home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Magnum One
/
Magnum One (Mid-American Digital) (Disc Manufacturing).iso
/
d13
/
stistics.arc
/
PAIRED-T.DOC
< prev
next >
Wrap
Text File
|
1988-07-24
|
2KB
|
50 lines
PAIRED T-TEST
AIM: This is the standard T-test which is to be applied to two samples of
data obtained from the same subject. For example, two drugs might be
applied on different occasions to a group of patients. Each patient will
then provide a measurement of response to drug A and drug B. The
variability due to differences between patients is thus eliminated, and one
considers only the difference in response to the two drugs on the part of
each patient.
Note that there must be equal numbers for treatment A and B. This is
obvious, and indeed, the program prompts for equal numbers.
EXAMPLE: Two drugs were tested in 8 patients for their ability to slow the
heart. The data is as follows:
--Heart Rate Change---
PATIENT DRUG A DRUG B
1 -22 -14
2 -14 -12
3 -36 -22
4 -28 -30
5 - 8 +10
6 -22 0
7 - 8 - 8
8 + 2 +24
The data can be input via a WordStar non-document file, as follows:
Sample data from GoldStein <-- Title, up to 65 characters
-22 -14 <-- Pair 1
-14 -12 <-- Pair 2
-36 -22 <-- Pair 3
-28 -30 .
-8 10 .
-22 0 .
-8 -8 .
2 24 <-- Last pair
REFERENCE: Goldstein, A. 1964 Biostatistics: An Introductory Text.
Macmillan Company, New York. Pages 59-61.
FORTRAN VERSION: Dr. Stanley Kaplan May 65
BASIC VERSION: Dr. Stanley Kaplan Jul 81
PASCAL VERSION: Dr. Stanley Kaplan Aug 85
UPDATES: Oct 86 (Version 2.00)
Feb 87 (Version 2.10)
Feb 88 (MS-DOS, Version 2.2)