PRINTDOC ·
TXT ·
23.1 KB ·
1984-03-13 ·
from IBM-PC-User-Group-Software-Library_Disk-36A-Apr-1987-EPISTAT
1
EPISTAT
Statistical Package
for the IBM Personal Computer
Version 3.0, 1984
Written by:
Tracy L. Gustafson, M.D.
Copyright 1984
2
INTRODUCTION
EPISTAT is a collection of programs written in BASICA for
statistical analysis of small to medium-sized data samples ( < 28
samples or variables and < 2000 total data entries per file).
The 25 programs in EPISTAT perform more than 40 common statistical
tests or functions and provide utilities for data entry, editing,
printing, graphing, sorting, selecting, transforming and crosstabs.
The programs are intended to be as self-explanatory and user-
friendly as possible. You do not need to memorize this guide
before using the programs. On the other hand, neither the programs
nor this manual purport to TEACH the proper use or interpretation
of statistics. The user must have some familiarity with the kinds
of data required and the underlying assumptions appropriate to each
statistical test.
For further explanations of tests, refer to:
1. Colton, Theodore. Statistics in Medicine. Little, Brown and Co.
Boston, 1974.
2. Fleiss, Joseph. Statistical Methods for Rates and Proportions.
John Wiley and Sons. New York, 1981.
3. Snedecor, George W. and Cochran, William G. Statistical Methods.
Iowa State Univ. Press. Ames, Iowa, 1978.
4. Schlesselman, James. Case-Control Studies. Oxford Univ. Press.
New York, 1982.
CAVEAT:
These programs have been tested extensively, but I cannot
guarantee that they will work correctly with every possible data set.
Incorrect results are usually due to errors in format or type of
data entered. If you believe you have discovered an error in the
programs, please write me. I intend to correct any bugs that are
brought to my attention.
It is good practice to regularly compare the results obtained
by programs in EPISTAT with results obtained by your previous method
of calculation. ANY unexpected result should be questionned and
double-checked by reference to tables or another method of
calculation.
3
INDEX TO EPISTAT
The following statistical tests and functions are available:
TEST or FUNCTION PROGRAM NAME
---------------- ------------
Analysis of variance (1 and 2-way)...................ANOVA
Bayes' theorem.......................................BAYES
Binomial distribution................................BINOMIAL
Chi-square test and distribvtion.....................CHISQR
Correlation coefficients.............................CORRELAT
F distribution.......................................ANOVA
Fisher's exact test..................................FISHERS
Linear regression analysis...........................LNREGRES
Mantel-Haenszel Chi-square test......................MHCHISQR
Mantel-Haenszel for multiple controls................MHCHIMLT
McNemar's test.......................................MCNEMAR
Mean, median and standard deviation..................DATA-ONE
Normal distribution..................................NORMAL
Poisson distribution.................................POISSON
Random sample generator..............................RANDOMIZ
Rank sum test........................................RANKTEST
Rates adjusted (direct and indirect).................RATEADJ
Sample size calculations..........,..................SAMPLSIZ
Signed rank test.....................................RANKTEST
Student's T-test and T distribution..................T-TEST
The following data-handling capabilities are provided:
DATA MANIPULATION PROGRAM NAME
----------------- ------------
Determine best test and program names................EPISTAT
Graph histograms.....................................HISTOGRM
Graph scattergrams...................................SCATRGRM
Perform data transformations.........................LNREGRES
Print data (sorted or input order)...................DATA-ONE
Print crosstab reports...............................XTAB
Select specific records..............................SELECT
Transfer data between EPISTAT files..................FILETRAN
Transfer data from FORTRAN to EPISTAT files..........FORTRANS
4
SYSTEM REQUIREMENTS FOR EPISTAT
MINIMUM OPTIMAL
IBM PC with 64K RAM IBM PC with 96K RAM
One 160K disk drive Two 320K disk drives
Monochrome monitor Color graphics adapter
BASICA Hi-res color monitor
BASICA
IBM, Epson, Okidata, or
Prowriter printer with
graphics capability
OVERALL PROGRAM DESCRIPTION
All calculations in EPISTAT are performed using single precision.
Although it may first appear that double precision would be more
appropriate for statistical tests, "double" precision makes little or
no real improvement in precision in these programs. Many of the
algorithms used to evaluate p values use trigonometric functions which
are calculated in single precision anyway. For best results, data
entries should be numbers between 1E+7 and 1E-7. Larger or smaller
numbers should be multiplied by an appropriate power of 10 before
entry and analysis in EPISTAT.
