Faking the Data on Changing Attitudes About Gay Marriage
Recently, there was a scientific study showing that people’s attitudes on gay marriage could be changed by interacting with a gay pollster whose life would actually be affected by the right to marry. It seemed logical that people would want to be more polite to the gay pollster, but the surprising result of the study was that the changes in attitude were long-lasting.
Now it turns out that the study was likely based on faked data. It has been withdrawn, and 538 warns us how easy it is to fall for hoaxes like this.
People who are willing to lie to push their political agenda cannot be trusted in any respect.
To me, the moral of this story is how easy it is for scientists to lie and get away with it. This fraud was only exposed because the principal author made extravagant funding claims and the debunkers were lucky enough go find another source for the data.
DRJ (e80d46) — 6/2/2015 @ 7:52 amThere are so many questions about this student researcher, but his conclusions were so popular that they were immediately and widely accepted. It reveals there is a disincentive to question popular results.
DRJ (e80d46) — 6/2/2015 @ 8:09 amA few notes:
1. (to DRJ): The initial problem was just that people don’t act this way. Broockman, the key debunker, was suspicious from the outset, though he gamely tried to replicate the study.
2. I don’t think LaCour’s primary problem was politics – he earlier did a study that said partisan lefties *and* righties consume primarily neutral political news. (This also has notes of fabrication.) He’s just a compulsive liar who would have done anything to get that tasty Princeton professorship. Now, as to people believing it…
3. Broockman (the debunker) is a gay liberal. Who practiced science.
4. LaCour’s response deserves a thousand words I won’t give it here. It is 23 pages of awesome doubling down. It’s the not-my-pants defense of a social scientist.
5. “Likely based on false data,” is an understatement. I’d bet my car and my house against a cheeseburger that this was fraudulent. I’ve read everything involved in this, and everything screams fraud.
JRM (de6363) — 6/2/2015 @ 8:11 amThere was also discussion recently about another ‘study’, on health benefits of chocolate in relation to weight loss.
I think in that case, the data was accurate and not faked. Problem is, there weren’t enough data points to be statistically significant, too many parameters tested on to allow cherry-picking of outcomes, and other things that were intentionally bad, scientifically and statistically-speaking.
And apparently, it was a form of honey-pot by these scientists, to show just how little journalism is done by actual Journalists; it took the comment sections for people to actually ask ‘What was the size of the groups tested’, instead of those purporting to do the news.
Another Anon (f43943) — 6/2/2015 @ 8:12 amBroockman claims he was initially impressed with LaCour’s research. His doubts came later, after he was unable to replicate the results and because of the funding anomalies.
DRJ (e80d46) — 6/2/2015 @ 8:23 amthey do this with climate change too
they just make stuff up
happyfeet (831175) — 6/2/2015 @ 8:25 amAnother Anon -‘I was just reading that article this morning. It was amazing the number of “peer reviewed” journals that published their study.
JD (3b5483) — 6/2/2015 @ 8:34 amSo he was trying for a big splash instead of something already covered to tedium by other dissertations. Can happen. Needed to distinguish himself. Happens. Faked the data. Happens. Figured he’d get away with it because of the subject–nobody wants to be a h8ter, and as Broockman describes, there’s an institutional culture against exposing somebody’s fraud. Poli sci is a small community and there’s no place in the business you can guarantee you won’t run into the guy, his fans, or people who just don’t like whistleblowers.
Richard Aubreyr (f6d8de) — 6/2/2015 @ 8:40 amSo, he figured he’d get away with it. Happens.
But this isn’t faking a newly-discovered Shakespeare sonnet.
The BFD here is the subject. If he could sell this, he could be on top of a multi-billion dollar persuasion industry.
Would it work? Of course not. But so what? Having guys on television yelling at you about the deals at Whosits Autos doesn’t work but they keep doing it,spending money on it. This dude was going to be rich.
The scientific enterprise in a world of proliferating journals and lots of money and prestige/influence in some quarters is like our government/society,
able to function only when there is an adequate public virtue to support and maintain it.
And there isn’t anymore.
