l'etranger (allogenes) wrote in stat_geeks,
l'etranger
allogenes
stat_geeks

Stats Software for Intro Class (of non-statisticians)


I am slated to teach a class in intro stats to psych majors this upcoming semester, and I would like some software to use consistently throughout the course. I refuse to use commercial software; having myself been trained extensively in SAS only to find it unavailable due to my current job, I don't want to create that situation for my students. In short, I'd like to give them something they can keep using as long as they want to use it.

I'd use R, if they were science students (more comfortable with computing/etc.) or stats majors, but these people are from psychology. So I need something easy to install, and which can be made easy (easier) to use. Don't get me wrong, I have used R in intro work before, but it was not smooth.

I would consider R with a front end. Is R-commander better than it used to be a couple years ago?

I have already rejected PSPP, as it has no graphics (right?), and Statistical Lab (Statistiklabor) as, well, I am having trouble getting that working well. It also lacks the needed detailed support in English that my students might require.

Any other ideas? Any users of OpenEpi or Gretl? Would these work for general stats?

Ideally I'd like the following: some tools for simple non-parametrics, resampling (wishful thinking?), the usual suspects in normal theory statistics with both 1 and 2 way ANOVA, Fisher's exact test (wishful thinking again?), lots of good graphics, relatively easy data transformations, and the ability to do simple simulations (this may have to be done elsewhere than the main package). Obviously I'd like it to be Free (beer) and/or free (libre) and/or open source. But I'll settle for anything the students can get for 0.00 USD legally and give to there friends.

I realize that what I want is simplified-R. Any ideas? Thanks in advance for any leads you can give me.

X-post-to: statisticians.
  • Post a new comment

    Error

    default userpic
  • 9 comments
I'd recommend SPSS if that's not what you were referring to in your post as PSPP. It's the standard software for soft sciences in the universities I've been to. Heavy use of drop down menus but covers a lot ( not sure about resampling). Barring that, maybe Minitab. I'm not that familiar with it but I know it's used for very basic courses. Another option is Stata, which is similar to R, though the syntax is simpler. Hope that helps or least gives you some stuff to google.
Thanks but all of those are commercial. I have decided that no matter what else, I am no longer going to require commercial software in classes. In this context it is immoral--generally it is a lesser product (in terms of capacity, the only exception being, possibly, SAS), it requires money, it is not open to inspection, and finally I hate training students to use software that is chained to university computers. If they are going to invest in learning software I want it to be available for them now and always.

By the way, PSPP is an open source version of the older SPSS (now apparently renamed PASW for some reason). People who like SPSS and want a free/open version should try it out. It does not currently do the most advanced stuff, but it does cover basics pretty well (except graphics, but someone else told me I am wrong about that).

The problem is that I work with psychological researchers a lot and their needs have expanded beyond what SPSS does easily. Plus I find SPSS/PASW and Stata encourage bad process when it comes to statistics. They encourage students to start with a specific test rather than starting with the data more organically. I am trying to remake the intro stats course for psych in away that gets free of that. I admit I am way out of the mainstream on this. :-)

Thanks for the comment!
SPSS was specifically developed for analysing survey data.

As a government statistician I generally find economic stats people tend to use SAS and social stats people use SPSS.
Yeah I have seen that, too.

SAS is not an option at the price point they set, unfortunately. Most of my economist friends use Stata and R these days, but I used to know some SAS experts in econ back in grad school. :-) All of these are, unfortunately, commercial.

Like I said to bast_veiled above, I have to go open-source/libre/free this time for a variety of reasons, some sound from a teaching standpoint some due to costs, etc. Plus I find commercial software immoral in this context.

Thanks!
In fact PSPP does have graphics. For some examples see the wiki at http://pspp.kiberpipa.org/wiki/doku.php
Really? Where are they in the menus? I found none.
To be more specific--if I want to generate a histogram from some data, where in the menus is that?
A contrasting view from another person slated to teach stats to psych majors this fall:

You do your students a great disservice if, for personal reasons, you decide against teaching them to use something that is standard in many fields. While your experience is regrettable, you're in fact doing something similar to your students: their jobs will probably not let them install whatever stats package they want onto job-related machines. The job will probably use SPSS or SAS.

If you're really afraid of their being hindered by learning a software package they later can't be accessed, teach them how to do it by hand. Seriously.
Unfortunately in the other discussion on this, I have made certain points clearer. Some of which directly address issues you raise. It is at: http://community.livejournal.com/statisticians/45172.html if you are curious.

SPSS/PASW is standard in the sense that English is often standard, it is in many cases a default. Standard sometimes implies a clear choice, sometimes it just implies history. I work with psychologists and neuroscientists every day, and have for well over 15 years. In that time SPSS has gone from being the only thing (well, excluding Stata and SAS) to a relatively minor player. I have also found most companies are realizing the savings in using R, and in the commercial world the only real contender is SAS, not SPSS or Stata. So the only real argument for SPSS is that the previous generation of psychologists are comfortable with it. And that is not really an argument at all.

You say it is a disservice to not use SPSS. I say that is wrong. I have used SPSS a lot, and taught with it often. I am not rejecting it for any personal reasons whatsoever. I am rejecting it for sound pedagogical reasons. I feel that the structure of its menus and the way it organizes data and limits analyses to certain styles leads students to a wrong-headed way of thinking about statistics. I am trying to break them of those bad habits before they get ingrained. SPSS forces a very unnatural and fixed style of data analysis.

At some point the people who know (those with statistical training) have to stand up to the people who don't know (psychologists with little training) and say "enough is enough." I crossed that line a few years ago, and have been having markedly increased success with my stats students since that time. In standardized testing, my students class average regularly comes out above the average for other classes taught by more traditional methods. Part of that is letting go of SPSS. Part of that has been moving away from the traditional psych stats course which is still 90%+ a 1950's style course. I suspect that part has been luck, too.

I say it is the greater disservice to use software just because granddad used it.

For what it is worth, the few students I have seen after my classes who have transitioned to SPSS have had little trouble, with the possible exception of asking why SPSS is so limited.

As for doing it by hand, I don't believe that is viable. Computers are not a short-cut nor are they calculators. Some forms of analysis completely depend on computation. To drop computers entirely from a first course means dropping all real-sized data sets, dropping randomization testing, dropping exact test results, dropping electronic reference tables, etc. Too much to give up. Computers are an integral part of research, not something to drop. For what it is worth, I have taught the purely "by hand" class, too, and that has always been a fiasco.

So I am afraid I have to disagree with you, but I hope you understand that I do so for good (pedagogical) reasons, not for any personal problems I may have. Though I have plenty! :-) Thank you for your thoughts on these matters, and I appreciate the input!