Compute key measures of spread: variance (population & sample), standard deviation, range, interquartile range (IQR), quartiles, coefficient of variation (CV). Interactive dot plot visualizes data distribution.
In statistics, dispersion (or spread) describes how much a dataset deviates from its central tendency. While measures like the mean provide a summary, they conceal critical information about variability. Two datasets can share identical means but have vastly different interpretations — one tightly clustered, the other widely scattered. Dispersion metrics are essential for risk assessment, quality control, inferential statistics, and machine learning feature analysis.
Population Variance (σ²) = 1/n ∑(xᵢ - μ)² | Sample Variance (s²) = 1/(n-1) ∑(xᵢ - x̄)²
Standard deviation = √variance, restoring original units.
IQR = Q₃ - Q₁ (robust to outliers) | Range = max(x) - min(x)
Coefficient of Variation (CV) = (s / |x̄|) × 100% (for sample) – unitless measure of relative dispersion.
Our calculator follows standard statistical conventions: Population variance divides by n (applicable when the dataset represents the entire population), while sample variance divides by n-1 (Bessel's correction) to provide an unbiased estimator of the population variance. Quartiles are computed using the inclusive median method (Tukey's hinges): Q₁ is the median of the lower half, Q₃ the median of the upper half, excluding the overall median if n is odd. The coefficient of variation (CV) is defined as the ratio of the sample standard deviation to the absolute sample mean, expressed as a percentage. It allows comparison of variability across datasets with different units or scales. The dot plot maps each data point along a horizontal axis, with a small vertical offset to reduce overlap, providing an intuitive visualization of density and spread.
Note on quantile calculation: Multiple methods exist for quantile calculation (e.g., (n+1)/4 method, Mendenhall and Sincich). This calculator employs Tukey's hinges, consistent with standard boxplot conventions. Results may differ slightly from other software; please refer to your field's guidelines when precise alignment is required.
A factory produces metal rods with target length 10.0 cm. Two machines produce samples: Machine A: [10.1, 9.9, 10.0, 10.2, 9.8, 10.1] → mean = 10.02, sample SD = 0.15, CV = 1.5%. Machine B: [9.7, 10.3, 9.8, 10.4, 9.6, 10.2] → mean = 10.00, sample SD = 0.32, CV = 3.2%. While means are almost equal, Machine B has double the standard deviation and CV, indicating inconsistent production. Using our dispersion calculator, quality engineers can identify problematic machines and reduce variability, saving costs and scrap material.