Retraction Watch

Tracking retractions as a window into the scientific process

Author retracts study of changing minds on same-sex marriage after colleague admits data were faked

with 65 comments

science coverIn what can only be described as a remarkable and swift series of events, one of the authors of a much-ballyhooed Science paper claiming that short conversations could change people’s minds on same-sex marriage is retracting it following revelations that the data were faked by his co-author.

[3:45 p.m. Eastern, 5/28/15: Please see an update on this story; the study has been retracted.]

Donald Green, of Columbia, and Michael LaCour, a graduate student at UCLA, published the paper, “When contact changes minds: An experiment on transmission of support for gay equality,” in December 2014. The study received widespread media attention, including from This American LifeThe New York Times, The Wall Street JournalThe Washington Post,  The Los Angeles Times, Science FridayVox, and HuffingtonPost, as LaCour’s site notes.

David Broockman and Joshua Kalla, graduate students at University of California, Berkeley, were two of the people impressed with the work, so they planned an extension of it, as they explain in a timeline posted online yesterday:

As we examined the study’s data in planning our own studies, two features surprised us: voters’ survey responses exhibit much higher test-retest reliabilities than we have observed in any other panel survey data, and the response and reinterview rates of the panel survey were significantly higher than we expected. We set aside our doubts about the study and awaited the launch of our pilot extension to see if we could manage the same parameters. LaCour and Green were both responsive to requests for advice about design details when queried.

Earlier this month, they began a pilot of their extension. They soon realized that

The response rate of the pilot study was notably lower than what LaCour and Green (2014) reported.

When Broockman and Kalla contacted the firm they thought had performed the original study upon which the Science paper was based,

The survey firm claimed they had no familiarity with the project and that they had never had an employee with the name of the staffer we were asking for. The firm also denied having the capabilities to perform many aspects of the recruitment procedures described in LaCour and Green (2014).

After finding several other irregularities, the pair contacted Green, who was concerned, and they also asked Yale political science professor Peter Aronow to join their work. By May 16, the team had found other irregularities, and sent them to Green, who reviewed them and on May 17 agreed that

a retraction is in order unless LaCour provides countervailing evidence. Green also requests this report be made public concurrently with his retraction request, if this request is deemed appropriate.

Over the next two days, Green confronted LaCour and told the team that LaCour had

confessed to falsely describing at least some of the details of the data collection.

Green then added a note on May 19 to his website saying the paper was retracted, and submitted a retraction letter to Science:

I write to request a retraction of the above Science report. Last weekend, two UC Berkeley graduate students (David Broockman, and Josh Kalla) who had been working on a research project patterned after the studies reported in our article brought to my attention a series of irregularities that called into question the integrity of the data we present. They crafted a technical report with the assistance of Yale professor, Peter Aronow, and presented it to me last weekend. The report is attached. I brought their report to the attention of Lynn Vavreck, Professor of Political Science at UCLA and Michael LaCour’s graduate advisor, who confronted him with these allegations on Monday morning, whereupon it was discovered that he on-line survey data that Michael LaCour purported to collect could not be traced to any originating Qualtrics source files. He claimed that he deleted the source file accidentally, but a Qualtrics service representative who examined the account and spoke with UCLA Political Science Department Chair Jeffrey Lewis reported to him that she found no evidence of such a deletion. On Tuesday, Professor Vavreck asked Michael LaCour for the contact information of survey respondents so that their participation in the survey could be verified, but he declined to furnish this information. With respect to the implementation of the surveys, Professor Vavreck was informed that, contrary to the description in the Supplemental Information, no cash incentives were offered or paid to respondents, and that, notwithstanding Michael LaCour’s funding acknowledgement in the published report, he told Professor Vavreck that he did not in fact accept or use grant money to conduct surveys for either study, which she independently confirmed with the UCLA Law School and the UCLA Grants Office. Michael LaCour’s failure to produce the raw data coupled with the other concerns noted above undermines the credibility of the findings.

I am deeply embarrassed by this turn of events and apologize to the editors, reviewers, and readers of Science.

Green tells Retraction Watch:

…Michael LaCour attended my summer workshop on experimental design in 2012 and proposed at that time a project that involved both canvassing and internet surveys.  It sounded to me too ambitious to be realistic for a graduate student but in principle worthwhile.  I later introduced him to Dave Fleischer, who heads up the LGBT canvassing operation in Los Angeles, and they struck up a collaboration.  Several weeks after the canvassing launched in June 2013, Michael LaCour showed me his survey results.  I thought they were so astonishing that the findings would only be credible if the study were replicated.  (I also had some technical concerns about the “thermometer” measures used in the surveys.)  Michael LaCour and Dave Fleischer therefore conducted a second experiment in August of 2013, and the results confirmed the initial findings.  Convinced that the results were robust, I helped Michael LaCour write up the findings, especially the parts that had to do with the statistical interpretation of the experimental design. Given that I did not have IRB approval for the study from my home institution, I took care not to analyze any primary data — the datafiles that I analyzed were the same replication datasets that Michael LaCour posted to his website.  Looking back, the failure to verify the original Qualtrics data was a serious mistake.

According to his website, LaCour will become an assistant professor at Princeton University in July. [Update: As of 8 a.m. Eastern on 5/20/15, that mention had been removed from his site, but it is still available on the Google cache version.] We’ve contacted him for comment, and will update with anything we learn.

