Newsgroup sci.stat.consult 22026

Directory

http://www.geocities.com/Paris/8892/gallery.htm

On Wed, 15 Jan 1997, Paige Miller wrote:
>
> Steven T Barlow wrote:
> >
> > I have two regression equations.  Both use the same IVs but different
> > DVs. Is there a test for significance for difference between the R^2 for
> > the two equations?
The first question that I have is, "Are the two regression equations
computed on the same sample?"
> In most situations that I am familiar with, I cannot even imagine why
> you would want to do such a test. I don't expect the IVs to have the
> same predictive ability for different DVs. Perhaps you could provide
> more details of what the 2 DVs are and why you want to know if the R^2
> are the same.
>
> And anyway, in direct answer to your question, no I don't know any
> statistical test; the only answers I am aware of are based upon subject
> matter knowledge -- a 50% R^2 may be great for DV 1, while a 51% R^2
> might be considered poor for DV 2.
Second, are you really interested in the difference between the two R^2s,
or are you interested in differences in the regression surfaces?  I don't
know a test to compare the two R^2s, but you could use multivariate
multiple regression to compare the regression coefficients for the two
DVs.  In all my years, I have only come across this type of question once.
One of my students was interested in trying to assess the degree to which
a set of aptitude variables similarly predicted achievement over content
measured in two different ways (essay vs multiple choice).
______________________________________________________________________________
William B. Ware, Professor and Chair                   Educational Psychology
CB# 3500                                               EMAIL:   wbware@unc.edu
University of North Carolina                           PHONE:   (919)-966-5266
Chapel Hill, NC      27599-3500                        FAX:     (919)-962-1533
URL:http://www.unc.edu/~wbware/
______________________________________________________________________________

Steven T Barlow wrote:
> I have two regression equations.  Both use the same IVs but different
> DVs. Is there a test for significance for difference between the R^2 for
> the two equations?
There is a nice little formula for a smaller situation in Hinkle, Wiersma,
& Jurs (2nd edition, but not 3rd edition) *IF* you only have one
predictor.  Sounds like you have more than one, so more complicated
approaches are warranted.
Here's what I'd do:  Create a path model in which all predictors covary
and all have direct paths to both dependent variables.  The dependent
variables should be standardized (which does take some liberties with the
underlying Wishart distributional assumptions, but...).  In this path
model, both DVs have error terms.  You may wish to allow these error
terms to covary, indicating that you believe that the two DVs may covary
above and beyond their mutual causal influences (i.e., the common set
of predictors).  The variances of the error terms represent 1-R^2 for both
of the DVs.  Before submitting your model to a structural equation
modeling program (LISREL, EQS, AMOS, CALIS, Mx, Ramona,...), constrain the
two error variances to be equal.  Run the model with and without this
constraint, do a chi-square difference test between the fit of the two
models to see if the constraint yields significantly worse fit.  If so,
then the two error variances must differ significantly, i.e., the 1-R^2
values differ significantly.  The error with the smaller variance has the
significantly larger R^2.  (Note: the comparison of two models may also be
conducted just on the constrained model using a LAgrange Multiplier test
available in some programs).
Good luck.
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm
Gregory R. Hancock                       "Just because all your friends are
Department of Educational Measurement,    doing this structural equation
   Statistics, and Evaluation             modeling thing doesn't mean you
1230 Benjamin Building                    have to.  If all your friends
University of Maryland                    jumped off a cliff...."
College Park, MD  20742-1115                                     -- my mom
phone: (301)405-3621     fax: (301)314-9245   e-mail: ghancock@wam.umd.edu
    Check out our graduate program in measurement, stats, and evaluation:
       http://www.inform.umd.edu:8080/EdRes/Colleges/EDUC/Depts/EDMS
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm

Karen Gilbert wrote:
> I am seeking references for a p-value correction described by Holm.
The original article is as follows:
Holm, S.  (1979).  A simple sequentially rejective multiple test
procedure.  Scandinavian Journal of Statistics, 6, 65-70.
Holm's method, as well as a number of more powerful options, are described
in a recent paper in Review of Educational Research.  At the risk of
self-promotion, the partial reference is below:
Hancock, G. R., & Klockars, A. J.  (1996).  The quest for alpha:
Developments in multiple comparison procedures in the quarter century
since Games (1971).  Review of Educational Research.
(I don't have the page numbers here at home, but it is the last 1996
issue.  If you want it and have trouble getting it, let me know.)
Good luck,
Greg
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm
Gregory R. Hancock                       "Just because all your friends are
Department of Educational Measurement,    doing this structural equation
   Statistics, and Evaluation             modeling thing doesn't mean you
1230 Benjamin Building                    have to.  If all your friends
University of Maryland                    jumped off a cliff...."
College Park, MD  20742-1115                                     -- my mom
phone: (301)405-3621     fax: (301)314-9245   e-mail: ghancock@wam.umd.edu
    Check out our graduate program in measurement, stats, and evaluation:
       http://www.inform.umd.edu:8080/EdRes/Colleges/EDUC/Depts/EDMS
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm

I am interested in studying some physiological processes over time, such
as GSR, blood pressure and skin temperature.  I'm looking for suggestions
as to how to statistically analyze such data.  Also, references which
deal with this would be appreciated.
Thanks.
Jean
-- 
Jean Petree
petrjean@cwis.isu.edu

                             41st Annual
                      Fall Technical Conference
                                 1997
                           Call for Papers
                "Mining Data for Quality Improvement"
                       Omni Inner Harbor Hotel
                         Baltimore, Maryland
                         October 16-17, 1997
Co-sponsored by:
   American Society for Quality Control
      - Chemical and Process Industries Division
      - Statistics Division
   American Statistical Association
      - Section on Physical and Engineering Sciences
Applied and expository papers are  needed  for  parallel  sessions  in
Statistics, Quality Control, and Tutorial / Case Study.
Detailed submission instructions are available on the Web
   http://www.sas.com/ftc97/
or you can request them from one  of  the  following  members  of  the
program committee:
   Susan L. Albin
   Department of Industrial Engineering
   Rutgers University
   PO Box 909
   Piscataway, NJ  08855-0909
   phone: 908-445-2238
   email: salbin@rci.rutgers.edu
   FAX: 908-445-5467
   Sharon Fronheiser (to whom paper correspondance should be addressed)
   Eastman Kodak Company
   151 Mill Hollow Crossing
   Rochester, NY  14626
   phone: 716-588-2014
   email: sharonf@kodak.com
   FAX: 716-722-4415
   Randy Tobias (to whom electronic correspondance should be addressed)
   SAS Institute Inc.
   SAS Campus
   Cary, NC  27513-2414
   tel: 919-677-8000 x7933
   email: sasrdt@unx.sas.com
   FAX: 919-677-8123
The submission  process will  start on  August 1, 1996 and conclude on
January 17, 1997.  Papers should be  strongly justified by application
to  a problem in  quality  control,  or  the  chemical,  physical,  or
engineering sciences.  The mathematical level of papers may range from
none,  to  that  of  the  Journal of  Quality  Technology, or  that of
Technometrics.
-- 
Randy Tobias          SAS Institute Inc.     sasrdt@unx.sas.com
(919) 677-8000 x7933  SAS Campus Dr.         us024621@interramp.com
(919) 677-8123 (Fax)  Cary, NC   27513-2414
   Faith, faith is an island in the setting sun.
   But proof, yes: proof is the bottom line for everyone.
                                                       -- Paul Simon

