So, here is the link to a paper which Keiths claims says something about income inequality, and I say is another example of the proliferation of shoddy science.
http://pss.sagepub.com/content/early/2015/03/19/0956797614567511.abstract
The highlight of the paper is this claim:
“We found that of the 40 search terms used more frequently in states with greater income inequality, more than 70% were classified as referring to status goods (e.g., designer brands, expensive jewelry, and luxury clothing). In contrast, 0% of the 40 search terms used more frequently in states with less income inequality were classified as referring to status goods.”
Where does one begin to critique the ridiculousness of this claim? 70% of the majority of searches are for luxury goods in some states, 0% of the most searched items in other states?
If one claims the difference in search patterns from one state to another is that dramatic, shouldn’t ones bs detector already be ringing alarm bells?
And what is considered a luxury good? What is the cut-off for equal states and unequal states? Did they decide the luxury terms before or after they viewed the data? Who do they claim is doing all this searching for luxury, the haves or the have nots?
The red flags are everywhere. Isn’t it likely that they had a conclusion that they wished to reach, and that they fulfilled their own prophecy?
New York has roughly 6 times the population of Mississippi. NY is 16th in per capita income, while Mississippi is 50th. NY is 88% urban, while Mississippi is 49%. NY is 22% foreign born, while Mississippi is 2%, a factor of ten difference. Yet after controlling for these significant differences, the two states have similarly high income inequalities and people in those states have searched for “Ralph Lauren” or “Yurman rings” approximately the same number of times, which is more than other states which may have similar demographics but less income inequality. That’s the point of the paper’s findings. The common denominator between states where people search for “Yurman” most frequently is income inequality, even when there are radical differences in population, income, lifestyle, etc.
I haven’t read the paper, and I don’t really care what phoodoo thinks, but can anyone explain the point of the paper. I don’t mean explain what it says, but rather explain why it would be considered important.
You are free to not consider it important. And I also do not consider it important.
But so what. I’m sure that there are things that I see as important that just bring yawns to others.
For those who don’t know (Phoodoo) regression tries to decompose a thing into constituent parts. By doing this you can isolate individual effects.
There are things about inequality that I consider important, but income taken in isolation is not one of them. I lean conservative, but not to the extent that I would deny all welfare and medical benefits to the poor and unemployed.
What I question is whether income disparity per se is important.
Hobbes,
Well, yes, they are saying that they control for these differences, what is the control? Do more foreign people make the income inequality higher or lower? Does rural life make it higher or lower? Why?
So, do they give a listing of the states rankings or not?
The state of Hawaii has a very low Gini co-effcient (its more equal) , but I am guessing many people there are foreign born. Does that make them more equal or less? Do people in Hawaii look at luxury goods less, I doubt it.
Furthermore, is a search about horses a luxury search? I guess it depends where you live. What about a search for diamond cutting? Pool cleaning supplies? Does the paper give a list of the most common terms?
Will you be writing a rebuttal and submitting it to the publishers of the paper, out of interest? If not, why not?
Neil Rickert,
And yet you reply here?
Its important because MOST published studies are bullshit. This looks like it is easily one of those studies. Clearly the authors had a theory they wanted to confirm when they started the study. The wanted to find that income inequality leads to people wanting more luxury items. Surely they were going to find a correlation one way or the other. If a state didn’t fit the pattern they wanted, they would probably find a way to change the status of that state or the status of the luxury terms (by, you know, “controlling” for differences), or by changing other parameters.
So Neil, I realize you have no skeptical inclinations. I am a bit more discerning than you “science skeptics”.
Foreign people don’t necessarily make inequality higher or lower. Rural life doesn’t necessarily make inequality higher or lower. The point of controlling for variables is to remove the effects of these factors, whatever they might be, so that one can learn about the effect of the factor one is studying, in this case inequality.
I’m afraid you are misunderstanding the concept of “controlling for variables” in research. You are implying here that it is some sort of fudge factor to get the answer you want. Rather, it is exactly the opposite. It is intended as a standardized method to unbias the results in an established and formal way that others can understand, duplicate and critique, in order to help make sure that you’ve actually measured what you claim you have measured.
