follow CCP

Recent blog entries
« The Next Frontier of Risk Perception: AI | Main | Combining Likert Categories and Embracing a Simulation-Based Mindset »
Saturday
Jul252009

yellow statistics

With apologies to Coldplay, here's a lament for all the "stargazers" out there:

Look at the stars,
Look how they shine for you,
It’s what you want them to do,
Yeah, they were all yellow.

We go along,
We write our articles for you,
And one of the things we must do,
Is make them all yellow.

So we make our models,
Oh we twist them all around,
Until they’re all yellow.

Cause you always look
For significance,
Asterisks look so beautiful,
Asterisks, I know you love them so,
I know you love them so.

I tortured data,
I did all for you,
Oh what a thing to do.
Cos you want it all yellow,

I made a table,
I made a table for you,
Oh what a thing to do,
And it was all yellow.

Significance,
Oh yeah you want asterisks,
You think it’s beautiful,
And you know for you,
I’ll ignore my data for you,
I’ll ignore my data.

Its true, look how they shine for you,
Look how they shine for you,
Look how they shine for,
Look how they shine for you,
Look how they shine for you,
Look how they shine.

Look at the stars,
Look how they shine for you,
It’s what you want them to do.

"Stargazers," for those not immersed in statistical analyses, are people who ignore just about everything in a statistical model except for the asterisks (stars) indicating levels of statistical significance in tables reporting regression results. 

Would a journal ever consider barring the use of tables and asterisks, instead requiring the reporting of practical effect estimates and confidence intervals?  I suspect not -- for reasons Alan Gerber, Don Green and David Nickerson reported years ago: people think statistical significance is more significant than it actually is. 

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>