 - Q on eigenvalues ...
: >I computed the eigenvalues and eigenvectors of the following
: >covariance matrix using Matlab and code from numerical recipes in 'c'.
: >
: >They both return the same eigenvalues, but the signs of the eigenvectors
: >of the 2 smallest eigenvalues (0.0238, 0.0782) are reversed. Can
: >someone shed some light on this for me?
 - one responder
: I ran into a similar problem back in the early 1980's, the signs of
: the eigenvectors from SAS were the reverse of the signs calculated
: in GENSTAT.
: I asked the local SAS consultant, and the answer was simple.  SAS,
: supposedly, determined if there were more positive or negative values
: in the elements of the vectors.  If there were more negative values,
: the program reversed signs to save printing all those minus signs.
: I thought he was bluffing, but never got a more satifactory answer.
: Different algorithms, perhaps?
 - If "v" is an eigen vector as a solution, then "-v" is also a solution.
You can use the signs that an algorithm spits out, or you can use some
rule about reversing all signs if most values, or most BIG values
(by some criterion)  are negative.
Rich Ulrich, wpilib+@pitt.edu

     In article <32d84448.82298631@news.otago.ac.nz>,
     agray@commerce.otago.ac.nz
     (Andrew Gray) writes:
     |>     I'm working on combining neural networks and fuzzy logic models
     |> with statistical techniques (regression and data reduction) for
     |> software metrics (for example, predicting development time based on
     |> the type and size of system).  While there has been a lot of work n
     |> neural-fuzzy, neural-genetic, fuzzy-genetic, etc. type systems ...
     AbTech may be able to help you.  Contact Norman Fink at
     normanf@abtech.com

Does anyone know of a good textbook on basic statistics using Excel
that  covers the use of the Add-In Data Analysis? 
The textbook should cover descriptive statistics, t-tests,
non-parametrical stat. up to ANOVA.
-- 
 Mats Sjoquist                 Voice  (+)46 18 174181
 Dept of Physiol & Med Biophys Fax    (+)46 18 553541 
 Biomedicum, Box 572           E-mail  MatsS@Physiology.uu.se
 S-751 23 Uppsala, Sweden      WWW     http://fysms2.medfys.uu.se

At Mon, 13 Jan 1997 15:27:44, Don  wrote:
>Hello,
>
>I computed the eigenvalues and eigenvectors of the following
>covariance matrix using Matlab and code from numerical recipes in 'c'.
>
>They both return the same eigenvalues, but the signs of the eigenvectors
>of the 2 smallest eigenvalues (0.0238, 0.0782) are reversed. Can
>someone shed some light on this for me?
>Thanks
>Don
I ran into a similar problem back in the early 1980's, the signs of
the eigenvectors from SAS were the reverse of the signs calculated
in GENSTAT.
I asked the local SAS consultant, and the answer was simple.  SAS,
supposedly, determined if there were more positive or negative values
in the elements of the vectors.  If there were more negative values,
the program reversed signs to save printing all those minus signs.
I thought he was bluffing, but never got a more satifactory answer.
Different algorithms, perhaps?
Cheers,
CWR
-------------------------              --------------------------
C.W. Ramm                              14762cwr@msu.edu
Forest Biometry                        ##########################
Department of Forestry                 I'd rather be chasing elk!
Michigan State University              ##########################
East Lansing, MI   48824-1222

   I recently came across some old data of mine from a previous research project. I discovered I could use this data to answer another research question that has been on my mind for some time (but had no time to start another research project for it).
   The question I have is about the proper experimental design I should
be using to analyze this previous research project. This problem is both
challenging and infuriating at the same time. I hope someone could help
me here. However, my research is about soil science, so I may have to
give a little background on this
research.
   In this research, I'm interested to determine whether the stability of
a soil's various aggregate sizes are equal to each other or not. All
soils are made up of aggregates of various sizes. You can think of an
aggregate as sort of a "granule", so it would not be incorrect to think
of a soil made up of granules of different sizes - some large granules
and some small granules. Please note that these aggregates or granules
are formed naturally; I did not force or treat a soil to have more or
less larger
aggregates.
   To meet this research objective, I identified 8 soil types/locations
as my soil sample collection. For each soil type, I took 10 samples
randomly from the field. And for each of that 10 samples, I separated the
soil samples into 6 aggregate sizes. I then measured the stability of
each aggregate
size.
   In other words, for each soil type, I have 10 soil samples which I
would later separate each soil sample into their respective 6 aggregate
sizes. I would then measure the stability of each aggregate size for
every soil sample from every soil type. This can be illustrated in the
figure
below:
                                               Aggregate size
                     1	2	3	4	5	6
              -----------------------------------------------------------------------------------------
Soil	      1   G1	   G1		G1	     G1 	   G1	 
G1
type	     2	 G2	  G2	       G2	    G2		  G2	 
G2
		3   .							 
 .
		4   .							 
 .
		5   .							 
		     .		
		6   .							 
 .
		7   .							 
 .
		8  G8	      ...	     ...	     ...	 
   ...		 
G8
where Gi is the i-th group of soil samples, where each group has 10 soil
samples.
   The questions are these:
       (1) what's the experimental design; that is, is there a way I can
determine whether the stability of a soil's 6 aggregate sizes are equal
to each
other?
       (2) can I treat the 8 soil types and 6 aggregate sizes as
treatments? To me, the soil types and aggregate sizes are more of a
"classification", rather than an intentional
treatment.
   I think I should analyze my experimental as a repeated measures design
because a soil sample is separated into its 6 aggregate sizes, and  the
stability of each aggregate sizes was then determined. So in a way, the
same soil sample is used for all 6 "treatments" (6 aggregate sizes).
Would this repeated measures design be effected because there is no
random assignment of "subjects" (or samples) in the treatments? Remember,
for every soil type, I randomly collected 10 soil samples. I didn't
randomly assign 80 samples (8 soils with 10 samples) into the treatments
-- this is
impossible.
   Any takers? I know this message is long, but thanks anyway if you read
this far
down.
 - Christopher Teh
-------------------==== Posted via Deja News ====-----------------------
      http://www.dejanews.com/     Search, Read, Post to Usenet

Mats Sjoquist  wrote:
> 
> Does anyone know of a good textbook on basic statistics using Excel
> that  covers the use of the Add-In Data Analysis?
> 
> The textbook should cover descriptive statistics, t-tests,
> non-parametrical stat. up to ANOVA.
One book I have seen is "Learning Business Statistics with
Microsoft Excel" by John L. Neufeld. (Prentice Hall 1997)
ISBN 0-13-234097-6
It covers ANOVA and regression, but leaves out the non-parametric
tests. I have found it to be quite thin and very short on examples
and exercises. On the other hand, it goes into great detail on the
Excel side.
-------------------------------------------------------
gus gassmann          (Horand.Gassmann@dal.ca)
School of Business Administration, Dalhousie University
Halifax, Nova Scotia, Canada , B3H 1Z5
ph. (902) 494-1844
fax (902) 494-1107
http://ttg.sba.dal.ca/sba/profs/hgassmann/hgassman.html

On 15 Jan 1997 22:32:41 GMT, tseck@gibbs.oit.unc.edu (Chiu Kit Jessica
Tse) wrote:
>I calculated the weighted proportion and standard error
>of a rare factor.
>
>I got the proportion as 0.0175 and
>      the standard error as 0.011895.
>
>The corresponding 95% Confidence Interval is
>   -0.0058, 0.0408
>
>How do I explain the negative lower C.I.?
>I know my math is correct.   
To be more precise, your _arithmetic_ is correct - your _statistics_
is not.
The method you're using, adding or subtracting 1.96*SE, doesn't work
well for this problem, because the method is (a) only an
approximation, even under the best of circumstances; and (b)
inherently symmetric; you need an asymmetric CI when p is close to
zero. 
Exact confidence limits for proportions can be found in many intro
stats texts, e.g., Zar JH (1984) Biostatistical Analysis. Englewood
Cliffs NJ: Prentice-Hall. (2nd ed) p.378.
Jay Weedon.

