Performing Data Analysis Tasks
Previous chapters in the book have shown you how to perform a wealth of common database tasks—everything from extracting information to creating special fields to hold your notes and data. However, IDEA can do a whole lot more. You might not need these special features every day, but it's good to know that they exist for those situations when you do. This chapter shows how to work with a number of advanced database features that include stratification, summarization, random record sampling, and gap detection. In addition, you consider how to work with pivot tables to view data in new ways, join databases to get the complete data picture, and work with monetary unit samples. Even if you won't use these features immediately, follow along with the examples so you become familiar with them. You never know when the need for such features will become apparent.
Stratification relies on the creation of value bands and placing the records from a database within these bands. Each value band represents a range of data and provides a means of categorizing the data in certain ways. You can stratify databases based on character, numeric, or date fields. IDEA supports up to 1,000 stratification bands. After you stratify the database, you can use the results for a number of tasks such as:
- Totaling the number of records in each band, then using the results to look for expected trends.
- Setting high and low cutoff values to test ...