Standard Deviation Calculator
What This Means
Your Data, Visualized
Each dot is one of your data points. The solid line marks the mean; the shaded bands mark 1 and 2 standard deviations from it.
If Your Data Is Roughly Bell-Shaped (Empirical Rule)
This only applies if your data follows a roughly normal distribution — treat it as a rough guide, not a guarantee.
- About 68% of values fall within
- About 95% of values fall within
- About 99.7% of values fall within
I have spent close to twelve years helping students and researchers get their statistics right. This is mostly at the dissertation and thesis stage, where a wrong formula does not just cost marks. It costs months. The standard deviation calculator on this page exists to stop exactly that from happening.
Most calculators online just give you a number and leave you to figure out the rest on your own. That is not how I have built the tool here. Enter your data above, pick sample or population, and you get the full working, not just the final figure. Below, I will walk you through what the number means, how it is actually calculated by hand, and where I see students and even working professionals get it wrong.
What Is Standard Deviation?
Standard deviation is a measure of how spread out a set of numbers is around its average value. A small standard deviation means your data points sit close to the mean. A large one means they are scattered further away.
That is the same definition used by NIST’s own Measures of Scale reference, one of the most widely cited statistics handbooks among practitioners. It is also, in my experience, the part most textbooks overcomplicate before they get to how to calculate standard deviation itself.
How to Use This Standard Deviation Calculator
Using the calculator on this page takes less than a minute, even if your data set has fifty or a hundred values in it. No sign-up, no payment, nothing to install. Enter your numbers and you get your answer.
- Enter your numbers. Separate them with commas, spaces, or line breaks. Decimals and negative values are both accepted.
- Choose sample or population. If you are not sure which one applies to your data, read the section further down before you decide. It changes the answer.
- Click calculate. You get the mean, the sum of squared deviations, the variance, and the standard deviation, along with the working for each step.
- Copy the working if you need it. Students preparing assignments or dissertation chapters can lift the step by step calculation directly, which most calculator tools online do not show at all.
Sometimes you already have the mean and the individual data points worked out separately. The standard deviation calculator with mean and data points option on this page skips straight to the deviation and variance steps. You will not have to redo the mean calculation again.
Standard Deviation Calculator With Steps: A Worked Example
Numbers make more sense with a real example, so here is one I use often when teaching this concept. Say a small café tracked the number of customers it served each day for one working week: 4, 8, 6, 5, 3.
The first step is always the same, no matter if you end up computing sample or population standard deviation.
Step 1: Find the mean. (4 + 8 + 6 + 5 + 3) ÷ 5 = 26 ÷ 5 = 5.2
This average is exactly what the Measures of Central Tendency Calculator gives you directly, if you would rather check the mean on its own first.
Step 2: Find each deviation from the mean. 4 − 5.2 = −1.2 8 − 5.2 = 2.8 6 − 5.2 = 0.8 5 − 5.2 = −0.2 3 − 5.2 = −2.2
Step 3: Square each deviation. 1.44, 7.84, 0.64, 0.04, 4.84
Step 4: Add up the squared deviations. 1.44 + 7.84 + 0.64 + 0.04 + 4.84 = 14.8
From here, the path splits depending on whether these five days represent your entire population of interest, or a sample drawn from a bigger period.
Sample Standard Deviation, Step by Step
If these five days are a sample meant to represent a longer trading pattern, divide the sum of squared deviations by n − 1 instead of n.
14.8 ÷ (5 − 1) = 14.8 ÷ 4 = 3.7 (sample variance) √3.7 = 1.92 (sample standard deviation)
This is the version to use any time your data is a subset of a larger group you are trying to draw conclusions about. It covers most survey data, most experiments, and most dissertation samples. If your goal is simply to find sample standard deviation online, without redoing this arithmetic by hand, the calculator above is built for exactly that.
Population Standard Deviation, Step by Step
If Monday to Friday is the entire population you care about, for instance you only ever wanted to know about that one specific week, divide by n instead.
14.8 ÷ 5 = 2.96 (population variance) √2.96 = 1.72 (population standard deviation)
Notice the population figure is always slightly smaller than the sample figure for the same data set. That gap is not a rounding error, and I will explain exactly why it exists in the next section. If you simply want to calculate population standard deviation from data you already know is a complete group, this is the only version of the formula you need.
Sample vs Population Standard Deviation: Which Should You Use?
This is the single most common question I get, and most online guides answer it in one confusing sentence and move on. I want to be more direct about it.
Use population standard deviation only when your data set is the complete group you are describing, with nothing left out. Use sample standard deviation when your numbers are drawn from a larger group and you are using them to estimate something about that bigger group. In academic research, in surveys, and in almost all business data, you are working with a sample, so the sample formula is the safer default.