Hi Clay:
Just curious. What does SPSS stand for?
Is this a stand-alone app, library, or OCX?
Dan Markham
Clay Helberg  wrote:
>Tim Dierauf wrote:
>> 
>> Greetings Group,
>> 
>> I am working with a company that is attempting to trend compressor
>> degradation with software using linear regression and Windows-95.   The
>> application is written in visual basic.  I would like to get them to use
>> some more robust diagnostics such as deleted residuals and/or Cook's
>> Distances along with an F-test to determine model strength.
>> I can write these routines for them, but would rather use a commercial
>> debugged package.  Does anybody know if there are any packages available
>> in DLL or Visual Basic callable forms that provide statistical
>> distributions such as the Normal, F, and Student-t?  Better yet,
>> regression analysis and diagnostics?
>> 
>> Thanks in advance,
>> Tim
>
>(Disclaimer: I work for SPSS, so my opinions may be biased....)
>
>Tim--
>
>I agree with you that a commercial package is the way to go--by the time
>you write your own code for the thing, the time you've spent on it will
>have cost the company far more than the cost of the commercial package.
>
>As for interfacing with VB, consider SPSS 7.5 for Windows. It provides
>OLE object links to most of the SPSS's features, and allows you to open
>SPSS data files, manipulate them, call statistical procedures, generate
>graphs and plots, and access the results, all from your VB (or C++, or
>Powerbuilder, or whatever) application. We will soon be releasing a book
>for developers filled with examples of VB, C++ and Powerbuilder code for
>building custom applications which access the procedures and features in
>SPSS.
>
>For more information, see .
>
>						--Clay
>--
>Clay Helberg         | Internet: helberg@execpc.com
>Publications Dept.   | WWW: http://www.execpc.com/~helberg/
>SPSS, Inc.           | Speaking only for myself....

chris@agri.upm.edu.my wrote:
:    In this research, I'm interested to determine whether the stability of
: a soil's various aggregate sizes are equal to each other or not. All
: soils are made up of aggregates of various sizes. You can think of an
: aggregate as sort of a "granule", so it would not be incorrect to think
: of a soil made up of granules of different sizes - some large granules
: and some small granules. Please note that these aggregates or granules
: are formed naturally; I did not force or treat a soil to have more or
: less larger
: aggregates.
  -- But then, as I picture it, an aggregate is the clod that results
when you've exerted X amount of force.  So, by definition, every
aggregate would have the same amount of 'stability'  if 'stability'
was the amount of force that it could withstand.  I suppose stability,
instead, could be related to the INCREASE in force, at that point,
which was required to break it into smaller pieces....
But it seems to me that soils would still, mainly, differ in the size
of their aggregates, rather than in the 'stability' of the aggregate.
Since you are wanting to analyse the latter rather than the former,
please tell me what I am misunderstanding.
  -  I can offer more thoughts, if you clear up these misconceptions.
Rich Ulrich, biostatistician                wpilib+@pitt.edu
http://www.pitt.edu/~wpilib/index.html   Univ. of Pittsburgh

I would like recommendations about SPC software. I want a system that 
will produce SPC charts for a line, thus it needs to be easy to use by 
operators, yet allow access to the data by others.
you comments appreciated
John

Mark4Flies wrote:
> 
> I am looking for information about regression models and techniques for
> cases where there is error in the independent X variable as well as the Y.
> My application is method comparison between different chemical assays.
> 
> The only solution I found in the clinical chemistry literature so far is
> Deming regression. This technique requires a separate determination of the
> error (such as standard deviation) for both variables. It uses the ratio
> of the X to Y error in the standard formulas to discount the Y variance.
> It seems to work well in my simulations but I have some misgivings. The
> results depend a lot on the ratio but getting good precision estimates in
> the real world can be difficult. I would prefer a method that obtains such
> information directly from the data itself. What do you know about Deming
> regression?
> 
> I remember another approach introduced in a SYSTAT newsletter (several
> years ago) which involved adapting the default "loss function" from a
> standard least squares form to include x error. In this case, the goal for
> the minimization algorthim is directly changed to address both errors. I
> don't have access to SYSTAT anymore but I do have JMP, MINITAB, LISP-STAT,
> and Microsoft Excel available to me. I can't find options to change the
> standard least squares goal to the one I want. Have you succeeded to use
> alternative goals? What other types or regression address my problem?
> 
> I would appreciate any direction you may offer about techniques,
> references, algorithms, or software solutions. I can translate most
> computer languages. I don't want a purely mathematical article; I trust
> the work and proof. I really need direct answers, if possible. I don't
> expect that all of my work has been done for me but some guidance would
> help a lot.
> 
> Mark Bailey
The citation classic for this problem is:
Bland, J.M. & Altman, D.G. (1986) Statistical methods for assessing
agreement between two methods of clinical measurement.  Lancet, 1986i,
pp 307-310.
Also 
Altman & Bland (1983) Measurement in medicine: the analysis of method
comparison studies.  Statistician vol 32 pp307-317.
The latest of their papers has a lot of references:
Bland & Altman (1995) comparing methods of measurement: why plotting
difference against standard method is misleading.  Lancet 1995 vol 346
pp1085-1087.
Other references are quoted in the above.  There is more recent work on
comparison of test methods using geometric mean regression, which is a
useful alternative to the above techniques, but I'd have to fax or mail
you the papers.  Please eMail direct if you require more information.
Peter Baxter

Dear List:
>  I am posting the following message for a colleague who does not have
>  internet access. 
Salary Range Information Needed
I have another year until I finish my Master’s degree in Applied Statistics 
(my Bachelor’s degree is in Math with a concentration in Statistics).
For a year and a half I have been working in clinical trials research.
My statistics experience to date includes using SAS for survival analysis
(LIFETEST,PHREG), regression (GLM, LOGISTIC, REG) and programming 
for manipulation of large data sets.  I have also attended the SAS course 
for MACROS, and I am somewhat familiar with using them in programs.
I also have VMS/VAX experience, and have used BMDP and STATXACT.
I am presently seeking a position as a statistician in the pharmaceutical
industry, and would greatly appreciate any information about the current
competitive salary range in the PA/NJ area (both what salary I can 
anticipate now and when I complete my MS program).  Also, I would 
really appreciate any general information that may be helpful. 
>  Please send replies directly to me at
>  Jacqueline_Cater@msn.com

KR-20 explicitly accounts for 
item difficulty.  KR-20 calculates
item variances as: 
p_i*(1-p_i)
where p_i is the proportion of
correct responses to item `i`
in the sample.
So, generally,variable item 
difficulty poses no problem
for KR-20.
if the items of the test
are not tau-equivalent, 
KR-20 will underestimate the
reliability of the test.
KR-21, however, is affected
by differential item 
difficulty. That is 
because it uses:
N*p_bar(1-p_bar)
where p_bar is the average
item difficulty.  So,  here
the assumption is that all
items are equally difficult.
Any standard psychometric text
should explain everything you
would want to know about KR-20
and KR-21.
HTH
Steve Gregorich
In article <32DBC7CE.4B35@earthlink.net>, haywood2@earthlink.net says...
>
>Kuder - Richardson 20
>This is cronbach's alpha for dichot. Variables
>I am interested in learning all there is to know about this measure of
>Internal consistency "reliability".
>How would questions of differing dificulty level effect this meassure? 
>That is if the questions on one test very in difficulty, how does this
>effect the KR-20.  
>
>Steve Blohm

