badalrath@GMAIL.COM wrote: >i want to know which hypothesis test has to be used to analyse data >like yes no types .Problem is given below > > > Organisation 1 > Q1 . Are all your needs fulfilled ? YES / NO. > Q2 . Have you ever compalined ? yes / no > > Organisation 2 > Are all your needs fulfilled ? YES / NO. > Q2 . Have you ever compalined ? yes / no > >Organisation 3 > Are all your needs fulfilled ? YES / NO. > Q2 . Have you ever compalined ? yes / no > > > I want to know how to propose and formulate and test >the hypothesis for above data.I want to combine the above 3 >organisations into 2 by taking 1st organisatin as one group and 2 and >3 organisation combined as another group and compare.

First off, I agree with everything Peter said. So let's go from there.

You need to write back to SAS-L (not to Peter or me) and explain more of your problem.

Where do your data come from? How are they gathered? (I mean, exactly how they are gathered.) This is crucial, since it will determine two things: (1) which proc you need to use; and (2) whether your results will be garbage or not. It sounds like you have survey data, so the care taken in selected the people for the survey, and the issues involved in who responded and who didn't, are going to be important in getting useful conclusions. Tell us how many people were asked, and how many responded, and what was done about the non-response.

What stat background do you have? The level of discussion may get higher than your comfort level. So telling the list this will help people know how to discuss the issues.

And finally, why are you asking the above questions? What will the results be used for? This may affect what we do with the data.

HTH, David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ Express yourself instantly with MSN Messenger! Download today - it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/