Hypergeometric Distribution Table
The hypergeometric distribution models the probability of k successes in n draws from a finite population of size N containing K successes, without replacement.
Probability Formula
P(X = k) = [C(K,k) × C(N-K, n-k)] / C(N,n)
Where:
- N = population size
- K = number of successes in population
- n = number of draws (sample size)
- k = number of observed successes
- C(a,b) = binomial coefficient “a choose b”
Small Population Tables
N = 20, n = 5 (drawing 5 from population of 20)
| k | K=2 | K=3 | K=4 | K=5 | K=6 | K=7 | K=8 | K=10 |
|---|---|---|---|---|---|---|---|---|
| 0 | 0.5526 | 0.3991 | 0.2817 | 0.1938 | 0.1295 | 0.0839 | 0.0524 | 0.0181 |
| 1 | 0.3947 | 0.4420 | 0.4302 | 0.3874 | 0.3298 | 0.2687 | 0.2098 | 0.1150 |
| 2 | 0.0526 | 0.1458 | 0.2167 | 0.2664 | 0.2955 | 0.3040 | 0.2924 | 0.2384 |
| 3 | - | 0.0132 | 0.0630 | 0.1246 | 0.1795 | 0.2260 | 0.2587 | 0.3117 |
| 4 | - | - | 0.0083 | 0.0263 | 0.0571 | 0.0957 | 0.1388 | 0.2154 |
| 5 | - | - | - | 0.0016 | 0.0085 | 0.0217 | 0.0480 | 0.1015 |
N = 50, n = 10 (drawing 10 from population of 50)
| k | K=5 | K=10 | K=15 | K=20 | K=25 |
|---|---|---|---|---|---|
| 0 | 0.3106 | 0.0636 | 0.0095 | 0.0010 | 0.0001 |
| 1 | 0.4313 | 0.2211 | 0.0643 | 0.0130 | 0.0019 |
| 2 | 0.2098 | 0.2962 | 0.1668 | 0.0580 | 0.0137 |
| 3 | 0.0442 | 0.2404 | 0.2463 | 0.1444 | 0.0517 |
| 4 | 0.0041 | 0.1225 | 0.2300 | 0.2190 | 0.1182 |
| 5 | - | 0.0438 | 0.1570 | 0.2307 | 0.1787 |
| 6 | - | 0.0102 | 0.0764 | 0.1726 | 0.1916 |
| 7 | - | 0.0018 | 0.0275 | 0.0891 | 0.1453 |
| 8 | - | - | 0.0058 | 0.0289 | 0.0727 |
| 9 | - | - | 0.0008 | 0.0058 | 0.0218 |
| 10 | - | - | - | 0.0006 | 0.0043 |
Quality Control Application
Sampling from lot of N=100 items with D defectives
Sample size n = 10
| k defects | D=5 | D=10 | D=15 | D=20 |
|---|---|---|---|---|
| 0 | 0.5838 | 0.3305 | 0.1821 | 0.0951 |
| 1 | 0.3394 | 0.4081 | 0.3686 | 0.2954 |
| 2 | 0.0702 | 0.2015 | 0.2953 | 0.3427 |
| 3 | 0.0065 | 0.0477 | 0.1312 | 0.2114 |
| 4 | - | 0.0112 | 0.0354 | 0.0898 |
| 5 | - | 0.0010 | 0.0061 | 0.0255 |
Sample size n = 20
| k defects | D=5 | D=10 | D=15 | D=20 |
|---|---|---|---|---|
| 0 | 0.3193 | 0.0951 | 0.0243 | 0.0054 |
| 1 | 0.4200 | 0.2683 | 0.1167 | 0.0421 |
| 2 | 0.2075 | 0.2918 | 0.2157 | 0.1237 |
| 3 | 0.0482 | 0.2067 | 0.2525 | 0.2171 |
| 4 | 0.0049 | 0.0981 | 0.2027 | 0.2544 |
| 5 | - | 0.0319 | 0.1171 | 0.2073 |
| 6 | - | 0.0071 | 0.0486 | 0.1189 |
Cumulative Probabilities P(X ≤ k)
N = 50, K = 10, n = 10
| k | P(X = k) | P(X ≤ k) | P(X ≥ k) |
|---|---|---|---|
| 0 | 0.0636 | 0.0636 | 1.0000 |
