Scribd
Upload
50 views14 pages

Chapter 4 - Preparing Data For Analysis: Bradley Paulse Office 1.87 (Cams) Brpaulse@uwc - Ac.za

This chapter discusses preparing data for analysis in SAS. It covers labeling variables, creating new variables using functions and calculations, using conditional statements like IF-THEN-ELSE, subsetting data using conditional statements, selecting or dropping variables, joining data sets by appending or merging, and sorting data sets. Examples provided illustrate these techniques.

Uploaded by

jkleinhans.jk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
50 views14 pages

Chapter 4 - Preparing Data For Analysis: Bradley Paulse Office 1.87 (Cams) Brpaulse@uwc - Ac.za

This chapter discusses preparing data for analysis in SAS. It covers labeling variables, creating new variables using functions and calculations, using conditional statements like IF-THEN-ELSE, subsetting data using conditional statements, selecting or dropping variables, joining data sets by appending or merging, and sorting data sets. Examples provided illustrate these techniques.

Uploaded by

jkleinhans.jk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Chapter 4 - Preparing Data for

Analysis

Bradley Paulse
Office 1.87 (Cams)
[email protected]
Labelling Variables

• Assign labels in the data step.

• Make sure that your label is descriptive of the variable.

• If the data set is permanent then the labelling is permanent.

• Assign labels using the label’ statement in the data step.

• Output the labels in the proc print step using the label’ option.

*dlabel.sas,

*dlabel2.sas

*dlabel3.sas
Creating New Variables

• Performed or calculated in the data step.

• Variables are created/assigned when the program is run.

• The order of numeric calculations is important.

• Constant variables can also be created.

*dcalc.sas

*dlabel3
SAS Functions

• Functions that can be used when assigning variables

• Form: variable = function(argument1, argument2,…);

• Some important functions:


o S = INT(x);
o S = MAX(x1, x2, x3, …);
o S = MIN(x1, x2, x3, …);
o S = SUM(x1, x2, x3, …);
o S = ROUND(x,round);
o S = INTCK(‘interval’,startdate, enddate);

*time.sas
SAS Functions

• Some important functions:


o S = INTCK(‘interval’,startdate, enddate);
o S = MDY(month, day, year);

*ddates.sas

ddates2.sas
Conditional Statements

• IF-THEN ELSE is a conditional statement that creates variables based on


the evaluation of the expression.

• If the criteria for the expression are not met then ELSE statement is
executed.
Conditional Statements

• See page 86 for a list of SAS comparison operators

• Multiple expressions can be joined using either AND or OR


Conditional Statements

• Can also stack the conditional statements by adding the ELSE-IF


statement.

• Missing values are assigned by a “.”

*dcondition.sas
Sub setting data sets

• Smaller data sets can be created from a larger main data set by using the
SET statement along with the conditional IF statement.

• Observations are only excluded from the output data set.


Selection of variables

• Variables can be included or excluded from the output data set using the
DROP and KEEP statement in data step
• The DROP and KEEP statement only applies to the output data set.
Joining data sets

• Data sets can either be joined by appending or merging data sets to


create a new data set.
• By appending data sets, observations are merely added to a new data set.
• By merging data sets, observations can joined by a common variable.

• To APPEND data sets, use the SET statement.


• To MERGE data sets, use the MERGE statement with the BY statement.
Joining data sets

*dappend1.sas and dmerge.sas


Sorting data sets

• Data sets can be sorted ascending or descending using the PROC SORT
procedure
• Data sets can are sorted according to a single or multiple variables

• The default sorting sequence if it is not specified is ascending


• PROC SORT can either assign a newly sorted data set or sort the current
data set

*dsort.sas
READ THROUGH CHAPTER 4

You might also like