[R-sig-teaching] thoughts on teaching with R

[Eric Lamb]

Andrew,
I am just finishing my first year teaching a graduate level stats course 
to Agriculture and other life sciences students. The course is 
practically oriented, and most of the students have taken 1 
undergraduate course previous to mine.

1) What are the instructional decisions that a person needs to make if 
they are going to be teaching statistics using R?

I think that the biggest decision is whether we are going to teach a lot 
of calculations by hand, or whether we are going to "black box" a lot of 
the mathematics and teach the process of data analysis.

2) What decisions have you yourself made and what were your reasons?

I have chosen to focus on data analysis as a process. I think for the 
average MSc. student it is most important that they be able to choose an 
appropriate method and apply it correctly than to understand the theory 
of statistics. I have chosen to use R rather than a program like SAS 
because it is so much more useful to the students.

Most of my students will be going to work for small companies after 
graduation that do not have the resources to purchase something like 
SAS. There was a bit of pushback from faculty who didn't like the idea 
that the students were leaning a program other than their favorite one, 
but they fully accepted the value of wider availability.



3) How do you teach with R? Do you have sessions on R and other sessions 
where content is taught? Is the computing fully integrated with the 
content? Or some combinationn?

More than anything I like running examples "live" in class using R. I 
notice a real difference in my students between a segment of powerpoint 
(passive, unresponsive...) to when I am running examples (engaged, 
suggesting ideas on how an anlysis should proceed...). I provide them 
with all of the scripts developed in and for class so they can re-run 
analyses later on their own.

The students have weekly assignments where they are expected to apply 
the tests they have seen in class to new datasets. I find the process of 
having them see an example and then figure out how to apply it in a new 
situation to be very effective.

I use Crawley's "The R Book" extensively. I like the books approach, 
integrating both theory and practice. The examples are also very 
suitable to my agriculture / life sciences students.

4) If you have the heterogeneous group of students (some going on to 
program in R, others just trying to get through, etc.) how do we deal 
with this? Do we need to have different types of assignments and 
materials for the different students?

Everyone has to do the same assignments. Different groups have different 
challenges, however. I have some students with an engineering background 
that are very comfortable with the idea of coding but have no background 
in things such as experimental design and messy data. Oh the other hand 
I have ag students who found the idea of telling a computer what to do 
by typing very intimidating. Within 4 weeks of hands on practice I found 
most students to be very comfortable with using the command line 
interfaces, and since then almost all discusson and questions have 
focussed on the use and interpretation of statistics.

I hope this helps. In general I have found R to be rewarding to work 
with, and the students have in general responded very well to it.

Cheers, Eric
-- 
Eric Lamb
Assistant Professor, Plant Ecology and Biostatistics
Dept. of Plant Sciences, University of Saskatchewan
http://homepage.usask.ca/egl388/index.html

4D68 Agriculture Building
306-966-1799
eric.lamb at usask.ca