Simpson's paradox refers to the reversal of relationship following the
collapsing over heterogeneity of multi-way contingency tables.  More
recently, it has come to refer to ANY change of relationship (i.e., magnitude
and sign) following such collapsing.  It can be seen to be a special case of
the more general problem of inappropriate cross-level inference, of which the
"ecological fallacy" and the "individualistic fallacy" are also special
cases.  It is also a very old problem, recognised at least as early as Yule
(1903).
An intriguing recent example is that provided by Wardrop (1995) in which he
argues that the much believed, but fallacious "hot-hand" in basketball arises
from just such collapsing.  That is, collapsed over heterogeneous shooters,
there is an apparent "hot-hand", but only at that aggregated level of
analysis; with shooters as the unit of analysis, no such relationship exists
(as Gilovich and Tversky have long maintained).
Anil Menon (Syracuse University, School of Engg. and Computer Science)
compiled this reference list and placed it on a "Simpson's Paradox" web-site
he had created.  Unfortunately, the web-site URL has changed (or no longer
exists):
Articles on Simpson's Paradox and Related topics
   Last updated: 19/03/96
   I got interested in Simpson's paradox while studying deception in
   Genetic Algorithms. Here is a list of articles that might be useful.
   Fortunately, John Vokey at the Department of Psychology, University of
   Lethbridge, was kind enough to post most of these references, saving
   me an ascii adventure. I have grouped the bibliography in several
   ways:
     * A Beginner's Guide,
     * Chronologically,
     * Alphabetically.
   Eventually, I may put up a topically organized list as well... Please
   inform me if I have let out any pertinent articles. Some related
   links are:
     * Simple example based on drug tests.
     * A news group discussion (may have been removed).
     * Graphical Methods for Categorical Data.
     _________________________________________________________________
A Beginner's Guide
   I would recommend that the newcomer start off with:
     * Authors : Blyth, C. R.
       Title : On Simpson's paradox and the sure thing principle.
       Source : Journal of the American Statistical Association, 67,
       1972, 364-381.
   For some real-life examples of Simpson's paradox, see Keyfitz's
   classic book.
     * Authors : Keyfitz, N.
       Booktitle : Applied mathematical demography, Wiley, New York, pp.
       385-391, 1977.
   My favorite analysis of Simpson's paradox is the one in Simon and
   Blume's excellent book:
     * Authors : Simon, C. P. and Blume, L.
       Booktitle : Mathematics for Economists, W. W. Norton and Company,
       New York, pp. 368-371, pp. 784-791, 1994.
   They explain it using Don Saari's results. The importance of his work
   in the study of ``social paradoxes'' cannot be over-emphasized. A good
   starting point to Saari's remarkable theorem is:
     * Authors : Saari, D. G.
       Title : The source of some paradoxes from social choice and
       probability.
       Source : Journal of Economic Theory, 41(1), 1-22, 1987
   Shyam Sunder's paper gives Yuji Ijiri's necessary and sufficient
   condition for Simpson's paradox to occur in the ``simplest possible
   case''. This condition is a special case of Saari's theorem, but is
   particularly clear and simple to use in practice. I had no idea
   accountants worried about such matters.
     * Authors : Sunder, S.
       Title : Simpson's reversal paradox and cost allocation.
       Source : Journal of Accounting Research, 21, 222-233, 1983.
   Finally, I urge the reader to take a look at Vaupel and Yashin's very
   readable paper on the pernicious effects of heterogeneity on
   statistical decision making. It reads like a Stephen King novel (and
   is also equally horrifying).
     * Authors : Vaupel, J. W. and Yashin, A. I.
       Title : Heterogeneity's ruses: some surprising effects of
       selection on population dynamics.
       Source : The American Statistician, 39(3), 176-185, 1985.
     _________________________________________________________________
Chronological Bibliography
The 1900's
   Authors : Yule, G. U.
   Title : Notes on the theory of association of attributes in
   statistics.
   Source : Biometrica, 2, 121-134, 1903.
The 1930's
   Authors : Thorndike, E. L.
   Title : On the fallacy of imputing the correlations found for groups
   to individuals or smaller groups composing them.
   Source : American Journal of Psychology, 52, 122-124, 1939.
The 1940's
   Authors : Deming, M. E. and Stephan, F. F.
   Title : On a least squares adjustment of a sampled frequency table
   when the expected marginal totals are known.
   Source : Annals of Mathematical Statistics, 11, 1940, 427-444.
   Authors : Lindquist, E. F.
   Title : Statistical analysis in educational research.
   Source : Boston: Houghton Mifflin, 1940.
   Authors : Deming, W. E. Title : Statistical adjustment of data.
   Source : New York: Dover Publications, Inc., 1943.
The 1950's
   Authors : Robinson, W. S.
   Title : Ecological correlations and the behavior of individuals.
   Source : American Sociological Review, 15, 351-357, 1950.
   Authors : Simpson, E. H.
   Title : The interpretation of interaction in contingency tables.
   Source : The American Statistician, 13, 238-241, 1951.
The 1960's
   Authors : Mosteller, F.
   Title : Association and estimation in contingency tables.
   Source : Journal of the American Statistical Association, 63, 1-28,
   1968.
The 1970's
   Authors : Goodman, L. A.
   Title : The multivariate analysis of qualitative data: interactions
   among multiple classifications.
   Source : Journal of the American Statistical Association, 65, 226-256,
   1970.
   Authors : Blyth, C. R.
   Title : On Simpson's paradox and the sure thing principle.
   Source : Journal of the American Statistical Association, 67, 1972,
   364-381.
   Authors : Bickel, P. J., Hammel, E. A., and O'Connell, J. W.
   Title : Sex bias in graduate admissions: Data from Berkeley.
   Source : Science, 187, 1975, 398-404.
   Authors : Bishop, Y. M. M., Fienberg, S. E., & Holland, P. W.
   Title : Discrete multivariate analysis: Theory and practice. Source :
   Cambridge, Massachusetts: The MIT Press, 1975.
   Authors : Gardner, M.
   Title : On the fabric of inductive logic and some probability
   paradoxes.
   Source : Scientific American, 234, 119- 124, 1976.
   Authors : Fienberg, S. E.
   Title : The analysis of cross-classified categorical data.
   Source : Cambridge, Massachusetts: The MIT Press, 1977.
   Authors : Keyfitz, N.
   Booktitle : Applied mathematical demography, Wiley, New York, pp.
   385-391, 1977.
   Authors : Knapp, T. R.
   Title : The unit-of-analysis problem in applications of simple
   correlation analysis to educational research.
   Source : Journal of Educational Statistics, 2, 171-186, 1977.
   Authors : Freedman, D., Pisani, R., and Purves, R.
   Title : Statistics.
   Source : W.W. Norton & Company, New York, 1978.
   Authors : Whittemore, A. S.
   Title : Collapsibility of multi- dimensional contingency tables.
   Source : Journal of the Royal Statistical Society, Ser. B., 40,
   328-340, 1978.
The 1980's
   Authors : Hintzman, D. L.
   Title : Simpson's paradox and the analysis of memory retrieval.
   Source : Psychological Review, 87, 398-410, 1980.
   Authors : Flexser, A. J.
   Title : Homogenizing the 2 X 2 contingency table: A method for
   removing dependencies due to subject and item differences.
   Source : Psychological Review, 88, 327-339, 1981.
   Authors : Martin, E.
   Title : Simpson's paradox resolved: A reply to Hintzman.
   Source : Psychological Review, 88, 372-374, 1981.
   Authors : Mantell, N.
   Title : Simpson's paradox in reverse.
   Source : The American Statistician, 36, 395, 1982.
   Authors : Saari, D. G.
   Title : Inconsistencies of weighted summation voting systems.
   Source : Mathematics of Operations Research, 7(4), 479-490, 1982.
   Authors : Shapiro, S. H.
   Title : Collapsing contingency tables -- a geometric approach.
   Source : The American Statistician, 36, 43-46, 1982.
   Authors : Wagner, C. H.
   Title : Simpson's paradox in real life.
   Source : The American Statistician, 36, 46-48, 1982.
   Authors : Kennedy, J. J. (1983)
   Title : Analyzing qualitative data. Introductory log-linear analysis
   for behavioral research.
   Source : New York: Praeger Publishers, 1983.
   Authors : Sunder, S.
   Title : Simpson's reversal paradox and cost allocation.
   Source : Journal of Accounting Research, 21, 222-233, 1983.
   Authors : Knapp, T. R.
   Title : Instances of Simpson's paradox.
   Source : College Mathematics Journal, 16, 209-211, 1985.
   Authors : Paik, M.
   Title : A graphic representation of a three-way contingency table:
   Simpson's paradox and correlation.
   Source : The American Statistician, 39, 53-54, 1985.
   Authors : Vaupel, J. W. and Yashin, A. I.
   Title : The deviant dynamics of death in heterogeneous populations.
   Source : Sociological Methodology, Tuma, N. B. (ed), pp. 179-211,
   1985.
   Authors : Vaupel, J. W. and Yashin, A. I.
   Title : Heterogeneity's ruses: some surprising effects of selection on
   population dynamics.
   Source : The American Statistician, 39(3), 176-185, 1985.
   Authors : Cohen, J. E.
   Title : An uncertainty principle in demography and the unisex issue.
   Source : The American Statistician, 41, 1986, 32-39.
   Authors : Saari, D. G.
   Title : The source of some paradoxes from social choice and
   probability.
   Source : Journal of Economic Theory, 41(1), 1-22, 1987
   Authors : Saari, D. G.
   Title : Symmetry, Voting and Social Choice
   Source : The Mathematical Intelligencer, 10(3), 32-42, 1988.
   Authors : Kaigh, W. D.
   Title : A category representation paradox.
   Source : The American Statistician, 43(2), 92-97, 1989.
   Authors : Wermuth, N.
   Title : Moderating effects of subgroups in linear models.
   Source : Biometrika, 76, 81-92, 1989.
The 1990's
   Authors : Freehling, J. S.
   Title : Simpson's paradox and database profiling.
   Source : Direct Marketing, 53(5), 26-27, 1990.
   Authors : Haunsperger, D. B. and Saari, D. G.
   Title : The lack of consistency for statistical decision procedures.
   Source : The American Statistician, 45(3), 252-255, 1991.
   Authors : Klay, M. P. and Wesley, L. P.
   Title : Simpson's paradox: a maximum likelihood solution.
   Source : SRI International Technical Report, No. 502, 1-11, 1991.
   Authors : Mittal, Y.
   Title : Homogeneity of subpopulations and Simpson's Paradox.
   Source : Journal of the American Statistical Association, 86(413),
   167-172, 1991.
   Authors : Abramson N. S., Kelsey S. F., Safar P., and Sutton-Tyrrell
   K.
   Title : Simpson's paradox and clinical trials: What you find is not
   necessarily what you prove.
   Source : Annals of Emergency Medicine 21, pp. 1480-1482, 1992.
   Authors : DeBlois, B. M.
   Title : Simpson's Paradox.
   Source : Mathematica Militaris, 3(1), 1992.
   Authors : Mehrez, A., Brown, J. R., and Khouja, M.
   Title : Aggregate efficiency measures and Simpson's paradox.
   Source : Contemporary Accounting Research, 9(1), 329-342, 1992.
   Authors : Rogers, A.
   Title : Heterogeneity and selection in multistate population analysis.
   Source : Demography, 29(1), 31-38, 1992.
   Authors : Gunter, B.
   Title : A trio of statistical double takes.
   Source : Quality Progress, 26(6), 84-86, 1993.
   Authors : Simon, C. P. and Blume, L.
   Booktitle : Mathematics for Economists, W. W. Norton and Company, New
   York, pp. 368-371, pp. 784-791, 1994.
   Authors : Wardrop, R. L.
   Title : Simpson's Paradox and the Hot Hand in Basketball.
   Source : The American Statistician, 49, 24-28, 1995.
     _________________________________________________________________
Alphabetical Bibliography
   Authors : Abramson N. S., Kelsey S. F., Safar P., and Sutton-Tyrrell
   K.
   Title : Simpson's paradox and clinical trials: What you find is not
   necessarily what you prove.
   Source : Annals of Emergency Medicine 21, pp. 1480-1482, 1992.
   Authors : Bickel, P. J., Hammel, E. A., and O'Connell, J. W.
   Title : Sex bias in graduate admissions: Data from Berkeley.
   Source : Science, 187, 1975, 398-404.
   Authors : Bishop, Y. M. M., Fienberg, S. E., & Holland, P. W.
   Title : Discrete multivariate analysis: Theory and practice. Source :
   Cambridge, Massachusetts: The MIT Press, 1975.
   Authors : Blyth, C. R.
   Title : On Simpson's paradox and the sure thing principle.
   Source : Journal of the American Statistical Association, 67, 1972,
   364-381.
   Authors : DeBlois, B. M.
   Title : Simpson's Paradox.
   Source : Mathematica Militaris, 3(1), 1992.
   Authors : Cohen, J. E.
   Title : An uncertainty principle in demography and the unisex issue.
   Source : The American Statistician, 41, 1986, 32-39.
   Authors : Deming, W. E. Title : Statistical adjustment of data.
   Source : New York: Dover Publications, Inc., 1943.
   Authors : Deming, M. E. and Stephan, F. F. Title : On a least squares
   adjustment of a sampled frequency table when the expected marginal
   totals are known.
   Source : Annals of Mathematical Statistics, 11, 1940, 427-444.
   Authors : Fienberg, S. E.
   Title : The analysis of cross-classified categorical data.
   Source : Cambridge, Massachusetts: The MIT Press, 1977.
   Authors : Flexser, A. J.
   Title : Homogenizing the 2 X 2 contingency table: A method for
   removing dependencies due to subject and item differences.
   Source : Psychological Review, 88, 327-339, 1981.
   Authors : Freedman, D., Pisani, R., and Purves, R.
   Title : Statistics.
   Source : W.W. Norton & Company, New York, 1978.
   Authors : Freehling, J. S.
   Title : Simpson's paradox and database profiling.
   Source : Direct Marketing, 53(5), 26-27, 1990.
   Authors : Gardner, M.
   Title : On the fabric of inductive logic and some probability
   paradoxes.
   Source : Scientific American, 234, 119- 124, 1976.
   Authors : Gunter, B.
   Title : A trio of statistical double takes.
   Source : Quality Progress, 26(6), 84-86, 1993.
   Authors : Goodman, L. A.
   Title : The multivariate analysis of qualitative data: interactions
   among multiple classifications.
   Source : Journal of the American Statistical Association, 65, 226-256,
   1970.
   Authors : Haunsperger, D. B. and Saari, D. G.
   Title : The lack of consistency for statistical decision procedures.
   Source : The American Statistician, 45(3), 252-255, 1991.
   Authors : Hintzman, D. L.
   Title : Simpson's paradox and the analysis of memory retrieval.
   Source : Psychological Review, 87, 398-410, 1980.
   Authors : Kaigh, W. D.
   Title : A category representation paradox.
   Source : The American Statistician, 43(2), 92-97, 1989.
   Authors : Kennedy, J. J. (1983)
   Title : Analyzing qualitative data. Introductory log-linear analysis
   for behavioral research.
   Source : New York: Praeger Publishers, 1983.
   Authors : Keyfitz, N.
   Booktitle : Applied mathematical demography, Wiley, New York, pp.
   385-391, 1977.
   Authors : Klay, M. P. and Wesley, L. P.
   Title : Simpson's paradox: a maximum likelihood solution.
   Source : SRI International Technical Report, No. 502, 1-11, 1991.
   Authors : Knapp, T. R.
   Title : The unit-of-analysis problem in applications of simple
   correlation analysis to educational research.
   Source : Journal of Educational Statistics, 2, 171-186, 1977.
   Authors : Knapp, T. R.
   Title : Instances of Simpson's paradox.
   Source : College Mathematics Journal, 16, 209-211, 1985.
   Authors : Lindquist, E. F. Title : Statistical analysis in educational
   research.
   Source : Boston: Houghton Mifflin, 1940.
   Authors : Mantell, N.
   Title : Simpson's paradox in reverse.
   Source : The American Statistician, 36, 395, 1982.
   Authors : Martin, E.
   Title : Simpson's paradox resolved: A reply to Hintzman.
   Source : Psychological Review, 88, 372-374, 1981.
   Authors : Mehrez, A., Brown, J. R., and Khouja, M.
   Title : Aggregate efficiency measures and Simpson's paradox.
   Source : Contemporary Accounting Research, 9(1), 329-342, 1992.
   Authors : Mittal, Y.
   Title : Homogeneity of subpopulations and Simpson's Paradox.
   Source : Journal of the American Statistical Association, 86(413),
   167-172, 1991.
   Authors : Mosteller, F.
   Title : Association and estimation in contingency tables.
   Source : Journal of the American Statistical Association, 63, 1-28,
   1968.
   Authors : Paik, M.
   Title : A graphic representation of a three-way contingency table:
   Simpson's paradox and correlation.
   Source : The American Statistician, 39, 53-54, 1985.
   Authors : Rogers, A.
   Title : Heterogeneity and selection in multistate population analysis.
   Source : Demography, 29(1), 31-38, 1992.
   Authors : Robinson, W. S.
   Title : Ecological correlations and the behavior of individuals.
   Source : American Sociological Review, 15, 351-357, 1950.
   Authors : Saari, D. G.
   Title : Inconsistencies of weighted summation voting systems.
   Source : Mathematics of Operations Research, 7(4), 479-490, 1982.
   Authors : Saari, D. G.
   Title : The source of some paradoxes from social choice and
   probability.
   Source : Journal of Economic Theory, 41(1), 1-22, 1987
   Authors : Saari, D. G.
   Title : Symmetry, Voting and Social Choice
   Source : The Mathematical Intelligencer, 10(3), 32-42, 1988.
   Authors : Shapiro, S. H.
   Title : Collapsing contingency tables -- a geometric approach.
   Source : The American Statistician, 36, 43-46, 1982.
   Authors : Simon, C. P. and Blume, L.
   Booktitle : Mathematics for Economists, W. W. Norton and Company, New
   York, pp. 368-371, pp. 784-791, 1994.
   Authors : Simpson, E. H.
   Title : The interpretation of interaction in contingency tables.
   Source : The American Statistician, 13, 238-241, 1951.
   Authors : Sunder, S.
   Title : Simpson's reversal paradox and cost allocation.
   Source : Journal of Accounting Research, 21, 222-233, 1983.
   Authors : Thorndike, E. L.
   Title : On the fallacy of imputing the correlations found for groups
   to individuals or smaller groups composing them.
   Source : American Journal of Psychology, 52, 122-124, 1939.
   Authors : Vaupel, J. W. and Yashin, A. I.
   Title : Heterogeneity's ruses: some surprising effects of selection on
   population dynamics.
   Source : The American Statistician, 39(3), 176-185, 1985.
   Authors : Vaupel, J. W. and Yashin, A. I.
   Title : The deviant dynamics of death in heterogeneous populations.
   Source : Sociological Methodology, Tuma, N. B. (ed), pp. 179-211,
   1985.
   Authors : Wagner, C. H.
   Title : Simpson's paradox in real life.
   Source : The American Statistician, 36, 46-48, 1982.
   Authors : Wardrop, R. L.
   Title : Simpson's Paradox and the Hot Hand in Basketball.
   Source : The American Statistician, 49, 24-28, 1995.
   Authors : Wermuth, N.
   Title : Moderating effects of subgroups in linear models.
   Source : Biometrika, 76, 81-92, 1989.
   Authors : Whittemore, A. S.
   Title : Collapsibility of multi- dimensional contingency tables.
   Source : Journal of the Royal Statistical Society, Ser. B., 40,
   328-340, 1978.
   Authors : Yule, G. U.
   Title : Notes on the theory of association of attributes in
   statistics.
   Source : Biometrica, 2, 121-134, 1903.
--
Dr. John R. Vokey, Associate Professor, Department of Psychology
University of Lethbridge, Lethbridge, Alberta, CANADA  T1K 3M4
mailto:vokey@hg.uleth.ca  http://www.uleth.ca/~vokey

