Science Fictions: Exposing Fraud, Bias, Negligence And Hype In Science by Stuart Ritchie (book review).
In the introduction to ‘Science Fictions: Exposing Fraud, Bias, Negligence And Hype In Science’, author and psychiatrist Dr. Stuart Ritchie points out that he isn’t anti-science, just against the process and faulty procedure of getting valid papers into print, often without verification that the evidence gathered is true or checked properly. Please be aware, science is a large subject covering not only physics, chemistry and biology but a whole list of other sciences and discoveries as well.
I should also point out that the emphasis of at the beginning of this book is not in the hard sciences but the soft science of psychiatric experiments where there is often a failure for others to repeat the tests and ensure similar results are given. Likewise, the experimenters also not noting everything that they did and whether their influenced the people being experimented on. The biases Ritchie points out that happened are, quite frankly appalling, more so as the revised results after being tested by other qualified people gave different answers and not widely distributed to the public. In other words, you are less likely to read about the retractions.
The experiment where students becoming abusive guards to the student prisoners is noted as being one of these by the way. I’m surprised no one was stripped of their degree in all of this because there certainly needs some sort of professional punishment and a reminder to others to follow procedure for evidence than making adjustments to suit their theories. For us SF writers who rely on up-to-date scientific information, this is also a sharp reminder to check your source material to ensure it is correct before applying it in our stories.
When it comes to outright fraud, this is a book that everyone should read. Not because all scientists are frauds but more because con-men and women see their honesty as easy marks. Throughout the many examples given here, especially in medical research, it does appear that everything is given the benefit of the doubt until whistleblowers and anyone checking the information spots something isn’t right. 40% of them are actually honest mistakes but, looking at the evidence here, things like photo-flopping (reversing a photo to look different) should have been spotted at the editorial reading phase.
It’s hardly surprising whistleblowers are frowned upon when university, institute or business reputation is put at stake than scientific accuracy. I take Ritchie’s comment that you can’t expect scientists to put their own research on hold to do such work but a rotating unit using specialists should reduce having con-artists among them and include professional magicians akin to James Randi who although aren’t scientists do have the ability to pay attention to such detail and how cons work to balance things out.
Scientists want to have the results of their research out first to ensure their name is associated but you would think if everyone is on hold and subject to copyright, protection should be offered until proven verification was given. You only need to think of the MMR false data and the number of children who died by not taking the treble vaccine to realise that something procedurally should been in the way. There is but it is explained further into the book. Oddly, with the current covid-19 crisis, I think more people are going to see how correct procedure is carried out than ever before and will certainly be suspicious of ‘instant’ inoculations without evidence that it offers protection for at least 90% of those inoculated.
Oddly, the one thing Ritchie neglects in the choice of research in his ‘Bias’ chapter and having a no evidence result is whoever is paying will deem a waste of money and want results. Therefore the scientist will select something either safe to get a result or find some information that shows the work was worthwhile. No wonder they either become tempted to manipulate probabilities to favour a particular result or ignore results that don’t fit their theories.
With the ‘Negligence’ chapter, the focus is on mistakes, especially in maths, that sees print. You would think the calculations would be more than double-checked and done independently. Back when I worked in science and computers were still coming up to speed, I was asked to find and use an obscure formula which I found in an encyclopedia of mathematics that gave examples so I wouldn’t make any mistakes. However, it turned out, the mathematical professor author had made the mistakes and agreed and apologised when I sent the proof to him. Mistakes can happen but checking isn’t always done.
Ritchie does point out that there is now an algorithm program called ‘statchek’ which will can scan for common mistakes and such and found a lot of errors across the board. What puzzles me is why don’t all of these papers have their maths checked through the program before being put into print? We can’t expect everything to be perfect but we should expect mathematical mistakes to be minimal. Oh and if you use name dates to identify points in your data, be aware that spreadsheets will turn them into numbers on a graph.
Ritchie’s chapter ‘Perverse Incentives’ actually focuses on something I queried earlier in this review but with more detail. Scientists show their worth in getting papers into print. Something I hadn’t come across is salami-slicing, essentially releasing a bit at a time than all in one go and back-referencing their own papers which gains point on an H-scale used to measure integrity, which points to popularity. The translation of this is to work the system to enhance their pay and career. Part of me can recognise why they do this but ethically it does need a better system that doesn’t put scientific research at risk.
The final chapter on ‘Fixing Science’ does point out that the physics template is something that should be followed for the other sciences as it rarely has any problems. Pre-registering your experiments and what you are investigating at least gives you some copyright protection. Ritchie also thinks that the sensationalising of science is raising expectations too high and not giving enough time for the research to be completed properly. I’ve seen this a lot with diabetes and it’s a standard joke among us Type Ones that even if there was a cure, it would still be 20 years of research before being tried on a human subject.
The biggest annoyance is 81 pages of notes at the back of the book. At least half of these are where his quotes came from and could have been included in the main text if for no other reason to distinguish any important asides he has to say and these are pertinent and make it easier to track. In the end, I was either noting in my head how many down before I needed to read and avoid book references or catching up every two pages. Having so many notes might work in a university thesis but I really wish professors would adopt a better scheme when releasing books to the general public. More so as many of the notes are important to read.
After reading this book, I do think it needs a follow-up showing all the right procedures to follow. I know it might possibly be a con-person’s dream to break such a system but when its in black and white on paper, there is no excuse to say I can’t find the information or rely on other people to remember all the procedures they should know about.
As you can tell from the length of this review, I have a lot to say and probably more than I’ve actually given here. There really is a lot to digest here and think about and probably the one book of the year that you need to read. From scientist to layperson, this is an important book to read and understand. I should point out that actual science is alive and well, just its interpretation of results and the need to be checked, double-checked and treble-checked by others to prove accuracy really needs to be heightened. Science papers need to lead by example and not challenged for being inaccuracy because it is something we really do need to be shown correct.
GF Willmetts
October 2020.
Buy from https://amzn.to/3jvUPpc in the UK
