This visualization workshop was generated by Claude (Anthropic) for the STAT 101 course materials. Every number on it is computed in your browser and recomputed on every knob change: every mean, the sample variance with the n − 1 divisor, the covariance, the correlation, the regression slope and intercept, the exaggeration factor, and the pooled and per-group slopes. Nothing on this page is sketched.
Anscombe’s quartet is the canonical published dataset — the eleven (x, y) pairs of each of the four sets are recovered verbatim from Anscombe’s 1973 table, not invented. The four sets were constructed to share their summary statistics; this page confirms that they do, live, to two decimals, while drawing the four scatterplots so you can see they look nothing alike. The two-bar values and the Simpson dataset are hand-chosen illustrative inputs; the arithmetic done on them is real.
Anscombe’s quartet: F. J. Anscombe, “Graphs in Statistical Analysis,” The American Statistician, vol. 27, no. 1, pp. 17–21, 1973 (Table 1). Sample variance uses the n − 1 (Bessel) divisor, the unbiased estimator; the “identical statistics” claim is stated for that estimator, so the divisor is load-bearing. Simpson’s paradox: E. H. Simpson, “The Interpretation of Interaction in Contingency Tables,” JRSS B, vol. 13, 1951.
Course demo — reached from the Labs & Demos hub; the page itself is English‑only for now. Built for STAT-101 Module 2. Three things a good chart makes obvious and a summary number hides: Anscombe’s quartet (four datasets, identical statistics, four different shapes), a truncated axis (how a 5% gap is made to look like a doubling), and Simpson’s paradox (a trend that reverses once you split by group). Every statistic is computed here, so “which chart is honest” has a number attached.