🌱 Observable Terms

Ordinal Categorical Quantitative Band Scales Axis Facet group transform - groups data by ordinal value bin transform - intended for use with the rect mark

Cell Bar Rect Dot

When a cell’s position represents a time period, such as a year, consider using a bar instead of a cell.

Bars also have the advantage of accurately representing time periods of varying length, such as months

Aa bar is used for a bar chart but a rect is needed for a histogram.

Bin

Bin transforms used for histograms or heatmaps

Because the bin transform operates on quantitative data, a rect is used instead of a bar, but you can use area and line.

bin tranforms

• freedman-diaconis - the Freedman–Diaconis rule
• scott - Scott’s normal reference rule
• sturges - Sturges’ formula
• a number representing the desired number of bins
• an array of n sorted thresholds for n + 1 bins
• a time interval (for temporal binning)
• a function to return the desired thresholds given an array of values

bin reducers

• count - the number of observations in each group
• sum - the sum of values
• proportion - like sum, but divided by the total
• proportion-facet - like sum, but divided by the facet’s total
• min - the minimum value
• max - the maximum value
• mean - the mean (average) value
• median - the median value
• variance - an unbiased estimator of population variance
• deviation - the standard deviation
• first - the first value (in input order)
• last - the last value (in input order)

stack

The stack transform stacks from a zero baseline by default, but this can be changed with the offset option. Four offset methods are supported:

• silhouette - centered, as proposed by Havre et al.
• wiggle - minimizing movement, as proposed by Byron & Wattenberg
• expand - normalized to proportions in [0, 1]
• null - a zero baseline