In SEER*Stat, a variable is a dictionary entry which contains format groupings assigned to data values (values of data fields in the tumor record). Variables provide a mechanism for grouping individual data values into one statistic ("Ages 65+" is a grouping commonly used for analyses by age). A unique dictionary of variables is associated with each SEER*Stat database. The following terms are used for the 4 types of SEER*Stat variables:
- Standard Variable - Standard variables are the variables distributed in a database, they cannot be modified or deleted in SEER*Stat. These are created when the SEER*Stat database is created by processing the text data via SEER*Prep. A default set of format groupings are defined for most standard variables. For example, the year of diagnosis variables in the SEER 9 databases are distributed with one grouping for all years combined (1973-2007) and separate groupings for each individual year. Some variables, like histologic type, do not have any default formats, just a range of unlabeled values.
- User-defined Variable - a variable that you define based on one variable. Typically, you would create a user-defined variable based on a Standard Variable. For example, the "1973-2007" grouping for year of diagnosis would be inappropriate if you selected only 1992-2007 for your analyses. You would need to create a user-defined version based on year of diagnosis with separate groupings for the individual years (1992, 1993,...,2007) and, if desired, a single grouping for "1992-2007" combined.
- Merged Variable - In essence, a Merged Variable is a type of user-defined variable. A Merged Variable is a variable whose format groupings can be based on 2 or more variables. As an example, this may be used to create groupings to show cancer in specific age-sex combinations: "Women <50", "Women 50+", "Men <65", "Men 65+".
- Calculated Variable - A calculated variable is a variable that is not coded in the database; that is, the field is not on the tumor record. SEER*Stat determines its values based on the values of other variables and system specifications. For example, Age at Prevalence Date is a calculated variable used in Limited-Duration Prevalence sessions. The values of this variable are determined from the prevalence date selected for the analysis and either date of birth or date and age at diagnosis. User-defined variables based on calculated variables are displayed in a separate folder labeled “Calculated.”
Using SEER*Stat Variables
The following tutorials will help you learn to use the different variable types in SEER*Stat sessions.
- Merged Variables:
- Rate Exercise 3
- Calculated Variables:
- Prevalence Exercise 1