In data management and database analysis, a data domain refers to all the unique

values which a data element may contain. The rule for determining the domain bo
undary may be as simple as a data type with an enumerated list of values.
For example, a database table that has information about people, with one record
per person, might have a "gender" column. This gender column might be declared
as a string data type, and allowed to have one of two known code values: "M" for
male, "F" for female and NULL for records where gender is unknown or not applicab
le (or arguably "U" for unknown as a sentinel value). The data domain for the ge
nder column is: "M", "F".
In a normalized data model, the reference domain is typically specified in a ref
erence table. Following the previous example, a Gender reference table would hav
e exactly two records, one per allowed value excluding NULL. Reference tables are
formally related to other tables in a database by the use of foreign keys.
Less simple domain boundary rules, if database-enforced, may be implemented thro
ugh a check constraint or, in more complex cases, in a database trigger. For exa
mple, a column requiring positive numeric values may have a check constraint dec
laring that the values must be greater than zero.
This definition combines the concepts of domain as an area over which control is
exercised and the mathematical idea of a set of values of an independent variab
le for which a function is defined. See: domain (mathematics).