Skip to content
FiftyCalc Free calculators for all 50 states

Normal Approximation Calculator

Test cases replayed : 3 of 3 match · Methodology · No outside source listed

Enter the number of trials, the probability of success and a number of successes to approximate the binomial distribution with a normal curve.

Result

Sample figures

P(X ≤ k) ≈ Φ((k + 0.5 − μ) ÷ σ)—

Mean μ = n p —

  • Standard deviation σ = √(n p (1 − p))—
  • P(X ≥ k) ≈ 1 − Φ((k − 0.5 − μ) ÷ σ)—
  • P(X = k) ≈ Φ((k + 0.5 − μ) ÷ σ) − Φ((k − 0.5 − μ) ÷ σ)—
  • Rule of thumb met?—

Estimate for planning only, not tax, legal, financial or medical advice. This calculator uses the method explained on this page.

Normal Approximation Calculator

Number of trials (n): enter 1 or more. Prefilled example: 100.

Probability of success (0 to 1): enter a number from 0 to 1. Prefilled example: 0.5.

Number of successes (k): enter 0 or more. Prefilled example: 55.

How to use this calculator

Fill in the fields in this order; each step takes you to its field.

  1. Number of trials (n): enter 1 or more.
  2. Probability of success (0 to 1): enter a number from 0 to 1.
  3. Number of successes (k): enter 0 or more.
  4. Read the result: P(X ≤ k) ≈ Φ((k + 0.5 − μ) ÷ σ); the panel also shows Mean μ = n p, Standard deviation σ = √(n p (1 − p)), P(X ≥ k) ≈ 1 − Φ((k − 0.5 − μ) ÷ σ), P(X = k) ≈ Φ((k + 0.5 − μ) ÷ σ) − Φ((k − 0.5 − μ) ÷ σ) and Rule of thumb met?.

How this calculator works

A binomial count X with n trials and success probability p is approximated by a normal distribution with μ = n p and σ = √(n p (1 − p)).

The continuity correction adds or subtracts 0.5 because the binomial is discrete: P(X ≤ k) ≈ Φ((k + 0.5 − μ) ÷ σ) and P(X ≥ k) ≈ 1 − Φ((k − 0.5 − μ) ÷ σ).

The approximation is usually accepted when n p and n (1 − p) are both at least 10.

The formula

  1. μ=n×p
  2. Sd=n×p×(1−p)
  3. Le=normcdf⁡(k+0.5−μSd)
Symbols used in the formula
SymbolMeaning
nNumber of trials
pProbability of success (0 to 1)
kNumber of successes
μMean μ = n p
SdStandard deviation σ = √(n p (1 − p))
LeP(X ≤ k) ≈ Φ((k + 0.5 − μ) ÷ σ)
normcdf(x, m, s)the probability that a Gaussian variable with mean m and standard deviation s is at most x (mean zero and standard deviation one without m and s).

Worked example

Number of trials (n)100
Probability of success (0 to 1)0.5
Number of successes (k)55
Mean μ = n p50
Standard deviation σ = √(n p (1 − p))5
P(X ≥ k) ≈ 1 − Φ((k − 0.5 − μ) ÷ σ)0.1841
P(X = k) ≈ Φ((k + 0.5 − μ) ÷ σ) − Φ((k − 0.5 − μ) ÷ σ)0.0484
Rule of thumb met?Yes: n p and n (1 − p) are both at least 10
  1. μ=100×0.5=50
  2. Sd=100×0.5×(1−0.5)=5
  3. Le=normcdf⁡(55+0.5−505)=0.864334

P(X ≤ k) ≈ Φ((k + 0.5 − μ) ÷ σ) 0.8643

Each step uses the unrounded value of the step before it; figures on this page are rounded for display.

Frequently asked questions

What is P(X ≤ 55) for 100 fair coin flips?

μ = 50, σ = 5, so P ≈ Φ(1.1) = 0.8643; the exact binomial value is 0.8644.

Why use the continuity correction?

Each whole count k covers the interval from k − 0.5 to k + 0.5 on the continuous normal curve; without it the approximation is noticeably worse.

Questions answered

Related calculators