Whether it can be regained only the Lord knows.
http://www.powerlineblog.com/archives/2015/05/behind-science-fraud-chapter-4.php
MD in Philly (f9371b) — 6/2/2015 @ 9:02 amIf the debunker is a gay liberal, is he going to get trashed like the NY hotelier for going against the approved interest group?
MD in Philly (f9371b) — 6/2/2015 @ 9:04 amHide the decline.
njrob (d094ac) — 6/2/2015 @ 9:24 amstrike out the word “likely”, it was faked.
seeRpea (b2f97d) — 6/2/2015 @ 9:24 am3. Broockman (the debunker) is a gay liberal. Who practiced science.
As DRJ pointed out, there are an awful lot of lessons to be drawn from this fiasco. Certainly two of them are the degree to which it is very easy to fake scientific data all the way up to getting a story published in a mainstream outlet as well as the temptation for scholars, even respected ones like Donald Green, to believe studies which confirm their own biases. But this situation also should alert us to the idea of being extra skeptical when researchers study areas where they have personal interests. From what I have read I think that LeCour (the fraudulent researcher) is gay, so he certainly had reason to want his hypothesis to be proven true. This confirmation bias is a real problem in areas such as environmental science, grievance studies, and most other ideas where political ideology overlaps with academic pursuits. It should also alert academia to the degree to which political ideology bullies researchers into not questioning certain results. Broockman was apparently advised that it wouldn’t be a good thing for him from a professional standpoint to debunk such a popular study. Do you think the higher ed cartel will be honest enough to do some self-examination on the degree to which they push for politically correct results?
JVW (8278a3) — 6/2/2015 @ 9:47 amI agree that this story has many lessons. One of the more minor but far-reaching lessons is that apparently some graduate schools don’t check applicants’ resumes. It appears that LaCour claimed to be a Magna Cum Laude graduate from UT-Austin but the Austin branch of UT doesn’t use Latin Honors, a fact that is published at the UT website. Instead, UT recognizes departmental honors and university honors, the latter of which are the equivalent of Latin Honors. Further, anyone can verify UT Austin degree information at the UT website. It reflects LaCour graduated with departmental honors but not with university honors.
DRJ (e80d46) — 6/2/2015 @ 10:36 amYes, “People who are willing to lie to push their political agenda cannot be trusted in any respect.”
Why does anyone give Ace a pass?
Scrutineer (b7d257) — 6/2/2015 @ 10:48 amWhy does anyone give Ace a pass?
Ace’s nervous breakdown (from a blogger’s perspective) isn’t quite the same as falsifying data for a scientific journal, but nice try at the moral equivalency, sport.
JVW (8278a3) — 6/2/2015 @ 10:58 amTo amplify that comment: When you go to a blog you expect opinion. That doesn’t mean people should lie, but there’s the understanding that the “fact” is shaded by the blogger’s ideology. Science is definitely not supposed to work that way.
JVW (8278a3) — 6/2/2015 @ 11:00 amScrutineer,
I can’t speak for Patterico but we’re all human and sometimes we lie to ourselves or others. Or maybe it’s because we see Ace’s admission for what it is — a recognition that GOP leaders are the real liars and Ace recognizing that is his redemption.
DRJ (e80d46) — 6/2/2015 @ 11:02 amWell, there’s science and then there’s social science. Different fields have different standards and different systems in place. So perhaps a broad generalization about “science” is not really applicable here.
In physics (the only one I’ve published in), if you come out with some staggering and long-sought-after result, you are going to have hordes of people going through your work with a fine-toothed comb, mostly because they want to build on what you’ve started. A fraud or well-intentioned, but spurious, methods will be exposed in short order because other people won’t be able to duplicate what you did.
However, if you do something that few people care about, and don’t make bold claims, you could probably totally make stuff up and get away with it for a long time. Because there is a limit to how much checking anyone can do on you. Auditing every paper is a bit like putting surveillance camera’s in everyone’s home–the amount of wrongdoing you will uncover is not worth the enormous investment of time and money spent to uncover it.