I have reviewed enough dissertation drafts to tell you that this is where students lose marks most often. They collect data from 150 out of a possible 10,000 customers, then run the population formula because it feels like the “complete” version. It is not. If your data does not cover every single unit of the group you are studying, it is a sample.
Why Sample SD Divides by n − 1 (Bessel’s Correction)
Dividing by n − 1 instead of n is called Bessel’s correction, and here is the plain reasoning behind it, not just the rule. A sample tends to sit a little closer to its own mean than the true population does, purely because the mean was calculated from that same sample. Dividing by a smaller number, n − 1, inflates the result slightly to correct for this and gives you a more honest estimate of the population’s real spread. That n − 1 figure also has a formal name worth knowing: it is your degrees of freedom.
I disagree with how most textbooks present this. They call it a “correction factor” and leave it as a black box formula to memorise. Once you see it as a bias correction rather than an arbitrary rule, it actually sticks in your memory.
Standard Deviation Calculator for Grouped Data
Most calculators online only accept a raw list of numbers, which is a real gap because grouped data is extremely common in survey research and classroom statistics. A proper standard deviation calculator for grouped data needs to account for frequencies, not just distinct values. Here is a small worked example so you can follow along even before typing anything into the tool above.
Say 20 students were asked how many books they read last month, and the answers were recorded as a frequency table:
| Books read (x) | Number of students (f) |
|---|---|
| 1 | 2 |
| 2 | 5 |
| 3 | 8 |
| 4 | 3 |
| 5 | 2 |
Step 1: Find the weighted mean. (1×2 + 2×5 + 3×8 + 4×3 + 5×2) ÷ 20 = 58 ÷ 20 = 2.9
Step 2: Find the squared deviation for each value, then multiply by its frequency. (1−2.9)² × 2 = 7.22 (2−2.9)² × 5 = 4.05 (3−2.9)² × 8 = 0.08 (4−2.9)² × 3 = 3.63 (5−2.9)² × 2 = 8.82
Step 3: Add these weighted squared deviations. 7.22 + 4.05 + 0.08 + 3.63 + 8.82 = 23.8
Treating all 20 students as the full population here: 23.8 ÷ 20 = 1.19, so population SD = √1.19 = 1.09.
How to Enter Grouped Data Correctly
The mistake I see most often with grouped data is students entering only the five distinct values (1, 2, 3, 4, 5) into a plain calculator and ignoring the frequencies entirely. That gives you the standard deviation of five numbers, not of twenty students, and the two answers are never the same. When your data comes as a frequency table, each value needs to be weighted by how many times it occurs before you calculate anything.
How to Interpret Your Standard Deviation Result
Knowing how to interpret standard deviation matters more than knowing how to calculate it, and it is the step most guides rush through. A standard deviation on its own is just a number until you connect it back to your data’s units and context.
A standard deviation of 1.92 customers per day means very little in isolation. Next to a mean of 5.2 customers, it tells you daily footfall swings quite a bit relative to the average.
The 68-95-99.7 Rule Explained
For data that follows a roughly normal, bell shaped distribution, there is a reliable pattern worth knowing. About 68% of values fall within one standard deviation of the mean, about 95% fall within two, and about 99.7% fall within three. This is sometimes called the empirical rule, and you can check it against your own data using the Empirical Rule Calculator.
I would add one caution here that most articles skip entirely. This rule only holds reasonably well for roughly normal data. Apply it to a skewed data set, such as household income or wait times, and the percentages stop being accurate.
What Counts as a “High” or “Low” Standard Deviation
There is no universal cutoff for a “high” standard deviation, and any article that gives you one flat number to compare against is oversimplifying it. What matters is the standard deviation relative to the mean. A useful way to compare this is the coefficient of variation, which is standard deviation divided by the mean, expressed as a percentage. Two data sets can have the exact same standard deviation and mean completely different things, depending on the scale of the numbers involved.
Standard Deviation vs Variance: What’s the Difference
Variance and standard deviation are calculated from the same set of steps, but they answer slightly different questions. Variance is the average of the squared deviations, while standard deviation is simply the square root of variance, brought back to the same unit as your original data.
This is the practical reason standard deviation gets used far more often in reporting. If your data is in rupees, dollars, or kilograms, variance is expressed in rupees-squared or dollars-squared, which is not a unit anyone can intuitively read. Standard deviation converts it back into a number you can actually place next to your mean and interpret.
Standard Deviation in Finance: Portfolio Risk & Stock Volatility
If a standard deviation calculator for finance is specifically what brought you to this page, this section is written for you. In investing, standard deviation goes by another name: volatility. It tells you how much a stock or fund’s returns swing around their own average, which is a direct measure of risk.
Say a stock recorded these monthly returns over six months: 2%, −1%, 3%, 0%, 4%, −2%. The mean return is 1%, and running the sample formula on these deviations gives a sample standard deviation of roughly 2.37% per month, which is the stock’s monthly volatility figure.
