View Full Version : stats software summary

Paul Devita
05-18-1994, 05:19 AM
calling all biomechanists:
-about 10 days ago i posted the following request:
> dear friends:
> -i am looking for a statistical software package that can do
> 1) one and two way ANOVAs; 2) repeated measures; 3) multiple
> regression.
> -my requirements are:
> 1) ibm compatible; 2) dos or windows driven; 3) software is menu
> driven where user selects particular analysis, enters data and
> results are generated. i do not want a sas type software in which the
> user must write the analytical code (eg. proc statements).
> -i would appreciate any suggestions on this matter and i will post a
> summary of them.
> -thank you.
thank you very much to the many people who took the time to respond to my reque
st. 11 statistical software were suggested that might meet my requirements.
SPSS and Systat were the two most common suggestions with 8 and 7 recommendatio
ns, each. the other 9 software suggestions each had 1 or 2 recommendations.
the responses to my request are as follows and thanks again.
paul devita
From: "Mark D. Grabiner"
Subject: Re: statistical analysis software

Hey Paul!!
About the stats package...it seems to me you will be giving up a lot in
order to meet your requisite simplicity criterion. For example, many of
the simple packages are not set up to account for missing data or unequal
cell sizes. I have been using SPSS since my early days as a graduate
student. I have it now for the PC and it does just about everything I
need...but, old dogs new tricks. I have seen some really simple packages,
some come free in cereal boxes but I have yet to see one that would cause
me to give up on old reliable. Jst lower the boom on your students and
tell them they have to learn it (it being whatever package you select).
Conversely, lower the boom on yourself and learn it. Alternatively, forget
about statistics entirely.

================================================== =====
From: abraham@utxvm.cc.utexas.edu (Larry Abraham)
Subject: Re: statistical analysis software


We are using the windows version of SPSS. It is easy to use and well
documented, though not suited for a novice in statistics. That is , you
have to know what you are doing. We have the basic as well as the advanced
and professional components. It is available through our institution at an
excellent educational discount.

Larry Abraham, EdD
Kinesiology & Health Education
The University of Texas at Austin
Austin, TX 78712 USA
(512)471-1273 FAX (512)471-8914

================================================== =============================
Subject: Re: statistical analysis software

Dear Paul :

I am a graduate student in Biomechanics. My name is Richard C.H.
Yang. I suggest two softwares of statistical analysis to you --
SPSS for WIN (or DOS ), SYSTAT ver.5.03. Now I use those for data analysis.
I think it will be good for use.

Hope that will help you a lot.

With best wishes,

Richard C.H. Yang
Div. of Biomechanics
Inst. of Biomedical Engineering
National Cheng-Kung University
================================================== ===================== =======


I've begun using SYSTAT...we have a generous site license such that the copies
of the software are free, but we must buy the manual. To indicate it's
relative ease of use, I haven't bought a manual, and I can get it to do pretty
much what I need. It is menu driven, DOS, and supports most normal designs.

let me know if you want more info.


================================================== ======================
From: Brian L Caster
Subject: Re: statistical analysis software
Hi, Paul. How are things in Illinois? We're plugging away out here at UO,
and I am rapidly nearing escape velocity with my dissertation project.
I saw your e-mail posting, and thought you might want to have a look at
the Systat statistical package. We purchased it here a couple of years ago,
although we've only put it to limited use. Systat uses code similar to
SAS, but it is interactive and menu driven, so you have the option of
either typing in the code as in SAS, or simply selecting from menus the
analyses to be preformed. Again, we haven't used it extensively. When
we first got it, we had some problems with the huge data sets that we
were dealing with, and opted just to keep going with SAS. But with the
beefy computers and RAM sizes available now, it would probably run
quite well. You might also try giving Tim Derrick, one of Joe Hamill's
doctoral students, a call. I know he has had some experience with Systat
as well. Good luck.
Brian Caster
================================================== =============================
From: Kenneth R. St. John
Subject: Re: statistical analysis software


