What the Heck is a Data Step?

It is a basic building block of SAS programming. It creates the data sets that are used in a SAS program’s analysis and reporting procedures.

The Data Step is represented by three components:

1. The input or raw data.

2. The data step statements – DATA statement, INPUT statement and more statements.

3. The output or SAS data set.

Temporary versus Permanent SAS Data Sets

The following is an example of a DATA step that creates the temporary data set WEIGHT_CLUB.

data weight_club;

input IdNumber Name $ 6--20 Team $ 22--27 StartWeight EndWeight;

datalines;

1023 David Shaw red 189 165

1049 Amelia Serrano yellow 145 124

1219 Alan Nance red 210 192

1246 Ravi Sinha yellow 194 177

1078 Ashley McKnight red 127 118

1221 Jim Brown yellow 220 ;

run;

Kiss My SAS ... part two

Traditional SAS Output

• Data Set
• SAS log
• A Report or Simple Listing
• Other SAS Files Such as Catalogs
• External Files or Entries in Other Databases

Data Set

Just what you think it is - it stores multiple data variables, descriptive elements ... names and arrangements of data. Can be temporary or permanent.

SAS Log

A log file ... a record of SAS statements entered and detail about the execution of your program.

A Report or Simple Listing

Grouping and summaries of data ... can display statistics. This can be similar to the example output in the first blog posting.

Other SAS Files Such As Catalogs

Can't be represented as tables or data values. Examples include key settings, letters produced by SAS/FSP software, displays produced by SAS/GRAPH software.

External Files or Entries in Other Databases

Using SAS/ACCESS software you can create and update files stored in databases such as Oracle.

Procedures

The Proc Statement

Two examples:

proc tabulate data=weight_club;

class team;

var StartWeight EndWeight Loss;

table team, mean*(StartWeight EndWeight Loss);

title 'Mean Starting Weight, Ending Weight,';

title2 'and Weight Loss';

run;

And then this produces a table output with the calculations performed.

So you see, this is declaring a tabulate data procedure and assigning it the name weight_club

The second example ... :

proc print data=weight_club;

title 'Health Club Data';

run;

This displays all of the data in the weight_club data set.

So really, you just need a listing of all SAS procedures in order to get anything knocked out.

