Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: R: testing normalitt of data


From   "Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: R: testing normalitt of data
Date   Mon, 29 Jun 2009 15:12:36 +0200

Dear El Hawary,
the standard t-test of a difference assumes that the data to be tested are
normally distributed and that the two groups have equal variance.
These requirements are seldom fullfilled (although the t-test is quite
robust to violation of the latter prerequisite), making decision based on
t-test significance possibly misleading. Should this be your case, you can
consider:
- including in your research report the interval estimate of the difference
rather than the p-value obtained from t-test;
- contrasting the t-test p-value with a bootstrap p-value and decide about
the significance of the difference between the two means. To my
knowledge,this approach is reported in Stata 9/2 SE manual and is
recommended in at least one textbook and two articles:

. textbook: Efron & Tibshirani. An Introduction to the Bootstrap. New York:
Chapman & Hall, 1993: 224;
. articles: 
 - Barber JA, Thompson SG. Analysis of cost data in randomized trials: an
application of the non-parametric bootstrap. Statistics in Medicine.
2000;19: 3219-236;
- Desgagné A, Castilloux AM, Angers JF, LeLorier J. The use of bootstrap
statistical method for the pharmacoeconomic cost analysis of skewed data.
PharmacoEconomics 1998; 13: 487-97.

HTH and Kind Regards,
Carlo
-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di El Hawary
Inviato: lunedì 29 giugno 2009 11.03
A: statalist@hsphsun2.harvard.edu
Oggetto: st: testing normalitt of data

can I use the t-test for independent samples when the skewness and kurtosis
of data are as follow
skewness: -1.232 & the std. error of skewness is .616/ .044 and the std.
error of skewness is .427
kurtosis: 2.195 & the std. error of kurtosis is 1.191/ -.505 and the std.
error of kurtosis is .833
I really need help and guidance to make decisions about using the t0test


      

*
*   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/



*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index