Example 10.3: BLS State and Area, Employment, Hours and Earnings Surveys

The DATASOURCE Procedure

Example 10.3: BLS State and Area, Employment, Hours and Earnings Surveys

This example illustrates how to extract specific series from a State and Area, Employment, Hours and Earnings Survey. The series to be extracted is total employment in manufacturing industries with respect to states as of March, 1990.

The State and Area, Employment, Hours and Earnings survey designates the totals for manufacturing industries by DIVISION='3', INDUSTRY='0000', and DETAIL='1'. Also, statewide figures are denoted by AREA='0000'.

The data type code for total employment is reported to be 1. Therefore, the series name for this variable is SA1, since series names are constructed by adding an SA prefix to the data type codes given by BLS.

The following statements extract statewide figures for total employment (SA1) in manufacturing industries for March, 1990:

   filename datafile 'host-specific-file-name' <host-options>;
   proc datasource filetype=blseesa out=totemp;
      where division='3' and industry='0000' and detail='1' and
            area='0000';
      keep sa1;
      range from 1990:3 to 1990:3;
      rename sa1=totemp;
   run;

Variations of women workers in manufacturing industries with respect to states can best be demonstrated on a map of the United States, as shown in Output 10.3.1.

Output 10.3.1: Map of the Series in the OUT= Data Set for FILETYPE=BLSEESA

Note the following for the preceding example:

The INFILE= option is omitted, since the fileref assigned to the BLSEESA file is the default value DATAFILE.
When the FROM and TO values in the RANGE statement are the same, only one observation for each cross section is extracted. This observation corresponds to a monthly data point since the INTERVAL= option defaults to MONTH.

Chapter Contents
Previous
Next
Top