Calculating Standard Deviation

Master the step-by-step process of calculating standard deviation for any data set.

Advanced25 minLesson
Printable worksheet

Definition

Calculating standard deviation follows a systematic 5-step process that measures how spread out data values are from their mean.
The Formula:
σ=∑(xi−xˉ)2n\sigma = \sqrt{\frac{\sum(x_i - \bar{x})^2}{n}}
The 5-Step Process:
  1. 1.Find the mean (xˉ\bar{x}): Add all values and divide by the count
  2. 2.Find each deviation: Subtract the mean from each value (xi−xˉx_i - \bar{x})
  3. 3.Square each deviation: (xi−xˉ)2(x_i - \bar{x})^2
  4. 4.Find the average of squared deviations (variance): ∑(xi−xˉ)2n\frac{\sum(x_i - \bar{x})^2}{n}
  5. 5.Take the square root: σ=variance\sigma = \sqrt{\text{variance}}
Why square the deviations? Squaring makes all deviations positive (so negatives don't cancel positives) and emphasizes larger deviations.

Try it now

What is the FIRST step when calculating standard deviation?

Worked Examples

Find the standard deviation of these test scores: 72,78,80,84,8672, 78, 80, 84, 86
1

Find the mean

xˉ=72+78+80+84+865=4005=80\bar{x} = \frac{72 + 78 + 80 + 84 + 86}{5} = \frac{400}{5} = 80 → Mean = 8080
2

Find each deviation from the mean

72−80=−872 - 80 = -8 78−80=−278 - 80 = -2 80−80=080 - 80 = 0 84−80=484 - 80 = 4 86−80=686 - 80 = 6 → Deviations: −8,−2,0,4,6-8, -2, 0, 4, 6
3

Square each deviation

(−8)2=64(-8)^2 = 64 (−2)2=4(-2)^2 = 4 (0)2=0(0)^2 = 0 (4)2=16(4)^2 = 16 (6)2=36(6)^2 = 36 → Squared deviations: 64,4,0,16,3664, 4, 0, 16, 36
4

Find the variance (average of squared deviations)

Variance=64+4+0+16+365=1205=24\text{Variance} = \frac{64 + 4 + 0 + 16 + 36}{5} = \frac{120}{5} = 24 → Variance = 2424
5

Take the square root

σ=24≈4.90\sigma = \sqrt{24} \approx 4.90 → SD ≈4.90\approx 4.90

Common Mistakes

Forgetting to square the deviations

Why it's wrong: Without squaring, negative and positive deviations cancel out, giving a sum near zero regardless of actual spread.

Correct: Always square each deviation. The sum of squared deviations for a spread-out data set will be large.

Dividing by n−1n-1 instead of nn (or vice versa)

Why it's wrong: Population SD uses nn; sample SD uses n−1n-1. Using the wrong formula gives incorrect results.

Correct: Use nn when you have all data (population). Use n−1n-1 when you have a sample of a larger population.

Forgetting to take the square root at the end

Why it's wrong: Stopping at variance gives units that are squared (like 'points squared'), which is hard to interpret.

Correct: Always take the square root of variance to get standard deviation in the original units.

Making arithmetic errors when summing squared deviations

Why it's wrong: With many values, it's easy to make calculation mistakes.

Correct: Create a table with columns: Value, Deviation, Squared Deviation. Check each calculation.

Interactive Visual

Dot Plot

Test scores: 72, 78, 80, 84, 86

7274767880828486

Each dot is one value (5 values in total).

Mean = 80

Tap a column to count its dots.

Dot Plot

Set B: 2, 6, 10, 14, 18

24681012141618

Each dot is one value (5 values in total).

Mean = 10

Tap a column to count its dots.

Interactive Sandbox

Expression Calculator

Try these:

History

No calculations yet

Practice Problems

15 problems
Problem 1 of 15
Easy
What is the FIRST step when calculating standard deviation?

