MENTAL MODEL · M2555

Anscombe's Quartet

TechnicalHigh supportStatistics

Included

Version 1.0.0 · Updated 2026-07-30

CORE DEFINITION

Four completely different datasets (with entirely different distributions of X and Y values) that share nearly identical statistical properties (mean, variance, correlation coefficient, regression line). However, when plotted, the four datasets look vastly different. -

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

The trap of blindly relying on summary data. - It demonstrates that 'averages' and 'statistical summaries' can lie. When reviewing financial reports or business data, one must never only look at summary statistics in Excel; one must demand to see distribution plots or scatter plots. Data analysis without visualizations is like the blind men and the elephant.

anchor

Anchor fast decisions

The quartet was constructed by statistician Francis Anscombe in 1973. The four datasets have nearly identical means, variances, correlation coefficients, regression slopes, and intercepts, but their distributions differ markedly (linear, curved, outlier, and high-leverage point). The mechanism illustrates that summary statistics compress the structure and anomalies of data; only lossless visualization can reveal the truth.

MINIMUM ACTION

In progress 0/1

Practice this model in one real situation:

In any data analysis, always plot scatter plots or distribution graphs before drawing conclusions. For regression results, check for high-leverage points or non-linear relationships. Use visualization to cross-validate summary numbers, rather than making decisions based solely on correlation coefficients or means.

Check to track your progress (stored locally)

Learning progress0%

account_treeGenealogyexpand_more

menu_bookReferencesexpand_more

Source support: Explicit

link
en.wikipedia.orghttps://en.wikipedia.org/wiki/Anscombe%27s_quartetZH · Explicit
verified

RELATED MODELS