How do I access and download ICPSR data sets?
The first step in accessing ICPSR data sets is to locate an appropriate study. You can do this by clicking on Find & Analyze Data, where there are options to search Everywhere, in Title only, in Author/Investigator only, or in Summary only.
The next step is to look at the abstract of the study. Information contained in the summary, time period, data source, sampling, and universe fields will help you to determine whether the study will meet your research objectives. Information contained in the data format field will help you to determine whether the data files can be used with your statistical software program. The collection notes field contains further information about the data set. Restrictions on use of the data are noted in the restrictions field.
If you choose to download the data set, you will first be prompted with information concerning authorization. If you already have a free ICPSR account, type in your email address and password, then click on the "Login" button. New users can select the option to Create an Account. Be sure that you are connecting to ICPSR from the UA network. The download page should acknowledge that you are from the University of Akron.
Your next step is to download the appropriate elements of the data set. The basic elements of a data set are codebooks and data files. Codebooks provide information on the structure, contents, and layout of a data file. Generally, codebooks include a title page, bibliographic citation, data collection description, and codebook body. The codebook body describes the content of the data file and provides information on each variable contained in the data file. Codebooks are available for downloading in compressed and uncompressed PDF format files.
Most data files are ASCII fixed-format files. These files consist of rows and columns of alphanumeric characters. Some older data files are in ASCII format but were created for use with OSIRIS statistical software. OSIRIS data files must be used in conjunction with OSIRIS dictionary files. Data files are available for downloading in compressed and uncompressed formats.
Many data sets also contain data definition statements for specific software programs (e.g., SPSS, SAS). Data definition statements (also known as control cards) contain the syntax or program code needed to read raw data into a statistical package, such as SPSS or SAS. Data definition statements link the columns of alphanumeric data contained in the ASCII data files to the variables described in the codebook. It makes sense to download these data definition statements whenever available. Data definition statements are available for downloading in compressed and uncompressed formats.
Some data sets also contain SPSS portable (export) files or SAS transport files. SPSS portable files are not specific to a particular SPSS version or computer platform. These files can be opened directly in SPSS statistical software without the use of data definition statements. SAS transport files are also not specific to a particular computer platform, and can be opened directly in SAS statistical software without the use of data definition statements. In both cases, it is not necessary to download either the data definition statements or the data file; however, it is still necessary to download or refer to the codebook in order to make sense of the data. SPSS export files and SAS transport files are available for downloading in compressed and uncompressed formats.
Once you have downloaded the appropriate files of the data set, you can begin using these with your statistical software package.