Stata and SPSS
Work in progress
A tale of two packages
I use both these packages, often daily. Proponents of each camp would probably say that's a bad thing. SPSS is too backward and inconsistent, Stata is too difficult to learn. Or something. If there's one key difference between the two it probably lies in the user community. Among my contacts, SPSS seems to be widely used by generalist social scientists, Stata more the province of economists.
Their respective web pages are quite revealing. SPSS invites you to "Discover secrets your spreadsheet can't tell you", "make your organisation more successful" and "increase customer value". Stata's site tells you it is "software for statistical analysis", used by "medical researchers, biostatisticians, epidemiologists, economists, [...] and other research professionals needing to analyze data."
A large chunk of users want to read some data, create some new variables and then analyse the results. This is how you might do a simple task in SPSS:
and this is how you might do it in Stata:
Both programes give you alternative ways to do each, and to write each. As it happens, Stata gave me results more quickly whilst the SPSS results were much easier to read (better labelling). I would defy anyone to say that either coding was inherently better than the other.
Of course there are some areas covered by Stata, but not by SPSS, and indeed the reverse, which will be decisive for some users. As will issues of cost and local availability.
SPSS resourcesUCLA SPSS Portal
Raynald's SPSS Tools, perhaps the best and largest guide to SPSS syntax, macros and scripts.
Internet Guide to SPSS for Windows written by Wolfgang Ludwig-Mayerhofer and good for starting out
Stata resourcesUCLA Stata Portal
Stata's website, with user support, books about statistics and stata, gift shop ...
Stata's Listserv archives
Internet Guide to Stata written by Wolfgang Ludwig-Mayerhofer, for those starting out