1. Guest, the 'News in Brief' for the week beginning 12th April 2021 is here.
    Dismiss Notice
  2. Welcome! To read the Core Purpose and Values of our forum, click here.
    Dismiss Notice

The Black Box of Meta-Analysis

Discussion in 'Health News and Research unrelated to ME/CFS' started by Sly Saint, Aug 28, 2019.

  1. Sly Saint

    Sly Saint Senior Member (Voting Rights)

    Messages:
    5,953
    Likes Received:
    48,903
    Location:
    UK
    (was going to post this on another thread I set up, but the title to that thread has been changed so am posting this separately)
    full title
    The Black Box of Meta-Analysis: Personality Change
    August 11, 2019

    https://replicationindex.com/2019/08/11/the-black-box-of-meta-analysis-personality-change/
     
    Peter Trewhitt, JaneL, Wonko and 12 others like this.
  2. rvallee

    rvallee Senior Member (Voting Rights)

    Messages:
    6,001
    Likes Received:
    48,601
    Location:
    Canada
    The analyses of the Cochrane reviews by Mark Vink show how completely broken the process of meta-analysis can be, perfect example of garbage-in-garbage-out when biased reviewers are allowed exemptions from norms. Well, that and the initial peer review that was pretty brutal and still saw the GET review be slapped with a "low bias, high effect" label. All because people didn't give a damn about us and thought it was harmless since "there is no disease", nevermind that this is an opinion and a shitty one at that.

    I have no idea how psychological research got granted so much affirmative action but that it happened this late and after so many harsh lessons about the value of the scientific method it has to rank as one of the top 10 blunders in the whole history of science, one that is filled to the brim with major disasters.

    I'm sick of all this pseudoscience, it's turning out to be as bad for humanity as the major world wars. What a freaking mess.
     
    Peter Trewhitt and JohnM like this.
  3. alex3619

    alex3619 Senior Member (Voting Rights)

    Messages:
    1,503
    Likes Received:
    11,839
    I have long considered that meta-analysis can reinforce problems if there are systemic methodological biases in a field. CBT/GET comes to mind.
     
    Last edited: Aug 29, 2019
  4. alex3619

    alex3619 Senior Member (Voting Rights)

    Messages:
    1,503
    Likes Received:
    11,839
    Precisely.
     
    JohnM likes this.

Share This Page