Simple Random Sample Probability Calculation

Estimate the probability of drawing a specific number of successes in a simple random sample without replacement. This calculator uses the hypergeometric model, the standard probability framework for simple random sampling from a finite population.

Expert Guide to Simple Random Sample Probability Calculation

Simple random sampling is one of the most important ideas in statistics. It gives every possible sample of a fixed size the same chance of being selected from a finite population. Because of that property, simple random sampling is the foundation for many survey designs, audit procedures, clinical selection methods, educational experiments, and quality control workflows. When analysts want to know the probability of drawing a certain number of observations with a given trait, they are almost always working with the probability model behind sampling without replacement: the hypergeometric distribution.

This calculator is designed to help you compute probabilities for that exact situation. If a population contains a known number of successes and failures, and you select a sample without replacement, the number of successes observed in the sample does not follow a binomial distribution. Instead, it follows the hypergeometric distribution because each draw changes the composition of the remaining population. That dependency matters. It is the reason simple random sample probability calculations are slightly more involved than independent-trial calculations.

Core idea: If your population size is N, the number of successes in the population is K, your sample size is n, and the observed number of successes in the sample is k, then the probability of exactly k successes is based on how many favorable samples exist divided by how many total samples are possible.

When this calculator should be used

You should use a simple random sample probability calculation when all of the following are true:

• You are sampling from a finite, known population.
• The sample is drawn without replacement.
• You know or can estimate how many members of the population are classified as successes.
• You want the probability of seeing exactly, at least, or at most a certain number of successes in the sample.

Typical use cases include selecting defective parts from a production lot, drawing households from a small municipality, choosing files for an audit, testing a subset of medications from a batch, or sampling student records from a school district where a known count shares a particular characteristic.

The formula behind the calculator

The exact probability of observing k successes in a simple random sample is:

P(X = k) = [C(K, k) × C(N – K, n – k)] / C(N, n)

Here, C(a, b) means the number of combinations of a items taken b at a time. The numerator counts all samples with exactly k successes and n – k failures. The denominator counts all samples of size n from the population of size N.

This formula is elegant because it mirrors the logic of simple random sampling. If every sample of size n is equally likely, then probability becomes a counting problem. How many equally likely samples satisfy your condition? That count, divided by the total number of possible samples, gives the answer.

Understanding each input

1. Population size (N): The total number of units in the population. In a quality-control problem, this may be the total number of products in a batch.
2. Number of successes (K): The number of population members having the trait of interest. For example, how many products in the lot are defective.
3. Sample size (n): The number selected from the population. This is the actual size of the simple random sample.
4. Target successes (k): The number of observed successes in the sample that you want to evaluate.
5. Probability type: Whether you want exactly k, at least k, or at most k successes.

Why without replacement matters

If you were sampling with replacement, each draw would be independent and the binomial model would be appropriate. But simple random sampling from a finite population is almost always without replacement. Once an item is selected, it cannot be selected again, and the pool changes. That means probabilities shift on each draw. For example, if a batch contains 60 defective units among 500 total units, drawing one defective item slightly lowers the defective rate in the remaining pool. Drawing a nondefective unit slightly increases it. The hypergeometric model accounts for that changing composition exactly.

Worked example: quality control lot sampling

Suppose a manufacturer has a lot of 500 components. Inspection records show 60 are defective. If an auditor selects a simple random sample of 40 components, what is the probability of finding exactly 5 defective items?

Set the values as follows:

• N = 500
• K = 60
• n = 40
• k = 5

The expected number of defectives in the sample is n × (K/N), which equals 40 × (60/500) = 4.8. So seeing 5 defectives is very close to the center of the distribution. As a result, the exact probability tends to be relatively high compared with more extreme values like 0, 10, or 12 defectives.

Scenario	Population Size (N)	Successes in Population (K)	Sample Size (n)	Target (k)	Probability Focus
Manufacturing audit	500 parts	60 defectives	40 sampled	5 defectives	Exactly 5 defectives in sample
School records review	1,200 students	180 with program participation	75 selected	12 participants	At least 12 program participants
Inventory check	220 units	18 damaged	25 inspected	2 damaged	At most 2 damaged units

Expected value and standard deviation

A strong probability analysis does not stop at one probability. You should also understand the center and spread of the distribution. For a hypergeometric random variable:

• Expected value: E(X) = n(K/N)
• Variance: Var(X) = n(K/N)(1 – K/N)[(N – n)/(N – 1)]
• Standard deviation: the square root of the variance

The finite population correction factor, (N – n)/(N – 1), is what distinguishes this spread from the spread of a binomial model. As the sample becomes a larger fraction of the population, the correction becomes more important and the sampling variability decreases. In plain language, when you sample a large share of a population without replacement, there is less uncertainty left in what remains.

