At the heart of his argument are two biases. I'm so glad he limits it to two - I get totally lost trying to keep on top of all the biases that psychologists introduce, so sticking to confirmation bias plus black and white thinking as the key errors to look out for, both in how we receive information ourselves and how others present it, is very helpful.
At the heart of the book is a ladder of levels of something like quality of information. These are statement, fact, data, evidence and proof. Edmans goes into plenty of detail on each rung - how we get, for example, from statement to fact, or data to evidence. Most of all, he demonstrates brilliantly how both those undertaking studies and those interpreting them and making use of them fail to make the leap from one rung to the next. Some of the examples are horrific - where, for instance, politicians make use of a totally opposite finding to the one in a paper, or where there is no finding at all in the study itself, yet the authors claim there is one. It's a catalogue of errors, both conscious (fraudulent) and unconscious (often dues to the biases mentioned above).
Although Edmans avoids getting over-technical I found the way he presented the content wasn't always easily approached - for instance, the way he breaks down the shades of grey that are the alternative to black and white thinking into moderate, granular and marbled is hard to get your head around. Similarly, when he is talking in details about various kinds of error he introduces, for example, the concept of an 'instrument' which 'causes the input to change, but for random reasons that having nothing to do with the output' - this isn't well introduced and needs better handling.
Edmans does a really good job in identifying all the problems in that ladder from statement to proof. I think he's less effective on giving ordinary people tools to deal with them. He accepts we can't all go into detail on the what a study really says, but his 'checklist for smart thinking' requires answers to about 17 questions, though admittedly he does then condense this into around four shortcuts. For me, a focus on source, evidence and quality is about the most we can expect most to manage. Edmans also covers studies and books/articles as key sources of information/misinformation but misses university press releases, which often employ hyperbole. Similarly, he could have made more of the replication crisis.
Despite these relatively minor negatives, this is a fascinating book that really does make it clear not only how difficult it is to be sure what a study shows, but also how difficult it is for those undertaking studies to get it all right. Recommended.
Paperback:   |
Review by Brian Clegg - See all Brian's online articles or subscribe to a weekly email free here
Comments
Post a Comment