Concept Architecture
Concept
Theoretically, the Chi-Square Distribution is a continuous probability distribution describing the distribution of the sum of squared independent standard normal random variables. It forms the theoretical basis of numerous statistical hypothesis tests involving variances, frequencies and goodness-of-fit, and is fundamental to likelihood-based statistical inference.
Mathematically, the chi-square distribution is parameterised by its degrees of freedom and is obtained by summing independent squared standard normal variables. Many statistical test statistics, including Pearson's chi-square test, likelihood ratio tests and Wald-type asymptotic procedures, follow a chi-square distribution under the null hypothesis.
In practice, the chi-square distribution is used to calculate p-values and critical values for hypothesis testing in health economics, epidemiology and clinical research. It supports analyses of contingency tables, model goodness-of-fit, regression diagnostics and likelihood-based comparisons used in economic evaluations and health technology assessment.
Purpose
Used to evaluate statistical significance, assess model fit, analyse categorical data, estimate variance-related quantities and provide the reference distribution for numerous statistical hypothesis tests.
Mathematical Formulae
Primary Formula
?� = ?Z?�
where Z? ~ N(0,1)
Supporting Formulae
?� ~ ?�(k)
E(?�) = k
Var(?�) = 2k
Pearson ?� = ?((O? ? E?)� / E?)
Related Mathematical Methods
Pearson chi-square test
Likelihood ratio test
Goodness-of-fit testing
Test of independence
Wald test
Likelihood ratio statistics
Generalised linear models
Example
A health economist compares observed and expected treatment uptake across four patient groups. The observed frequencies are compared with expected frequencies under the null hypothesis using:
?� = ?((O? ? E?)� / E?)
Suppose the calculated statistic is ?� = 10.52 with 3 degrees of freedom. Comparing this value with the chi-square distribution yields a p-value below 0.05, indicating statistically significant differences between observed and expected frequencies.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| CHISQ.DIST.RT | =CHISQ.DIST.RT(10.52,3) | Calculate p-value from chi-square statistic |
| CHISQ.INV.RT | =CHISQ.INV.RT(0.05,3) | Obtain critical value |
| CHISQ.TEST | =CHISQ.TEST(B2:E3,G2:J3) | Perform Pearson chi-square test |
| SUM | =SUM(B2:E5) | Calculate observed totals |
VBA (Optional)
Automate chi-square hypothesis testing and goodness-of-fit analyses across multiple health economic datasets.
Sources
Agresti A. Categorical Data Analysis.
Casella G, Berger RL. Statistical Inference.
Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes.
Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
Related Concepts (2)
Library
Publications
1
Statistical Analysis of Cost-Effectiveness Data — Willan & Briggs, 1st Edition ed., 2006 (John Wiley & Sons)
A synthesis of statistical methods for analysing cost-effectiveness data, including net-benefit regression, confidence intervals for the ICER, cost-effectiveness acceptability curves, and covariate adjustment. Part of the Wiley Statistics in Practice series.
BookView source →
Frequently Asked Questions (6)
What is the chi-square distribution?
A continuous probability distribution arising from the sum of squared standard normal variables, underlying tests such as goodness of fit.
Source: Pearson 1900
What role does the chi-square distribution play in statistical tests?
The chi-square distribution arises from summing squared standard normal variables, and it provides the reference against which several common tests are judged. When a test statistic measuring the discrepancy between observed and expected counts is computed, its value is compared with this distribution to see how surprising it would be under the null hypothesis. Tests of goodness of fit and of association in contingency tables rely on it in just this way. Supplying the yardstick for such tests is its role. Kirkwood and Sterne (2003) describe this distribution.
Source: Kirkwood & Sterne 2003
How is the chi-square distribution used?
The chi-square distribution is used as the reference distribution for tests whose statistics follow it, including the chi-square goodness-of-fit test, which compares observed with expected frequencies; the chi-square test of independence in contingency tables; and likelihood ratio tests. The test statistic is compared with the chi-square distribution to obtain a p-value. So the chi-square distribution is used to assess whether observed data depart from what a hypothesis predicts, by referring a computed statistic to the distribution with the appropriate degrees of freedom, which is why it is central to categorical data analysis and to a variety of other tests in statistics.
Source: Pearson 1900
What are degrees of freedom in the chi-square distribution?
Degrees of freedom in the chi-square distribution are the parameter determining its shape, corresponding to the number of independent squared standard normal variables summed, or in tests to the number of independent pieces of information underlying the statistic. More degrees of freedom make the distribution less skewed and shift it rightward. So degrees of freedom define which chi-square distribution applies, and in tests they are derived from the structure of the data, such as the number of categories minus constraints, which is why identifying the correct degrees of freedom is necessary for referring a test statistic to the right chi-square distribution and obtaining a valid p-value.
Source: Pearson 1900
What tests rely on the chi-square distribution?
Tests relying on the chi-square distribution include the chi-square goodness-of-fit test, assessing whether observed frequencies match expected ones; the chi-square test of independence or association in contingency tables; likelihood ratio tests, whose statistics are approximately chi-square in large samples; and tests concerning variances. So a range of tests rely on the chi-square distribution as the reference for their statistics, particularly those involving categorical data and comparisons of observed with expected counts, which makes the distribution important in the analysis of frequencies and associations and in the large-sample testing of models through likelihood ratio statistics.
Source: Pearson 1900
What are the properties of the chi-square distribution?
The chi-square distribution takes only non-negative values, is right-skewed, and has a shape determined by its degrees of freedom, becoming more symmetric and approaching normality as the degrees of freedom increase. Its mean equals its degrees of freedom. So the chi-square distribution's properties include non-negativity, right skew that lessens with more degrees of freedom, and a mean equal to the degrees of freedom, and these characteristics, together with its origin as a sum of squared standard normals, explain its role as the reference distribution for tests based on squared deviations and on likelihood ratios.
Source: Pearson 1900
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 12 Dec 2025
Content version: 1.0.0
Canonical Identity
- Term code
- HE-ES-SA-024
Stable URI · Machine-readable · Resolvable · CC BY 4.0