1. Patients launch $1.27 million crowdfunding campaign for ME/CFS gut microbiome study.
    Check out the website, Facebook and Twitter. Join in donate and spread the word!
ME/CFS and the Magic of the Canine Factor
There's been plenty of research indicating that having pets is good for your health. I never really noticed any particular benefits to having cats, though that may have had more to do with my cats. They've been fairly indifferent to my presence and we've shared a live-and-let-live...
Discuss the article on the Forums.

"The British amateur who debunked the mathematics of happiness"

Discussion in 'Other Health News and Research' started by Tom Kindlon, Jan 19, 2014.

  1. Tom Kindlon

    Tom Kindlon Senior Member

    Messages:
    254
    Likes:
    785
    http://www.theguardian.com/science/2014/jan/19/mathematics-of-happiness-debunked-nick-brown

    [..]

    On whistle-blowing:

    On the specific problem:
    This is why replication is so important. Also, sometimes one can test within the data e.g. dividing up the data using a training set.


    Another reason why one needs outsiders as whistle-blowers:
  2. Bob

    Bob

    Messages:
    8,044
    Likes:
    10,275
    England, UK
    http://www.theguardian.com/science/2014/jan/19/mathematics-of-happiness-debunked-nick-brown
    snowathlete, Simon, Valentijn and 3 others like this.
  3. alex3619

    alex3619 Senior Member

    Messages:
    7,181
    Likes:
    11,237
    Logan, Queensland, Australia
    Do read the comments, especially the highly recommended ones..
  4. SOC

    SOC Moderator and Senior Member

    Messages:
    5,289
    Likes:
    6,276
    USA
    Sadly, this type of pseudo-mathematical claptrap in not occasional, but common. As a graduate student in engineering, I was horrified to find our psychology faculty and grad students using a simple statistical package with so little understanding of the statistics or software that they were making huge errors just in the input. They didn't even bother to read the manuals; they just asked the (totally ignorant) person at the next desk. Their understanding of the output was even worse, so the conclusions they drew were beyond outrageous. What they called evidence was simply a lot of Garbage Out. It made me sick.

    And just for giggles -- I had business dealings with the psychology academic (yes one of those faculty members at my U) who was called "a leading authority in the psychology of successful relationships". The man had the social/interpersonal skills of a rock and his personal life was not exemplary of any understanding of successful relationships. What a laugh... if it weren't so academically horrifying. He's one of the reasons I am suspicious of the integrity of psychology and psychiatry.
    Last edited: Jan 19, 2014
    Roy S, Little Bluestem, Tito and 6 others like this.
  5. SOC

    SOC Moderator and Senior Member

    Messages:
    5,289
    Likes:
    6,276
    USA
    ETA: For the record -- I ha veknown a couple of psych clinicians who were wonderful people with amazing skills in helping others. One probably saved me from a life of misery. So I don't think they're all worthless. ;) Unfortunately, I've known far, far more charlatans and downright evil people working as therapists, psychologists, and psychiatrists. Sad, really, that the profession doesn't police itself better.
    Last edited: Jan 19, 2014
  6. Roy S

    Roy S former DC ME/CFS lobbyist

    Messages:
    437
    Likes:
    432
    Illinois, USA
    That's a good article. It's a little surprising that even after that they have to go through more hoops to get another response published.

    "After initially being turned down, Brown, Sokal and Friedman went through American Psychologist's lengthy appeals procedure and won the right to reply to Fredrickson's reply. They are currently working on what is certain to be a very carefully considered response."
  7. anciendaze

    anciendaze Senior Member

    Messages:
    870
    Likes:
    933
    Psychology and psychiatry are not the only fields with seriously flawed statistical reasoning. You can find errors in logic all over medical statistics. Convenient assumptions about normal or Gaussian distributions lie behind almost all parametric statistics. When it is necessary to justify the assumption, researchers often appeal to the Central Limit Theorem, though few have gone through proofs of this theorem to see if it applies. There are several preconditions for the theorem, though different proofs may use different sets of these. The broadest ones are that the separate distributions being combined to give the one you want to justify: have well-defined means, bounded variance, and that they combine additively. It is this last "obvious" assumption which I've addressed in a number of posts. There are in fact theorems about combining distributions by multiplication which yield distributions far from "normal". These often fail to have well-defined variance or standard deviation.

