SPSS: Does not play well with others
SPSS does some pretty cool things, and I have written about some of them here. However, there are also some annoyances. First of all, it is one of the hardest packages to use moving datasets from one format or platform to another. The regular SPSS dataset, the one that ends in .sav crashes many mail and file transfer programs when you try to email or upload it.
Recognizing this problem, SPSS created the .por (for portable) file that makes it easier to move files across systems. These files can be easily uploaded or downloaded. I routinely move mine from a Mac to a Unix server to a Windows machine. For students who are used to clicking on an Excel file and having it download, it can be a bit daunting when you click on a .por file it comes up with gibberish that looks something like this.
`1233 SPSS some stuff 12397346 9400- -9774 45556 “ ~~~ 34455
for an entire page.
And yes, I know the solution is to just right-click on the file name and then you can select “Save Target As” or “Save Link As” and then save the file to disk. For once, the problem cannot be blamed on Internet Explorer (the archetypal does-not-play-well-with-others) and Microsoft’s on-going efforts to take over the world, because this gibberish-appearing trick happens with every browser.
Incredibly, SPSS does not even play well with itself. I discovered this less than lovely characteristic when students sent me .spo output files from SPSS 15 and I could not read these because I had not installed the legacy viewer that comes with SPSS 16 for Windows.
Well, I guess no one is perfect.
Lately, I have been playing with Stata, which, although it is not going to replace any time soon as my all favorite statistical software, is starting to grow on me, like that unattractive, ill-behaved child in a class that slowly charms himself into being the teacher’s pet. Stata: Bart Simpson of statistical packages.
I guess that would make SAS, Lisa and SPSS Maude Flanders.
SPSS sav files are binary files. There is nothing special about that. Most modern mail systems handle these fine. I email them routinely. I don’t know of any reason why they should be harder to email than SAS or Stata or, for that matter, Excel files.
And it should never cause a problem with ftp as long as you select binary mode for the transfers.
SAV files are themselves portable across all the supported SPSS platforms.
As for spo vs spv files, SPSS Version 16 was a massive rewrite of the entire frontend, including a switch to a portable Java technology so it could be released simultaneously on all platforms. The old spo format used the Microsoft compound document format and caused many problems over the years. The new format is XML based and will be much more compatible across versions and platforms. The Legacy Viewer on the installation CD provides the bridging technology for the old files.
Regards,
Jon Peck
Well, I said that SPSS .sav files crash mail programs because I have had it happen on several computers. In fact, I was wondering if it was just me, and I called over one of my co-workers, he tried it on my computer, suspecting I had not mastered the intricacies of the Fetch program on the Mac (hmph). It crashed again. I gave him a copy on a flash drive, he tried it on his computer, same problem. We tried it on a couple of other computers in the lab and it didn’t work on any of them. We tried emailing it and that didn’t work. So, we said, “Well, I’ll be damned,” and went back to work. Contrary to appearances, we do occasionally do some work.
That was months ago. So, when I got your comment, I tried it again and now it works! Why the difference? I don’t know. We have all kinds of security protections up the wazoo (technical term) and sometimes things don’t work here that work other places. At first I thought, “Oh, they fixed it with SPSS 17,” but I went to a computer where I had SPSS 16 installed, tried it there and it works fine.
My only conclusion is that somewhere in the latest rounds of upgrades, reinstalls and patches we did of everything between semesters, whatever was causing the problem got fixed.
Thanks a lot for posting, though,because if you hadn’t, I never would have tested it again.
I haven’t had a chance to play with SPSS 17 too much but since I am teaching five Intro to SPSS workshops in March and April, I guess I better get cracking.