On Google-Scholar search on AUTHOR and KEYWORD (Part II)
Reef Fish
Google-Scholar search on AUTHORS on REGRESSION related topics (Part I)
I deliberately changed the SUBJECT so that when reading the groups
from Google, the parts would not overlap each other, so that only 1 of
the parts can be seen at one time (under Google Groups), because
there will be other parts. Furthermore, the discussion under
different
parts can be more focused on what's in the specific parts.
I am giving only the tabulation of TWO extremely prominent authors
and statisticians, and what Google-Scholar search on them yielded
on the TOPICS in which they are known by every statistician:
John Tukey, or John W. Tukey (the "Father" of Data Analysis") as
a discipline apart from Statistics, or "Exploratory Statistics" as
upposed to "Confirmatory Statistics".
http://stat.bell-labs.com/who/tukey/index.html
Frederick Mosteller (F Mosteller) -- a well-known Professor-Author-
Research at the Harvard Statistics and BIostatistics Departments.
Mosteller has written substantive statistical BOOKS on many areas
of statistics related to Regression and Data Analysis methods.
http://www.umass.edu/wsp/statistics/tales/mosteller.html
Mosteller is also on the Selection Committee of the MacArthur Fellows.
http://www.macfound.org/site/c.lkLXJ8MQKrH/b.959463/k.9D7D/Fellows_Program.htm
The topics (keywords) I'll tabulate on the Google-Scholar search
results are:
MR Multiple Regression
RD Regression Diagnostics
DA Data Analysis
IDA Interactive Data Analysis
LM Linear Models
ALM Applied Linear Models
MB Model Building
The counts are those given by Google-Scholar, to include the author's
BOOKS, published articles, the count of the number of books and
papers (by the authors as well as by OTHER authors) who had cited
their work in the keyward areas.
MR RD DA IDA LM
ALM MB
JW Tukey 11 1 145 2 7
0 3
John Tukey 11 2 59 2 7
0 3
Frederick Mosteller 5 1 27 0 3
0 2
F Mosteller 7 1 71 0
3 0 2
Note that the search counts could differ substantially depending on
which of the author's name to use, as in the Data Analysis category.
I find the small number of counts in EVERYONE of these categories
to be surprisingly low, especially when the count is supposed to
include articles by OTHER authors who cited the keyword.
This part is still ANOTHER illustration of the NON-scholarly result
of the searches. If my undergrad had come back fom the library
with that kind of count on those keywards I would have told such
an undergrad back to the library for more (perhaps 10 times more)
before they should come back.
With these examples, perhaps some of the interested readers can
APPEND to the tabulated counts those that are found by Google-
Scholar searches on other statistical authors because this is all the
time I have now until I am in Cozumel tomorrow. I have many more
pressing things to do (like packing :-)) before leaving for the trip.
-- Reef Fish Bob.
Reef Fish
Reef Fish ha escrito:
> This is the second part, following (Part I).
>
> Google-Scholar search on AUTHORS on REGRESSION related topics (Part I)
>
> I deliberately changed the SUBJECT so that when reading the groups
> from Google, the parts would not overlap each other, so that only 1 of
> the parts can be seen at one time (under Google Groups), because
> there will be other parts.
>
> I find the small number of counts in EVERYONE of these categories
> to be surprisingly low, especially when the count is supposed to
> include articles by OTHER authors who cited the keyword.
>
> This part is still ANOTHER illustration of the NON-scholarly result
> of the searches.
> With these examples, perhaps some of the interested readers can
> APPEND to the tabulated counts those that are found by Google-
> Scholar searches on other statistical authors because this is all the
> time I have now until I am in Cozumel tomorrow. I have many more
> pressing things to do (like packing :-)) before leaving for the trip.
>
> -- Reef Fish Bob.
Since nobody took up the suggestion of tabulating counts on OTHER
statisticians by Google-Scholar search, I'll provide the Google-Scholar
of ONE statistician I know, to use as yet ANOTHER illustration of how
USELESS the Google-Scholar search is!
