| Date: | Wed, 16 Apr 1997 09:32:16 -0400 |
| Reply-To: | Dianne Rhodes <drhodes@FENIX2.DOL-ESA.GOV> |
| Sender: | "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU> |
| From: | Dianne Rhodes <drhodes@FENIX2.DOL-ESA.GOV> |
| Subject: | SPSS and SAS history |
|---|
When I first started doing statistical programming in 1978, I used SPSS,
the Statistical Package for the Social Sciences. It was a FORTRAN based
product, and used the same conventions as FORTRAN. That is your program
had to be typed in columns 7 thru 72, etc. It was not very flexible. SAS,
however, was PL/1 based, and free format. It was also much cheaper on I/O
than SPSS. Even though SAS charged a 25% surcharge to use the product, it
was cheaper to debug and run than SPSS. SAS was also much more proactive
in marketing and providing free training. I took a week long 40 hour
course in SAS at their expense. This may explain some of the divergence in
SAS versus SPSS users. We used SPSS X 10 on a VAX around 1985 and it would
bring the system to its knees. Now I have a user who runs a PC version and
uses it mostly for the graphics capability. We have SAS for the PC but I
have not had the opportunity to try and train these guys to use SAS instead
of Lotus, Harvard Graphics, and the other legacy applications they are
comfortable with.
************************************************************
Dianne Louise Rhodes
drhodes@fenix2.dol-esa.gov
Computer Data Systems, Inc.
US Department of Labor, Office of Workers Comp
Washington, DC 20210
************************************************************
|