Open wider Phoodoo, you can fit your other foot in 😉
phoodoo,
You show no signs of having read the paper, you haven’t identified a single flaw in the study, and the only reason you started this thread is because you completely misinterpreted the abstract:
A classic phoodoo flameout.
You are obviously skeptical of this and other research, since you think “most published studies” are BS, which is fine. My point is not to defend this particular article, which is neither my field of expertise nor my interest, but rather simply to explain to the best of my knowledge what they did.
However, I think there is a flaw in your skepticism here, to the extent that you imply that the authors had a result in mind at the outset and then manipulated their methods and data until they got what they wanted — in other words, they were dishonest or wrong, whether intentionally or unwittingly. If most published studies are BS, it is more likely because they are not very important or not very interesting, not because they are dishonest or wrong. That’s not to say there isn’t dishonest or wrong research, but it is the exception, not the norm.
The reason “most published studies” aren’t dishonest or wrong is the same reason that we see scandalous headlines about the dishonest or wrong research — it’s not that easy to get away with. Researchers have to publish their methods and data, which makes it possible for others to check their work. Bad work gets called out, and few researchers, especially ones working in more mundane academic areas that don’t vie for Nobel prizes and big pharma contracts, are willing to risk their careers in order to “confirm” their own pet notions by fudging their results or not doing their due diligence to try to get it right.
Hmmm. Do you ever consider that you have a personal bias that makes you believe this peculiar statement? Consider this statement is actually not true, but that you wish to believe it true because it supports a larger prejudgement of yours? I’m don’t want to say which prejudgement motivates you to believe such a silly thing, but I invite you to look into your own heart.
You do have a bias of some sort. We all do, all humans do. Which is rather the point of science, after all: humans collectively creating a system of mental/physical tools to eliminate — or at least reduce — the bias of any one limited observer.
Collectively. *Gasp* Is that perhaps the underlying source of your personal bias against scientific studies? That they are the result of a long “collective” process? And perhaps, you being a hard-working right-thinking American, you cannot acknowledge any achievements which are not exemplars of individualism and exceptionalism. Well, in that case, all modern science is right out the window. Too bad, there goes that computer you’re using.
Think about it, if you can. It’s not too late for you to ferret out your own biases — not to destroy them altogether, that’s not possible — but to reduce their distortion of your own thought process which is currently causing you to believe and claim such stupid things.
Well, Hawaii has a rather low Gini co-efficient, both before and after adjusting for confounding factors. At 17.9% foreign born, it’s above the national average (12.9%), but not dramatically.
Your “doubt” is noted, and given the weight it deserves.
If you want to complain about the effect of the adjustment, you should look at California, Texas, and Florida, which have high raw Gini’s, but relatively modest residual Gini’s. New York’s score was lowered too, from very high, to just high.
Sooooo, according to Phoodoo’s nefarious fraudulent researcher hypothesis, the authors needed to move California, Texas, Florida and New York down a notch because (according to phoodoo-logic) residents of these states do not search for high status items online.
You should definitely publish that.
P.S. Accusing total strangers of fraud just because you don’t understand their work is rather offensive, BTW.
Nobody else really cares. That’s what I find most amusing about all of you and your ilk’s “methods”. You think it’s wrong but can’t actually say why but have no reverse gear? OK. Sounds like the minimum required for your average ID supporter so far.
You’ll be a discerning army of one and will have made no difference at all.
KF will still be mumbling about FIASCO when he’s 100 and he’ll be the only one doing so. BA77 will probably the closest you can get to a conscious chinese room in another 50 years, pasting messages he’s incapable of understanding but appearing to be on topic still. JoeG will be acting as the ID super-ego out on display for all to see. Where’s it got you in the last 10 years? Imagine that for another 100. Unless you actually write what you think is wrong and support it, in detail, who cares?
100 times zero is zero.
And the most amusing thing of all is that you actually have a way to make them care. You, I, everyone can do it.
All you have to do is build an unassailable array of supportable rebuttals and publish it. Then, when no response is forthcoming, for what response can there be to such a masterpiece, people will have no choice but to accept your claims.
