- Big p = “Not Yet”
A large p-value doesn’t crown the null; it says today’s evidence can’t clear your risk line (alpha). Treat it as “not yet.” Tighten measures, strengthen design, or raise n. Then rerun.
- Small p, Small Claim
A tiny p-value rejects “no effect.” It doesn’t prove a big effect, a useful effect, or a true story. Pair p with effect size and context before you ship.
- Alpha Is a Business Choice
Alpha is your risk budget for false alarms. Alpha = .05 is a habit, not a law. Set it to match stakes, scrutiny, and harm.
- Pick Sides Before Data
One-sided or two-sided? Decide in advance and write it down. Switching after peeking turns science into theater.
- Power Fuels Progress
Underpowered studies waste time: likely null, unclear signal. Plan for power (effect size, alpha, variance, n) so a real effect can be seen.
- Don’t P-Hack, Precommit
Peeking, cherry-picking, and optional stopping shrink p-values and your credibility. Precommit analyses, stop rules, and outcomes.
- Significance Isn’t Significance
“Statistically significant” ≠ important. Ask: Does it change policy, practice, or price? If not, it’s trivia wearing a tux.
- Replicate or Reconsider
One result is a spark, not a sunrise. Independent replications, similar designs, converging evidence… that’s how claims grow up.
Share this post:
on Facebook
