Key Takeaways
- Standard deviation tells how spread out your data is from the mean (average value).
- You can calculate it with a simple pattern: mean → differences → squares → average → square root.
- Use sample (n−1) or population (N) based on what your data represents.
Standard deviation can feel confusing because it has many steps and symbols. You’re not behind. You just need a clear method. You’ll learn a simple process and see worked examples.
Quick answer
Standard deviation is a number that shows how far data values usually are from the mean (average value).
Mini-steps: (1) find the mean, (2) subtract the mean from each value, (3) square, (4) average (use n−1 for a sample), (5) square root.
Core idea:
- Variance = average of squared distances from the mean.
- Standard deviation = square root of variance.
What standard deviation means
- Mean (average value): add the numbers, divide by how many numbers.
- Standard deviation (how spread out the data is): bigger means “more spread,” smaller means “more clustered.”
A quick interpretation rule
If the standard deviation is 0, all values are the same.
If it is large, scores vary a lot (some low, some high).
👉 Use our Standard Deviation Calculator to compute sample or population standard deviation from your data in seconds, then come back here to understand what the result means.
Symbols you’ll see
| Symbol | Meaning | When used |
|---|---|---|
| x | a data value | always |
| x̄ | sample mean (average) | sample problems |
| μ | population mean | population problems |
| n | number of values in a sample | sample problems |
| N | number of values in a population | population problems |
| s | sample standard deviation | sample problems |
| σ | population standard deviation | population problems |
OpenStax uses s for sample standard deviation and σ for population standard deviation.
Sample vs population (pick the right one)
This choice changes the “divide by” step.
- Population: you have every value in the group you care about → divide by N.
- Sample: you have only some values, and you want to estimate the whole group → divide by n − 1.
Excel matches this idea:
- STDEV.S uses n−1 (sample).
- STDEV.P uses N (population).
Why sample standard deviation divides by n−1
When you compute the mean from a sample, the data gets “pulled toward” that mean.
So if you divide by n, the spread you compute is usually a bit too small.
Dividing by n−1 fixes that by making the spread a little bigger, so it’s a better estimate of the population spread.
Think: sample = guess the whole group → need a small correction.
Why population standard deviation divides by N
If you truly have the whole population, you are not estimating anything.
You are describing the real spread of that actual group.
So you divide by the actual count: N.
Think: population = no guessing → no correction needed.
Common confusion
Confusion: “Why not always use n−1?”
Fix: Because if you truly have the whole population, using n−1 would over-correct and make the spread too big.
Step-by-step: How to calculate standard deviation (by hand)
The beginner-safe process
- Find the mean (average value).
- Find each deviation: value − mean. (Deviation = how far a value is from the mean.)
- Square each deviation. (Squaring makes negatives become positive.)
- Add the squared deviations.
- Divide:
- sample: divide by n − 1
- population: divide by N
This result is variance.
- Square root the variance. That is the standard deviation.
What to write on your paper (mini-checklist)
- Mean = ____
- Deviations: (x − mean) for each x
- Squared deviations: (x − mean)²
- Sum of squares = ____
- Variance = ____ ÷ (n−1 or N)
- Standard deviation = √(variance)
Quick self-check (fast and useful)
- Standard deviation is never negative.
- If your data values are close together, SD should feel small.
- If you add the same constant to every value (like +10), SD stays the same (spread didn’t change).
Worked examples (step-by-step)
Example 1 (very easy)
Data: 5, 5, 5, 5
- Mean = 5
- Deviations: 0, 0, 0, 0
- Squared deviations: 0, 0, 0, 0
- Sum = 0
- Variance = 0
- Standard deviation = √0 = 0
Why: no spread at all.
Example 2 (typical homework level, sample SD)
Data: 2, 4, 4, 4, 5, 5, 7, 9 (treat as a sample)
- Mean: (2+4+4+4+5+5+7+9) ÷ 8 = 40 ÷ 8 = 5
- Deviations (x − 5): −3, −1, −1, −1, 0, 0, 2, 4
- Squares: 9, 1, 1, 1, 0, 0, 4, 16
- Sum of squares = 9+1+1+1+0+0+4+16 = 32
- Sample variance = 32 ÷ (8−1) = 32 ÷ 7 ≈ 4.571
- Sample standard deviation = √4.571 ≈ 2.14
Interpretation: Values are usually about 2.14 away from the mean (5), give or take. (It’s a “typical distance,” not a rule.)
