A small sample scoop beside a much larger bowl of colored beads. Original AI-generated artwork produced for GreenCalc; not a measured diagram or empirical data display.
The hat marks an estimate from observations
In a binary sample, p-hat is the observed success count divided by the number of trials. With 42 successes in 100 observations, p̂ = 0.42, or 42%. Population p is the model parameter being estimated; it is not automatically equal to 0.42 just because the sample happened to produce that fraction.
State what counts as a trial and a success
Define the outcome before counting. A trial might be one independently sampled component and success might mean passing a specified check. All trials must use the same outcome definition. Duplicate observations of the same component are not automatically independent new trials, and excluding inconvenient failures changes the estimand or biases the record.
Read the plug-in standard error conditionally
Under an independent constant-probability Bernoulli model, the plug-in standard error is √[p̂(1−p̂)/n]. For 42 out of 100, it is approximately 0.04936, about 4.936 percentage points. It describes estimated sampling variation under that model, not label accuracy or a correction for selection bias.
Choose estimation tools according to the question
The P-Hat Calculator reports the unweighted fraction and plug-in standard error without inventing a confidence interval. Survey weights, clustering and finite-population sampling can require other methods. The existing Confidence Interval Calculator has its own estimation scope and should be used only when its assumptions match the task. Read zero successes and endpoint standard errors before interpreting a boundary result as population certainty.
Check the model behind your result
Use the P-Hat Calculator for the supported calculation and review its stated assumptions before interpreting the output. Find related tools in the Statistics category, or browse the GreenCalc homepage for other calculation topics.
By GreenCalc Editorial Team · Updated 2026-10-26