Mathwords logoMathwords

Continuity Correction — Definition, Formula & Example

Continuity correction is the practice of adding or subtracting 0.5 to a whole-number boundary when you use a smooth normal curve to estimate probabilities for a count. It treats each whole number kk as the interval from k−0.5k-0.5 to k+0.5k+0.5, so the area you measure matches the bar you meant to include.

Let XX be an integer-valued random variable, typically X∼Binomial(n,p)X \sim \text{Binomial}(n, p), approximated by a continuous random variable Y∼N(μ,σ2)Y \sim N(\mu, \sigma^2) with μ=np\mu = np and σ=np(1−p)\sigma = \sqrt{np(1-p)}. A continuity correction replaces each event about XX with an event about YY whose boundary is shifted by 12\tfrac{1}{2}. The shift includes or excludes the full unit interval [k−12, k+12][k - \tfrac{1}{2},\, k + \tfrac{1}{2}] that represents the value kk.

Key Formula

P(X≤k)≈P(Y<k+0.5),P(X≥k)≈P(Y>k−0.5),P(X=k)≈P(k−0.5<Y<k+0.5)P(X \le k) \approx P(Y < k + 0.5), \quad P(X \ge k) \approx P(Y > k - 0.5), \quad P(X = k) \approx P(k - 0.5 < Y < k + 0.5)
Where:
  • XX = The discrete random variable, such as a binomial count
  • YY = The approximating normal random variable with the same mean and standard deviation as X
  • kk = An integer boundary value of X

Worked Example

Problem: You flip a fair coin 100 times. Use the normal approximation with continuity correction to estimate the probability of getting at most 45 heads.
Step 1: Let XX be the number of heads, so X∼Binomial(100,0.5)X \sim \text{Binomial}(100, 0.5). Find the mean and standard deviation. Both np=50np = 50 and n(1−p)=50n(1-p) = 50 are at least 10, so the normal approximation is reasonable.
μ=np=100(0.5)=50,σ=np(1−p)=25=5\mu = np = 100(0.5) = 50, \qquad \sigma = \sqrt{np(1-p)} = \sqrt{25} = 5
Step 2: Apply the correction. "At most 45" includes the whole bar for 45, which reaches up to 45.5.
P(X≤45)≈P(Y<45.5)P(X \le 45) \approx P(Y < 45.5)
Step 3: Standardize the boundary and read the area from a standard normal table.
z=45.5−505=−0.9,P(Z<−0.9)≈0.1841z = \frac{45.5 - 50}{5} = -0.9, \qquad P(Z < -0.9) \approx 0.1841
Answer: P(X≤45)≈0.1841P(X \le 45) \approx 0.1841. The exact binomial probability is also about 0.1841. Without the correction you would get z=−1z = -1 and a probability of about 0.1587, which is noticeably too low.

Why It Matters

In AP Statistics and introductory college statistics, you often estimate binomial probabilities with a normal curve. Without the correction, those estimates can be off by several percentage points. The same idea appears as Yates's correction in chi-square tests on 2×2 tables.

Common Mistakes

Mistake: Shifting strict inequalities the same way as non-strict ones, for example writing P(X<45)≈P(Y<45.5)P(X < 45) \approx P(Y < 45.5).
Correction: First rewrite the event using whole numbers: X<45X < 45 means X≤44X \le 44. Then apply the correction to get P(Y<44.5)P(Y < 44.5). Ask yourself whether the bar at the boundary value belongs in the event.

Related Terms