Definition of Control Variable


A control variable is a variable that is held constant or whose impact is removed in order to analyze the relationship between other variables without interference. For example, suppose the relationship between age and frequency of delinquent activity is first investigated for male students, then separately for female students. The variable sex has been treated as a control variable, its effect being removed. In short, it is a variable whose effect must be neutralized or controlled.

Ways of Handling a Control Variable

  • One way of handling a control variable to eliminate its effect is to hold it constant. Another is to let it vary, but perform analysis on subsets of the data, within which the control variable does not vary, as in the delinquency case above. This is possible if the control variable is discrete.

    If the control variable is continuous and must vary, then an analysis of covariance (ANCOVA) can remove the variable's effect.

    If bias can be removed by better randomizing the populations, this would be preferable to compensating for such bias mathematically.

A Distinction

  • A control variable, in its common usage, is not a controlled variable, which is an independent variable. An independent variable is varied to see what the resulting dependent variable is. A control variable is the contrary, a parameter whose effect needs elimination, not measuring.

In Computer Science

  • Some fields use the term "control variable" with slightly different meaning. For example, in computer science, a control variable is a program variable used to regulate the flow of the program. For example, it could control how many times a loop is executed. Another could measure the amount of data input, and not allow the program to proceed until a query for more information is satisfied.

In Control Theory

  • Control theory is an application of mathematics to engineering systems. In control theory, the control variable is also called the choice variable. Control theory is a broad field, and so is its concept of control variables. It can be a variable whose effect is to be eliminated, or instead a variable to be changed as deemed necessary by observation of the system output.

    For example, if the system is a car, throttle position (control variable) is varied based on feedback from the engine torque (output). If the system is a chemistry project, the variables whose effects need removal may be numerous, such as catalyst concentration, temperature, partial pressures and pH.


  • If the control variable is continuous, analysis of covariance could remove its effect. By accounting for a shared variable between populations, and fitting the populations separately, the effect of the shared variable can be removed.

    For example, three groups may receive three treatments. Fitting each separately as control variable "age" varies turns a hypothesis test into a comparison of linear fits, instead of the usual comparison of survival averages, and removes the effect of age differences between the three groups.


Promoted By Zergnet



You May Also Like

Related Searches

Check It Out

How to Build and Grow a Salad Garden On Your Balcony

Is DIY in your DNA? Become part of our maker community.
Submit Your Work!