Update, 10:30 a.m. Eastern, 5/20/15: Broockman tells Retraction Watch he agrees that this was remarkably swift and transparent:

I think there’s a couple reasons why. First, the study’s findings had huge implications for people who were trying to advance the cause of equality and have changed how advocates do their work. Every minute we knew the truth and did not disclose it really was a lie by omission to the advocates out there. There was some element of time sensitiveness. Second, the nature of the claims made in the article about the scope of the study — the survey’s scale, its funding, the number of bills that would have to have been incurred to pay for it, etc. — meant that we expected it would have been straightforward to produce at least one piece of evidence that something among these many claims occurred that would inject a shred of doubt into the suspicions. I would guess that Don realized that if not one such piece of evidence could be mustered after 48 hours it was very unlikely that anything satisfactory was going to ultimately materialize.

Update, 2 p.m. Eastern, 5/20/15: This post’s popularity crashed our servers, and we have now upgraded. Apologies for the interruption. In the meantime, we have heard from Science, who sent this comment from editor in chief Marcia McNutt, noting that the journal will be posting an Expression of Concern:

Thank you for your query about the possible retraction of the study, “When contact changes minds: An experiment on transmission of support for gay equality,” published in Science by Michael J. LaCour and Donald P. Green in December of 2014. Science takes this case extremely seriously and will strive to correct the scientific literature as quickly as possible.

No peer review process is perfect, and in fact it is very difficult for peer reviewers to detect artful fraud. Fortunately, science is a self-correcting process; researchers publish work in the scholarly literature so that it can be further scrutinized, replicated, confirmed, rebutted or corrected. This is the way science advances.

Dr. Green was informed about the study’s irregularities over the weekend. He submitted a request for retraction to Science yesterday, Tuesday, 19 May, after his co-author, LaCour, admitted that some of the details of the data collection were falsely described in the published report. At this time, our Editorial staff is assessing the report. Given the fact that the Dr. Green has requested retraction, Science will move swiftly and take any necessary action at the earliest opportunity. In the meantime, Science is publishing an Editorial Expression of Concern to alert our readers to the fact that serious questions have been raised about the validity of findings in this study.

We thank those who attempted the replication and pointed to the possible irregularities. It allowed the author to look more carefully into possible problems with the original study.

We also heard from Princeton. A spokesperson tells us:

As you’ve correctly noted, at this time the individual is not a Princeton University employee. We will review all available information and determine next steps.

And LaCour tells us:

I’m gathering evidence and relevant information so I can provide a single comprehensive response.  I will do so at my earliest opportunity.

Update, 10 a.m Eastern, 5/22/15: LaCour sent us (and others) this message:

I will supply a definitive response (including a timeline of events) on or before May 29, 2015. I appreciate your patience, as I gather evidence and relevant information, since I only became aware of the allegations about my work on the evening of May 19, 2015, when the not peer-reviewed comments in “Irregularities in LaCour (2014),” were posted publicly online.

I must note, however, that despite what many have printed, Science has not published a retraction of my article with Professor Green. I sent a statement to Science Editor McNutt this evening, providing information as to why I stand by the findings in LaCour & Green (2014). I’ve requested that if Science editor McNutt publishes Professor’s Green’s retraction request, she publish my statement with it.

Hat tip: Lila Guterman

Like Retraction Watch? Consider supporting our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post.

Written by Ivan Oransky

May 20th, 2015 at 7:09 am

Comments
  • Neuroskeptic May 20, 2015 at 7:33 am

    “Several weeks after the canvassing launched in June 2013, Michael LaCour showed me his survey results. I thought they were so astonishing that the findings would only be credible if the study were replicated … Michael LaCour and Dave Fleischer therefore conducted a second experiment in August of 2013, and the results confirmed the initial findings.”

    An important reminder that replication is not proof of integrity! If you can fake one result, you can fake two.

  • Emil O. W. Kirkegaard May 20, 2015 at 8:49 am

    Interesting how Institutional Review Boards in this case caused a worse outcome because they made the co-author unwilling to analyze primary data. I wonder how common that situation is? Why do they require ‘ethical review’ for analyzing survey data? It makes no sense.

    • Ken May 22, 2015 at 6:33 am

      I serve on an ethics committee and this is something I don’t usually see a problem with. Possibly the problem was that the nature of the trial meant there had to be identification information in the data files but that could be sorted out either by Qualtrics de-identifying the files or giving the co-investigator ethics approval.

  • Bashir May 20, 2015 at 9:00 am

    At every institution I’ve been at the co-author getting the ok from IRB to view the data would be trivial. This seems like an odd detail to me. Why didn’t the co-author just get the official ok?

    • egrl May 26, 2015 at 6:12 pm

      This appears to be his “excuse” for not doing what he should have done—especially in a study with such far-reaching implications.

  • Michael May 20, 2015 at 9:54 am

    As noted in the post, you can find a detailed explanation of these data irregularities at http://web.stanford.edu/~dbroock/broockman_kalla_aronow_lg_irregularities.pdf

  • Salamano May 20, 2015 at 10:05 am

    This is an interesting scenario to stack up against the anything-but “swift” or “remarkable” tale of the Lewandowsky papers on conspiracy ideation among climate change deniers and what-not…

    http://retractionwatch.com/2013/03/28/why-publishers-should-explain-why-papers-disappear-the-complicated-lewandowsky-study-saga/

    As part of the parallel, an inquisitive grad student (Jose Duarte) was among the people skeptical of this work. But, unfortunately for science, none of this has ever made it to the laudatory level that the instant turnabout seen with LaCour (2014).

    I guess when push comes to shove, “science” isn’t equally self-correcting after all. Some causes apparently matter just that much.