Gabriel Hanna (64d4e1) — 6/2/2015 @ 11:21 amOne more thing, Scrutineer:
I don’t put the same importance on honesty as Patterico does. Yes, I try to be honest because I care about my integrity and reputation, and I would rather be around other people who share that view. But I know sometimes I’m going to be wrong and make mistakes, and I also know that in today’s climate there is a fine line between lying and mistakes — as we learned from Bush and the Iraq WMDs.
Thus, what matters to me is people getting in the arena and letting everyone vigorously debate ideas. That’s how we get to the truth. I don’t read Ace or Patterico because they are always right or truthful. I read them because they are always provocative and make me think.
DRJ (e80d46) — 6/2/2015 @ 11:24 amGood point, Gabriel Hanna. It’s my impression Broockman was impressed with LaCour’s work and wanted to build on it, perhaps because he saw it as a way to similarly enhance his career prospects.
DRJ (e80d46) — 6/2/2015 @ 11:26 am@DRJ: One of the groups I was in was involved in chasing a White Whale, that’s how I know. Every six months or so another group would claim to have caught it, and the refutations would follow soon after. I do not believe that White Whale has been caught yet.
The peer-reviewers in that field have had to learn the IR spectrum of oil very well, because that’s what a significant percentage of startling new results were-oil contaminating the sample.
Gabriel Hanna (64d4e1) — 6/2/2015 @ 11:33 amDRJ, you could not more harshly insult Ace’s intelligence than by suggesting he only just this year figured out “GOP leaders are the real liars.”
Your point about Bush/WMDs makes me think we’re talking about different things. Honest mistakes (even those motivated by wishful, ideological thinking) are one thing. Consciously lying for “the team” is something else. Ace’s statement reminded me of Stephanopolous’ wonderful announcement that Clinton “kept the promises he meant to keep.”
JVW, Ace to his credit did not offer the feeble defense that “there’s the understanding that the ‘fact’ is shaded by the blogger’s ideology.” Yes, there’s a difference between an op/ed and straight news. No, Ace did not confess to mere opinionating.
Scrutineer (b7d257) — 6/2/2015 @ 11:52 amTo me, it’s not about when Ace realized GOP leaders were lying to us. It’s about when he decided he’d had enough. We all deal with that on our own timetable.
As for lying versus mistakes, my point is that it doesn’t matter to the recipient of information whether the information is the result of a lie or a mistake. What matters is whether the information is right or wrong. That’s why I care about getting the information into the arena — so we can figure out the truth.
DRJ (e80d46) — 6/2/2015 @ 12:17 pmRegarding blogging, I think it is the best policy to say only things that you believe. If you are convinced you will do more harm than good by saying what you believe, then the very most you are justified in doing is keeping silent.
I cannot imagine expressing opinions I do not hold nor saying things I do not believe to be true.
Gabriel Hanna (64d4e1) — 6/2/2015 @ 12:43 pmI’ve blogged in the past. As a rule, it wasn’t about my opinion, it was about putting stories out there for discussion. That’s one reason I like commenting more than blogging, because I can say what I think.
I like Ace because he isn’t shy about stating his opinions but he also engages in hyperbole and rhetorical devices, and he changes his mind. It seems to me that he may have thought GOP leaders were lying long before he said something. I can understand not quibbling about every political detail until you reach the tipping point.
DRJ (e80d46) — 6/2/2015 @ 1:03 pmThe entire point is that he was shy about stating his opinion. He peddled lies to promote the GOP instead of sharing his real views. I would love to read what he really thinks, but I doubt he’ll ever tell us because he wants to protect his niche at the right edge of the Overton window.
Scrutineer (b7d257) — 6/2/2015 @ 1:19 pmyou are a very cynical scrutineer indeed
happyfeet (831175) — 6/2/2015 @ 1:21 pmScrutineer,
Then it’s an easy call. Don’t read Ace.
I’m sure there are thousands of bloggers who always say exactly what they think and never hold back, not even once, for any reason. Of course, that doesn’t mean they will always be right, does it?
DRJ (e80d46) — 6/2/2015 @ 1:28 pmIt’s like when someone describes AIDS as a “pandemic” you can pretty much just ignore them.
