This normal distribution calculator finds probabilities under the bell curve for any mean and standard deviation — below, above, between two values — and runs backwards: give it a probability, get the cutoff value.
Probabilities Are Areas: z First, Then Φ(z)
Every normal question reduces to standardizing (z = (x − µ)/σ — the job of our Z-Score Calculator) and reading the standard normal CDF Φ. This page wraps that pipeline with your own µ and σ, in both directions. Part of our Statistics Calculators collection.
Worked Examples (Verified)
- IQ-style defaults (µ=100, σ=15), x=115: z = 1.00 → P(X < 115) = 0.8413 — the classic "one sigma up = 84th percentile"
- P(X > 130) on the same curve: z = 2 → 0.0228 — the top 2.28%
- Between 85 and 115 (±1σ): 0.6827 — the empirical rule's first band, exact here
- Inverse: cutoff for the top 5% of N(100,15): z = 1.645 → x ≈ 124.7 — how "gifted program" thresholds get computed (usually rounded to 125 or 130 by policy, a human overlay on exact math)
When the Bell Curve Lies
- Incomes, city sizes, file sizes and most biological counts are right-skewed (log-normal-ish): normal arithmetic there manufactures impossible negative values and fantasy percentiles.
- With small samples, the t-distribution's heavier tails replace Φ — that's the T-Test Calculator's machinery.
- "Normal" in a problem statement may mean *assumed* normal — this calculator computes consequences of the assumption, not its truth; check skew before trusting tails.
- Central limit theorem comfort: *means* of samples normalize fast even when raw data don't — which is why sampling distributions legitimately use this page.
Related Tools
- Z-Score Calculator — the standardization step alone
- T-Test Calculator — means with unknown σ
- Statistics Calculators — the full collection
Frequently Asked Questions
How do you find a probability from a normal distribution?
Standardize the value (z = (x − µ)/σ), then read the standard normal cumulative probability Φ(z). For x=115 with µ=100, σ=15: z=1, Φ(1)=0.8413, so 84.13% of values fall below 115. This calculator does both steps with your µ and σ.
What does the inverse normal calculation do?
It runs the pipeline backwards: given a target probability (e.g. 0.95), it finds the z and then the x that leaves exactly that much area to its left. This is how percentile cutoffs, exam thresholds and specification limits are derived from a distribution.
What are the 68-95-99.7 rule numbers exactly?
68.27% within ±1σ, 95.45% within ±2σ, 99.73% within ±3σ. The classroom rounding (68/95/99.7) is safe; this calculator's outputs use the exact Φ values (2σ two-tailed exceedance is 4.55%, not exactly 5% — hypothesis tests use the exact figures from the P-Value Calculator).
Normal distribution vs standard normal distribution?
The standard normal is the special case µ=0, σ=1. Every other normal curve is a rescaling of it — which is why one Φ function serves all inputs: standardization maps any (µ, σ) problem onto the standard curve, and destandardizing maps answers back.
Can I use this for sample means?
Yes — enter the standard error (σ/√n) as the "standard deviation" and the population mean as µ. That's the sampling-distribution use behind confidence intervals and z-tests, justified by the central limit theorem for reasonable n.
Is this normal distribution calculator free?
Yes — free, instant, no sign-up.