    The old joke about normal distributions is that mathematicians think this is an experimental fact, while experimentalists think it is a mathematical guarantee. Both are mistaken.

    I addressed some of these issues in a series of posts here. Just to avoid confusion, I want to explain now that I'm talking about two different ways in which I know things combine by multiplication: probability of survival and efficiency of operation. That second combination likely also applies to multiple defects in systems with redundancy, though I haven't gone through the details in that case.

    As one mathematically sophisticated correspondent said "Ha! I can't even imagine trying to explain something like this to the medical doctors!!" I'm still working on it.
    Snow Leopard, Simon and Valentijn like this.
  8. Simon

    Simon

    Messages:
    1,337
    Likes:
    4,204
    Monmouth, UK
    My two favourites:

    Though as @anciendaze says, psychology and psychiatry are not the only fields with problems with maths and stats, as shown by the excellent blog: 6 Shocking Studies That Prove Science Is Totally Broken | Cracked.com

    Roy S, biophile, alex3619 and 2 others like this.
  9. Little Bluestem

    Little Bluestem Senescent on the Illinois prairie, USA

    Messages:
    2,506
    Likes:
    1,856
    Midwest, USA
    A very lucrative brain-fart.
    Roy S and Valentijn like this.
  10. Firestormm

    Firestormm Guest

    Messages:
    5,824
    Likes:
    5,974
    Cornwall England
    Does CBT represent a 'cookie-jar'? Arguably it does given the widespread use of it across medicine these days and the sub-contracting to those 'specialists' outside of e.g. the NHS.

    She just basically invented her own method. Is that worse than inventing your own data?

    Left me wondering about the use of subjective outcome measures... ;)

    Ah, yes:

    And, 'there may be all manner of cultural and personal reasons why an individual or group' might feel better today than yesterday, less fatigued today than yesterday, or that a particular intervention or even chat with a clinician has 'helped' their overall wellbeing.

    I do think it is important to allow patients input into any measure of outcomes, but only to the extent that such measures are secondary to objective ones. Claimed improvements to 'Wellbeing' or 'Quality of Life' or the 'Helpfulness of a therapy' should encompass much more of an evidence base than self-report alone.

    I am a kid from the 80's and 90's. For me 'the power of positive thinking' was debunked ages ago. It was overhyped and oversold - but the same assumptions still persists in psychology. Too many assumptions and theories and too little actual science.

    Said it before and I'll say it again - the claims made about psychological interventions are too ambitious for the evidence and PACE was no different in this regard. We need a better way of determining 'feeling better' after therapy and determining whether or not and indeed how such an intervention can impact on disability.

    That said, I just had a really nice cup of coffee whilst reading that article from the Guardian at long last and feel quite relaxed and happy.

    I will now be doing some work. BUT am I returning to that work - and not resting or doing something more pleasurable - because I am feeling happy and less fretful about my symptoms? Hmm.... :p
    Last edited: Jan 26, 2014
  11. Snow Leopard

    Snow Leopard Senior Member

    Messages:
    2,297
    Likes:
    1,755
    Australia
    There is nothing wrong with promoting novel analytical methods if they seem to be useful. There is however something wrong with writing a paper using methods you don't understand and having it reviewed by people who don't understand it.
    Little Bluestem, SOC and Valentijn like this.
  12. Roy S

    Roy S former DC ME/CFS lobbyist

    Messages:
    437
    Likes:
    432
    Illinois, USA
    This is from last year --
     
    "Here is the modus operandi of the positivity lady. She goes to a scientific field and picks up the jargons. Then she uses those jargons to write a complex paper, whose conclusion has something to do with human well-being, positive emotions, etc. Neat, isn’t it? Did we say that the technical terms are used in meaningless way to impress the naive reader?"
    "In the meanwhile, positivity lady moved on to new field of research – genomics. Her paper linking positivity with gene expression came out in PNAS a week back. Check the last sentence (emphasis ours), if not anything else."
     
    http://www.homolog.us/blogs/Blog/2013/08/05/tragedy-of-the-day-pnas-got-duped-by-positivity-lady/
    there is also a link to a relevant James Coyne blog:
    http://blogs.plos.org/mindthebrain/...-health-by-pursuing-meaning-versus-happiness/
     
     
     
     
     
     

See more popular forum discussions.

Share This Page