Re: Statistical Conventions in Social Science papers?


Kevin E. Thorpe wrote:
On Dec 30, 6:24 am, "Reef Fish" <large_nassua_grou...@xxxxxxxxx> wrote:

The error of omission was in NOT stating the Alternative Hypothesis,
which is critical in the DEFINITION of the p-value.


As I explained earlier this year, the difference is in whether
or not you approach this in a Fisherian way or a pure NP way.

I consider my statistical education about as "old fashioned" as one
can be without being labeled as a dinosaur. Fisher's way was
already out-of-style long before I started statistics. It is
definitely
out of style NOW.

In the case of continuous data, the strict inequality does not
affect the computation of a p-value, but for discrete data it
does, which is why NP suggests randomization on the
boundary.

That is the nonessential part of the error. To take ALWAYS the
one-sided alternative and call it a "significance level" is an ERROR,
a major BLUNDER, no matter how you argue it.

But the examples carefully selected by David Winsemius were much
worse than that! Below are the reasons for my statement, spelled out:

David's examples (which I recognize) are both from
descriptions of Fisher-type significance testing, not NP
hypothesis testing. As I pointed out, Fisher did not
require a formal alternative hypothesis.

That is why he is THE dinosaur. What about the one-tailed
vs two-tailed tests?

I don't mind the discussion of the historical perspective of
how hypothesis testing evolved, but to pull out those anti-
que ideas and quote them as if they are APPLICABLE in
today's statistics environment, is at best a diberate distortion,
and at worst a peddling of ERRONEOUS concepts as is
correctly practiced by all reputable authors today.

I know you disagree, that is fine. In reality, what is now
taught is often a hybrid of NP and Fisher theory. You
yourself have rejected it in favour of Bayesian approaches
anyway, suggesting there are multiple approaches to the
problem.

But that's a different issue becuase BOTH theories are
conceptually DEFECTIVE, but that's not the same as saying
I accept the Fisherian version as VALID when it can't even
distinguish a one-tail from a two-tailed test, and it disregards
the Alternative Hypothesis.

In the end, the Fisherian way NEVER had a DEFINITION of
a p-value.

It is a gross error on the part of all, who thinks that what Cox
and Hinkley and others call the p_obs a p-value.

Read this VERY carefully,

A p-value does NOT exist in the Fisherian context

It exists, and is meaningful ONLY when it works in conjection
with an Alternative Hypothesis to be able tell what "more
extreme" means.

In the end, what you are pointing out is that David Winsemius,
and possibly yourself, are using the INAPPRORIATE concepts
and definition AS IF what is NOT a p-value in the Fisher sense
is the current definition of a p-value.

That is WRONG - no matter how you cut it. A p-value does
NOT exist in the Fisherian framework. Significance testing,
yes, but in a very restricted and narrow way.

That is the bottom line.

-- Reef Fish Bob.

Kevin

.

Relevant Pages

  • Re: Statistical Conventions in Social Science papers?
    ... which is critical in the DEFINITION of the p-value. ... Alternative Hypothesis is a NECESSARY and indispensible ... The notion of a p-value DOES NOT EXIST in Fisherian ... for continuous OR discrete test statistics. ...
    (sci.stat.math)
  • Exact Level of Significance
    ... The notion of a p-value is that it is NOT necessary to pre-determine ... any alpha level to test a hypothesis, ... a p-value DOES NOT exist in the Fisherian way ... The idea of randomization is NOT for p-values, ...
    (sci.stat.math)
  • Re: Two nit-picks re definition of p-value (Was: goodness of fit ?)
    ... RFB> is the last place anyone wants to learn Statistics! ... We are talking about p-value definition. ... alternative hypothesis: true probability of success is not equal to ...
    (sci.stat.math)
  • Re: Two nit-picks re definition of p-value (Was: goodness of fit ?)
    ... RF> It depends on the STATEMENT of the Alternative Hypothesis. ... DIFFERENT Alternatives (for hightly discrete cases, ... "more extreme" definition, defined by Ha, and Ha alone. ... But the STATEMENT and DEFINITION of p-value ...
    (sci.stat.math)
  • Re: Two nit-picks re definition of p-value (Was: goodness of fit ?)
    ... or describe it as "more extreme than the observed ... definition of a p-value. ... It depends on the STATEMENT of the Alternative Hypothesis. ... What conclusions after observing an X = 0. ...
    (sci.stat.math)