Talk:Friedman test

From Wikipedia, the free encyclopedia

This article is within the scope of WikiProject Statistics, which collaborates to improve Wikipedia's coverage of statistics. If you would like to participate, please visit the project page.

"n wine judges rate k different wines. Are the ratings consistent?" doesn't seem like an application of the Friedman test ... the test compares the mean rank between wines but these could be identical even though the ratings between judges are highly inconsistent.

Thom

This Version of the Friedman test is no longer used. The version currently used recommended by conover is explained quite nicley here: http://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/friedman.htm

It is also fully explained with full Citations in David Hull's paper "Using Statistical Testing in the Evaluation of Retrieval Experiments" The key difference is the t-distribtution is now used rather than Chi-Squared

Why is the Friedman test referred to as equivalent to a 2-way ANOVA? It is only a one-way repeated ANOVA, or am I missing something?

Why is the Friedman test referred to as equivalent to a 2-way ANOVA? It is only a one-way repeated ANOVA, or am I missing something? [forgot to sign previously] Rickogorman 18:26, 8 March 2007 (UTC)