Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: Using prtest for large obs, in conjunction with oneway


From   Steven Young <youngazn@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Using prtest for large obs, in conjunction with oneway
Date   Mon, 17 Jun 2013 23:01:25 -0700

I am a bit new to statistics. I have a sample of about 2400 with 3
variables, each with 2 or 3 levels associated with them. As I was
reading online, I saw that z-tests should be used in favor of t-tests
when n>30. After finding that prtest is the z-test equivalent, I
proceeded to use it.

I have been using .oneway to check for equal variance in the
population, and then had continued to conduct a t-test. If I found
that Bartlett's chi2 was significant, I rejected the equal variance H
and used unequal welch with my ttest. Do I need to do the same when
using prtest, ie run .oneway to check for equal variance? Or would I
forgo that and just compare the data with prtest alone?

Although the documentation also says bitest is preferable, it does so
by mentioning "smaller" samples. Reading the documentation on bitest,
it seems to be more useful for actual probability values, whereas I
have data means to test. What do you say?

Thank you.

Steven Young
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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