How to interpret exact, at least, and at most probabilities

These three output types serve different decision goals:

• Exactly k: Useful for textbook problems and precise event modeling.
• At least k: Common in threshold decisions, such as the chance of detecting at least a minimum number of defects.
• At most k: Useful for acceptance sampling and compliance checks, where too many observed failures would trigger concern.

For example, an auditor may ask, “What is the probability that our random review finds at least 12 noncompliant files?” That is a cumulative probability. A quality manager may ask, “What is the probability that we find at most 2 damaged units?” That is also cumulative, but on the lower tail of the distribution.

Realistic comparison table

The table below shows how changing the sample fraction can alter interpretation, even when the population success rate is similar. These are realistic finite-population examples frequently seen in operations, audit, and institutional research.

Case	Population Rate	Sample Fraction	Expected Successes	Why It Matters
Batch inspection: 60 defectives in 500, sample 40	12.0%	8.0%	4.8	Moderate sample fraction, finite population correction is noticeable but not dominant.
Program evaluation: 180 participants in 1,200, sample 75	15.0%	6.25%	11.25	Sample is small relative to population, so hypergeometric and binomial results may be close.
Inventory damage review: 18 damaged in 220, sample 25	8.18%	11.36%	2.05	Larger sample fraction makes the without-replacement model more important.

Common mistakes to avoid

1. Using the binomial distribution when sampling without replacement: This is the most common error. If the sample fraction is not negligible, the binomial approximation can be meaningfully off.
2. Entering a target value outside the feasible range: The number of successes in the sample cannot exceed the sample size or the number of successes in the population.
3. Ignoring finite population structure: In small or medium-sized populations, exact finite population methods are often preferable to large-sample approximations.
4. Confusing percentages and counts: The calculator needs actual counts for N, K, n, and k.
5. Forgetting that “at least” and “at most” are cumulative: These probabilities sum across multiple values, not just one value.

How the chart helps decision-making

The distribution chart provides more than visual appeal. It shows the full range of plausible success counts in the sample. If your selected k falls near the peak, the event is relatively common. If it falls in the far tail, the event is rare. This visual context is especially useful for executives, auditors, researchers, and students who need to explain why an outcome is ordinary or unusual under the assumptions of simple random sampling.

When to use approximations

In some settings, analysts approximate the hypergeometric distribution with a binomial or normal distribution. That can be reasonable when the population is very large relative to the sample and when the event rate is not too extreme. However, if your data come from a finite list, roster, lot, or batch, exact hypergeometric probabilities are usually better. Since modern browsers can compute them instantly, there is little reason to rely on rough approximations unless you are working analytically by hand.

Practical applications across industries

• Public administration: Randomly sampling claims, permits, or files from a finite administrative inventory.
• Healthcare operations: Reviewing charts from a known panel of patients to estimate the likelihood of finding cases with a specific condition.
• Manufacturing: Acceptance sampling from a finite production lot where the defect count is known or modeled.
• Education: Selecting student records from a finite school population to assess program participation or test accommodations.
• Research design: Teaching finite-population sampling theory using exact probabilities rather than asymptotic approximations.

Authoritative references for further study

If you want a deeper statistical treatment of simple random sampling and finite-population inference, review these authoritative sources:

• U.S. Census Bureau glossary and survey methodology resources
• Penn State STAT 506 resources on sampling theory
• NCBI Bookshelf overview of sampling methods in research

Final takeaway

Simple random sample probability calculation is fundamentally about counting how many valid samples satisfy your condition out of all equally likely samples from a finite population. The correct model is hypergeometric because the sample is drawn without replacement. By entering the population size, the number of successes in the population, the sample size, and the target number of successes, you can compute exact or cumulative probabilities, compare the result with the expected value, and visualize the full distribution. This makes the calculator useful not just for classroom exercises, but for real-world decision support in auditing, quality control, surveys, inventory review, and institutional analysis.