This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: i for hypotheses H i (1 < i < n) under the assumption H o that all hypotheses H i are false, and if the individual test critical values T critical < α/n, then the underlying experimentwide critical value is < α. In equation form, if… Pr(T i passes  H o ) < α/n, for 1 < i < n, then the probability that some test values pass is… Pr(some T i passes  H o ) < α . Another depiction of the critical alpha values can be found using the Sidak correction . Like Bonferroni, though less conservative, an adjusted critical value is used; in this case α s = 1 – (1 – α) 1/N , where N is the number of tests to be performed....
View
Full Document
 Summer '08
 Staff
 Sociology, Critical Point, Multiple comparisons, Critical phenomena, Bonferroni correction

Click to edit the document details