely capable statistical brain and not your biased political one. If (and it is only a hypothesis at the moment) it turns out that the partisan distribution of car dealers is 88/12 while the distribution of closed dealers is, say, 98/2 that is absolutely a statistically significant difference and one of the biggest political stories of the last decade."
*Absolutely* a statistically significant difference? Just out of curiosity, can you explain which statistical test you would apply to this data (88/12 vs 98/2)? Because if that's all you got, you will find it difficult to reject the null hypothesis.
"It seems to me that the GOP was defeated so soundly in '08 and '06 that they find themselves now looking for a Golden Ticket back to prominence. Or maybe, to snatch Obama's golden ticket. They are so hungry for a scandal..."
This is what is known as an ad hominem attack. You lose.
"The point that all you right wing hacks seem to be missing is that most car dealers (as Nate has shown here) tend to be Republican."
We're not missing it at all. Read my lengthy previous comment and try to wrap your mind around what was said.
"Even if there is a difference in campaign donations between closed and open auto dealers, this does not imply causality."
If no reason for the difference is provided, yes, it does imply causality.
"The democratic contributors might just be better dealers with better community networks, hence doing better in crisis..."
Nope. Some of the dealerships that were closed were among the most successful, profitable dealerships.
"*Absolutely* a statistically significant difference? Just out of curiosity, can you explain which statistical test you would apply to this data"
If the closed dealerships are disproportionately GOP and non-Obama-Democrat donors (as compared to the ones that are allowed to remain open), it is pretty self-explanatory.
The RW effort to make a mountain out of a molehill in this case (or a molehill out of randomness) is typical of what they've been trying to do to Obama since last summer: make one criticism after another, one argument after another, one label after another (antiAmerican, socialist, fascist) in hopes that something -- anything -- might stick.
None of it has.
In the meantime, this search for something with staying power has done nothing but weaken and discredit the conservative movement. They are not putting forward any kind of coherent world view -- it's almost all mudslinging and random "gotcha" attacks -- while the vast majority of people can just look at what Obama and his administration are trying to do and realize that these attacks on him are arbitrary.
We see the same stupid effort to discredit Sotomayor, taking statements out of context (and without a comparative reference, e.g., to what other future judges may have said or done prior to their appointment to the SC). It's counterproductive in this case because it is likely to further alienate the Right -- and the GOP -- from the growing Latino population.
This lack of a coherent strategy is related to Nate's article of a few weeks ago about the RW's need for some more real thinking -- and thinkers. And I'm not referring to RW radio hosts.
The Obama admin has done a very good job of making the image of Limbaugh and Newt and Cheney the "face" of the GOP -- as opposed to the GOP Congressional leadership (where in hell has McConnell been, for example?) or to the RNC leadership.
So what will the next attack on Obama be from the right? It can hardly be predicted -- just some further futile random jab. And that will make the conservative leadership look even more like the teabag crazies.
Just two quick comments:
88% vs 92% is statistaclly insignificant. For those who keep spouting if it is like 82% vs 98% it means something is up, well it isn't 82 vs 98 so sit down.
If someone says they are an auto dealer they are not an employee. Duh.
Facts are facts, which modern day Republicans have a difficulty understanding. It is like they live on their own little island. There are Dems like this too, who during Bush's term believed every single conspiracy theory regarding the Bush admin. The difference between the two is the Rep whackos are the ones running their party and they drive away all who don't participate in their mass delusion.
""*Absolutely* a statistically significant difference? Just out of curiosity, can you explain which statistical test you would apply to this data"
If the closed dealerships are disproportionately GOP and non-Obama-Democrat donors (as compared to the ones that are allowed to remain open), it is pretty self-explanatory."C'mon, humor us and explain it.
Long before Chrysler ever entered bankruptcy, its executives in many quoted places in the media said the company's fortunes would sink or swim based on volume of sales -- not just those of Chrysler but the whole industry. If I recall, their make-or-break line was 10 million units sold in a year, or thereabouts. They said if the industry as a whole sold that many cars, they would seize their fair share of the market, and survive.
So I have to believe that, as numerous sources have confirmed, the dealerships being closed are those with the least volume. Those with the highest sales volume are remaining open.
Politics, and even the profitability, of each individual dealer don't appear to be playing a role.
Oh, and I suspect that the dealers that did survive have body shops. Insurance repairs are quite lucrative, particularly if they contract with insurance companies directly. This I think would be interesting to find out.
Also, I wonder how much politicking within the company the dealers that survived did?
Was Nordelli still in charge when the actual decisions were being made?
I would like to know more about the decisions, not because there is any merit to the donor charge, but because I would not be surprised if there were back room deals going on between execs and the dealers.
I am sure some of the brouhaha is from dealers who are no doubt devastated to be losing their life's work and Chrysler has not provided very much by the way of explanations. Many are just grasping for some reason, when it may just be that the dealer 5 miles away has drinks with the decisions makers when he is at the corporate offices and sends them all Omaha Steaks at Christmas. Or something else. I do think Chrysler does owe their dealerships some explainations.
The author doesn't even really need the "bad math" to explain such an obvious answer, when sound logic tells you everything is true in Nate's last paragraph.
Jim, given the number of dealers (n) I'm qite certain that a statistical test such a chi-square would show a high level of statistical significance for an 88/12 dealer differential compared to a 98/2 closing differential. I suspect that statistical significance would remain even with, say, a 95/5 closing differential.
George and UseYourBrain,
I believe the linked to articles state that 25% of dealerships were to be closed. Therefore, assuming on average an equal donation from each dealer, if 92% of those closed were Republican dealers, and from the total distribtion of dealers 88% are Republican we can calculate the distribution for open dealerships.
I think the following is correct, but let me know if I have made an error.
Let x be the percent of dealerships that are open.
Open + closed = total.
0.75x + 0.25*0.92 = 0.88
x = 0.867
Thus, if the assumption of near equal donations to politcal contributions holds, the number of open Republican dealerships would be about 87%. Very close to the 88% average for all dealerships, and implying no foul play.
Lucille @ 10:28:
Next, your list fails to distinguish between owners and employees; the former are the most salient factor in this discussion.Those people would likely have listed their occupation as "salesperson" or "mechanic", not "dealer."
Next?
I don't disagree with you, Patrick. All I said was that IF the closing ratio was 95/5 or higher that I suspect you'd find that to be a statistically significant difference. I think it is worth examining the data to find out if there was a statistically significant difference because the consequences of political motivation in the selection of dealerships to close are awesomely negative.
Nate Silver: still the man.
"The RW effort to make a mountain out of a molehill in this case (or a molehill out of randomness)..."
How do you know whether it is a mountain or a molehill? The point is that without an analysis of the difference between the "closed" list and the "open" list, it is foolish to assert what the author is trying to assert here.
"... is typical of what they've been trying to do to Obama since last summer: make one criticism after another, one argument after another, one label after another (antiAmerican, socialist, fascist) in hopes that something -- anything -- might stick."
I still cringe when I see someone who doesn't realize who Obama is. You are completely ignorant of his life history and apparently aren't paying attention to his policies either. Both reveal him to be, unquestionably, a Marxist. His twenty-year apprenticeship under a racist, anti-American "preacher" who he referred to as his mentor reveals something additional about him. His Hyde Park mansion bought with the help of the criminal slumlord Tony Rezko in exchange for political favors says something additional about him. I suggest you do what you should have done before you cast your vote - educate yourself about who this man is.
"We see the same stupid effort to discredit Sotomayor, taking statements out of context"
There is nothing taken out of context in the criticisms you are referring to. She plainly stated that federal court is where policy is made. She plainly stated that her status as a Hispanic woman conveys some possibility of better judgement than that of a white male. There is nothing unclear about her statements.
"The Obama admin has done a very good job of making the image of Limbaugh..."
Yes, the government really showed that private citizen! *confetti*
"88% vs 92% is statistaclly insignificant. For those who keep spouting if it is like 82% vs 98% it means something is up, well it isn't 82 vs 98 so sit down."
You don't know what the numbers are. That is the point. The article doesn't even touch on the numbers that are crucial to the determination the author is trying to make.
"So I have to believe that, as numerous sources have confirmed, the dealerships being closed are those with the least volume. Those with the highest sales volume are remaining open."
I don't know who "confirmed" this for you, but some of the dealerships that are being forced out of business are among the most successful, profitable ones in the country.
"I believe the linked to articles state that 25% of dealerships were to be closed. Therefore, assuming on average an equal donation from each dealer, if 92% of those closed were Republican dealers, and from the total distribtion of dealers 88% are Republican we can calculate the distribution for open dealerships."
Where did you get 92% as the percentage of closed dealerships that are Republican-owned? Where do you get 88% as the percentage of Chrysler dealerships overall that are Republican-owned? And why do you not notice that, even assuming these numbers, you are admitting that the closed dealerships are disproportionaly Republican-owned - 92% compared to the baseline of 88%?
explain the McLarty/Bob Johnson dealerships, all of theirs remain open and all of their competition(republicans) were closed.
Coincedence?
92% and 88% were the values Nate provided in his article. I stated an assumption "assuming on average an equal donation from each dealer," which I used to estimate the number of Republican donating dealers.
I have not made efforts to get specific Chrysler number, I was just doing a quick 'back of the envelope' calculation based on the numbers already provided.
The 88% vs 92% difference has already been discussed in detail by others.
You have to control for market-size as well. There could be sound, non-ideological reasons for closing dealerships in small markets (having mainly to do with economies of scale). But those smallvilles tend to be whiter and more Republican than the nationwide average.
So the relevant question is, grouping dealerships by volume and market size, is there a pattern to the closed dealerships versus the ones kept open? E.g., in markets with populations below 50,000, are the remaining dealerships more Democrat than the closed ones?
"92% and 88% were the values Nate provided in his article. I stated an assumption "assuming on average an equal donation from each dealer," which I used to estimate the number of Republican donating dealers.
I have not made efforts to get specific Chrysler number, I was just doing a quick 'back of the envelope' calculation based on the numbers already provided."
Exactly... we don't know what the percentage of Chrysler dealers who are GOP donors is, and even going with your assumptions, the "closed" list is disproportionately GOP donors.
To reiterate my point to the author - saying that the phenomenon of the "closed" list consisting almost exclusively of GOP donors can be dismissed because most dealers are Republican, as your article attempts to do, is silly. If an analysis of both the "closed" and the "open" lists reveals that GOP-donor-owned dealerships occupy the same percentage of entries on both lists, then okay, no foul. But if an analysis of both the "closed" and the "open" lists reveals that GOP-donor-owned dealerships occupy a larger percentage of the "closed" list than the "open" list, then this raises some serious questions. Let's see an analysis (and it better be complete, or we will be here calling you on it yet again).
"You have to control for market-size as well. There could be sound, non-ideological reasons for closing dealerships in small markets"
If a dealership is profitable, it is profitable. Some of the dealerships being forced out of business are quite successful. Closing them results in a loss of profit. They are being closed anyway. With no explanation. Under pressure from the Democrat administration's "automotive task force" - which is the only reason this is happening, as there is no sound business rationale for it on Chrysler's end. 2+2=?
I don't know if anyone has pointed this out in the comments yet but I think there's reason to suspect dealers are even more Republican.
Keep in mind there's a difference between an upscale dealership that sells new cars and one that sells only used cars yet the searches for 'auto dealer' 'car dealer' etc. cannot make that distinction. I wouldn't be surprised if Democratic donating car dealers tend to be in the lower end of the market (higher density urban areas tend to have lots of small car lots and many used car dealers). The true Obama donating car dealership might be very unusual indeed.
@useyourbrain: make use of your screenname and wrap your head around normalizing the political affiliation data using the total profitability equation *first*. I'm sure Chrysler could make big waves on the news without GOP help if they were told to close essential profit centers. We have to learn Chrysler's formula for determining the closings first. This includes not just raw sales data, but service/warranty/insurance/etc work, which factors into overall profitability.
"The true Obama donating car dealership might be very unusual indeed."
Speculation.
"I'm sure Chrysler could make big waves on the news without GOP help if they were told to close essential profit centers."
Considering that Chrysler is under the thumb of Obama's "automotive task force", I very seriously doubt it. What I said is a fact - some of the dealerships being closed are very successful.
@useyourbrain: I worked in HVAC, so I can tell you a thing or two about businesses that "appear" profitable, but bleed the supplier and surrounding support networks well enough to kill overall profitability.
@UseYourBrain:
You seem a smart guy. Instead of getting all up in a bunch, why not do the analysis that you seem to expect everyone else to do, and post your results? I'm sure everyone would love to see them.
Elementary probability lesson for wingnuts: let's say Barack Obama (because he is unquestionably a bad man) has been using a particular coin as a randomizer to pick which dealerships to close. Further suppose that you have managed to get hold of the coin.
