29 réponses à ce sujet

[Dermain]

That's the beauty of statistics.  You can make them say just about anything you want.

And that's why I abandoned my psych PhD plans.

 

Yes, but in doing so you open yourself up to criticism that could be dangerous to your career. 

[leighzard]

Yes.  Truth.  Always a chance you take when manipulating data or misusing statistical tests.

There are more subtle ways of manipulating data though.  Researchers may not even realize they're doing it, like excluding participants as outliers.  Was there some other factor that you missed that made them that way?  Plus no one reports negative findings because journals don't want to publish articles without significant results, unless it's about an abysmal failure to replicate.  These things contribute to the public view of science.  Plus the media only reports the glamorous studies.

[Dermain]

Yes.  Truth.  Always a chance you take when manipulating data or misusing statistical tests.

There are more subtle ways of manipulating data though.  Researchers may not even realize they're doing it, like excluding participants as outliers.  Was there some other factor that you missed that made them that way?  Plus no one reports negative findings because journals don't want to publish articles without significant results, unless it's about an abysmal failure to replicate.  These things contribute to the public view of science.  Plus the media only reports the glamorous studies.

 

And the ones that make odd claims.

 

I recall the media reported one study that said that men were "stupider" because the "researchers" analyzed the "winners" of the Darwin awards. Since the Darwin award winners are mostly men, the researchers claimed that this showed that men were stupider than women. There are other explanations than just "men are stupid k, lol", but I am unsure if that was mentioned in the studies discussion section. I would assume not.

[TheChosenOne]

Thinking?

[mybudgee]

I hope Bunz posts in here so we can use him as a living example...