Start by checking the model
A count X has a binomial model when there is a fixed number n of trials, every trial has two defined outcomes, trials are independent, and the success probability p is constant. Then
X ~ Binomial(n, p)
and
P(X = k) = C(n,k)p^k(1−p)^(n−k).
Defining “success” clearly matters because it determines both X and p.
Translate words before calculating
Endpoint errors are among the easiest mistakes to prevent.
| Wording | Event |
|---|---|
| at least k | X ≥ k |
| more than k | X > k, or X ≥ k+1 |
| at most k | X ≤ k |
| fewer than k | X < k, or X ≤ k−1 |
| between a and b, inclusive | a ≤ X ≤ b |
Write the event first, and only then choose a calculator or sum.
Direct sums and complements
An upper tail can be calculated directly:
P(X ≥ k) = Σ[x=k to n] P(X=x).
It can also be calculated with the complement:
P(X ≥ k) = 1 − P(X ≤ k−1).
Notice the k−1. Subtracting P(X ≤ k) would incorrectly omit P(X=k) from the desired tail.
Similarly,
P(X > k) = 1 − P(X ≤ k).
Use whichever side requires fewer terms or matches the cumulative-distribution function available on your tool.
A fresh example
A device passes a calibration check with probability 0.80, independently of the other devices. Twelve devices are checked. Let X be the number that pass, so X ~ Binomial(12, 0.80).
To find the probability that at least 10 pass, write
P(X ≥ 10) = P(X=10) + P(X=11) + P(X=12).
The three terms are approximately 0.283468, 0.206158, and 0.068719, so
P(X ≥ 10) ≈ 0.558346.
The same result comes from 1 − P(X ≤ 9).
A reliable check routine
- Confirm the four binomial conditions.
- State what one success means.
- Define
Xas a count and recordnandp. - Translate the wording with the correct inclusive or exclusive endpoint.
- Calculate and verify that the result lies between 0 and 1.
A probability near a tail should also make qualitative sense relative to the mean np; this is a reasonableness check, not a substitute for the exact calculation.
Related question
Apply this knowledge
Use the concept guide to understand the reasoning, then return to the complete question and worked answer.
Assume That a Procedure Yields a Binomial Distribution: 50-Student Homework ProbabilitySources
These references support the core concepts and interpretation boundaries explained above.