Written and maintained by CASRAI Editorial Board
Published
On this page
Who this is for: researchers who have a column of numbers in Excel and need to report its mean, spread and shape. After reading you can produce the summary two ways: in one step with the Analysis ToolPak, or cell by cell with worksheet functions that recalculate when the data changes. A short pandas section covers the Python equivalent.
What descriptive statistics in Excel are used for in research
Descriptive statistics summarise a sample: where its values centre (mean, median, mode) and how far they spread (standard deviation, variance, quartiles). Microsoft describes the Analysis ToolPak’s Descriptive Statistics tool as one that “generates a report of univariate statistics for data in the input range, providing information about the central tendency and variability of your data.” Report these before any inferential test so readers can see the data you analysed. For the concepts and how to report them, see descriptive statistics.
Install or access (official source only)
The Descriptive Statistics tool lives in the Analysis ToolPak add-in, which ships with Excel but is switched off by default. Per Microsoft Support (Use the Analysis ToolPak to perform complex data analysis, read 2026-11-08):
- Windows: File > Options > Add-Ins. In the Manage box choose Excel Add-ins, select Go, tick Analysis ToolPak, then OK.
- Mac: Tools > Excel Add-ins, then tick Analysis ToolPak.
- Once loaded, the command appears as Data Analysis on the Data tab.
The worksheet functions in the next section need no add-in. The full add-in walkthrough, including all of its tools, is in Excel Analysis ToolPak.
Step-by-step: summarise a column
Route 1: the Data Analysis tool
- Put the values in one column with a header in the first row.
- Select Data > Data Analysis, choose Descriptive Statistics, then OK.
- Set the input range to the column and choose where the report should go.
- Select OK. Excel writes a block of univariate statistics to the output location.
The tool writes fixed values. If you edit the data later, run it again.
Route 2: worksheet functions
Functions recalculate when the data change, and you choose exactly which statistics appear. The table maps each statistic to its function, using the descriptions in Microsoft’s statistical functions reference (read 2026-11-08).
| Statistic | Excel function | What Microsoft says it does |
|---|---|---|
| Count | =COUNT(A2:A101) |
Counts how many numbers are in the list of arguments |
| Mean | =AVERAGE(A2:A101) |
Returns the average of its arguments |
| Median | =MEDIAN(A2:A101) |
Returns the median of the given numbers |
| Mode | =MODE.SNGL(A2:A101) |
Returns the most common value in a data set |
| Minimum / maximum | =MIN(A2:A101), =MAX(A2:A101) |
Smallest and largest value in a set |
| Sample standard deviation | =STDEV.S(A2:A101) |
Estimates standard deviation based on a sample |
| Sample variance | =VAR.S(A2:A101) |
Estimates variance based on a sample |
| Quartiles | =QUARTILE.INC(A2:A101,1) (use 3 for the third quartile) |
Returns the quartile of a data set, based on percentile values from 0 to 1 inclusive |
| Skewness / kurtosis | =SKEW(A2:A101), =KURT(A2:A101) |
Skewness of a distribution / kurtosis of a data set |
Sample or population? STDEV.S estimates the standard deviation from a sample and uses the n−1 method; it ignores logical values and text in the sample. If your data are the whole population, Microsoft directs you to STDEV.P instead. STDEVA differs by counting logical values and text. Most research samples call for STDEV.S. See standard deviation and Excel statistics functions for more, and how to calculate the median in Excel for a single-statistic walkthrough.
Worked example: three values, Excel and pandas side by side
Take the values 1, 2 and 3. The pandas documentation for DataFrame.describe shows this series returning count 3.0, mean 2.0, std 1.0, min 1.0, 25% 1.5, 50% 2.0, 75% 2.5 and max 3.0. The same values come from Excel’s functions:
| Statistic | Excel formula | Result | pandas describe() row |
|---|---|---|---|
| Count | =COUNT(A1:A3) |
3 | count |
| Mean | =AVERAGE(A1:A3) |
2 | mean |
| Standard deviation | =STDEV.S(A1:A3) |
1 | std |
| Minimum | =MIN(A1:A3) |
1 | min |
| First quartile | =QUARTILE.INC(A1:A3,1) |
1.5 | 25% |
| Median | =MEDIAN(A1:A3) |
2 | 50% |
| Third quartile | =QUARTILE.INC(A1:A3,3) |
2.5 | 75% |
| Maximum | =MAX(A1:A3) |
3 | max |
The match on std holds because both use the sample (n−1) form, and the match on the quartiles holds because QUARTILE.INC is the inclusive method. Check which quartile function you use before comparing numbers across software; a different quartile definition can give different values for small samples.
Descriptive statistics in Python (pandas)
In pandas, DataFrame.describe() (and Series.describe()) generates the summary in one call. By default it analyses numeric columns only and returns count, mean, std, min, 25%, 50%, 75% and max; the percentiles default to 25, 50 and 75, and the 50% row is the median (pandas documentation, read 2026-11-08).
import pandas as pd
s = pd.Series([1, 2, 3])
s.describe()Output and how to read it
- Mean vs median: a mean far from the median suggests a skewed distribution or outliers. See left- vs right-skewed data.
- Standard deviation: read it alongside the mean; it is in the same units as the data.
- Quartiles: the first and third quartiles bracket the middle half of the data; see the five-number summary.
- Static vs live: report block output is a snapshot, while functions update. This is an editorial point about workflow, not a Microsoft statement.
Troubleshooting
- No Data Analysis button: the add-in is not loaded. Repeat the install steps above.
- Text or logical values in the range:
STDEV.Signores them, whereasSTDEVAincludes them, so the two can disagree on the same cells. - Standard deviation looks too small: check you did not use the population function
STDEV.Pon a sample. - pandas shows no text columns: by default
describe()covers numeric columns only.
Free alternatives
| Route | What it gives you | Source |
|---|---|---|
| Worksheet functions in Excel | Individual statistics that recalculate; no add-in needed | Microsoft statistical functions reference |
pandas describe() |
count, mean, std, min, quartiles, max in one call | pandas documentation |
For a statistical package route, see descriptive statistics in SPSS.
FAQ
How do I get descriptive statistics in Excel?
Load the Analysis ToolPak (File > Options > Add-Ins on Windows), then choose Data > Data Analysis > Descriptive Statistics. Alternatively use functions such as AVERAGE, MEDIAN and STDEV.S.
Should I use STDEV.S or STDEV.P?
Use STDEV.S when your data are a sample of a larger population; it uses the n−1 method. Use STDEV.P only when the data are the whole population.
How do I do descriptive statistics in Python?
Call describe() on a pandas DataFrame or Series. It returns count, mean, standard deviation, minimum, quartiles and maximum for numeric columns.
Sources
- Microsoft Support, Use the Analysis ToolPak to perform complex data analysis, read 2026-11-08.
- Microsoft Support, STDEV.S function, read 2026-11-08.
- Microsoft Support, Statistical functions (reference), read 2026-11-08.
- pandas documentation, pandas.DataFrame.describe, read 2026-11-08.
