Suppose that your data file has two different kinds of records, family records and person records. How do you read the data so that the family information is included for each person?
Here is an example dataset with two kinds of records: Family records and person records. The data are organized such that the family record comes first and all the person records for that family follow it. The family records are the shorter data lines and the person records are the longer ones.
06470 1 1 32161 232 0 19082 230 1 07470 1 0 11231 240 1 08470 1 0 43711 227 0 09470 1 0 40221 213 1 41162 222 0 16173 224 1 10470 1 1 30111 220 0 36222 211 1 11470 1 0 21751 217 0 33962 210 1 32143 226 1 |
Here are the codebooks for the family and person records.
family record: person record:
column 1-5 family id column 1-4 person id
column 7 record type column 5 person number
(1 = family) column 7 record type
column 9 group (2= preson)
column 8-9 age
column 11 male
The following syntax example reads and displays in the data from the two different types of records. The file type is nested and the number in column seven indicates which record type the data belong to. There is a separate data list for each record type.
file type nested record=7. record type 1. data list / famid 1-5 group 9. record type 2. data list / personid 1-4 pernum 5 age 8-9 male 11. end file type. begin data 06470 1 1 32161 232 0 19082 230 1 07470 1 0 11231 240 1 08470 1 0 43711 227 0 09470 1 0 40221 213 1 41162 222 0 16173 224 1 10470 1 1 30111 220 0 36222 211 1 11470 1 0 21751 217 0 33962 210 1 32143 226 1 end data. list.
Here is what the final dataset looks like.
FAMID GROUP PERSONID PERNUM AGE MALE 6470 1 3216 1 32 0 6470 1 1908 2 30 1 7470 0 1123 1 40 1 8470 0 4371 1 27 0 9470 0 4022 1 13 1 9470 0 4116 2 22 0 9470 0 1617 3 24 1 10470 1 3011 1 20 0 10470 1 3622 2 11 1 11470 0 2175 1 17 0 11470 0 3396 2 10 1 11470 0 3214 3 26 1 Number of cases read: 12 Number of cases listed: 12