The independent variable is what you change; the dependent is what you measure; controlled variables are held constant.
Step 1: Something to Notice
Watch first. The explanation comes later.
A trial finds that patients given a new drug improve. A second trial, in which some patients unknowingly receive a sugar pill, finds the same improvement in both groups.
What did the first trial actually measure?
Step 2: Find Out
Identify the single variable being tested and everything that must be held constant for the result to be attributable.
Step 1 — Predict
A trial with no control group finds improvement. What can be concluded?
Choose what you think will happen. You cannot see the experiment until you do — guessing first is what makes it worth watching.
Step 3: So Here Is Why
Now the explanation, after you have seen it happen.
The control group
A group treated identically except for the variable being tested. Without it there is nothing to compare against.
Control group A comparison group identical except for the tested variable.
Confounding variables
Anything that differs between groups alongside the treatment. A confounder makes a result uninterpretable rather than merely weaker.
Confounding variable An uncontrolled difference that could explain the result instead.
Blinding
Expectation changes what people report and what observers record, so neither should know the allocation.
Replication and sample size
One organism proves nothing and one study rarely settles anything. Repetition separates a real effect from chance.
Controls are what make a result mean anything
Without a group treated identically except for the variable under test, any difference could be caused by anything. The control is not a formality — it is what converts an observation into evidence.
Biological variability demands replication
Two organisms treated identically respond differently. Sample sizes in biology have to be large enough that individual variation does not masquerade as an effect, which is why n = 1 proves nothing.
Step 4: A Common Mistake
Lots of people think
“If the treatment group improved, the treatment worked.”
Step 5: Why Biology Is Harder to Control Than Physics
The same idea somewhere new.
Organisms differ from each other, respond to being studied, and cannot usually be reset and run again. That is why biology leans so heavily on large samples, randomization and statistics — not because biologists are less rigorous, but because the noise is genuinely larger and cannot be engineered away.
Step 6: Think It Through
Practice makes it stick.
The Sunny Shelf
Problem 1 of 2
Treated plants grew taller and were also nearer the window. What can you conclude?
Both Groups Improved
Problem 2 of 2
Drug and placebo groups improve equally. What does that show?
Designing It Properly
1 of 5
What is a confounding variable?
2 of 5
Why blind an assessor?
3 of 5
Which threats remain even in a well-designed trial?
4 of 5
Why does biology rely so heavily on statistics?
5 of 5
What does a control group provide?
Step 7: Quick Check
Show what you know.
Question 1 of 1
The treatment group improved. What can you conclude without a control?
Step 8: Explain It in Writing
Claim, evidence, then reasoning.
The question
A drug trial without a control group reports improvement. Explain what it can and cannot establish, and design a study that would settle it.
Fill in all three boxes. The reasoning box is the one that matters most.
What You Found Out
- Change one variable, measure another, hold everything else constant.
- A control group provides the comparison that makes a causal claim possible.
- A confounding variable makes a result uninterpretable, not merely weaker.
- Blinding, randomization, sample size and replication address the threats that remain.