Why It Matters

Knowing how to calculate standard deviation by hand is essential for:
  • Understanding the concept: The calculation reveals WHY standard deviation measures spread
  • Checking calculator results: Verify that technology gives correct answers
  • Exams and tests: Many standardized tests require manual calculation
  • Small data sets: Quick mental calculation without technology
Real applications:
  • Quality control: A factory calculates SD to check if products are consistently sized
  • Sports analytics: Compare player consistency using calculated SD of performance
  • Finance: Calculate investment risk by finding SD of returns

Real World Applications

Quality Control in Manufacturing

Factory managers calculate standard deviation to ensure products meet specifications consistently.

Example:

A factory produces bolts that should be 1010 mm long. If σ=0.1\sigma = 0.1 mm, 95% of bolts are within ±0.2\pm 0.2 mm of target. If σ=0.5\sigma = 0.5 mm, there's too much variation.

1Try It Yourself

Five bolts measured: 9.8,10.1,10.0,9.9,10.29.8, 10.1, 10.0, 9.9, 10.2 mm. The target is 1010 mm.

Calculate the standard deviation. Is the production consistent?

Step 1: Write the mathematical expression

First find the mean of the measurements:

Comparing Student Performance

Teachers calculate SD to understand how varied student scores are on tests.

Example:

If a class has a mean of 75 with SD = 5, most students score between 70 and 80. If SD = 15, scores range widely from 60 to 90.

2Try It Yourself

Quiz scores: 8,6,9,7,108, 6, 9, 7, 10. Calculate the standard deviation.

How much do scores typically vary from the average?

Step 1: Write the mathematical expression

Mean = (8+6+9+7+10)/5(8+6+9+7+10)/5:

Key Takeaways

  • 1Standard deviation calculation follows 5 steps: mean → deviations → square → average → square root
  • 2Squaring deviations prevents negative values from canceling positive ones
  • 3Variance is the average of squared deviations; SD is its square root
  • 4Population SD divides by nn; sample SD divides by n−1n-1
  • 5A higher SD means more spread; SD = 0 means all values are identical

Frequently Asked Questions

Use nn (population SD) when your data includes the entire group you're studying. Use n−1n-1 (sample SD) when your data is a sample from a larger population. The n−1n-1 correction gives a better estimate of the true population spread.
Use nn (population SD) when your data includes the entire group you're studying. Use n−1n-1 (sample SD) when your data is a sample from a larger population. The n−1n-1 correction gives a better estimate of the true population spread.
Squaring has mathematical advantages: it's differentiable (useful in calculus), gives more weight to larger deviations, and connects to the Pythagorean theorem and Euclidean distance. Absolute values work but don't have these properties.
No! Since we're taking the square root of a sum of squared numbers, standard deviation is always zero or positive. SD = 0 only when all values are identical.

Glossary

Standard deviation
A measure of how spread out data values are from the mean
Variance
The average of squared deviations; the square of standard deviation
Deviation
The difference between a data value and the mean: xi−xˉx_i - \bar{x}
Mean
The average of all data values: sum divided by count
Population
The complete set of all individuals or items being studied
Sample
A subset of a population used to represent the whole

Formula Card

Population Standard Deviation

σ=∑(xi−xˉ)2n\sigma = \sqrt{\frac{\sum(x_i - \bar{x})^2}{n}}

Use when data represents the entire population

Sample Standard Deviation

s=∑(xi−xˉ)2n−1s = \sqrt{\frac{\sum(x_i - \bar{x})^2}{n-1}}

Use when data is a sample from a larger population

Variance

σ2=∑(xi−xˉ)2n\sigma^2 = \frac{\sum(x_i - \bar{x})^2}{n}

The square of standard deviation (before taking the root)

Today's challengeOne problem, three tries

More in This Topic

Range and Interquartile Range

Intermediate25 min

Standard Deviation

Advanced25 min

Identifying Outliers

Intermediate25 min