Thursday, July 22, 2010

Biased Research on Perceptions of Autism

Over at Astrid's Journal, Astrid is talking about a recent study that claimed to look at "lay perceptions" of autism. While she does have a point that there is a divide in the autism community in how autism is portrayed, the study she cites is absolute nonsense.

The "study", and I use that term loosely, gave ten "lay" people (people not involved in the autism world) a "semi-structured" interview about autism and then looked at the data using "interpretative phenomenological analysis" and "discourse analysis". Yes, those are real things.

Based on their analysis, the authors came up with the following conclusion -
Participants demonstrated clear views (although not necessarily correct) about the nature, origins and manifestations of autism. They drew upon their understanding of normal child development to make these assessments and they made judgements about the ability of individuals with autism to attain social independence. It is recommended that interactions between lay people and people with autism may help develop positive conceptualisations of autism.
I have not read the study, nor do I intend to, because by the time you actually read the study (or the conclusion), you have missed the point. Where any discussion of this "study" should have stopped is at the point where they point where you see that the results are based on the opinions of just 10 people.

To be fair, Astrid does note that "the study group was very small (only ten participatins), so no generalizations should be made" but I think that is the wrong way to think about the "study". Ten random people cannot be representative of anything, especially a population as large as the "lay" autism population (think billions).

So, the question is then what are these ten peoples' opinions being used to represent? Or to borrow a line from the person who writes The Last Psychiatrist, what do the authors want to be true?

To help answer that question, I would direct your attention back to an earlier "study" written by the same two people that looks suspiciously familiar. In this "study" the authors gave a "semi-structured" interview to 9 young people with high functioning autism who were "capable of providing a verbal account of their perceptions of autism and diagnosis experiences" and looked at the data using "interpretative phenomenological analysis".

Notice the setup. Another infinitesimally small group whose opinions were analyzed and taken to mean something. The conclusion from this "study"?
In relation to the existing literature, it is suggested that the effects of diagnosis, or disclosure of diagnosis, from the perspective of the person with autism be given greater consideration.
So, what do the authors want to be true? They want to people with autism (presumably the high functioning ones) be given a greater voice in describing their autism and "lay" people to interact with them so that they can develop a positive attitude towards autism.

OK, fair enough, people with autism are people just like everyone else and have the right to have their voices heard. But what do we do with the vast number of children with autism who aren't able to communicate well enough to provide their perspectives of autism? Do we let the young, high functioning adults give a perspective for them?

Should we keep on ignoring their needs while singing kumbaya and thinking happy thoughts about autism?

The real questions are how did this nonsense get published in the first place and how did the authors get away with suggesting that they are talking about the other people's perceptions of autism. If they wanted to voice their opinions about autism, they should write articles that state that these are their opinions and publish them as such.

We don't need more junk pretending to be science.

References (and I use that term loosely)

Huws JC, Jones RS. Diagnosis, disclosure, and having autism: an interpretative phenomenological analysis of the perceptions of young people with autism. J Intellect Dev Disabil. 2008 Jun;33(2):99-107. PubMed PMID: 18569397. DOI: 10.1080/13668250802010394

Huws JC, Jones RS. 'They just seem to live their lives in their own little world': lay perceptions of autism . Disability & Society, 25:3:331-344. DOI: 10.1080/09687591003701231


  1. Well Mr Jabberwock,

    "I have not read the study, nor do I intend to"

    and then you go on to say

    "The real questions are how did this nonsense get published in the first place"

    How do you know it's nonsense if you haven't read it? One cannot have it all ways - one does not confutate by refusing to engage with one's subject matter.

    "Where any discussion of this "study" should have stopped is at the point where they point where you see that the results are based on the opinions of just 10 people."

    How do you know it's a "study" and not a study, if you haven't read it?

    Where any reading of your essay should've stopped is at the statement "I have not read the study, nor do I intend to".

    Nevertheless I will continue... ...

    If a study of 10 people dosed with cyanide claims that 100% of the participants suffered adverse effects - is it useless?

    Off the top of my head, Leo Kanner's 1943 paper "Autistic Disturbances of Affective Contact" related the case histories of 6 children. Does that make it useless?

    The authors of the paper may well be propogandizing, but then it seems, so are you.

    Pot, Kettle, Black.


  2. I think you may have missed the point of what I was saying. I am not saying that all studies of only 10 people are worthless.

    What I am saying that a "study" about perceptions that only includes the opinions of 10 people is worthless.

    Think of it this way, if you look at polls or surveys that are done for a variety of topics, such as politics, how much weight would you give to a poll that only asked ten people?

    I think that an opinion poll with only ten people is almost completely worthless and has such as high margin of error that you can't draw any conclusions from it.

    Now, that is a poll, this is a study that is supposed to be grounded in evidence based medicine. The bar for evidence in a study is much higher that is it for a simple poll.

    As the current "study" doesn't even meet the threshold of evidence for a simple opinion poll let alone the higher standards of evidence based medicine, how can we take it to be something other than just the authors opinions masquerading as science?

  3. Given that neither of us has read the paper, further discussion of it would be illegitimate from a Logician's perspective and indeed mine.

  4. Uhm, OK.

    I never thought of my opinion as being illegitimate before, but whatever.

  5. I would certainly not want to inhibit you expressing your opinion. It's just I'm working on a post inspired by the events you have just posted on and don't really have the time to do the necessary background reading/thinking to have a sensible discussion.

    I'm sure there will be plenty of opportunities for us to lock horns in the future.


  6. So how does you not having time to put into a discussion (which I can completely understand), make a discussion "illegitimate"?

    Because I was thinking that the phrase "strategic retreat" might have been more appropriate.

  7. Do not disseminate:

    You have not read the paper.
    I have not read the paper.


    Discussion of the paper is futile.

    I am British, we retreat by claiming the Potomac is useless for trade, anyway.