Techniques by Category

EDA techniques divide into graphical (visual pattern detection) and quantitative (numerical summary) methods. This taxonomy organizes all techniques by the analysis problem they address. See also Analysis Questions for guidance on when to apply each category.

Distributional Analysis

Technique	Type	Purpose
Histogram	Graphical	Shape, modality, spread
Box Plot	Graphical	Median, quartiles, outliers
Normal Probability Plot	Graphical	Normality assessment
Probability Plot	Graphical	General distributional fit
PPCC Plot	Graphical	Optimal shape parameter
Q-Q Plot	Graphical	Compare two distributions
Anderson-Darling Test	Quantitative	Formal goodness-of-fit
Chi-Square GOF Test	Quantitative	Categorical/binned fit
K-S Test	Quantitative	Distribution-free GOF
Skewness and Kurtosis	Quantitative	Shape measures

Location (Central Tendency)

Technique	Type	Purpose
Histogram	Graphical	Visual center
Box Plot	Graphical	Median display
Measures of Location	Quantitative	Mean, median, mode
Confidence Limits	Quantitative	Interval estimation
Two-Sample t-Test	Quantitative	Compare two means
One-Factor ANOVA	Quantitative	Compare multiple means
Multi-Factor ANOVA	Quantitative	Factorial mean comparisons

Spread (Dispersion)

Technique	Type	Purpose
Box Plot	Graphical	IQR visualization
Histogram	Graphical	Range and shape
Measures of Scale	Quantitative	Std dev, variance, range, IQR
Bartlett’s Test	Quantitative	Equal variances (normal data)
Levene Test	Quantitative	Equal variances (robust)
Chi-Square SD Test	Quantitative	Single variance test
F-Test	Quantitative	Two-variance comparison

Randomness

Technique	Type	Purpose
Run-Sequence Plot	Graphical	Trend, shift detection
Lag Plot	Graphical	Serial dependence
Autocorrelation Plot	Graphical	Correlation structure
Autocorrelation	Quantitative	Lag-k correlation values
Runs Test	Quantitative	Formal randomness test

Model Adequacy

Technique	Type	Purpose
6-Plot	Graphical	Multi-panel diagnostic
4-Plot	Graphical	Rapid assumption check
Scatter Plot	Graphical	Residual patterns
Normal Probability Plot	Graphical	Residual normality

Outlier Detection

Technique	Type	Purpose
Box Plot	Graphical	1.5xIQR rule
Normal Probability Plot	Graphical	Tail departures
Grubbs’ Test	Quantitative	Single-outlier hypothesis test

Multivariate and Time Series

Technique	Type	Purpose
Scatterplot Matrix	Graphical	Pairwise relationships
Star Plot	Graphical	Multivariate profiles
Contour Plot	Graphical	3D surface projection
Conditioning Plot	Graphical	Conditional relationships
Spectral Plot	Graphical	Frequency decomposition
Complex Demodulation	Graphical	Amplitude/phase extraction