Example 3 (exam-style: same mean, different spread)
Set A: 4, 5, 5, 6
Set B: 1, 5, 5, 9
Both have mean = 5.
But Set B is more spread out, so Set B has a bigger standard deviation.
That is what standard deviation is for: same average, different spread.
Practice set (with answer key)
Directions
For each set, find the mean and the sample standard deviation (s). Round SD to 2 decimals.
- 3, 3, 3, 3
- 1, 2, 3, 4, 5
- 10, 12, 14, 14
- 5, 7, 7, 7, 9
- 8, 9, 9, 10, 14
- (Trap) 2, 2, 2, 20
Answer key (SD is sample SD):
- mean 3; s = 0
- mean 3; s ≈ 1.58
- mean 12.5; s ≈ 1.92
- mean 7; s ≈ 1.41
- mean 10; s ≈ 2.35
- mean 6.5; s ≈ 8.05 (big spread because of 20)
Short solutions (2 items only):
- #2: deviations from mean=3 are: −2, −1, 0, 1, 2; squares sum = 10; variance = 10/4 = 2.5; SD = √2.5 ≈ 1.58.
- #6: one outlier (20) makes squares huge, so SD jumps a lot.
Common mistakes
| Mistake | Why it happens | Do this instead |
|---|---|---|
| Using n when it should be n−1 | You forgot sample vs population | If it’s a sample, divide by n−1. |
| Forgetting to square root at the end | You stopped at variance | Variance first, then SD = √variance. |
| Getting a negative SD | You averaged deviations (not squared) | Square deviations so negatives don’t cancel. |
| Rounding too early | Early rounding changes the final SD | Keep more digits until the end. OpenStax warns this improves accuracy. |
| Mixing up variance vs SD | They sound similar | Variance is in “squared units.” SD goes back to the original units. |
If you get stuck…
- “I don’t know what formula to use.”
Ask: “Is this all the data (population) or just some (sample)?” Then pick N or n−1. - “My answer is negative.”
Standard deviation can’t be negative. You probably averaged raw deviations instead of squared deviations. - “I keep mixing up steps.”
Use the 5-word chain: mean → subtract → square → average → root. - “My calculator gives a different value.”
Check rounding. Also check if your calculator is using sample SD or population SD. - “My units don’t match.”
Your SD should have the same unit as the data (points, minutes, cm). Variance is the one with squared units.
If you’re still stuck, review spread basics (range/variance) first, then ask your teacher for one checked example you can copy as a template.
Why standard deviation is used
Standard deviation is used because it describes spread around the mean in one number.
It is a common “measure of dispersion” (dispersion = spread) in many fields, including research and data reports.
If you have a sample, its SD (s) can be used to estimate the population SD (σ) under common sampling conditions.
Next steps
- standard deviation in Excel — quick way to compute SD with spreadsheet functions.
- variance vs standard deviation — learn the difference so you don’t stop one step early.
- measures of dispersion — compare range, IQR, variance, and SD.
How we know
- Definitions align with standard references that describe SD as spread around the mean.
- The step process matches widely used classroom instruction (mean → deviations → square → average → root).
- Sample vs population rules match OpenStax and Microsoft’s function notes (n−1 vs n).
- Rounding guidance matches OpenStax’s accuracy note.
Use this the right way
Use this guide to learn the method and check your understanding. Don’t just copy answers.
If you’re using standard deviation in a research paper, cite your source for definitions and methods, and write your own explanation.
Study tools that can help
- If you get distracted easily: use one clean notebook page per problem and box each step (mean, deviations, squares, etc.).
- If your work gets messy: try a structured notebook like the ones here: Best High-Quality Notebooks for Math Notes.
- If calculations take too long: use a calculator or spreadsheet after you can do one full example by hand.
References
Khan Academy. (n.d.). Standard deviation: calculating step by step. https://www.khanacademy.org/math/statistics-probability/summarizing-quantitative-data/variance-standard-deviation-population/a/calculating-standard-deviation-step-by-step
OpenStax. (2023). Introductory statistics 2e: 2.7 Measures of the spread of the data. https://openstax.org/books/introductory-statistics-2e/pages/2-7-measures-of-the-spread-of-the-data