  • Dave Burton May 21, 2015 at 7:23 am

    When linking to pages in Google’s cache, it is important to make copies, because they don’t last long. The link above, to the cached version of LaCour’s page which said he would be an Assistant Prof. at Princeton in July, is already obsolete. However, I found a cached equivalent page, and saved it in several places:

    1. https://archive.is/IkWPV (16 May version, from Yahoo)

    2. https://archive.is/NRkvB (20 May version, from Google)

    3. http://www.burtonsys.com/mikelacour_experiments_cached.html (copy of the 20 May Google version)

  • Kevin May 21, 2015 at 9:20 am

    And this on Princeton’s News site:

    http://www.princeton.edu/csdp/news/archive/?id=14943

  • Theresa Defino May 21, 2015 at 9:30 am

    “Given that I did not have IRB approval for the study from my home institution, I took care not to analyze any primary data — the datafiles that I analyzed were the same replication datasets that Michael LaCour posted to his website.”

    So…did his IRB deny the protocol or was it ever even asked to review it? And what IRB did approve it?

    And what difference does IRB approval mean, anyway? So…all the exempt research in the world doesn’t involve the senior, in this case, co-author, reviewing real data? Scapegoating?

  • Blunderqueen May 21, 2015 at 10:14 am

    @ Emil O. W. Kirkegaard et al.
    I do agree with your comment and I’d add: whether an IRB approval is required to work out primary data why the Senior Author decided going ahead despite this?
    Most peculiar: why at the end of the article published on Science the reader can find the statement “This research was approved by the UCLA institutional review board, case no. 12-001750.” ? (ACKNOWLEDGMENTS paragraph) and here https://www.socialscienceregistry.org/trials/611/history/3433

  • Marco May 21, 2015 at 11:11 am

    Blunderqueen, LaCour is at UCLA. Green is at Columbia.

    Theresa, I have to admit some ignorance about IRB’s, but I do not think it is all that easy to get IRB approval after the study has already been performed, and notably elsewhere. Also, I assume that IRB approval would then have had to come from Columbia. Just imagine that the Columbia IRB had some problems with the proposed set-up!

  • NeuroVoice May 21, 2015 at 11:17 am

    I’m can’t speak with certainty about Columbia, but typically, for a researcher to work with data at an institute s/he would need IRB approval at his/her institute. Some IRBs are more stringent than others and so approval is not automatic just because it has been granted by another institute’s IRB. In this case, the data was collected through UCLA where the grad student was, with UCLA’s IRB approval. Green would have needed approval from Columbia’s IRB to work with the data himself. This is not a difficult process though; it should have been easy enough to put in a data transfer agreement in place. Then again, PIs typically do not look at raw data even with their own grad students.

  • NotClareFrancis May 21, 2015 at 2:42 pm

    Sorry to muckrake, but I have had a few tedious days trying to teach a student how to collect meaningful raw data and reading this has really put me in an angry mood.

    Apparently this paper won the 2015 Pi Sigma Alpha Award for Best Paper, presented at the Midwest Political Science Association Annual National Conference.

    I gathered this information from LaCour’s CV on his website and from his Twitter account. Will this award now be returned or rescinded?

  • The Watcher May 21, 2015 at 3:10 pm

    Surely since LaCour’s PHD states Spring 2015, it would mean that it is based on the work of this fraudulent retracted paper. Is University of California going to be looking at the recent award of a PhD contain this work. His PhD dissertation was “When Persuasion Works, Lasts and Spreads: Evidence From Three Longitudinal Field Experiments on Gay Equality and Abortion”, with a committee which includes Don Green who has just retracted this Science paper.

  • Testosterone May 21, 2015 at 3:13 pm

    One has to wonder if LaCour acted alone in the creation of the charts, diagrams, personal stories, etc.

  • qqaaqq May 21, 2015 at 4:13 pm

    Assuming that there is not, in fact a quite miraculous rebuttal from Mr. LaCour (whose dissertation’s legitimacy seems to be quite tenuous at this point, enough so for me to personally retract his title for the moment), it would seem that we have on our hands two heros: David Broockman and Joshua Kalla. These folks did some solid work. Equally important as doing a study is confirming it or taking it down. However, when they go to apply for jobs, they will have nothing to show for this effort… in fact, they will be worse off for having “wasted” a lot of time not producing a paper.

    I suggest that when a flaw is uncovered like this that leads to a retraction, the discovering authors should be awarded the opportunity to publish their work thus far as an article in the original journal as well as their commentary on the flaws in the original article. Further, it would be great if journals would publish online only “replication” studies giving reward to those who confirm.

    • egrl May 27, 2015 at 1:20 am

      Excellent points. But don’t hold your breath, when the pomp and circumstance of “journals” is involved.

  • Tom Andrews May 21, 2015 at 4:26 pm

    I found out about this issue here. So LaCour has a high level PhD from UCLA, he gets to allegedly lie in a research paper, and he receives rich rewards for his (non)scholarship. Wow. I went to UCLA myself, I became a software developer, and I have had the devil’s time trying to find a job because of the economy. Oh yeah – I made sure to upgrade my skills, too. When I publish my own papers re: R & D – mostly as blog articles – I make damn sure my sample software works as advertised. I always had and have this fear that if I publish apparent / alleged lies the way our friend LaCour APPARENTLY / ALLEGEDLY publishes, I will never have a chance to get a job again. Instead, I follow those old-fashioned rules and I get to slowly starve to death. Based on l’affaire LaCour, maybe I should try the LaCour approach.

    • egrl May 27, 2015 at 1:28 am

      Tom,
      As the editor of Science, McNutt, said, “Right now he’s going to have such a black cloud over his head that it’s going to haunt him for the rest of his days.”

