### What types of weights do SAS, Stata and SPSS support?

Types of weights

There are several types of weights that you might find or create in a data set.

• probability weights - Perhaps the most common type of weights are probability weights.  These weights represent the probability that a case (or subject) was selected into the sample from a population.  These weights are calculated by taking the inverse of the sampling fraction.  For example, if you have a population of 10 widgets and you select 3 into your sample, your sampling fraction would be 3/10 and your pweight would be 10/3 = 3.33.  You frequently find this type of weight in survey data.
• frequency weights - Frequency weights are whole (i.e., integer) numbers that tell the program how many cases each case represents.  It is a kind of short cut:  if you have five rows of data that are identical, you can use a frequency weight with a value of 5 and spare yourself having to input the same row five times.
• importance weights - Importance weights are just what you think they should be - they are weights that indicate how "important" a case is.  There is no standard way of calculating this type of weight.
• analytic weights - Analytic weights are perhaps the least common type of weight.  This type of weight is used when the cases are actually an average.  If the averages are based on different numbers of observations (for example, some averages are based on three observations and others are based on 30 observations), some cases (averages) are measured with more precision than others, and you want the more precisely measured cases to have a greater weight than the less precisely measured cases.  The more measurements used in the average, the more precise the average will be.  The weights are proportional to the inverse of the variance.

In SAS

You need to read the documentation for the proc that you are using to determine what kind of weight will be used with the weight statement.  The weight statement used in one proc might assume frequency weights while another assumes probability weights.  Some procs will handle the weights differently depending on the values of the weight variable.  For example, if all of the values of the weight variable are integers (whole numbers), SAS will assume that you have a frequency weight.  If you specify a different weight variable that has decimals, then the proc will assume that you have a probability weight.   If you cannot tell from the documentation which type of weight will be used, you will either need to do some experimenting or contact SAS technical support.

In SPSS

In Stata

Stata recognizes all four type of weights.  You can specify which type of weight you have by using the weight option after a command.  Note that not all commands recognize all types of weights.  If you use the svyset command, the weight that you specify must be a probability weight. You can find out more about using weights in Stata by seeing help weight.

The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California.