Autism, the frog book, antivaxxers and other bad science

The Frog Book

This blog started last year when “the frog book” was frequently mentioned on Twitter, someone complaining the book was called Tuesday and they got the test on a Wednesday.

Now I have no qualifications in psychology, but I am a bit of a maths whizz, so I thought I’d look at this. The book Tuesday by David Wiesner is a graphic novel with no words apart from Dates and times. What the International Journal of Speech-Language Pathology have to say about it can be found at

Method: Participants were children with ASD (n = 11) and TD [Typically-developing] controls (n = 17). Both groups were aged 9–15 years and were matched for expressive and receptive language skills and non-verbal intelligence. Narratives were analysed for local structure elements (length, fluency, errors, semantics and syntax), cohesion and global elements (story grammar and internal state language).

Do you see the problems here? I see two very big ones.

  • N = 11 is not a significant number for a scientific study. It is fine for the initial study but scientific results need a sample size of several hundred. No conclusions can be drawn from this small sample size. The conclusion of the report, “The present findings provide evidence that children with ASD exhibit subtle story generation impairments,” is meaningless in my opinion. That is mathematics, not psychology, by the way.
  • A sample with a top age of 15 years old, regardless of sample size, does not show any correlation in adults. This book is being used in adult diagnosis and based purely on mathematics, I think it should not be used.

A lot more research is needed before picture only books, such as Tuesday, are used in diagnosis. There are other things not covered in the report: What is the chance of missing a diagnosis? What is the chance of false positives?

The book may be a useful tool for diagnosis of adults and children, or it may not. We need research with a decent sample size: Until we have that it is meaningless.

As for the book, it is a decent graphic novel. I like it, I’m a fan of graphic novels. But as a diagnostic tool based on a sample size of 11, forget it.

Autistic people are more likely to divorce?

A broken egg shell, split through the word marriage

Rubbish, absolute rubbish.

Anyone saying that or that the divorce rate among autistic people is 80% are wrong. The data they are using is 10 years out of date. A study using a sample size of over 2,000 shows that the divorce rate among autistic people is 25%, just as it is in the rest of the population. This is science, and mathematics, done properly.

Kennedy Krieger Institute. “80 percent autism-divorce rate debunked in first-of-its kind scientific study.” ScienceDaily. ScienceDaily, 19 May 2010.

The prevalence and characteristics of mental health crisis in children, adolescents,and young adults with autism spectrum disorder.

Ann Memmott on Twitter October 2019:

Who would like to suggest what’s wrong with this brand new research, on autistic children/young people/young adults and their mental health, average age nearly 14 yrs old?… the paper in question

There are big problems here:

Parents completed the Mental Health Crisis Assessment Scale,” A study which included autistic people up to 25 years old was completed by the parents, usually the mother. The reports say that 99% of responders were the mother. Why not ask the people concerned themselves? Another paper which noted that quite a few parents get it wrong when they guess.…

83% male, 86% Caucasian,” That’s not a very representative group either of the population nor of autistic people. Where are the females? This is a 2019 report, autism as a disorder in boys is no longer sais as significant numbers of females were missed because the present symptoms of autism differently.

Really they need to go back and get a representative sample. There’s nothing wrong with asking parents either, especially where the autistic person has communication difficulties. But there is a need to ask the autistics themselves. Take time to listen to them, it is worth it. One of the autistic people I communicate with on Twitter is non-verbal, unable to speak and takes around 30 minutes to write a 140 character tweet, yet comes across as very intelligent. Lack of communication does not equate with lack of mental capacity, something that is often missed.

Trigger Warning: ABA, ableism, institutionalized child abuse

Here comes the contentious bit: ABA therapy (Applied Behaviour Analysis). This turned up on Twitter on the day before I planned to write this, I like a week for each blog unless I’m doing a train of thought.

Behavior analysts train an autistic adult to stop talking about “music, age, specific dates, times, months, days of the week… numbers larger than 10, mathematical terms… alphabet or alphabetical order, specific names, & colors”? ABA in 2020

Note: in this study, autistic adults (total N=5) were trained to never show any sign of autism (e.g. “distracting nonvocal behavior” “inappropriate laughter” “perseverative utterances”) in a conversation.

N = 5.

Yes FIVE, only five. How can anyone take a sample size of five seriously for anything. But it is supposed to be taken seriously.

For the view of an ex-ABA therapist and why she left after a year, along with a breakdown of why she got out this is a great blog. I got the heading to this section from there, hence the American spelling. The conclusion is that she quit because of evidence from autistic adults that have had ABA therapy as children that the therapy causes a lot of stress in later life because you are concentrating too much on trying to behave like a non-autistic person that it tires you out. ABA may make parenting easier. but it is not in your best interest.

“Abandon all hope, anti-vaxxers who enter here.”

I should not need to say this, but here goes: There is no link between the MMR vaccine and autism. In fact there is no link between any vaccination and autism. Sorry to those who refuse their children the MMR, but there you are.

It starts in the 1998 with a report by Andrew Wakefield. Wakefield said in a report that there was a ling between the MMR vaccine. He was wrong. Very wrong. He ended up being struck off as a doctor because when more accurate evidence came in he refused to stop saying there was a link and because of malpractice. Wakefield was working for the lawyers in a court case where a boy had started showing signs of autism after having the MMR shot, and they were wanting evidence. Wakefield did not mention the conflict of interest in his report.

The report had three other problems:

  1. The small sample size (n = 12)
  2. Uncontrolled design
  3. The speculative nature of the conclusions

Shortly after the paper was published 10 of the co-authors (out of 12) retracted because “no causal link was established between MMR vaccine and autism as the data were insufficient”. Small sample size and no control is OK for a preliminary study, but not to be taken as fact.

The Lancet, who published the Wakefield report found that the authors of the report had done nothing wrong and did not retract it until 2010 when a large scale report with a large sample size in Denmark plus other investigations had found that there was not a link. And investigative journalism in the British Medical Journal had exposed Wakefield’s report as a fraud intended to make money.

I find it is sad that it was journalism and not academic diligence that exposed the fraud . There seems a lack of academic diligence in some autism research still. We need better.

Tell me what you think

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s