What Is a One-Sample T-Test?
A one-sample t-test is a statistical test used to compare the mean value of one sample with a specific reference value. This reference value may come from a known population average, a clinical benchmark, a target standard, or a theoretically meaningful number.
Unlike an independent samples t-test or a paired samples t-test, the one-sample t-test does not compare two groups. Instead, it answers one focused question:
Is the average value in this sample significantly different from a chosen reference value?
This test is commonly used in medical research, psychology, education, public health, nutrition, and other fields where researchers want to compare one group against a known or expected value.
Why Is This Test Useful?
In many studies, researchers already have a meaningful value for comparison. For example, a national average score, a recommended clinical threshold, a standard laboratory value, or a target treatment goal may be available.
Suppose a researcher knows that the average score on a standardized test is 100. She collects scores from students at one school and finds that their average score is 106. At first glance, the students seem to perform better than the national average.
However, samples naturally vary. A difference between 106 and 100 may reflect a true difference, but it may also be caused by random sampling variation. The one-sample t-test helps determine whether the observed sample mean is far enough from the reference value to be considered statistically significant.
How Does the Test Work?
The one-sample t-test compares the sample mean with the hypothesized or reference value. It calculates a t-statistic, which reflects how far the sample mean is from the reference value relative to the variability in the sample.
In simple terms, the test considers three pieces of information:
- The sample mean.
- The reference value.
- The variability of the observations in the sample.
If the sample mean is very different from the reference value and the sample data are not highly variable, the t-statistic will be larger. A larger absolute t-value suggests stronger evidence that the population mean may differ from the reference value.
The test also produces a p-value. The p-value tells us how likely it would be to observe a sample mean as far from the reference value as the one in the data if the true population mean were actually equal to that reference value.
A common significance level is 0.05. If the p-value is less than 0.05, the result is usually considered statistically significant.
How Is It Different from a Z-Test?
The one-sample t-test is similar in purpose to a one-sample z-test. Both tests compare a sample mean with a known or hypothesized value.
The key difference is whether the population standard deviation is known.
A z-test requires the true population standard deviation. In real research, this value is rarely known. Most of the time, researchers only have the standard deviation calculated from their sample.
The one-sample t-test uses the sample standard deviation to estimate the population variability. Because this estimate introduces additional uncertainty, the test uses the t-distribution instead of the normal distribution.
When the sample size becomes large, the t-distribution becomes very similar to the normal distribution. For this reason, the difference between a t-test and a z-test becomes smaller as sample size increases.
What Does the Result Mean?
A statistically significant one-sample t-test suggests that the sample mean is unlikely to have come from a population with the specified reference mean.
For example, if a group of patients has an average laboratory value that is significantly higher than a clinical reference value, the result suggests that this patient group may differ from the reference population.
A non-significant result means that the data do not provide strong enough evidence to conclude that the sample mean differs from the reference value. It does not prove that the true mean is exactly equal to the reference value.
As with other statistical tests, the p-value should not be interpreted alone. Researchers should also consider the size of the difference, confidence intervals, sample size, measurement quality, and practical or clinical importance.
Key Assumptions
Several assumptions should be considered before using a one-sample t-test.
1. One sample of observations
The test is used for one group of observations. Each value should come from the same study sample or population of interest.
2. Continuous outcome variable
The measured variable should be continuous, such as blood pressure, body weight, test score, biomarker concentration, reaction time, or daily calorie intake.
3. Independent observations
Each observation should be independent. One subject’s value should not determine or influence another subject’s value.
For example, measurements from different individuals are usually independent. Repeated measurements from the same individual are not independent and require a different method.
4. Approximate normality
The data should be approximately normally distributed, especially when the sample size is small.
When the sample size is moderate or large, the one-sample t-test is generally robust to mild departures from normality because the sampling distribution of the mean becomes approximately normal.
5. No severe outliers
Extreme outliers can strongly influence the sample mean and standard deviation. Before using a one-sample t-test, it is good practice to inspect the data for unusually large or small values.
If the data are highly skewed or contain serious outliers, a non-parametric alternative such as the one-sample Wilcoxon signed-rank test may be considered.
When Should You Use It?
Use a one-sample t-test when:
- You have one group of observations.
- The outcome variable is continuous.
- You want to compare the sample mean with a specific reference value.
- The reference value is meaningful for the research question.
Common examples include:
- Comparing a school’s average test score with a national average.
- Comparing patients’ mean blood pressure with a clinical target.
- Comparing average daily calorie intake with a recommended value.
- Testing whether the mean change score differs from zero.
- Comparing a laboratory measurement with a known reference standard.
When Should You Not Use It?
A one-sample t-test is not appropriate when you want to compare two separate groups. In that situation, an independent samples t-test should be used.
It is also not appropriate when the same subjects are measured twice and you want to compare before-and-after values. In that case, a paired samples t-test is more suitable.
If the outcome is categorical, such as positive/negative or yes/no, a one-sample proportion test or binomial test may be more appropriate.
If the data are ordinal or strongly non-normal, a non-parametric method may be a better choice.
A Simple Example
Suppose a nutrition researcher wants to know whether the average daily calorie intake of university students differs from the recommended value of 2,000 calories.
She collects data from 50 students and finds:
- Mean daily calorie intake = 2,180 calories
- Standard deviation = 420 calories
- Reference value = 2,000 calories
She performs a one-sample t-test and obtains:
- t = 3.03
- p = 0.004
Because the p-value is less than 0.05, the result is statistically significant. The researcher concludes that the average daily calorie intake in this group is significantly different from the recommended value.
Since the sample mean is higher than 2,000, the result suggests that these students consume more calories on average than the recommended level.
How to Report the Result
A one-sample t-test result can be reported like this:
“The sample mean was significantly different from the reference value, t(df) = value, p = value.”
For example:
“The mean daily calorie intake was significantly higher than the recommended value of 2,000 calories, t(49) = 3.03, p = 0.004.”
A more complete report may include:
- Sample size.
- Sample mean.
- Standard deviation.
- Reference value.
- Mean difference.
- Confidence interval.
- t-statistic and degrees of freedom.
- p-value.
- Effect size.
Summary
The one-sample t-test is a simple and useful method for comparing one sample mean with a meaningful reference value. It is especially helpful when researchers want to know whether a study sample differs from a known standard, target value, or population average.
To use the test correctly, researchers should make sure the outcome is continuous, observations are independent, the data are reasonably normal, and severe outliers are not driving the result.
The one-sample t-test is easy to understand and widely applicable, but its result should always be interpreted together with the size of the difference and its practical or clinical importance.