Nominal Variable Definition

A nominal variable is a variable that is not continuous and has no order. It is part of the typology of variables developed by Stanley Smith Stevens, which also includes ordinal, interval and ratio variables.

Formal Definition

A nominal variable can be transformed in any way that neither combines nor splits levels of the variable. That is, the levels of the variable are simply labels, with no inherent meaning.

Examples

Some common examples of nominal variables are marital status (single, married, widowed, separated, divorced, other), hair color (red, blond, brown, gray, white) and racial group (African-American, white, American Indian/Alaskan Native, Asian, native Hawaiian/Pacific Islander).

Nominal Variables Treated Other Ways

Some variables may be nominal in some purposes and not in others. For example, hair color might, for some purposes, be looked at as the wavelength of that color. Racial group might be ordered in terms of stigma attached to that group.

Other Variables Treated as Nominal

Education would usually be treated as having an order, such as "grade school only," "some high school," "high school graduate," "some college," "college grad," "some post-college education" and "graduate degree." People who went to college have more education than those who did not. But it might be useful to treat education as nominal: "graduated and stopped," "stopped without graduating," "still in school."

Benefits

The principal benefit of treating a variable as nominal is that it makes no assumptions about the variable.

References

• "Science"; On the theory of scales of measurement; S.S. Stevens; 1946