In article <32DD26B6.7BCF@ny.ubs.com>, David Rothman  wrote:
> given:
> 
> 1. an observed time series which is stationary according to an ADF test
> 2. can be reasonably fitted with a low order ar representation 
> 3. a limited number of initial observations ( say 100), but each day
>    one obs gets added and the series will grow in length until a time
> when
>    the series ceases to be stationary and is then discarded (the
> stationarity
>    will be tested on either the whole series, or on a sliding (100 day)
>    window concept - im not sure which...yet)
> 
> 
> question:
> i need a way to estimating say the 5% tails of the underlying population
> of the Y's.  looking @ the tail pts of the sample doesnt do me much
> good, nor does assuming normality of the dist and using a simple
> variance
> idea.
> 
> therefore i can :
> (1) flesh out the distribution each day by simulating the ar process
> with
>     estimated parameters (over a 100 day moving window) and then get
>     the tail pts (id rather monte carlo it than calcing sigma directly
> for 
>     other reasons).
> 
> or
> 
> 2. bootstrap ==> given its a time series (and that its serially
> correlated),
> i guess i am forced to bootstrap by blocks.  this makes it a bit more
> difficult.
> 
> or 
> 
> 3. ???
> 
> 
> comments?  any help wud be appreciated....thanks, dave
> 
> ps. i actually have n time series (all with a small # of obs) and need
> to get a methodology robust enuf to generalize.
Have you thought about a chained-R technique factor analysis?
Jay Lee
-- 
Jay Lee, Teaching Fellow
Department of Health & Human Performance
College of Education
University of Houston
Houston, Texas 77204-5331
USA

