Some terms are of importance in multi dimensional data models:

• Granularity describes the level of detail of an info cube. A cube can be detailed only to the level of the villages, it could also be detailed to the level of the households.
• Drill down is an OLAP process, which zooms into detail. E.g. a researcher could drill further into the data, if he splits up the data of theprevious example into age-groups.
• Roll up is the process of further aggregating the data. A roll up of the previous example could take out the dimension ‘Sex’ and aggregate the data according to the dimensions ‘Town’ and ‘Ethnic Group’.
• Slicing is taking one category out of the cube, to analyse it seperately.
• Dicing is accordingly taking out several categories from a cube.

With the aid of these processes, it is possible to create specific output tables from all the potential outputs, which are contained in the info cube.