Re: st: Re: specifying difference in Ho of ttest

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 06:15 PM 1/14/2004 -0600, Scott Merryman wrote:
> Isn't this just calculating the difference and then using -ttest diff = #-
>
> Example:
>
> . sysuse auto
> (1978 Automobile Data)
>
> . gen diff = disp - length
>
> . ttest diff = 10
That should work, and is more straightforward than what I tossed out 
before.  A matched pairs problem can be converted into a single sample 
t-test problem in this way.

One thing that occurs to me: Scott and I both interpreted this as a matched 
pairs problem.  But, if the original author meant "diff" to stand for the 
difference between 2 group means (i.e. an independent samples t-test) 
rather than the difference between the means of two variables (matched 
pairs) then neither Scott's nor my approach is right.  So, if you 
hypothesized that group 1's mean was 10 points higher than group 2's, you'd 
do something like

gen score2 = score
replace score2 = score2-10 if group==1
ttest score2, by(group)

-------------------------------------------
Richard Williams, Associate Professor
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/