Skip to content

purpose of list

15 messages · Robert W. Hayden, Mark Daniel Ward, Liviu Andronic +4 more

Original
Robert W. Hayden
# 
I had some minor role in the creation of this list.  At times I have
complained that members have been too harsh in redirecting queries
whose proper home seemed to me debatable.  But as time has gone on I
have to admit that too much (most?) of the traffic on this list is
about how to make R work rather than how to use it in education.  So,
trying to think positively, let me toss out some general EDUCATIONAL
questions.  How are people using R for educational purposes?  What do
people think of the various GUI or alternate (e.g., spreadsheet)
interfaces?  Is there anything that makes R as easy to use for
beginners as, say, Minitab?  What about using R for educational
simulations?  Is R the tool for replacing what George Cobb calls a
"Ptolemaic curriculum"?

       http://repositories.cdlib.org/uclastat/cts/tise/vol1/iss1/art1/ 

If anyone wants to respond it might help to briefly describe your
student audience.  While I am attracted to the power of R, many of the
folks who ask me about R are attracted by the price.  That is
especially important in a number of contexts I work with:


my own online courses where we have no computer labs for
students who must get their own software, often paying full
single-copy prices

public high schools in the U.S. where student computer access at
school is limited and it is very helpful to give students something
they can install on a computer at home

people teaching in less wealthy nations
------->  First-time AP Stats. teacher?  Help is on the way! See

 http://courses.ncssm.edu/math/Stat_Inst/Stats2007/Bob%20Hayden/Relief.html
	    
  Robert W. Hayden in the old library at  212 Main Street (P. O. Box 450)
  North Troy, VT 05859  phone (802) 988-2587  web site http://statland.org/      
  email  bob statland.org  (add your own "@" and save me some spam)
Mark Daniel Ward
# 
Dear Bob,
    Thank you for your question to the list.  I'm using R to teach 
courses on Computing with Data.  I am in the Department of Statistics at 
Purdue University (in West Lafayette, Indiana, USA), although of course 
my opinion does not represent our entire department.
    I believe that R is a great tool for teaching undergraduates and 
graduate students.  I teach these Computing with Data classes at both 
levels.  As you might guess from the title, I do not focus on 
statistical methods, but instead I focus on the ability to manipulate 
data that is very large and thus cannot be handled by less-powerful 
software.
    It might be helpful to note that I teach R alongside other 
technologies (in the same course), including UNIX tools, bash shell 
scripting, regular expressions, SQL, XML, and a brief introduction to 
Perl....  although R remains the main topic in my course.  I spend about 
1/3 of the time in the course on R.
    Our students are fortunate to have access to other software, 
including (for instance) S-PLUS, but I really enjoy using open source 
technologies in my classes to the greatest possible extent.
    I hope that helps give a perspective about one use of R in the 
classroom.
Mark
Robert W. Hayden wrote:
Liviu Andronic
# 
Dear Robert,
On Tue, Jun 30, 2009 at 1:32 PM, Robert W. Hayden<hayden at mv.mv.com> wrote:
I am not sure that this is a proper answer to your e-mail, but I will
mention a recent discussion [1] on r-sig-gui that would---I
believe---also be interesting to this list. It is about Deducer [2], a
new R-GUI built on top of JGR [3], intended to be in some respects
similar to SPSS or Minitab.

Personally, as a student and a self-taught novice in R, I believe that
JGR and Rcmdr are individually (and combined) most helpful to
beginners in grasping the basics of R, of course apart from the
introductory web sites [4] and beginner-friendly documentation. There
is also playwith [5] for graphics manipulation. I'd be keen to add
Deducer on the list, when the project matures.
Departing from the "doing statistics" objective, LyX [6] is most
helpful in writing Sweave reports without the additional burden of
(properly) learning LaTeX.
One important note is that all the mentioned applications are
cross-platform, and relatively easy to install.

Best regards,
Liviu

[1] http://www.mail-archive.com/r-sig-gui at stat.math.ethz.ch/msg00465.html
[2] http://cran.r-project.org/web/packages/Deducer/index.html
[3] http://cran.r-project.org/web/packages/JGR/index.html
[4] http://www.statmethods.net/
[5] http://cran.r-project.org/web/packages/playwith/index.html
[6] http://www.lyx.org/
Graham Smith
# 
Rob,
As a casual observation I am now in my third year of teaching basic
statistics to biology/ecology students with a mix of Minitab and R.
Outside the formal teaching, I have used R with selected students for
6 or 7 years (students who want to work at home, students with
Macs/Linux at home, and my personal dissertation students).