  • Name May 21, 2015 at 5:28 pm

    Journals can publish work whose sole purpose is to discredit possibly falsified data. A great example is the publication by Janssen et al (at Nature) showing that the now-infamous C3b crystal structure by Murthy and colleagues was likely not genuine. So, it can and does happen on occasion:

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2504710/

  • Ken May 22, 2015 at 6:20 am

    I was surprised that the funding of the survey had not been an issue, but according to his CV http://polisci.ucla.edu/sites/default/files/lacour_cv.pdf there was nearly half a million in grants. Pretty impressive for a grad student. Then I looked at the Hewlett Foundation for the $50,000 grant using gay for a search term http://www.hewlett.org/grants/search?order=field_date_of_award&sort=desc&keywords=gay&year=&term_node_tid_depth_1=All&program_id=All and didn’t find it. The other foundations don’t have recent data so can’t be checked, but it would be worth asking if any of them had actually funded the work.

    One thing that becomes obvious to any statistician is that the moment people start being evasive about where there data has come from, it usually means something unethical.

    • The Watcher May 23, 2015 at 4:29 am

      I agree with this point, I would assume the huge amount of funding this graduate student has received was not transferred in to his personal bank account, but instead into Unversity coffers. So there must very simple evidence that the funds were used to pay the external commercial survey company to do the survey of 11K+ people in the snowball door to door survey. This would have taken a huge amount of time and a very large people resource, Qualtrics would also easily also show evidence of what they had done for the research paper.

  • scotus May 22, 2015 at 6:58 am

    It is very likely that the Columbia IRB would have determined that a post hoc analysis of de identified data was exempt from their oversight. However, this would have required an application which could have presumably been done most simply by reformatting the UCLA IRB protocol. The main reason for IRB oversight at UCLA is that the original data set was supposed to have identified the respondents to the survey including information about their sexual orientation.
    Of course the irony of all this is that there is no need to protect imaginary subjects of a made up trial..

  • scotus May 22, 2015 at 7:02 am

    Did anyone spot this update on Lacour’s web site? Apparently he “stands by” his findings which seems pretty mind boggling in light of the fairly comprehensive takedown and the Green retraction request.

    Statement: I will supply a definitive response on or before May 29, 2015. I appreciate your patience, as I gather evidence and relevant information, since I only became aware of the allegations about my work on the evening of May 19, 2015, when the not peer-reviewed comments in “Irregularities in LaCour (2014),” were posted publicly online.

    I must note, however, that despite what many have printed, Science has not published a retraction of my article with Professor Green. I sent a statement to Science Editor McNutt this evening, providing information as to why I stand by the findings in LaCour & Green (2014). I’ve requested that if Science editor McNutt publishes Professor’s Green’s retraction request, she publish my statement with it.

    • Conrad Seitz MD May 23, 2015 at 10:59 am

      Yes, it takes a lot of nerve to “stand by” your findings when someone has discovered that the company you allegedly used to conduct the survey says it has no employee who goes by the name you stated in the paper… and they never received the half million dollars it cost to do the survey… I’d like to see how he explains that.
      Then, it does take a lot of nerve to fake a study in the first place, especially when your PhD and the rest of your career depend on it.

  • scotus May 22, 2015 at 7:28 am

    Also, no mention of a retraction, the Green letter to science and Lacour’s statement in the May 22 issue of Science, at least as published online today.

  • lester May 22, 2015 at 7:37 am

    The central issue is not: do people support gay marriage in a poll, scientific or not. The question is: should the choice over the definition of marriage be left up to judges or the people of the individual states.

    Science in the USA is no longer science it has become political. And political decisions are being forced on the people by other people seeking judges that agree with their political opinions.

    • Tony Mitchell May 22, 2015 at 10:33 am

      Lester, two immediate thoughts. First, you are entirely correct when you state that this topic falls under the nature of politics more than it does science.

      But any political discussion requires an informed populace and so opinion polls are a necessity in the discussion. It becomes a problem when we manipulate the data for our own purposes or when the sampling technique is faulty (case in point – the polls that declared Thomas Dewey defeated Harry Truman in the 1948 Presidential election; the poll was taken via telephone calls at a time when the number of individuals with telephones was limited and predominately Republican. Those who voted for Truman were not polled.)

      The outcome of this issue is that it will poison the water for those who try to conduct future research on the mind-set of the populace (consider the comments of one potential Republican candidate for President who implied a certain degree of arrogance by those who accept the polls concerning climate change).

    • herr doktor bimler May 23, 2015 at 6:48 pm

      That may be your central issue, but it it is peripheral to the study itself, about how easily people will change their minds and their opinion-poll responses.

  • Chester White May 22, 2015 at 4:17 pm

    My wife is a research scientist. It is a lifelong ambition to get a SCIENCE paper, and Green got one without even partially knowing the data that went into it.

  • Tom May 23, 2015 at 9:46 am

    As long as the “study” presents the correct political bias, facts are irrelevant. This is the elephant in the room when it comes to academia.

    • Conrad Seitz MD May 23, 2015 at 11:09 am

      Since we’re in a “scientific” blog here, that statement requires some evidence to back it up. Please at least cite an example of a study with the “correct political bias” since according to your thesis, there aren’t any studies with the “incorrect” political bias.
      Note that I’m not stating that your thesis is incorrect, I’m asking for data to back it up. So that I can argue with my cubicle-mates that what they’re doing is reinforcing political correctness instead of doing real science (which some of them already argue.)