If you toss the coin 10 times, and you get heads six of them, you haven't proven that the coin (process) was loaded. Now you haven't even proven a statistical disparity yet (as Nate demonstrates). But even if you have, that's a necessary condition to prove your allegation, not a sufficient one.
captcha: matio, as in "matio, come on now, brother, put it, put it where it's at now"
If a dealership is profitable, it is profitable.The profitability of a local dealership has no necessary connection with the costs Chrysler incurs in maintaining the relationship with that location, especially if Chrysler can sell just as many cars, and service their warranties more cheaply, with fewer locations.
If there were no costs (including opportunity costs) to maintaining the relationship between the manufacturer and the dealership, Chrysler/Obama wouldn't be able to "close" them, even if they cared to.
Low volume of new car sales and growth potential were the reasons they were closed. The "back end" of the store (parts and service)a body shop or used car volume mean nothing to the car makers. Chrysler makes money on new car sales and financing, hence the low volume, low potential stores were closed. Not political, no back room deals made, just low man out!
The burden of proof that there is discrimination against GOP donors in the axing of dealerships rests with those who claim that there is (a) such a correlation between donating to the GOP and the probability of being axed, and (b) that it is causal.
Nate's analysis, though crude in one sense (e.g., it didn't look at the size of the donations), nonetheless suggests that proving (a) (above) is going to be difficult, given that there is only a 4 pct. point difference in the proportion who contribute to the GOP among those who were axed and those who were not. And even if this were a "statistically significant" difference in proportions, proving (b) (causality) still remains to be done. Absent the discovery of some recordings or documents regarding the negotiations, it will likely be impossible.
For a couple of days, then, this is just another random shot taken at the Obama administration by the Right, built on flimsy evidence at best. It would be far better, IMO, for the discussion to focus on the issue of how many (if any) dealerships needed to be axed, and using what criteria, and not on simply attacking the outcome as politically motivated.
Citing that 90% of dealerships are GOP donors does not address the argument in any way. Allow me to illustrate. Agreeing with the premise that 90% of dealerships overall are GOP donors - if 98% of the dealers on the "closed" list are GOP donors, and only 82% of the dealers on the "open" list are GOP donors, the numbers are showing a clear disparity based on political affiliation. If 90% of the dealers on both lists are GOP donors, the numbers are not showing a disparity.
But knowing that the proportion of GOP donors on the "closed" list is 92%, and the proportion for all dealers is 88%, tells us that the proportion on the open list is somewhere around 87-88%, if the "closed" list is 25% of the size of the "open" list. The only way this is not so is if Chrysler dealers have a different proportion of GOP-to-Dem donors than the sample Nate pulled.
I'm not sure what this proves.
The list you provide adds up to just over 1000 dealers.
Last numbers I find (for 2002) lists 21,000+ dealerships in the US.
Did the other 20,000 make no donations? Is your list a statistically random sample?
The 88% could very well be true but without knowing how those lists were made I don't see how one can assume the other 20,000 automatically follow suit with the sample.
Let me slightly echo what Juris is saying. Its fascinating that the right seek to find almost criminal activity in the Obama administration, rather than to debate the point about whether these dealerships should be closed and how to deal with that tragedy.
That tells you a lot about the Republican Party and the mess it is in. When people decry Limbaugh for saying he wants Obama to fail, the problem is not that he is being nasty to Obama, but that he is being unthoughtful to those caught in the middle of it all. By wanting Obama to fail, Limbaugh is saying he wants people to be unemployed, he wants people to have the homes foreclosed etc. Its that level of unfeelingness that is killing the GOP right now.
The right's interest in this Chrysler story is about scoring political points and alleging unconstitutionality, not about helping out these poor dealers.
The rights attacks on Sonia Sotomayor have little to do with protecting the court, more about embarrasing a woman who has fought her way up from the projects of the Bronx to the doorstep of the US Supreme Court.
That is why people aren't listening. Thats what people are attacking when they call out 'partisan politics'. What people want is a genuine and honest debate, not point scoring and embarrassing the opposition. (Don't get me wrong here, there are times when the Democrats get into the gutter to debate politics, though I sense in general we progressives are less happy to do so!)
1. Do people really think that people are paid to sit down and work out precisely which dealers are Republican so that they can be closed down at a disproportionate ratio? I mean get real. It's not worth the effort.
2. Maybe they should be disproportionately closed so they know what it's like to live in the real world and not conservative fantasyland.
From the misleading use of probability theory department:
http://directorblue.blogspot.com/2009/05/dealergate-statistical-evidence-that.html
Let's say I toss a coin 5 times. The first toss is heads, the next four are tails. Using Doug Ross' statistical theory, the odds that I wind up with that exact sequence (one heads followed by four tails) are one in 32 (just over 3%). The coin must be loaded!
Ross is making a fundamental statistical error, since to take the Arkansas example, the probability (he also conflates odds and probability) for any given dealer remaining open while the two others are closed is .75 * .25 * .25, or < 5%. His "statistical analysis" proves nothing.
To do this right, you would need to run a logistic regression that includes the stated criteria that Chrysler said would be used to pick which dealers are open, and add a term for donations, and see if which party was donated to (or amount of donations) was a statistically significant predictor of the odds of closure, when adjusted for all the other variables. Simple to do if you have all the data.
Josh
"I worked in HVAC, so I can tell you a thing or two about businesses that "appear" profitable, but bleed the supplier and surrounding support networks well enough to kill overall profitability."
So we are speculating that they may not have been profitable, and when I point out that some very successful and profitable dealerships are being forced out of business, you defend your speculation with further speculation that they may still be unprofitable even though they appear profitable and you have no evidence to convince me otherwise.
I should also point out that dealerships don't bleed Chrysler - dealerships have to purchase their inventory outright, and they don't get to bill Chrysler for their operating expenses. The costs to Chrysler of having a dealership open do exist, but they are a drop in the bucket. As in, reps to communicate with the dealers, most of which is done via telephone and email.
"You seem a smart guy. Instead of getting all up in a bunch, why not do the analysis that you seem to expect everyone else to do, and post your results?"
I don't have to do the analysis to point out that it is sorely lacking in this article.
"If you toss the coin 10 times, and you get heads six of them, you haven't proven that the coin (process) was loaded."
That's a tough metaphor to swallow. Coin tosses are random by nature, whereas a decision process like this is going to be based on certain factors. A higher rate of heads than tails in coin tosses requires no explanation. A higher rate of GOP donors on the "closed" list than on the "open" list would require an explanation, and, absent a good explanation of what factors were used to justifiably arrive at these decisions, it raises some serious questions.
"The profitability of a local dealership has no necessary connection with the costs Chrysler incurs in maintaining the relationship with that location, especially if Chrysler can sell just as many cars, and service their warranties more cheaply, with fewer locations."
You don't sell just as many cars from one profitable dealership as you do from two.
"If there were no costs (including opportunity costs) to maintaining the relationship between the manufacturer and the dealership,...
I didn't say there were no costs, I said that we are talking about some very successful and profitable dealerships, the kind that create a high volume of sales and generate enough profit for Chrysler to cover their opportunity costs very easily.
"... Chrysler/Obama wouldn't be able to "close" them, even if they cared to."
I'm not sure where you get that idea. They can stop any dealership from calling themselves an official Chrysler dealership and selling new cars.
"Low volume of new car sales and growth potential were the reasons they were closed."
You have looked at some of the highly successful dealerships being forced out of business and determined that they are all short on new car sales and growth potential?
I should also point out that we are missing a key point here. Market forces will put the ill-performing dealerships out of business. Any dealership doing badly enough to actually create a negative money flow from Chrysler would have already gone out of business long ago.
"The burden of proof that there is discrimination against GOP donors in the axing of dealerships rests with those who claim that there is (a) such a correlation between donating to the GOP and the probability of being axed, and (b) that it is causal."
True, but raising a question requires no burden of proof. The question has been raised and has not been answered.
"Nate's analysis, though crude in one sense (e.g., it didn't look at the size of the donations), nonetheless suggests that proving (a) (above) is going to be difficult, given that there is only a 4 pct. point difference in the proportion who contribute to the GOP among those who were axed and those who were not."
First, no. The 4% difference is between the percentage of entries on the "closed" list that are GOP donors and the overall percentage of dealers that are GOP donors, not between the percentage of entries on the "closed" list that are GOP donors and the percentage of entries on the "open" that are GOP donors.
Second, the baseline number he is using is the percentage of GOP donors out of all dealerships, not the percentage of GOP donors out of Chrysler dealerships, which is sloppy to put it nicely, especially considering that a real analysis of this requires no such baseline number in the first place - it simply requires a comparison of the percentages on the "closed" list and the percentages on the "open" list.
"proving (b) (causality) still remains to be done."
Sure. But absent other believable factors being claimed and confirmed, it is the 800-pound-gorilla in the room.
"But knowing that the proportion of GOP donors on the "closed" list is 92%, and the proportion for all dealers is 88%, tells us that the proportion on the open list is somewhere around 87-88%, if the "closed" list is 25% of the size of the "open" list. The only way this is not so is if Chrysler dealers have a different proportion of GOP-to-Dem donors than the sample Nate pulled."
That is a pretty big if, considering that it is the baseline number from which you are attempting to extrapolate everything. And even using those numbers, once again, if the "closed" list is 92% GOP donors, and the "open" list is 87-88% GOP donors, this raises a question as to what factors were used in deciding which dealerships would be forced out of business by this textbook fascist policy.
"That tells you a lot about the Republican Party and the mess it is in. When people decry Limbaugh for saying he wants Obama to fail, the problem is not that he is being nasty to Obama, but that he is being unthoughtful to those caught in the middle of it all. By wanting Obama to fail, Limbaugh is saying he wants people to be unemployed, he wants people to have the homes foreclosed etc."
He wants no such thing. This is another example of an impressionable liberal buying into, and now repeating, a blatant misrepresentation made by partisan hacks like James Carville working from within our White House. Limbaugh was very clearly saying (I have read the transcript) that he wanted Obama to fail in advancing his Socialist agenda because he believes it is bad for the country. Yet you allowed someone to manipulate you into believing that he wants Obama to fail even at the expense of the country.
"The rights attacks on Sonia Sotomayor have little to do with protecting the court, more about embarrasing a woman who has fought her way up from the projects of the Bronx to the doorstep of the US Supreme Court."
The "attacks" on this woman are a factual discussion of her plainly stated beliefs - that judges make policy and that a Hispanic woman can make better judgements than a white male. Get off the kool-aid for ten minutes and think for yourself.
"though I sense in general we progressives are less happy to do so!"
You are a "progressive"? Do you understand that the progressive movement is literally a front for Communists? If so, and you still claim this label, you don't warrant any more of time time. If not, then I suggest you look into exactly what it is that you are identifying yourself with.
"Do people really think that people are paid to sit down and work out precisely which dealers are Republican so that they can be closed down at a disproportionate ratio? I mean get real. It's not worth the effort."
Closing dealerships allows the competitors that remain open to capitalize. If the ones closed are disproportionately GOP donors, the ones that remain are simultaneously less likely to donate to the GOP and more likely to have money to donate due to the influx of new customers. Worth the effort indeed.
"Maybe they should be disproportionately closed so they know what it's like to live in the real world and not conservative fantasyland."
It was precisely people like you who sat by while every tyrant in history, from Stalin to Mussolini to Hitler, rode roughshod over the freedoms of others and eventually their own.
Would the accusers please set aside their statistical fallacies for a moment?
Right wing commentators have blamed Obama for closing the dealerships. Actually, Chrysler planned to close the dealerships at least as far back as February, 2008. Chrysler's plan was called Project Genesis. Obama had nothing to do with that. He wasn't even in office yet.
http://www.carseek.com/news/february2008/Chrysler-announces-Project-Genesis/
Chrysler and GM will be liquidated, 100% of their dealers put out of business, their factories taken apart and shipped to China, India, Brazil, or maybe sold for scrap, and their workers can kiss their pensions and health benefits good-bye.
Oh wait, sorry, I was living in the real world for a moment. Back now. Whose gets to spin the wheel for the next round of Candyland Economics?
Before too many people lose their minds, may I point out something very simple?
Pres.Obama raised an unprecedented amount of money during the election. I don't think he's jealous of the donations.
The economy sucks (although getting better) and Obama is not stupid. Deliberately targeting some dealerships for purely partisan reasons is just a dumb idea.
My point:
Why in God's name would he or anyone in his administration give a rat's ass about targeting and closing a bunch (although a tiny % of the total) of GOP-staffed/managed dealerships?
I saw somewhere (not here) someone comparing this to the partisan-based firings from the DOJ during Pres.Bush's term. Bull.
Those at least had some form of political gain to be had and a purpose (albeit a bad one).
Is there a rational reason for this sort of targeting in CAR DEALERSHIPS?
Rattners wife, is a Top DNC fundraiser
So the guy making closing/shutting decisions wife knows all about the political donations game.
the case of the McLarty/Bob Johnson Dealers is very damning to Obama. The only way to figure this out though is not this silly post but to go through all Chyrsler dealers in question, one by one, and document their political donations and leanings. Then analyze, and only then.