All EPISTAT programs are written so that as much pertinent
information about the test as possible can fit on the final screen.
This feature allows a summary printed copy to be produced simply by
pressing <Shift-PrtSc>. This will work any time there is a pause in
the program display. Six programs, "DATA-ONE", "HISTOGRM",
"RANDOMIZ", "SCATRGRM", "SELECT", and "XTAB" produce printed reports
without using <Shift-PrtSc>. In these, follow program instructions
to route output to your printer.
EPISTAT is the introductory program in the EPISTAT package.
DATA-ONE is the major data entry, editing, and printing program.
Most of the programs in EPISTAT can evaluate data entered and saved
using DATA-ONE. Many of the programs can, in addition, evaluate
summary data. The programs marked with a star (*) below can
evaluate data entered in DATA-ONE. Non-starred programs provide
their own data entry routines.
The EPISTAT disk should be placed in drive A (or other default
drive) when loading any program because "EPIMRG" and "EPISETUP.DAT"
are used by every program. Once a program is running, EPISTAT can
be removed from drive A if necessary.
5
INDIVIDUAL PROGRAM DESCRIPTIONS
(1) "EPISTAT"
This introductory program lists the available programs and aids
the user in selecting the best statistical test. It also allows one
to specify hardware configuration and colors for a color monitor.
Choose colors 7,0,0 if you have a monochrome monitor connected to
the color/graphics adapter. If yours is not one of the listed printers,
check your printer's codes for the typeface you want. For example,
the code for elite type on the Prowriter is ESC "E". If you press
Escape then E, the display will show the decimal ASCII codes: 27 69.
An alternate method is to press <Alt> and enter the decimal code on
the numeric keypad. Press <Enter> when the complete code is entered.
"DATA-ONE" *
DATA ENTRY:
This is the central keyboard data entry program for the EPISTAT
package (for non-keyboard data entry, see FILETRAN and FORTRANS).
Initial data entry (Option 1) first asks you to name your samples or
variables. Then type in the data, pressing <Enter> twice after each
entry. The maximum number of samples or variables (S) allowed is
28 with a color adapter and 7 with a monochrome adapter. The maximum
number of records in each sample is 2000/S. A blank record can be
entered by pressing <Enter> then key F2. To exit, press <Enter> then
key F10. The mean, median and (n-1) standard deviation are then
displayed. When you return to the main menu, SAVE your datafile to
disk (Option 5) for future modification or use by other programs
in the EPISTAT package.
Although all entries in a datafile are treated as numbers by
DATA-ONE, it is possible to enter characters (names) in a record.
Characters will be treated as zeros in calculations. Nevertheless,
it improves data readability to use the "Sample 1" column for record
or case names. Thus, DATA-ONE allows one to specify a name for each
column (variable) and each row (case) in the datafile.
DATA MODIFICATION:
APPEND (Option 2) allows one to add more observations to a sample
at a later session. EDIT (Option 3) allows one to delete or replace
incorrect data entries and to change sample or variable names. When
you return to the main menu, SAVE modified data to disk again.
PRINTING DATA:
To view or review a datafile, a printout to screen or printer can
be selected (Option 4). To print a datafile exactly as it was keyed in,
request the printout in INPUT order. DATA-ONE can also print the
data SORTED by any selected sample. Only numeric data is sorted by
DATA-ONE, so it will not alphabetize a character field. Blank records
are not sorted, either.
SAVING DATAFILES and LOADING DATAFILES:
SAVING data (Option 5), writes your data to disk in a sequential
file for later editing, review, or use by another program. DATA MUST
BE SAVED TO DISK before it can be used by other programs in EPISTAT.
Since EPISTAT must be in drive A: (or other default drive) to begin,
you will probably want to SAVE datafiles on drive B. To do so,
precede each datafile name with B: (e.g. B:TESTDATA). Do not enclose
filenames in quotation marks.
6
(3) "ANOVA" *
Provides ONE-way and TWO-way analysis of variance. One-way ANOVA
compares the means of 3 or more samples. Two-way ANOVA compares the
combined effects of 2 variables on a third (ROW and COLUMN effects).
All samples in two-way ANOVA must have the same number of elements.
ANOVA prints sample means, (n-1) variances and sums of squares.
