Re: Testing the Equality of Two Population Proportions


Jerry Dallal wrote:

I am NOT going to read an obscure paper in a 2nd rate journal by a
third or 4th rate author just because a professor who made an error
on his webpage couldnt articulate a freshman level SPECIAL CASE.
on using Z to test Ho: p1 = p2 = ZERO.

A very, very special case.

All other relevant reasons that I'll be too busy the next couple of
weeks
are beside the point, in YOUR evasion and absence of even a single
valid response to WHY you ignore the Ho condition other than you made
a mistake (a careless one) on your webpage.

That is excusable. Very excusable.

Your EVASION of an answer is not.

You EXCUSE for anyone to read the Am Stat paper to answer a simple
Freshman Level question taught in ALL Freshman textbooks is the
most laughable excuse I've seen since I have been reading this group,
and I've seen plenty of laughable excuses, from just Afonso alone, the
#1 who point to your webpage mistake. And Afonso #2 pointed to the
same mistake for support, not knowing Afonso #1 had already done so
and had been thoroughly debunked. Now comes the webpage professor
himself, NOT able to make a simple retraction or correction of his
error,
and unable to give an explanation other than pointing to a pointless
paper
in the American Statistician.

Is that how you actually TEACH?

That is a pretty sad commentary of the American System of "higher
education" I must say.

-- Reef Fish Bob.


There's nothing to retract. You asked for references. I gave you
references. YOU asked for them. You got them. NOW READ THEM!!!

Perhaps someone else would like to read them for you and defend your
position. We shall see.

No, Jerry. Your Humpty Dumpty Theatre of the Absurd days are over.


Watch this simple history of this thread, Professor Dallal.

I had asked for an explanation to a VERY specific problem of
testing Ho: p1 - p2 = ZERO, using a large sample Z approximation,
using a pooled estimate of the variance, as it should, as opposed
to the error in your webpage, referenced by two different Afonsos.

You gave a whole bunch of irrelevan books to the SPECIFIC issue.
in your Opening Post, plus the Am Stat paper.
Not a SSINGLE word of explanation.

The rest of the thread was nothing more than your continued
EVASION.

I asked for a SPECIFIC explanation of the Ho: p1 - p2 = ZERO
case using a large sample Z test as to WHY you deviate from
the standard textbook treatment of that topic, again.

You gave NO EXPLANATION.

There's nothing to retract. You asked for references. I gave you
references. YOU asked for them. You got them. NOW READ THEM!!!

I asked for NO reference. If I had asked for any reference to the
SPECIFIC Ho: p1-p2 = ZERO using Z, it would be an elementary
textbook reference, as in a first course of statistics.

You kept pointing to the American Statistician article.

1. That's not an elementary textbook.
2. That's Jerry Dallal's evasion of an explanation.
3. I didn't ask for it. It's your FAILURE to give an explanation
for your own ERROR on the webpage cited by Afonso.

Perhaps someone else would like to read them for you and defend your
position. We shall see.

I don't need anyone to read it for me. I had stated my position.

Jerry Dallal had NO DEFENCE for his ERROR of NOT using
the pooled variance for the Ho: p1-p2 = ZERO case,

or explain why he use two estimates for p when "Ho is TRUE"
plains says p1 = p2 = p = ONE p which is required in the
definition of alpha and p-value.

NO ONE needs to speak for Jerry Dallal's inability to explain
his own error, or his NON-response as his only remaining
evasion tactic.

I repeat for you, Jerry, what I had already said, while you
ignored it and kept digging deeper and deeper into the hole
you had already dug:

You EXCUSE for anyone to read the Am Stat paper to answer a simple
Freshman Level question taught in ALL Freshman textbooks is the
most laughable excuse I've seen since I have been reading this group,
and I've seen plenty of laughable excuses, from just Afonso alone, the
#1 who point to your webpage mistake. And Afonso #2 pointed to the
same mistake for support, not knowing Afonso #1 had already done so
and had been thoroughly debunked. Now comes the webpage professor
himself, NOT able to make a simple retraction or correction of his
error, and unable to give an explanation other than pointing to a
pointless paper in the American Statistician.

Is that how you actually TEACH?

That is a pretty sad commentary of the American System of "higher
education" I must say.

A VERY SAD Commentary, Jerry.

Jerry Dallal> YOU asked for them. You got them. NOW READ THEM!!!

I asked for an explanation. I got NONE, and nothing but
evasion from Jerry.

Now stop spreading your "chicken manure" (I am borrowing a fond
expression of my mentor, who Jerry Dallal knew, the late Jimmie
Savage, who likes to use) to describe the kind of un-scholarly
substance, now being used by Jerry Dallal. If Jerry's name were
"Bliss", I am sure Savage would have said, "Ignorance is Bliss" too.

I had told Jerry, more than once privately, that I would treat his
ERRORS
no differently from I treat the errors by anyone else in this forum.
So,
this should come as no surprise to Jerry. Or to be precise, "Jerry
asked for it".

..-- Bob.

.

Relevant Pages