Obama was making the decisions, not Chyrsler.
http://www.reuters.com/article/rbssConsumerGoodsAndRetailNews/idUSN2632731920090526
[i]A lawyer for Chrysler dealers facing closure as part of the automaker's bankruptcy reorganization said on Tuesday he believes Chrysler executives do not support a plan to eliminate a quarter of its retail outlets.
Lawyer Leonard Bellavia, of Bellavia Gentile & Associates, who represents some of the terminated dealers, said he deposed Chrysler President Jim Press on Tuesday and came away with the impression that Press did not support the plan.
[b]"It became clear to us that Chrysler does not see the wisdom of terminating 25 percent of its dealers," Bellavia said. "It really wasn't Chrysler's decision. They are under enormous pressure from the President's automotive task force."[/b]
[/i]
"Right wing commentators have blamed Obama for closing the dealerships. Actually, Chrysler planned to close the dealerships at least as far back as February, 2008. Chrysler's plan was called Project Genesis. Obama had nothing to do with that. He wasn't even in office yet."
Had they already decided which dealerships to close, or was this decision made under the watchful eye of Obama's "automotive task force"?
"Pres.Obama raised an unprecedented amount of money during the election. I don't think he's jealous of the donations."
Undercutting the Republicans' fundraising base is a major goal of Obama and the DNC. Don't kid yourself.
"The economy sucks (although getting better)"
It's not getting better, it's still shrinking and we are still losing jobs.
"Why in God's name would he or anyone in his administration give a rat's ass about targeting and closing a bunch (although a tiny % of the total) of GOP-staffed/managed dealerships?"
As I said, money (either directly, looking forward, or indirectly, thanking prior DNC contributors). And 25% is not tiny.
"the case of the McLarty/Bob Johnson Dealers is very damning to Obama."
It is, and you really have to have your head in the sand to not realize it. These people with DNC connections are being allowed to keep all of their dealerships open, while in each area their competitors' dealerships are being closed. Random chance does not account for this kind of "luck".
Well, if you want to thoroughly put this to bed, you should compare how much was donated to obama alone, because when the meme was started one of the accusations is if you gave to any of his "enemies" republicans or democrats included, you were more likely on the close list.
The right in the US have pinned all their hopes on Obama being a stupid criminal. That's why they chase these hail-mary stories as fast and as hard as they can. The post comparing this strategy to Wonka's "golden ticket" was right on the money.
If it turns out that Obama is merely a pragmatic, moderate politician (which is what the reality-based evidence points to over and over again), then they have *nothing* to offer.
We've seen this binary world-view play out on the Republican side for the last 8 years. I can't imagine what will change it.
To do this right, you would need to run a logistic regression that includes the stated criteria that Chrysler said would be used to pick which dealers are open, and add a term for donations, and see if which party was donated to (or amount of donations) was a statistically significant predictor of the odds of closure, when adjusted for all the other variables. Simple to do if you have all the data.Problem, you have to account for autocorrelation: If 88% of dealer donations are Republican then right off the bat there is something a bit unusual about the 12% of Democratic donations. Those dealerships are in a *different type of market*. You're only going to see a perfect 88%-12% split on an unbiased closed list if the political donations have nothing to do with the market conditions of the dealership....but that's a heroic assumption. Heavily Democratic areas tend to be more urban, more dense, have greater household income, higher land prices etc.
I should also point out that dealerships don't bleed Chrysler - dealerships have to purchase their inventory outright, and they don't get to bill Chrysler for their operating expenses. The costs to Chrysler of having a dealership open do exist, but they are a drop in the bucket. As in, reps to communicate with the dealers, most of which is done via telephone and email.A few things:
1. Dealerships bleed on financing. The company borrows cheap and allows dealers to buy on credit.
2. The dealerships bleed on support, transportation costs, as well as cannibalization (undercutting other dealerships etc.).
3. Companies that are not in bankruptcy like Ford and Honda have a much leaner dealership base. Dealerships have passed special laws that make it very hard to close them outside of bankruptcy. Now maybe the secret to profitability is *more* dealerships, but the evidence doesn't seem to point in that direction. Speculation? Possibly but I'm not seeing much in way of an argument against closing dealerships.
4. If dealerships are truely 'costless' then how is Chrysler closing them? Are you telling me if a dealer who is 'closed' calls up Chrysler and says they will immediately wire them $2M+ cash for a hundred new cars Chrysler will refuse to deliever them? If this were true the shareholders and bondholders would be fighting the closings tooth and nail in bankruptcy court.
What a brilliant plan! Let we the government, in our evil attempts at socialism, take over a failing company, then take deliberately partisan steps that will financially hamper the very company we seek to own and control, thus ensuring it will be that much closer to complete failure.
Nothing says "Look how well socialism works" like setting out to destroy the company you're diabolically seeking to own and control.
Who knew they were such masterminds?
"And 25% is not tiny."
No, 25% of Chrysler is not tiny. But the % of actual dealerships in the country IS tiny.
"Undercutting the Republicans' fundraising base is a major goal of Obama and the DNC. Don't kid yourself."
I don't doubt that - but is putting a few tens-of-thousands of employees out of work in a scenario where they're almost sure to blame you (the government) and thus donate anything they can to your political opponents REALLY the best way to do that? Must be that fabled Obama mastermind at work.
"raising a question requires no burden of proof. The question has been raised and has not been answered."
I therefore raise the question that you are not, in fact, a human being, but are really an alien sent to disrupt the commentary on a statistical blog for some overly-complicated and obscure reason whose grand plan is, no doubt, to destroy the United States. I have absolutely no proof nor rational reason to suspect this, but the burden of proof now lies upon you to disprove my absurd question.
See how silly that particular line of reasoning is? In order to expect a reasoned response, you have to give a reasoned suspicion/question - otherwise it deserves to get ignored.
"If it turns out that Obama is merely a pragmatic, moderate politician (which is what the reality-based evidence points to over and over again)"
Let's see.
Entire life spent as a student of, and ally to, a long list of Communists and Socialists. Twenty years spent as an apprentice of a racist anti-American "preacher" who he referred to as a mentor. Fifteen years spent owning a Hyde Park mansion bought with the help of the criminal slumlord Tony Rezko. Political career launched in the living room of an unrepentant domestic terrorist.
Nationalizing the banks, auto industry, and healthcare. Raising taxes across the board while claiming to cut them. Spending us into more debt with pork-laden bills in three months than the entire prior history of our country combined. Demonizing private citizens who dissent against the government. Proposing a second military just as big as our current military to be deployed across the United States. Walking all over our second amendment rights. Breaking his campaign promises by the dozens immediately upon taking the presidency. Bowing down to the leader of the Muslim world. Apologizing for America and legitimizing dictators around the world while alienating our best ally. Nominating extremist liberal judges and filling his cabinet with liberal tax cheats and conflicts of interest. Taking such an extreme stance on abortion that he not only supports partial birth abortion, he believes that a baby who survives an abortion should be allowed to die after birth with no medical treatment.
"Moderate"? Wake up. Wake up. WAKE UP.
"1. Dealerships bleed on financing. The company borrows cheap and allows dealers to buy on credit.
2. The dealerships bleed on support, transportation costs, as well as cannibalization (undercutting other dealerships etc.).
3. Companies that are not in bankruptcy like Ford and Honda have a much leaner dealership base. Dealerships have passed special laws that make it very hard to close them outside of bankruptcy. Now maybe the secret to profitability is *more* dealerships, but the evidence doesn't seem to point in that direction. Speculation? Possibly but I'm not seeing much in way of an argument against closing dealerships."
Once again, any dealership that is performing so poorly as to create a negative money flow from Chrysler would have gone out of business long ago and would not have to be forced out of business.
"4. If dealerships are truely 'costless' then how is Chrysler closing them?"
They can close any dealership they want to (or are forced to by Obama's "automotive task force") close. Or to be more accurate, they can make them stop displaying the Chrysler sign as an official dealership, stop selling Chryslers as "new", stop performing warranty repair work, etc.
"What a brilliant plan! Let we the government, in our evil attempts at socialism, take over a failing company, then take deliberately partisan steps that will financially hamper the very company we seek to own and control, thus ensuring it will be that much closer to complete failure."
Marxists do not necessarily care whether one of the companies they are taking over fails. They can continue to prop it up with monopoly money printed by the privately owned federal reserve and collaterized by the ability of this private bank to take the interest on this debt from us, the citizens, under penalty of imprisonment.
"No, 25% of Chrysler is not tiny. But the % of actual dealerships in the country IS tiny."
We are only talking about Chrysler here, and in the context of Chrysler, putting 25% of the dealerships out of business is a big deal.
"I don't doubt that - but is putting a few tens-of-thousands of employees out of work in a scenario where they're almost sure to blame you (the government) and thus donate anything they can to your political opponents REALLY the best way to do that? Must be that fabled Obama mastermind at work."
Let's see. Cost the GOP about 800 big donors in exchange for making a few thousand people who have no jobs or money want to donate to them? Seems like a good deal to me.
"I therefore raise the question that you are not, in fact, a human being, but are really an alien sent to disrupt the commentary on a statistical blog for some overly-complicated and obscure reason whose grand plan is, no doubt, to destroy the United States. I have absolutely no proof nor rational reason to suspect this, but the burden of proof now lies upon you to disprove my absurd question."
Unfortunately for your asinine metaphor, there is a rational reason to suspect that GOP donors occupy a disproportional number of spots on the "closed" list as compared to the "open" list. It is the fact that nearly every single one of them is a GOP donor. If you want to counter that by comparing the proportion of GOP donors on the "closed" list with that of the "open" list, fair game. But don't pretend to refute the idea when you do nothing of the sort, which is precisely what this article attempts.
use your brain said
'Limbaugh was very clearly saying (I have read the transcript) that he wanted Obama to fail in advancing his Socialist agenda because he believes it is bad for the country'
-----------------------------
Wow, do I go down the quoting out of context/hypocrisy route (espeiocally as UYB then uses quotes out of context arguments against Judge Sotomayor later on.) No. Because Limbaugh himself has quoted himself many many times on this out of context. He is happy, nay gleeful to be quoted out of context on this. He is more than happy to quote himself as saying he wants Obama to fail. But even so, the results of Obama's policies failing are unemployment, house foreclosures, etc etc etc. Limbaugh shows little or no sympathy with those who would suffer the results of Obama failing. THATS the point, not exactly Limbaugh's quote.
---------------------
UYB continues
'The "attacks" on this woman are a factual discussion of her plainly stated beliefs - that judges make policy and that a Hispanic woman can make better judgements than a white male. Get off the kool-aid for ten minutes and think for yourself.'
---------------------
Ok firstly there is the quoting in context thing, which apparently vexes you with Limbaugh, but you are happy to do yourself to Judge Sotomayor. [She actually said that judges make policy, but should be careful not to, and was pointing out that judges should be careful to think about there own prejudices when making judgements). But secondly, none of that has anything to do with Judge Sotomayor's legitimacy to sit on the bench. (Especially given that Justice Alito gave a similar response to the cultural quote in his testimony to the Senate judiciary committee during his confirmation hearings). Yet again you kind of prove my point that the right aren't interested in furthering policy discussion and would prefer personal attacks.
Use Your brain added
'Do you understand that the progressive movement is literally a front for Communists?'
-------------------------
Seriously? Your accusing the whole of the Democratic Party as being 'literally a front for the communists'? Wow I hadn't realised the fifties had come around again so quickly.
And for your benefit, the Progressive Movement in the US is a century old movement. Theodore Roosevelt was a good progressive.
Try taking your own advice and using your brain
Nate is right. And the conservatives/Republicans are making an absolute stink about nothing. Chrysler does have a plan for closing dearlerships--whether the plan is the right plan or not is not at issue. From a Freep.com article from May 15th Chrysler is proposing to close 789 dealerships. These dearlerships only made up 14% of the sales in 2008--not exactly go getters. Secondly around 50% sold less than 100 vehicles in 2008 and 44% of these dealerships were paired with a non-Chrysler automaker. Moreover, with these 789 dearlerships, only 62% of all the Chrysler dealerships sold just Chrysler products; with the closing of the 789 dealerships it will go up to 84%.
So given the nonsense arguments for blaming Obama, any statistical analysis would have to look at these varibles before concluding that more Republican dearlerships were closed than non-Republican.
Lastly, it seems so basic: if you have dealership A that sells 250 Chrysler cars a year and dealership B that sells 100 cars a year, and you cut back on the amount of autos you are going to manufacture--lets say in this instance from 350 per year to 200 per year, you would obviously keep dealership A open and close dealership B. This is what Chrysler is trying to do on a large scale.
As for dearlerships that are challenging Chrysler in court--it is based on how Chrysler made the decision to close, not some political plot.
This has got to be the funniest reach ever. Honestly, it surpasses BirthGate.
First of all, evidence of this sort of shenanigans leaves an indelible statistical footprint. Given the dataset, it should be trivial to prove some element of abuse. Therefore, the long we go without such evidence, the greater the chance this is just another right wing tinfoil sale.
Of course, I think most of us realize that waiting for proof of this is like waiting for the Whitey tape.
