When reading papers talking about the statistics of something or another. I find it difficult to understand how some people can use extremely complex statistics and yet miss over the fact that their standard deviations are way out of character for what the average is. If you have a test that is standardized to have a mean of 100 and a standard deviation of 50 and you report a test with a mean of 180 and a standard deviation of 200 or more, then I think it's more than likely that you're missing value is defined as 999 and you forgot to tell your stats program that.
So in this case, is it worse that they reported way-off-base results? Or would it be worse if they had just made it up and made up closer-to-understandable numbers? And how on earth did reviewers not catch this?
No comments:
Post a Comment