[R] R GUI for undergraduate lab class?

[Ranjan Maitra]

Hi Louise,

I tried using Deducer (graphical frontend to R) in my introductory
class (Stat 105 at Iowa State University) for  civil and construction
engineers and this was a roaring success this year. The class itself
has a wee bit of software experience, which has previously been done
using JMP. Most instructors (graduate TAs, typically) haven't
particularly been enamoured of using JMP, so I decided to explore the
possibility of using a GUI version of R. 

I am not in general a great fan of any kind of GUI for my own
work: however, it makes sense for this kind of class and I wanted a
one-click solution for the Windoze user which would make it possible
for them to use R with minimal typing. (Windoze users are primarily
what made up the students in this this class.)

I looked at a few R packages: pmg (or Poor Man's GUI, which appeared to
not have been updated since 2009, so I am not sure how active this
project currently is) and Rcmdr (R commander). Both these packages
served my purpose for a limited amount, but were not able to do all the
aspects of regression needed in the class (in pmg's case, could not
calculate the residuals, but in the case of RCmdr's case, could not
provide a residuals plot without the entire plethora of diagnostics
also showing up) so I decided to look at other alternatives. Also,
installing Rcmdr (with its dependencies) was not particularly easy from
the student's perspective (in my view).

I found my solution in Deducer which actually has its own website:
www.deducer.org. For Windoze users, it is a one-click
installation. The beauty of this software, in my mind, is that it can do
a lot of things (including quite involved calculations)  and has its
entire suite of graphics built on the ggplot package. Being a GUI, one
has to look around for how to do particular things (which may not
always be in the place that I might thing is intuitive, but it is
there). The manual is generally decent: however, the graphics part of
the manual is perfunctory at best. And it uses Java, a resource hog I
generally abhor.

I sent the students the e-mail (below) on how to install (and come
prepared for the class last week so that we could do a demo and
examples). They did not have much issues. In fact, in class, I got
exactly one complaint: "Can't all this be done in excel?"
-- I completed this question for him: "... and also on pen and paper"
-- but other than that, most students have taken to it and turned in
their homeworks without much fuss. In fact, some have liked it and
explored with it. 

In our introductory classes, we have typically used JMP. But if some of
us prefer using an open-source option that is not encumbered by
patents and proprietary licenses but have been bothered by R's learning
curve, using it through Deducer might be a possible option.


Btw, here is the e-mail I sent my class on how to go about installing
Deducer (note that the instructions are from late January):



Dear students,

For Stat 105, we will be demonstrating Deducer which is
available at http://www.deducer.orgfor installation. Deducer is a
graphical frontend to the statistical software R.

If you are installing to Windows, please use:

http://neolab.stat.ucla.edu/cranstats/Deducer-R-2.15.0-win.exe

This will install all you need in a one-click operation.For other OS's,
please use:

http://www.deducer.org/pmwiki/index.php?n=Main.MacOSXInstallation

for MacOSX while for Linux, please use:

http://www.deducer.org/pmwiki/index.php?n=Main.LinuxInstallation

For both Mac and Linux, you will need to install R prior to instaling
Deducer and JGR. We will go through a demonstration of Chapter 4 using
Deducer in class tomorrow.

I hope that this helps!

Many thanks and best wishes,
Ranjan


On Sat, 12 Jul 2014 13:47:08 -0700 Louise Stevenson
<louise.stevenson at lifesci.ucsb.edu> wrote:

> Hi,
> 
> I'm working on a new set of simple, ecological modeling exercises for our
> campus' undergraduate Introductory Biology lab series. The students work
> with simple population models by looking at graphs and seeing how changing
> parameter values and initial population sizes changes how the populations
> fluctuate through time. Does anyone know of an existing R GUI or any other
> interface that would be good for an undergraduate setting? Basically I want
> something that shows the students the model's output as graphs and lets them
> change parameter values but the equations/coding itself is hidden such that
> they can't change any of that. I want what Populus can do (a fantastic
> program written for this exact purpose, info here:
> http://www.cbs.umn.edu/research/resources/populus) but I want to be able to
> upload data so the students can compare model outputs to real data and I
> can't figure out how to get Populus to plot data. Any help would be very
> much appreciated! Thank you!
> 
> Sincerely,
> Louise Stevenson
> Graduate Student, University of California, Santa Barbara
> Department of Ecology, Evolution and Marine Biology
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 


-- 
Important Notice: This mailbox is ignored: e-mails are set to be
deleted on receipt. Please respond to the mailing list if appropriate.
For those needing to send personal or professional e-mail, please use
appropriate addresses.

____________________________________________________________
FREE 3D MARINE AQUARIUM SCREENSAVER - Watch dolphins, sharks & orcas on your desktop!