It also evaluates a known F value. (Snedecor, pp. 258-338)
(4) "BAYES"
Using Bayes' theorem, this program calculates the rates of false
positive and false negative tests given different sensitivities,
specificities and outcome incidences. Using the formula in a different
way, it calculates the prior probability of several outcomes given a
positive test. (Fleiss, p. 5)
(5) "BINOMIAL"
The binomial distribution allows calculation of the probability
of an observed number compared to the expected. It assumes the variable
is dichotomous and has an equal probability of occurring in each trial.
This program calculates the ONE-tailed probability of the observed
number and all more extreme situations. For example, in the case of
2 heads in 10 tosses of a coin, the ONE-tailed probability includes the
sum of the probabilities for 0,1 and 2 heads. (Colton, p. 151)
(6) "CHISQR"
The Chi-square program evaluates a table of data or a known
chi-square value. 2 by 2 tables are evaluated using Yates' correction
and the odds ratio and its confidence limits are calculated using
Cornfield's method (Schlesselman, p. 175,177). A Chi-square test
for trend can also be performed. (Sclesselman, p. 201)
(7) "CORRELAT" *
Pearson's correlation coefficient and Spearman's rank correlation
assess the relationship between paired variables. The probability
of a given Pearson R value is evaluated using the T distribution.
(Colton, p. 212)
(8) "FILETRAN" *
On occasion, it may happen that you want to compare 2 samples
or variables that are in separate datafiles. Or you may have a data
set with more than 28 variables that you split between two or more
datafiles. Since EPISTAT programs only allow analysis of samples
that are in the same datafile, FILETRAN allows you to transfer
samples between two datafiles. You may create a new datafile by
selecting one sample from DATAFILE #1 and another from DATAFILE #2.
FILETRAN can also combine two samples by APPENDING one to the other.
7
(9) "FISHERS"
Fisher's exact test evaluates 2 by 2 tables of discrete variables.
It is particularly valuable when the Chi-square test is inappropriate
because the expected value for a cell is < 5. However, this program
can evaluate some tables where A+B+C+D > 200.
(10) "FORTRANS"
If your data was previously entered into a FORTRAN or other SDF
sequential card image file, FORTRANS may be able to transform it into
an EPISTAT datafile. You must know the record length, appropriate
column numbers, number of decimal places and missing value code.
(11) "HISTOGRM" *
The histogram program graphs a data sample according to user
specifications on the high resolution graphics screen. To obtain
a printed copy on the IBM, Epson, Okidata or Prowriter (specified in
"EPISTAT") press key F1. Press F10 to return to the program.
(12) "LNREGRES" *
Linear regression analysis calculates the least-squares regression
line for paired samples. It then uses the T distribution to determine
if the calculated slope is significantly different than zero. (Colton
p. 199) LNREGRES also provides a variety of data transformations.
Transformed data can be saved to disk for future use or printout.
(13) "MHCHISQR"
The Mantel-Haenszel Chi-square test evaluates the relationship
between two discrete variables while controlling for the effect of
a third variable. It also calculates an odds ratio and 95% confidence
limits. (Schlesselman, pp. 183,206)
(14) "MHCHIMLT" *
The Mantel-Haenszel Chi-square test for multiple controls compares
a case sample with 2 or more matched control samples, and calculates
a probability and an odds ratio. (Fleiss, p. 125) MHCHIMLT can
evaluate summary data or raw data entered using DATA-ONE. If using
DATA-ONE, data should be coded as "1" for factor present, and "0" for
factor absent in each case and control sample.
(15) "MCNEMAR"
McNemar's test (paired Chi-square test) evaluates 2 by 2 tables
of paired discrete variables using Yates' correction and calculates
an odds ratio and 95% confidence limits. (Schlesselman, p. 210)
(16) "NORMAL" *
The normal distribution has innumerable uses in statistics. This
program specifically addresses three situations: (1) It compares
a sample mean to a population mean. (2) It calculates the proportion
of samples that would be expected to fall in any given range under
the normal curve. (3) It calculates the two-tailed probability
associated with any given value of z.
8
(17) "POISSON"
The Poisson distribution applies to dichotomous variables when
the number of successes can be counted, but the number of failures
cannot. This program calculates a ONE-tailed probability.
(18) "RANDOMIZ"
This random sample generator aids in the selection of random
samples for several purposes. It can provide a random subset of a
larger population, or it can assign cases randomly to independent or
paired groups for case-control studies.
