book call

Current Japanese Micro-Season

Loading...

Loading...

Loading...

Loading...

“We conclude that apparent effects of growth mindset interventions on academic achievement are likely attributable to inadequate study design, reporting flaws, and bias.”

Trey Causey

In another sense, my conclusion is positive on mindset interventions in that, given that any average effects will be small, the lack of statistically significant average effects in small or even moderately-large studies does not have to imply that mindset interventions don’t work; it just says that they only work in some settings, and individual effects will mostly be small.

Andrew Gelman hardly needs more examples of how to think carefully about experimental results, especially regarding treatment effects, but this stuck out to me as a particularly clear example of nuance in practice.

Unfortunately “statistically significant” is well-known to be abused as a proxy for “true”, even by those who know better (don’t get me started on “stat sig samples”) but I think the opposite is true as well. No significant effect is not the same thing as “no effect exists”.

I honestly have no idea how to hold two thoughts in my mind at the same time, and yet I do:

  1. We should be teaching way more basic statistics and probability to students, beginning in high school or earlier.
  2. A little bit of statistics knowledge can be dangerous and instill a lot of false confidence in how to interpret research.

Subscribe for new posts

ESC
Type to search...
↑↓ to navigate to select