Definition · Plain-language
Control variable
A control variable is any factor that a researcher deliberately keeps constant throughout a study, so it cannot influence the results or obscure the effect being measured.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
Holding factors constant
A control variable is anything the researcher keeps the same for every condition so that it cannot vary with, and therefore cannot explain, the outcome. If a study tests how fertiliser type affects plant growth, factors such as light, temperature, water, soil, and pot size are control variables: kept identical across all plants. By fixing them, the researcher ensures that any difference in growth can be attributed to the fertiliser (the independent variable) rather than to some other influence. Controlling variables is one of the main ways a study reduces alternative explanations for its results.
Control variables, confounders and the control group
These three terms are easily confused but are distinct. A control variable is held constant by design. A confounding variable is an extraneous factor that was not controlled and that influences both the independent and dependent variables, biasing the result — in effect, a control variable that got away. The control group is something different again: the group in an experiment that does not receive the treatment, providing a baseline for comparison. A well-designed study identifies likely confounders in advance and converts them into control variables.
Why control variables matter
Control variables underpin internal validity — the confidence that the independent variable, and not some other factor, produced the observed effect. Methods of control include holding conditions physically constant, standardising procedures, restriction (studying only one level of a variable), matching participants, randomisation, and statistical control during analysis. Reporting which variables were controlled, and how, is part of a transparent and reproducible methods section, in keeping with reporting standards that ask authors to describe how potential sources of bias were addressed.
Key facts
At a glance
- Definition: a variable deliberately held constant so it cannot affect the outcome
- Purpose: isolate the effect of the independent variable on the dependent variable
- Supports: internal validity and reproducibility
- Not the same as: the control group (a group, not a held-constant factor)
- If uncontrolled: it can become a confounding variable and bias results
- How to control: hold constant, standardise, restrict, match, randomise, or adjust statistically
Common misconceptions
What people often get wrong
Often heard: A control variable and a control group are the same thing.
Actually: They are different. A control variable is a factor kept constant across all conditions; a control group is the set of participants who do not receive the treatment, used as a baseline for comparison.
Often heard: Control variables are the variables you are trying to measure.
Actually: No — the variables of interest are the independent and dependent variables. Control variables are the background factors you hold constant precisely so they do not interfere with measuring that relationship.
Often heard: If you control some variables, you have eliminated all bias.
Actually: Not necessarily. You can only control variables you have anticipated; unmeasured or unknown factors may still confound results, which is why randomisation and careful design complement explicit controls.
