what does this mean? (Stats)

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

Moderators: honeev, Leonid, amiradm, BioTeam

Post Reply
Posts: 260
Joined: Tue Mar 21, 2006 9:33 am

what does this mean? (Stats)

Post by biology_06er » Wed May 05, 2010 8:10 am

Hi there,

Everytime I read articles I never understand the statistical methods used...example this sentance:

"in threshold analyses the authors found that those in the top decile of total zinc intake (15mg+/day) were significantly less likely to develop early AMD (RR 0.54; 95% CI 0.30-0.97 P=0.041)...what does relative risk actually mean..i have googled these terms but still dont really understand so hoping someone here can explain simply for me. I read that the lower the P value (less than 1 means its statis. sig??)...is that right?....as for RR/CI can somone please explain..

also...the sentance "antiox intakes were analysied as continuous variables (per standard deviation) and as tertiles, the lowest tertile was the referent group". what is a tertile...it doesn't even exist on dictionary.com :S...does it just mean group as in the highest tertile for zinc (highest group having the highest intake of zinc??)

Thanks and hopefully someone can explain the above,

User avatar
Inland Taipan
Inland Taipan
Posts: 5694
Joined: Mon Sep 14, 2009 7:12 pm

Post by JackBean » Wed May 05, 2010 9:01 am

significance shown by p-value differers. As clinical trial can be happy for anything under 0.1, that's significant for them, usually it's 0.01 or 0.001 ;) Of course, the lower, the more realiable is it. Best is, if you get 0 (so low, that it can't be expressed as 1E...)

Cis or trans? That's what matters.

Post Reply

Who is online

Users browsing this forum: hildajovin and 11 guests