You have probably noticed I’m not much into R (though I’m slowly coming around to it). It goes back to when I was in my graduate statistics program, where we were required to use SPlus (R’s parent language—as far as I can tell, it’s the same thing, but with customer support).
We were given a half hour tutorial and an incomprehensible text, and sent off to figure it out how to use SPlus on graduate level stats.
Not fun.
And since I was already fluent in SAS, SPSS, and BMDP (may it rest in peace), I resisted SPlus. A lot.
I actually wish R had been around, and I wish all the great resources for learning it that exist now, existed then.
Here’s one of them, created by our very own R instructor, David Lillis. An R tutorial series that will get you started with R. Enjoy.
If you want to learn more about the syntax and techniques for data analysis and graphics using R, check out our upcoming 6-hour online workshop: Intro to R!
The resources to learn R now make it feasible for people like me (non-statisticians) to pick it up. The online documentation provided is way too terse. One web site I like for R is , not to be confused with John Quick’s page you link too above.
I switched from Data Desk 5 (rest in peace) as a graduate student to Stata 7 as a post-doc in 2002. At the time I tried R, but found the lack of documentation at that time a road block. Stata’s documentation is really good, by comparison.
I notice Stata is not feature much on this site. Stata 11 is a wonderful product, btw. I know find a combination of Stata and R the most useful tools for me. I use Stata for data manipulation, analysis, and graphics, and I use R for things not in Stata, and simulation.
Hi Dave,
I agree that Stata is a wonderful program. I don’t feature it much because I just haven’t used it as much, and don’t use it regularly.
And because Stata’s documentation IS so good, people don’t need help in the same way they do for something like SPSS, which has pretty bad manuals. 🙂
I would be very open to guest post submissions on Stata, though.
Karen
John’s approach to simple main effects following between subjects two-way ANOVA seems a little conservative, in that he is using only the data subsets he creates for the error, not the error from the original model. I made a comment on this on his web site, where I suggested this approach with repeated measures ANOVAs, but not between subjects ANOVAs, might be better? Just a thought.
Hi Dave,
It is a bit conservative, and the usual method of dealing with simple effects is to use the model MSE, as you suggest. (Geoffrey Keppel’s book Design And Analysis has a nice chapter on this).
It’s trickier, though in repeated measures ANOVA, because there isn’t a single MSE for the whole model. Within and between subjects effects have different error terms. It’s another reason that mixed models work better for repeated measures data–it doesn’t have that problem.
Karen