What does a sampling distribution of the sample mean describe?
Quick answer
A sampling distribution of the sample mean describes the values the average could take across repeated samples of the same size, drawn from a population using the same sampling design. It is a distribution of averages, not of the individual observations in one sample. Its shape and spread depend on the population and sampling setup. For random samples of fixed size, its standard deviation is the population standard deviation divided by the square root of the sample size. [S1]
The question, the details, the next step.
I understand that a sample mean is an average, but what does its sampling distribution describe?
It describes how the sample mean would vary if you repeatedly took samples of the same size from the same population using the same sampling method. Each sample produces one average; put those averages together and you have the sampling distribution of the mean. It is about variation in the statistic, not variation among the individual values in just one sample. The sampling setup matters, so specify the population, sample size, and how samples are selected. [S1]
Can you show that with a small example I can calculate myself?
Suppose, just for illustration, that a population has four activity times: 10, 20, 30, and 40 minutes. Imagine taking every possible sample of two different people, without replacement. There are six pairs:
- 10 and 20: mean 15
- 10 and 30: mean 20
- 10 and 40: mean 25
- 20 and 30: mean 25
- 20 and 40: mean 30
- 30 and 40: mean 35
The sampling distribution consists of the six resulting means: 15, 20, 25, 25, 30, and 35 minutes. The value 25 appears twice because two different pairs produce it. By contrast, the population consists of the four individual times, and any one sample contains just two of them. These numbers and this sampling plan are hypothetical; they illustrate the idea rather than describe real activity data.
Does that mean the sampling distribution is always bell-shaped, or that a sample mean is certain to match the population mean?
No. Do not assume it is always bell-shaped: its exact shape depends on the population and sampling setup. A normal approximation has conditions; the cited explanation discusses the role of sample size, not that every sampling distribution is normal. Nor does each sample mean have to equal the population mean. In the example, the six means vary, even though their average is 25 minutes—the same as the average of the four population values. That equality here follows from this particular complete set of equally represented pairs; it is not a guarantee for one sample. [S1]
What should I write when I need to interpret one in a statistics problem?
State what varies, what stays fixed, and what the distribution contains. For example: “For repeated samples of two activity times selected without replacement from the hypothetical population {10, 20, 30, 40}, the sampling distribution of the sample mean contains the averages 15, 20, 25, 25, 30, and 35 minutes. It describes variation in sample averages, not the spread of individual activity times in one sample.”
For a real problem, replace the hypothetical values with the stated population and sampling plan. If the problem gives only one sample, you can calculate its mean, but you cannot list the full sampling distribution from that sample alone. The standard deviation formula applies to random samples of fixed size as described above; the shape or numerical spread for a particular population requires the relevant details. [S1]
A practical checklist
Name the population the samples come from.
State the sample size and how samples are selected, including whether replacement is used when specified.
For each possible sample in a small example, calculate its mean.
Treat the resulting means—not the original individual observations—as the sampling distribution.
Interpret the distribution as variation across repeated samples, and avoid claiming a shape or value not supported by the setup.
A few more useful answers
How is a sampling distribution different from a sample’s data?
A sample’s data are the individual observations collected once. A sampling distribution of the mean is made from the means of repeated samples taken under the same setup.
What does the standard deviation of the sampling distribution measure?
It measures how much sample means vary across the specified sampling process. For random samples of fixed size, the standard deviation is the population standard deviation divided by the square root of the sample size. [S1]
Can I find the sampling distribution from one observed sample?
One sample gives you one mean, not all the possible means. To construct the full distribution, you need the population and sampling plan, or an appropriate model with enough information.
Sources
References used for this guide.
- [S1] OpenStax Introductory Statistics 2e: The Central Limit Theorem for Sample Means openstax.org · Checked