Hi,
I am a new user.  I am professional statistical analyst, but I don't know how I 
will ask some question.  Could you help me please.
Best regards
Mirjana Stojanovic

Chiu Kit Jessica Tse (tseck@gibbs.oit.unc.edu) wrote:
: I calculated the weighted proportion and standard error
: of a rare factor.
: I got the proportion as 0.0175 and
:       the standard error as 0.011895.
: The corresponding 95% Confidence Interval is
:    -0.0058, 0.0408
: How do I explain the negative lower C.I.?
: I know my math is correct.   
: Could someone please help me understand this.
  -- You are using a formula that treats proportions as if they
are linear, since your CI is symmetric around the point estimate 
of the mean (.0175).  That formula is never exact, and you have
run into one of the examples where it is obviously, grossly, "in
error" by extending beyond (0,1).
The stat text by Zar has one accessible formula which is based on
more reasonable assumptions about the extremes.
Rich Ulrich, biostatistician                wpilib+@pitt.edu
http://www.pitt.edu/~wpilib/index.html   Univ. of Pittsburgh

HI,
My name is Mirjana Stojanovic. I am profesional statistical analyst. I realy like 
to ask some question and to communicate with people but I am new member and I don't 
know how.  Could you please help me to start. I find that this news groups is 
exactly for me, I am loking almost evry day, but I don't know how I will ask 
somebody some question.
Best regards
MIRJANA STOJANOVIC

Hello,
I have a copy of MSP, a program for Mokken Scale
developed by DISC, University of Amsterdam.
I have an old copy version 1.50 (1988) and
I have lost the connection with them and they
send me nothing in all this years.
Please, can anyone give some address to contact
about MSP and in general about software for IRT
Thanks in advance
Jose

One of the best books on math I have got is "The Mathematics of Games"
by John D. Beasley.   Golf and football (soccer I think) are a few of
the subjects covered in that book.
Mats Liljedahl
mats.liljedahl@mbox200.swipnet.se

Hi,
I am a new user.  I am professional statistical analyst, but I don't know how I 
will ask some question.  Could you help me please.
Best regards
Mirjana Stojanovic