1800 Sherman Avenue
Evanston, IL 60201-3793

************************************************** ******************
Kenneth R. St. John, Assistant Professor Voice: (601) 984-6199
Orthopaedic Research and Biomaterials Fax: (601) 984-6014
University of Mississippi Medical Center Fax: (601) 984-5151
2500 North State Street ASTM PROFS: COP004
Jackson, MS 39216-4505 Internet: stjohn@fiona.umsmed.edu
************************************************** ******************

================================================== =============================

I just received SigmaStat from Jandel Scientific. So far I think it is
wonderful as it guides you through the tests and provides a written
interpretation of the results. YOu can get it from Jandel on a 90
day trial - 415-453-6700.

Joel Vilensky
================================================== =============================

Dear Paul,

Have you considered STATISTICA 4.5 (Windows version) from StatSoft?
A collegue is using it and has "scrapped" SPSS for PC for good!

Gilmen Smyth
Gilmen Smyth
Centre universitaire Saint-Louis-Maillet
165 boul. Hebert
E3V 2S8
Tel.: 506-737-5050 (5173)
FAX.: 506-737-5373
E-Mail: gilmen@cuslm.ca

Dear Paul:

Try Minitab. I used Minitab for windows (pull down menus) which can read
in data separated by commas or tabs (or from spreadsheets like quattro
pro) for my thesis analyses. It allows ANOVA of various forms with
balanced and unbalanced systems and Multivariate analyses of variance.

Good luck
Nancy Black __|~_)_I__)_|~_
Ecole de genie )_ __)_|_)__ __)
Universite de Moncton | )____) | EMAIL: blackn@umoncton.ca
Moncton, New Brunswick \\___|____|____|____// FAX: (506) 858-4082
E1A 3E9 CANADA \\ // PHONE: (506) 858-4079
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
From: Rob Neal
Date: Wed, 11 May 1994 06:12:48 EST5EDT
Subject: Re: statistical analysis software
Priority: normal
X-mailer: WinPMail v0.99 (WRB2).

You wrote ...

Try Quattro Pro for Windows (Version 5 I think). It has a remarkable
array of statistical functions, is fully Windows compatible and very
easy to use. The ability to graph results and functions afterward
within the same package is very good.


Robert Neal, PhD
Department of Human Movement Studies
The University of Queensland

ph 61 7 365 6240
FAX 61 7 365 6877
================================================== =============================


We've had good success using Microsoft Excel. The newer versions
have some fairly powerful statistical functions. Hope this information
proves useful.

Brent Parks
Drexel/Hahnemann Biomechanics Lab
================================================== =============================

During my graduate work, I used Systat for Windows. It is a menu-driven
program with a spreadsheet-like interface for data analysis. It was
originally a command-oriented program, but a graphical front end was
added. That allows you to see the commands that the menu executes if
you are interested. I am not a statistician, but the program seemed
to be very thorough. It came with a number of what I considered to be
porely written manuals. If you have any more questions, just send


************************************************** ***********************
* Robert S. Cargill II * Use OS/2, it's real nice! *
* rcargill@biochem.dental.upenn.edu * *
* Post-doctoral Fellow * *
* My opinions, my spelling errors! * *
************************************************** ***********************

================================================== =============================

There are several software packages available that fit your particular needs.
I have used the following:
by SomeWhare In Vermont, Inc.
P.O. Box 215
Montpelier, VT 05502
This is a DOS application and is very cheap! I paid $25 and it does
everything that you want.

by Jandel Scientific
2591 Kerner Blvd
San Rafael, CA 94901
(415) 453-6700
This can be obtained in DOS or WINDOWS versions. It is very easy to use, but
is expensive ($500) and has limited analysis capabilities. The most notable
is the lack of ANCOVA.

You may also want to look at SYSTAT for WINDOWS. I have very limited
experience with it, but it is very powerful and a step up from SPSS or SAS.

