Non significant statistical trends. Why are they reported?

I’m working on a slew of new posts but wanted to just throw this out there. Reading through a paper this morning, I was struck by the fact that for the 4th time this week, I’ve come across a study that reports on or mentions some non-signifiant statistical trend in the data (the latest I’ve come across touts a p value of .07). Why are these trends reported at all? They’re very misleading and most certainly only reported when they suggest support for a given hypothesis (I haven’t noticed too many papers reporting trends that would go against the central hypothesis). Why set a threshold at all if you’re going to report stats that exceed it? Am I off base here?

Advertisements

One thought on “Non significant statistical trends. Why are they reported?

  1. I agree, if its not statistically significant, then why bother even mentioning it, however its like saying ‘there is one black jellybean in every pack of a hundred jellybeans’ however, that one black jellybean is there in every pack, so it is important to note that I suppose.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s