I quickly googled the question and the answer is: it is always true that if the confidence intervals do not overlap, the statistics are statistically significantly different.

However, it is not necessarily true that they are not significantly different if the confidence intervals do overlap. The means are significantly different when X1 – X2 = 1.96*sqrt(SE1^2+SE2^2) and there is no overlap between the CI when X1 – X2 = 1.96*(SE1+SE2). Please also see this infobox from Cornell University.

Best, Ben ]]>

I have a question regarding the interpretation of the confidence intervals. Are two predicted values significant from each other only when their respective confidence intervals don’t intersect or also when the confidence interval of one value does not include the other value?

I hope I expressed myself clearly enough.

Best,

Magdalena ]]>