Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Switching from SPSS to Stata


From   John Antonakis <john.antonakis@unil.ch>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Switching from SPSS to Stata
Date   Sun, 19 Jul 2009 11:34:03 +0200

Hi Yves,

I avoid SPSS like the plague! I used SPSS for a few years (and was quite competent with it when I dumped it); however, since having switched to Stata I have never had to use SPSS again. There are so many thing that one can do in Stata that one can't do in SPSS. Basic things, for example (this reflects what I knew of SPSS until 3 or so years ago): robust standard errors for non i.i.d. (identically & independently distributed) residuals, postestimation tests (linear and non-linear), selection models (Heckman type), censored models, simultaneous equations (though SPSS has not bought-out Amos, which can be bundled with SPSS; however, compared with programs like Mplus or LISREL, Amos is no good), models with constrains, robust-regression, quantile regression, and many other things (not to mention graphing). Of course, some of the stuff can be programmed in SPSS; however, the syntax is not as straightforward as Stata. Stata updates are very regular and free (Official updates are updates to Stata as it was originally shipped from StataCorp). There are the ado updates (user-written routines) centrally organized (at SSC--Statistical Software Components or the Boston College Archive: http://www.repec.org) and easily installed directly from Stata. Finally, the Stata community is not only very helpful; it is also much more methodologically savvy and sophisticated than the Spss community. Most Spss users generally estimate quite simple models (or come from an experimental background where knowledge of advanced statistics is not necessary), so, in general, they cannot help you as much as do Stata users.

When you make the jump to Stata you will not regret it.

HTH,
J.

P.S. ton anglais est parfait (et beaucoup mieux que mon français)--alors il n'est pas nécessaire de t'excuser !

____________________________________________________

Prof. John Antonakis
Associate Dean Faculty of Business and Economics
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland

Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305

Faculty page:
http://www.hec.unil.ch/people/jantonakis&cl=en

Personal page:
http://www.hec.unil.ch/jantonakis
____________________________________________________



On 18.07.2009 22:55, Yves_Therriault@ssss.gouv.qc.ca wrote:
Dear Statalisters,

I'm working as a research analyst in public health since 1990. I've always
used Spss for my statistical analysis (the current version I have is the
outdated 12.1 release). I always work with the Spss syntax language, almost
never with the GUI (maybe one or two times a year).

Earlier this year, I had the opportunity to take a look at an evaluation
copy of Stata 10. Although I'm not a statistician, it seems to me that
Stata is a far more superior statistical software that Spss. One of the
things I like about Stata, it's that Stata is a "complete package"
relatively to the Spss base software (well, it's true that it is now known
as PASW ...). In order to get with Spss what Stata offers, one has to buy
rather expensive add-on modules. One of the features I like the most about
Stata, is the possibility to analyse data from complex surveys (like the
Canadian Community Health Surveys - CCHS). It's not possible to do that
with the Spss base version. Stata also appears to me to have a much better
online help and documentation. And, to the say the least, the Stata support
staff is terrific.

In my humble opinion, the only advantage I see in Spss over Stata is that
the Spss output is nicer than Stata's. It's easy to edit an Spss output
(insert a title for example) and distribute it as an PDF file to those who
haven't Spss installed on their computer.

I'm particulary interested to hear about the experience of former strong
Spss users who decided to switch to Stata. Beside the fact Stata has much
more statistical procedures, is more powerful, and is much less expensive
to buy (as we don't have to purchase add-on modules to get "a complete set"
of statistical procedures), for what reasons did you decide to do so ?  I
would be very grateful if you could give me a few reasons that could help
me to convince the administration department here to purchase Stata instead
of the next release of Spss (version 18.0)

Thank you very much.

Kind regards,

P.S. Sincere apologies for the very bad English grammar and syntax.

Yves Therriault, Ph. D.
Agent de recherche
Surveillance de l'état de santé de la population (Santé publique)
Direction de la planification, des ressources informationnelles et
financières
Agence de la santé et des services sociaux de la Côte-Nord
Téléphone    (418) 589-9845   Poste 2312 ; Téléc. (418) 589-8574
Courriel :  yves_therriault@ssss.gouv.qc.ca


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index