This sample proportion calculator converts a count of successes x and a sample size n into the statistic p̂, reported as a simplified fraction, a decimal, and a percentage, together with its standard error. A second mode treats a population proportion as known and calculates normal-approximation probabilities for where p̂ is likely to fall.

Use it when you need to describe one binary outcome in one sample—converted or not, defective or not, yes or no—or to answer textbook questions about the sampling distribution of p̂. It reports a point estimate and probabilities; it does not run a hypothesis test or build a confidence interval.

02Input

How to Use the Sample Proportion Calculator

Choose a task. Point estimate and standard error describes an observed sample; sampling-distribution probability predicts p̂ from a known population proportion.
For a point estimate, enter the number of successes x and the sample size n. Define what counts as a success before looking at the data.
For a sampling-distribution probability, enter the population proportion p as a decimal, the sample size, and a lower bound, an upper bound, or both for p̂.
Read the result with its large-count check. The check reports whether the normal approximation behind the standard error and probability is reasonable for these counts.

Counts must be whole numbers with 0 ≤ x ≤ n. Proportions and bounds are decimals: enter 0.46, not 46, for 46 percent.

03Calculation

Sample Proportion Formula and Standard Error

p̂ is the number of successes divided by the sample size:

p̂ = x / n

The calculator also reduces x / n to its simplest fractional form, so 138 / 300 is reported as 23/50 alongside the decimal 0.46. The standard error of p̂ estimates how much the statistic varies from sample to sample:

SE = √[p̂(1 − p̂) / n]

In the sampling-distribution mode the roles reverse: the population proportion p is treated as known, the sampling distribution of p̂ is modeled as approximately normal with mean p and standard error √[p(1 − p) / n], and each bound is converted to a z-score before reading a standard-normal probability.

Point estimatep̂ → p

The sample proportion p̂ is the point estimate of the unknown population proportion p. A point estimate is a single best guess; it carries no uncertainty statement by itself, which is why the calculator always reports the standard error next to it.

04Concepts

Point Estimate, Standard Error, and Sampling Distribution

Three linked ideas explain everything the sample proportion calculator reports. Each one answers a different question about the same ratio.

ADescribep̂ = x / n

Point Estimate of the Population Proportion

p̂ summarizes the sample you actually observed and is the standard point estimate of the population proportion p. It is a statistic, not the parameter: a different random sample would give a different p̂.

BVarySE = √[p̂(1 − p̂)/n]

Standard Error of p̂

The standard error measures the typical sample-to-sample variation of p̂. It shrinks with the square root of n: quadrupling the sample size halves the standard error, which is why precision gains flatten out.

CPredictp̂ ~ N(p, SE)

The Sampling Distribution of p̂

When the large-count condition holds, p̂ from repeated random samples is approximately normal, centered at p. That model turns questions like “how likely is p̂ to reach 0.46?” into standard-normal areas.

05Worked example

Sample Proportion Calculator Examples, Both Modes

A preplanned random sample checks 300 orders and finds 138 that shipped late.

Sample proportion: p̂ = 138 / 300 = 23/50 = 0.46, or 46 percent; the other 162 orders shipped on time.
Standard error: SE = √[0.46(1 − 0.46) / 300] = 0.028775.
Large-count check: 138 successes and 162 failures both exceed 10, so normal-based follow-up methods are reasonable to consider.

Now suppose the long-run late rate is believed to be p = 0.50, and the question is how likely a random sample of 300 is to produce p̂ of 0.46 or more.

The sampling distribution of p̂ has mean 0.50 and standard error √[0.50(0.50) / 300] = 0.028868.
Standardize the bound: z = (0.46 − 0.50) / 0.028868 = −1.385641.
Probability: P(p̂ ≥ 0.46) = 1 − Φ(−1.385641) ≈ 0.917072.

A sample proportion at or above 0.46 would be quite common under that model; about 92 percent of such samples qualify.

Under p = 0.50 the sampling distribution centers at 0.50 with SE ≈ 0.0289; P(p̂ ≥ 0.46) ≈ 0.917.

06Validity

Conditions and Common Mistakes

The arithmetic is exact; the interpretation depends on the study design. Review these points before using output from the sample proportion calculator in either mode:

Representative sample: p̂ estimates the population proportion only when the sample is random or defensibly representative. A precisely calculated proportion from a biased sample is still biased.
Independent observations: one observation should not determine another. When sampling without replacement, the usual guideline keeps the sample below about 10 percent of the population.
Large counts: the point-estimate mode checks x ≥ 10 and n − x ≥ 10; the sampling mode checks np ≥ 10 and n(1 − p) ≥ 10. When a count falls short, exact binomial methods are more defensible than the normal approximation.
Decimals, not percentages: entering 46 instead of 0.46 for a bound or proportion is the most common input error. The calculator rejects values outside [0, 1] rather than guessing.
Statistic vs. parameter: p̂ never becomes p by rounding. Any claim about the population needs an interval or test, not just the point estimate.

07Interpretation

Reading p̂, Its Fraction, and Its Probabilities

Is the sample proportion the same as the population proportion?

No. p̂ is calculated from one observed sample; p is the fixed, unknown population value. p̂ is the point estimate of p, and the standard error describes how far such estimates typically fall from the parameter.

Why does the calculator report a fraction like 23/50?

The reduced fraction is the exact value of x / n with no rounding, which is useful for homework answers and for spotting simple underlying ratios. The decimal and percentage are the same number in more familiar units.

Which standard error does each mode use?

The point-estimate mode substitutes p̂ into √[p(1 − p) / n] because p is unknown. The sampling-distribution mode substitutes the stated population value p itself. The two standard errors are close whenever p̂ is close to p, but they answer different questions.

Can I get the probability that p̂ lands in a specific range?

Yes. Enter both a lower and an upper bound in the sampling mode, and the calculator reports the normal-approximation probability between them, along with the z-score for each bound. Leave one bound blank for a one-sided probability.

08Method

Method, Limitations, and Statistical Sources

This calculator reports the exact ratio x / n and, for probabilities, the uncorrected large-sample normal approximation to the sampling distribution of p̂. It does not apply a continuity correction, compute exact binomial probabilities, build confidence intervals, or test hypotheses. For those next steps, use the site’s proportion confidence interval calculator or one proportion z test calculator, and keep the method consistent when comparing outputs with other software.

The definitions and conditions follow the NIST/SEMATECH e-Handbook of Statistical Methods and Penn State STAT 200’s lesson on the sampling distribution of the sample proportion. The ≥ 10 large-count guideline reported by the calculator matches those references. Educational references do not replace a statistical analysis plan for high-stakes decisions.