How To Calculate Mean Of Variable In Stata

Stata Mean Calculator

Use the calculator below to compute a mean from raw values, preview the equivalent Stata command, and visualize the data. Then read the expert guide for a practical, analyst-level explanation of how mean calculations work in Stata using summarize, tabstat, egen, and related commands.

Interactive Mean Calculator

Expert Guide: How to Calculate Mean of Variable in Stata

If you are learning how to calculate mean of variable in Stata, the good news is that Stata makes the process fast, flexible, and highly reliable. In most cases, calculating a mean takes only one command. The more important skill is understanding which command to use, how missing values are handled, when to calculate means by group, and how to save those means for later analysis or reporting. This guide walks through the complete workflow from beginner basics to more advanced analyst use cases.

The mean, also called the arithmetic average, is the sum of all valid observations divided by the number of valid observations. In Stata, the simplest way to calculate the mean of a variable is to use summarize. For example, if your dataset contains a variable named income, you can type summarize income in the Command window. Stata will report the number of nonmissing observations, the mean, standard deviation, and the minimum and maximum values.

Key idea: Stata excludes missing numeric values by default when calculating a mean. That behavior is usually what researchers want, but you should always confirm how much missing data exists before interpreting the result.

The fastest way to calculate a mean in Stata

For a single variable, this is the standard approach:

summarize variable_name

If your variable is age, the command becomes:

summarize age

Stata then returns a table with five major outputs:

• Obs: the number of nonmissing observations used in the calculation
• Mean: the arithmetic average
• Std. dev.: the standard deviation
• Min: the smallest observed value
• Max: the largest observed value

This command is ideal when you need a quick descriptive statistic and want a clean summary. It is also one of the first commands many Stata users learn because it is intuitive and works across nearly every quantitative dataset.

How Stata actually computes the mean

The formula is straightforward:

Mean = (sum of all nonmissing values) / (number of nonmissing values)

Suppose you have five observations for a variable called score: 70, 75, 80, 85, and 90. The sum is 400, and the number of observations is 5, so the mean is 80. If one value were missing, Stata would exclude it and divide only by the remaining valid observations. That means if your values were 70, 75, 80, 85, and missing, the sum of valid values would be 310 and the count of valid observations would be 4, so the mean would be 77.5.

Useful Stata commands for mean calculation

Although summarize is the default choice, it is not the only option. Here are the most useful commands depending on your objective:

1. summarize for a quick descriptive summary of one or more variables.
2. tabstat when you want a customized table with selected statistics such as mean, median, and standard deviation.
3. mean when you want estimation output, confidence intervals, and postestimation compatibility.
4. egen when you want to create a new variable containing a mean, especially by group.
5. collapse when you want to reduce the dataset to group-level means.

Here are practical examples:

summarize income

tabstat income, statistics(mean sd min max n)

mean income

egen mean_income = mean(income)

bysort region: egen region_mean_income = mean(income)

Calculating the mean for multiple variables

Stata can summarize several variables at once. If you want the means for income, age, and hours, use:

summarize income age hours

This is especially useful during the exploratory phase of analysis when you want a quick picture of central tendency across your key measures. However, if your variables are on different scales, remember that comparing their means directly may not be substantively meaningful without context.

Calculating mean by group in Stata

One of the most common real-world tasks is finding the mean of a variable within categories, such as average income by sex, average test score by school, or average wage by industry. There are several ways to do this. A very common pattern is:

bysort groupvar: summarize outcome

For example:

bysort gender: summarize income

This gives separate summaries for each level of gender. If you want to store the group mean in a new variable for every observation in that group, use:

bysort gender: egen mean_income = mean(income)

That command is extremely valuable when you need the group mean for later modeling, graphing, or comparison. For example, you might calculate each student’s deviation from the school average or compare each employee’s wage to the average wage in their occupation.

Using the mean command instead of summarize

The mean command deserves special attention because it is not just descriptive. It treats the mean as an estimand and reports standard errors and confidence intervals. That makes it useful when your goal is inferential rather than purely descriptive. For example:

mean income

If your data come from a sample and you want to discuss uncertainty around the estimated population mean, this command can be more appropriate than summarize. Researchers working with survey data may go further and use survey-prefixed estimation commands to produce design-corrected means.

Real statistics example from Stata’s sample auto dataset

Stata’s well-known sample auto dataset is often used to teach descriptive statistics. The table below shows real summary values commonly produced from that dataset, illustrating how means look in practice.

