Concept Architecture
Concept
Theoretically, the Binomial Distribution is a discrete probability distribution that describes the number of successful outcomes obtained in a fixed number of independent Bernoulli trials, each having the same probability of success. It represents one of the fundamental probability distributions in statistical inference and is founded on Bernoulli trial theory. The concept exists to model binary outcomes such as disease occurrence, treatment response or diagnostic success when the number of trials is fixed.
Mathematically, the Binomial Distribution assigns probabilities to each possible number of successes using combinations together with the probabilities of success and failure. The distribution is completely characterised by the number of trials and the probability of success on each trial. Its expected value and variance describe the central tendency and dispersion of the distribution and provide the basis for statistical inference.
In practice, the Binomial Distribution is applied to binary outcome data arising in clinical trials, epidemiological studies, diagnostic accuracy research and health economic evaluations. It underpins exact confidence intervals, hypothesis tests, Bayesian analyses and simulation models involving dichotomous clinical events.
Purpose
Used to model binary outcomes, estimate event probabilities, support statistical inference for proportions and underpin clinical, epidemiological and health economic analyses involving dichotomous events.
Mathematical Formulae
Primary Formula
P(X = k) = C(n,k) ? p? ? (1 ? p)�??
where:
- X = number of successes
- k = observed number of successes
- n = total number of trials
- p = probability of success
- C(n,k) = n! / [k!(n ? k)!]
Supporting Formulae
E(X) = np
Var(X) = np(1 ? p)
SD(X) = �[np(1 ? p)]
Related Mathematical Methods
- Bernoulli Distribution
- Bernoulli Trial
- Binomial Test
- Beta Distribution
- Beta-Binomial Distribution
- Normal Approximation to the Binomial
- Maximum Likelihood Estimation
- Bayesian Analysis
Example
A vaccine trial enrols 20 participants. The probability that an individual develops protective immunity is estimated to be 0.80.
The probability that exactly 18 participants develop immunity is:
P(X = 18)
= C(20,18) ? 0.80?? ? 0.20�
= 190 ? 0.80?? ? 0.04
� 0.1369
The probability of observing exactly 18 successful immunisations is therefore approximately 13.7%.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| BINOM.DIST | =BINOM.DIST(18,20,0.8,FALSE) | Calculates the probability of exactly 18 successes. |
| BINOM.DIST | =BINOM.DIST(18,20,0.8,TRUE) | Calculates the cumulative probability of up to 18 successes. |
| COMBIN | =COMBIN(20,18) | Calculates the binomial coefficient. |
| SQRT | =SQRT(B2*C2*(1-C2)) | Calculates the standard deviation of a binomial distribution. |
VBA (Optional)
A VBA macro can automate binomial probability calculations, confidence interval estimation and simulation of binary clinical outcomes across multiple study scenarios.
Sources
- Casella G, Berger RL. Statistical Inference. 2nd ed.
- Mood AM, Graybill FA, Boes DC. Introduction to the Theory of Statistics. 3rd ed.
- Rice JA. Mathematical Statistics and Data Analysis. 3rd ed.
- Agresti A. Categorical Data Analysis. 3rd ed.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
Related Concepts (2)
Library
Publications
1
Bayesian Methods in Health Economics — Gianluca Baio, 1st Edition ed., 2012 (Chapman & Hall / CRC Press)
An overview of Bayesian statistical methods for the analysis of health economic data, covering economic evaluation concepts, statistical cost-effectiveness analysis, Bayesian computation and MCMC, and applied health economic evaluation.
BookView source →
Frequently Asked Questions (6)
What is the binomial distribution?
A discrete probability distribution describing the number of successes in a fixed number of independent trials with the same success probability each.
Source: Casella G, Berger RL. Statistical Inference. 2nd ed. Duxbury; 2002.
What does the binomial distribution describe?
The binomial distribution describes the number of successes in a fixed number of independent trials, each with the same probability of success, such as the number of heads in ten coin tosses or of responders among ten treated patients. It gives the probability of every possible count of successes, which underlies inference about proportions, since an observed proportion is a count of successes over a total. Its shape depends on the number of trials and the success probability. Counting successes across independent trials is what it captures. Kirkwood and Sterne (2003) describe this distribution.
Source: Kirkwood & Sterne 2003
What are the assumptions of the binomial distribution?
The binomial distribution assumes a fixed number of trials; that each trial has only two outcomes, success or failure; that the probability of success is the same on every trial; and that the trials are independent of one another. If these assumptions hold, the count of successes follows a binomial distribution. So the binomial distribution rests on fixed, independent, identical binary trials, and its validity depends on these conditions, meaning that if the success probability varies between trials or the trials are dependent, the distribution may not apply, which is why these assumptions are checked when using it to model counts of binary events.
Source: Casella & Berger 2002
How is the binomial distribution used?
The binomial distribution is used to model and analyse the number of successes in binary outcome data, such as the number of events among a fixed group, to calculate probabilities of particular counts, to construct confidence intervals for proportions, and to underpin tests and models for binary data, including logistic regression. So the binomial distribution is used wherever counts of binary events out of a fixed number arise, providing the basis for inference about proportions and for statistical methods dealing with yes-or-no outcomes, which makes it one of the foundational distributions in the analysis of health and other data involving events among individuals.
Source: Casella & Berger 2002
What are the properties of the binomial distribution?
The binomial distribution has a mean equal to the number of trials times the success probability, and a variance equal to the number of trials times the success probability times one minus the success probability. It is discrete, taking whole-number values from zero to the number of trials, and it becomes approximately normal when the number of trials is large and the probability not too extreme. So the binomial distribution's properties include a defined mean and variance determined by its parameters and an approach to normality in large samples, which is why, for large numbers of trials, normal approximations are sometimes used for inference about proportions derived from it.
Source: Casella & Berger 2002
How does the binomial distribution relate to proportions?
The binomial distribution relates to proportions because the sample proportion, the number of successes divided by the number of trials, is derived from a binomial count, so inference about a proportion is based on the binomial distribution. The distribution provides the probabilities of different counts, and hence of different observed proportions, allowing confidence intervals and tests for proportions. So the binomial distribution underpins the analysis of proportions, since an observed proportion of events reflects a binomial count, and understanding its distribution is what allows valid estimation and testing of the underlying probability of success, such as a response or event rate.
Source: Casella & Berger 2002
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 11 Dec 2025
Content version: 1.0.0
Canonical Identity
- Term code
- HE-ES-SA-015
Stable URI · Machine-readable · Resolvable · CC BY 4.0