Data Dictionaries are an RML data model that capture details on the field level about the data in a system or systems. During the requirements phase, the focus is not on actual data in the database or technical design required to implement the business data objects within the database. Instead the focus is on how the business users categorize fields into business data objects.

Capturing all of this information in a Data Dictionary allows you to look at the properties of all the data in your system in a very analytical, centralized, and structured fashion.  As you can see from the models provided, each row of the table is an individual field and each column is a property of that field.  

It is important to note, this model is most applicable when the system has a database back end. It should also be used anytime you have business data objects with fields that have validation criteria. 

Learn more about requirements models in our book Visual Models for Software Requirements (Microsoft Press, 2012) Joy Beatty and Anthony Chen. (See Chapter 21 for Data Dictionary)