# st: Re: Excuse me, just a question about chi-square test.

 From Maarten buis To stata list Subject st: Re: Excuse me, just a question about chi-square test. Date Thu, 12 Mar 2009 14:02:42 +0000 (GMT)

```--- On Thu, 12/3/09, Xu Kaikuo wrote to me privately:
> I saw your discussions on
> http://www.stata.com/statalist/archive/2007-09/msg00575.html,
> so I think you could provide some clue to my question. This
> is a abstract of my problem, but not the real one.
> Assume we have two kind of fridges A and B, and a function
> set F = {f1,f2...,fn} and each function has its corresponding
> weight W = {w1,w2,...,wn}, I want test whether difference exist
> between A and B on cover the functions in F.
>
> 1 I know that without the weight, we could do it like this:
> First create the 2*2 table
>                           A               B
> covered functions         a                c
> uncover functions         b                d
>
> Then use chi-square test.
> But with the weight, what should I do.
>
> 2 How to test whether A covers more functions than B
> considering the weight.

First of all it is better to respond to the list than to
individual members privately, as is discussed on the
statalist FAQ, point 5:
http://www.stata.com/support/faqs/res/statalist.html#private

As to question 1: -tab- allows weights, see: -help tab-.

As to question 2: In that case you probably want to do a
logistic regression, with fridge type as an explanatory
variable.

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://home.fsw.vu.nl/m.buis/
-----------------------------------------

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```