scatter plot

scatter plot is a specific type of graph: a plot using Cartesian coordinates to display values from two variables. Specific→general per Law 9.

A scatter plot IS a specific kind of plot: it displays quantitative data as a collection of points on Cartesian coordinates. Per Law 9, a competent speaker would call a scatter plot 'a plot'. Plot is the nearest kind — the edge runs from specific to general.

A scatter plot maps pairs of measured values; remove measurement and there are no coordinates to plot.

Bubble chart is a specific kind of scatter plot: it displays data points as bubbles on x-y axes where a third dimension controls bubble size. A competent speaker would call a bubble chart 'a scatter plot.' Scatter plot is older (0.31 vs 0.35 epoch) and is the nearest kind.

A scatter plot matrix IS a grid (matrix) of scatter plots — each cell is a scatter plot showing the relationship between two variables. The whole is the matrix layout; each constituent part is a scatter plot.

TESTED INSTANCE_OF: qq plot IS a specific kind of scatter plot. A QQ plot displays pairs of quantiles from two distributions as points on a 2D scatter — x-axis plots quantiles of one distribution, y-axis plots quantiles of another. It uses exactly the scatter plot glyph (points in Cartesian coordinates), differing only in what the axes encode. The nearest kind is scatter plot.

TESTED INSTANCE_OF: a funnel plot IS a specific kind of scatter plot used in meta-analysis. It plots effect sizes against a measure of variance/precision, with confidence interval lines — the visual core is scattered points with error bars. A competent speaker would call a funnel plot 'a scatter plot of study effects.'

A residual plot IS a specific kind of scatter plot: it plots residuals (errors) against fitted values or predictors as a scatter of points. A competent speaker would call a residual plot 'a scatter plot' — the points encode fitted value on x-axis and residual on y-axis.

INSTANCE_OF: A strip plot IS a specific kind of scatter plot — it plots individual data points at their quantitative values, with one axis being categorical. A competent speaker would call a strip plot a type of scatter plot.

beeswarm plot IS a specific kind of scatter plot that avoids overplotting by arranging data points along one axis to reduce overlap. A competent speaker calls a beeswarm plot 'a kind of scatter plot.' Tested: scatter plot is older/more foundational, so beeswarm plot INSTANCE_OF scatter plot is correct direction.

A pair plot is a specific kind of scatter plot: it arranges scatter plots in a grid to show pairwise bivariate relationships between multiple variables. Per Law 9, 'X is a specific kind of Y' maps to INSTANCE_OF. A pair plot is fundamentally a structured collection of scatter plots, each showing two variables.

A rug plot displays individual data points as marks along a single axis, showing the distribution of one variable. The removal test: remove scatter plot (the concept of plotting individual points on axes) and a rug plot has no operational mechanism — it is a 1D scatter plot variant. Direction: rug plot (1D scatter) to scatter plot (general point-plotting concept).

A rug plot is a specific type of scatter plot: dots plotted along a single axis with a density dimension. A competent speaker would say 'a rug plot is a scatter plot.' Nearest kind test: scatter plot is the general class, rug plot is a specific variant with one fixed axis.