Don’t Wash That Data!

[From the article “Embrace the unknown” by Chris Ferguson, published in The Psychologist
“Consider the basic premise ‘Does X cause Y?’ It’s at the root of almost any question of interest to the general public or policy makers. Does cognitive-behavioural therapy treat PTSD? Does the TV show 13 Reasons Why cause suicide in teens? Can implicit racism be tested for, and does training reduce racism in society? …it is often in the interest of professional guilds – the advocacy organisations that represent psychology and other sciences – to give simple answers.”
“Science laundering is the washing away of inconvenient data, methodological weaknesses, failed replications, weak effect sizes, and between-study inconsistencies. The cleaned-up results of a research field are then presented as more solid, consistent and generalisable to real-world concerns than they are.”
“As one recent example, Jean Twenge and colleagues (2018) released a study, covered widely in the press, linking screen use to youth suicides. However, another scholar with access to the same dataset noted in an interview that the magnitude of effect is about the same as for eating potatoes on suicide (see Gonzalez, 2018: effect sizes ranged from r = .01 to .11 depending on outcome).”
“Some readers may be thinking, ‘Isn’t it better to attempt to apply psychology to important societal issues even if the evidence available falls short of being conclusive? How certain do we really need to be before we stop fretting about overselling the value of our science?’”
“I take an unapologetically hard line on this: honesty must be a fundamental facet of scientific communication. We cannot and should not sweep under the rug inconvenient data, methodological weaknesses or tiny effect sizes for the sake of an appealing narrative, no matter how heartfelt that narrative may be.”
To read more, click here.

Leave a Reply

Please log in using one of these methods to post your comment: 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

%d bloggers like this: