How to Check if Data Is Normally Distributed in Excel

Checking if your data follows a normal distribution is a fundamental step in many statistical analyses, but it can feel complicated. It doesn't have to be. This article will guide you through three practical methods to test for data normality directly within Microsoft Excel, using visualizations and simple formulas.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

What is a Normal Distribution (and Why Does it Matter)?

You’ve probably seen a normal distribution before - it's the classic "bell curve." Data that is normally distributed is symmetrical, with the most frequent data points clustering around the average. As you move away from the average, the data points become less common.

In a perfect normal distribution:

• The curve is perfectly symmetrical.
• The mean (average), median (middle value), and mode (most frequent value) are all the same.
• About 68% of the data falls within one standard deviation of the mean, 95% within two, and 99.7% within three.

Why do you need to check for this? Many of the most common statistical tests, like t-tests and ANOVA, are built on the assumption that your data is normally distributed. If you run these tests on data that is highly non-normal (for instance, heavily skewed to one side), your conclusions could be misleading or outright wrong. It’s like trying to put a square peg in a round hole - the results won’t be reliable. Taking a few minutes to check for normality is a small price to pay for confidence in your data analysis.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Method 1: Visual Inspection with a Histogram

The quickest way to get a feel for your data’s distribution is to visualize it with a histogram. This chart groups your data into ranges (called "bins") and displays the frequency of data points in each bin as a bar. If the bars roughly form a bell shape, your data might be normally distributed.

Step 1: Enable the Data Analysis ToolPak

Excel has a powerful but sometimes hidden feature called the Analysis ToolPak. If you haven't used it before, you'll need to enable it first - you only have to do this once.

• Go to File > Options.
• In the Excel Options window, click on Add-ins.
• At the bottom, next to "Manage," make sure Excel Add-ins is selected and click Go....
• Check the box for Analysis ToolPak and click OK. A new "Data Analysis" button will appear on the Data tab of your Excel ribbon.

Step 2: Create the Histogram

Let’s say you have a single column of sales data you want to analyze. Here’s how to create a histogram:

1. Click on the Data tab and then click on Data Analysis.
2. From the list that appears, select Histogram and click OK.
3. In the Histogram dialog box, configure the following:
4. Click OK.

Excel will generate a frequency table and a histogram chart. Examine the chart: does it look like a bell? Does it peak in the middle and slope down symmetrically on both sides? While this method is subjective, it gives you a great initial impression of your data's shape. If the chart is heavily leaning to one side (skewed) or has two peaks (bimodal), it's a clear visual cue that your data is not normal.

Method 2: Calculating Skewness and Kurtosis

If looking at a chart feels a bit too subjective, you can use built-in Excel formulas to get quantitative measures that describe your distribution's shape. The two most important measures are Skewness and Kurtosis.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Understanding Skewness

Skewness measures the asymmetry of your data distribution. A perfectly symmetrical, normal distribution has a skewness of 0.

• Positive Skew: A skewness value greater than 0 means the distribution has a long tail extending to the right. The bulk of the data is on the left side. An example could be individual income, where most people have lower-to-middle incomes and a few have very high incomes.
• Negative Skew: A value less than 0 means the distribution has a long tail to the left. The bulk of the data is on the right side. An example could be retirement age, where most people retire later in life.

In Excel, the formula is simple:

=SKEW(A1:A100)

(Replace A1:A100 with your actual data range.)

A good rule of thumb: If your skewness value is between -0.5 and 0.5, your data is fairly symmetrical. If it's between -1 and -0.5 or 0.5 and 1, it’s moderately skewed. Anything beyond -1 or 1 is considered highly skewed.

Understanding Kurtosis

Kurtosis measures the "tailedness" of your distribution. It indicates how heavy the tails are and how sharp or flat the peak is compared to a normal distribution.

• Zero Kurtosis (Mesokurtic): A normal distribution has a kurtosis of 0.
• Positive Kurtosis (Leptokurtic): A value greater than 0 indicates a "pointy" distribution with a sharp peak and heavy tails, meaning more outliers.
• Negative Kurtosis (Platykurtic): Less than 0 suggests a flatter distribution with fewer outliers.

The Excel formula for kurtosis is:

=KURT(A1:A100)

(Again, replace A1:A100 with your data.)

A good rule of thumb: Kurtosis values between -1 and 1 generally suggest the data's shape is close enough to normal for many statistical tests.

By calculating both Skewness and Kurtosis, you get a quick, numerical assessment. If both are near zero, your data is likely normally distributed.

Method 3: Creating a Q-Q Plot

The Quantile-Quantile (Q-Q) plot is one of the most robust ways to visually check for normality. While it sounds complex, it's simple in principle: it compares the quantiles of your data against those of a theoretical normal distribution. If your data is normal, the points on the plot will form a nearly straight diagonal line.

Excel doesn't have a built-in Q-Q plot feature, but you can create one with a few calculations.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Step-by-Step Guide to a Q-Q Plot in Excel

Assuming your data is in column A (cells A2 to A101, with A1 as header):

1. Rank your data: Sort your data from smallest to largest in column B (copy column A and sort A-Z).
2. Calculate the rank: In column C, assign ranks 1 through n. Enter 1 in C2 and drag down, or fill sequentially.
3. Calculate the percentile rank: For each data point, compute (rank - 0.5) / n. For example, in D2:
4. Find the theoretical Z-scores: In column E, calculate:
5. Create the scatter plot: Highlight your sorted data in column B (Y-values) and Z-scores in column E (X-values). Insert > Chart > Scatter.

Interpreting the Q-Q Plot

You’re not looking for a bell shape here. Instead, you want a straight, diagonal line.

• If points hug the 45-degree line, it suggests normality.
• Deviations, especially at the ends (tails) or the formation of an S or C shape, indicate skewness or heavy/light tails, suggesting non-normality.

What To Do If Your Data Isn't Normal

Finding out your data isn't normal isn’t a dead end. Here are some options:

1. Data Transformation: Apply functions like logarithm, square root, or reciprocal to make data more normal. Often effective with skewed data.
2. Use Non-parametric Tests: Statistical tests that do not assume normality, such as the Mann-Whitney U Test, can be alternatives.
3. Proceed with Caution: For large samples (often N > 30 or N > 50), the Central Limit Theorem states that the sample mean tends to be normally distributed, so some analyses remain valid despite non-normal data. However, always interpret with understanding of assumptions.

Choosing the right approach depends on your data and goals, but testing for normality helps inform your decisions.

Final Thoughts

Confirming your data’s distribution is a key step before more advanced statistical analysis. Excel provides straightforward tools—from histograms and skewness/kurtosis formulas to the visual check of a Q-Q plot—so you can confidently proceed knowing your data's shape.

While these methods are powerful, remember that data collection and cleaning are ongoing processes, and manual checks across platforms can be time-consuming. At Graphed, we built tools to streamline this. By connecting sources like Shopify, Google Analytics, or Salesforce directly, our platform's AI handles integration, visualization, and insights—so you can ask complex questions in plain English and get answers in seconds, not hours.