Calculate Mean Of Variables In Stata

Use this premium calculator to compute the mean from a list of values, then apply the same logic inside Stata with the commands explained below.

Mean Calculator

Chart Preview

The chart plots each observation and overlays the overall mean, making it easy to see which values sit above or below the average.

In Stata, this type of visual check is useful after computing descriptive statistics so you can connect the numeric mean to the shape of the underlying data.

How to Calculate Mean of Variables in Stata

If you need to calculate the mean of variables in Stata, the good news is that Stata gives you several reliable ways to do it. The right method depends on what you mean by “variables.” In some projects, you want the mean of one variable across all observations. In others, you want the mean for multiple variables side by side. In still other cases, you want a row wise mean across several variables for each observation. Those are similar statistical ideas, but the commands are not identical. Understanding that distinction is the key to working efficiently and avoiding mistakes.

At its core, the mean is simply the arithmetic average. You add all valid values and divide by the number of valid observations. In Stata, missing values are usually excluded from mean calculations unless you explicitly recode or replace them. That default is extremely helpful, because it prevents empty or missing data from artificially lowering or raising the result. If you are summarizing survey responses, exam scores, sales totals, or clinical measurements, this behavior generally matches standard statistical practice.

The basic idea is simple: if a variable contains 10, 12, 14, and 16, the mean is 13 because the sum is 52 and there are 4 valid observations.

Most Common Stata Commands for Means

The most common command for quickly checking a mean is summarize. It produces the number of observations, the mean, the standard deviation, and the minimum and maximum values. If your goal is a fast descriptive summary, this is usually the best starting point.

summarize income

If you want means for several variables at once, you can list them together:

summarize income age hours_worked

Stata will display each variable on its own row with a mean in the output table. For many analysts, this is the fastest way to inspect the center of multiple measures before moving on to regressions or visualizations.

Another useful command is mean. It is designed more directly for mean estimation and can be especially useful when you want confidence intervals or more formal output.

mean income

You can also apply conditions with if or subsets with in:

mean income if gender == 1 summarize income if region == “West”

This is essential when you need subgroup averages. For example, you may want the mean wage for full time workers only, or the mean blood pressure among patients over age 50. Stata makes this straightforward.

Calculating the Mean of Multiple Variables

There are two related but different tasks analysts often confuse:

• Column mean: the mean of each variable across all observations.
• Row mean: the average across several variables for each observation.

For column means, use summarize, mean, or tabstat. For row means, use egen with rowmean().

egen exam_avg = rowmean(test1 test2 test3 test4)

This creates a new variable called exam_avg that contains the average of the listed variables for each row. This is especially common in education, survey scoring, finance, and healthcare datasets where multiple related indicators need to be combined into one composite average.

One major advantage of egen rowmean() is that it handles missing values sensibly. If one score is missing but the others are present, Stata will compute the average from the available values. If all included variables are missing for that observation, the row mean remains missing.

Worked Example with Real Numeric Output

Suppose you have the following five observations for a variable named sales. The mean is computed as the total divided by the count of nonmissing values.

Observation	Sales	Running Total
1	120	120
2	135	255
3	150	405
4	145	550
5	130	680

The total is 680, and there are 5 observations, so the mean is 136.0. In Stata, either of the following would return that result:

summarize sales mean sales

Now imagine a student dataset with four test variables: test1, test2, test3, and test4. If one student scored 80, 75, 90, and 95, then the row wise mean is 85.0. In Stata, you would compute it like this:

egen student_avg = rowmean(test1 test2 test3 test4)

This creates a new variable you can summarize later:

summarize student_avg

Comparison of Stata Commands for Mean Calculation

Task	Recommended Command	What It Returns	Example Result
Mean of one variable	summarize sales	Count, mean, standard deviation, min, max	Mean = 136.0
Mean of one variable with formal estimation	mean sales	Mean plus standard error and confidence interval	Mean = 136.0
Means for several variables	summarize sales profit expenses	One row per variable	Sales = 136.0, Profit = 28.4
Row wise mean across variables	egen avg = rowmean(test1 test2 test3)	New variable storing row average	Obs 1 average = 81.7
Grouped mean	bysort region: summarize sales	Separate mean by group	North = 142.3, South = 128.8

Grouped Means in Stata

Very often, you do not want one overall mean. You want the mean within categories such as region, gender, treatment status, year, or department. There are multiple ways to do this. One common option is to combine bysort with summarize.

bysort region: summarize sales

This tells Stata to sort by region and then produce summary statistics separately for each subgroup. If you want a cleaner table focused on means, tabstat is excellent.

tabstat sales, by(region) statistics(mean n sd)

This produces a compact output that is often easier to present in a report. When you are comparing categories, using tabstat can save time and improve readability.