CrustyB (69f730) — 6/2/2015 @ 1:28 pmBy all means, Scrutineer, write a post about Ace if it means that much to you. While entertaining and thought provoking, his willingness to sway with the wind has always left me wanting.
JD (3b5483) — 6/2/2015 @ 1:36 pmThat’s all well and good, Scruitneer, but your original comment asked why were jumping all over the authors of this study while seemingly not caring about Ace’s admissions of mendacity. Do you understand our point that lying in a “scientific” study conducted under the auspices of a tenured professor at an accredited institution of higher education and involving a prestigious scientific magazine is far more significant than lying on a political opinion blog, or do we have to explain to you why?
JVW (8278a3) — 6/2/2015 @ 1:38 pmBecause there is a limit to how much checking anyone can do on you.
I personally know of a couple of medical studies that are fraudulent and were proven to be so at major medical meetings but the same studies are quoted today in medical literature. The fellow who published these studies was well known and his manipulation of data was well known among his colleagues.
One example of an attempt to replicate the results.
Overall (P = 0.42) and corrected (P = 0.25) survival did not differ significantly among the treatment groups although in every analysis the survival data of the no-touch isolation group were superior. The data do suggest a limited benefit of the no-touch isolation technique. This observation is important since the morbidity and mortality of surgery were equal in both groups.
This is known as the proverbial British courtesy. It made no difference in results yet it is quoted all the time. I was present when the original study was shown to be fraudulent. The author, a very well known surgeon, had used time-life tables to correct mortality for his study group but HAD NOT USED THEM for the control group. The control group had more deaths, of course, but not from cancer.
He also fraudulently reported on breast cancer results.
Mike K (504df3) — 6/2/2015 @ 1:45 pmYes, they do. A lot.
But they make up crap in every area of scientific publishing.
Publish or perish. At best about half of of the “peer reviewed” studies you read in the academic journals may be true. But you’ll never know which half.
I’m sure it’s higher in the journals of Climastrology.
http://www.powerlineblog.com/archives/2015/05/behind-science-fraud-chapter-6.php
Steve57 (f41960) — 6/2/2015 @ 3:05 pmI think he meant “queer reviewed” journals.
Hoagie (b482c1) — 6/2/2015 @ 3:12 pm@34 – Hoagie, I think he was referring tot he “chocolate” study, not the “gay marriage” study.
Bill M (906260) — 6/2/2015 @ 5:45 pmWhenever a liberal uses that word, he should be required to put quotation marks wround it:
The “science” is settled.
Bill Nye the “Science” Guy.
The vast majority of “scientists” agree.
arik1969 (02de93) — 6/2/2015 @ 6:29 pmMike K.-
MD in Philly (f9371b) — 6/2/2015 @ 7:00 pmYes, I read somewhere that even when a study has been retracted it is often quoted in further studies and the corrections are ignored.
I guess with the proliferation of journals it’s easy to do these days.
I have a comment that’s been in moderation for hours. I’ll just pass along the links.
http://investmentwatchblog.com/1-researcher-faked-171-papers-and-is-responsible-for-7-of-all-retractions-since-1980/#zOlPytkApHTzCXHu.99
Steve57 (f41960) — 6/2/2015 @ 7:08 pmhttp://www.powerlineblog.com/archives/2015/05/behind-science-fraud-chapter-6.php
Anybody who isn’t a skeptic is a gullible fool. I came across an article in The Lancet (UK equivalent of the New England Journal of Medicine) a while back. It observed that maybe half of peer reviewed articles in scientific journals are actually good science. I.E. not just fatally flawed if not outright fraudulent. The problem is you never know which half.
Steve57 (f41960) — 6/2/2015 @ 7:12 pm97 %ers pay for the 3 %ers crying and feet stomping.
mg (31009b) — 6/2/2015 @ 7:22 pmSteve.
My irony detector just exploded. It was the Lancet which published that grand article saying myriads of thousands died in Iraq because of the US invasion, wasn’t it? And they were questioning the generality of studies? Chutzpah has a new definition.
