Independent and Dependent Variables, Explained With Examples

In experimental design, an independent variable is a variable that changes so that researchers can observe downstream effects of that change. In some cases, experimental variables are intentionally manipulated variables that researchers set to different values as part of a study. In other cases, researchers cannot directly manipulate independent variables, but they still study how these variables affect the outcome of the experiment.

Sometimes researchers use different terminology to describe independent variables. When researchers perform a linear regression, they often call independent variables "right-side variables," since they appear on the right side of a linear regression. They can also be called predictor variables (as they help researchers predict the outcome of an experiment) or explanatory variables (since they help explain final results).

Independent variables fall into two categories: experimental variables and subject variables.

1. 1. Experimental variables: An experimental independent variable is one that researchers can manipulate to test downstream outcomes. Experimental variables are also called controlled variables due to researchers' ability to control them.
2. 2. Subject variables: Subject variables cannot be controlled by researchers. Despite this, they are still used as experimental inputs because they can help answer research questions. For instance, if researchers are examining high school standardized test scores from different regions, they cannot manipulate which region a student comes from. Yet they can still use those regional backgrounds to group test subjects at the outset of their study.

To understand how independent variables manifest in science, consider the following hypothetical studies.

• Plant growth study: Imagine researchers want to study the effects of fertilizer dosage on plant growth. The amount of fertilizer they give each plant would be an independent experimental variable. It will presumably affect a downstream phenomenon, such as how much each plant grows.
• Math test results: Imagine researchers want to analyze the standardized math test scores of students who took honors-level algebra and those who took standard algebra. The students' class selections would be independent subject variables. The researchers cannot manipulate which class each student had taken, but they can still study whether this independent variable causes changes in the students' standardized test scores.

In experimental design, a dependent variable is a responding variable that changes based upon input values from an independent variable. Some researchers call a dependent variable an “outcome variable” or “response variable” because the value of the dependent variable is intrinsically linked to upstream changes in the independent variable.

To properly follow the scientific method, a study's research design must only change one variable at a time—leaving all other aspects of an experiment intact. This lets researchers specifically observe how the change in one variable affects other measurements in the experiment. This means that researchers do not directly manipulate dependent variables. Still, they expect these variables to fluctuate as a result of the independent variable changes.

To illustrate dependent variables in real life, consider the following two examples:

• Plant growth study: In a hypothetical plant growth study, the independent experimental variable is the amount of fertilizer fed to each plant sample. The dependent variable in such an experiment would be the recorded growth of each specimen. If all other aspects of the plant treatment remain the same—the amount of water, size of the container, amount of sunlight, amount of time growing—it is reasonable to assume that the recorded plant growth directly fluctuates as a result of the independent variable: fertilizer.
• Math test analysis: In this study, a student's coursework background is the independent variable. Researchers specifically choose students to study based upon what kind of algebra class they had taken—regular or honors. The dependent variable would then be the standardized test scores that each cohort of students ended up earning. The researchers cannot manipulate these test scores; they can only observe them after having selected the input group.

Use the following rubric to spot the difference between independent and dependent variables.

• Manipulated or observed: Ask yourself if the variable in question can be manipulated or hand-selected by researchers, or whether it is merely observed and measured in the course of the experiment. Manipulated and selected variables are always independent variables. Observed and recorded variables are dependent variables. (Although subject variables cannot be controlled by researchers, they are still treated as independent variables because they affect the dependent variables.)
• Graphing: When graphed on an X-Y coordinate plane, independent variables tend to be inputs that appear on the X-axis (horizontal axis). Dependent variables are output variables, and they appear on the Y-axis (vertical axis).
• A third type of variable: Sometimes researchers encounter a third type of variable that is neither independent nor dependent. These are called confounding variables. They affect experimental results in unseen ways. In a sense they are unforeseen independent variables that researchers may not have anticipated. So when classifying variables, it isn't always a binary choice. Some variables don't fit neatly into the category of independent or dependent variables, and they may be best described as confounding variables.

Get the MasterClass Annual Membership for exclusive access to video lessons taught by science luminaries, including Terence Tao, Bill Nye, Neil deGrasse Tyson, Chris Hadfield, Jane Goodall, and more.