Data cleansing

How to Use SAS - Lesson 5 - Data Reduction and Data Cleaning

• Report rights infringement
• published: 15 Jan 2012
• views: 58760

Bitcoin donations are welcome: 1GGV3gbJeA83FWmz9hDfPri8EuqcUtodXy Mike's SAS Tutorials Lesson 5 This video series is intended to help you learn how to program using SAS for your statistical needs. Lesson 5 introduces the concept of data reduction (also known as subsetting data sets). I discuss how one can subset a data set (i.e. reduce a data set's number of observations) based on some criteria using the IF statement in the DATA STEP, or using the WHERE statement in a PROC STEP. I also discuss using the KEEP, DROP, and RENAME statements for reducing data to only a handful of the original variables (i.e. reduce a data set's number of variables). Furthermore, I show how one can label variables so that descriptive information can be presented in output and value formats so that specific values are easy to understand. Finally, I provide basic examples of each of these for three hypothetical data sets. Helpful Notes: 1. There are two places you can reduce the data you analyze; in the DATA STEP, and in the PROC STEP. 2. To subset data in the DATA STEP, use the IF statement. 3. To subset data in the PROC STEP, use the WHERE statement. 4. Another way to reduce data is to eliminate variables using a KEEP or DROP statement. This method is useful if you are creating a second data set or analytic version of your main dataset. 5. The RENAME statement simply changes a variables name. Today's Code: data main; input x y z; cards; 1 2 3 7 8 9 ; run; proc contents data=main; run; proc print data=main; run; /* 1. Reduce data in the DATA STEP using a simple IF statement */ data reduced_main; set main; if x = 1; run; proc print data=main; run; proc print data=reduced_main; run; /* 2. Reduce data in the PROC STEP using a simple WHERE statement */ proc print data=main; where x = 1; run; proc print data=main; run; proc print data=reduced_main; run; /* 3. Reduce data in the DATA STEP by KEEPing only the variables you do want */ data reduced_main; set main; KEEP x y; run; proc print data=main; run; proc print data=reduced_main; run; /* 4. Reduce data in the DATA STEP by DROPing the variables you don't want */ data reduced_main; set main; DROP y; run; proc print data=main; run; proc print data=reduced_main; run; /* 5. Clean up variables using the RENAME statement within a DATA STEP */ data clean_main; set main; rename x = ID y = month z = day; run; proc contents data=main; run; proc contents data=clean_main; run; /* 6. Clean up variables using a LABEL statement within a DATA STEP */ data clean_main; set clean_main; label ID = "Identification Number" month = "Month of the Year" day = "Day of the Year"; run; proc contents data=main; run; proc contents data=clean_main; run; /* 7. FORMAT value labels using the PROC FORMAT and FORMAT statements */ PROC FORMAT; value months 1="January" 2="February" 3="March" 4="April" 5="May" 6="June" 7="July" 8="August" 9="September" 10="October" 11="November" 12="December"; run; data clean_main; set clean_main; format month months.; run; proc freq data=clean_main; table month; run;