Cartes-fiches

different kinds represented by differentiating symbols

- colour hue

- orientation

- texture

Example: different lines for different roads, Flow Map of winds

different ranks/magnitudes represented by proportionally sized symbols, using:

- size

- colour value

- sometimes colour hue

different lines for different magnitudes, absolute data, total value!

Example: quantitative flow maps, e.g. petrol flow

types: radial, network, distributive (siehe S. 12)

design issues:

emphasis is on difference in magnitude flows

increse figuere gorund contrast

appropriate projection

arrows if necessary

comprehensive unambiguous legend

Summary S. 16

-shape

-color hue

-orientation

-texture

gives answer to question: what and where is at this location?

different ranks/ magnitudes of real and conceptual pint features

using variables:

-size

-colour value (and colour hue i.e. yellow vs. brown)

emphasis is on quantity, not necessarily on the exact location

absolute data, total values

What, where and how much is at this location?

proportional point symbol map = graduated symbol map

point symbol types:

- mimetic (symbol looks like real thing)

- pictographic (bildhaft)

- geometric

area of symbol is proportional to data being mapped! Symbol area represents raw totals (absolute numbers)

classed or unclassed scaling

1. continuous proportional area scaling for circles

area size is in proportion to the radius of the circle

other possibilities:

2. range graded scaling (data are classed, scaled to midpint of class)

3. psychological scaling (problem with just noticeable difference, the more similare circles are, the more difficult is it to differentiate the circles. )

compensation for underestimation of circle difference:

power law function: R= K*S^n n= .8747, truth: n=1

Further problems: ebbinghaus effect

Design Issues: clear ground figure contrast, consider minimal/maximal simbol size, rearrange overlapping symbols

also multivariate versions in combinatin with other data possible!

arrangement of magnitude at pint locations

- shows relative locations

- focus on feature density

2 types: one-to-one and one-to-many dot maps

pros: easy to understand, show density depending on additional data, original data can recovered

cons: tendious to construct by hand, computer methods often assign dots randomly, use subdivision of enumeration units to correct, original values are difficult to extract if optimized for display

Dessign Issues:

- selection of number of dots: 2-3 dots for lowest magnitude, find maximum coalescence for highest magnitude

- selection of dot values: logical

.adjust dot density to map scale: not too detailed, not too general

- adjust dot size