Mark W. Cornwall, PhD, PT
Northern Arizona University
Dept of Physical Therapy
P.O. Box 15105
Flagstaff, AZ 86011
(602) 523-1606

================================================== =============================
You may want to look into a software package called 'Systat'. I have
extensive experience with it and have been quite happy with it. It comes in
either Mac, MS-dos or windows versions. In addition to being menu-driven,
you can also write your own commands and submit them in a batch if you want
to be more sophisticated, but you can almost do every thing you need to
without ever writing commands too. Another package which also received high
popularity ratings from my colleagues is the latest version of 'Statview',
but I'm not sure whether it comes in PC versions. It's very user-friendly
and certainly can handle all the basic statistical analysis such as ANOVAs
and Regression Analysis, but it does not have the capability of doing
non-linear model-fitting or parameter optimization as Systat does.
Therefore, Systat is still my first choice.

Good Luck!

************************************************** ******************************
:::::::Joyce Fung, PT, Ph.D.:::::::::==================================== =======
:::::::Research Associate::::::::::::=======Tel: (503) 229-7390=================
:::::::R. S. Dow Neurol Sci Inst:::::=======Fax: (503) 229-7229=================
:::::::1120 N.W. 20th Ave::::::::::::=======e-mail: joyce@nsi.lhs.org===========
:::::::Portland, OR 97209::::::::::::================================= ==========
************************************************** ******************************

================================================== =============================


since you mentioned SAS, you are probably aware of SPSS for DOS or Windows.
In case you're not: SPSS for Windows does the calculations you are looking
for. You can enter data in a spreadsheet and select the appropriate test
from menues. No code has to be entered. Since SPSS is one the most used
statistic packages worldwide, it could be recommended, but it is very
expensive, depending on the analyses modules you need. The _basic_ version
costs about >$1000 or so (don't know the US prices).
Stefan Kolodzie e-mail:
Institute for General Psychology kolodzie@ze8.rz.uni-duesseldorf.de
Heinrich-Heine-University CompuServe: 100331,3354
40225, Duesseldorf Germany Phone: +49-211-311-3057
================================================== =============================
Subject: RE: statistical analysis software

Try SPSS V6 for windows.....

================================================== =============================

I have been using SPSS for Windows. It easily imports various data files and
spreadsheet data from Lotus, Excel etc using the Windows cut and paste (DDE).
It includes all the statistical analysis techniques you could ever want
including multivariate ANOVA with repeated measures and multiple regression.
The graphics package is quite good as well. All functions are accessed via
the menuing system (no commands or macros to enter).

I hope this helps
================================================== ===================
Robert Newton Email: run1@psu.edu
Center for Sports Medicine Telephone: Int+ 1 814 865 7107
The Pennsylvania State University Facsimile: Int+ 1 814 865 7077
117 Ann Building, University Park, PA 16802
================================================== ===================

Paul - We have had very good results in terms of functionality and
ease of use, with a DOS-based package called ECSTATIC. We bought ours
in 1988 for $50.00. Vendor: SomeWare in Vermont
PO Box 215
Montpelier, VT 05602

For a Windows package, take a look at BMDP's New System for Windows.
It provides menu-driven stats with a spreadsheet-like data interface.
A very neat package for $79.00 until sometime in May. Vendor:
BMDP Statistical Software
1440 Sepulveda Blvd.
Suite 316
Los Angeles, CA 90025

Happy Computing !

- Roy Bechtel, PT
University of Maryland

e-mail: rbechtel@cosy.ab.umd.edu
================================================== =============================

You might try:

GraphPad InStat, Instant Statistics
GraphPad Software
10855 Sorrento Valley Road # 203
San Diego, CA 92121 USA
phone: int+1-619-457-3909
FAX: int+1-619-457-8141

Good luck!

Henk Schamhardt
================================================== =============================

Dear Paul,
the statistical package that we use (DOS or WIN. version,
WIN. version recommended) is SYSTAT.
It is easy to learn.
Menu driven.
It can be used in Batch mode as well that is what I prefere.
It can import various types of data, (ASCII, and from spreat sheets).
It is a complete program with additional modules available, such
as for path analysis etc.
Good luck with your search.

