IBIS-Q System Documentation - Preparing SAS Datasets
Step-by-Step for Preparing a SAS DatasetThe IBIS System requires SAS and that each dataset is in SAS format. For more information about the needed SAS software see the IBIS-PH Overview page.
- All datasets need to be in SAS format. If you are using a rate measure, such as Mortality Crude Rate, the Mortality and the Population datasets both need to be in SAS format.
- Decide what variables will be used and how to display them on interface. For example, if you want age you need to decide whether it will be displayed in 11 age groups, 4 age groups, etc.
- Check variables availability in dataset, if using a rate the variable must be available in both datasets. The variables can be named differently in each dataset as long as they are coded the same. For example, Mortality Crude Rate, if you want to use 11 age groups then a variable with 11 age groups must be available in both Mortality data and Population data.
- If using a PROC Sum procedure in your SAS function files you need to have a variable such as "x" for every record that has a value of 1. This provides a variable for SAS to sum. If you are not using the PROC Sum procedure (as in the survey modules) then this summing variable is not needed. In the Population data there is a variable called "popcount" that gives a population number for the other variables in that record and this "popcount" is used as the summing variable.
- Regroup variables if needed. For example, if you have a weight or age variable, but you want to use weight or age categories (instead of continuous weights or single years of age) on the interface then you need to regroup those variables into the needed categories (0-4 years old, 5-10 years old, etc.).
- Make variables numeric if possible
To see an example of a Population data with the "popcount" variable or Mortality data with the "x" variable see the Dataset Contents page. Both the "x" and the "popcount" variables are used by the PROC Sum procedure in SAS. You can also see variables that are regrouped, such as the agepop variable. The agepop variable has been regrouped to use 11 age groups: less than 1, 1-4, 5-14, 15-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75-84, 85+.