First set-up

Following is a simple set-up that produces four tables from a data file which contains 500 characters per line in fixed format ASCII. The four tables contain accumulated figures from the 60 lines (respondents) in the data file.

start control,

* First example using ASCII data

c=first.asc,

cep,

finish control,

start data,

serial number in 1-5,

card characters binary 500,

!

ds $sex=$8/1,2,

xt='Sex of respondent', 

x='Sex\Male;Female', 

!

ds $q1=$9/1-3,e,

xt='Rating of test product', 

x='Rating\Good<v1>;Average<v0>;Poor<v-1>;No answer', 

!

dm $q2=$10/1,$11/1,$12/1,

xt='Products eaten every day', 

x='Every Day\Bread;Meat;Potatoes', 

!

di $q3=$13-14,

xt='Cups of tea drunk daily', 

!

ds $sq3=$q3/0,1..5,>5,e,

xt='Cups of tea drunk daily', 

x='Tea\None;Up to 5;Over 5;Faulty response', 

!

dc $q4=$15-18,

xt='Title', 

!

ds $sq4=$q4/'Mr ','Mrs ','Miss','Ms ',e,

xt='Title', 

x='Title\Mr.;Mrs.;Miss;Ms.;Other', 

!

finish data,

start tables,

!

%jh='Example tables',

f=lpp46/spl120/psv/clw6/urh/uch,

!

t #1=$q1*$sex,

t #2=$q2*$sex,

t #3=$sq3*$sex,

t #4=$sq4*$sex,

!

finish tables,

 

The first table in the output Tables file will look something like this -

 

Example tables Page 1

Table 1

Rating of test product

by Sex of respondent

 

Sex

-------------

Total Male Female

 

Total 60 24 36

 

Rating

------

 

Good (1.0) 22 10 12

37% 42% 33%

 

Average (0.0) 21 7 14

35% 29% 39%

 

Poor (-1.0) 13 3 10

22% 13% 28%

 

No answer 4 4 -

7% 17% -%

 

Mean score .2 .4 .1

 

The other three tables will be similar except that the rows will be different depending on which variable was used as the stub.

This set-up contains three stages which all begin with START and end with FINISH -

Please see the following sections:

Control stage

Data stage

Tables stage