Statalist The Stata Listserver

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

Re: st: simple question to tests

From   Maarten buis <>
Subject   Re: st: simple question to tests
Date   Mon, 8 May 2006 09:10:08 +0100 (BST)

--- Stephan Brunow <> wrote:
> I run a regression and afterwards I did a test for
> I am (very often) confused, what the probability (here 0.0000) means.
> Does it mean that the Ho can not be rejected on a 5 percent level?

There are two (more or less) equivalent ways of thinking about
- It is a 'degree of surprise'. If you believed that the null
hypothesis is true than the probability of finding the data you have
found is <.00005. That data is so surprising that you start to
reconsider your null hypothesis.
- You have chosen a significance level, which is the minimum
probability of accidentally rejecting a true null hypothesis that you
find acceptable. You are basing your inference on a radom sample, so
'accidents' can (and will) happen. For instance you may find 5%
probability of accidentally rejecting a true null hypothesis
acceptable. Once you have choses the significance level you compute the
p-level and if the p-level is less than the siginificance level you
reject the null hypothesis. If you do so consistently throughout your
scientific career than 95% of your conclussion will be correct
(assuming that the only source of mistakes are normal fluctuations that
occur with random sampling)


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

visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z214

+31 20 5986715

Switch an email account to Yahoo! Mail, you could win FIFA World Cup tickets.
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index