Sentry Page Protection

Please Wait...

Data Analysis [10-15]

---

Output Data Set
(Proc Means)

The output data set requires special attention when having classification variable(s).
​
Example

Data Income; Input ID $ Gender $ Edu :$15. Income; Datalines; FID8001 Male Bachelor 56000 FID8002 Female Master 80000 FID8003 Male Bachelor 45000 FID8004 Male PhD 103000 FID8005 Male Bachelor 98000 FID8006 Female Bachelor 68000 FID8007 Male Highschool 46000 FID8008 Male Bachelor 70000 FID8009 Female Highschool 30000 FID8010 Female Bachelor 60000 FID8011 Male Bachelor 84000 FID8012 Female Master 89000 FID8013 Male PhD 76000 FID8014 Male PhD 66000 FID8015 Male Bachelor 84000 FID8016 Female PhD 85000 FID8017 Male Highschool 65000 FID8018 Male Bachelor 150000 FID8019 Female Master 48000 FID8020 Female Master 49000 ; Run;

The INCOME data set contains 4 variables:

• FID: ID Number
• Gender: Gender
• Edu: Education
• Income: Income Level

A social researcher wants to find out if gender and education play a role in income discrepancy.

The mean and standard deviation of the population income are computed for male and female at the 4 education levels.

Example

Proc Means Data=Income;
Var Income;
Class Gender Edu;
Output out = Stat Mean=Mean STD=STD;
Run;

Note: both the GENDER and EDU are listed as the classification variables.

​Let's take a look at the output data set:

The output data set contains 4 combinations of the classification variables:

(1) _TYPE_ = 0

When _TYPE_ = 0, the statistics are computed without any classification variable. ​

There is only 1 observation showing the overall results for the entire population.
​

(2) _TYPE_ = 1

When _TYPE_ = 1, the statistics are computed using only Education as the classification variable.

There are 4 observations showing the results for the four education level: Bachelor, High school, Master, and PhD.

Gender is not used in the classification.
​

(3) _TYPE_ = 2

When _TYPE_ = 2, the statistics are computed using only the Gender as the classification variable.

There are 2 observations showing the results for male and female.

Education is not used in the classification.
​

(4) _TYPE_ = 3

Finally, when _TYPE_ = 3, the statistics are computed by using both the Gender and Education as the classification variables.

The statistics are computed across all of the classification variables.

The output data set could look very confusing when involving more classification levels.

Pay good attention to the variable _TYPE_ when examining the results!

---

Exercise

Locate the CARS data set from the SASHelp library.

Compute the Mean and Standard Deviation of Horsepower of the cars, using both the MAKE and TYPE as the classification variables.

Create an output data set for the results.

Next

---

Need some help? 

Get Hint

HINT:
You might want to subset the data set to include only the classification levels you need.

Get Solution

SOLUTION:
Proc Means Data=SASHelp.cars noprint;
Var Horsepower;
Class Make Type;
Output Out = Cars2 Mean=Mean STD=STD;
Run;

Data Cars3;
Set Cars2;
if _Type_ = 3;
Run;

or

Proc Means Data=SASHelp.cars nway noprint;
Var Horsepower;
Class Make Type;
Output Out = Cars2 Mean=Mean STD=STD;
Run;

---

Fill out my online form.

---