> MR RD DA IDA LM ALM MB
>
> JW Tukey 11 1 145 2 7 0 3
> F Mosteller 7 1 71 0 3 0 2
>
RF Ling 8 7 20 3 5
1 3
Bob Ling 0 0 0 0 0
0 0
To draw any kind of conclusion about these statistical authors'
contribution to the statistical topics (or compare their contributions)
is AS ABSURD AS using Google-Scholar for the bean counting in
the first place! Google-Scholar is USELESS -- which is actually
the reason Richard Ulrich used it for Cohen. Richard Ulrich had
a count of 2 under MR and 0 in the other categories.
As mentioned earlier, now I know why Ulrich kept calling me Bob
Ling even though I NEVER post with that name OR used that name
in my professional publications! It's just another one of Ulrich's
devious schemes of ad-hominem attack on me, because TWO
different readers in sci.stat.math had posted (and I learned it from
THEM) that the proper way to do a Google search for my
professional work is to use the WEB search option under Google,
and search the name "RF Ling" in quotes OR the name "Robert
F Ling" in quotes.
On the other hand, if anyone wishes to search for KEYWORDS
of what I have posted in these newsgroups, all one has to do is
to go to the "advanced groups search" option of Groups Google,
and search for the keywards and author "Reef Fish", with or
without the quotes, and one would find what I had posted with
the keywards under all the different posting IDs of Reef Fish.
THAT is by far the most important reason I use the posting name
of Reef Fish which I have used since 1992, because even though
my posting IDs (email addresses) had changed many times, the
same author "Reef Fish" would find ALL of them.
For example, if you use that advanced search for AUTHOR
Reef Fish and keywords "linear models", you would get
Resultados 1 - 10 de 44 for su consulta
or 44 threads, many of them featuring Richard Ulrich's various
ERRORS and BLUNDERS under the topic of "linear models".
There is hardly any statistical topic you would not find something
I've posted in the sci.stat.* groups since I began posting in them
in 2005.
If no one is doing the Google-Scholar count on the topics above
on their favorite statistician or some well-known statisticians on
those topics, then I'll do them for George Box, John Neter and
NR Draper (famous Regression and Linear Models authors),
and RD (Dennis) Cook and Sanford Weisbreg (well-known for
their regression and regression diagnostics work).
In each and every case, it'll show how INADEQUATE and USELESS
the Google-Scholar search is!
-- Reef Fish Bob.
Richard Ulrich
Google Scholar.
- Google Scholar, or any citation index based on content,
is far better at finding NAMES than at finding subject matter
of the citations. It doesn't give instructions for using subject
matter along with the name, but that happens to work, as a default.
Those interested in a citation reference may read here.
http://scholar.google.com/intl/en/scholar/help.html
Any count in the hundreds, in Google Scholar, is very high,
since "1000" is quite a decent count for citations in the indexes
that you have to pay for, and Google is seeing only a tenth or so
of that number -- from what I remember reading earlier.
Any count on a *subject* in Google Scholar is probably
"high" if it is more than a dozen, since the subject is usually
missing. Try it out yourself. Bob has gone crazy, getting
sidetracked by that feature. I was intrigued by how *very*
high it was for J. Cohen but I did not intend to suggest that
"subject" was any preferred way to use the tool. Last week
was the first time that I used it for 'subject' and the short counts
were immediately apparent. The main use should be by searching
for the author, or perhaps for a title.
On 28 Jul 2006 14:40:42 -0700, "Reef Fish"
<Large_Nass...@yahoo.com> wrote:
>
> Reef Fish ha escrito:
[snip; Bob is going to give counts on Tukey and Mosteller]
> >
> >
> > The topics (keywords) I'll tabulate on the Google-Scholar search
> > results are:
> >
> > MR Multiple Regression
> > RD Regression Diagnostics
> >
> > DA Data Analysis
> > IDA Interactive Data Analysis
> >
> > LM Linear Models
> > ALM Applied Linear Models
> >
> > MB Model Building
> >
> > The counts are those given by Google-Scholar, to include the author's
> > BOOKS, published articles, the count of the number of books and
> > papers (by the authors as well as by OTHER authors) who had cited
> > their work in the keyward areas.
> >
> > MR RD DA IDA LM ALM MB
> >
> > JW Tukey 11 1 145 2 7 0 3
> > John Tukey 11 2 59 2 7 0 3
> >
> > Frederick Mosteller 5 1 27 0 3 0 2
> > F Mosteller 7 1 71 0 3 0 2
> >
> >
> > I find the small number of counts in EVERYONE of these categories
> > to be surprisingly low, especially when the count is supposed to
> > include articles by OTHER authors who cited the keyword.
