Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Comparing results from two separate t-tests


From   Geoff Wallace <[email protected]>
To   [email protected]
Subject   Re: st: Comparing results from two separate t-tests
Date   Wed, 12 Mar 2008 16:26:06 -0400

I took the advice of both Maarten and David to try a regression route and this definitely provides a lot more flexibility and indeed gives the same substantive results.

I pretty much followed Maarten's suggestion of incorporating the interaction effect and that does the trick for testing the hypothesis of any difference between liberals and conservatives on the treatment without have to do multiple tests.

Thanks again to both of you!

Cheers, Geoff

Maarten buis wrote:

Assuming that you have two dumies called treatment (1=treatment,
0=control) and liberal (1=liberal, 0=conservative), and that your
outcome variable is called y then you want:

anova y treatment liberal treatment*liberal

or:

xi: regress y i.treatment*i.liberal

The two are actually the same.

If you want to read more about this, than the term you are looking for
is "interaction effect".

Hope this helps,
Maarten



--- Geoff Wallace <[email protected]> wrote:

Dear All,

I have had problems trying to test for whether the treatment effects
for two groups of respondents (conservatives and liberals) in a social science experimental questionnaire I am using are significantly different. I used the -ttest- command to obtain the difference of
means across the treatment and control group first for the conservative respondent subsample and then repeated the procedure for the liberal respondent subsample. I should also note that I used the –unpaired- option for each command since the data are not paired. The output
from the two -ttest- commands yields a separate treatment effect for each subsample. However, I am having trouble figuring out how to use Stata
to perform a corresponding ttest on the two separate difference of
means. I tried using -est store- after each -ttest- command, but as I expected

this option does not appear to work with –ttest-.

Any help would be much appreciated.

Best, Geoff

--
Geoffrey P.R. Wallace
Ph.D. Candidate
Department of Government
Cornell University

214 White Hall
Ithaca, NY 14850
t.202.302.4209
f.607.255.4530

*
* 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/


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

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


___________________________________________________________ Rise to the challenge for Sport Relief with Yahoo! For Good
http://uk.promotions.yahoo.com/forgood/
*
* 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/
--
Geoffrey P.R. Wallace
Ph.D. Candidate
Department of Government
Cornell University

214 White Hall
Ithaca, NY 14850
t.202.302.4209
f.607.255.4530
www.people.cornell.edu/pages/gw46/
<http://www.people.cornell.edu/pages/gw46/index.htm>
*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index