Overview
Why experiments use comparison groups
A comparison group helps investigators determine whether an observed change is associated with the treatment or independent variable being tested.
Without a suitable comparison, it may be difficult to separate the treatment effect from natural variation, environmental changes, measurement differences, or other influences.
Group definitions
Control group vs experimental group
Control group
Provides the baseline
The control group normally does not receive the treatment or receives a standard reference condition.
Experimental group
Receives the treatment
The experimental group receives the condition, treatment, or independent-variable level being tested.
Experimental logic
The groups should differ mainly in one factor
Controlled comparisonExperimental group − control group = estimated treatment effect
This comparison is strongest when important conditions are kept consistent and the treatment is the main planned difference between groups.
Design process
Six steps for setting up comparison groups
Define the treatment
Identify the condition, substance, procedure, or variable level being tested.
Choose a comparison condition
Decide what untreated, standard, baseline, or reference condition will provide a meaningful comparison.
Assign the groups
Place subjects or samples into control and experimental groups using a fair method.
Keep other conditions consistent
Ensure that groups differ mainly in the treatment being investigated.
Measure both groups equally
Use the same instruments, timing, units, and procedures for every group.
Compare the outcomes
Evaluate whether the experimental group differs meaningfully from the control group.
Worked example
Testing fertilizer and plant growth
Research questionHow does fertilizer treatment affect the weekly height increase of bean plants?
- Control group: plants receiving no fertilizer
- Experimental group: plants receiving the fertilizer
- Dependent variable: weekly height increase
- Controlled variables: species, soil, water, light, pot size, temperature, and investigation duration
Chemistry example
Testing a catalyst
Control condition
Reaction without catalyst
The reaction is performed under the same conditions without adding the catalyst.
Experimental condition
Reaction with catalyst
The same reaction is performed with a defined amount of catalyst.
Control types
Positive and negative controls
Negative control
Expected to show no treatment effect
A negative control helps identify background changes, contamination, or effects unrelated to the treatment.
Positive control
Expected to produce a known response
A positive control helps confirm that the method, equipment, or detection system can produce the expected result.
Fair assignment
Reduce bias when forming groups
Groups should be as similar as practical before the treatment begins. Random assignment can reduce systematic differences between groups.
- Use comparable subjects, samples, or materials.
- Avoid assigning stronger or larger samples to one group intentionally.
- Use sufficient replication when possible.
- Apply identical measurement procedures.
- Record any unavoidable baseline differences.
Treatment levels
Experiments may use several experimental groups
Multiple experimental groups can receive different treatment levels, concentrations, durations, or conditions. This helps reveal whether the response changes gradually or reaches an optimum.
Example designControl: 0% fertilizer → Group 1: 1% → Group 2: 2% → Group 3: 3%
Design limits
Not every investigation uses a separate control group
Some investigations compare several naturally occurring conditions, measure a physical relationship, or use each sample as its own baseline.
The important requirement is a meaningful comparison that allows the research question to be answered.
Common mistakes
Comparison-group problems to avoid
- Changing more than the treatment between groups.
- Using different measurement methods for each group.
- Selecting groups that differ before treatment.
- Using too few samples or trials.
- Failing to define the control condition clearly.
- Assuming any difference must be caused by the treatment.
Related resources
Strengthen the experiment design
Identify the Independent, Dependent, and Controlled Variables, review the complete Steps of the Scientific Method, and organize the final report using the Lab Report Template.
Questions and answers
Control and experimental groups FAQ
What is a control group?
A control group provides a comparison condition and normally does not receive the treatment or independent variable being tested.
What is an experimental group?
An experimental group receives the treatment, condition, or level of the independent variable being investigated.
Does every experiment need a control group?
Not every investigation requires a separate control group, but a suitable comparison condition is often essential for interpreting cause and effect.
Can an experiment have more than one experimental group?
Yes. Multiple experimental groups may receive different treatment levels so the investigator can compare a range of effects.