You should try to map the dealrship closings to the congressional district presidential map as shown at http://www.swingstateproject.com/showDiary.do?diaryId=4161 It might be worthwhile to see if dealers in "red" districts (voted for McCain) were closed more than those in "blue" districts (voted for Obama) I will be unable to do this myself but perhaps your crack research staff my be able to do this.
"He is happy, nay gleeful to be quoted out of context on this."
Sure, he doesn't mind the attention and resulting boost in ratings. That doesn't change the fact that his position was misrepresented and you bought into it.
"But even so, the results of Obama's policies failing are unemployment, house foreclosures, etc etc etc. Limbaugh shows little or no sympathy with those who would suffer the results of Obama failing."
No, Limbaugh believes that the things you are describing will be the results of Obama's policies being enacted. Limbaugh quite literally said that he wanted Obama to fail in advancing his Socialist agenda for the good of the country. Not that he wanted Obama to fail even if that meant the failure of the country.
"Ok firstly there is the quoting in context thing, which apparently vexes you with Limbaugh, but you are happy to do yourself to Judge Sotomayor. [She actually said that judges make policy, but should be careful not to, and was pointing out that judges should be careful to think about there own prejudices when making judgements)."
Garbage. She is not being quoted out of context in any way. I have seen the video of her talking about judges making policy. She said flat out that it was her view that judges make policy. It wasn't subtle and there was no other context. Then she joked around about the fact that she knows you're not supposed to say this and the cameras were there. Cameras being there wouldn't be a problem if she hadn't been saying the controversial thing she was saying. And I don't see how saying that a Hispanic woman would make better judgements than a white male constitutes a warning that judges should be careful to think about their own prejudices.
You seem to think it is opposites day. She says making policy is the role of federal judges, and then jokes about how she shouldn't say this because the cameras are there and it's controversial to say. You believe she is actually saying she doesn't believe this. She says a Hispanic woman will make better judgements than a white male. You believe she is actually saying they are equal and we should be careful about our prejudices.
"But secondly, none of that has anything to do with Judge Sotomayor's legitimacy to sit on the bench."
If she doesn't understand that the role of a judge is to enforce the law on the books, and not to make policy, that certainly does have something to do with her ability to perform properly as a judge (which is probably the reason 60% of her decisions are overturned on appeal).
"Seriously? Your accusing the whole of the Democratic Party as being 'literally a front for the communists'? Wow I hadn't realised the fifties had come around again so quickly."
I honestly feel bad every time I have to tell this to people who consider themselves "progressives", but it is a simple fact that today's "progressive" movement is a front for Communism in the United States. Progressives for Obama is an umbrella group that brings together the members of several known Communist and Socialist groups, and its parent group is chaired by two Communists named Mark Rudd and Jeff Jones who are both longtime Communists and allies of William Ayers. The progressive poster boy, Barack Obama, has spent his entire life as a student of, and ally to, a long list of Communists and Socialists. You can research this quite easily.
I supplied proof that the plan to close Chrysler dealerships started at least as far back as February, 2008. Obama could not possibly have had anything to do with that, since he wasn't in office. Closing the dealerships was Chrysler's idea.
Jonathan said:
"Obama was making the decisions, not Chyrsler."
http://www.reuters.com/article/rbssConsumerGoodsAndRetailNews/idUSN2632731920090526
[i]A lawyer for Chrysler dealers facing closure as part of the automaker's bankruptcy reorganization said on Tuesday he believes Chrysler executives do not support a plan to eliminate a quarter of its retail outlets.
Lawyer Leonard Bellavia, of Bellavia Gentile & Associates, who represents some of the terminated dealers, said he deposed Chrysler President Jim Press on Tuesday and came away with the impression that Press did not support the plan.
[b]"It became clear to us that Chrysler does not see the wisdom of terminating 25 percent of its dealers," Bellavia said. "It really wasn't Chrysler's decision. They are under enormous pressure from the President's automotive task force."[/b]
I provided proof that the plan started back in 2008. Did somebody put an article on the web dated February, 2008 as part of a conspiracy to fake people into thinking it was Chrysler's plan? Of course not.
Compare my proof to the veracity of a lawyer. Ha! The only thing the lawyer even did say was that it was his "impression" that Chrysler was against the closures. He had no definite knowledge. And further down in the article Chrysler denies what the lawyer says:
"Our position is that the market can't support the number of
dealers that are out there," said spokeswoman Carrie McElwee.
"This has been our plan for more than 10 years to combine
Chrysler, Dodge and Jeep under one roof."
It should also be noted that the lawyer for the dealers that Jonathan cites is hoping to prevail with a 5th amendment claim. 5th amendment claims can only be prosecuted against the government, not private entities. So the lawyer has to make up a story about the government ordering Chrysler to close the dealerships. Otherwise, he doesn't have a case.
That's not a compelling argument. We can easily differentiate 88% and 92% with a large sample size. (Just as we can differentiate between a batting average of .280 and .320)
Chrysler is closing 1/4 of it 3500 dealerships. If 88% are Republican donors, than the probability that 92% or more of the dealerships that close are Republican donors is 1/12480. (Binomial Distribution)
To put it another way, we are 3.6 standard deviations away from the mean.
Their may very well be confounding factors (for example in the aspect of people who are donors to both party), but those would need to be examined and explained before the argument could be dropped. Probably the best way would be to independently examine the fates of large Democrat donors among the car-dealers, as there are few enough that the effort would take less time.
Regards,
David
"I supplied proof that the plan to close Chrysler dealerships started at least as far back as February, 2008. Obama could not possibly have had anything to do with that, since he wasn't in office. Closing the dealerships was Chrysler's idea."
Was the decision regarding which dealerships to close made before Obama took office, or was it made under the thumb of Obama's "automotive task force"?
"Chrysler is closing 1/4 of it 3500 dealerships. If 88% are Republican donors, than the probability that 92% or more of the dealerships that close are Republican donors is 1/12480. (Binomial Distribution)
To put it another way, we are 3.6 standard deviations away from the mean."
Bingo.
Useyourbrain is almost a parody. Stop feeding the troll, or he's never going to stop.
I supplied proof that the plan to close Chrysler dealerships went back at least as far as February, 2008. UseYourBrain asked:
"Had they already decided which dealerships to close, or was this decision made under the watchful eye of Obama's 'automotive task force'"?
They didn't have an exact list but they knew what kinds of dealerships they wanted to close.
Here's a recent quote from Chrysler:
" The decision about cutting dealers took into consideration
factors like location, customer satisfaction, and sales
potential, she said. Nearly half of the terminated dealers also
carry non-Chrysler brands, and most rely on used vehicles for
the bulk of their sales."
From the February 2008 article:
"By reducing the number of unprofitable dealers operating in close proximity to one another, Chrysler hopes to push the advantage of marking its three brands under one roof "to get the full lineup of Chrysler Corp. everywhere" said Press."
So it looks like there wasn't a specific list before Obama came along, however there were plans for the TYPES of dealers to be shut. The final list mirrored the plan. Since 1/4 of dealerships were closed, and they were drawn from poorly performing dealerships, it would have been difficult to sort out donor types without shaking up the original criteria.
Matt: OK
Hi Nate... this is Doug Ross.
Your post is a bit out-of-date.
Please address the RLJ and Lithia dealership situations that I described here. Check out the maps and then let me know the probability of that happening the way Chrysler said it happened.
http://directorblue.blogspot.com/2009/05/dealergate-statistical-evidence-that.html
There's more where that came from.
Preparing it now.
Best, Doug
"Useyourbrain is almost a parody. Stop feeding the troll, or he's never going to stop."
When somebody says something like this, it tells me that not only do they have no idea what I am talking about, but they never will. Because they will simply never look into it for themselves and instead will accept whatever CNN/MSNBC/Factcheck/Snopes tells them. It is easier for some to live in ignorance than confront unpleasant realities.
"Since 1/4 of dealerships were closed, and they were drawn from poorly performing dealerships"
I think it is worth pointing out again that some of the dealerships being closed are high successful.
* highly *
They can close any dealership they want to (or are forced to by Obama's "automotive task force") close. Or to be more accurate, they can make them stop displaying the Chrysler sign as an official dealership, stop selling Chryslers as "new", stop performing warranty repair work, etc.You have failed to address the questions:
1. The statistics do not support the contention that an unusual number of GOP dealers were closed and Dem. dealers were spared.
2. MORE THAN ONE Democratic donating dealer was on the list *despite* the fact that the best numbers available tell us that very few dealerships would be Blue rather than Red. The partisan hack simply decided not to count them.
3. Dealers DO cost their companies money. That is why the more profitable firms have fewer dealers and the less profitable ones are saddled with many dealers. That is also why dealers, just like the unions, spent a lot of money lobbying and getting state laws passed making it hard for the manufacturers to close them. If dealers didn't cost anything, why would they have gotten laws passed to protect them? Why would they worry about getting closed if they were either costing the manufacturers nothing or were helping them be profitable?
Marxists do not necessarily care whether one of the companies they are taking over fails.No company that has been taken over has failed. The only companies that have been taken over are ones that already failed. The gov't has not been trying to buy controlling stakes in Microsoft or Wal-Mart. A company needing a bailout or in bankruptcy has, by definition, gotten there because it *failed*.
At this point you've established yourself as troll who doesn't deserve serious attention, but I'm writing this response for those who might still think you're making a serious argument.
Doug, I addressed your statistical analysis above. Any specific outcome by definition was an exceedingly rare event, just as flipping any specific distribution of 10 coin flips (in order) is going to be an exceedingly rare event. So I do not think your statistical analysis proves anything. Moreover, did you randomly pick four metro areas, or did you pick metro areas where you knew a democratic donor had survived. I could do the exact same thing; find a Republican dealer that survived, look at the specific distribution of dealership closures and conclude that it was a statistically rare event, and that Obama must be biased towards Republicans.
To do this analysis in an unbiased way, you would set up a logistic regression equation that evaluated closure as your outcome variable, and put in a number of predictors: specifically the stated criteria used to decide whether a dealership would stay open or be closed, and the political donations of the dealership. If you can assemble the dataset, the relevant statistics can be run in about 2 seconds on any statistical software. If you email me, I'll do it.
Josh
1/4 of 3500 dealers is 875. If 88% of dealers are Republican, you expect 770 Republican dealers to close. Large number / Gaussian statistics gives a 1-sigma deviation of square root(770) = 28, about 3.6%.
So you expect about 2/3 of the time from 84.4 - 91.6% of the closures to be Republicans, and 90% of the time about 81% - 95% to be Republicans.
If 92% of those closed are Republicans, it is entirely consistent with the closures being randomly chosen. To show bias requires then demonstrating that there was a business need to actually close a significantly larger percentage of dealers who happened to be non-Republicans.
And I'll say once again, the closing were not 'random'...or more precisely...there's no reason to think that there's no correlation between the party a dealer donates too and the 'unbiased' factors that lead to him being closed. It is clear a Democratic donating dealer is VERY rare. There is no reason to think that the demographics of a Blue donating dealer are those of the more normal Red dealers.
I'm rather amazed at the statistical ignorance being shown both by you Nate, and by your commenters. The relevant probability distribution here is the Hypergeometric. Using your numbers, and assuming 3000 dealerships, we get a 3.671535e-05 chance that the picks were by random chance. If we assume 3500 dealerships, the chance drops to 8.738693e-06.
Sorry, that's not even close to "reasonable random chance."
(You can see my work here.
Ah! C'mon you lunatic right wingers - are you still trying to run with this even though it's been proven beyond any doubt that there's nothing here? God Bless! You kids are DESPERATE!
Ronald Said:
If 88% of dealers are Republican, you expect 770 Republican dealers to close. Large number / Gaussian statistics gives a 1-sigma deviation of square root(770) = 28, about 3.6%.
So you expect about 2/3 of the time from 84.4 - 91.6% of the closures to be Republicans, and 90% of the time about 81% - 95% to be Republicans.
If 92% of those closed are Republicans, it is entirely consistent with the closures being randomly chosen. To show bias requires then demonstrating that there was a business need to actually close a significantly larger percentage of dealers who happened to be non-Republicans."
Why would you use Gaussian statistics?
This is a classic binomial distribution. It's variance is Npq=875*.88*.12=92.4
Standard Deviation is 9.6.
You can can get a brief primer on this at:
http://en.wikipedia.org/wiki/Binomial_distribution
Or I could refer you to some textbooks or lecture notes on the subject.
Your standard deviation is calculated as for a Poissonian process, which is an... interesting choice.
Regards,
David
gregq:
You are taking the 88% figure to be a bible at showing the split among Chrysler dealers on the whole. It was a quick number that Nate was able to generate looking at all company dealers donations to see how much a 92% figure should stand out and look disconcerting.
Since his figure was 88%, we quickly see how unmenacing a 92% figure is, given all the unknowns as well.
In other job types, 92% would be amazing to see, and would lead to a lot of questions.
If 92% of dealer money went repub, among both dealers that are "Chrysler" and "Closed", and a quick look at all dealers money went 88% Repub, you realize that trying to chase this theory is rather pointless.