A client of mine once compared two mutual funds before investing his year-end bonus. Fund A showed a standard deviation of 4%, Fund B showed 6%, and he was ready to pick Fund A purely on that basis. When we pulled the mean returns, Fund A averaged 3% with that 4% swing, while Fund B averaged 11% with its 6% swing.
Fund B was actually the more efficient risk-reward trade, not the riskier pick his gut suggested. This is exactly why comparing raw standard deviation numbers without their means is a mistake I keep correcting.
How to Calculate Portfolio Standard Deviation
A portfolio standard deviation calculator needs to do more than average two numbers, and this is where most free tools online get it wrong. Portfolio risk depends on covariance, meaning how the two assets move in relation to each other, not just their individual standard deviations. You can check this relationship first using the Covariance Calculator.
Take a two-asset portfolio, split 50-50, where one asset has a standard deviation of 10%, the other 15%, and the correlation between them is 0.3. The naive average of the two SDs would suggest a portfolio risk of 12.5%.
Run the actual portfolio formula, and the real figure comes out closer to 10.2%. The two assets do not move perfectly together, and that imperfect correlation is exactly what reduces overall risk. This is the mathematical reason diversification works, and no simple averaging of individual volatilities will ever show you that.
Standard Deviation for Students & Researchers: Reporting It Correctly
This page works as both a standard deviation calculator for students and a standard deviation calculator for research projects. Just get the sample versus population choice right before you write anything down.
If you are working on a dissertation, thesis, or research paper, standard deviation almost always needs to be reported alongside your mean. Label it clearly as sample or population too, since reviewers do check this. I have sent back more draft chapters than I can count for this exact reason. Either the SD version was not stated, or its decimal precision did not match the rest of the results table.
One client of mine last year ran her entire results chapter using the population formula, on survey data from just 220 respondents out of a target population of several thousand. The numbers were not wrong exactly, but a reviewer flagged the choice immediately, because it signalled a misunderstanding of the sampling design rather than a genuine methodological choice. We corrected the formula and reran the numbers. The change to her conclusions was small, but the credibility damage from that first reviewer flag would have been much harder to undo.
Reporting Standard Deviation in APA Format / Dissertations
APA style expects standard deviation to be reported as SD, typically in parentheses next to the mean, for example M = 5.20, SD = 1.92. Keep your decimal precision consistent across the whole results section, generally two decimal places, following the APA Style numbers and statistics guide unless your discipline’s own convention says otherwise.
If you are also reporting a standardised effect size such as Cohen’s d, note that it uses standard deviation directly in its formula. You can cross check that calculation using the Cohen’s d Calculator.
Standard deviation on its own answers “how spread out is my data.” Once you move to interpreting individual scores against that spread, you are really asking a z-score question. Work that out using the Z-Score Calculator.
FAQ
What’s the difference between sample and population standard deviation?
Population standard deviation covers every single unit in the group you are studying, and divides by n. Sample standard deviation covers a subset used to estimate the wider group, and divides by n − 1, which gives a slightly larger, more conservative result.
Why do we divide by n − 1 for sample standard deviation?
A sample’s own mean sits closer to its own values than the true population mean would. Dividing by the smaller number, n − 1, corrects for this bias and gives a more accurate estimate of the population’s real spread. This adjustment is called Bessel’s correction, and the n − 1 figure is also known as the degrees of freedom.
What does a standard deviation of 0 mean?
It means every value in your data set is identical, with no spread at all. This is rare in real data outside of controlled experiments or constant values.
How do I calculate standard deviation in Excel or Google Sheets?
Use =STDEV.S() for a sample or =STDEV.P() for a population in Excel. Google Sheets uses the equivalent STDEV and STDEVP functions. The underlying maths is identical to the manual steps shown earlier on this page.
What is considered a “good” standard deviation?
It depends entirely on your mean and the scale of your data. Comparing the coefficient of variation across data sets is more meaningful than comparing raw standard deviation figures directly.
How is standard deviation different from standard error?
Standard error describes how much your sample mean would vary if you repeated the study. It is calculated by dividing the standard deviation by the square root of your sample size. The two answer different questions and should not be used interchangeably in a results section.
Can I calculate standard deviation for grouped or frequency table data with this tool?
Yes. Enter each value weighted by its frequency, or use the grouped data option directly if the calculator provides one. Entering only the distinct values without their frequencies gives you the wrong answer, exactly as shown in the worked example above.
How do I check if two data sets have significantly different standard deviations?
Run an F-test for two samples, or Levene’s test if your data is not normally distributed. Both compare variances directly and give you a p-value, which a simple side-by-side comparison of two SD numbers cannot do.
Can standard deviation help reduce variation in a business or production process?
Yes. In quality control, a shrinking standard deviation over time means a process is becoming more consistent. Track it on a control chart alongside the mean, and investigate any point where SD spikes suddenly, since that usually signals a process problem rather than random noise.