A variable, in general, is a value that changes.

More specifically, it is a value that changes as a result of direct intervention (independent variable) or a change in another variable (dependent variable).

A variable may be numerical or classificatory, such as gender. In market research, key variables must be identified.

For example, an airline might wish to measure its competitive position according to the attitudes of air travelers. The key independent variables that might influence their attitudes (attitude being the dependent variable) would include reservation service, baggage handling, check-in procedures, seat comfort, and so forth. These would have varying values, depending upon consumer opinion of them.

In performance marketing promotions, both the characteristics of the promotion and consumer response to it are variables. To test a promotion, one or more variables such as offer or copy are changed, to measure resultant changes in response.

In this example, offer and copy are independent variables, and response is the dependent variable.

It is best, when testing, to change only one independent variable per test, so that differences in the dependent variable can clearly be attributed to that change. If more than one independent variable is changed, it will not be clear whether the difference in response is caused by one or more of the changes or, also, whether the change caused by each variable is in the same direction. One change may increase response; the other may decrease it.