Yet you’ve failed to respond to the specific technical points raised in this thread, just like they don’t exist. So it seems your way of dealing with rebuttals is just to ignore them.
Which once again leads me to the conclusion that you are a poe!
I think it’s quite possible that ID is actually all reverse-poe’s unaware of each other. No group of people can really be that lacking in self-awareness?
What sort of person skims over technical explanations that are only there because they themselves said something and then just carries on without addressing them?
An ID supporter! It’s a self-fulfilling prophecy. They believe in ID because of their ability to skip over evidence they don’t like. So that’s who. phoodoo.
It’s hilarious how Phoodoo gripes about his perceived problems with a paper based on its extract, largely because he doesn’t understand statistical methods but wont tell us how old *he* thinks the universe is. He’s a one trick pony like Joe Gallien and makes ID look bad (!) like Joe Gallien.
Hobbes,
Sorry, but this does not seem like an answer to my question at all. Does having many foreign born residents make a state more or less equal? Does having more city dwellers make a state more or less equal? Why?
DNA_Jock,
DNA Jock, if you have information about the study, why don’t you just show it. Is there some need for secrecy? Where is the list of the states and the adjustments?
Alan, please remove the array of guano on this thread, thank you.
Hobbes,
Did you read this link?
http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124
So you either believe the finding in this study or wrong, or you believe that most published research is wrong. If you believe the author of this study is wrong, why is he wrong? Is he intentionally wrong?
Hobbes, so far you (nor DNA Jock) nor Kieths (who hasn’t read it) have not provided any information from the study. Is it not there, or are you just not providing it?
phoodoo:
I’ve got a better idea, phoodoo. Why don’t you learn to take what you routinely dish out?
Have you read the paper, Phoodoo? If you don’t paste the last line of it, we’ll assume you haven’t and we can all move on. Thanks.
phoodoo,
What do you think Hobbes has been spoon-feeding you this entire time?
Do you know how ridiculous you look criiticizing a paper you haven’t read and whose abstract you don’t even understand?
keiths,
I was actually asking him to remove the guano for your sake Keiths. Omagain thinks your claim of reading the study is bullshit, since you can’t back up that claim, so I didn’t want to have to argue with him.
phoodoo:
Good. Then I cancel your request.
OMagain:
He’s already there. Whenever I ask him to state an argument in his own words, he folds like origami.
Good grief, phoodoo. I haven’t read the study. I took the data that Hobbes kindly provided for your education, and compared it with the publicly available Gini data for the US States.
Why on earth should I read the paper in order to prove wrong some random goofball who (as I explained on the Veblen goods thread) clearly cannot understand the abstract, has a proven track record of being impervious to evidence, and (based on our previous interactions) does not even understand what a p-value is?
SIWOTI
LMAO
eta: sorry folks, wordpress ate my links: the p-value debacle was on the “Yes, Lizzie, Chance is Very Often an Explanation” thread 12/29/13…
This comment sucks big time. This is way worse than phoodoo’s usual shitty-but-tolerable self-expression..
I found this aggravating enough to look up what phoodoo has distorted here:
phoodoo to keiths, 4/14 2:36pm :
One post later, OMagain responds:
Immediately, phoodoo fires back:
(Sorry, folks, for not linking to each of these comments from the thread; dunno how to do that, but more work than it’s worth anyways, I’m sure. But it’s just a quarter-way down the first page if you want to doublecheck.)
No one took issue with it at the time, but phoodoo’s comment sucked then and it sucks even worse now that phoodoo is inventing this:
Pants on fire, phoodoo, pants on fire!!
Thanks for letting me know what I think. Your selective misreading is telling, yet another demonstration of what it takes to be an ID supporter despite, not because of, the evidence.
As hotshoe has demonstrated your dishonesty I’ll leave it at that.
But try to remember, you don’t speak for me or say what I think.
DNA_Jock,
What DATA did hobbes provide about the adjusted values for the states? He simply said they adjusted for some variables, he didn’t say in what way the adjustments were used and how much they weighted them. How do you know that a 17% foreign born population wouldn’t cause a state to be shifted 20 spots? You don’t. How do you know if Hawaii residents are considered to live in a rural environment? You don’t? How do you know what the average income figure for the population they used was?