    • herr doktor bimler May 23, 2015 at 6:53 pm

      As long as the “study” presents the correct political bias

      The bias presented in the paper was the idea that people are relatively susceptible to changing their minds on an issue when confronted with personal testimony. This is an optimistic view (in light of all the other work on our tolerance for cognitive dissonance and our obduracy in the face of evidence), and that optimism might well have assuaged suspicions and smoothed the paper’s path to publication.

      But why is it a specifically *political* bias?

  • Christian Scientist May 23, 2015 at 10:37 am

    I am trained scientist who also have strong Christian convictions in the tradition of luminaries like James Maxwell or Allan Sandage. I must admit that I didn’t have strong opinions on the issue of gay marriage one way or another when the issue began to be discussed in the media in the aftermath of the Massachusetts case.

    As I have become more mature and thought about having children of my own, I began to see the proposition of actively denying children of a mother or a father preposterous and antithetical to what I understood to be the way nature shows children should be raised (regardless of one’s religious beliefs). Still, the publication of studies like this one made me be cautious not to share my opinions with colleagues, specially after what happened to Brendan Eich.

    Glad to see the retraction, but I think that those who care about ethics in the scientific enterprise should go farther, affirming the right of people to hold to whatever views they have on marriage (or not marriage) without having to be labeled bigots.

    In the end, shutting down public debate doesn’t resolve matters, it creates resentment. Those of us who believe that gay marriage is an abomination exist everywhere, only we don’t talk about it out of fear of the people who brought down the career of Brendan Eich. That cannot be good for a thriving intellectual community.

    • Ryan May 23, 2015 at 8:13 pm

      Same-sex marriage is not the same debate as raising children in a same-sex household — you don’t need to be married to do that.

    • A. P. May 24, 2015 at 1:25 am

      Well, if we want to bring nature into it, the vast majority of species on earth do not have a male and female equally raise offspring, let alone a “mother” and “father”. Personally, I managed to become the first doctor in my family despite being raised by a single parent, so I don’t see what’s so horrible about being raised by two loving parents of the same sex/gender.

      This study was not particularly pro-gay, and the use of its retraction as a point of action by the far right speaks to the nonsense of the far right more than to anything about the LGBT movement.

      The idea that people can have any view they wish without being labeled bigots is hilarious. I don’t care if it creates resentment. Nobody cares about your feelings. The fact is that gay people have always existed and will continue to exist, and permitting them marriage will not harm anything that exists now but may improve some inequalities that are present for currently existing gay people, as well as children who are cared for by gay people. If your opinion is so horrible that you fear for your career, maybe you should rethink your opinion.

      • Christian Scientist May 24, 2015 at 10:53 am

        I think that this sentence of yours “if your opinion is so horrible that you fear for your career, maybe you should rethink your opinion” summarizes the sad state of affairs in which we found ourselves.

        I am not going to change my opinion, because I do believe that children do best when everything else being equal (ie, we are not talking about dysfunctional environments here) they are raised by their biological mother and biological father. Everything is suboptimal.

        You did well, good for you. I was raised by both my biological parents – together with my siblings- and became the first guy in the family to get a doctorate from an Ivy League school. In the process I also got the love and support of my biological mother and father.

        So very sad, really, that because people like you we are forced to shut up.

        You say that gay people exist, sure. We, the people who defend marriage as a union between one man and one woman also exist, everywhere and in every place. Shutting down debate will not make us go away either, we will just self segregate with like minded people, which is what has already happened in my own city.

        • Ryan May 24, 2015 at 11:53 am

          It’s a bit like lamenting that because you harbour white supremacist beliefs your career might suffer if you publicly share them. Some opinions just aren’t very popular — no getting around it.

          You complain about shutting down debate, yet the amusing thing is that the very topic you’re debating revolves around shutting down peoples’ rights to live the lives they wish to. If it’s so detrimental to human life, natural selection will soon establish that.

          Who are you or anybody else to determine in advance how people should be able to live? That is the biggest problem with bigotry (sorry if you don’t like the word but it has a definition and isn’t just a slur): an unwillingness to permit those who are different to live freely. If the choice is robbing people of their freedom, or people like yourself self-segregating themselves into minority populations, I know which I’d choose.

          • Christian Scientist May 24, 2015 at 2:30 pm

            It is not like white supremacy (a reference that I find offensive because I am not white). In California, the majority decided to keep the definition of marriage that had existed in the state since it was founded: one man and one woman. Prop 8, that amended the state’s constitution, was passed in 2008 to correct the abuses of an activist California Supreme Court that had invalidated our first vote in the matter: prop 22 (which didn’t have the status of constitutional amendment) passed in 2000 by the same voters that voted in favor of Al Gore. The same majority of Californians that voted overwhelmingly for Obama in 2008, voted in favor of Prop 8 (although with a lower majority). In Santa Clara County, Prop 8 got 46% of the vote. However, the county is home to a sizable number of Asian immigrants that cannot vote (25% according to some estimates) that also tend to be conservative socially even though they vote Democratic when they become citizens. I am pretty sure that a majority of Santa Clara County people (not voters) would support keeping marriage as a union between one man and one woman. The Brendan Eich affair reminded all of us that technical excellence alone is not anymore the only thing that matters in this Valley which is supposed to be a meritocracy. His critics were unable to find a single instance in which he had discriminated anyone (LGBT or not) in his job. Still, he got fired. We got the message and we shut up at work about these matters.

          • Christian Scientist May 24, 2015 at 2:36 pm

            By the way, we do not want to tell anybody how to live their private lives. In California, two consenting adults who happen to be brothers or a mother and a daughter cannot marry even as gay marriage is legal. Are the people who support gay marriage telling them how to live their lives?

