SAS 9.4 Language Reference, 6th Edition

Understanding BY Groups

BY Groups with a Single BY Variable

The following figure represents the results of using a single BY variable, zipCode, in a

DATA step. The input data set,

zip contains street names, cities, states, and ZIP codes.

The groups are created by specifying the variable zipCode in the BY statement. The

DATA step arranges the zipcodes that have the same values into groups.

The figure shows five BY groups that are created from the examples Example Code 22.1

on page 461 and Example Code 22.2 on page 462..

The first BY group contains all observations with the smallest value for the BY variable

zipCode. The second BY group contains all observations with the next smallest value

for the BY variable, and so on.

Figure 22.1 BY Group Using a Single BY Variable (ZipCode)

Example Code 22.1 Create the Zip Data Set

data zip;

input zipCode State $ City $ Street $20-29;

datalines;

85730 AZ Tucson Domenic Ln

85730 AZ Tucson Gleeson Pl

33133 FL Miami Rice St

33133 FL Miami Thomas Ave

33133 FL Miami Surrey Dr

33133 FL Miami Trade Ave

33146 FL Miami Nervia St

Understanding BY Groups 461

33146 FL Miami Corsica St

33801 FL Lakeland French Ave

33809 FL Lakeland Egret Dr

;

You can then specify the BY variable in the DATA step using the following code:

Example Code 22.2 Sort and Group the zipCode Data Set by a Single Variable

proc sort data=zip;

by zipcode;

run;

data zip;

set zip; by zipcode;

run;

proc print data=zip noobs;

title 'BY-Group Uing a Single Variable: ZipCode';

run;

BY Groups with Multiple BY Variables

The following figure represents the results of processing the zip data set with two BY

variables, State and City. This example uses the same data set as in

Example Code 22.1

on page 461, and is arranged in an order that you can use with the following BY

statement:

by State City;

The figure shows three BY groups. The data set is shown with the BY variables State

and City printed on the left for easy reading. The position of the BY variables in the

observations does not affect how the values are grouped and ordered.

The observations are arranged so that the observations for Arizona occur first. The

observations within each value of State are arranged in order of the value of City. Each

BY group has a unique combination of values for the variables State and City. For

example, the BY value of the first BY group is AZ Tucson, and the BY value of the

second BY group is FL Lakeland.

462 Chapter 22 • BY-Group Processing in the DATA Step

Figure 22.2 BY Groups with Multiple BY Variables (State and City)

Here is the code for creating the output shown in the figure Figure 22.2 on page 463 :

Example Code 22.3 Create the Zip Data Set

/* BY Groups with Multiple BY Variables */

data zip;

input State $ City $ Street $13-22 ZipCode ;

datalines;

FL Miami Nervia St 33146

FL Miami Rice St 33133

FL Miami Corsica St 33146

FL Miami Thomas Ave 33133

FL Miami Surrey Dr 33133

FL Miami Trade Ave 33133

FL Lakeland French Ave 33801

FL Lakeland Egret Dr 33809

AZ Tucson Domenic Ln 85730

AZ Tucson Gleeson Pl 85730

;

Example Code 22.4 Sort and Group the zipCode Data Set by Multiple BY Variables

proc sort data=zip;

by State City;

run;

data zip;

set zip;

by State City;

run;

proc print data=zip noobs;

title 'BY Groups with Multiple BY Variables: State City';

run;

Understanding BY Groups 463