Simon, Steve, PhD <[EMAIL PROTECTED]> wrote in sci.stat.edu: > >http://www.childrens-mercy.org/stats/definitions/pvalue.htm > >Any comments from edstat-l readers would be appreciated.
Mostly I like it. One sentence did bother me, though: "a large p- value is evidence in support of the null hypothesis while a small p- value is evidence against the null hypothesis." Does a large p-val actually constitute evidence in support? I thought it was more the absence of evidence against. (If you can't come up with evidence that a defendant is guilty of a crime, that does not mean he's innocent. He might just have covered his tracks very well.) Even if my point is valid, maybe what you have written is okay in a treatment that is trying to keep things simple; what do others think? (You do have an inconsistency, though. A bit later you say "A large p-value should not automatically be construed as evidence in support of the null hypothesis.") I think "what sized difference" could be "what size difference". There are three typos on the referenced page, <http://www.childrens-mercy.org/stats/library/pvalueci.asp> (1) The period is missing after Hopkins' middle initial. (2) Punctuation and space are missing after Thompson's name. (3) I get a "broken image" icon in the upper left corner, which I guess means a typo in an <img src="...">. Could you add an online reference to a power test to that page? -- Stan Brown, Oak Road Systems, Cortland County, New York, USA http://OakRoadSystems.com/ "Walrus meat as a diet is less repulsive than seal." -- Harry de Windt, /From Paris to New York by Land/ (1904) . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
