Stats help please

Debate and discussion of any biological questions not pertaining to a particular topic.

Moderators: honeev, Leonid, amiradm, BioTeam

Post Reply
Posts: 1
Joined: Thu May 07, 2009 5:06 pm

Stats help please

Post by realdeal » Thu May 07, 2009 5:31 pm

This should be an easy question

3 columns
Flow rate

some of the values of columns cased and uncased are zero I'm trying to show the difference between the two what test should i use I have tried man-Whitney T tests and anaova but none seem to give any results

I'm using mini tab & Excel

Posts: 672
Joined: Fri Mar 30, 2007 4:23 am

Re: Stats help please

Post by blcr11 » Fri May 08, 2009 8:20 pm

You sure you don't really have a "two-column" problem with "cased" flow rates vs. "uncased" flow rates; unless the "flow rate" column is some kind of control? I'm imagining the question as: is there a significant difference between cased and uncased flow rates (whatever those are)? If there are only two categories, you don't need ANOVA (rather, ANOVA and staightforward t-tests should give you the same result). Is there a reason to prefer the non-parametric Mann-Whitney over the parametric t-test? Do you know the underlying distribution to be seriously non-normal? The non-parametric test should always be valid no matter what the underlying distribution, but non-parametric tests are less powerful than their corresponding parametric tests when the underlying distribution is normal.

Post Reply

Who is online

Users browsing this forum: Bing [Bot], Google [Bot] and 12 guests