Carol Cronin writes:
The new Wolfram Statistics Course Assistant App, which was released today for the iPhone, iPod touch, and iPad. Optimized for mobile devices, the Wolfram Statistics Course Assistant App helps students understand concepts such as mean, median, mode, standard deviation, probabilities, data points, random integers, random real numbers, and more.
To see some examples of how you and your readers can use the app, I’d like to encourage you to check out this post on the Wolfram|Alpha Blog.
If anybody out there with an i-phone etc. wants to try this out, please let me know how it works. I’m always looking for statistics-learning tools for students. I’m not really happy with the whole “mean, median, mode” thing (see above), but if the app has good things, then an instructor could pick and choose what to recommend, I assume.
P.S. This looks better than the last Wolfram initiative we encountered.
the whole "mean, median, mode" thing does seem a _throw back_ like putting the internal combustion engine in front of the driver because thats where the horse was.
When I was in Oxford, I tried to argue all summaries were a throw back, as we can easily deal with all the distribution or raw data. David Cox pointed out that I was missing the need to split data into that for estimating versus that for checking the model (this requires summarization) – but in the main I think I was right.
Some of us tried to provide a new perspective on statistics based on current symbolic computational tools to Wolfram – but with little impact.
Work by David Andrews and Jamis Stafford though was published in a book – though now no longer in need for Mathematica program.
Really do like their position on not wasting student brain cells learning to do mathematical equation manipulations. Very disappointed in my daughter's recent first year uni calculas course.
K?
This was hard to find – Andrews, D. F. and Stafford, J.E. (2000). Symbolic Computation for Statistical Inference, Oxford University Press: Oxford.
Even without the Jamis (James) typo.
K?
Perhaps this is all too obvious to say, but
1. Mean, median and mode need to be taught if only because they are widely used outside the introductory course.
2. It is hard to see how regression and other models could be taught without building on the idea of the mean.
3. Keith's point that we should look at the entire distribution is clearly correct too.
I imagine that Keith's version of the introductory course would echo Felix Klein in being elementary statistics from an advanced standpoint. Good luck with the hypothetical course evaluations!
Nick: Flattering to be compared to Felix Klein but I am actually aiming more at Tristan Needham's Visual book http://usf.usfca.edu/vca//
> if only because they are widely used outside the introductory course.
Yes, very important reality, but same arguments were/are given for teaching COBOL and SAS.
> how regression and other models could be taught without building on the idea of the mean.
Expecially since regression is exactly a (weighted) mean, but why not start with P(f(x)) in general and then narrow to Normal(u + Bx,sigma) and other variants. And of course I am heading back to displaying these in those horrid plots I always shoe horn in ;-)
Historically it went the other way, probability models were found that justified the seemingly good (to Fisher "Good for What?") summaries (see Keynes in paricular )http://www.stat.columbia.edu/~cook/movabletype/mlm/JustHistory.pdf
> hypothetical course evaluations!
Fortunately just hypothetical for now, but building a productive and pragmatic intro stats course is no easy task for any group and it would need to be iterated over numerous years and classes to get it anywhere near less wrong enough to be taken seriously by others.
K?
Always being wrong – just got some add email from Wolfram.
They seem to have upgraded statistics (i.e. the math and algebra components) since I last checked.
Still though mostly what masters level/beginning Phd students work very hard at learning how to compute (expectations and probabilities under various assumed distributions).
Now that Mathematica is likley much better at that than they ever likely will be …
K?