Keith McCormick

I have used and taught K-means cluster for many years. I have never worried too much about sort order, but a recent experience made me revisit it. I was running a simple data set through all the IBM SPSS Statistics (SPSS) Cluster Techniques, and also ran them all through IBM SPSS Modeler (Modeler). The short […]

I think the team in the Texas A&M Stats department have done a great job in putting together this material, and providing it free on the web.  One caution is that is seems to be version 10.0.
I personally find that listening to all of them would be a bit rough in that you are just […]

I have purchased two easy, basic SPSS books.  They are both popular.  One is Pallant's "Surviving SPSS".  The other is Andy Field's.  I prefer the Field book.  Both have been brought to class by participants.  I will be writing complete reviews. One thing that concerns me is that these books are often REALLY out of […]

In anticipation of 16.0 release later this year, I am very intrigued with R.  I just found this link.  Let me know what you think. I haven't jumped in yet, but it seems very helpful.
http://oit.utk.edu/scc/RforSAS&SPSSusers.pdf 

I think the only good answer these days is both.  I recently worked with a couple of several folks whose co-workers use SPSS syntax exclusively.  I think they are missing out on some new features by doing so.
Here is a short, partial list:  Transpose, Restructure, Identify Duplicate Cases (all in the DATA menu) and Visual Binning, Date and Time Wizard […]

I recently had an exchange on SPSSX-L where someone asked about this link.  I suggested the "stat's coach" in SPSS's help for SPSS users.  But … I like the chart regardless, and it comes in handy when teaching multivariate stats.  Check it out.
http://www.datastep.com/StatisticalSelectionGuide.pdf
Another intersting version of the same thing was also on SPSSX-L.  My apologies […]

I thought this story was interesting because it illustrates that something that seems difficult like odds ratios (which appears as Exp(B) in the results) can make for an interesting article.  Explaining what we do is at least as difficult as the analysis. "A 50 percent lower risk of death" was a .5 odds ratio is someone's […]