I am still waiting for hobbes to provide ANY data about the states rankings and adjustments, or to see if the paper even provides this information. Obviously Keith could speak up in defense of the paper he referenced, but since he hasn’t even read it himself, he also can’t.
BLOCK QUOTING the delusional Omagain.
Verbatim from the paper:
Information used in the regression analysis (see Methods: Variables and estimation) can be obtained from the U.S. Census Bureau. Exact instructions for accessing each dataset are provided below.
Urban population:
Go to http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t. In the “topic or table name” box type “Urban/Rural” and press “Go”. From the newly opened list of datasets, choose the 2010 113th Congress 100% Data titled “Urban and Rural” with ID “P2”. Click “download”. In the downloaded file named“DEC_10_SF1_P2_with_ann.csv”, Column E / Column D gives the “percent urban” measure we used.
Income inequality (GINI index):
Go to http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t. In the “topic or table name” box type “B19083: GINI INDEX OF INCOME INEQUALITY”. Using the “Geographies” menu, select “All States within United States and Puerto Rico”. The “Gini index of income inequality” (2012 ACS 5-year estimates) contains GINI index for each U.S. State for year 2012 (downloaded file named “ACS_12_5YR_B19083_with_ann.csv”).
Household income:
Go to http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t In the “topic or table name” box type “Household Income 2012”. Select “All States within United States and Puerto Rico” from the “Geographies” dropdown menu. Choose the 2012 ACS 5- year estimates dataset called “Household Income in the past 12 months (in 2012 Inflation- adjusted dollars)”. Click “download”. Mean income is in column IR of the downloaded file named “ACS_12_5YR_DP03_with_ann.csv”.
Percent of people who are foreign born:
Go to http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t and in the “topic or table name” box type “Selected social characteristics in the United States”. Choose “All States within United States and Puerto Rico” under “Geographies” tab and download the “2012 ACS 5-year estimates” dataset (File ID DP02). The downloaded file is named
“ACS_12_5YR_DP02.csv”. Use the estimates of foreign born population and total population to calculate percentages for each U.S. State.
Total population:
Go to http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t. Search for “Population total”. Select “All States within United States and Puerto Rico” under “Geographies” tab. Download “total population” (2013 ACS 5-year estimates) (File ID B01003). Population figures are in the downloaded file named “ACS_13_5YR_B01003_with_ann.csv”.
You should read the paper then. I don’t hold out much hope that you will be able to understand the adjustments they made, given your failure to understand the abstract. I note that you have evidently NOT yet read the paper, from which I conclude that YOU share my opinion of your likely ability to understand it. Hobbes HAS told you how the adjustments were used.
Well, Hawaii ranks #10 on the raw Gini list, and #5 on the residual Gini list, so it hasn’t shifted 20 spots. See, that was easy. I did look at the % foreign born for those states that moved the most (Texas, Florida, California); did you?
If you actually knew anything about multi-factorial regression, you would immediately recognize that the individual adjustments may be meaningless, but the residuals will be uncorrelated with any of the factors adjusted for, which is all that the authors needed to achieve.
If you, phoodoo, want to understand the relative importance of the different adjustments, what would you need to do first, in order to determine whether the co-efficients for the individual factors (“how much they weighted them” in your parlance) have meaning? This is multifactorial regression 101: if you cannot answer this question, you won’t be able to extract the information you desire.
Hobbes,
Did someone in that post you state what the adjustments were? I think I asked a pretty straight forward question earlier, and if you answered it forgive me, but I don’t see where.
How did they weight these variables? Does having more foreign born people make a state more or less equal? Does having a rural population make a state more or less equal?
It seems all that you have shown is where they got their numbers from, not how they applied them?
Also, do you agree or disagree with the link I provided you about most published studies being wrong?
Hobbes,
Sorry, I just saw your post one of your posts I missed earlier (which had the state information on it). I passed it when I first opened my browser, and it hadn’t appeared yet.