In article <5bm7jv$dml@usenet.srv.cis.pitt.edu>,
  wpilib+@pitt.edu (Richard F Ulrich) wrote:
> 
> chris@agri.upm.edu.my wrote:
> 
> :    In this research, I'm interested to determine whether the stability of
> : a soil's various aggregate sizes are equal to each other or not. All
> : soils are made up of aggregates of various sizes. You can think of an
> : aggregate as sort of a "granule", so it would not be incorrect to think
> : of a soil made up of granules of different sizes - some large granules
> : and some small granules. Please note that these aggregates or granules
> : are formed naturally; I did not force or treat a soil to have more or
> : less larger
> : aggregates.
> 
>> wpilib+@pitt.edu replied:
>>   -- But then, as I picture it, an aggregate is the clod that results
>> when you've exerted X amount of force.  So, by definition, every
>> aggregate would have the same amount of 'stability'  if 'stability'
>> was the amount of force that it could withstand.  I suppose stability,
>> instead, could be related to the INCREASE in force, at that point,
>> which was required to break it into smaller pieces....
>> 
> But it seems to me that soils would still, mainly, differ in the size
> of their aggregates, rather than in the 'stability' of the aggregate.
> Since you are wanting to analyse the latter rather than the former,
> please tell me what I am misunderstanding.
> 
>   -  I can offer more thoughts, if you clear up these misconceptions.
> 
> Rich Ulrich, biostatistician                wpilib+@pitt.edu
Soil aggregates are not made up of homogenous materials, meaning that a
4-mm aggregate is different than a 2-mm aggregate in more ways than just
their size. Because the composition (or constituents) of aggregates
varies according to size, I was wondering if the stability of various
aggregate sizes would still be equal with each other despite their size
and composition differences -- this is what my research is
about.
Thanks for your response. Looking foward to your help.
-------------------==== Posted via Deja News ====-----------------------
      http://www.dejanews.com/     Search, Read, Post to Usenet

Thomas Hinders wrote:
> 
> A Doctor friend of mine is recording patient family information with regards
> to the occurrence of cancer in family groups.  Currently he records the
> information and then plots it on paper using symbols to reflect male and
> female and annotating the symbols with lines, or other marks, to indicate that
> cancer was present, cured or if death was attributed to the cancer.
> 
> Makes for a VERY effective visual display of  the information, and is much
> easier for to see the effect of family (i.e. genetics) on cancer in a family.
> 
> The Doctor is recording the information on paper, using a template of sorts to
> draw the symbols, and he has asked me to look into the availability of
> programs (Win31/95 based) that would allow him to enter the information from
> his PC and create a family chart, both on screen (probably pretty easy) and
> print-out on a laser quality printer (probably NOT so easy).
> 
> He is happy to use whatever symbols the program provides,or to create his own
> and use them within the program.
> 
> Does anyone in this group have any suggestion....most of the family tree type
> programs are for more typical "who was my great great grandfather" and not esp
> suited to this type of information recording.
> 
> Please reply directly to my email address, our news server unstable.
> 
> Thanks in advance......................
> 
> Tom Hinders & Dr. George Gowen
> 
> Thomas_Hinders@lotus.com
There is a product called "Plot," (a.k.a. "Spyglass Plot")which I've
nerver seen or used, that advertises that it's easy to use and highly
customize-able. There's another one called "PlotIt." Both are found in a
catalogue called "Engineering Solutions Direct" - 1-800-898-9044 (USA)

Richard Scott wrote:
> 1) Is this a feasible area of research?  I have been working on it in a
> basic way given my own limited resources and time for the past few years.
> 2) Following from 1) is that at all useful, employment-wise do you think?
Two words . . . "sports gambling" . . . where do you think the SPREADS
come from? Hold your nose and you'll get rich. Hello Las Vegas . . .

In response to a request for a good introductory book on nonparametric
statistics, I would recommend 
Conover, WJ (1980) Practical Nonparametric Statistics, 2nd Ed,  published
by Wiley
Dan

Hi,
you can ask any question the same way you asked this question.
So, go ahead and ask.
Dan

Fisher’s exact test for testing independence in a 2x2 contingency table 
is available on-line at http://nlh10.nlh.no/~matfola/fisher.htm
Oyvind Langsrud
==========================================
HOMEPAGE: http://nlh10.nlh.no/~matfola/
==========================================

Jose Garcia de Abreu  wrote:
>Hello,
>
>I have a copy of MSP, a program for Mokken Scale
>developed by DISC, University of Amsterdam.
>
>I have an old copy version 1.50 (1988) and
>I have lost the connection with them and they
>send me nothing in all this years.
>
>Please, can anyone give some address to contact
>about MSP and in general about software for IRT
>
MSP is now in version 3.
iec ProGAMMA distribute the programme in Europe
You will find more info at
Http://www.gamma.rug.nl
>
>Thanks in advance
>Jose
--Baard
----
Baard Liaboe, Department of Sociology, University of Bergen, 
N-5007 Bergen, Norway - E-mail: Bard.Liabo@sos.uib.no
Phone : +47 55 58 91 67- Fax: +47 55 58 91 99

Hi,
I am a new user.  I am professional statistical analyst, but I don't know how I 
will ask some question.  Could you help me please.
Best regards
Mirjana Stojanovic

Gordon Johnson wrote:
> 
> 101614.1440@compuserve.com wrote:
> 
> >Hello out there,
> We're here ( or hereabouts)!
> >I am looking for population census data and data for
> >population estimates of countries, administrative areas
> >and especially of towns and places. Until now I only
> >found some data from the US (though it seems not complete,
> >because I could not find population estimates eg for
> >Metairie, Louisiana; Citrus Heights, California, or
> >places of Hawaii like Hilo).
> 
> >Is there anyone who can help me to find sources,
> >especially on the web? Please email me.
> 
> >regards
> >S. Helders
> **I am not clear wherether this is a request for current,
> present-day data, or historical population data, this having
> appeared in a genealogy newsgroup. Can you clarify, as it makes a
> big difference about possible sources.
> 
> Gordon Johnson's homepage -
> http://www.wintermute.co.uk/~kinman/
> (With Scottish genealogical goodies)
Check ot the KINDS Project at: http://www.midas.ac.uk/kinds/
-- 
=======================================================================
 Andrew Peter Johnston     Telephone: +44 (0)161-745 5443
 Research Assistant        Facsimile: +44 (0)161-745 8169
 IT Institute              Telex: 668680 (Sulib)
 University of Salford     E-Mail: a.p.johnston@iti.salford.ac.uk
 Salford M5 4WT ENGLAND    http://www.salford.ac.uk/iti/apj/andy1.html
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