Means with Conditions

You can calculate a mean for only part of the dataset by using an if qualifier. This is especially useful in data cleaning and subgroup analysis.

summarize income if age >= 25 & age <= 54 mean bmi if smoker == 1

These commands restrict the calculation to the observations that satisfy the condition. This is a better practice than manually deleting rows or creating temporary files when you only need a conditional average.

Weighted Means

In some datasets, observations do not all represent the same amount of information. Survey microdata, for example, often require weights because one respondent may represent many people in the population. In that case, the simple unweighted mean may not be appropriate. Stata supports several weight types depending on the design of your data.

mean income [aw=weightvar] summarize income [aw=weightvar]

Before applying weights, always check the survey documentation. Weight choice is a methodological decision, not just a coding detail. If you work with public data, review official guidance from sources such as the U.S. Census Bureau or the CDC NHANES weighting tutorial. For Stata learning support, UCLA provides a widely used resource at UCLA Statistical Methods and Data Analytics.

How Missing Values Affect the Mean

Missing values are one of the most important issues in mean calculation. Stata generally excludes missing observations from the denominator. That is usually what you want, but not always. If a missing value actually means zero in your study design, then leaving it as missing would overstate the mean. On the other hand, replacing genuine missing values with zero can badly distort results. The right choice depends on the data generating process.

For example, suppose a spending variable contains 100, 150, and a missing value. If the missing entry means “not reported,” the proper mean across valid values is 125. If the missing entry actually means “no spending,” then the substantively correct average might be 83.33 after recoding the missing to zero. Stata can handle either approach, but the analyst must make the decision based on domain knowledge.

Using collapse to Create Mean Based Datasets

If your goal is not just to display means but to transform the dataset into aggregated averages, use collapse. This replaces the current dataset with grouped summary values, so it should be used carefully.

collapse (mean) sales profit, by(region year)

After running this command, each remaining row represents a region year combination, and the variables contain means rather than original observation level data. This is powerful for building summary datasets, dashboards, or chart inputs.

Common Mistakes When Calculating Means in Stata

1. Confusing row means with column means. Use summarize or mean for variable means across observations, and egen rowmean() for means across variables within each row.
2. Ignoring missing data rules. Missing values are excluded by default, which is often correct but should still be verified.
3. Using the wrong subgroup filter. Small syntax mistakes in an if condition can produce a valid but unintended mean.
4. Forgetting weights in survey data. An unweighted mean may be misleading when the dataset requires design weights.
5. Overwriting the dataset with collapse. Always save a copy first or use preserve and restore.

Best Practice Workflow

• Inspect the variable with codebook or summarize, detail.
• Check for missing values and impossible values.
• Run summarize for a quick mean.
• Use mean if you want inferential output.
• Use egen rowmean() for averages across several variables in one row.
• Use tabstat or bysort when comparing group means.
• Document any recoding or weighting choices.

Practical Example You Can Reproduce

Suppose your data contains three performance variables: quality, speed, and accuracy. You want both the overall mean of each variable and a combined score for each employee. A clean workflow would be:

summarize quality speed accuracy egen performance_avg = rowmean(quality speed accuracy) summarize performance_avg tabstat performance_avg, by(department) statistics(mean n sd)

This sequence gives you variable level means, an employee level composite mean, and department level comparisons. That is a strong pattern for operational analytics, HR scorecards, student assessment files, and many business intelligence tasks.

Final Takeaway

To calculate the mean of variables in Stata, start by deciding whether you need a mean across observations, a mean by subgroup, or a row wise mean across several variables. For a quick average, use summarize. For formal mean estimation, use mean. For a new variable based on the average of several columns, use egen rowmean(). For grouped summaries, use bysort or tabstat. Once you understand those use cases, Stata becomes extremely efficient for descriptive analysis.

The calculator above helps you verify the arithmetic behind the mean before you run your Stata code. Enter sample numbers, inspect the chart, and compare the computed average to the command patterns shown here. That combination of numerical intuition and command level fluency is what makes your analysis both faster and more accurate.