Introduction to SASIntroduction to SASoptions ls=65 nodate; data class; input name $ sex $ age height weight; datalines; JOHN M 12 59.0 99.5 JAMES M 12 57.3 83.0 ALFRED M 14 69.0 112.5 WILLIAM M 15 66.5 112.0 JEFFREY M 13 62.5 84.0 RONALD M 15 67.0 133.0 THOMAS M 11 57.5 85.0 PHILIP M 16 72.0 150.0 ROBERT M 12 64.8 128.0 HENRY M 14 63.5 102.5 JANET F 15 62.5 112.5 JOYCE F 11 51.3 50.5 JUDY F 14 64.3 90.0 CAROL F 14 62.8 102.5 JANE F 12 59.8 84.5 LOUISE F 12 56.3 77.0 BARBARA F 13 65.3 98.0 MARY F 15 66.5 112.0 ALICE F 13 56.5 84.0 proc print; proc means; variables height weight;When you have entered these lines, save the file:
It is advisable to view the LOG first to ensure that the job has run without problem. We do this by selecting the Log window (Click in the Log Window if it is visible, otherwise, choose Log from the Window memu.) The following file is displayed:
NOTE: Copyright (c) 1989-1993 by SAS Institute Inc., Cary, NC, USA.
NOTE: SAS (r) Proprietary Software Release 6.10 TS019
Licensed to YORK UNIVERSITY, Site 0009300001.
NOTE: The SAS System for Microsoft Windows, Release 6.10 Limited Production
1 options ls=68 nodate;
2 data class;
3 input name $ sex $ age height weight;
4 datalines;
NOTE: The data set WORK.CLASS has 19 observations and 5 variables.
NOTE: The DATA statement used 2.68 seconds.
24 proc print;
NOTE: The PROCEDURE PRINT used 1.32 seconds.
25 proc means;
26 variables height weight;
NOTE: The PROCEDURE MEANS used 0.81 seconds.
To see the results of the SAS procedures we have used, select the Output Window (Click in the Output Window if it is visible, otherwise, choose Output from the Window memu.) It should look like this:
The SAS System 1
OBS NAME SEX AGE HEIGHT WEIGHT
1 JOHN M 12 59.0 99.5
2 JAMES M 12 57.3 83.0
3 ALFRED M 14 69.0 112.5
4 WILLIAM M 15 66.5 112.0
5 JEFFREY M 13 62.5 84.0
6 RONALD M 15 67.0 133.0
7 THOMAS M 11 57.5 85.0
8 PHILIP M 16 72.0 150.0
9 ROBERT M 12 64.8 128.0
10 HENRY M 14 63.5 102.5
11 JANET F 15 62.5 112.5
12 JOYCE F 11 51.3 50.5
13 JUDY F 14 64.3 90.0
14 CAROL F 14 62.8 102.5
15 JANE F 12 59.8 84.5
16 LOUISE F 12 56.3 77.0
17 BARBARA F 13 65.3 98.0
18 MARY F 15 66.5 112.0
19 ALICE F 13 56.5 84.0
The SAS System 2
Variable N Mean Std Dev Minimum Maximum
--------------------------------------------------------------------
HEIGHT 19 62.3368421 5.1270752 51.3000000 72.0000000
WEIGHT 19 100.0263158 22.7739335 50.5000000 150.0000000
--------------------------------------------------------------------
When you tell SAS to print your job the output is routed to a printer automatically assigned to you when you logon. If you want your output printed at a different printer you must select a different printer with the Print Manager application.
The file EXAMPLE1.SAS is stored on the server in the directory N:\psy3030\examples. You can tell SAS to include this file in your own program with the statement
%include 'N:\psy3030\examples\example1.sas';
SAS will then search for EXAMPLE1 SAS, and insert that file exactly as if
it were present in your own program.
So, if you wanted to carry out a regression analysis of the CLASS data set, you could make your own file, CLASSREG.SAS, which looked like this:
%include 'N:\psy3030\examples\example1.sas'; Proc Reg data=CLASS; Model WEIGHT = HEIGHT; Id NAME;This way, you don't have to clutter up your directory with copies of files that are stored on the server. Moreover, if the version on the server is changed, you always get the current version.
