P-Value Interactive Learning Games

P-Value Interactive Learning Games

P-Value Interactive Learning Games

Learn statistical concepts through interactive scenarios

The 10‑second idea

A p‑value is the chance of getting results at least as extreme as yours if there were truly no real effect (the null). Smaller p = harder to blame on luck.

Always read p‑values with effect size and a 95% CI.

Quick glossary

  • Null hypothesis: no real difference/effect
  • Effect size: how big the difference is
  • 95% CI: plausible range for the true effect

Try it yourself

n = 40
Δ = -0.8
SD = 1.2
Iterations = 2000

We simulate two groups, compute the observed difference, build the null by shuffling labels, then compute a two‑sided p‑value.

Null world vs. your result

Null distribution Observed difference
Observed difference
—
Approx. 95% CI
—
Two‑sided p‑value
—
Quick read
—

Copy‑ready reporting sentence

—

Story: The 0.05 Line

Scene 1 — Friday Formulary Meeting

Riverside Wound Center debates HydroX. Vendor slide: “Time‑to‑closure improved by 0.4 days (95% CI 0.1–0.7), p=0.04.” Cost: +$58/patient. The committee approves based on p<0.05. Six months later: outcomes unchanged; spend +$74,000; nurses note no real‑world benefit.

Takeaway: statistically significant but trivial effect → costly decision.

Scene 2 — Same Team, Different Decision

OxyPatch pilot: closure at day 28 is 56% vs 40% (risk diff +16%, 95% CI −1% to +33%), p=0.07, +$8/patient. Rather than dismissing it, the team runs a pragmatic ward trial. Result: +12% closures (95% CI +3% to +21%), p=0.02, fewer painful changes, and net savings. OxyPatch is added with clear use criteria.

Takeaway: p≈0.07 ≠ “no effect.” Weigh effect size, CI, patient impact, and costs; gather local data when promising.

Cartoon placeholder

Drop your image here (replace the src of the <img> below) — or keep the SVG sketch.

0.05 "Does anyone see me?" Effect Size = 0.4 days AUDIT Costs ↑ Patient outcomes ↔ Effect size • 95% CI • p=0.07 Pragmatic trial → +12% closures

🎯 P‑value Games Arena

Score: 0 / 5

Game 1 — Guess the Evidence

Trial shows p=0.001 and mean difference = 0.2 days (95% CI 0.1–0.3). What does this mean?

Game 2 — Keep or Reject?

New burn dressing: −1.5 days (95% CI 0.2–2.8), p=0.06, cost +$4/patient. Adopt?

Game 3 — True or False?

1️⃣ A p‑value tells you how likely the data are if the null is true.

2️⃣ A p‑value tells you the probability the null is true.

3️⃣ p=0.05 means "about 1 in 20 by luck."

4️⃣ A low p‑value always means clinical importance.

Game 4 — Match the p‑Value

A. "We're quite sure it's real, but effect is small."

B. "This result could easily be by chance."

Game 5 — Fix the Report

Which version correctly reports results?

Your Results

Final Score: 0 / 5

Game 1: Coin Toss Challenge

Test if a coin is fair by flipping it multiple times and calculating the p-value for your results.

Game 2: Treatment Effect Simulator

Compare treatment and control groups to see if the observed difference is statistically significant.

Game 3: P-Value & Confidence Intervals

Explore the relationship between p-values and confidence intervals by adjusting sample size and effect size.

Coin Toss Challenge

You suspect a coin might be biased. Flip the coin multiple times and calculate the p-value to test the null hypothesis that the coin is fair (50% heads, 50% tails).

?

Results

Total Flips: 0

Heads: 0

Tails: 0

Proportion of Heads: 0

P-Value: -

Conclusion: The p-value represents the probability of seeing results this extreme if the coin were fair.

Treatment Effect Simulator

You're testing a new drug. Compare recovery rates between treatment and control groups to determine if the drug has a statistically significant effect.

Treatment Group (Received Drug)
Recovery Rate: 70%
Control Group (Received Placebo)
Recovery Rate: 50%

Statistical Analysis

Observed Difference
0%
P-Value
-

Interpretation: A p-value less than 0.05 suggests the difference is unlikely due to chance alone.

P-Value & Confidence Intervals

Adjust sample size and effect size to see how they influence p-values and confidence intervals.

Sample Size 100
Effect Size 0.5

Results

P-Value
0.05
95% Confidence Interval
[0.4, 0.6]

As sample size increases, confidence intervals narrow and p-values become more precise.