Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.



simple question
Posted:
Jul 4, 2012 7:53 AM


Hello all. I have a simple stats question that's been bugging me for about a year now. In several papers I have noticed that some authors will compare two numbers and find significant difference between the numbers with a pvalue < .05, and will then go on to say much about this in the discussion. What they fail to mention is the 95% confidence intervals for these values overlap despite the statistical significance. My question is: What is the best way to interpret data in these situations (namely, where there is a statistically significant difference between the values, but their errorbars overlap)?
Thanks.



