Controlled Experiment
A controlled experiment compares an experimental group with a control group that differs only in the one variable being tested.
Definition
A controlled experiment is a fair test: two setups are identical except for one thing, so any difference in the results must come from that one thing. It compares an experimental group, which receives the change, against a control group, which does not, while every other factor (the control variables) is held the same for both. The design isolates a cause by making the independent variable the only systematic difference between conditions, and it is judged by how confidently the observed effect can be attributed to that manipulation alone. Where possible, random assignment spreads unknown influences evenly across groups.
Example
Two identical pots of bean seeds get the same soil, light, and shelf position; one gets fertilizer and one does not, so any height difference is due to the fertilizer. Testing salt and freezing, a student puts $100\ \text{mL}$ of plain water and $100\ \text{mL}$ with $10\ \text{g}$ of salt in identical cups in the same freezer; the plain water freezes at $0^\circ\text{C}$ and the salt water near $-6^\circ\text{C}$. A stricter fertilizer test grows 12 seedlings per condition from one seed batch, randomly assigned, rotated daily on the same shelf, and watered by measured volume; after 21 days the fertilized mean is $14.6\ \text{cm}$ versus $10.9\ \text{cm}$, with too little overlap to blame chance.
Key Insight
The pot without fertilizer is just as important as the pot with it, because "taller" means nothing without a comparison. "Controlled" refers to the variables you hold steady, not to how neatly you work; a messy but properly controlled experiment beats a tidy one with two variables changing at once. The control group is a claim in itself, "here is what happens without the treatment," and if it surprises you the experiment has already taught you something.