Also, even if statistical relevance is reached against some generic body of dealers to compare, there are about 1,000 better explanations than deliberate acts by the WH based on donations to the GOP.
You have to start looking at sale volume, or location, or rural/urban, you have unknowns like internal lobbying within the company by dealers, whether certain areas or states with different political slants also had different dealership density within the same overall population size (i.e. do the "red" slanted areas that tick well above 88% have more Chrysler dealers per capita than other areas, and perhaps then higher closing % of all dealers?)
etc etc etc to really understand what the 92% figure should be stacked against.
But given Nate's findings that overall dealership donation already was incredibly slanted to Repub, and at 88%... there doesn't seem to be a whole lot of point to keep charging at the windmill on this one.
Gregg is correct. I rescind my previous posts and hide in shame. (If my thesis adviser found out, I might be out of a job). The correct distribution is hypergeometric, because binomial distribution is only with replacement. I haven't run his numbers, but they seem accurate, and certainly the probably in that distribution is lower than for my numbers.
So what do you say, Nate?
I haven't learned the math Gregg is using but I don't see how it applies. If we are talking about the GOP donating dealers the RW researches found there was ZERO chance they were chosen randomly. The researchers played around the list and picked out some double names. Double names would be owners of chains of dealerships, who would be wealthier than the members of the group as a whole.
My research, though not perfect, its been many years since I was indoctrinated by lib professors, finds that (4) 22.2% of the dealers on the closed list donated to the GOP. (1) 5.3% donated to Democrats. Nobody donated to both.
METHODOLOGY
Not being able to sleep, I took the number of closed dealerships at 789 and used a net random number generator to give me 20 random numbers. I only got 19, because one number came up twice.
I took a list of closed dealerships and counted to the rank on the list of the random selected numbers. This didn't work out perfectly, because my list only had 741 names. But none of my random numbers was above 739, so I ignored that.
I copied first and last names, along with middle initials. The data also gave city of the dealership.
I then ran the names through the search engine at OpenSecrets.org. I cast a wide net. First I searched for all the matches of last name in the entire state the dealer was located in. From that list I used only records where the first name matched the dealer's. Then, if I found a middle initial or a dealer occupation or a city match I counted that person as the dealer in question.
Like I said, out of the 19 I got 4 who donated to the GOP (total $75,800) and 1 (total $750) Democratic.
CONCLUSION:
This really needs a comparison sample from the dealers who did not close. If I still can't sleep tonight I may do one. But with the data from closed dealers sample I don't see anything remarkable, except that I think the data the RW researchers used is very suspect. Since there are 3 times as many open dealers, taking a sample would be lots of work.
If anybody wants they can E-mail me at creeksneakers2@aol.com and I'll send them the data.
Of course, all of this is silly anyway because I proved by a preponderance of evidence that the selections were made by Chrysler and not Obama in the first place.
Not reading through all the comments, so I dunno if someone else has already said this, but you should search on "auto sales," "Automotive sales" etc. too.
Hmm, I guess the assumption that 92% and 88% of the money donated corresponded to number of republicans was a bad bad assumption. If very few donated, and there is a large variance in donor numbers, we'd need to do more math to make a statistically relevant statement.
I think you need to do more numbers than CreakerSneekers did (or just do his run 1000 times or some such), to get a decent answer, but that would be the reasonable way.
OK. I've just finished a control sample from the open group to compare to my scores in the closed group above. Out of 32 scores, 4 (12.5%) donated to Republicans and 0 donated to Democrats.
METHODOLOGY
There were too many open dealerships to use random numbers.. So I used a systematic sampling strategy. There were 127 pages of PDF listings. I used the fourth entry on each fourth page, except I only counted 3 pages for the last score. The total was 32 scores.
Research on OpenSecrets.org was conducted the same way as in the closed dealer sample, except one record with a different first name but all other data matching was counted as R in the open group.
As before, E-mail me if you wish to see the data.
CONCLUSION:
I don't what can be made of any of this. I would have suspected a higher portion of donors in the open dealers since they probably sell more cars and have more money. The opposite turned out. A paranoid mind would conclude that GOP donors were more likely to be chosen for closure than for staying open. I don't know if there is any statistical significance to my data. I do know lib college professor indoctrinators always repeated, "Correlation is not causation."
I hope this helps put this silly hoax to rest.
ADDENDUM
I'd also like to add that between both groups there were 8 GOP donors and 1 Dem, so that's about the same GOP/Dem proportion predicted at the outset of this blog.
UYB
Either we can all quote out of context or none of us can. Sotomayor in both cases has been quoted whilst making a far wider point than you claim she is making. The quote about wise Latina judges is clipped out of a far wider point saying that judges use their own life experiences to help them make judgements but should be cautious about doing so. Justice Alito made the self same point during his confirmation hearings.
Judge Sotomayors comments about judges making policy is again part of her saying that judges should be careful not to, as you would know if you had seen the whole answer and not just the. Lip a right wing nut wanted you to see.
Then again Limbaugh did say he wanted Obama to fail.
It's worth noting that Chrysler dealerships would likely have even higher ratios than car sellers as a whole, since you rule out the people selling Prius cars, who are more likely to be liberal.
Of course the one thing everyone missed is the obvious. With about 90% of dealers breaking Republican the easiest thing for Obama to have done is simply let the company fall apart in bankruptcy. Then instead of 800 or so closed dealerships, you'd have closer to 3000. Sure a few Obama donating ones would get hit but many more GOP ones would go down.
Have a look at the last paragraph in this Freedomworks article.
It reads "...In 1995, Democratic districts received an average of $35 million more in federal largesse than Republican districts, which seems roughly fair given that Democratic districts have more people in need of government aid. By 2001, the gap had not only reversed, it had increased nearly twentyfold, with GOP districts receiving an average of $612 million more than Democratic ones. Justifying this shift, then- Majority Leader Dick Armey said, “To the victor goes the spoils..."
http://www.freedomworks.org/news/manservant
Oops, never mind. I had misread the article.
"The statistics do not support the contention that an unusual number of GOP dealers were closed and Dem. dealers were spared."
You don't seem to be adding anything to the debate here. What we do know is that the "closed" list contains a disproportionate number of GOP and non-Obama-Democrat donors as compared to the "open" list. We also know that a DNC-connected dealer that operates six dealerships is being allowed to stay open in every single area in which they operate while their competitors in each area are being forced out of business - calculate the probability of that being due to chance.
"MORE THAN ONE Democratic donating dealer was on the list *despite* the fact that the best numbers available tell us that very few dealerships would be Blue rather than Red. The partisan hack simply decided not to count them."
Non-Obama-Democrat donors, yes.
"Dealers DO cost their companies money."
Already addressed thoroughly.
"Why would they worry about getting closed if they were either costing the manufacturers nothing or were helping them be profitable?"
Ask the dealers who are highly profitable and were closed anyway.
"No company that has been taken over has failed. The only companies that have been taken over are ones that already failed."
Brilliant.
"At this point you've established yourself as troll who doesn't deserve serious attention"
At this point you've established yourself as a script kiddie who applies internetz labels to people with whom you disagree. WTFPWN!
"Ah! C'mon you lunatic right wingers - are you still trying to run with this even though it's been proven beyond any doubt that there's nothing here?"
What an insightful comment.
"Either we can all quote out of context or none of us can."
None of us can.
"The quote about wise Latina judges is clipped out of a far wider point saying that judges use their own life experiences to help them make judgements but should be cautious about doing so."
That's great, but her statement that a Latina woman presents the possibility of better judgement than a white male stands on its own, and the context you are describing does not change its meaning. This is the difference between out of context and not out of context.
"Judge Sotomayors comments about judges making policy is again part of her saying that judges should be careful not to, as you would know if you had seen the whole answer and not just the."
I have seen the video of her statement. I have seen what led into the statement (talking about why people want to get on the federal appelate courts), I have seen the statement itself (federal appelate courts are where policy is made), and I have seen what followed (joking that the cameras are there and she shouldn't say that, she doesn't support it, with a smirk on her face while the people in the room laugh). Nothing about the leadup or the joking that followed changes her quite serious statement that the courts are where policy is made.
"Then again Limbaugh did say he wanted Obama to fail."
I have already addressed this thoroughly, and you are not addressing what I said. You lose.
"With about 90% of dealers breaking Republican the easiest thing for Obama to have done is simply let the company fall apart in bankruptcy."
Then he wouldn't own them.
You don't seem to be adding anything to the debate here. What we do know is that the "closed" list contains a disproportionate number of GOP and non-Obama-Democrat donors as compared to the "open" list.We know nothing of the sort, in fact, the evidence seems to point in the opposite direction.
We also know that a DNC-connected dealer that operates six dealerships is being allowed to stay open in every single area in which they operate while their competitors in each area are being forced out of business - calculate the probability of that being due to chance.Dealerships were not choosen by 'chance' but by strategy. Clearly they want to reduce the number of dealerships in an area so that the ones that remain have less competition.
Non-Obama-Democrat donors, yes.Which blows your 'chance' idea out of the water. The only baseline numbers we have is an estimate of 88% of all donations going to Republicans and 12% going to Democrats by dealers. We have no numbers telling us that when dealers do donate to a party, how often they donate to the top of the ticket versus candidates below the top of the ticket. Lacking any numbers, I used a 50-50 split in my estimates but we really don't know. It's quote possible that dealers put 90% of their donations into local and regional races and only 10% in national ones. If that's the case, finding an Obama donating dealer by chance would be much less likely than simply finding a Democratic donating dealer.
Ask the dealers who are highly profitable and were closed anywayYou mean the dealers who say they were highly profitable. And when mass layoffs happen most of the workers laid off will tell you they were doing great jobs and contributed more to the company than they were collecting in pay.
Brilliant.Thank you, it's an important distinction. It is a general rule in business that if you want to make all the decisions you have to keep yourself solvent. If you fail then other people will start making decisions for you whether they be angry creditors, shareholders attempting to buy out the company, bankruptcy judges, or 'angels' who bailout the company. Like I said, let us know if Obama wants to take a controlling interest in functioning companies like Microsoft, Ford, Wal-Mart etc.
"We know nothing of the sort, in fact, the evidence seems to point in the opposite direction."
Even this author's numbers, insufficient as they are, show that the percentage of GOP donors on the "closed" list is higher than the percentage of GOP donors overall.
"Dealerships were not choosen by 'chance' but by strategy. Clearly they want to reduce the number of dealerships in an area so that the ones that remain have less competition."
Try to understand what I am saying. We are talking about six DNC-supporter-owned dealerships in six different areas. All are being allowed to stay open. In every area in which they operate, their competitors are being forced out of business. If you don't see anything fishy here, your eyes are sewn shut.
"The only baseline numbers we have is an estimate of 88% of all donations going to Republicans and 12% going to Democrats by dealers. We have no numbers telling us that when dealers do donate to a party, how often they donate to the top of the ticket versus candidates below the top of the ticket."
I wasn't talking about top of the ticket vs lower. I was talking about Obama's Democrat rivals for the presidency.
"You mean the dealers who say they were highly profitable."
One of them is ranked in the top 2% nationwide. And I love how you leap to dismiss every profitable dealer that is being closed as a liar. Good show.
"Thank you, it's an important distinction."
Don't thank me. When you make a patently contradictory statement, and I say "Brilliant.", I am insulting you. I would think anyone with a functioning brain would have picked up on this.
@gregq:
Your hypergeometric distribution assumes all dealers donate money, which is a pretty poor assumption to make. If we assume that:
1) There are 3,000 dealers
2) Only 10% of dealers make donations3) A donating dealer is 88% likely to be a Republican
then out of 300 donating dealers (DDs), 36 are Democrats, and 264 are Republicans. If we further assume (and this is kind of a reach) that:
4) A DD is just as likely to be closed as a non-DD
then 25% of 300, or 75, DDs were closed. Our final assumption:
5) 92% of DDs were Republican
then 69 Republican DDs, and 6 Democratic DDs, were closed. Under these assumptions, phyper(6,36,264,75) gives us a probability of 15.2%, or not statistically significant.
If we change Assumption (2) to be 20%, then phyper(12, 72, 528, 150) yields 5.15%, and still does not satisfy the 95% confidence level.
Even this author's numbers, insufficient as they are, show that the percentage of GOP donors on the "closed" list is higher than the percentage of GOP donors overall.Not in any statistically relevant way.
Try to understand what I am saying. We are talking about six DNC-supporter-owned dealerships in six different areas. All are being allowed to stay open. In every area in which they operate, their competitors are being forced out of business. If you don't see anything fishy here, your eyes are sewn shut.In other words one person owns six dealerships and Chrysler eliminated competition against that one man? Two rather obvious things stand out:
1. The owner must be pretty good at running a dealership if he has six of them.
2. Chrysler would rather deal with one person to cover six dealerships (esp. one good person) rather than six different people who may have varying levels of quality.