Variable	Observations	Mean	Standard Deviation	Minimum	Maximum
price	74	6165.26	2949.50	3291	15906
mpg	74	21.30	5.79	12	41
weight	74	3019.46	777.19	1760	4840

If you type sysuse auto, clear followed by summarize price mpg weight, you can reproduce these summary values directly in Stata. This is a useful benchmark because it helps you verify that you understand how Stata presents means and related statistics.

Comparing means by category

Another valuable use case is comparing means across groups. In the same auto dataset, analysts often compare domestic and foreign cars. Means can reveal meaningful pattern differences quickly.

Group	Variable	Approximate Mean	Interpretation
Domestic cars	mpg	19.83	Domestic models average lower fuel efficiency.
Foreign cars	mpg	24.77	Foreign models average higher fuel efficiency.
Domestic cars	price	6072.42	Average listed price is slightly lower than foreign cars in this sample.
Foreign cars	price	6384.68	Average listed price is somewhat higher in this sample.

You can generate those grouped means with:

tabstat mpg price, by(foreign) statistics(mean n)

This format is often better than a basic summary when you need to compare categories side by side in a polished output table.

How missing values affect the mean

Missing values are one of the most important practical issues in any mean calculation. In Stata, numeric missing values are represented by . and extended missing values such as .a, .b, and so on. By default, Stata excludes them from mean calculations. This is often sensible, but it can hide a data quality problem if many observations are missing.

A good workflow is:

1. Calculate the mean.
2. Check the observation count used.
3. Inspect how many values are missing.
4. Decide whether the resulting mean is still representative.

For example:

summarize income

count if missing(income)

If many cases are missing, the reported mean may reflect only a subset of the sample. In applied work, that can materially change interpretation.

Saving the mean into a scalar or variable

Sometimes you need the mean for later programming or reporting. After summarize, Stata stores results in memory. The mean is available as r(mean). For example:

summarize income display r(mean)

You can also save it to a scalar:

summarize income scalar avg_income = r(mean)

This is useful in do-files, automation scripts, and reproducible reports. If instead you want a variable that contains the overall mean on every row, use:

egen avg_income = mean(income)

Weighted means in Stata

In many datasets, a simple unweighted mean is not the correct statistic. Survey data, administrative files, and aggregated records often require weights. Stata supports several weight types. A common example is:

mean income [aw=weightvar]

Always verify which weight type is appropriate for your data and research design. Using the wrong weight can produce a misleading estimate. If you are working with complex survey data, use Stata’s survey commands after setting the design with svyset.

Common mistakes when calculating means in Stata

• Calculating a mean on a string variable instead of a numeric variable.
• Ignoring missing values and failing to check how many observations were excluded.
• Interpreting a mean for a heavily skewed variable without also checking the median.
• Forgetting to use weights when the data require them.
• Using grouped means without verifying whether the grouping variable is coded correctly.
• Collapsing the dataset before saving the original file.

When mean is the wrong summary statistic

The mean is powerful, but it is not always the best measure of center. For strongly skewed variables such as household income, medical spending, or home prices, a few very large observations can pull the mean upward. In those cases, analysts often compare the mean with the median and selected percentiles. In Stata, a practical command is:

tabstat income, statistics(mean median p25 p75 sd n)

If the mean and median differ substantially, that is a sign the distribution may be skewed. It does not make the mean wrong, but it does mean you should interpret it carefully and perhaps report multiple statistics.

Recommended workflow for professional analysis

1. Confirm the variable is numeric with describe.
2. Inspect coding and missing values with codebook or misstable summarize.
3. Calculate the mean with summarize or mean.
4. Compare with the median and spread if the variable may be skewed.
5. Compute group means if your research question is comparative.
6. Save the result in a scalar, matrix, or generated variable if needed downstream.
7. Document the exact command in your do-file for reproducibility.

Authoritative references and further learning

For stronger statistical grounding and command guidance, review these authoritative resources: UCLA Statistical Methods and Data Analytics for Stata, NIST Engineering Statistics Handbook, and Princeton University Stata training materials.

Bottom line

To calculate the mean of a variable in Stata, the simplest command is usually summarize variable_name. If you need confidence intervals, use mean. If you need group-specific means stored in a new variable, use egen with bysort. If you need a publication-friendly table, use tabstat. The mechanics are simple, but good analysis depends on checking missing values, understanding the variable distribution, and choosing the right command for the job. Once you master those pieces, mean calculation in Stata becomes a fast, dependable part of your workflow.