3 ms·
These are the 'damn lies' form of statistics. It starts out showing 2D representations, then switches to a linear format. It also uses sideways scrolling which
by karmakaze 3y ago
These are the 'damn lies' form of statistics. It starts out showing 2D representations, then switches to a linear format. It also uses sideways scrolling which is cumbersome than vertical which would be far more natural (but people are too used to scrolling far in this direction so let's do something awkward).
It's hard for me to say what 'dimensionality' accurately portrays magnitude of wealth. Showing us say 2d projections (pictures) of 3D cubes might correlate, but I suspect that it more of a logarithmic than cube root relationship.
- BHSPitMonkey 3y agoHow did you determine that? I saw no indication that the scale had ever changed (from the starting definition of 1 pixel per $1,000). At a certain point it tells you how much money a vertical column of pixels represents, but that's still based on the original scale (factored by the fixed height of the shaded region).
- karmakaze 3y agoThe first three areas are shown as square-shaped rectangles. Then at $1b, it switches to using a fixed height thus the 'size' is a linear dimension $amount/height rather than sqrt($amount). Technically all these have the correct number of pixels, but we can perceive a larger magnitude range when laid out in squares rather than long rectangles with a limited/fixed height. There's nothing special about using a line, square, or cube to illustrate amounts. However if I were trying to have people actually get a sense of a large range I would use cubes as that's the highest dimension we can intuitively grasp. e.g. picking up a metal object of visible volume weighs about as much as we expect.