The connections between data and metadata can be quite complex. One fairly common example of the complicated nature of data systems—and why metadata are so important—is the concept of "instantiation," which refers to how a piece of data may change in presentation or format without changing in meaning at various "instants" during its life cycle within a data system. For example, note the flow of gender data that might occur in a typical education organization's data system:
|Collection format (on the form):||Sex = Boy or Girl|
|System entry interface (how the data are entered):||Sex = M or F|
|Data warehouse format (how the data are stored):||Sex = 1 or 2|
|Data report (how users see the data):||Sex = Male or Female|
Such change in nomenclature for the same information throughout a single system requires a mechanism for data users and managers to identify element code lists at each instance, regardless of the changing source and target. A robust metadata system is the most systematic, reliable, and accurate way of accomplishing this vital data management task.