| 1 | 0.2211 | 0.2847 | 0.9364 |
| 2 | 0.2962 | 0.5809 | 0.7153 |
| 3 | 0.2404 | 0.8213 | 0.4191 |
| 4 | 0.1225 | 0.9438 | 0.1787 |
| 5 | 0.0438 | 0.9876 | 0.0562 |
| 6 | 0.0102 | 0.9978 | 0.0124 |
| 7 | 0.0018 | 0.9996 | 0.0022 |
| 8 | 0.0004 | 1.0000 | 0.0004 |
Mean and Variance
| Parameter | Formula |
|---|---|
| Mean (μ) | n × K/N |
| Variance (σ²) | n × (K/N) × (1 - K/N) × (N-n)/(N-1) |
| Standard Dev (σ) | √Variance |
Finite Population Correction
The factor (N-n)/(N-1) is the finite population correction, which reduces variance when sampling a large fraction of the population.
Comparison: Hypergeometric vs Binomial
| Aspect | Hypergeometric | Binomial |
|---|---|---|
| Sampling | Without replacement | With replacement |
| Population | Finite | Infinite |
| Probability | Changes each draw | Constant |
| When to use | n/N > 0.05 | n/N < 0.05 |
Rule of thumb: Use binomial approximation when n < 0.05N (sample is less than 5% of population)
Example Problems
Example 1: Quality Control
Problem: A lot of 50 items contains 10 defectives. If 5 items are sampled, what’s the probability of exactly 2 defectives?
Solution:
- N = 50, K = 10, n = 5, k = 2
- P(X = 2) = [C(10,2) × C(40,3)] / C(50,5)
- P(X = 2) = 0.2098
Example 2: Card Drawing
Problem: In a deck of 52 cards (13 hearts), what’s P(exactly 3 hearts in 5 cards)?
Solution:
- N = 52, K = 13, n = 5, k = 3
- P(X = 3) = 0.0815
Example 3: Acceptance Sampling
Problem: A lot of 100 items has 15 defectives. Accept if sample of 10 has ≤ 1 defective. What’s P(accept)?
Solution:
- P(X ≤ 1) = P(X=0) + P(X=1)
- P(X ≤ 1) = 0.1821 + 0.3686 = 0.5507
Lottery/Selection Problems
Committee Selection
Problem: 8 men and 6 women available. Select committee of 5. P(exactly 2 women)?
Solution:
- N = 14, K = 6 (women), n = 5, k = 2
- P(X = 2) = [C(6,2) × C(8,3)] / C(14,5)
- P(X = 2) = 0.4196
Lottery Numbers
Problem: 49 numbers, 6 winning. Buy ticket with 6 numbers. P(matching exactly 3)?
Solution:
- N = 49, K = 6, n = 6, k = 3
- P(X = 3) = 0.0177 (about 1 in 57)
When to Use Hypergeometric
✓ Sampling WITHOUT replacement
✓ Finite population
✓ Sample size > 5% of population
✓ Quality control sampling
✓ Card/lottery problems
✓ Committee/selection problems
Related Distributions
| Situation | Use |
|---|---|
| Sample with replacement | Binomial |
| Large population (n/N < 0.05) | Binomial approximation |
| Multiple categories | Multivariate hypergeometric |
Related Resources
- Hypergeometric Lesson - Complete guide
- Binomial Table - With replacement
- Probability Calculator - Calculate probabilities
- Combinations Calculator - Calculate C(n,k)