            Marriage is not about telling people how to live their private lives, it is about what kinds of unions the state decides to promote. The sexual complementarity of men and women makes keeping marriage as a union between one man and one woman very rational. In fact, that is the only type of union (with variations as to the number, age and blood ties of the people who can take part on it) that existed in human history until the Netherlands decided otherwise in 2000.

          • Ryan May 24, 2015 at 3:11 pm

            Christian Scientist
            By the way, we do not want to tell anybody how to live their private lives. In California, two consenting adults who happen to be brothers or a mother and a daughter cannot marry even as gay marriage is legal. Are the people who support gay marriage telling them how to live their lives?
            Marriage is not about telling people how to live their private lives, it is about what kinds of unions the state decides to promote. The sexual complementarity of men and women makes keeping marriage as a union between one man and one woman very rational. In fact, that is the only type of union (with variations as to the number, age and blood ties of the people who can take part on it) that existed in human history until the Netherlands decided otherwise in 2000.

            All right, you’ve convinced me. I can now definitely see that the fabric of society will break down if we allow people of the same sex to marry or — god forbid — raise children. Two fathers or mothers is incomparably worse than any alternative, even being raised by a single parent or no parent at all is preferable. Thank you for your very rational insights.

            In other news, my neighbours in Ireland just overwhelmingly voted in same-sex marriage. An excellent national decision.

          • Marco May 24, 2015 at 3:58 pm

            Brendan Eich did not get fired, he resigned. People criticized his support for Prop 8, which is their good right. Some others called for a boycot of Mozilla; again, their good right. Eich decided he could not be an effective leader and thus stepped down; his good right. He was not fired, plain and simple.

          • Salamano May 24, 2015 at 7:21 pm

            Whatever anyone’s feelings on this overall subject, Eich’s situation was not a matter of a simple choice to resign, no matter what you think it looked like on paper.

      • Christian Scientist May 24, 2015 at 10:55 am

        Here is an example of self segregation http://www.sfgate.com/bayarea/article/S-F-area-s-gay-population-tops-survey-San-6148970.php . The North Bay is inhabited by people like you, while the South Bay is inhabited by people like me.

        • Nicole May 30, 2015 at 2:15 am

          1. This problems with this study had nothing to do with the gay marriage debate. Had the study been valid, the methodology would have been adopted by people on both sides of this and other, unrelated issues.

          2. You say you are a scientist, but base your opinions on your “feelings” about how children should be raised. Most studies show children raised by same-sex couple are as well adjusted as those raised by opposite sex couples.

          3. The South Bay is one of the most liberal areas of the country, on gay marriage and almost every other issue (both in polls of voters and in polls of residents.) As you pointed out, Prop. 8 failed in Santa Clara County when the rest of California passed it. (And Prop 8 would never pass in California today. Like the country as a whole, Californians are much more accepting of gay marriage now. In fact, it passed in 2008 only due to a massive ad campaign, financed by out-of-state interests, which made absurd claims such as: grade-school children would be taught explicit gay sexual content in state sex-ed classes if Califonians voted against 8. No one would believe those lies in 2015.)

    • Neuroskeptic May 26, 2015 at 4:19 am

      But you are talking about it.

    • Conrad Seitz MD May 27, 2015 at 3:05 pm

      What was it that Galileo supposedly muttered after being forced to shut up by the Inquisition? “And yet it moves.”

  • Felipe G. Nievinski May 23, 2015 at 6:51 pm

    Would open data policies have helped prevent data fabrications such as in this case (presumably)?

  • Terri Angelo May 24, 2015 at 6:09 pm

    Question: Was it only the surveys of the persistence of the change in attitude that were allegedly faked? As near as I can tell from the reports I have read, the initial change in attitude reported by the canvassers was valid. Or was there no significant change in attitude?

    • herr doktor bimler May 26, 2015 at 11:13 pm

      Part of the problem is that the survey company have no record of conducting the surveys, and the funding bodies have no record of giving LaCour the money to conduct them.

  • M.E. Tobin May 26, 2015 at 9:53 pm

    How does an exercise social science number-crunching end up being published in Science anyway?

  • M.E. Tobin May 27, 2015 at 2:14 pm

    Lacour co-authored an earlier paper with Aronow, “Double Sampling with Nonignorable Missing…” Has anyone looked into that research, especially given Aronow’s participation in the refutation of the Science piece?

    • Marco May 27, 2015 at 3:22 pm

      Actually, that paper contains one case with actual data: that of the retracted Science paper!

  • SL Wood June 8, 2015 at 8:41 pm

    Not clear why Dr Green is getting such a free pass on criticism. Sure he acted quickly when he was alerted to the data irregularities but why was he an author in the first place? From what I have understood Lacour approached him with a finished study and analysis? Is this correct? If so, in what way did he qualify for authorship in the first place???

    • G. Bowser June 9, 2015 at 7:04 pm

      Often established scholars like Dr. Green provide additional support and guidance, copy editing, as well as help pilot it through peer review. I’m not sure if Dr. Green provided similar support or simply signed off to allow his name to appear on it. Vox put it best: there’s no way this shows up in Science without Dr. Green in the byline.

      That said, Dr. Green should have been much more resolute in the matter. These data represented a seismic shift in the understanding of how people form political opinions. To me, that seems well worth the headaches of jumping through IRB’s hoops.

      On the other hand, as a early career PhD myself, it’s hard to fathom that Mr. LaCour would be so cavalier about his future and that of his peers. This damages the credibility of academia and the social sciences in particular. As it stands, Mr. LaCour’s actions and Dr. Green’s carelessness provided a sledgehammer to those wanting to topple the ivory tower.

