Hi
On Sat, 14 Aug 1999, Michael LAVIN wrote:
This year I am teaching our senior seminar capstone course. We
require the seniors to propose and design an original study but
they have the option to collect and analyze real data or analyze
simulated data. The majority opt for the simulated approach. Do
any of you know a computer package that can generate data that
would allow those "simulation" students to analyze virtual data
thus lessening the burden on our relatively small subject pool?
Jim replies:
An article by Jeff Miller in a recent _Teaching of Psychology_
described such a program. Jeff can be reached at
[EMAIL PROTECTED]
For simple designs, this can also be done quite easily in SPSS
(or other statistical packages).
The following program illustrates the process for a 4 group study
in which one group is sampled from a population with mu 10 units
(1 sigma) higher than the others (a pretty robust effect). The
SEED in line one determines what specific numbers are generated.
This is an old example from a few years ago, but I think I had
described a study with 4 techniques of teaching reading. COND
defines the groups (group 4 was a Phonics group in the
hypothetical example) and READ is the dependent variable. I have
a bunch of these for t-tests, anovas, simple regression, and
multiple regression, but have never put them in any order. I use
them for student assignments, with each student adding a number
distinct to them to the SEED.
SET WIDTH = 80 SEED = 4000000
INPUT PROGRAM
. LOOP subj = 1 TO 32
. COMPUTE cond = TRUNC((subj-1)/8)+1 /* cond = 1, 2, 3, or 4
. COMPUTE read = RND(NORMAL(10)+50) /* mu = 10, sigma = 10
. END CASE
. END LOOP
. END FILE
. END INPUT PROGRAM
IF (cond = 4) read = read + 10 /* add 10 to mu for cond 4
LIST
Best wishes
Jim
============================================================================
James M. Clark (204) 786-9313
Department of Psychology (204) 774-4134 Fax
University of Winnipeg 4L02A
Winnipeg, Manitoba R3B 2E9 [EMAIL PROTECTED]
CANADA http://www.uwinnipeg.ca/~clark
============================================================================