I wasn't talking about top of the ticket vs lower. I was talking about Obama's Democrat rivals for the presidency.the only problem is that the original author not only counted Democratic donars to Obama rivals but also donars who simply donated to Democrats running for Senate and other positions. In other words, Obama supposedly went on a crazy jihad against not only Republicans but also fellow democrats who supported his rivals BUT ALSO a crazy jihad against Democrats whose only crime was not giving him money.
Your statement though exposes you as the partisan hack that you are. I know you weren't talking about the top of the ticket versus the lower. If you cared about the truth you would because it is very important. If dealers put most of their money against lower races, then you would almost certainly expect to find McCain donars on the list but an Obama donar would be a very rare bird.
Why? Say they split 90-10 in favor of lower races. Well if the split between Republicans and Democrats is 90-10 then for every 100 donars you're going to see 90 GOP donations and 10 Democrats. But if the top ticket split is 90% you're going to see 9 McCain donations but only 1 Obama.
NOW I didn't talk about dealers but donars. How many dealers donate to begin with? Certainly not all of them. So to get 100 donars you're going to have to choose more than 100 dealers. For example, if only 50% of dealers donate you need 200 dealers before you can expect 100 donars and one Obama donar. If only 25% donate you'll need 400 dealers.
One of them is ranked in the top 2% nationwide. And I love how you leap to dismiss every profitable dealer that is being closed as a liar. Good show.If he is such a great dealer then wouldn't other companies like Honda, Ford, and Toyota be itching to sign him up? And what does this mean 'ranked in the top 2%'? Ranked by who? Using what criteria? Where is this list? Is it open to public inspection? Statements like yours are automatically suspect because anyone with even a little knowledge of the field knows that there are multiple metrics to measure the quality of a business.
Don't thank me. When you make a patently contradictory statement, and I say "Brilliant.", I am insulting you. I would think anyone with a functioning brain would have picked up on this.Your insult is brilliant.
republicans are worried about car dealerships, but they did not give a damn about the abuses in the Justice Department when a litmus test was in place and any liberal, Democrat was not considered for state AG based on that alone.
grasping at straws.
JeffP
Very right, we don't know how many dealers donate money. If the portion is very low (like 10%) then even among 800 dealers in the closed list you're going to get only a few doners and since most doners go GOP seeing only one Obama donation (or even zero) says basically nothing.
I would take it a step further and propose that below the top donations be ignored. A dealer may donate to local races simply for practical purposes (such as getting support from the mayor in a fight with the zoning board) that have nothing to do with party. I would ask:
1. Out of 100 random dealers, how many donate to national races?
2. Out of 100 random NEW CAR dealers, how many donate to national races?
3. Out of 100 NEW CAR dealers WHO DO DONATE, what is the split between GOP, Democratic and both (and 3rd party donations)?
We don't have these numbers but to even begin to show any statistical significance we are going to have to assume #1 and #2 is very, very high.
"Not in any statistically relevant way."
Oh, I see. So, far from the evidence pointing in the opposite direction, it does show that they are disproportionately GOP donors, but not in a way you consider to be statistically relevant. Thanks for clarifying.
"In other words one person owns six dealerships and Chrysler eliminated competition against that one man?"
Yes. Not a single one of this DNC-supporter's six dealerships are being closed, and all of his competitors in all six areas are being closed.
"1. The owner must be pretty good at running a dealership if he has six of them."
What a reach. You can be just as good at running a dealership if you only have one.
"2. Chrysler would rather deal with one person to cover six dealerships (esp. one good person) rather than six different people who may have varying levels of quality."
I would think that, if political considerations by Obama's "automotive task force" weren't involved, Chrysler would rather make the most money possible. Unless it can be demonstrated that every single one of this DNC-supporter's six dealerships in different areas was legitimately superior to every single competitor, and every single competitor was performing badly enough to warrant forcibly putting them out of business, it reeks of foul play. You clearly have a clothespin on your nose, but I don't.
"the only problem is that the original author not only counted Democratic donars to Obama rivals but also donars who simply donated to Democrats running for Senate and other positions. In other words, Obama supposedly went on a crazy jihad against not only Republicans but also fellow democrats who supported his rivals BUT ALSO a crazy jihad against Democrats whose only crime was not giving him money."
Or they were planning on closing a certain number of dealerships already, but his donors received preferential treatment under pressure from his "automotive task force" while GOP donors and even in some cases DNC supporters who didn't donate to him were left to fend for themselves or were even shut down regardless of the fact that they are highly successful. And it's spelled "donor".
"If he is such a great dealer then wouldn't other companies like Honda, Ford, and Toyota be itching to sign him up?"
I don't normally humor speculative attempts at rebuttal, but I will say that this guy operates in Long Island and the other manufacturers already have very successful dealerships there.
"And what does this mean 'ranked in the top 2%'? Ranked by who? Using what criteria? Where is this list? Is it open to public inspection? Statements like yours are automatically suspect because anyone with even a little knowledge of the field knows that there are multiple metrics to measure the quality of a business."
The talking heads from Chrysler have had ample opportunity (and have been asked directly about this particular dealer) to object to his statement. They don't.
"Your insult is brilliant."
I assumed we were both out of the second grade and therefore am surprised you don't realize that "I'm rubber and you're glue" isn't quite the zinger it used to be.
"republicans are worried about car dealerships, but they did not give a damn about the abuses in the Justice Department"
Give me a break. US Attorneys serve at the will of the President. End of story. And I found the uproar over that very odd considering that Clinton fired nearly all of the US Attorneys in place when he took office - when one of them was investigating him at the time. But to the point at hand - comparing the firings of political appointees by the person authorized to fire them with the forcible closing of privately owned businesses involving a Marxist regime is asinine.
@useyourbrain: "I don't have to do the analysis to point out that it is sorely lacking in this article."
Can we go ahead and declare this the "GOP Godwin" statement?
What I said is a simple fact. I don't have to do the statistical analysis to point out that it is lacking in this article.
@UYB:
I agree with you insofar as Nate's initial analysis is not rigorous and is back-of-the-envelope. However, most of your points have been anecdotal, making them subject to selection bias and statistically useless. Also, the more rigorous analysis that I applied above based on Nate's assumptions fails to reject the null hypothesis that there is no favoritism in which dealerships are being closed.
...And it doesn't help that your writing style is disjointed, abrupt, and abrasive, and makes a LOT of qualitative assumptions about the President that are dubious.
UYB: I would very interested to hear if you were in up in arms when Samuel Alito was nominated for the same reasons you are annoyed with Sotomayor. If not, I would also like you to explain why Alito's comments are not offensive, but Sotomayor's are (excuse the length).
---
U.S. Senate Judiciary Committee Hearing on Judge Samuel Alito's Nomination to the Supreme Court
U.S. SENATOR TOM COBURN (R-OK): Can you comment just about Sam Alito, and what he cares about, and let us see a little bit of your heart and what's important to you in life?
ALITO: Senator, I tried to in my opening statement, I tried to provide a little picture of who I am as a human being and how my background and my experiences have shaped me and brought me to this point.
ALITO: I don't come from an affluent background or a privileged background. My parents were both quite poor when they were growing up.
And I know about their experiences and I didn't experience those things. I don't take credit for anything that they did or anything that they overcame.
But I think that children learn a lot from their parents and they learn from what the parents say. But I think they learn a lot more from what the parents do and from what they take from the stories of their parents lives.
And that's why I went into that in my opening statement. Because when a case comes before me involving, let's say, someone who is an immigrant -- and we get an awful lot of immigration cases and naturalization cases -- I can't help but think of my own ancestors, because it wasn't that long ago when they were in that position.
And so it's my job to apply the law. It's not my job to change the law or to bend the law to achieve any result.
But when I look at those cases, I have to say to myself, and I do say to myself, "You know, this could be your grandfather, this could be your grandmother. They were not citizens at one time, and they were people who came to this country."
When I have cases involving children, I can't help but think of my own children and think about my children being treated in the way that children may be treated in the case that's before me.
And that goes down the line. When I get a case about discrimination, I have to think about people in my own family who suffered discrimination because of their ethnic background or because of religion or because of gender. And I do take that into account. When I have a case involving someone who's been subjected to discrimination because of disability, I have to think of people who I've known and admire very greatly who've had disabilities, and I've watched them struggle to overcome the barriers that society puts up often just because it doesn't think of what it's doing -- the barriers that it puts up to them.
So those are some of the experiences that have shaped me as a person.
many of the donors don't ist their occupation, or don't list car dealer as their first occupation. For example, Robert Johnson, founder of BET (and car dealership owner) mentioned above.
[url]http://www.opensecrets.org/indivs/search.php?capcode=bhb5g&name=johnson,%20Robert&employ=&cand=&state=CT&zip=&all=N&old=N&c2008=Y&c2006=Y&c2010=Y&sort=N&page=&page=1[/url]
That's a shit ton of money there, looks like just over 140K from him alone to different DNC causes, and a maxed out contribution to Obama.
(some aren't him on that page, but most are), There are several different occupation terms, none of which are "car dealer" "auto dealer" or any term included in half-ass "control group."
Convenient how he's left out, along with the ~141K he brought to the table.
Just including that one guy changes 88% to 80, so even using that flawed metric, you'd see a statically significant problem. IF 92% of dealerships closed are GOP donors, and only 80% are GOP donors overall thats plenty of evidence of a political hitjob. Just by adding one guy.
"...And it doesn't help that your writing style is disjointed, abrupt, and abrasive,"
In the eye of the beholder, I suppose.
"and makes a LOT of qualitative assumptions about the President that are dubious."
Such as?
"If not, I would also like you to explain why Alito's comments are not offensive, but Sotomayor's are (excuse the length)."
Because Alito actually said what those on the left are trying to pretend Sotomayor said - that his background shapes his view. He did not say that someone of one race and sex presents the possibility of better judgment than someone of the opposing race and sex. He spoke for himself and that was that. His statement was not racist and sexist. Understand?
Oh, I see. So, far from the evidence pointing in the opposite direction, it does show that they are
disproportionately GOP donors,In order to say something is disproportionate one must know the proportions in the population. We do not
know the proportions but we do know for you to even approach a level of disproportionate you need rather
implausible proportions (such as almost all dealers make political donations and almost all make
donations to national rather than local races).
6 dealerships:
What a reach. You can be just as good at running a dealership if you only have one.You *could* be just as good, but the man with 6 working dealerships is *almost certainly* good at running
dealerships. What you are missing is the pretty obvious fact that the man
with 6 almost certainly has a proven track record.
What a reach. You can be just as good at running a dealership if you only have one....Sounds like a lot of speculation backed up by nothing. While we don't know the necessary proportions, a
good eyeball estimate would say the number of Obama doners that it would be sensible to see in a random
sample of 800 would probably be in the single digits. So try to find some evidence that there was
specific lobbying for certain dealers to stay off the list by the task force. So far you've only
presented a statement from a lawyer saying closing dealerships was desired by the task force (duh,
everyone who knows anything about the industry says taht).
I don't normally humor speculative attempts at rebuttal, but I will say that this guy operates in Long
Island and the other manufacturers already have very successful dealerships there.:
You mean this guy in the top 2% is surrounded by other top 2% guys? What are the odds that so many top
people end up within spitting distance of one another!
The talking heads from Chrysler have had ample opportunityMost likely they know that to do so would open the door up to debating with every dealer on the closed
list who, no doubt, will have a half dozen reasons why *they* shouldn't be on the list. When companies
do mass layoffs do they engage in protracted debates with canned employees?
c5367ss
many of the donors don't ist their occupation, or don't list car dealer as their first occupation. For
example, Robert Johnson, founder of BET (and car dealership owner) mentioned above.This is a good example of the problem of autocorrelation. On the one hand, Johnson is a big Democratic
donar. On the other hand, Johnson clearly has an excellent knack for business and knows how to market to
the black community. On the other other hand, Johnson is involved with lots of dealerships other than
Chrysler. So imagine the decision to keep him off or put him on the list. Put him on the list and you
not only could loose a star dealer but could end up putting him firmly in the competition's camp. His
media connections and high influence with a niche market could result in him engaging in a market war
against whoever you decide to leave as the Chrysler dealer in his areas. Even if some of the people who
lost out to him are technically 'better' dealers, there's a strong business argument to choose to keep
him that has nothing to do with his donations.
The critics here are making a lot of amazing assumptions. One of them is that there is no correlation
between unbiased traits of dealerships that were kept (such as high market share in a local African
American community) and their political donations. This quite frankly is not the case of something that can be effectively answered by statistical analysis and trying to do so is simply lazy. If you want to dig into the list go ahead and dig and see who put it together and at what level did the auto task force review it.
Please correct me if I'm wrong - I have a psychologist's understanding of statistics (not good...)
If you are comparing the # of dealerships closed, wouldn't it be more valuable to compare the # of those dealerships who leaned democrat and the number who leaned republican than the total $ from republicans/democrats?
Based on the control group above, I believe that puts the numbers something like this:
Republican donors = 836 (86%)
Democrat donors = 136 (14%)
This, again, is rough, but it seems to me more accurate than the fairly random $ figure. Robert Johnson brings the $ figure under a great deal of speculation with the way that he immediately dominates the ratio.
