See All Boards Subscribe To Thread
Statistics Public Forums
Teaching Tools and Tips: Testing if Stat was taught with R
hvu132 Feb 13 ReplyFlag
I've been teaching Intro to Stat using graphing calculators for a long time, but I don't like the fact that outside of the classroom nobody uses graphing calculators to do Stat with huge data sets. I'm thinking of switching to R, so if you have taught with R, can you help me with this:

Only a few classrooms at my college (Coastline Community College in CA) have computers at every seat. And even so, I can't patrol the computers to make sure students don't cheat, like chatting with someone or googling for the answer. And we don't have the software that turns off those features.

Have you found a solution for teaching with R but giving tests without having access to R? Also I don't think I can test my students on theory only.

Hao-Nhien Vu
mine894 Feb 14 ReplyFlag
This is indeed tricky. The solution I've settled on is to not test students on R during the closed book exam but instead assess that knowledge by weekly labs and a project. R coding is rarely done without Googling anyway in my experience, so I think asssessment that allows for that makes sense.

This still leaves things like finding probabilities under a curve for p-value. I do this either with tables or more by avoiding the p-value and using the critical value method and providing a sketch of the distribution with critical values marked.
hvu132 Feb 21 ReplyFlag
wjackson Nov 10 ReplyFlag
Hi Hao,

I am fully incorporating the use of R into my Statistics course. I do not give in class exams, but rather fully assess students by take home projects, similar to how Mine described. This approach also has the benefit of allowing for much richer and in depth questions I can ask them that come from interesting datasets.


Your User ID will appear with your posts.