Case and frequency form data

Most data are available in case form with data for each sampling unit. Sometimes data are not available for each unit but are already summarized by counting the frequency of occurrences of each value, called frequency form data.

Whenever possible you should record data in case form. Case form data can be reduced to frequency form, but frequency form data cannot be reconstructed into case form. You may want to use frequency form data when working with a large amount of data from a database, where you can save computing resources by letting the database server tabulate the data.

Case form dataset layout

Each column is a variable (Height, Eye color), each row is a separate case (Subject) with the values of the variables on that case.

Subject (optional) Height Eye color
1 175 Blue
2 180 Blue
3 160 Hazel
4 190 Green
5 180 Green
6 150 Brown
7 140 Blue
8 160 Brown

Frequency form dataset layout

Each column is a variable (Eye color) and a seperate column for the number of cases (Frequency), each row is a combination of categories with the frequency count.

Eye color Frequency
Brown 221
Blue 215
Hazel 93
Green 64
Note: Examples of dataset layouts are included in the Statistical Reference Guide.
Related concepts