I usually recommend using SPSS as your primary package. Although there 
are some procedures that are in other packages that are not in SPSS, on 
those occasions it is easiest to do your data prep, cleaning, and 
exploration on SPSS and then save  the file in whatever format is needed.

SPSS has major advantages in the human factors aspects.  Human factors 
such as readability, comprehensibility, consistency in language usage, 
make a major difference in the time and effort needed to use them. Based 
on my experience consulting to social scientists, accountants, and 
statisticians for the past 30 years, using the straight syntax approach, 
you can save 15 to 20% of the time it takes to write, debug, test, 
verify, and quality review SPSS procedures compared to SAS. Using the 
iterative GUI-then-paste approach really helps the development of 
analyses and the time saved should be even greater than this. As an 
extreme, I have written a 30 line SPSS syntax to do the same data 
manipulation as an 850 line SAS syntax. (The SAS syntax was written by a 
programmer with 25 years experience.) Remember that data prep, cleaning, 
and exploration take the vast majority of the time in the data portion 
of a project.

If you are planning to develop new stat procedures, it is easier to 
write them using all the data manipulation, preparation, missing data 
handling routines, etc. that are already in SPSS.

The availability of syntax is important in government work because of 
the need for documentation and quality review.

The total cost of use is more important than the original purchase price.

Hope this helps.

Art
[EMAIL PROTECTED]
Social Research Consultants
University Park, MD USA
(301) 864-5570

Dalby, James WLAP:EX wrote:
> I am curious to know what people on EdStat think about various statistical
> software packages.  Which would you say performs the best?  We are thinking
> of purchasing several copies of one program for the various offices around
> our province sometime this year.  We want to analyze all kinds of
> environmental monitoring data, such as water chemistry data, benthic
> invertebrate data, etc.  Software currently being discussed by our staff
> include JMP and Statistica.  Thanks for sharing your first-hand experiences.
> 
> 
> Dr. James E. Dalby, Jr., Aquaculture Biometrician
> BC Ministry of Water, Land, & Air Protection
> 2080A Labieux Road
> Nanaimo, British Columbia
> CANADA, V9T-6J9
> Phone (250) 751-7246, Fax 751-3103
> [EMAIL PROTECTED]
> http://wlapwww.gov.bc.ca/epd/epdpa/industrial_waste/agriculture/agri_fishf.h
> tm
> <http://wlapwww.gov.bc.ca/epd/epdpa/industrial_waste/agriculture/agri_fishf.
> htm> 
> 
> "The best thing about being a statistician is that you get to play in
> everyone's backyard".     --- John W. Tukey (American statistician,
> 1915-2000)
> 
> 
> .
> .
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> .                  http://jse.stat.ncsu.edu/                    .
> =================================================================

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to