Old man needs help with stats - Chi-square + Wilcoxon Test?

Everything on bioinformatics, the science of information technology as applied to biological research.

Moderators: honeev, Leonid, amiradm, BioTeam

Post Reply
Posts: 1
Joined: Wed May 10, 2006 1:31 pm

Old man needs help with stats - Chi-square + Wilcoxon Test?

Post by Mooreanut » Wed May 10, 2006 3:03 pm

Hi there. Can anyone give me some help?

THE SITUATION: A friend had found plant sp F to be very common in a valley where she works. I didn't think it was as common in the valley where I work so I ran the same kind of transects (randomly chosen 2m-wide belt transects) across 'my' valley and counted sp F.

THE RESULT #1: She had found 69 individuals in 6,974 m2 whereas I found only 4 individuals in 8,374 m2.

QUESTION #1 – I would like to say that these results are significantly different. I've used the Chi-square test with the Yates correction on these data - is that OK? The observed were 69 and 4, and the expected were the 73 individuals arranged according to the area surveyed.

RESULT #2: While doing my transects I also noted any sp L I came across. So I have the counts for two species along my transects:
total number of sp L = 99;
total number of sp F = 4; and
number of transects = 8. [Sp L was more common along 5 of the transects and both species were zero along the other three.]

QUESTION #2: I would like to say that sp L is significantly more common than sp F in my valley. How do I do that? I have used a Wilcoxon Test for unpaired data. Is that right?

Any help/comments/insights will be GREATLY appreciated! Cheers.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest