If you searched for what is the mean of the sample means and you want to calculate it yourself rather than just understand the theory, this guide is for you. Here, what is the mean of the sample means is treated as a practical calculation: you will learn a clear step-by-step method, work through fully solved examples with small numbers, then move on to larger examples with realistic data, and finally learn how to check your answer and avoid the usual mistakes. The companion concept guide explains why the method works; this guide focuses on doing it, with every arithmetic step shown so that you can follow along with a calculator and reproduce each result.
Before the examples, here is the method in brief. To find what is the mean of the sample means, first list the samples you are working with. Second, calculate the mean of each sample by adding its values and dividing by the number of values in that sample. Third, add all of those sample means together. Fourth, divide that total by the number of samples. The result is the mean of the sample means. When all samples have the same size, there is a useful shortcut: you can add every value in every sample together and divide by the total number of values, and you will get the same answer. The worked examples below show both routes, so you can see why they agree and when the shortcut does not apply.
The step-by-step method for what is the mean of the sample means
Step one is to organise your data clearly. Write each sample on its own line, and label it Sample 1, Sample 2, and so on, so that you never mix values between samples. Step two is to find each sample mean: add the values in a sample, count how many values it contains, and divide the total by that count. Write each sample mean down with at least two decimal places, because rounding too early is a common source of small errors. Step three is to add the sample means together. Step four is to divide by the number of samples, not by the number of individual values, and that distinction is the single most common mistake in this topic. The divisor in the final step is the count of samples, and remembering that one rule will save you from most errors.
Step five is to check your answer for sense. The mean of the sample means must lie between the smallest and the largest sample mean: an average of several numbers can never be outside their range. If your answer is larger than every sample mean, or smaller than every sample mean, you have made an arithmetic slip. A second check is to compare your answer with the overall average of all the individual values when samples are equal in size; the two should match exactly. A third check is to look at whether the answer is plausible given the data: if all your samples cluster around 50, a mean of the sample means of 90 is plainly wrong. These quick checks take seconds and catch most mistakes before they reach an exam script or a report.
Worked example 1: small numbers, fully shown
Suppose three samples are drawn, each of size 4. Sample 1 is 6, 8, 10, 12. Sample 2 is 4, 8, 12, 16. Sample 3 is 10, 10, 14, 18. Start with Sample 1: the values add to 36, and 36 divided by 4 gives a sample mean of 9. Next, Sample 2: the values add to 40, and 40 divided by 4 gives a sample mean of 10. Then Sample 3: the values add to 52, and 52 divided by 4 gives a sample mean of 13. Now add the sample means: 9 plus 10 plus 13 equals 32. Finally, divide by the number of samples, which is 3, and 32 divided by 3 equals 10.67, to two decimal places. So what is the mean of the sample means in this example? It is 10.67.
Check the answer with the shortcut. Add every individual value together: Sample 1 totals 36, Sample 2 totals 40, and Sample 3 totals 52, giving a grand total of 128. There are 12 individual values in total, and 128 divided by 12 equals 10.67, matching the first calculation. The shortcut works here because every sample has the same size, 4. Notice also that the answer 10.67 lies between the smallest sample mean, 9, and the largest, 13, as it must. Working through an example this small, where every step can be checked by hand, is the best way to build confidence before tackling larger data sets where a calculator does the heavy lifting but the method stays exactly the same.
Worked example 2: samples of different sizes
Now consider samples of different sizes, where the shortcut needs care. Sample 1, of size 2, is 20 and 30, giving a mean of 25. Sample 2, of size 3, is 10, 20, and 30, giving a mean of 20. Sample 3, of size 5, is 40, 50, 60, 70, and 80, giving a mean of 60. To find what is the mean of the sample means, add the three sample means, 25 plus 20 plus 60 equals 105, and divide by the number of samples, 3, giving 35. That is the unweighted mean of the sample means: each sample counts equally, whatever its size. It is the answer most textbook questions ask for when they use this phrase.
Now compare that with the pooled average of all individual values. The grand total is 50 plus 60 plus 300, which equals 410, and there are 10 individual values, so the pooled average is 41. The two answers, 35 and 41, differ, and both are correct answers to different questions. The mean of the sample means treats each sample equally, while the pooled average treats each individual value equally, so the large sample of high values pulls the pooled average up. When a question asks what is the mean of the sample means, give 35, the average of the sample means, and do not pool the values unless the question explicitly asks for the overall mean. Examiners set this trap often, and the worked numbers above show exactly how it works.
Worked example 3: realistic data with decimals
Real data rarely comes in tidy whole numbers, so here is an example with decimals, similar to what you might meet in a survey or a laboratory report. Four samples of daily rainfall, in millimetres, are recorded over short periods. Sample 1 is 2.4, 3.1, 1.8, and 4.2, which total 11.5 and give a mean of 2.875. Sample 2 is 3.6, 2.9, 3.3, and 3.0, which total 12.8 and give a mean of 3.2. Sample 3 is 1.5, 2.0, 2.5, and 2.8, which total 8.8 and give a mean of 2.2. Sample 4 is 4.0, 3.8, 4.4, and 3.6, which total 15.8 and give a mean of 3.95. Adding the four sample means, 2.875 plus 3.2 plus 2.2 plus 3.95, gives 12.225, and dividing by 4 gives 3.05625, which rounds to 3.06 millimetres.
Two practical lessons emerge from this example. First, keep full precision until the final step: if you had rounded each sample mean to one decimal place before averaging, you would have averaged 2.9, 3.2, 2.2, and 4.0, giving 3.075, which rounds to 3.08, a slightly different answer. The difference is small here, but in graded work it can cost marks, so carry at least two extra digits through the working and round only at the end. Second, the answer again sits between the smallest sample mean, 2.2, and the largest, 3.95, confirming the range check. When you report a result like this, state the units, rainfall in millimetres, and the number of samples, so that a reader can judge what the number represents without hunting through your working.
Worked example 4: every possible sample from a tiny population
The most instructive example of what is the mean of the sample means uses every possible sample from a tiny population, because it shows the theory in action. Take the population 3, 6, and 9. Its population mean is 6, since 3 plus 6 plus 9 equals 18 and 18 divided by 3 equals 6. Now list every possible sample of size 2 drawn with replacement: 3 and 3, 3 and 6, 3 and 9, 6 and 3, 6 and 6, 6 and 9, 9 and 3, 9 and 6, and 9 and 9. That is 9 samples. Their means are 3, 4.5, 6, 4.5, 6, 7.5, 6, 7.5, and 9. Add those means together: 3 plus 4.5 plus 6 plus 4.5 plus 6 plus 7.5 plus 6 plus 7.5 plus 9 equals 54. Divide by 9 samples, and the result is exactly 6, the population mean.
This example is worth studying because it demonstrates the central result with numbers you can verify in minutes: the mean of the sample means equals the population mean when every possible sample is included. In real research you never list every possible sample, because populations are far too large, but the principle is the same. Your handful of samples gives an approximation to this ideal average, and the more samples you take, the closer your average of sample means comes to the population mean. Teachers use this tiny-population example precisely because it makes an abstract theorem visible, and students who work through it once rarely forget what the phrase means afterwards. Try changing the population to 2, 5, and 11, and confirm for yourself that the same equality holds.
| Example | Samples | Sample means | Mean of sample means |
|---|---|---|---|
| 1: small whole numbers | 3 samples of 4 | 9, 10, 13 | 10.67 |
| 2: unequal sizes | Sizes 2, 3, 5 | 25, 20, 60 | 35 |
| 3: rainfall decimals | 4 samples of 4 | 2.875, 3.2, 2.2, 3.95 | 3.06 |
| 4: every possible sample | 9 samples of 2 | 3 to 9, as listed | 6, exactly the population mean |
For background reading on this site, see guides to defining the mean in maths, to the meaning of a stratified sample, and to the meaning of probability in maths terms. Those guides cover the vocabulary that appears in sampling questions, while this article covers the arithmetic itself.
Common mistakes when working out what is the mean of the sample means
The most frequent mistake is dividing by the wrong number in the final step: dividing the total of the sample means by the total number of individual values instead of by the number of samples. That error gives a far smaller answer, and it is easy to spot because the result will usually fall below the smallest sample mean. The second mistake is rounding too early, as shown in the rainfall example, where early rounding shifted the answer by 0.02. The third mistake is mixing values between samples, which happens when data is copied in a hurry; labelling each sample and checking each total against the original values prevents it. The fourth mistake is confusing the mean of the sample means with the pooled mean when sample sizes differ, as shown in example 2; reading the question wording carefully is the only defence, and the phrase “mean of the sample means” always means averaging the sample means themselves.
A fifth mistake is forgetting units and context in the final answer. A bare number such as 35 means little on its own, while “35 marks, the mean of three sample means” tells the reader exactly what was calculated. A sixth mistake is presenting too many decimal places, which implies false precision: with data measured to one decimal place, reporting a mean to five decimal places overstates accuracy, and two or three decimal places are usually enough. Finally, watch out for missing values: if a sample has a blank entry, decide before calculating whether that entry is genuinely missing or simply zero, because treating a missing value as zero drags the sample mean down. Good statistical practice records such decisions explicitly, so that anyone checking the work can follow them.
Frequently asked questions
What is the mean of the sample means, in calculation terms?
In calculation terms, what is the mean of the sample means is found by calculating each sample mean, adding those means, and dividing by the number of samples. That is the whole method, and every example in this guide follows it.
Do I divide by the number of samples or the number of values?
Divide by the number of samples. The number of individual values matters only when you are finding a single sample mean, or when you are deliberately calculating the pooled average of all values, which is a different quantity.
When can I use the shortcut of pooling all the values?
You can pool all the values and divide by the total count only when every sample has the same size. With unequal sample sizes, pooling gives the pooled mean, not the mean of the sample means, and the two answers will differ.
How many decimal places should I keep?
Keep full precision through the working, carry at least two extra digits, and round only the final answer, usually to two or three decimal places. Rounding early is a common cause of small but avoidable errors.
How do I check my answer quickly?
Check that your answer lies between the smallest and largest sample means, and, for equal-sized samples, check it against the pooled average of all values. If either check fails, rework the arithmetic before moving on.
A final practice note on what is the mean of the sample means
If you are practising for an exam, try this routine: pick three small samples, calculate each sample mean, add those means, and divide by three, and you have answered what is the mean of the sample means for that data set. Then check your answer against the range of the sample means and, for equal-sized samples, against the pooled average of all the values. Repeating that routine with different numbers, including decimals and unequal sample sizes, builds the speed and accuracy that exam questions reward. Remember that what is the mean of the sample means always means averaging the sample means themselves, never pooling the raw values unless the question asks for the pooled mean.
The most valuable habit in this topic is showing your working clearly: label each sample, write down each sample mean, and state the final divisor, the number of samples. Examiners give credit for method as well as for the final answer, and clear working makes your own checking far easier. With that habit, what is the mean of the sample means becomes one of the most reliable marks available in a statistics paper, because the method never changes, whatever numbers the question supplies.
Conclusion
To conclude, what is the mean of the sample means is a calculation you can now carry out with confidence: find each sample mean, add those means, and divide by the number of samples, checking that your answer falls within the range of the sample means. The small-number example, the unequal-size example, the decimal example, and the every-possible-sample example each teach a different lesson, from the basic method to the traps that catch unwary calculators. Practise with your own numbers, keep your working organised, round only at the end, and the phrase will stop being a tongue-twister and become a routine tool in your statistics work.