Serious question. How does a researcher know if a study has bern confirmed, refuted, retracted, or otherwise followed up on. There is a citatíon service which tells lawyers if cited cases have been overruled, cited as precedent in other jurisdictions. Is there a similar index in medical research?
kishnevi (9c4b9c) — 6/2/2015 @ 7:31 pmkishnevi, I’m not quite sure what you’re driving at. So I’ll approach it two ways.
There is a site called Retraction Watch that attempts to keep track of articles that have been retracted as either fraudulent or false.
According to the three minute video embedded in the Power Line post there’s at least one company that offers as a service to prospective research paper authors to try and replicate the experiment(s) on which a proposed journal article is based. And as an independent, unbiased finder of fact report its results.
Of course, that service isn’t offered for free but may be worth it to Universities, etc., which would like to preserve their reputations.
Steve57 (f41960) — 6/2/2015 @ 8:12 pmWhat I had in mind was an equivalent to Shepard’s…and looking for a descriptive link found something like the answer to my question, the Science Citation Index.
kishnevi (9c4b9c) — 6/2/2015 @ 8:31 pmI wasn’t looking simply for retractions, but more general followups by other studies.
Despite the festishists on the left, science does not denote a collection of settled facts. Science is a collection of questions with unknown answers. It is a way to find those answers through empirical truth. If anyone says “the science is settled,” they know nothing of science regardless of how many scientists say so.
This is what bothers me about the climate change folk. They have drawn a line in the sand that says this scientific query has an answer and it is absolute. It is and it can not be changed or challenged.
Logically, when that line is drawn, any potential solution is void unless it follows the line already established by science.
As a semi-educated person, I am prone to think that science is conducted in an objective environment because science has allowed me to live far beyond my ancestors’ average age of death.
Regardless, in science, I cast a wary eye on scientists who know “the truth” and those who follow them because, you know, that leads to sadness and pain.
Ag80 (eb6ffa) — 6/2/2015 @ 9:35 pm@kishnevi:ooking for a descriptive link found something like the answer to my question, the Science Citation Index.
I wasn’t looking simply for retractions, but more general followups by other studies.
Scitation is one tool, but it doesn’t really substitute for being well-read in the literature pertaining to your field, because Scitation too big. But checking Scitation for relevant references is part of what you’re expected to do before writing a paper.
This is a large part of peer review. Good peer reviewers have read a lot of the literature and they will point out references that bear on your paper. Generally this means you’ll end up needing to cite these papers to get yours accepted. If one of your references is discredited your peer reviewers, if they catch it, will include this in their comments along with their suggestions for what you’ll need to change after taking it out.
Gabriel Hanna (748221) — 6/2/2015 @ 9:57 pmWe’re all acquainted with the saying “First reports are frequently wrong” (especially in the context of military success and defeat) “we need to wait (or look again in) 24 hours.”
Same rule in science. First articles are frequently wrong. Flawed. Incomplete. Not reproducible. Give it a couple of weeks (or months.) It seems to me a lot of what is published as science these days just isn’t. Publish or perish has given way to Publish, Collect the Grant, and Go On.
Self-deception is a flaw in everyone’s makeup, and if you find someone saying they don’t have it, you know they’re deeply flawed in that way.
htom (4ca1fa) — 6/2/2015 @ 10:42 pmPeople who are willing to lie to push their political agenda cannot be trusted in any respect.
This is debatable but doesn’t have much to do with LaCour in any case. He appears to have been lying to promote his career not his political agenda if any.
James B. Shearer (23c542) — 6/3/2015 @ 12:08 amPerhaps both, Mr.Shearer? We do not know one’s inner motives.
Hoagie (b482c1) — 6/3/2015 @ 4:31 amthey do not believe in the gay marriage?
we will go to their houses and say hey please to believe in the gay marriage!
it’s science!
they’re so screwed hah! gay agenda ftw
happyfeet (831175) — 6/3/2015 @ 5:04 amYou keep telling yourself that, Shearer.
JD (82bdb7) — 6/3/2015 @ 6:01 am