 In the first year I use Minitab and in the second year I use R.
Classes are around 30 students in first year and 20 in second year.

In all three years, unprompted by me, I have had between five and
seven students come to me after the R sessions asking why we hadn't
used R in first year. They find the command line more direct and
immediate than Mintab, which, it seems, they had found confusing. This
year I asked the whole  2nd year class to vote on whether they would
have preferred R to Mintab in 1st year. Everyone voted for R.

This year I also offered R sessions to the first year, in addition to
the Minitab sessions. Six people came to these sessions and all used R
in preference to Mintab for the later marked assignments. They could
have used either program. Again they all said they would rather have
used R for the main teaching sessions.

Next year I am dropping Mintab,and switching to R, but three points.

1. We only have a very limited tme for the stats and only do very
simple things, which are easy in R.
2. The stumbling block in changing to R has not been the students, but
the staff, who don't want to learn a command line stats program, and
are confident the students "won't like it" BUT SPSS license are up for
renewal, staff seemed reassured by my anecdotal evidence for students
liking R and now a days almost everyone seems to know someone in their
field who is using R, or teaching with R. So a switch to R as the
standard looks on the cards.
3. But, we are re-organising and merging with the School of Social
Sciences who I understand are super glued to SPSS, so it may be
interesting to see what happens.

Graham

BTW I do introduce Rcmdr and BiodiversityR to the students so they
know there are GUI alternatives if they get stuck, and some do use
these, but most seem to simply email me the code to sort out where
they have gone wrong.
Hadley Wickham
# 
This matches my experience in a course which I taught both Excel and
R.  Most students preferred R because it was much harder to follow
what I was doing in the GUI - where exactly was I clicking, was it a
right or left click, etc.  With R you see everything I type and it's
very easier to reproduce.  It's also much faster and easier to produce
a page of commented R code that allows students to reproduce all the
important steps, compared to recording a screencast to show the steps
in Excel.

Hadley
Mark Daniel Ward
# 
I'll follow-up on Hadley's comment by noting that I always post the 
complete R transcript of our class session, so that the students can 
download it and use it.  I also add lots and lots of comments to the 
file (after class is over), so that they can remember what we did in 
class.  They seem to like this feature of my class.
Mark
hadley wickham wrote:
Graham Smith
# 
Hadley /Mark
Mark Daniel Ward wrote:
I'm glad to see my experience isn't unique.

I find the ability to email me a bit of code that isn't working a
fanatastic feature for me. It has always been a nightmare trying to
diagnose via email what some one is doing wrong with Minitab/Excel.

Graham
Hadley Wickham
# 
I used to do before I became depressed by the huge number of mistakes
I made!  Now I tend to come up with the commented R code first, then
build the lecture around that.  I'll still do some live exploratory
programming, but all the important stuff will be available online.

Hadley
On Tue, Jun 30, 2009 at 12:58 PM, Mark Daniel Ward<mdw at purdue.edu> wrote:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Mark Daniel Ward
    

    

      
      #
    

  


  

      
Dear Hadley,
    Yes, yes, I make lots and lots of mistakes too.  I do everything 
"live", in front of the students.  (I think that it makes them happy to 
see the professor making some mistakes, because they see that everyone 
makes mistakes from time to time!)
    After the lecture is over, while I am adding lots of comments, I 
also take the time to correct any mistakes that I made.  This process 
took a LOT of time the first time that I did it, but I think that it was 
worthwhile.  I'm weighing whether to do it again this fall.....  and I 
think that I will.  It should be easier the second time around.
Mark

        
hadley wickham wrote:

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Mark Daniel Ward
    

    

      
      #
    

  


  

      
Dear Graham,
    Actually, I love to communicate with the students by email, but I 
find it a nightmare when they email me any code!  Sometimes a 1-minute 
verbal explanation will take several paragraphs to clarify by email.
    I try to be available almost all of the time in my office, so that 
they can drop by with occasional questions, even outside of office 
hours.  I also use a once-per-week lab session to answer questions, and 
I've noticed that the students talk to each other during such labs..... 
and they frequently answer each other's questions, which is a blessing.
Mark

        
Graham Smith wrote:

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Albyn Jones
    

    

      
      #
    

  


  

      
On Tue, Jun 30, 2009 at 02:29:35PM -0400, Mark Daniel Ward wrote:

            

      
      
      
    

        
I have often done exactly the same, and students have expressed their
appreciation.  They also are likely complain when I don't do this, no
matter how simple the example presented...

albyn

  
  


  


      

    

    

      
      

        


  

        

      Hadley Wickham
    

    

      
      #
    

  


  

      
I spent a lot of time focussing on code as communication to make
electronic interaction as easy as possible.  Personally, I think this
is very important as the majority of my programming work is in
collaboration with others spread around the world, who I might see in
person at most once a year.  I agree with Mark that encouraging
cooperation within class makes a big difference.  There will usually
be a few students who catch on really quickly and can help the others
- a good reason for group projects.

