Describe what are the roles of dimensional analysis and OLAP in the interpretation of data?
Dimensional analysis and OLAP are somewhat synonymous methods of browsing data. Primarily this refers to data in "multidimensional" cube structures or, at least, modeled dimensionally (using star or snowflake approaches). The general flow of a query session using dimensional analysis or OLAP is:
- What measures are you after? This selects the fact table and the measure.
- What dimensions do you want to filter the results by?
- What values within those dimensions (any level) do you want to filter the results by?
If properly done, dimensional modeling inherently implies exactly how the data can be accessed and should be interpreted because it defines the relationships amongst the data that can be traversed using the three steps above.