Standard deviation appears in statistics textbooks, scientific papers, financial reports, quality control data, and exam grade distributions — yet many people have only a vague sense of what it actually means. "It's something to do with spread" is about as far as most recollections go. That's a shame, because once you genuinely understand it, standard deviation becomes one of the most useful statistical tools in your mental toolkit.
What Is Standard Deviation?
Standard deviation (usually abbreviated to SD or σ — that's the lowercase Greek letter sigma) measures how spread out a set of data is around its average (mean). A small standard deviation means the data points cluster closely around the mean. A large standard deviation means they're spread widely.
For example: two classes both average 65% on an exam. Class A has scores ranging from 60-70%. Class B has scores from 20-100%. Both have the same average — but vastly different standard deviations. This distinction matters enormously when making decisions based on data.
You can calculate standard deviation using our grade calculator for academic results, or apply the formula below to any dataset.
The Formula (Step by Step)
Let's use a simple dataset: exam scores of 5 students: 55, 60, 70, 75, 90
Step 1: Find the mean (average)
(55 + 60 + 70 + 75 + 90) ÷ 5 = 350 ÷ 5 = 70
Step 2: Find the difference between each value and the mean, then square it
- 55 − 70 = −15 → squared: 225
- 60 − 70 = −10 → squared: 100
- 70 − 70 = 0 → squared: 0
- 75 − 70 = 5 → squared: 25
- 90 − 70 = 20 → squared: 400
Step 3: Find the mean of these squared differences (this is the variance)
(225 + 100 + 0 + 25 + 400) ÷ 5 = 750 ÷ 5 = 150
Step 4: Take the square root of the variance
√150 ≈ 12.25
Standard deviation = 12.25. This tells us that, on average, scores in this group differ from the mean by about 12 marks.
Population vs Sample Standard Deviation
Here's a subtlety that trips people up. In step 3, we divided by n (the number of data points). This is the population standard deviation — used when your dataset is the entire population you're measuring.
When your dataset is just a sample of a larger population (which it usually is in practice), you divide by n−1 instead. This is the sample standard deviation — it's a slightly better estimate of the true population spread when working from limited data. In Excel or most calculators, the STDEV function uses n−1 (sample). STDEVP uses n (population).
Our percentage calculator can help with related statistical work when comparing figures.
The 68-95-99.7 Rule (The Normal Distribution)
When data follows a normal distribution (that classic bell-curve shape), standard deviation has a beautiful property:
- 68% of data falls within 1 standard deviation of the mean
- 95% falls within 2 standard deviations
- 99.7% falls within 3 standard deviations
With a mean of 70 and SD of 12.25, 68% of students would score between 57.75 and 82.25; 95% would score between 45.5 and 94.5.
This rule is why "within two standard deviations" is often used in quality control and scientific research as the boundary for "normal" or "acceptable" variation.
Real-World Uses of Standard Deviation
- Finance: measures investment risk — higher SD means more volatile returns
- Manufacturing: quality control checks that product dimensions stay within acceptable SD of the target
- Medicine: determines whether a clinical result is statistically different from normal
- Education: standardised test scoring and grade normalisation
- Weather: comparing how variable temperatures are in different locations
Once you understand that standard deviation is simply "how spread out" a dataset is, you'll start noticing it everywhere — and understanding data discussions far more clearly as a result.
Further reading: Khan Academy's statistics section has excellent visual explainers for standard deviation. Learn standard deviation visually at Khan Academy.