Statistics Calculator
Free online statistics calculator. Find mean, median, mode, range, variance, standard deviation, and sum from any data set with step-by-step working.
Statistics Calculator
Comprehensive Guide: Descriptive Statistics, Variance & Standard Deviation
Learn statistical dispersion, sample vs. population variance, Bessel’s correction (n−1), Gaussian distributions, and Six Sigma quality metrics.
1. What is Descriptive Statistics & Measures of Dispersion?
While central tendency metrics (such as the arithmetic mean or median) describe the numerical center of a data cluster, they convey nothing about how widely dispersed the individual observations are. Two separate classrooms can both record an identical average test score of 75%, yet in Classroom A every single student scored between 73% and 77%, while in Classroom B individual scores fluctuated wildly between 20% and 100%.
Descriptive statistics quantifies this spread through rigorous mathematical indices: variance, standard deviation, interquartile range (IQR), and mean absolute deviation (MAD).
2. Standard Deviation & Variance Formulas: Sample vs. Population
A critical foundational distinction in statistical science is whether your dataset represents an entire exhaustive population (N) or a sampled subset (sample of size n):
Sample Variance (Bessel's Correction): s² = Σ(xᵢ − x̄)² ÷ (n − 1)
Standard Deviation: σ = √(σ²) | s = √(s²)
Why divide by (n − 1)? When calculating dispersion from a random sample, using the sample mean x̄ systematically underestimates the true population dispersion because sample deviations are calculated around their own center. Dividing by n − 1 (known as Bessel's correction) restores mathematical unbiasedness to the variance estimator.
3. Graduated Step-by-Step Worked Problems
Problem 1: Step-by-Step Sample Variance & Standard Deviation
Dataset (Sample of n = 5 test runs): {4, 8, 6, 5, 7}
Step 1: Compute sample mean: x̄ = (4 + 8 + 6 + 5 + 7) ÷ 5 = 30 ÷ 5 = 6.0.
Step 2: Calculate squared deviations from mean:
• (4 − 6)² = (−2)² = 4
• (8 − 6)² = (+2)² = 4
• (6 − 6)² = (0)² = 0
• (5 − 6)² = (−1)² = 1
• (7 − 6)² = (+1)² = 1
Step 3: Sum of Squared Deviations (SS): 4 + 4 + 0 + 1 + 1 = 10.0.
Step 4: Sample Variance (s²): SS ÷ (n − 1) = 10.0 ÷ (5 − 1) = 10 ÷ 4 = 2.5.
Step 5: Sample Standard Deviation (s): s = √2.5 ≈ 1.5811.
Result: Sample variance is 2.5, sample standard deviation is 1.58.
Problem 2: Population vs. Sample Comparison
If the same dataset {4, 8, 6, 5, 7} represented the entire complete population of 5 production machines in a facility:
• Population Variance: σ² = SS ÷ N = 10.0 ÷ 5 = 2.0.
• Population Standard Deviation: σ = √2.0 ≈ 1.4142.
Notice that sample standard deviation (1.58) is slightly larger than population standard deviation (1.41) to correct for sample estimation uncertainty.
4. The Empirical Rule (68–95–99.7%) & Quality Engineering
In normally distributed (Gaussian) populations, standard deviation defines exact probability bands:
- 68.27% of all data falls within 1 standard deviation of the mean:
[μ − 1σ, μ + 1σ]. - 95.45% of all data falls within 2 standard deviations:
[μ − 2σ, μ + 2σ]. - 99.73% of all data falls within 3 standard deviations:
[μ − 3σ, μ + 3σ].
Six Sigma Industrial Quality: In precision aerospace and semiconductor manufacturing, processes are engineered to operate at 6σ tolerance limits, allowing no more than 3.4 defective parts per million opportunities (DPMO).
5. Frequently Asked Questions (FAQ)
Why do we square the deviations instead of just taking the average distance?
The sum of raw deviations from the mean is always identically zero: Σ(xᵢ − x̄) = 0, because negative and positive distances cancel each other out completely. Squaring forces all quantities to be positive while giving greater weight to large outliers.
What is the difference between standard deviation and standard error?
Standard deviation (s) describes the variability of individual data points within a dataset. Standard error of the mean (SEM = s / √n) measures how accurately the sample mean estimates the true population mean, decreasing as sample size increases.
Can standard deviation ever be negative?
No. Because variance is the sum of squared real numbers divided by a positive count, variance is always non-negative (s² ≥ 0). Its principal square root, standard deviation, is therefore always zero or positive. It equals zero only if every single observation in the dataset is identical.