Re: Afonso tried to drag Jerry Dallal down his own GUTTER.

Reef Fish wrote:
"Luis A. Afonso" (kept dragging Jerry into his gutter) wrote:

Jerry NEVER admitted that it was wrong.

But he NEVER gave a reason WHY he didn't pool the variance when p1 = p2
and there is only ONE unknown variance.

Because it's not necessary! All I need is a statistic whose distribution is known under H0. The separate 'p' statistic satisfies the requirement. I hope you're not inferring that I'm claiming it's the *right* way to do the test. I'm not. I'm saying it is *a* way to do the test. What surprised me when I first saw it was the Robbins conjecture cited by Snedecor & Cochran there there may be situations where the separate p statistic outperforms the pooled statistic.

It's similar to the large sample t test that uses separate variance estimates rather than a pooled estimate. The test remains valid even if the population variances are equal. In fact, I believe Paul Velleman presented only the separate variances test in his ActiveStats.

>
Plenty of bibliography is added to base Jerry's argumentation!!!!!!

ALL of them supported the pool variance, including Hoel, Dixon and
Massey, and those cited in the paper he cited! They were discussing
something ELSE, not whether it should pool or not.

YOu never read that paper did you?

HOWEVER this discussion is restricted to him and Jerry.

They why are YOU dragging Jerry into your corner -- Jerry never made
any of YOUR mistakes except the one not pooling. YOu made at least
half a dozen other errors NOBODY would make.

What you dragged out of Jerry's post had ALREDY been debunked by
me. In particular, NONE of those non-specific references addressed
the PARTICULAR test of the hypothesis Ho: p1 = p2.

That is the ONLY Ho for which pooling of the variance is necessary.

YOu should let Jerry speak for himself. You are dragging him down
YOUR GUTTER when he didn't commit YOUR atrocities.

I explained Jerry's ERROR in my post you cited. In particular, later
I point out that the article

Also, see Eberhardt & Fligner (1977), The American Statistician, 31,
151-155

was the only that ALL of the cited authors Hoel, Dixon and Massey,
etc. used the POOLED variance -- just as I said that EVERY elementary
textbook on the subject of using Z to test Ho: p1 = p2 used the
pooled variance.

While it's true that many text present only the pooled 'p' version of the test, I know of none that calls the separate 'p' test wrong (other than the one that misinterpreted the E&F paper by focusing on *small sample* properties).

Jerry gave NO REASON why didn't use it, other than pointing the the
authors and article which did NOT deal with that specific topic.

I never said I didn't use it. I said that both tests are valid.


HOWEVER, THIS IS NOT THE POINT HERE

The point is Luis A. Afonso (BOTH of them) are ignorant about
Hypothesis Testing.

_____licas (Luis A. Afonso)

With the re-explanation by me above, and the explanation I already
gave Jerry, you should SHUT UP! Let Jerry speak for himself --
and he did, and did plenty.

You are just leaning on his NAME to try to raise your own status,
in the same manner Richard Ulrich was trying to do so on one of
Jerry's other faux pas in the interpretation of correlation
coefficients
in a multiple regression.

Sorry. I recall no such faux pas, which doesn't mean I'm incapable, but you'll have to point it out to me.

-- Reef Fish Bob.

******************************************************************************************************************************************
Date: Nov 7, 2006 10:19 AM
Author: Jerry.Dallal@xxxxxxxxx
Subject: Re: Testing the Equality of Two Population Proportions


Reef Fish wrote:
Jerry.Dallal@xxxxxxxxx wrote:
Snedecor & Cochran, 7-th, p 125
Freedman Pisani & Purves, section 27.2
Robbins & van Ryzin, p 192
Fisher & van Belle, p 187
Ott & Longnecker 5-th, p 484

Also, see Eberhardt & Fligner (1977), The American Statistician, 31,
151-155
I am sure Jerry Dallal's references have to do with what I called an
"error" in testing Ho: p1 - p2 = 0 with the form of the test
statistic
Z = (p1^ - p2^)/ sqrt( var )

WITHOUT using the pooled variance p^( 1-p^)(1/n1 + 1/n2)

which is found in EVERY elementary textbook I've used in 25 years
of teaching the FIRST course in statistics (at least half a dozen
different books of that level).

The REASON: The TEST STATISTIC for TESTING Ho must
incoporate (whenever appropriate) the fact that Ho is TRUE.
I know of no such principle. The only requirement is that the
distribution of the test statistic be known when the null hypothesis is
true.

When Ho is TRUE, p1 = p2 , so that there is only ONE unknown
p. Therefore, why should one NOT use the pooled p^ for the
variance in the test statistic by the common p^ = (x1+x2)/(n1+n2)?

The REASON for using the pooled variance is NOT whether one
form of the approximation is better than the other (whatever that
means), but to satisfy the definition of "alpha" and "p-value"
associated with the test Ho: p1 = p2. For Ha: p1-p2 > 0,

alpha = Pr ( TEST STAT > c | Ho is true)

p-value = Pr (TEST STAT > observed Z* when Ho is true).