On Thu, 16 Jan 1997 12:01:23 -0600, chris@agri.upm.edu.my wrote:
I'm no stats expert but perhaps a spli-plot arrangement would work.
Main plots would be the 8 soil types
Sub plots would be the 10 random samples
The final groupings would be the actual analyzed data
Here is the first rep of the trial. I don't know if you replicated the
whole thing? Obviously it is better to design the trial and then take
the data but this may work for you. SAS or some similar program should
handle the analysis
If you had replications the model might look something like this:
BLOC MP BLOC*MP SP MP*SP (can't remember if this will give the right
residual)
PLOT  BLOC MP       SP	GROUP
1	1	1	1	1
2	1	1	9	2
3	1	1	2	2
4	1	1	5	3
5	1	1	4	1
6	1	1	10	5
7	1	1	7	3
8	1	1	3	6
9	1	1	8	2
10	1	1	6	3
11	1	7	6	4
12	1	7	4	1
13	1	7	5	6
14	1	7	9	4
15	1	7	2	5
16	1	7	1	3
17	1	7	3	1
18	1	7	7	2
19	1	7	8	4
20	1	7	10	3
21	1	8	10	2
22	1	8	7	1
23	1	8	9	1
24	1	8	2	1
25	1	8	3	4
26	1	8	8	3
27	1	8	4	5
28	1	8	6	6
29	1	8	1	6
30	1	8	5	3
31	1	5	6	2
32	1	5	2	4
33	1	5	3	6
34	1	5	1	5
35	1	5	9	1
36	1	5	8	2
37	1	5	7	4
38	1	5	10	2
39	1	5	5	5
40	1	5	4	2
41	1	4	6	3
42	1	4	4	3
43	1	4	3	5
44	1	4	8	4
45	1	4	10	6
46	1	4	9	2
47	1	4	7	1
48	1	4	2	3
49	1	4	5	4
50	1	4	1	1
51	1	6	4	5
52	1	6	2	3
53	1	6	6	2
54	1	6	5	4
55	1	6	8	1
56	1	6	3	5
57	1	6	1	5
58	1	6	10	4
59	1	6	7	6
60	1	6	9	3
61	1	2	3	2
62	1	2	5	4
63	1	2	7	5
64	1	2	10	3
65	1	2	6	2
66	1	2	4	4
67	1	2	9	1
68	1	2	8	6
69	1	2	1	4
70	1	2	2	3
71	1	3	1	4
72	1	3	9	4
73	1	3	3	3
74	1	3	6	2
75	1	3	5	6
76	1	3	7	5
77	1	3	2	2
78	1	3	10	3
79	1	3	8	1
80	1	3	4	3
>   I recently came across some old data of mine from a previous research project. I discovered I could use this data to answer another research question that has been on my mind for some time (but had no time to start another research project for it).
>
>   The question I have is about the proper experimental design I should
>be using to analyze this previous research project. This problem is both
>challenging and infuriating at the same time. I hope someone could help
>me here. However, my research is about soil science, so I may have to
>give a little background on this
>research.
>
>   In this research, I'm interested to determine whether the stability of
>a soil's various aggregate sizes are equal to each other or not. All
>soils are made up of aggregates of various sizes. You can think of an
>aggregate as sort of a "granule", so it would not be incorrect to think
>of a soil made up of granules of different sizes - some large granules
>and some small granules. Please note that these aggregates or granules
>are formed naturally; I did not force or treat a soil to have more or
>less larger
>aggregates.
>	
>   To meet this research objective, I identified 8 soil types/locations
>as my soil sample collection. For each soil type, I took 10 samples
>randomly from the field. And for each of that 10 samples, I separated the
>soil samples into 6 aggregate sizes. I then measured the stability of
>each aggregate
>size.
>
>   In other words, for each soil type, I have 10 soil samples which I
>would later separate each soil sample into their respective 6 aggregate
>sizes. I would then measure the stability of each aggregate size for
>every soil sample from every soil type. This can be illustrated in the
>figure
>below:
>
>                                               Aggregate size
>                     1	2	3	4	5	6
>              -----------------------------------------------------------------------------------------
>Soil	      1   G1	   G1		G1	     G1 	   G1	 
>	
>G1
>type	     2	 G2	  G2	       G2	    G2		  G2	 
>	
>G2
>		3   .							 
>			
> .
>		4   .							 
>			
> .
>		5   .							 
>		     .		
>		6   .							 
>			
> .
>		7   .							 
>			
> .
>		8  G8	      ...	     ...	     ...	 
>   ...		 
>G8
>
>where Gi is the i-th group of soil samples, where each group has 10 soil
>samples.
>
>   The questions are these:
>
>       (1) what's the experimental design; that is, is there a way I can
>determine whether the stability of a soil's 6 aggregate sizes are equal
>to each
>other?
>
>       (2) can I treat the 8 soil types and 6 aggregate sizes as
>treatments? To me, the soil types and aggregate sizes are more of a
>"classification", rather than an intentional
>treatment.
>
>   I think I should analyze my experimental as a repeated measures design
>because a soil sample is separated into its 6 aggregate sizes, and  the
>stability of each aggregate sizes was then determined. So in a way, the
>same soil sample is used for all 6 "treatments" (6 aggregate sizes).
>Would this repeated measures design be effected because there is no
>random assignment of "subjects" (or samples) in the treatments? Remember,
>for every soil type, I randomly collected 10 soil samples. I didn't
>randomly assign 80 samples (8 soils with 10 samples) into the treatments
>-- this is
>impossible.
>
>   Any takers? I know this message is long, but thanks anyway if you read
>this far
>down.
>
> - Christopher Teh
>-------------------==== Posted via Deja News ====-----------------------
>      http://www.dejanews.com/     Search, Read, Post to Usenet
James E. Blatta
Manager, Agronomix Software, Inc.
P.O. Box 67, Portage la Prairie, MB
CANADA R1N 3B2
blatta@agronomix.mb.ca

Hi,
I am a new user.  I am professional statistical analyst, but I don't know how I 
will ask some question.  Could you help me please. Also I have licenced copy of 
SPSS 7.5,  and I need help some times with spme explanation.
Best regards
Mirjana Stojanovic

tseck@gibbs.oit.unc.edu (Chiu Kit Jessica Tse) wrote:
>I calculated the weighted proportion and standard error
>of a rare factor.
>
>I got the proportion as 0.0175 and
>      the standard error as 0.011895.
>
>The corresponding 95% Confidence Interval is
>   -0.0058, 0.0408
>
>How do I explain the negative lower C.I.?
>I know my math is correct.   
>
>Could someone please help me understand this.
>Thanks a lot.
>
The formula you must have used is the one given in most introductory 
textbooks.  You solve the following for p, the "population" proportion.
(phat-p)^2 < Chi*phat*(1-phat)/n
where phat is the sample proportion and chi is the appropriate percentage 
point from a chi-square dist. with 1 df (use 3.84 for 95% CI).
You are estimating the variance using the sample 
estimator, phat(1-phat)/n.  
Particularly when the proportion is small, the sample estimator of the 
variance does not perform well...coverage probability is not correct and 
intervals are typically too wide.
Now, instead of solving the above inequality, you could use one proposed 
by Quesenberry and Hurst (1964), Goodman (1965) and discussed by Fleiss 
(1981) Stat. Meth. for Rates and Proportions.
(phat-p)^2 < Chi*p*(1-p)/n
Expand to a quadratic and solve for p.  You won't run into the "negative" 
estimates.  This interval performs very well in practice.  It is formed 
by inverting the "Pearson's chi-square" goodness-of-fit test.
Despite the warnings by Fleiss and others, texts still advocate the 
simpler formula using p(1-p).  The QH intervals are not difficult to 
calculate and perform very well for binomial samples.
Hope this helps,
Warren.

Daniel B. Markham wrote:
> Just curious. What does SPSS stand for?
This has evolved over the years as our market has changed. Back in the
old days when our product was mainly used by social science researchers,
it was Statistical Package for the Social Sciences. Now that we've
broadened our customer base to include more business statistics, I think
they've changed it to Statistical Product and Service Solutions. Anyway,
to most people in the statistics community, SPSS has become a name of
its own, in much the same sense that "IBM" and "AT&T;" have.
> >As for interfacing with VB, consider SPSS 7.5 for Windows. It provides
> >OLE object links to most of the SPSS's features, and allows you to open
> >SPSS data files, manipulate them, call statistical procedures, generate
> >graphs and plots, and access the results, all from your VB (or C++, or
> >Powerbuilder, or whatever) application. We will soon be releasing a book
> >for developers filled with examples of VB, C++ and Powerbuilder code for
> >building custom applications which access the procedures and features in
> >SPSS.
> Is this a stand-alone app, library, or OCX?
SPSS is a stand-alone application, but the libraries (DLL's) are
accessible through OLE. Of course, if you're developing applications for
redistribution which use SPSS libraries, you'll need to either license
SPSS for redistribution, or make sure that your customers all have their
own copies of SPSS to use with your application.
As for the licensing arrangements, I don't really know any of those
details. I'll see if I can find out how it works from the appropriate
people, and post a summary.
						--Clay
--
Clay Helberg         | Internet: helberg@execpc.com
Publications Dept.   | WWW: http://www.execpc.com/~helberg/
SPSS, Inc.           | Speaking only for myself....