The continuity correction factor is used in statistics. A common type of distribution in statistics is a normal distribution, sometimes called a bell curve. It's useful to fit data into a bell curve shape to make predictions. When you use a normal distribution to approximate a binomial distribution (such as flipping a coin), you use a continuity correction factor to account for the difference between the two distributions.
The continuity correction factor is a series of statements that tells you whether to add 0.5 or subtract 0.5 to the number in your sample.
If P(X=n) use P(n -- 0.5 < X < n + 0.5)
If P(X>n) use P(X > n + 0.5)
If P(X≤n) use P(X < n + 0.5)
If P (X<n) use P(X < n -- 0.5)
If P(X ≥ n) use P(X > n -- 0.5).
Which statement you use depends on which value you have for the probability and how many items are in your sample, n. You're usually given these numbers as part of a question in statistics. For example, in the statement,"62 percent of a sample of 500 12th graders actually attend school," the probability is 62 percent and the sample size is 500.