(19) "RANKTEST" *
Two non-parametric tests of significance are performed by this
program. They are appropriate for small samples which are clearly NOT
normally distributed. They also specifically apply when quantitative
variables are not available but qualitative ranks are. The RANK SUM
TEST compares 2 independent samples. The SIGNED RANK TEST compares the
medians of paired samples. RANKTEST calculates the TWO-tailed
exact probability associated with the various rank sums. Note that
for samples larger that 20 observations, the latter calculation can
take several minutes. (Colton, pp. 219-222)
(20) "RATEADJ" *
The rate adjustment program will adjust sample rates by either
the direct or indirect methods.(Colton, pp. 47-51) For the direct
method, the datafile must include the study sample rates and the
standard population figures. For indirect adjustment, the datafile
used must include the study population figures and the standard
population rates. For indirect rate adjustment, RATEADJ evaluates
the probability of the observed number of cases using the ONE-tailed
Poisson distribution for small numbers, or the Chi-square
distribution for large numbers.
(21) "SAMPLSIZ"
The sample size program calculates the approximate sample sizes
required to achieve statistical significance given certain specified
levels of certainty. Adjustments are made if the user desires more
than one control per case. (Schlesselman, p. 168)
For a survey: TP = total population pi = population proportion
d = maximum acceptable error in sample proportion
n = [ z(a)*SQR(pi*(1-pi)) / d ] squared
N = n / (1+n/TP)
For a paired case-control study: (Colton, p. 161)
N = [(z(a)*SQR(pi*(1-pi)) + |z(b)|*SQR(PT*(1-PT))) / (PT-pi)] squared
For an unpaired case-control study: (Fleiss, p. 41)
[(z(a)*SQR(2*pi*(1-pi)) + |z(b)|*SQR(PT*(1-PT)+PC*(1-PC))]
N = [-----------------------------------------------------------] squared
(PT - PC)
9
(22) "SCATRGRM" *
The scattergram program graphs paired variables according to
user specifications on the hi-res graphics screen. To add the linear
regression line, press key F5. To obtain a printed copy on the IBM,
Epson, Okidata or Prowriter (specified in "EPISTAT"), press key F1.
Press key F10 to return to the program.
(23) "SELECT" *
This program allows the user to select any combination of
records for printout. It can also create a new disk datafile that
is a select subset of the original. One can select on any variable
with "AND" and "OR" specifications. As many as 10 selection criteria
can be set at one time. SELECT assumes that "AND"s are in parentheses.
For example:
"SELECT IF Sample #1>10 AND Sample #2=1 OR Sample #1<Sample #3"
is interpreted as meaning:
"SELECT IF (Sample #1>10 AND Sample #2=1) OR Sample #1<Sample #3"
(24) "T-TEST" *
The Student's T-test compares the means of two samples. The
program provides both paired and unpaired T-test calculations.
Variances (n-1) are displayed and, for independent samples, the
equality of variances is tested to be sure that the assumptions
of the T-test are met.(Snedecor, p. 116) T-TEST will also
evaluate a known T value.
(25) "XTAB" *
The crosstab program generates 1,2 or 3-way crosstab reports.
It allows the user to specify the crosstab criteria as well as a name
for each row and column so that the report will be readable and
easily interpreted.
NOTICE
---------------------------------------------------------------------
Users may copy EPISTAT and distribute it to others on the following
conditions:
1. The programs are not modified in any way.
2. Individual programs are not distributed separately.
3. No fee is charged for copying or distribution.
---------------------------------------------------------------------
10
====USER-SUPPORTED SOFTWARE====
The concept of user-supported software is based on three
principles:
1. The value and utility of a software package is best assessed
by each user on his or her own system with his or her own data.
Only after using a program can one determine whether it serves
one's personal applications, needs, and tastes.
2. The creation of independent personal computer software requires
a substantial commitment of time and effort. Rather than
duplicate this effort time after time, the computing community
can and should support individual creative efforts.
3. By encouraging users to copy programs, rather than spending
large sums on copy-protection, authors can supply quality
software at reduced cost. Users will support useful programs.
If after using EPISTAT, you find it of value, your contribution
in any amount will be appreciated ( $25 suggested ).
Send contributions to:
Tracy L. Gustafson, M.D.
1705 Gattis School Road
Round Rock, Texas 78664
Thank you, and good luck.