Hobbes,
So back to the original point. Does Michigan qualify as a state with greater income equality? Whilst Oregon qualifies as a state with less income inequality?
Poor phoodoo. You started this thread in an attempt to save face, and you’ve ended up humiliating yourself — again.
The data for the raw Gini and the residual Gini after regression, ranked from lowest inequality to highest inequality. You can see how Michigan and Oregon compare on both scales, and you can see how much movement there was of states based on the controlling for other variables. Some moved, some didn’t.
State Gini Index State Residual
Alaska 0.4132 Utah -0.02769
Utah 0.4197 California -0.02413
Wyoming 0.42 Alaska -0.02396
New Hampshire 0.428 Idaho -0.02376
Idaho 0.4281 Wyoming -0.02121
Hawaii 0.4294 Hawaii -0.0204
Iowa 0.4299 Iowa -0.0188
Wisconsin 0.4336 Wisconsin -0.01693
Vermont 0.4347 Nevada -0.01616
Nebraska 0.4357 Indiana -0.0146
Delaware 0.4373 Nebraska -0.01277
Indiana 0.4396 Washington -0.01073
Montana 0.4398 New Hampshire -0.01015
Maine 0.44 Vermont -0.0087
South Dakota 0.4417 Montana -0.00778
Minnesota 0.442 Texas -0.0076
Nevada 0.4434 Maine -0.00758
Washington 0.4437 Delaware -0.00565
Maryland 0.4444 Minnesota -0.00551
Kansas 0.4454 Florida -0.00443
North Dakota 0.4481 South Dakota -0.00372
Oregon 0.4517 Kansas -0.00357
Ohio 0.455 Ohio -0.00335
Missouri 0.4551 Arizona -0.00298
Michigan 0.4554 Michigan -0.00289
Colorado 0.4559 Oregon -0.00249
Arizona 0.4571 Maryland -0.00137
Oklahoma 0.4593 Missouri 0.0026
West Virginia 0.4596 Pennsylvania 0.00328
Virginia 0.4606 North Dakota 0.00543
Pennsylvania 0.4611 Illinois 0.00594
Arkansas 0.4618 Oklahoma 0.00597
Rhode Island 0.4634 North Carolina 0.00648
South Carolina 0.464 Arkansas 0.0065
New Mexico 0.4663 Colorado 0.00711
Kentucky 0.4666 West Virginia 0.00766
North Carolina 0.4666 South Carolina 0.01014
New Jersey 0.4669 Virginia 0.01107
Illinois 0.4681 New Mexico 0.01111
Alabama 0.4705 New Jersey 0.0114
Tennessee 0.4706 Georgia 0.012
Georgia 0.4719 Kentucky 0.01286
Massachusetts 0.4741 Rhode Island 0.01408
Texas 0.4741 Tennessee 0.01539
California 0.4751 Alabama 0.01693
Florida 0.476 Mississippi 0.02257
Mississippi 0.4765 Massachusetts 0.02431
Louisiana 0.479 New York 0.02584
Connecticut 0.4846 Louisiana 0.02771
New York 0.5005 Connecticut 0.04252
keiths,
Not only have you not read the paper, you haven’t even contributed ONE bit of information to the discussion. An amazing fail on your part.
Hobbes,
I think you didn’t quite understand my question. For the purpose of the results, was Michigan considered a state with equality or inequality? What about Oregon? They are next to each other in the rankings.
Are you jealous because that’s normally your role?
phoodoo,
Lying about your opponents won’t help you save face.
keiths,
Feel free to answer the question about Michigan…oh, wait , I forgot, you haven’t read the paper, you can’t contribute…..nevermind.
phoodoo,
Again:
Lying about your opponents won’t help you save face.
Lizzie’s first rule:
Assume all other posters are posting in good faith.
For example, do not accuse other posters of being deliberately misleading
Alan Fox,
Alan, there is a whole lot of guano here, that could be removed so we can actually make a fair assessment of the scientific credibility of the paper.
phoodoo,
Is that a formal request?
phoodoo,
Don’t forget. Your guano request was made on my behalf, and I’ve already canceled it.
Alan Fox,
Ok sure