Mark observation separately
Write what changed on a graph before deciding why it changed. Evidence comes before interpretation.
Field notes
What you record may be the only way home.
Every 4physics field physicist carries the same rule: a notebook is not a place to hide uncertainty. It is where you preserve what was seen, what was changed, and what your model still cannot explain.
Control the unknown
Change one condition at a time. A dramatic result is not useful if its cause cannot be traced.
Give a conclusion its boundary
State the condition that makes your conclusion work, then deliberately search for the point where it fails.
Leave uncertainty visible
Separate a measurement, a model prediction, and a question that remains unanswered.