data visualization

Data visualization is a specific kind of visualization — it applies the general practice of making information perceptible to the specific domain of data and information. Per Law 9: 'a competent speaker would call data visualization a visualization.' The direction is correct: data visualization (specific, domain-limited) → visualization (general, applies to all perceptible representations).

Data visualization IS a specific kind of information design — it is the application of information design principles (visual encoding, hierarchy, layout, legibility) specifically to numerical and tabular data. A competent speaker would call data visualization a form of information design. Direction: specific→general under Law 9.

Data visualization is designed and maintained for the sake of communication — its purpose is to make information legible through visual encoding. The test is for-whose-sake: visualizations exist to communicate complex data to human readers. Servant (data visualization) points at master (communication). This is teleology (Law 8d), not dependency.

Error bar is a specific kind of data visualization encoding that represents quantitative uncertainty. Direction specific→general per Law 9. A competent speaker classifies error bars as a data visualization technique.

A faceted visualization IS a specific kind of data visualization: it creates multiple coordinated views (facets) of the same or related data, each filtered by categorical variable values. A competent speaker calls it a visualization technique. Direction: specific→general. The facets grid and small multiple are specific instances of this general approach.

A dot density map IS a specific kind of data visualization — it uses dot density to encode quantitative data spatially. Competent speakers call it 'a visualization.'

Parallel coordinates IS a specific kind of data visualization — multiple vertical axes with lines connecting values. Competent speakers call it 'a chart' or 'a visualization.'

A swarm plot IS a specific kind of data visualization — scatter plot with jittering to avoid overlap. Competent speakers call it 'a visualization.' Direction: specific→general.

A chart IS a specific kind of data visualization — it encodes data into a structured visual form. A competent speaker would call a chart a type of data visualization. Specific→general INSTANCE_OF to nearest kind.

Direction tested: INSTANCE_OF (specific→general). A dashboard IS a specific kind of data visualization — a consolidated visual display aggregating multiple data points for monitoring. A competent speaker would call a dashboard 'a form of data visualization.' Nearest kind.

Small multiples (Tufte's technique of displaying parallel small views of data) is designed for the sake of data visualization — it enables comparison across subsets of data, which is the core purpose of visualizing data. By design and sustained practice in statistics and information design.

A treemap is a visualization technique designed for the sake of data visualization: it displays hierarchical data as nested rectangles sized by quantitative values. Per Law 8d: its designed purpose is to further data visualization's operation.

A waterfall chart is a visualization built for the sake of data visualization: it displays sequentially accumulated positive and negative values as floating bars. Per Law 8d: its designed purpose is to further data visualization's operation.

A dot plot is a visualization technique built for the sake of data visualization: it displays individual data points as dots on a categorical axis. Per Law 8d: its designed purpose is to further data visualization's operation.

A rug plot is built and maintained for the sake of data visualization: it is a visualization technique whose designed purpose is to display data in a way that reveals patterns and distributions. Per Law 8d, the servant (rug plot) points at the master (data visualization). The note shows the purpose is by design — rug plots are a recognized statistical visualization tool.

A pivot table is a specific kind of data visualization — it summarizes and groups data in a tabular format that conveys quantitative patterns through structured layout.

A violin plot is built and maintained for the sake of data visualization — its designed purpose is to show the distribution and density of data. The servant (violin plot) points at the master (data visualization).

Error bars are built into charts and plots to communicate the reliability or variability of the displayed data. They serve data visualization by encoding uncertainty visually so viewers can assess data quality.

A chord diagram visualizes connections and flows within a dataset — its purpose is to make relational data visible, serving data visualization.

A contour plot maps 3D surfaces onto 2D using isolines to visualize data patterns. It is built for the sake of enabling data visualization — its designed purpose is to further the operation of data visualization, just like chord diagrams, error bars, and violin plots.

A stem plot (or stem-and-leaf plot) is a data visualization technique that displays quantitative data by splitting each value into a stem and a leaf, enabling visual inspection of distribution shape and individual values simultaneously.

A spider chart (radar chart) is a data visualization technique for displaying multivariate data on a 2D plane with axes starting from the same point, enabling comparison of multiple quantitative variables across categories.

A calibration plot serves data visualization by providing a diagnostic channel through which modelers can visually inspect whether predicted probabilities match observed frequencies, translating statistical assessment into an interpretable visual form.

Bubble charts are specifically designed to visualize multivariate data using circles positioned in two dimensions with area encoding a fourth variable — their entire purpose is to serve data visualization by making complex relationships perceptible at a glance.

A streamgraph is built and maintained for the sake of data visualization — its designed purpose is to display multiple data series as flowing stacked areas over time, making patterns and trends visually apparent. The servant (streamgraph) points at the master (data visualization). This is not DEPENDS_ON: data visualization would still operate without streamgraphs; streamgraphs are one tool among many for that purpose.

A facet grid is designed to serve data visualization by enabling the display of multiple subplots, each showing a different subset of data conditioned on one or more categorical variables. Its purpose is to reveal patterns and interactions across dimensions. This is SERVES (for whose sake? data visualization), not DEPENDS_ON (a facet grid doesn't need data visualization to function — it IS a data visualization tool).

Visual encoding IS a specific kind of data visualization: it maps data attributes to visual properties (position, color, size) to represent information. A competent speaker would call visual encoding a form of data visualization.

A beeswarm plot is designed for the sake of effective data visualization — it arranges points to reduce overplotting while preserving scatter-plot semantics.

parallel coordinates plot DERIVED_FROM data visualization: edge-test confirms data visualization (epoch 0.24) is older/more-foundational than parallel coordinates plot (epoch 0.97). Parallel coordinates plots are a specific visualization technique built upon the broader concept of data visualization.