Standard Deviation Explained: Formula, Steps & Real Examples

Standard deviation tells you how spread out data is from the average. A low standard deviation means data points cluster tightly around the mean; a high one means they’re scattered widely. It’s the most commonly used measure of variability in statistics, finance, science, and quality control.

The Formula: Population vs Sample

Population standard deviation (when you have data for the entire group):

σ = √[Σ(xi - μ)² / N]

Sample standard deviation (when you have a subset of the population):

s = √[Σ(xi - x̄)² / (N - 1)]

The only difference is dividing by N (population) vs N-1 (sample). The N-1 correction (called Bessel’s correction) accounts for the fact that a sample tends to underestimate the true variability of the population.

When to use which:

• Population: Test scores for your entire class, all products in a batch, complete census data
• Sample: Survey of 500 people representing a city of 1 million, subset of manufacturing output

Step-by-Step Calculation

Dataset: Test scores of 7 students: 72, 85, 90, 68, 95, 78, 82

Step 1: Find the mean

x̄ = (72 + 85 + 90 + 68 + 95 + 78 + 82) / 7 = 570 / 7 = 81.43

Step 2: Calculate each deviation from the mean, then square it

Value	Deviation (xi - x̄)	Squared
72	-9.43	88.93
85	3.57	12.74
90	8.57	73.44
68	-13.43	180.37
95	13.57	184.14
78	-3.43	11.76
82	0.57	0.33

Step 3: Sum the squared deviations

Σ = 88.93 + 12.74 + 73.44 + 180.37 + 184.14 + 11.76 + 0.33 = 551.71

Step 4: Divide by N-1 (sample) and take the square root

s = √(551.71 / 6) = √91.95 = 9.59

The standard deviation is 9.59 points, meaning scores typically fall about 9.6 points above or below the mean of 81.43.

Interpreting Standard Deviation: The 68-95-99.7 Rule

For normally distributed data (bell curve):

• 68% of data falls within 1 SD of the mean
• 95% falls within 2 SDs
• 99.7% falls within 3 SDs

In our example (mean 81.43, SD 9.59):

• 68% of students score between 71.84 and 91.02
• 95% score between 62.25 and 100.61

Practical applications:

• Finance: A stock with 15% average return and 20% SD is much riskier than one with 12% return and 8% SD
• Manufacturing: If a bolt should be 10mm ± 0.1mm, a process with SD = 0.03mm produces 99.7% within spec
• Grading: Curved grades often assign A to scores >1.5 SD above the mean

Standard deviation is your baseline tool for understanding data variability. When someone says results are “statistically significant,” they’re essentially saying the observed difference is larger than what random variation (measured by SD) would explain.

Calculate instantly with our Standard Deviation Calculator.