In article <5vmb9i$21h@netsrv2.spss.com>, David Nichols <nichols@spss.com> wrote: >In article <341714E2.1E61266F@odin.cc.pdx.edu>, >Marc R. Feldesman <feldesmanm@pdx.edu> wrote: >>Forgive the cross-posting but I am trying to solicit honest opinions >>from a variety of sources. I will be teaching a computer-based >>statistical methods course next spring to Anthropology seniors and >>graduate students, all of whom will have had a two-term introductory >>survey of statistics. My course will be more applied and will focus on >>methods of particular use to students in archaeology and biological >>anthropology (i.e. anova, regression (bivariate and multiple linear), >>some non-parametrics, discriminant analysis, cluster analysis, principal >>components analysis, multidimensional scaling, as well as EDA. >> >>Our on-campus computing facilities are fairly limited for the size of >>our student body. Students typically find themselves waiting several >>hours for a free system. We have SAS available in one lab, SPSS >>available in several labs, Minitab widely available across campus (but >>universally despised by most faculty including this one). We had SYSTAT >>but did not renew our site license when utilization dropped >>significantly below critical threshholds. >> >>Thus, I am working toward a solution that would have most students buy >>their own software. Student versions are available for SPSS, SYSTAT, >>and Minitab. Our university will NOT increase the size of our SAS >>license to allow students to "purchase" a license. SPSS and SYSTAT have >>"graduate packs" (full versions at a significant reduction from list >>price, but still expensive - about $150 - 160 without manuals). SPSS' >>student version 7.5 has everything I need BUT is Win95 only, requires 16 >>MB RAM, and has a significant footprint. Minitab's student version has >>virtually no multivariate routines, and Systat's latest student version >>(6.0.1) will not do factor analysis, multidimensional scaling, or >>cluster analysis. All the student versions are about $60. >> >>I am leaning toward SPSS' student version but the hardware requirements >>will severely limit many of our students, thus forcing them into the >>labs I'm trying to elude. >> >>Anyone been down this path before and have any experiences they'd like >>to share before I make a decision that I'm likely to be crucified for. >>(BTW, I've used all these packages and am fairly comfortable working >>with any of them. My personal opinion is not really relevant here). >> >>Thanks for any information you might have to offer. >> >>To reply by private email, simply remove the "nospam" from my email >>address above. > >In considering the Student Version rather than the Grad Pack, remember >that ALSCAL (multidimensional scaling) is not included, and that any >specifications that require syntax will not be available in the >Student Version. > >-- >----------------------------------------------------------------------------- >David Nichols Senior Support Statistician SPSS, Inc. >Phone: (312) 329-3684 Internet: nichols@spss.com Fax: (312) 329-3668 >-----------------------------------------------------------------------------

A note on hardware requirements: the University of Chicago is recommending to students for the 1997-98 school year that those who plan to do just word processing and basic tasks like email have at least 32 MB of RAM, while those who plan to do number crunching should have at least 64 MB. SPSS looks positively ultra-efficient in that context (I have 16 MB on my PC at home, and SPSS runs very nicely). Of course, one could also say that this is a clear sign that the U of C expects people to have more money than common sense . . .

-- ----------------------------------------------------------------------------- David Nichols Senior Support Statistician SPSS, Inc. Phone: (312) 329-3684 Internet: nichols@spss.com Fax: (312) 329-3668 -----------------------------------------------------------------------------