Re: st: sign test output

[Maarten Buis <maartenlbuis@gmail.com>]

On Fri, Jan 18, 2013 at 1:34 PM, Nahla Betelmal wrote:
> This is what initially made me go for the sign test instead of t-test
> as I first thought that if the data is not normally distributed, then
> use non-parametric test, but from previous discussions in statalist ,
> no agreement on that!

I have seen a surprisingly great level of agreement on this list on
this issue: it depends on what you want to do and what your data looks
like.

> So, now I am confused how to interpret the results. Using the sign
> test: there is sufficient evidence that earnings are managed upwardly.
> Using the t-test: earnings are not managed at all!! and this is what I
> meant by irrationality in previous email.

The first step would be to start thinking about whether you want your
results in terms of means or medians. As soon as I see the word
rationality I think expected value, and in that case I would thus tend
to the mean, but it is your theory and your paper so you need to
decide.

-- Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   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/