SAS: PROC FREQ with Examples

This tutorial explains how to use PROC FREQ with examples in SAS.

What does PROC FREQ do?

The PROC FREQ procedure is used to summarize categorical variables in SAS. It calculates count/frequency and cumulative frequency of categories of a categorical variable. It can also produce bar charts and tests for association between two categorical variables.

Create a Sample Dataset

The program below creates a sample SAS dataset which will be used to explain examples in this tutorial.

data example1;
input x y $ z;
cards;
6 A 60
6 A 70
2 A 100
2 B 10
3 B 67
2 C 81
3 C 63
5 C 55
;
run;

The created dataset looks like below -

X	Y	Z
6	A	60
6	A	70
2	A	100
2	B	10
3	B	67
2	C	81
3	C	63
5	C	55

Example 1 : To check the distribution of a categorical variable (Character)

Suppose you want to see the frequency distribution of variable 'y'.

proc freq data = example1;
tables y;
run;

The TABLES statement tells SAS to return n-way frequency and cross-tabulation tables and computes the statistics for these tables.

It answers a question 'which category holds the maximum number of cases'. In this case, the category 'C' contains maximum number of values.

Example 2 : To remove unwanted statistics in the table

Suppose you do not want cumulative frequency and cumulative percent to be displayed in the table. The option NOCUM tells SAS to not to return cumulative scores.

proc freq data = example1;
tables y /nocum;
run;

If you want only frequency, not percent distribution and cumulative statistics.

proc freq data = example1;
tables y /nopercent nocum;
run;

Example 3 : Cross-Tabulation ( 2*2 Table)

Suppose you want to see the distribution of variable 'y' by variable 'x'.

proc freq data = example1;
tables y * x;
run; 

The output of the above SAS program is shown in the image above.

Example 4 : Show Table in List Form

Suppose you do not want output to be shown in tabular form. Instead, you want final analysis to be displayed in list form (See the image below)

proc freq data = example1;
tables y * x / list;
run;

The forward slash followed by LIST keyword produces the list styled table.

Example 5 : Hide Unwanted Statistics in Cross Tabulation

proc freq data = example1;
tables y * x / norow nocol nopercent;
run;

The NOROW option hides row percentage in cross-tabulation. Similarly, NOCOL option suppresses column percentage.

Example 6 : Request Multiple Tables

Suppose you want to generate multiple crosstabs. To accomplish it, you can run the command below-

proc freq data = example1;
tables y * (x z) / norow nocol nopercent;
run;

The tables y*(x z) statement is equivalent to tables y*x y*z statement. In this case, it returns two tables - y by x and y by z.

Example - tables (a b)*(c d); is equivalent to tables a*c  b*c  a*d  b*d;

Example 7 : Number of Distinct Values

The NLEVELS option is used to count number of unique values in a variable.

proc freq data = example1 nlevels;
tables y;
run;

In this case, it returns 3 for variable Y.

Example 8 : Use WEIGHT Statement

The WEIGHT statement is used when we already have the counts. It makes PROC FREQ use count data to produce frequency and cross-tabulation tables.

Data example2;
input pre $ post $ count;
cards;
Yes Yes 30
Yes No 10
No Yes 40
No No 20
;
run;
proc freq data=example2;
tables pre*post;
weight count;
run;

Example 9 : Store result in a SAS dataset

Suppose you wish to save the result in a SAS dataset instead of printing it in result window.

proc freq data = example1 noprint;
tables y *x / out = temp;
run;

The OUT option is used to store result in a data file. NOPRINT option prevents SAS to print it in results window.

Example 10 : Run Chi-Square Analysis

The CHISQ option provides chi-square tests of homogeneity or independence and measures of association between two categorical variables.  Also it helps to identify the statistically significant categorical variables that we should include in our predictive model. All the categorical variables with a chi-square value less than or equal to 0.05 are kept.

proc freq data = example1 noprint;
tables y * x/chisq;
output All out=temp_chi chisq;
run; 

Example 11 : Generate Bar Chart and Dot Plot

The bar chart can be generated with PROC FREQ. ​To produce a bar chart for variable 'y', the plots=freqplot (type=bar) option is added. By default, it shows frequency in graph. In order to show percent, you need to add scale=percent. The ODS graphics ON statement tells SAS to produce graphs. Later we turn it off.

Ods graphics on;
Proc freq data=example1 order=freq;
Tables y/ plots=freqplot (type=bar scale=percent);
Run;
Ods graphics off;

Similarly, we can produce dot plot by adding type=dot. See the implementation below-

Ods graphics on;
Proc freq data=example1 order=freq;
Tables y/ plots=freqplot (type=dot);
Run;
Ods graphics off;

Example 12 : Include Missing Values in Calculation

By default, PROC FREQ does not consider missing values while calculating percent and cumulative percent. The number of missing values are shown separately (below the table). Refer the image below.

Proc freq data=sashelp.heart;
Tables deathcause;
Run; 

By adding MISSING option, it includes missing value as a separate category and all the respective statistics are generated based on it.

Proc freq data=sashelp.heart;
Tables deathcause / missing;
Run;

Example 13 : Ordering / Sorting

In PROC FREQ, the categories of a character variable are ordered alphabetically by default. For numeric variables, the categories are ordered from smallest to largest value.

To sort categories on descending order by frequency (from largest to smallest count), add ORDER=FREQ option.

Proc freq data=sashelp.heart order = FREQ;
Tables deathcause / missing;
Run;

It is generally advisable to show distribution of a nominal variable after sorting categories by frequency. For ordinal variable, it should be shown based on level of categories.

To order categories based on a particular FORMAT, you can use order = FORMATTED option.

Related Posts

SAS Tutorials : Top 100 SAS Tutorials

Spread the Word!
Share Share Tweet

About Author:

Deepanshu founded ListenData with a simple objective - Make analytics easy to understand and follow. He has over 10 years of experience in data science. During his tenure, he worked with global clients in various domains like Banking, Insurance, Private Equity, Telecom and HR.

While I love having friends who agree, I only learn from those who don't
Let's Get Connected Email LinkedIn