NWK
for the data files from Neter, Wasserman, and Kutner.
To include one of these files (say, NWKP0403.SAS), use the form
%include nwk(nwkp0403);
Let's say it looks like this:
JOHN M 12 59.0 99.5 JAMES M 12 57.3 83.0 ALFRED M 14 69.0 112.5 WILLIAM M 15 66.5 112.0 JEFFREY M 13 62.5 84.0 RONALD M 15 67.0 133.0 THOMAS M 11 57.5 85.0 PHILIP M 16 72.0 150.0 ROBERT M 12 64.8 128.0 ... (more data lines) ALICE F 13 56.5 84.0
We can use SAS to analyse or manipulate this data by creating a different file containing only SAS statements. We create a new file called EXAMPLE2 SAS
To start a new program, clear the Program Editor window, as follows: Select the Program Editor window, then choose Clear text from the Edit menu. Then type in the following statements:
filename indata 'a:data.raw'; data class; infile indata; input name $ sex $ age height weight; proc print; proc means; var height weight;
Notice the difference between this SAS program and our previous example. A FILENAME statement has been added at the beginning of the program to inform the SAS supervisor where to find our data. We indicate that our data is in a file named DATA.RAW on the A: drive, and that we are going to refer to that file as INDATA. Whenever we use the name INDATA in our program the SAS supervisor will now know that we are referring to our data file: A:DATA.RAW.
The FILENAME statement associates a logical name, INDATA, with the physical name, A:DATA.RAW, of a file. Generally, any file can be specified in this manner, and any logical name can be used to refer to it in the program (eg. we could have used the name EXCITING instead of INDATA). However note that 8 letters is the maximum length for a label in SAS.
The second alteration to the previous example is the INFILE INDATA; statement which is used to tell the SAS supervisor that the data must be read from an external raw data file. Note - it replaces the CARDS; statement in our original example and is placed before the INPUT statement.
Once your file has been set up in this manner you "run" or execute it in the same manner as our first example.
libname save 'a:'; data save.class; input name $ sex $ age height weight; datalines; JOHN M 12 59.0 99.5 JAMES M 12 57.3 83.0 ALFRED M 14 69.0 112.5 WILLIAM M 15 66.5 112.0 JEFFREY M 13 62.5 84.0 RONALD M 15 67.0 133.0 THOMAS M 11 57.5 85.0 PHILIP M 16 72.0 150.0 ... (more data lines) ALICE F 13 56.5 84.0 proc print; proc means ; var height weight;
When this job is submitted to SAS, a file CLASS.SD2 is created in the directory (A:\ specified in the LIBNAME statement. This is a SAS system file and contains the data created by the data step starting DATA SAVE.CLASS.
Note that the LIBNAME statement associates a physical directory with the first part of the dataset name (SAVE).
You can use SAS system files to create new data files or you
can invoke them directly in a PROC statement without need of
employing a DATA step. In the first example below a SAS system
file is employed to create a new data file containing data for
males only to be used in a plotting procedure.
libname save 'a:'; data males; set save.class; if sex='M'; proc print; proc plot; plot weight*height=sex;
The following output appears in the Output Window. (The output from the PROC PLOT step is not shown.)
OBS NAME SEX AGE HEIGHT WEIGHT
1 JOHN M 12 59.0 99.5
2 JAMES M 12 57.3 83.0
3 ALFRED M 14 69.0 112.5
4 WILLIAM M 15 66.5 112.0
5 JEFFREY M 13 62.5 84.0
6 RONALD M 15 67.0 133.0
7 THOMAS M 11 57.5 85.0
8 PHILIP M 16 72.0 150.0
9 ROBERT M 12 64.8 128.0
10 HENRY M 14 63.5 102.5
If there is no need to alter the data in the file, statistical analyses can be performed by calling the data directly from the PROC statement. Below is an example:
libname save 'a:'; proc print data=save.class; proc means data=save.class; var height weight:
This performs our original analysis (EXAMPLE1); there is no need for a DATA step since the data file has already been created by SAS in an earlier job.