> >
> > This part is still ANOTHER illustration of the NON-scholarly result
> > of the searches.
- "NON-scholarly result"? -
No, that illustrates Bob's unwillingness to figure out what Google-
Scholar is good for, and then to use it for what it is good for.
First, look for the author. Consider variations on the author's name
and variations on the title of the book or article if you want to,
but the 'primary' count shows *something*. "Hundreds" is good,
and thousands is even better. The number of *hits* for the name
you give can be indicative, too, if the results are not confounded
by too many people who have published with the same searchable
name. The largest counts are (usually) for textbooks
I found it interesting to think about the number of hits for
a name, and the number of citations for a reference. It does say
something about the credibility of the author, and the uniqueness
of the test, or the elegance of the advice, and so on.
Tukey has 2144 'primary' citations for his textbook on exploratory
data analysis, and there are 4000 *hits* for "j-tukey" which
seem to mostly be him. One book can show up twice, or even
dozens of times, if "citers" manage to use different editions or
spellings or mis-spellings. Tukey is well represented in the
database. His article with Cooley has 1367 citations, the only
article I've noticed today with over 1000. (Having written that,
I looked up N-Mantel. He has 3 articles with 1773 [group of 3],
1756 [group of 2], and 2005.)
"F-Mosteller" has 2530 hits (not citations), most of which seem to
be him, if numbers 1-20 and 81-100 are representative. The
largest citation count I see is 454, for his textbook with Tukey.
Draper & Smith have 5179 citations listed for their textbook. Huge.
Neter has 3327+830+639 citations for his textbooks on linear
models. Huge.
JD Cook has 1049 citations for finite elements, 330 for another
book, 626 for his book with Weisberg. Weisberg has 1007
for one other book.
All the folks Bob mentioned, above, are well-cited, though none
of them out-poll Cohen, whom Bob wants to disparage. Here are
some others that I looked up, for further comparison.
The Snedecor-Cochran textbook, which has slews of particular
tests, has 12,327 citations.
The textbook by Zar has one of the highest citation counts I've
Googled, 17,616. Zar gives many tests, and he also provides good
historical contexts and references for them.
Agresti has 3540 and 1060 for books on categorical data analysis.
Frank Harrell has 472 citations for his book on logistic regression.
Searle, variance components, has 870.
[snip]
>
> As mentioned earlier, now I know why Ulrich kept calling me Bob
> Ling even though I NEVER post with that name OR used that name
> in my professional publications! It's just another one of Ulrich's
> devious schemes of ad-hominem attack on me, because TWO
Bob is inventing again.
His paranoia? - a bit. It is some more of that blather from the
top of his brain, that he never has figured how to censure.
No superego, a Freudian psychiatrist would say.
"Not a lick of common sense," I'll say.
He followed that with another repetition of his failed reasoning --
My use of "Bob Ling" does not hamper, in the slightest, anyone's
search on "Reef Fish". It does allow Bob's casual acquaintance
(who knows Bob is a statistician who spends a lot of time on-line,
and little more) to find his less-than-graceful appearances here.
- "Stranger-searching" is justification number 3, these days.
I think it never occurred to me until Bob posted his inane assertion
about "searching". I've mentioned the other two reasons recently
enough to omit them here.
[snip, about how to search for "Reef Fish", etc.]
--
Rich Ulrich, wpi...@pitt.edu
http://www.pitt.edu/~wpilib/index.html
Reef Fish
Richard Ulrich ha escrito:
> Here is a commentary for the benefit of everyone curious about
> Google Scholar.
One example is sufficient to counter any number of pages of
commentary on how INEFFICTIVE and MISLEADING Google-
Scholar is, compared to just any ordinary Google-web search,
with the appropriate keywords!
John Neter, the co-author of the widely used textbook, "Applied
Linear Models" which I have used for nearly 30 years through
several editions, had ZERO count, when searched with the
keyword "applied linear models".
RF Ling, under the identical search, had ONE count, and the
count was on his SIAM Review of a different book, but having
cited the book "Applied Linear Models" by Neter et al.
I cannot think of ANY search, under ANY circumstances that
could give as silly and absurd results as the one cited above.
-- Reef Fish Bob.