For most homeworks, I printed out their submitted code and marked it
like you'd mark an essay, with marks for the equivalent of punctuation
and sentence structure. Little things like indentation make a big
difference to readability.  It took a few weeks, but the quality of
their code improved considerably (so did mine!)

Hadley

        
On Tue, Jun 30, 2009 at 1:34 PM, Mark Daniel Ward<mdw at purdue.edu> wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Graham Smith
    

    

      
      #
    

  


  

      
Mark

            

      
      
      
    

        
I agree, but

a) I only lecture part time, (I do ecological consultancy the other
part). I have a "very" flexible office hours, which often don't fit in
with student hand in dates, so I feel I need to give email support
b) I also run a part time Masters in Ecological Impact Assessment,
which relies on teaching being concentrated into five long weekends.
So email is the only way I can give ongoing support for the statistics
(or any of the other) parts of the course.

However, I agree with you about email support being a mixed blessing
and I often find that an hour will have vanished as I try and work out
the best way of explaining something in writing, which face to face
would have been relatively simple. But it is still a lot easier with a
bit of code in front of me instead of vague descriptions about what
menu steps they followed,  and what boxes they may or may not have
ticked along the way.  Sending me the data set and the code is, on
balance, a lot easier.

Graham

  
  


  


      

    

    

      
      

        


  

        

      Mark Daniel Ward
    

    

      
      #
    

  


  

      
I've seen Hadley's rubrics for his assignments.  They are really excellent!
Mark

        
hadley wickham wrote:

            

      
      
      
    

        

      
      
    

        

      
      
    

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Greg Snow
    

    

      
      #
    

  


  

      
My teaching situation is a bit different from the others who have responded so far.  I do teach an occasional class at the university, but my main job is with a group of hospitals and doctors doing their statistics.  As part of that I teach some classes within the hospital system to doctors/nurses/whoever.  These tend to be very different students from the undergrads at the university (anyone else ever had a student use the excuse "I was performing surgery" for missing class?).

These classes are less focused on how to do the stats (they will have me do them for them), but more on the concepts so that they can understand the literature that they read, work with me better in designing studies, and do better at writing/presenting the results.

The university where I teach occasionally uses some web based stats program for the intro class (even simpler than minitab in what it can do).  I keep trying to talk them into using R (through the Excel interface or RCmdr), but so far have made little progress.

In both situations I am not teaching the students how to use R, but I still use it to demonstrate various concepts.  I mostly use simple examples (plots, quick numbers) or those from the TeachingDemos package (that is what it was written for (disclosure of possible conflict of interest in promoting the package: I believe that the package author/maintainer raids my fridge more often than he should)).  Even though the students are not learning R itself, they appreciate the concepts illustrated.  The mainly GUI based illustrations could be written in any language (but using R means that I can customize them and more advanced students can look at the code themselves and learn more).

The R-based code usually does not phase the students even though they have not leaned R themselves yet.  Those that do not understand it just look at the results, but many do understand the main ideas.  For example, at the university, the standard course materials use the 1970 draft lottery as an example for showing scatterplots, correlation, relationships and other concepts.  As a bit of a teaser there is a slide that asks if the relationship could be due to chance (this is before we get into the inference part of the course).  We don't come back and answer that teaser until the last week of class (when we have done hypothesis testing an regression).  But I like to show the students a simple simulation at the time of the teaser with simple code such as:

            

      
      
      
    

        
[1] 5

            

      
      
      
    

        
[1] 533

            

      
      
      
    

        
2.5%      97.5% 
-0.1021715  0.1003527

It takes less than a minute to explain that 1:366 is the numbers from 1 to 366 and that the sample function randomizes the order.  The replicate function and the rest of the code is fairly self explanatory even for people who don't know any programming, but this example clearly shows that the amount of correlation in the 1970 draft lottery was unlikely due to random chance as well as what values we would expect to see by chance.  Some of the students will stay after class and ask about R, I explain that it is not part of the official class, but that if they are interested here is the website and additional info ...

So, I think that R is a great tool for teaching statistical concepts even when R is not the statistical package being used for the course itself.