************************************************** ********
************************************************** ********
** **
** Vassilios Vardaxis **
** McGill University, Physical Education Department **
** 475 Pine Ave, West, P.Q. H2W-1S4, Montreal CANADA **
** **
** Email: invx@musicb.mcgill.ca **
** Fax: (514) 398-4186, Phone: (514) 398-4188 or 4184 **
** **
************************************************** ********
Hey bud, just ask. Get SPSS for Windows, Student Version. About $75. It can
do the above and more, very simply. Command lines are pulled down from the
windows, guided by the software Ie correct ordering of commands). to try it
out, call SPSS and they will send you, a research methods/statistics
instructor, a free student version. Trust me, it works; it is simply a
scaled down version of the full SPSS for Windows package (it limits # of
cases to under 200, or something like that, but it basically offers all of
the procedures).
Regarding your request, why didn't you ask for software that generates
the data, and writes and reviews the paper, too? Regards, STM
Steve McCaw
Dept of HPERD
5120 Illinois State University
Normal, Ill 61790-5120
Phone (309)438-5307 Fax: (309)438-5559 e-mail SMCCAW@ILSTU.EDU
================================================== =============================

I just threw away some literature on a new product from MathSoft.
Contact them and have them send you info on their package.
I do not have their number. Try calling 800 directory assis.

Jim Foto
Paul W. Brand Biomechanics Lab
Carville, La
================================================== ======================

SPSS for windows is very user-friendly, comprehensive and menu-driven
package and would probably be most suitable for your needs.

Anne Mannion, University of Bristol

================================================== =============================


Centinela Hospital Biomechanics Lab
================================================== =============================

I don't know if this will help but have you looked at Excel? I know that it
has ANOVA capabilities although I'm not sure about repeated measures and
whatever else you need. Just a thought. ---
************************************************** ******************************
Scott Draper Phone: (515)294-6520
1132 Veterinary Medicine Home: (515)232-2075
Biomedical Engineering Department Fax: (515)294-3932
Iowa State University e-mail:
Ames, IA U.S.A. 50011 sdraper@iastate.edu
************************************************** ******************************

I am no statistician but you may want to try

xlisp-stat for microsoft windows 3.0, version 2.1 alpha 6

It was written by Luke Tierney, School of Statistics, University of
Minnesota and is public domain.



Deakin University - Rusden Campus
662 Blackburn Rd, CLAYTON. Victoria. Australia. 3168.
Fax (03) 244-7407
Phone [03] 244-7244

================================================== =============================
In-Reply-To: Your message of Wed, 11 May 94 09:47:16 CST

BMDP Statistical Software, Inc.
1440 Sepulveda Blvd.
Los Angeles, CA 90025

(213) 479 - 7799
Fax: (213) 312 - 0161

I have BMDP PC90 version. No 800 number to my knowledge.
Dan Becque
used to use BMDP at Amherst, he may be able to help you going.

PHONE (919) 757 - 4564
From: acc@bihobl2.bih.harvard.edu (Amy C. Courtney)
Subject: Re: statistical analysis software

One such software package is RS1 (or Cornerstone) both by BBN
Software Products of Cambridge, MA

It works well. RS1 can't do two-way ANOVAs, but Cornerstone may be
able to. These packages can also plot ASCII data, and do descriptive
statistics and graphing, although RS1 graphs are not publication quality.

Last I heard, I think SAS now has a menu-driven version now too.
================================================== =============================


We are working with SPSS for windows now and I guess this is exactly
what you are looking for. The big advantage (which is possibly a
disadvantage) compared to the SPSS PC plus version is that you don't
have to write the analysis code by yourself any longer. It is totally
driven by the menu and you don't have to care for the right data
format any more.
Axel Knicker

German Sports University Cologne
Institute for athletics and gymnastics
Carl Diem Weg 6
50927 Cologne

The adress is:
SPSS inc.
444 N. Michigan Ave
Chicago, Illinois 60611