Excel KURT Function

Q: Does Excel's KURT function return excess kurtosis?

Yes. KURT returns excess kurtosis, which subtracts 3 from the raw Pearson kurtosis. This means a normal distribution scores approximately 0, so you interpret results relative to 0 rather than relative to 3.

The Excel KURT function calculates the kurtosis of a dataset, a measure of how heavy-tailed or peaked the data are relative to a normal distribution. Type =KURT(number1, [number2], ...) and pass at least four numeric values, and Excel returns the excess kurtosis — so a perfectly normal distribution scores roughly 0.

Kurtosis is a measure of whether the data are heavy-tailed or light-tailed relative to a normal distribution. In simpler terms, it helps to understand the extremity of outlier values in a dataset. With this function, Excel users can effectively gauge the peakedness of data, which is crucial in fields such as finance, research, and quality control. At LearnExcel.io, we aim to provide trusted advice that demystifies complex functions like KURT, making them accessible for everyday use.

This function falls under Excel Statistics Functions, the family of tools that also includes STDEV for standard deviation and the closely related SKEW function for measuring asymmetry.

Excel KURT Syntax

=KURT(number1, [number2], ...)

The KURT function syntax includes several arguments, where number1, number2, etc., represent the set of values for which you want to calculate the kurtosis. You can input up to 255 arguments, allowing for the analysis of large datasets. In practice you usually pass a single cell range, for example =KURT(A2:A101).

Excel KURT Parameters

Each parameter in the KURT function represents a value in your dataset:

number1: This is required and represents the first number (or range) in your dataset.
[number2], …: These are optional. You can have up to 254 additional values. Each represents another number or range in your dataset.

It’s crucial that your dataset has at least four values because the function cannot calculate kurtosis with fewer data points — fewer than four returns a #DIV/0! error. Also, remember that non-numeric values, empty cells, and logical values within a referenced range are ignored.

What Kurtosis Measures and How to Interpret It

Kurtosis describes the tailedness of a distribution — how much of the data sits in the extreme tails versus near the center — and is often informally linked to how sharply peaked the distribution is. Crucially, Excel’s KURT function returns excess kurtosis, meaning it subtracts 3 from the raw (Pearson) kurtosis so that a normal distribution is used as the baseline. That gives you three cases to read:

Positive value (leptokurtic): heavier tails and a sharper peak than a normal distribution. Expect more extreme outliers — common in financial returns.
Negative value (platykurtic): lighter tails and a flatter peak than a normal distribution. Outliers are rarer than in a normal bell curve.
Approximately zero (mesokurtic): the dataset’s tails behave roughly like a normal distribution.

Because Excel reports excess kurtosis, you compare the result against 0 rather than against 3. A KURT value of 0.0 does not mean “no kurtosis” — it means kurtosis comparable to a normal distribution.

Return Value

The KURT function returns the excess kurtosis of the dataset. A high positive value indicates a dataset with more outlier values (heavy tails), whereas a negative value suggests fewer outliers (light tails). The return value helps in understanding the distribution’s tail behavior, which is a fundamental aspect of statistical analysis and pairs naturally with summary measures like the mean and standard deviation.

Worked Example

Consider a dataset entered in cells A2:A11: 3, 4, 5, 2, 3, 4, 5, 6, 4, 7. To calculate its kurtosis, you would use either of these equivalent formulas:

=KURT(A2:A11)

=KURT(3, 4, 5, 2, 3, 4, 5, 6, 4, 7)

This returns a small negative value (around -0.6), telling you the distribution is slightly flatter than a normal curve — its tails carry fewer extreme values than a bell curve would. If you swapped one of the middle numbers for a far-off value such as 25, the result would jump well above 0, reflecting the new heavy tail. That sensitivity to extreme points is exactly what makes KURT useful for spotting outlier-prone data.

KURT vs SKEW

KURT and SKEW are complementary shape measures, and analysts usually report them together:

KURT answers how heavy are the tails? — it measures peakedness and the prevalence of outliers, regardless of which side they fall on.
SKEW answers which way does the distribution lean? — it measures asymmetry, with positive skew indicating a longer right tail and negative skew a longer left tail.

A dataset can be highly skewed but have near-normal kurtosis, or symmetric (zero skew) yet sharply peaked (high kurtosis). Reading both alongside the mean absolute deviation or the correlation coefficient gives a fuller picture of your data than any single statistic alone.

Use Cases

Common use cases for the KURT function include:

Financial analysis to assess the risk of extreme deviations in returns.
Quality control to identify when a process may be producing items with extreme characteristics.
Research to understand the distribution of data collected in experiments.

Best practices involve ensuring your dataset is large enough to warrant a kurtosis calculation and remembering that the KURT function is sensitive to outliers.

Troubleshooting Common Errors

Users may encounter errors if:

#DIV/0! — the dataset has fewer than four values, or all values are identical (zero variance). Kurtosis cannot be calculated in either case. Add more data points or confirm the values actually vary.
#VALUE! — this typically appears when you type a non-numeric value directly as an argument, for example =KURT(3, "x", 5, 6). Note that text inside a referenced range is ignored rather than throwing an error; the #VALUE! error comes from non-numeric arguments passed directly to the function.
Unexpected results — non-numeric values, blank cells, and logical values inside a range are silently skipped, so a range that looks like it has enough entries may have fewer than four valid numbers. Verify your range contains only the numeric data you intend.

To avoid these issues, always ensure your dataset is correctly prepared and contains at least four numeric values.

Compatibility

The KURT function is available in all current versions of Excel, ensuring broad compatibility. It also sits within the wider toolkit of Excel math functions and statistics functions you can combine for analysis. For the best experience and to avoid unexpected errors, always ensure your version of Excel is updated.

Frequently Asked Questions

Does Excel’s KURT function return excess kurtosis?

Yes. KURT returns excess kurtosis, which subtracts 3 from the raw Pearson kurtosis. This means a normal distribution scores approximately 0, so you interpret results relative to 0 rather than relative to 3.

How many data points does KURT need?

At least four numeric values. With three or fewer values — or when every value is identical — KURT returns a #DIV/0! error because the kurtosis formula divides by terms that become zero.

What does a positive KURT value mean?

A positive value (leptokurtic) means the distribution has heavier tails and a sharper peak than a normal distribution, so extreme outliers are more likely. This is common in datasets like financial returns.

What does a negative KURT value mean?

A negative value (platykurtic) means the distribution has lighter tails and a flatter peak than a normal distribution. Extreme outliers are rarer than they would be in a normal bell curve.

What is the difference between KURT and SKEW in Excel?

KURT measures tailedness and peakedness (how outlier-prone the data are), while SKEW measures asymmetry (which direction the distribution leans). They describe different aspects of distribution shape and are usually reported together.

Why does KURT ignore some cells in my range?

KURT skips text, blank cells, and logical (TRUE/FALSE) values inside a referenced range. If too many cells are skipped, you can drop below the four-value minimum and trigger a #DIV/0! error even though the range appears full.

Conclusion

Understanding the KURT function in Excel allows for deeper statistical analysis of your data, providing insights into the presence and extremity of outliers. Whether you’re analyzing financial risks, performing quality control, or conducting research, KURT can be an invaluable tool in your analytical arsenal. Remember, practice makes perfect. We encourage you to experiment with this function in your own spreadsheets to grasp its full potential. Happy analyzing!