Controlled variables
Open the Fair-test control room room
This article opens a matching guided lab. Follow the on-screen prompts, make a prediction, and connect the model result back to this topic.
01 // Why they matter
A controlled variable is anything besides the independent variable that could affect the dependent variable, so you deliberately hold it constant across every trial. Controlled variables are what make a test fair: if only one thing is different between your trial groups, then any difference in the result can be blamed on that one thing.
Leaving a variable uncontrolled creates a confound — a second explanation for your result that you can no longer rule out. If two plant pots get different amounts of fertilizer but also sit in different windows, you cannot tell whether fertilizer or sunlight caused a height difference.
02 // Worked example: osmosis egg lab
In the osmosis egg lab, the independent variable is the soaking liquid (water or corn syrup). Everything else has to match between the two eggs.
| Starting egg size | similar mass before soaking |
| Volume of liquid | 250 mL for every egg |
| Temperature | room temperature for both |
| Soak time | 24 hours for both |
Formula A controlled variable is anything held the same on purpose so it cannot explain the result.
03 // When a variable is hard to control
- You cannot control every possible variable. List the ones most likely to affect your specific dependent variable, and control those first.
- If a variable truly cannot be controlled (like outdoor temperature in a field study), record it instead, so you can check afterward whether it lines up with an odd result.
- A control group is a different tool from a controlled variable — a control group is a whole trial that gets no treatment, used for comparison. See control group vs experimental group.
04 // Choose controls that matter to the outcome
A controlled variable is a condition kept the same across the compared trials because it could influence the dependent measurement. In a plant-light test, pot size, soil type, water amount, plant species, and measurement schedule may matter. The list should come from the mechanism being studied, not from a requirement to keep every imaginable detail identical. Focus on factors with a plausible path to changing the outcome.
Write down how each condition will be controlled. 'Same water' is not specific enough if students use different amounts; '20 mL every morning, measured with the same cylinder' is reproducible. Some conditions cannot be perfectly controlled, such as room temperature changing through the day. Record them and decide whether they are likely to confound the comparison or need to be measured and included in the interpretation.
05 // Recognize a confound when two factors change together
A confound is a second factor that offers an alternative explanation for the observed result. If one plant receives more light and more water than another, the experiment cannot separate the effects of light from water. The result may still be real, but it does not answer the original question cleanly. Controlled conditions reduce these alternative explanations by making the planned comparison more specific.
Use a preparation checklist, labels, and a consistent schedule to prevent accidental changes. If a spill, delay, or equipment failure changes a condition, record it at the time. Do not silently 'correct' only one sample after the fact. The class can decide whether to repeat the trial, analyze it separately, or keep it with a limitation, depending on how much the event affected the data.
06 // Controls are not the same as a control group
Controlled variables are individual conditions held steady, such as temperature or volume. A control group is a comparison group that does not receive the treatment or receives a standard treatment. For example, in a plant-fertilizer experiment, keeping pot size constant is controlling a variable; plants receiving no fertilizer may form the control group. A study may have many controlled variables and one control group at the same time.
Some fair tests compare two treatment levels without a separate no-treatment control if the research question is a direct comparison. Explain why the comparison is suitable. If the control group's handling differs from the experimental group, the treatment is no longer the only systematic difference. Match routines such as measurement, time, and environment whenever possible.
07 // State the limit of control
Perfect control is rare outside simple classroom models. Investigators may not be able to control weather, individual biology, or every difference between samples. Random assignment, replication, blinding, or statistical adjustment can reduce some biases, but each method has limits. A clear report names important uncontrolled factors instead of implying that the experiment isolated one cause with absolute certainty.
Before beginning, ask a partner to find one plausible alternative explanation for the predicted result. Then decide whether a controlled variable, measurement, or design change can address it. This quick review often catches overlooked factors such as starting size, position under a lamp, order of testing, or instrument calibration. A well-designed experiment does not eliminate all uncertainty; it makes the remaining uncertainty visible.
08 // Use a control checklist during every trial
Turn each important controlled condition into a visible check: measured liquid volume, same container, same timing rule, same starting sample range, and the same instrument. A checklist makes repeated trials more consistent than relying on memory. Assign one person to prepare and another to verify when a group has several tasks, then rotate roles if time allows.
If a condition changes unexpectedly, record which sample and trial were affected. Decide whether to repeat or retain the observation based on a rule that applies to all groups, not on whether the value supports the prediction. This preserves evidence while making the impact of imperfect control available for discussion.
09 // Questions people ask
What is the difference between a controlled variable and a control group?
A controlled variable is a condition (like temperature or soak time) that is held the same across every trial. A control group is an entire trial or group that receives no treatment at all, used as a baseline to compare against.
How many controlled variables should an experiment have?
There is no fixed number — list every factor besides your independent variable that could plausibly change your dependent variable, and control as many of those as you reasonably can.
What happens if I forget to control a variable?
Your result has a confound: a second possible explanation you cannot rule out. The experiment can still run, but the conclusion becomes weaker, because you cannot say for certain which variable caused the change.