LOGO
General Discussion Undecided where to post - do it here.

Reply to Thread New Thread
Old 02-23-2008, 10:57 AM   #1
PhotoSHOPadob

Join Date
Oct 2005
Posts
328
Senior Member
Default Statistical Test Question
Maybe a t-test?
But no, wait this would require a normal distribution
PhotoSHOPadob is offline


Old 02-23-2008, 11:22 AM   #2
ElegeExcest

Join Date
Oct 2005
Posts
430
Senior Member
Default
@Eli, That's what I know. What I want to know is if this has a statistical significance.
ElegeExcest is offline


Old 02-23-2008, 11:51 AM   #3
gomosopions

Join Date
Oct 2005
Posts
467
Senior Member
Default
Simple Binominal test with true propablility of 29% for smoking.
gomosopions is offline


Old 02-23-2008, 03:36 PM   #4
MediconStop

Join Date
Oct 2005
Posts
353
Senior Member
Default
Which software do you use?
MediconStop is offline


Old 02-23-2008, 06:15 PM   #5
pobrierce

Join Date
Oct 2005
Posts
572
Senior Member
Default
The best test that you can do is: Chi-square test

A t test is inappropriate for this one.


http://en.wikipedia.org/wiki/Chi-square_test
pobrierce is offline


Old 02-24-2008, 12:28 PM   #6
UKkoXJvF

Join Date
Oct 2005
Posts
453
Senior Member
Default
UKkoXJvF is offline


Old 02-24-2008, 06:48 PM   #7
AlexanderPalamayr

Join Date
Oct 2005
Posts
498
Senior Member
Default
Meanwhile I found some chi^2-like test which did the thing.

Thanks to all of you who contributed constructively.
AlexanderPalamayr is offline



Reply to Thread New Thread

« Previous Thread | Next Thread »

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 

All times are GMT +1. The time now is 04:29 PM.
Copyright ©2000 - 2012, Jelsoft Enterprises Ltd.
Search Engine Optimization by vBSEO 3.6.0 PL2
Design & Developed by Amodity.com
Copyright© Amodity