Freedman, David H. (2010). Wrong: Why experts keep failing us – and how to know when not to trust them. Little, Brown and Company, New York, 295p.
- Bias and Corruption
- Irrational Thinking
- Pandering to the audience
- Lack of oversight
The trouble with scientists
- Studying the wrong mammal
- Tossing out inconvenient data
- Moving the goalpost
- Being confounded
- Juggling the numbers
- Being paid to get it wrong
Studies that prove the hypothesis (positive studies) are more likely to a) be submitted and b) be published.
Lack of contrarian studies does not imply that the hypothesis holds.
Types of study and their perceived worth
According to reporters and scientists
- Observational study: Interesting but untrustworthy
- Epidemiological study: Somewhat trustworthy if it is large and done well
- Meta-analysis or review study: Trustworthy
- Randomized controlled trial: Very trustworthy if large – the goal standard of evidence
We prefer someone who is sure of being right than someone who freely raises counterarguments to improve our shared knowledge, even though we recognize that he who is blindly sure is often wrong.
Knowing that blind faith in a solution may be a social desirability bias on the part of those answering and wanting to show that they know the least likely answer in their minds is probably what the research is aiming for.
Foolproof title formula for being read
The [6 to 13] [tips, secret, rules, etc.] for [aspect the reader would like to master]
What we look for in expert advice
- Dramatic claims
- Retroactive fixes
The idiocy of crowds
The Condorcet theorem
Even with a marginally positive chance of each individual coming up with the right binary solution, the larger the group, the more likely the correct conclusion will be reached.
The Hitchcock Effect
From one of the old TV show, an individual sends many predictions in waves to a large group of people, each following wave only receiving further predictions if the initial one turned out right. In the end, the main character, by luck, received all the correct predictions and thinks the sender a prophet.
Looking back to experts to validate who has been right in the past has no bearing on who may be right in the future
Experts and organizations
Book types of expert-inspired paradigms
The authors place a number of winning companies or CEOs under a microscrope, distilling what management principles theses role models offer that losers don’t
- In search of excellence
- Good to great
- First, break all the rules
- The breakthrough company
- Talent is overrated
Authors observe or derive a new strategy, trend, or management technique that will determine which business will succeed in the coming years, showing how winning companies are already taking advantage of the new thinking
- Competing on analytics: the new science of winning
- The world is flat: a brief history of the twenty-first century
- Wikinomics: how mass collaboration changes everything
- The future of management
Any change introduced by management tend to invite at least a brief improvement in the output of workers involved in the study.
According to a 2002 research in HBR, tend to be :
- Falsely encouraging
- In tune with the zeitgeist
- Novel, but not radical
Experts and the media
Study, both valid and not, are often misinterpreted, misquoted or not fact checked by the media.
The wildfires that can be started by this distortion are impressive.
The effect of the internet
More information is available, but most aren’t equipped to differentiate between what is good and what isn’t. Swamped under that volume, most will go for the most popular advice, which is often flawed.
Characteristics of less trustworthy expert advice
- Simplistic, universal, and definitive
- Geared toward preventing a future occurrence of a prominent failure or crisis
Characteristics of expert advice we should ignore
- Mildly resonant
- Lots of positive attention
- Other experts embrace it (put into perspective, not ignored)
- Appears in a prestigious journal
- Supported by a big, rigorous study
- Experts backing it boast impressive credentials
Characteristics of more trustworthy expert advice
- Doesn’t trip the other alarm
- Negative findings
- Heavy on qualifying statements
- Candid about refutational evidence
- Provides some context for the research
- Provides perspective