Supreme Court rules that tests of statistical significance are inadequate

For serious.  In a recent case the U.S. Supreme Court heard a product-liability case where a drug company failed to disclose to the public a nasty side effect of a new drug.  The company argued that it did not need to disclose the side effect because in clinical trials the correlation between the two was not statistically significant.  The Court ruled (unanimously) that a company cannot rely solely on the metric of statistical significance for matters like this.

Then the article goes on to discuss the debate over statistical significance.  The upshot: even correlations that are not statistically significant should be reported if the effect size is large.

File this under “Always report your effect sizes and statistical power”


About Jake Wobig

I teach international relations and comparative politics at Wingate University in Wingate, North Carolina
This entry was posted in Methodology, Statistics. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ 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