Further, the numbers above count individuals, not dealerships as a whole. Seeing the way that dealerships/owners tend would be more informative than seeing the way that employees tend (in my mind, Auto Dealer/Car Dealer almost certainly refers to a sales rep, not an owner.)
Finally it should be brought up that Fundrace cuts off donors around $200. So, if we're counting employees of dealerships or the amount of $ from employees of dealerships, it should be noted that Obama had more donors in 2008 at a lower average $ amount. This could mean that donors of <$200 could swing toward Obama, emphasizing the importance of weighting the dealers/owners differently than individuals.
My point is that while the control group Nate put together is interesting, it certainly doesn't put the "conspiracy theory" to rest. If people are willing to put time into crunching the numbers, why not let them do so and then judge their methods and findings afterward?
"In order to say something is disproportionate one must know the proportions in the population."
Going by the author's numbers, which I agree are lacking, the GOP donors are disproportionately represented on the "closed" list.
"You *could* be just as good, but the man with 6 working dealerships is *almost certainly* good"
Sure - but out of the six different areas in which he operates, it is almost certain that at least one or more of his many competitors operates a comparable or better dealership than the one he runs in that area. To assume that none of the other dealerships in any of the areas in which he operates deserved to stay in business is wishful thinking on the part of a DNC apologist.
"Sounds like a lot of speculation backed up by nothing."
No, it's a fact that you can be just as good at running a dealership if you own one as a guy who owns more than one. If you take a given dealer that owns one dealership and compare him to the guy that owns six, is it probable that the guy who owns six is better? Sure. If you take dozens of dealers in six different areas and compare them to the same guy, is it probable the he is better than every single one of them in every single area? No.
"You mean this guy in the top 2% is surrounded by other top 2% guys? What are the odds that so many top people end up within spitting distance of one another!"
I have no idea what you are talking about here.
"Most likely they know that to do so would open the door up to debating with every dealer on the closed list who, no doubt, will have a half dozen reasons why *they* shouldn't be on the list. When companies do mass layoffs do they engage in protracted debates with canned employees?"
If an employee who is being laid off enters his boss's office, points out that he is performing in the top 2% nationwide and wildly outperforming many of the people who are keeping their jobs, and asks why he is being laid off - yes, the company will give him an answer, unless they are a piece of shit company with horrible management or they have ulterior motives.
"This is a good example of the problem of autocorrelation."
Not at all. It simply smells rotten when you have a major DNC guy getting to keep all of his dealerships in every single area under the watchful eye of the Democrat White House's "automotive task force" while all of his competitors in every single area are being put out of business without a firm answer as to why. It warrants further investigation, as much as you may dislike this fact.
And please, next time you compose your post in notepad, turn off word wrap.
@TJ
The point is the author of the original theory should have done that. Is it sound to make claims on limited information and then say "wait while I see if it's true."
"automobile dealer" 138 Rep 20 Dem total 158
"automovitve dealer" 11 Rep 2 Dem total 13
Total contributors: 172
According to http://www.usatoday.com/money/autos/2008-10-20-auto-dealerships-credit-crisis-loans_N.htm, there are about 20,000 new car dealerships in the US
What does this mean? It means only a tiny percentage of dealerships contributed anything to the 2008 race. Even 1% would be 200 dealerships and that wouldn't take into account that some of those 172 people may work at the same dealership....
Take 800 closed dealerships. Say 10% of dealerships contributed anything in 2008. That's only 80 contributors. 90% contribute to Rep, 10% to Dems. That's 8 expected democratic donors and 72 expected Republican donors. Say of people who donate, 50% donate only to local or regional races and 50% donate to national ones. That would be only 4 Obama donors one would expect to find.
Of course, the one dealer is just found 'so far'. I'm not convinced at all that there aren't other dealers who have donations to Obama and got closed, and while we don't have an estimate for how many donate to the top of the race we do have one for who donates to Dems in general and a few have been found.
This story is now totally discredited unless someone has NEW data.
If I didn't want to be generous and I used 1% as the portion of dealers who donate 800 closed dealers would result in 8 donors and 10% as Democratic donors would yield an expected value of 0.8 Obama donors!
Sorry messed up on the math
"auto dealer" 546 Rep 72 Dem total 618
"car dealer" 141 Rep 42 Dem total 183
"automobile dealer" 138 Rep 20 Dem total 158
"automovitve dealer" 11 Rep 2 Dem total 13
Total contributors: 972 / 20,000 = 0.0.486 or 5%
5% * 800 dealers = 40 who made donations * 10% Dem donations = 4 Dems
So we expect to find less than 5 Democratic donors and that includes not only Obama donors but also any type of Democratic donor whether you're talking about a rival of Obama like Hillary or a Senate race or whatnot.
Jeff P,
You are correct, I am assuming that essentially all dealership owners are donors. I am also assuming that all Democrat donors are created equal (a generous assumption for Obama in this case).
If you'd like to provide some real numbers for how many of the dealers are donors, I'll be happy to add them to my calculations. Lacking that, we can't make any calulations, which makes Nate's post even more dishonest than it originally appeared.
because someone who knows as much stats as he does knows that 88% v. 92% is an absolutely meaningless statement when you don't know the size of the pool.
Note that at 25% donors, we're talking ~3%, which does pass a 95% confidence interval test.
In short, Nate's blither dismissal was utterly unsupported by the facts, and anyone who wants honest government wants this examined.
Partisan hacks, of course, want this swept under the rug, because it might show Obama in a bad light.
@Troy
I'm with you. Doug Ross (who I believe is the originator of "Dealergate") should have done the hard math before trying to start a media frenzy. Reporting his results without some semblance of statistical rigor is disconcerting at best.
Still, Nate has also made a claim based on limited and faulty information. "There's no conspiracy here, folks -- just some bad math," sounds pretty final to me. At least Doug Ross is still doing the math and we have hope he will flesh out his work and allow it to be properly peer-reviewed. Nate has already drawn his conclusion based on curious premises and put the matter to rest. Whose process requires the greater revision?
@Boonton
Are the 10% donors/1% donor figures you're using from Fundrace? Who says that donors met the criteria to be included there? Who says they all filled in their job title as one of the four searched? Saying that the theory is, "totally discredited" based on your quick math seems a bit strong.
TJ
A fair point but it doesn't do much to help your argument. Say the site missed half those who gave. Double the contribution rate to 10% and you get 8 expected Dem dealers. You're not even breaking into double digits here!
And while fundrace may miss some who put another job title it is also overcounting cases where two or more dealers work at one site and make contributions.
2008 was a bad year economically and the car companies have been in bad shape for a while. What % of dealers is it reasonable to expect to have been making donations to political campaigns? If you used the general population I can tell you you're not breaking 10%
@TJ
But what about the fact that a statistical difference, if discovered, doesn't result in any of the causation claimed?
A statistical difference found doesn't mean you get to assume why the difference exists.
I honestly think that, given my large assumptions about the locations of American dealerships, that you would have more "red" dealers in Chrysler than overall (more "red" dealers in domestic vs all dealers), and that you would probably want to close more "red area" dealers than "blue area" dealers if closing 25%.
My assumptions that result in the above are based upon an idea that it is likely that such "red areas", such as a town of 250k in Texas, might average a higher amount of Chrysler dealers than an equal 250k area in Oregon, Washington, etc. That higher density results in a higher % likelihood of closure, and if those owners are more likely to be red than the overall dealership owners of the US, you should find the small subset that donated would be incredibly Repub.
Basically, that is just 1 of 1,000 possible reasons that one could surmise as a better explanation than WH interference in Chrysler, even if you found a discrepency between 2 dealership groups.
For a conspiracy theory to make sense, usually it has to make sense that the conspirators would want to do what they are claimed to do.
The WH gets little to no benefit from the "blue" dealers, as evidenced by their lack of donations. So going to bat for them, and risking all the political capital they have for their benefit, makes little sense.
Also, the red dealers, while they do donate, also do not seem to be a largely important group, certainly not one that would be worthy of this effort.
At least conspiracy theories about Obama being born in Africa give Obama a reason for his actions, lying about his birth allows him to run for president. That makes sense, even if the evidence is terrible that he actually did it.
In this new theory, Obama goes out of his way, risks lots, to target a bad threat, to give benefits to friends that provide none.
No Chicago politician would be that dumb at politics.
---
In the meantime, both Chrysler and the WH agree that Chrysler decided the individual dealers to cut...
I randomly selected 19 dealers out of 789 closed dealers and found 4 (22.2%) donated to the GOP and 1 (5.7% ) donated to Democrats.
I compared this to a random sample of 32 of 2392 open dealers and found that 4 (12.5%) donated to the GOP and 0 donated to Democrats.
Are any of these samples large enough to be statistically representative of the populations they are drawn from? Is the difference in the outcomes significant?
I've already shown by a preponderance of evidence that the dealerships chosen for closure were selected by Chrysler, not Obama, so all this seems moot to me.
I also just found out that the chairman of the new Chrysler Group was a member of G.W. Bush’s election organization and has consistently contributed to Republicans running for Congress and the White House.
Can anybody who understands statistics comment on my study? I wrote up more details in previous posts
Please note where data come from Huffington Post?! or non biased source?
Old Russian saying...You can tell same lie 1000 time but not change truth!
Difference between USSR Communist media and USA "mainstream media"
In Russia government make media say what they want - even if lie.
In USA "mainstream media" try make government what they want - even if lie..
.....eventually they become same thing?!
I Igor produce Obama Birth Certificate at www.igormaro.org
What total nonsense.
The statistics mean nothing.
Even if the % of closures for each side innocently falls EXACTLY along the lines of % of each side of donors, this neither proves NOR disproves any bias. Even if the closures fall 100% on Republican donors, this neither proves NOR disproves any bias.
All this does is indicate a statistical likely hood for bias based on an expected statistical probability. Just like a xerox of a xerox, a likelyhood of a probability has degraded value. There are too many other variables to consider, including the distribution of these dealers across communities. Dealerships are not balls in an urn.What HAS value is an honest disclosure, under oath if necessary, of the criteria used by those who made the decisions.
Even if the question only arose because of what turns out to be tinfoil hat lunacy, this does NOT invalidate the value of the question itself:
"Is good criteria being used in a decision which affects a taxpayer investment?"Why are some so reluctant to seek an answer to this question?
Wait a minute, UseYourBrain, you mean Obama had a scary black preacher? Tell me more.
About the article: I'm not sure I follow all the methodological stuff, but given the sample size, a discrepancy of a couple percent doesn't seem to indicate hostile intent. If it is like 99% as opposed to 91% (out of a population of 88%) that smells a little iffy to me, but in the latter case, hell, just because you donate to the Republicans doesn't mean you're automatically some kind of blessed job creator.
I'm out of town this weekend, so my responses will be brief and incomplete:
@gregq:
If you'd like to provide some real numbers for how many of the dealers are donors, I'll be happy to add them to my calculations.Would love to, but I'm not confident in my ability to obtain a reliable data set. I wouldn't chalk up Nate's conclusion to dishonesty, though -- I'd say he jumped to the conclusion too hastily. It's a mistake, not a lie or deception.
@TJ:
You're correct about proportions of donors being more useful than the sum of the donations, but given that they're fairly close, I think it's a reasonable proxy.
@UYB:
Going by the author's numbers, which I agree are lacking, the GOP donors are disproportionately represented on the "closed" list.Please re-read my analysis -- it refutes this statement, based on the enumerated assumptions.
@creeksneakers2:
Without even crunching the numbers, I can tell you that those sample sizes are nowhere near large enough.
@Khyris:
Even if the % of closures for each side innocently falls EXACTLY along the lines of % of each side of donors, this neither proves NOR disproves any bias. Even if the closures fall 100% on Republican donors, this neither proves NOR disproves any bias.Well, that's just dumb. The whole point of statistical analysis is to determine whether there's something there. What you're saying is that it's possible to
(1) Have bias and still not affect the results negatively (the "EXACTLY along the lines" scenario), in which case, why does anyone care.
(2) Have the "100% on Republican donors" scenario, which is in fact the very definition of bias: "A tendency or preference towards a particular perspective, ideology or result".
I tried to post this earlier this week but for some reason it didn't show up.
A correction is warranted: Nate's calculation is based on calculating the amount of money donated to Republicans and Democrats. If you're trying to determine the proportion of GOP-donating dealers vs. Democratic-supporting dealers, you should focus on the number of dealers who contributed to each, not on the absolute amounts.
Here are the numbers based on Nate's research: 836 dealers he found gave to the GOP. 136 donated to the Democratic Party. Therefore, 84% (not 88%, as Nate stated) of the dealers cited are Republican supporters. And 16% are supporters of the Democrats.
Please use these numbers in the future when discussing this issue.
yeah, see, use your brain, you didn't read the entire speech even though you insist that you did. because ms. sotomayor was very very clear in the rest of the speech that her experiences would give her a wider breadth of understanding that while important, could not overwhelm the force of law. that is, a pool of experience to draw on, not overwhelm. which is precisely what Alito was saying.
http://www.weeklystandard.com/weblogs/TWSFP/Sotomayor-La%20Raza.pdf (see page 92).
