Skip to content

DOE teaching suggestions?

4 messages · Erin Hodgess, Spencer Graves

Erin Hodgess
# 
Dear R People:

I will be teaching an undergraduate Design of Experiments class
in the Spring Semester.  It will be very much an applied course.

My question, please:  has anyone used R for a course like this, please?

I've tried Rcmdr for a regression course and just plain command
line for a time series course.

Should I use Rcmdr, or teach them to use the command line, OR is there
something else, please?

Thanks in advance!

Sincerely,
Erin Hodgess
Associate Professor
Department of Computer and Mathematical Sciences
University of Houston - Downtown
mailto: hodgess at gator.uhd.edu
Spencer Graves
# 
Hi, Erin: 

      Are you planning to have them design and conduct an actual 
physical experiment as part of the class?  You may know that Bill Hunter 
(the second Hunter of Box, Hunter & Hunter) wrote articles about doing 
this, and I found it extremely helpful.   Things happen with real 
physical experiments that can't be duplicated with any kind of computer 
simulation. 

      I think I've gotten good results assigning team projects of their 
own choosing.  I found it necessary to have "design review" 
presentations in the middle of the class.  These presentations give you 
feedback on their understanding of the class material to that date.  
They also give you an opportunity to suggest improvements before they 
actually do the experiment. 

      This is not what you asked, but I hope you find it useful, anyway. 
      Best Wishes,
      Spencer Graves
Erin Hodgess wrote:
Spencer Graves
# 
Hi, Erin:

	  Also, have you seen the "BHH2" package, companion to Box, Hunter and 
Hunter (2005) Statistics for Experimenters, 2nd ed. (Wiley)?

	  Hope this helps.
	  Spencer Graves
############################
      Are you planning to have them design and conduct an actual
physical experiment as part of the class?  You may know that Bill Hunter
(the second Hunter of Box, Hunter & Hunter) wrote articles about doing
this, and I found it extremely helpful.   Things happen with real
physical experiments that can't be duplicated with any kind of computer
simulation.

      I think I've gotten good results assigning team projects of their
own choosing.  I found it necessary to have "design review"
presentations in the middle of the class.  These presentations give you
feedback on their understanding of the class material to that date.
They also give you an opportunity to suggest improvements before they
actually do the experiment.

      This is not what you asked, but I hope you find it useful, anyway.
      Best Wishes,
      Spencer Graves
Erin Hodgess wrote:
Spencer Graves
# 
Hi, Erin:

	  Also, have you seen the "BsMD" package ("Bayes Screening and Model 
Discrimination"), also discussed in Box Hunger and Hunter (2005).

	  Spencer Graves
##################
	  Also, have you seen the "BHH2" package, companion to Box, Hunter and
Hunter (2005) Statistics for Experimenters, 2nd ed. (Wiley)?

	  Hope this helps.
	  Spencer Graves
############################
      Are you planning to have them design and conduct an actual
physical experiment as part of the class?  You may know that Bill Hunter
(the second Hunter of Box, Hunter & Hunter) wrote articles about doing
this, and I found it extremely helpful.   Things happen with real
physical experiments that can't be duplicated with any kind of computer
simulation.

      I think I've gotten good results assigning team projects of their
own choosing.  I found it necessary to have "design review"
presentations in the middle of the class.  These presentations give you
feedback on their understanding of the class material to that date.
They also give you an opportunity to suggest improvements before they
actually do the experiment.

      This is not what you asked, but I hope you find it useful, anyway.
      Best Wishes,
      Spencer Graves
Erin Hodgess wrote: