Complete SAS Programming Guide - Learn SAS and Become a Data Ninja - Selecting Variable(s) for Your Index

The video tutorial discusses the importance of being selective when creating indexes in SAS, as they consume resources and take up disk space. It emphasizes the need to choose variables that are often used for subsetting data and are discriminant, meaning they select a small, specific set of observations. The tutorial advises against using non-discriminant variables like age or sex for indexing. It also demonstrates how to use the frequency procedure to determine if a variable is a good candidate for indexing, focusing on those with frequencies in the 1-15% range.

Why is it important to be selective about which variables you create an index on in SAS?

• Indexes consume resources and take up disk space.

• Indexes are automatically created for all variables.

• Indexes are always beneficial regardless of the variable.

• Indexes do not consume any resources.

What is a key consideration when choosing a variable for a composite index?

• The variable should be a text field.

• The variable should be numeric.

• The variable should be frequently used for subsetting.

• The variable should be rarely used.

What does it mean for a variable to be discriminant in the context of SAS indexing?

• It selects a large portion of the data set.

• It selects a very specific small set of observations.

• It is used for sorting data.

• It is a variable with many unique values.

Which of the following is NOT a good candidate for a discriminant variable?

• Social Security number

• Patient ID

• Age

• Employee ID

What frequency range indicates a good candidate for indexing?

• 16 to 20%

• 21 to 25%

• 26 to 30%

• 1 to 15%