Political Science Data Services : Data Formats

Data can be found in a variety of formats. In many cases, data will be reported in partial form, that is, as a subset of a larger data set. This data is often presented in the form of tables. For example, the following table contains data on one variable, the number of American children living in single-parent families:

Children in Single-Parent Families

In other cases, data will be provided in datasets. Datasets usually are much larger, containing a wider set of variables, and typically you can't see them on screen on the data archive or website. You will usually have to open these datasets in a program like SPSS or Excel.

When you find data you want, you will need to be able to use it. In some cases, you may just want to include data from a table in your own paper. In other cases, you may wish to analyze the data.   

If you just want to include the data, you can paste it into a Word document, or else create a table in Word and enter the data values in the table. However if you want to analyze the data, you will need to get it into a suitable computer program. Excel and SPSS are probably the two most common programs that you might want to use.

Getting Data into an Analysis Program
There are two basic ways to get data into an analysis program. Some data producers make datasets available, most commonly in a generic format -- for example, comma-delimited data -- or else in a proprietary format, like SPSS or Excel data files. In most cases, these datasets can easily be "read into" the data analysis program you want to use.   

Most programs that use datasets (again, Excel and SPSS are excellent examples) will read and write datasets in a wide variety of formats, so this makes your task a lot easier. If not, specialized programs like StatTransfer can take datasets in one format and change them into another format.  

Alternatively, if you have data that is only made available in table form, you can manually copy it into your analysis program. If your data is in electronic form, you can usually copy and paste it into your analysis program using standard cut and paste commands. Excel, in particular, is good at recognizing table formatting and retaining it in its spreadsheets. The original table may look a bit different (for example, it may have background colors in the cells) but you can usually modify this in Excel.