---
title: "One-Sample Test"
description: "How Statsig uses one-sample tests to compare an experiment group's metric against a fixed reference value rather than against a control group."
product: general
last_updated: 2025-09-18
token_estimate: 498
---
# One-Sample Test

> For AI agents: a documentation index is available at [/llms.txt](/llms.txt). Append `.md` to any page URL for markdown, or send `Accept: text/markdown`.

## One-sample tests (fixed-value test)

A one-sample test compares a single sample of data against a known or hypothesized value to determine whether there is a statistically significant difference. Unlike A/B tests that compare two groups, one-sample tests evaluate whether a single group differs from a specific benchmark, target, or historical baseline.

## When to use one-sample tests

One-sample tests are useful for comparing a single group against a known value:

- **Single Group Events**: When only one group can trigger certain events (for example, feature usage or error types), compare against an expected baseline.
- **Algorithm Testing**: Test whether an algorithm performs better than random (for example, whether a success rate differs from 50%).

## Statistical considerations

One-sample tests provide a way to make statistical inferences about whether observed data differs significantly from a hypothesized value. The test helps determine whether any observed difference is due to random variation or represents a true change in the underlying process.

## Enable fixed-value baseline comparison

1. Go to the setup page of an experiment.

![Experiment setup screen highlighting metrics section](https://docs.statsig.com/images/experiments/statistical-methods/methodologies/one-sample-test/742634e0-0db9-44f9-b849-0a205f604a76.png)

2. Click the metric name.

![Metric name dropdown showing configure options](https://docs.statsig.com/images/experiments/statistical-methods/methodologies/one-sample-test/3a645771-4771-480e-a263-15a6af951284.png)

3. Select **Use Fixed Baseline as Control**.

![Fixed baseline control modal for one-sample test configuration](https://docs.statsig.com/images/experiments/statistical-methods/methodologies/one-sample-test/9c7e85d6-e0b0-40bb-a22c-a96d7084d3e7.png)