    • Felipe G. Nievinski June 9, 2015 at 7:54 pm

      “there’s no way this shows up in Science without Dr. Green in the byline” — so much for editorial review.

  • Dave Burton June 9, 2015 at 10:53 pm

    Nicole
    1. This problems with this study had nothing to do with the gay marriage debate. Had the study been valid, the methodology would have been adopted by people on both sides of this and other, unrelated issues.
    2. You say you are a scientist, but base your opinions on your “feelings” about how children should be raised. Most studies show children raised by same-sex couple are as well adjusted as those raised by opposite sex couples.
    3. The South Bay is one of the most liberal areas of the country, on gay marriage and almost every other issue (both in polls of voters and in polls of residents.) As you pointed out, Prop. 8 failed in Santa Clara County when the rest of California passed it. (And Prop 8 would never pass in California today. Like the country as a whole, Californians are much more accepting of gay marriage now. In fact, it passed in 2008 only due to a massive ad campaign, financed by out-of-state interests, which made absurd claims such as: grade-school children would be taught explicit gay sexual content in state sex-ed classes if Califonians voted against 8. No one would believe those lies in 2015.)

    1. True, mostly.

    2. False.

    3. True.

    Re: #2. By far the highest quality, most comprehensive study to date of this issue was the Regnerus New Families Structure study:
    http://dx.doi.org/10.1016/j.ssresearch.2012.03.009

    Key findings from the study were that (the rest of this is quoted from an article about it), “…compared with off-spring from married, intact mother/father homes, children raised in same-sex homes are markedly more likely to:

    Experience poor educational attainment
    Report overall lower levels of happiness, mental and physical health.
    Have impulsive behavior
    Be in counseling or mental health therapy (2xs)
    Suffer from depression (by large margins)
    Have recently thought of suicide (significantly)
    Identify as bisexual, lesbian or gay
    Have male on male or female on female sex partners (dramatically higher)
    Currently be in a same-sex romantic relationship (2x to 3x more likely)
    Be asexual (females with lesbian parents)
    As adults, be unmarried; much more likely to cohabit
    As adults, more likely to be unfaithful in married or cohabiting relationships
    Have a sexually tramsmitted infection (STI)
    Be sexually molested (both inappropriate touching and forced sexual act)
    Feel relationally isolated from bio-mother and -father (Although lesbian-parented children do feel close to their bio-mom – not surprisingly – they are not as close as children with a bio-mom married to father)
    Be unemployed or part-time employed as young adults
    As adults, currently be on public assistance or sometime in their childhood
    Live in homes with lower income levels
    Drink with intention of getting drunk
    Smoke tobacco and marijuana
    Spend more time watching TV
    Have frequency of arrests
    Have pled guilty to minor legal offense

    Editor’s note: Citing the Regnerus study without mentioning the severe criticism of it would be incomplete: http://www.washingtonpost.com/news/volokh-conspiracy/wp/2015/05/10/new-criticism-of-regnerus-study-on-parenting-study/

    • Dave Burton June 9, 2015 at 11:01 pm

      Clarification: w/r/t #3, what I mean to agree with is Nicole’s statement hat the South Bay is a very liberal area. That is true. But her digression which followed, it which she accused Prop. 8 supporters of “lies,” was incorrect. The campaign for Prop 8 was not at all dishonest; what was dishonest was the claims by Prop 8 opponents that the supporters were dishonest.

    • Marco June 10, 2015 at 6:29 am

      I believe this will be a more appropriate response (scientifically speaking) than a WP article:
      http://dx.doi.org/10.1016/j.ssresearch.2015.04.005

      This study showed that Dave Burton’s “children raised in same-sex homes are markedly more likely to” is simply not supported by the evidence, because Regnerus did not even identify whether these children were actually being raised in a same-sex home!

    • Dave Burton June 10, 2015 at 10:30 am

      The December 15, 2012 Science News has a feature article entitled “Families in Flux,” here:
      http://www.sciencenews.org/view/feature/id/346695/description/Families_in_Flux

      The full article is paywalled, but the first page is here:
      http://onlinelibrary.wiley.com/doi/10.1002/scin.5591821221/abstract
      and the full text, sans illustrations, is here:
      http://tinyurl.com/IAEAFamFlux

      Here is what the article has to say re same-sex parenting:

      Even the most steadfast same-sex relationships ignite controversy when kids enter the picture. About 100,000 gay couples in the United States are raising children amid a fierce national debate about what effects, if any, such family arrangements have on child development.

      A 2005 brief issued by the American Psychological Association concluded that children raised by same-sex parents do just as well socially and emotionally as kids with a mom and a dad at home.

      But the 59 studies referenced in the brief contain too many flaws to say anything about same-sex parenthood, concluded family studies professor Loren Marks of Louisiana State University in Baton Rouge in the July [2012] Social Science Research. Most of the research cited by the APA consisted of small numbers of affluent, self-selected gay parents who weren’t compared with heterosexual parents of similar backgrounds, he says. Among other problems, measures of children’s well-being varied from one study to another and few investigations tracked kids beyond childhood.

      A study aimed at providing a clearer look at the adjustment of kids with same-sex parents, appearing in the same issue of Social Science Research, set off its own wave of disputes. Sociologist Mark Regnerus of the University of Texas at Austin analyzed data from a nationally representative sample of nearly 3,000 people, ages 18 to 39, surveyed in 2011 and early 2012.

