Counts first, formula second
Count successes describes a device that passes each of several independent checks with the same chance. X is the number of passes. You choose the calculation for an event, such as exactly 2 or at least 3, work it out, and then say whether the model still fits when the procedure changes.
Before any formula, write the counts the event keeps: exactly 2 keeps 2; at least 3 of 6 keeps 3, 4, 5 and 6. Both ends are in. Each kept count k then adds C(n, k) × p^k × (1 − p)^(n − k): the number of ways to place k passes among n checks, times the chance of one such way.
When the kept list is long, the counts left out are shorter: at least 3 of 6 is everything except 0, 1 and 2, so it is 1 minus their sum. Check the boundary in that form too. The 3 belongs to the event, so it stays out of the part you subtract.
1List the counts the event keeps, both ends included.
2Add C(n, k) × p^k × (1 − p)^(n − k) for each kept count.
3Or subtract the counts left out from 1.
Exactly 2 of 4
Four checks, each passed with probability 75%. Exactly 2 passes keeps one count, 2.
There are C(4, 2) = 6 ways to place two passes among four checks, and each way has the chance 0.75^2 × 0.25^2. So P(X = 2) = 6 × 0.5625 × 0.0625 = 0.2109375, about 0.2109.
At least 3 of 6
Six checks, each passed with probability 75%. The event is that X is at least 3.
Try it
Which calculation gives P(X ≥ 3)?
- 1 − (Σ[k=0…2] C(6, k) × 0.75^k × 0.25^(6−k))
- 1 − (Σ[k=0…3] C(6, k) × 0.75^k × 0.25^(6−k))
Show the answer
1 − (Σ[k=0…2] C(6, k) × 0.75^k × 0.25^(6−k))
Right. At least 3 keeps 3, 4, 5 and 6, so the counts left out are 0, 1 and 2. Their sum taken from 1 is about 0.9624.
Enable JavaScript to answer interactively and save your place.
Practice Count successes