this is the definition of Type I error, or false positive. googl

google type I error or find any intro statistics book and you will understand it better. the person upstairs was bluffing or didn't take any stat courses. the probability of type I error (false test significant difference when it's not true in reality) in a statistical test is equal to the alpha level you set (usually 0.1, 0.05, or 0.01). The more lenient you set alpha (such as 0.1), the more likely you will encounter type I error.

所有跟帖： 

• as to how it could happened it your specific study, first check -Kamioka- ♀ (2338 bytes) () 06/26/2012 postreply 20:15:34