Instructions for Using the IPUMS-CPS
Data Extraction System

The IPUMS-CPS data extraction system allows researchers to fashion extracts of the data oriented to their own specific research needs and available computing resources. In practice, researchers never require all variables and all cases from a sample. Researchers can design subsamples incorporating a subset of variables and surveys pertaining to the specific population(s) of interest to them.

The Extract Interface

The extract procedure involves a login screen followed by a series of 4 web pages, with the contents of each depending on selections made on the previous page. On each page, you make the desired (and required) selections and click the gray bar to advance to the next screen.

Creating a New Extract

On the first extract page, you are prompted for your e-mail address, which acts as your password and provides us with a means of contacting you and constructing a unique file name for your extract output.

To use the extract system, you must first register. Users are automatically registered when they agree to all conditions for use.

You can log in as "guest" in order to examine the extract interface. The system will not actually create a data extract, however, until you have gone through the registration process.

Step 1 -- Sample Selection

In Step 1 of the extract procedure, you define the general characteristics of your desired extract.

Choose the preferred file structure for your extract: rectangular ("flat," all household information attached to respective household members) or hierarchical (household record followed by person records). The system defaults to rectangular format, which is the overwhelming choice of researchers.

All data are produced in unix-compressed format. Most uncompression software has no difficulty with the files. We have noted, however, that older versions of Netscape (versions 3 and 4) corrupt the compressed files unless specific advanced features are implemented. We suggest using a newer version of Netscape or an alternative browser.

The system produces only ASCII column-format data, but it will generate SAS, SPSS, or Stata command files to facilitate reading the data into one of those statistical packages. The command files contain the column locations of variables, variable labels, and value labels for categorical variables.

Finally, you select the particular sample or combination of samples you want.

Step 2 -- Variable Selection

In Step 2, you select which variables you want to include in your extract. Only those variables available for the particular samples selected in Step 1 are displayed as options. If you have selected multiple survey years, all variables occurring in any of the specified samples are displayed. Some variables have a second check box allowing you to select cases based on the value of the variable.

On the right-hand side of variable selection page, there is a column for each survey year selected on the previous (sample selection) page. Only columns for selected years are displayed, with the symbols showing the availability of each variable across years. An "X" indicates that the variable is available for that sample.

Selected variables have a checkbox to the right of their variable label. Clicking here allows you to filter out cases based on particular values for these variables. (For example, you can use case selection on the SEX variable to choose only females for your extract.) The details of case selection are handled in the next screen.

Step 3 -- Case Selection

Step 3 provides for case selection. Only those variables chosen for case selection in Step 2 will appear on this page. Case selection will limit the extract to include only cases that contain the selected values for the listed variables.

You can choose either "and" or "or" selections to filter across variables using case selection. You also have the option of selecting only those individuals with the selected characteristics or entire households containing any individual with the selected characteristics.

On this page you may also choose to include data quality flags. Only flags associated with the variables selected on the previous screen appear as choices.

Step 4 -- Extract Summary

In the final step, you review your selections on a summary screen. Use the buttons near the bottom of the screen to jump to particular points in the extract process to alter your selections. When you are satisfied with your extract design, submit it for processing. The system will inform you via e-mail when the extract is completed and provide instructions for downloading the files. For each extract, you receive data, codebook, and command files. To access the files, click on the "Download Extracts" link on the left navigation bar on the main CPS page.