Anscombe's Quartet (1973) & Datasaurus Dozen (2017)
All FOUR datasets below have IDENTICAL statistics:
Yet they are COMPLETELY DIFFERENT when plotted!
Created by statistician Francis Anscombe in 1973, this quartet demonstrated that numerical summaries alone are insufficient for understanding data. Each dataset has 11 points and identical summary statistics, but reveals completely different patterns when visualized.
📖 Source: Anscombe, F. J. (1973). "Graphs in Statistical Analysis." The American Statistician, 27(1), 17-21.
In 2017, Justin Matejka & George Fitzmaurice created 13 datasets with nearly IDENTICAL statistics but wildly different shapes - including a dinosaur!
All 13 datasets have: Mean X ≈ 54.26, Mean Y ≈ 47.83, Correlation r ≈ -0.064
Created using a simulated annealing algorithm, these datasets demonstrate that identical summary statistics can describe vastly different patterns. The original "Datasaurus" was created by Alberto Cairo, and extended to 12 more shapes by Matejka & Fitzmaurice.
📖 Source: Matejka, J., & Fitzmaurice, G. (2017).
"Same Stats, Different Graphs: Generating Datasets with Varied Appearance and Identical Statistics through Simulated Annealing."
CHI 2017.
🔗 Data: TidyTuesday GitHub
📦 R Package: datasauRus by Steph Locke & Lucy D'Agostino McGowan (MIT License)