home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Between Heaven & Hell 2
/
BetweenHeavenHell.cdr
/
500
/
474
/
ems.arc
/
PC-PLAN.DOC
< prev
next >
Wrap
Text File
|
1986-02-25
|
6KB
|
169 lines
PC-PLAN
Randomization Plans
Version 1.0
December 12, 1985
Gerard E. Dallal
USDA Human Nutrition Research Center on Aging
at Tufts University
711 Washington Street
Boston, MA 02111
and
Tufts University School of Nutrition
132 Curtis Street
Medford, MA 02155
PC-PLAN generates randomization plans. The resulting file
can be used as input to a word processing program for touch
up and entry into a report.
NOTICE
Documentation and original code copyright 1985 by Gerard E.
Dallal. Reproduction of material for non-commercial purposes
is permitted, without charge, provided that suitable
reference is made to PC-PLAN and its author.
Neither PC-PLAN nor its documentation should be modified in
any way without permission from the author, except for those
changes that are essential to move PC-PLAN to another
computer.
INSTALLATION
PC-PLAN was written for the IBM-PC but few changes should be
needed to install it on another computer. The first DATA
statement initializes the variables
IIN -- input unit number (screen)
IOUT -- output unit number (screen)
IWOUT -- external output file unit number
The only major change involves the routine that seeds the
random number generator. The three seeds are obtained from
the computer's internal clock. The subroutine SEEDER will
probably have to be tailored to the user's particular
computer. All three seeds should be integers between 1 and
30,000, exclusive.
DESCRIPTION
PC-PLAN generates randomization plans in a form that lends
itself to easy editing with a word processor for
incorporation into memos and reports. The program is driven
by prompts to the user. The terminology is as follows:
The number of BLOCKs is the number of experimental units
defined by variables that are not subject to randomization.
For example, in a study in which men and women are to be
analyzed separately, the number of groups is 2 and their
names are "men" and "women". If there is more than one
classification variable, the units should be thought of as
being classified by a single variable whose categories are
all possible combinations of the categories of the original
variables. For example, if an experiment is to measure the
effect of drinking coffee on both men and women, the number
of groups is 4: "male drinkers", "male nondrinkers", "female
drinkers", and "female nondrinkers".
The number of TREATMENTS is the number of sets of
experimental conditions that will be applied to subject. For
example, if there are three diets under investigation, the
number of treatments is 3. If there are many experimental
factors being randomized, the number of treatments is the
number of possible combinations of the original factors. For
example, if a subject is given either a high protein or low
protein diet and is subjected to one of three levels of
exercise, the number of treatments is 6: "low protein/low
exercise", "low protein/moderate exercise", "low protein/high
exercise", "high protein/low exercise", "high
protein/moderate exercise", "high protein/high exercise",
PC-PLAN allows the user to specify whether the treatment for
each subject should be selected (1) totally at random, (2) so
that an equal number of each treatment are assigned, or (3)
so that prespecified numbers of each treatment are assigned.
PC-PLAN can also permute treatments within subject. For
example, if each subject is to undergo three different
treatments, PC-PLAN will generate a random permutations of
the three treatment.
RANDOM NUMBER GENERATOR
PC-PLAN uses the random number generator of Wichmann and Hill
(1981) as modified by McLeod (1985).
CONSTRAINTS
PC-PLAN makes provision for up to 20 blocks, 400 subjects per
blocks, 20 treatments when assigned, 5 treatments when
permuted. Block labels may contain up to 40 characters,
treatment labels up to 15.
REFERENCES
McLeod, A. Ian (1985). Remark AS R58. A remark on algorithm
AS 183. An efficient and portable pseudo-random number
generator. Appl. Statist.,34,198-200.
Wichmann, B.A. and I.D. Hill (1981). Algorithm AS 183. An
efficient and portable pseudo-random number generator.
Appl. Statist.,31,188-190.