A Listing of SAS Procedures

Learning Edition Task	SAS PROC	SAS Product
Mixed Models	MIXED	SAS/STAT
Generalized Linear Models	GENMOD	SAS/STAT
Split Column	PROC TRANSPOSE	Base SAS
Stack Column	PROC TRANSPOSE	Base SAS
Utility Procedure	APPEND	Base SAS
Utility Procedure	COMPARE	Base SAS
Utility Procedure	CONTENTS	Base SAS
Utility Procedure	COPY	Base SAS
Correlations	CORR	Base SAS
Utility Procedure	DATASETS	Base SAS
Create Formats	FORMAT	Base SAS
Utility Procedure	FORMAT	Base SAS
One-Way Frequencies	FREQ	Base SAS
Table Analysis	FREQ	Base SAS
Calculate Summary Statistics	MEANS	Base SAS
Utility Procedure	OPTIONS	Base SAS
Utility Procedure	OPTLOAD	Base SAS
Utility Procedure	OPTSAVE	Base SAS
Utility Procedure	PLOT	Base SAS
Utility Procedure	DELETE	Base SAS
List Data	PRINT	Base SAS
Utility Procedure	PRTDEF	Base SAS
Rank	RANK	Base SAS
Utility Procedure	REGISTRY	Base SAS
Utility Procedure	REPORT	Base SAS
Utility Procedure	SORT	Base SAS
Append Table	SQL	Base SAS
Query Builder	SQL	Base SAS
Standardize variables in a SAS data set	STANDARD	Base SAS
Utility Procedure	SUMMARY	Base SAS
Summary Tables	TABULATE	Base SAS
Transpose	TRANSPOSE	Base SAS
Distribution Analysis	UNIVARIATE	Base SAS
ARIMA Modeling and Forecasting	ARIMA	SAS/ETS
Regression Analysis with Autoregressive Errors	AUTOREG	SAS/ETS
Prepare Time Series Data	EXPAND	SAS/ETS
Basic Forecasting	FORECAST	SAS/ETS
Utility Procedure	PDLREG	SAS/ETS
Regression Analysis of Panel Data	TSCSREG	SAS/ETS
Scatter (3D)	G3D	SAS/GRAPH
Surface	G3D	SAS/GRAPH
Utility Procedure	G3GRID	SAS/GRAPH
Utility Procedure	GANNO	SAS/GRAPH
Bar	GCHART	SAS/GRAPH
Donut	GCHART	SAS/GRAPH
Pie	GCHART	SAS/GRAPH
Contour	GCONTOUR	SAS/GRAPH
Utility Procedure	GOPTIONS	SAS/GRAPH
Area	GPLOT	SAS/GRAPH
Bubble	GPLOT	SAS/GRAPH
Finance	GPLOT	SAS/GRAPH
Line	GPLOT	SAS/GRAPH
Line-multi interpolation support	GPLOT	SAS/GRAPH
Scatter (2D)	GPLOT	SAS/GRAPH
Radar	GRADAR	SAS/GRAPH
Capability Analysis	CAPABILITY	SAS/QC
cdf Plot	CAPABILITY	SAS/QC
Comparative Histogram	CAPABILITY	SAS/QC
Histogram	CAPABILITY	SAS/QC
P-P Plot	CAPABILITY	SAS/QC
Probability	CAPABILITY	SAS/QC
Q-Q Plot	CAPABILITY	SAS/QC
Pareto	PARETO	SAS/QC
Box Chart	SHEWHART	SAS/QC
c Chart	SHEWHART	SAS/QC
Control Charts	SHEWHART	SAS/QC
Individual Measures	SHEWHART	SAS/QC
Means and Ranges	SHEWHART	SAS/QC
Means and Standards	SHEWHART	SAS/QC
np Chart	SHEWHART	SAS/QC
p Chart	SHEWHART	SAS/QC
u Chart	SHEWHART	SAS/QC
One-Way ANOVA	ANOVA	SAS/STAT
Utility Procedure	BOXPLOT	SAS/STAT
Canonical Correlation	CANCORR	SAS/STAT
Cluster Analysis	CLUSTER	SAS/STAT
Discriminant Function	DISCRIM	SAS/STAT
Factor Analysis	FACTOR	SAS/STAT
Utility Procedure	FASTCLUS	SAS/STAT
Linear Models	GLM	SAS/STAT
Life Tables	LIFETEST	SAS/STAT
Logistic Regression	LOGISTIC	SAS/STAT
Nonlinear Regression	NLIN	SAS/STAT
Non-Parametric One-Way ANOVA	NPAR1WAY	SAS/STAT
Proportional Hazards	PHREG	SAS/STAT
Principal Components	PRINCOMP	SAS/STAT
Linear Regression	REG	SAS/STAT
Sampling	SURVEYSELECT	SAS/STAT
Utility Procedure	TREE	SAS/STAT
T Test	TTEST	SAS/STAT

Get Your SAS On or ... Get Your SAS in Gear

Rules for SAS

SAS statements end with a semicolon ; - just like in C#

Case does not matter unlike C#

Continuation is allowed, don't split words though

Some Sample SAS

This lil' procedure produces a report that displays the values of the variables in the SAS data set WEIGHT_CLUB

proc print data=weight_club;

title 'Health Club Data';

run;

The first line, the print procedure, tells the program to display the variables in a simple, organized form. The title slaps a title on the top of the report and run ... well .... that runs the program.

Here is the sample output:

Health Club Data Obs Id Number Name Team Start Weight End Weight Loss 1 1023 David Duke red 189 179 10 2 1049 Dan Quayle yellow 155 144 11 3 1219 Ronnie Reagan red 171 170 1

In order to do some calculations, you need to use the TABULATE procedure.

Proc means procedure.

proc tabulate data=weight_club;

class team;

var StartWeight EndWeight Loss;

table team, mean*(StartWeight EndWeight Loss);

title 'Mean Starting Weight, Ending Weight,';

title2 'and Weight Loss';

run;

And the results ...

MeanStartWeight EndWeight LossTeamred175.33158.3317yellow169.5150.519

You can see that a class, team is declared in this procedure and three variables; StartWeight, EndWeight and Loss.

Notice how it is not necessary to put commas between the variable names.

The next statement performs the mean, or average) calculation ... mean*( StartWeight EndWeight Loss)

Finally, we get some more titles and the run statement.