I should hope your myopia would not overwhelm the clear import and context of the statements.
Apologies for being off topic.
UYB,
The difference between your using Judge Sotomayor's quotes out of context, and my out of contexting Limbaugh is that your out of context quoting of Sotomayor is central to your case. My case is that Limbaugh is not showing any regard to those who suffer if 'Obama fails', and thats my point. He is positively gleeful and proud of his statement, but never interjects something along the lines of 'Look if the President fails many many people may lose there work or see the value of there savings decline or have there homes foreclosed. I'm sorry about that.' My point isn't about the exact nature of Limbaugh's quote, its that he is playing politics on his show, not really being interested in the results. (Obviously thats my judgement on Limbaugh, but I do believe it to be true, and if it were not I would expect Limbaugh to be ... well humbler.) The main point being that the right as a whole seems more concerned about playing a game right now than the consequences of policy. Its not so much the party of no, as the party of not interested (in the results). Its all about stopping the Dems and claiming whatever small wins that it can get, not about doing whats best for the people of the country.
(FWIW I would suggest this is not unusual for opposition parties throughout the world.)
Nate was making a simple point:
Why are people discussing a 92/8 split among Chrysler closed dealers, without a figure to compare to? It seems people are comparing 92/8 against nothing, or basically against 50/50 assumptions, and stating something is wrong despite no reason to assume so. The assumption was simply 92/8 seems wrong in and of itself.
So he did a quick and dirty look at some contribution figures, to help show people the original assumption is silly (not even getting into its impossibility due to the fact that the WH did not decide on individual dealers), and showed a 84/16 split among all dealers in the lists he grabbed, which was a 88/12 money split.
None of these figures can actually be directly compared, the 84/16//88/12 figures from his pulling of donation data are for ALL dealers, under 4 name types. This cannot be compared with closed dealers of the company Chrysler directly.
All it shows is that we should expect the split to be slanted, and thus the original anger at the split was misplaced.
Add in the fact that this would require a conspiracy by the WH and Chrysler to both lie about who made the decisions, add in the lack of any evidence from people trying to compare the open and closed Chrysler lists, and add in the lack of causality a statistically relevant difference would show, and you see why he was so dismissive.
Even if you found a disparity between closed and open Chrysler dealers, you still are very far away from being able to even reasonably guess that the WH was behind the difference.
No one is even that far, and so this ranks currently at the level of conspiracy theory required to believe a Kenyan native fake a birth certificate to become the POTUSA.
No evidence of a statistical difference between closed and open Chrysler, only anecdotal such as Dem dealer chains surviving in some areas.
No evidence to explain the host of alternative explanations for a statistically releveant difference, if there was a statistical difference.
No evidence that both Chrysler and the WH are lying, and the WH did decide the dealer closings through their auto task force.
No rational explanation on why the WH would even do this if they could, as the numbers show the Dem dealers are not contributing friends worth protecting.
--
This issue is a non-starter from the get go, and has not progressed beyond that.
This person called "useyourbrain" is spouting long-disproven talking points from campaign season. This person needs to be banned from commenting here.
Here are the numbers based on Nate's research: 836 dealers he found gave to the GOP. 136 donated to the Democratic Party. Therefore, 84% (not 88%, as Nate stated) of the dealers cited are Republican supporters. And 16% are supporters of the Democrats.The problem, though, comes in how many dealers don't donate at all. It seems like 95% or so do not donate and the number might even be higher since we do not have a good handle on how many of those 972 who did donate represent different dealerships versus people working/owning the same dealership. Given 20,000 dealerships in the country only a tiny minority make political donations (at least in 2008)
So in the closed list you're not going to see 84% donating to Republicans you're going to see 95% not donating at all. Only among the 5% that do, will there be a split between Republicans and Democrats and we are talking single digit numbers in terms of expected Democratic donations.
Again 800 dealers * 5% who donate = 40
40 who donated * 16% = 6.4 Democratic dealers.
So we expect to see 6 or 7 Dems and we see a few so far. One Obama and several who gave to other Dems. This is statistical nonsense. We have exactly what one would expect from a normal sample of 800 dealers that was unbiased in regards to political contributions. If the hypothesis was correct, that Obama slanted the list not only in favor of Democratic contributors but in favor of his own contributors what exactly did he do? Save two or three dealers whose total contribution couldn't be more than a few thousand to begin with?
David complained about my use of Gaussian statistics. It is simple and gives about the right answer. http://en.wikipedia.org/wiki/Hypergeometric_distribution even points to the Gaussian as a good approximation for n < m, N and p not close to 0 or 1. The variance of the hypergeometric distribution is 23 instead of 28 for the numbers given. My earlier conclusion is unaffected. There is no statistically valid evidence in the overall numbers that the choice of Republicans vs non-Republicans was other than random.
Anyone not familiar with math reading this should understand the following: the important thing is not the probability of an individual result. The important thing is whether you are reasonably close to the expected result.
Here is a more basic analysis.
Car dealers who have been selling cars are staying open. Dealers with marginal sales, high cost to finance their inventory, and pay for their advertising are closing. The notion that Democrat or Republican factors into this is really absurd.
The problem is Red ink.
@Jeff P
No. Wrong. Let me fix this for you:
"What you're saying is that it's possible to
(1) Have bias and still not affect the results negatively to the extent that the bias exists beyond doubt. The lack of completely undeniable interference does not prove the lack of ANY interference.(2) Have the "100% on Republican donors" scenario, which is NOT in fact the very definition of bias because this is merely A RESULT and bias exists in the PROCESS: "A tendency or preference towards a particular perspective, ideology or result"."
This particular result is not inherently exclusive to one biased process, but any number of politically biased or unbiased processes of which we do not know the one that was chosen.
If the closures fell on 100% Republican donors, that RESULT does not prove existence of bias in the PROCESS. The improbability of such a result coming from a truly random process implies a probability of a process bias existing and nothing more. If you flip a coin 100 times and they all come up heads, then the statistical likelihood that the coin has a physical bias progressively increases, but this is not a "proof" of bias. In general, flipping a coin is NOT a biased process.
For there to be a proof (and I use that word in its mathematical sense) of bias in the decision, we need to know the criteria of the decisions and the individuals who made them. The rest of this is just mathematical masturbation by career poll-spinners on both sides that, believe me, real mathematicians find cheap and tiresome.
It is simple and gives about the right answer. http://en.wikipedia.org/wiki/Hypergeometric_distribution even points to the Gaussian as a good approximation for n < m, N and p not close to 0 or 1.I think you're confused; that page mentions neither Gaussian nor Poisson (including its recent history). Anyway, one indication that the Gaussian isn't a good approximation to the Hypergeometric in this case is that they, um, disagree. Since everybody stipulates that the hypergeometric distribution is the most accurate, let's look at the predicted variance for it:
Your calculation: 23 (or maybe 529: you called 23 the variance, but put it alongside a standard-deviation result).
My calculation:
variance = [n(m/N)(1-(m/N))(N-n)]/(N-1)
substituting:
N (total) = 3500
n (drawn) = 875
m (defective a.k.a. Republican) = 0.88 * 3500 = 3080
we get: [875(3080/3500)(1-(3080/3500))(3500-875)]/(3500-1)
= 69.31980565875963
The standard deviation is the square root of that, or about 8.3258516 .
So closing 875 dealers and having 92% of them be Republican is ((.92 - .88) * 875)/8.3258516 = 4.20 standard deviations from what you'd expect. That's a lot of standard deviations.
A more direct estimate of the chances of a sample of 875 (out of a population of 3500 things 88% of which are R) coming up 92% or more R is about 1 in 100,000. (For comparison, you can get >= 91% about 1 in 1818 times, >= 90% about 1 in 57 times, and >= 89% better than 1 in 5 times.)
1 in 5 should generate a shrug, 1 in 57 has a p of under 0.02 so it might get you published in a social-science journal, but 1 in 1,818 is past the point where a reasonable society starts issuing subpoenas.
With government “buying” into an industry causes one to perceive political influences. For an administration that pledged transparency…it sure is lacking on all fronts.
IMHO I am trying to look at this with a common sense approach. First of all, auto dealers have made a huge investment in their stores. Aside from the infrastructure, the cost of purchasing the franchise rights, vehicle stock and parts, and the general overhead most dealers have community ties.
Secondly, the dealers purchase the cars from the manufacturer on credit and pay interest for them to sit on their lots. The manufacturer has received their money, the vehicles are not on consignment. There is an incentive paid by the manufacturer, for dealers to sell more vehicles, however, it is already figured in the cost of the vehicle to the dealer. In other words, the dealer can recoup a percentage of his cost of the vehicle if certain factors are met.
Third, people are in business to make money and turn a profit. If a dealership does not turn a profit it is usually sold. How many times have you seen names change on car dealerships? ( A) they are either in financial trouble and sell outright, B) they are owned by a huge auto group and “lend” the name to a local sports figure (who happens to be at the top of his/her game) and will change names when the celebrity falls from grace, or C) another family member takes over the store from the group.) Now who is to say what that “profit” should be? Not all dealers can be “mega” dealers, as long as the expenses are met, the lights are still on, and people are getting paid, and the dealership is turning a profit…who cares how much they earn? It is really any manufacturers business how much money a dealer makes? As long as their obligations are met, who should care?
Many questions have to be asked. Why are dealerships being terminated that are in the upper performance brackets, as rated by Chrysler? Why hasn’t clear cut guidelines been established and published to cull the dealerships? Why hasn’t Chrysler set up a program to assist in the redistribution of the vehicles/parts in inventory of the terminated dealers? Since, every dealer has a numerical rating as to performance, why wasn’t the lowest dealerships cut? Since the dealerships themselves have not been informed why they were cut, this alone shows the lack of transparency.
In one article in the paper, there was a quote (sorry I can not find it right now) however, it alluded to the fact, that reducing dealerships would in fact generate more income for the manufacturer and selected dealers. In other words, reduce the competition among dealerships and raise prices. No longer will consumers have a viable option of “shopping” for the best deal.
Perception clouds transparency.
So closing 875 dealers and having 92% of them be Republican is ((.92 - .88) * 875)/8.3258516 = 4.20 standard deviations from what you'd expect. That's a lot of standard deviations.
But 92% of them are not Republicans (or at least not making donations to Republicans). It's 92% of THOSE WHO MADE DONATIONS! Most dealers, closed or not, didn't make donations to anyone in 2008
your post is helpful and informative
web design India
キャッシング
網頁設計
クレジットカード 現金化
ダンボール
留学
水 通販
ウィークリーマンション
障害者
有料老人ホーム
看板 製作
レーザー脱毛
フランチャイズ
I recently came accross your blog and have been reading along. I thought I would leave my first comment. I dont know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.
Patricia
http://forextradin-g.net
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店經紀,
酒店打工經紀,
制服酒店工作,
專業酒店經紀,
合法酒店經紀,
酒店暑假打工,
酒店寒假打工,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店工作,
酒店打工經紀,
制服酒店經紀,
專業酒店經紀,
合法酒店經紀,
酒店暑假打工,
酒店寒假打工,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店工作,
酒店打工經紀,
制服酒店經紀,
酒店經紀,
菲
梵,
From cell phones users to see the specific situation of occupational segmentation in 2009, accounting for 19.5% of students dropped 21.2 percent over last year, other types of occupations than those last year, the proportion of Internet users cheap cell phones increase. White collar crowd from last year's 29.2% increase to 38.9% this year, accounting for 9.7 percentage points up to replace the student groups cellphone users as one of the biggest occupational hierarchy; blue-collar crowd from last year's 13.9% to 18.9% this year, accounting for rose by 5.0 percentage points, showing that mobile phones users by a group of students to the occupational groups a significant trend in the development. Ereli advice that, cheap cell phones and mobile phone users Internet users monthly income distribution of age, education, occupational distribution has strong correlation with high spending capacity of white-collar workers and some students in the crowd will be a huge cell phone china online potential consumer groups.
By comparing the traditional Internet users, Internet users to iResearch found that the traditional white-collar-based, cell phones wholesale, corporate general staff accounted for 18.9%, higher than the 5.6% of the wholesale cell phones users accounting; and discount cell phones users in the years students and blue-collar workers accounted for significantly more than the traditional Internet users, respectively, accounting for 19.5% and 18.9%, higher than the traditional Internet users Students and blue-collar workers accounted for 7.8% and 5.1% respectively.
The survey found that consumer 3G wholesale china from the crowd of view, the buyer 25 to 40 years old mainly white-collar workers, accounting for about 40%, followed by consumer groups of students, accounting for about three into. According to statistics, 3G wholesale products in sales, compared with a 2G mobile phone sales are still a wide gap between, but since June has been, 3G mobile phones increase in the average monthly buy products for more than 50%, "11" period due to holiday business, the increase of more than 150%. Pk that the "11" after the peak sales of 3G handsets likely to usher in more stable growth.
Post a Comment