In BOTH cases, "when Ho is TRUE" is imbedded in the definition.

So, the question can be asked from the other direction, that
if we assume p1 = p2, then WHY do you use two DIFFERENT
estimate for p in the variance formula which as only ONE
unknown p?
Because for large samples, the distribution of the test statistic with
two different estimates of P is known when the null hypothesis is true
and it may be more powerful in some cases when the null is false.

THAT's the crux of the issue.

I noticed that Jerry did not give the publication dates of those
books, and only the date of the American Statistician article
WITHOUT the specific discussion of the point I made above
and elsewhere in my "Hypothesis Testing" topics about this
particular problem of testing the EQUALITY of two independent
p's with an approximate Z. (Large Sample case).

The two different z's, z1 and z2 Jerry used in his webpage
on the problem are both perfectly valid, for constructing
CONFIDENCE INTERVALS.

In C.I., the only relevant assumption is that the Statistic is
approximately normal. There is NOTHING in the construction
of confidence intervals for (p1-p2) that assumes p1 = p2! :-)

The textbooks I am referring to are NEARLY ALL (I am
learning to be careful on that :-)) and ALL of those ones
that were used in my university in ALL the courses I've
taught at the FIRST COURSE level, to all majors. Those
include English majors, and other liberal arts majors for
which the one course is their last statistics course.

The period covered 1975 - 1999. The reason I am
particularly familiar with this particular problem is that
the same test is used in several DIFFERENT first courses,
for math majors, for engineering majors, for nursing majors,
and for liberal arts majors. They ALL had this approach
in common, and they ALL make sense, because otherwise
it would be rather difficult to make sense about p-values
and alpha, while ignoring the clause "when Ho is TRUE".

The thing about elementary statistics texts is that they are
elementary, perhaps too much so sometimes. It's easier to give the
pooled version, which agrees with the chi-square statistic, than to go
into the issues we are discussing here.

More important, do you know of any text, elementary or otherwise, that
recommends against the separate p^ version? I know of one, but its
justification is a MISINTERPRETATION of the 1977 American Statistician
paper I cited.

Snedecor, Cochran, Robbins, and Freedman are GIANTS in the field.
While no one is perfect, I find it much more likely that the four of
them writing in three separate texts are more likely to be correct than
any one of us writing here. I strongly suggest that anyone who wants
to comment further begin by reading the references (especially the 1977
American Statistician paper). I have no interest in discussing the
point further with anyone who has not.

I recall Jerry Dallal and I had a rather extended discussion
of the definition and operational meaning of "p-value"
and Jerry kept citing the book by two statistician who
did not even include Ha as part of the definition of p-value,
which would make it impossible to tell what "more extreme"
means in;

p-value = Pr (TEST STAT is "more extreme" than the
observed value of the TEST STATISTICS
when Ho is TRUE).

It's all related and consistent. I think I am quoting Jerry
Dallal correctly when I said,

RF> p-value = Pr (observing something more extreme
RF> than the observed T* when Ho is true).

and Jerry had a one-word response:

RF> irrelevant
The "irrelevant" was in response to your email to me over the weekend.
If we are in agreement that the separate p^ statistic has a standard
normal distribution when H0 is true, then there is no need to focus on
P values or any other particular use of that distribution other than
that we can construct a test with specified alpha. Diverting the
discussion to P values is a distraction.

I think that pretty much summed up our disagreement. I
think the meaning and definition of p-value is RELEVANT,
to make the use of p-value consist and equivalent to the
use of a fixed alpha level test.

In BOTH cases, it IS relevant to know "when Ho is TRUE"
is part of the definition.
-- Reef Fish Bob.

******************************************************************************************************************************************

.

Relevant Pages

  • Afonso tried to drag Jerry Dallal down his own GUTTER.
    ... and there is only ONE unknown variance. ... They why are YOU dragging Jerry into your corner -- Jerry never made ... Jerry gave NO REASON why didn't use it, ... Also, see Eberhardt & Fligner, The American Statistician, 31, ...
    (sci.stat.math)
  • Re: Need Help Determining the "True" Mean of a Sample
    ... > I'm a software engineer, not a statistician, so please forgive my ... > The distribution for these samples is such that about 75% of the ... only 4% as much of the total variance. ... of dropping immediately to zero when the N is under 58. ...
    (sci.stat.math)
  • Re: How to test means when theres no variance?
    ... women using a T-test. ... pay of men is greater than the mean pay of women at a two standard ... the variance in both samples was zero. ... The company probably should bring in a statistician who has ...
    (sci.stat.math)
  • Re: Test of two Independent Proportions: How to Do It RIGHT.
    ... Reef Fish states unequivocally that the test statistic for comparing ... two proportions MUST be based on the pooled estimate of the variance. ... As the comment in The American Statistician points out, ...
    (sci.stat.math)
  • To copy is too much
    ... Bob Ling the ichthyologist disguised in statistician had posted that the formula of the variance between two populations ... This is precisely the guy that full of boast intend to be considered a statistician, ...
    (sci.stat.math)