      More than 600,000 couples identified themselves as same-sex in the 2010 U.S. census, more than five per 1,000 households. About 17 percent of these couples were raising children belonging to one of the spouses (by birth, marriage or adoption).
      Credit: Gary Gates and Abigail Cooke/The Williams Institute/UCLA
      Regnerus compared 163 individuals who reported that their mothers had a past or current gay partner with 919 participants who grew up with a biological mother and a father. Those from families headed by a gay mother displayed poorer mental and physical health, were more often unemployed and reported more problems in their current romantic relationships. A separate group of 73 participants had gay fathers, but that was too few to make any confident assertions.

      Regnerus acknowledges that the results do not show that a gay mother’s sexual orientation or sexual behavior causes emotional and social problems in her children. The findings could reflect the fact that many gay mothers had previously been married to men and gotten divorced, so their children had contended with parental breakups while kids with opposite-sex parents often had divorce-free upbringings.

      What’s more, social stigma about gay parenthood might have made the wonder years even tougher for kids with lesbian mothers.

      Regnerus’ study swiftly came under attack. A letter published in the November Social Science Research, signed by 200 social scientists and scholars, criticized the work as muddled and misleading. Aside from the apples-and-oranges comparison of people with previously divorced gay parents and offspring of never-divorced heterosexual parents, Regnerus didn’t confirm parents’ sexual orientation, the letter asserts. As a result, it is unknown how many parents in Regnerus’ same- and opposite-sex households were bisexual and whether that has made a difference for the children.

      Better investigations are in the works, says Allen, of Simon Fraser. Small numbers of same-sex parents and the many paths that bring children into families with two men or two women as parents present big research challenges. Allen is currently analyzing census data on Canadian families, including about 1,400 with same-sex parents.

      “We’ve got a long way to go before we can say whether children are better off, the same or worse off in same-sex families compared to intact biological families,” Allen says.

      When Prof. Allen completed his study, his results largely confirmed Regnerus’ results. There’s an article about it in OneNewsNow, here:
      http://onenewsnow.com//culture/2013/01/06/the-truth-about-children-with-gay-parents

      Summary: Canadian economist Douglas Allen and two American economists reanalyzed the data used by sociologist Michael J. Rosenfeld, and published their findings in the journal Demography. They found that children raised by same-sex couples do worse in school than children raised by heterosexual couples, even cohabitating heterosexual couples, and much worse than children raised by married heterosexual parents.

      The OneNewsNow article reported that the Allen paper in Demography is paywalled, but for a while it was available without a paywall, here:
      http://link.springer.com/article/10.1007/s13524-012-0169-x/fulltext.html
      or http://link.springer.com/content/pdf/10.1007%2Fs13524-012-0169-x
      The paywall is back, but there’s a copy of the paper on Prof. Allen’s web site, here:
      http://www.sfu.ca/~allen/RosenfeldCommentDemography.pdf

      Allen et al reported that even “children who are being raised by a heterosexual cohabiting couple are about 15% more likely to be making normal progress in school than children being raised by same-sex couples,” and concluded that, “Together, these findings are strikingly different from those of the original study—and the differences are large enough to be noteworthy. With respect to normal school progress, children residing in same-sex households can be distinguished statistically from those in traditional married homes and in heterosexual cohabiting households. The magnitude of the differences is large enough to be relevant for current and future policy debates…”

      Demography also published a defense by Rosenfeld of his conclusions, here:
      http://link.springer.com/article/10.1007/s13524-012-0170-4/fulltext.html
      or http://link.springer.com/content/pdf/10.1007%2Fs13524-012-0170-4

      That one is paywalled, too, but the gist of Rosenfeld’s defense was that, “Allen et al. reached the conclusion that children in same-sex-couple families fare worse in school by including all children regardless of how long the child has lived with the family… and by including adopted and foster children along with the head of household’s own children…”

      In other words, according to Rosenfeld, the lack of family stability in homosexual households accounts for the poorer outcomes of their children. Since he could not admit that such instability is intrinsic to disordered relationships, he blamed the law, and in his concluding sentence he recommended legalizing same-sex marriage: “perhaps the logical public policy prescription would be to extend marriage rights to same-sex couples in the United States.”

      Prof. Allen also weighed in thoughtfully on the Regnerus study, June, 2012, here:
      http://web.archive.org/web/20141112075327/http://www.nationalreview.com/articles/302749/regnerus-debate-douglas-w-allen

      • Marco June 10, 2015 at 2:44 pm

        Dave, this ignores the much newer study for which I provided a link. Linking to the coverage by the AFA (onenewsnow.com is their news outlet) does not strengthen your argument.

        • Dave Burton June 12, 2015 at 7:09 pm

          Marco, that paper is paywalled, but from the abstract it appears that it ignored the Allen study.

          • Marco June 13, 2015 at 3:17 am

            The abstract tells you nothing about what it did or did not say about the Allen study (who himself has openly stated that homosexuals will go to hell unless they repent, so excuse me for not being quite convinced that his study was done any better than that of Regnerus).

          • Dave Burton June 13, 2015 at 5:17 pm

            Marco, I know nothing about Allen’s religious views, and don’t think they are relevant to this discussion. Before doing his own study he wrote, “We’ve got a long way to go before we can say whether children are better off, the same or worse off in same-sex families compared to intact biological families.” That sounds like someone with an open mind, to me. I don’t think it is reasonable to dismiss his conclusions just because you don’t like his religion. (Likewise, you shouldn’t dismiss articles from OneNewsNow, either.)

            Do you have that paywalled paper? Will you send it to me, please? http://tinyurl.com/dabmail

  • Post a comment

    Threaded commenting powered by interconnect/it code.