Teach & case
Help a near-peer reason through the idea — when you explain it, you understand it best — then read the real decision it comes from.
Help Hattie design the test
What you've shown you can teach
A real case · you decide
Hattie's plant experiment
Hattie — a student testing whether a new fertilizer makes plants grow taller for a science fair
Hattie wants to prove her fertilizer works. Her plan: give the fertilized plant more sun and more water too, "to really give it the best chance." After two weeks it's taller — proof the fertilizer works?
The fertilized plant got MORE fertilizer, MORE sun, AND more water than the control. It grew taller — but which of the three caused it? The design can't say.
A controlled experiment changes ONE thing (the fertilizer) and keeps everything else — sun, water, soil, pot, seed type — identical between the two plants. Then any difference is attributable to the one variable.
Hattie's plant got more fertilizer, sun, AND water, and grew taller. Does that prove the fertilizer worked?
To fix it, what should differ between the fertilized plant and the control?
Hattie re-runs it: same sun, same water, only the fertilizer differs — across five plants each. That's Hattie's whole thing: change one variable, hold the rest, and let the result actually mean something.