Preparing Effective Maps

Preparing Effective Maps 24 January 2002 What makes a map communicate accurately, clearly, and persuasively? Considerations in Preparing a Map Spatial aggregation o Determine the appropriate spatial level to display (e.g., state, county, tract, block group, town, ZIP code) Data processing o Data extraction and adjustment o Covered in database and census lectures o Often a substantial task Classification of data o Organizing data for display Symbolization o Deciding how to represent features Other elements of a map o Features to improve readability and clarity Map as a Graphical Tool for Presenting Information Maps "tell a story" The story may change depending on the items listed above Data Processing and Normalization The data you need may not be available in Raw counts need to be adjusted for variations in: o land area o total population o total housing units o others Converts magnitude data (counts, sums) into intensity data (rate, percentage, average) Examples: Magnitude Intensity Population of a census tract Population / area of tract = population density of a census tract Count of housing units in a block group Housing units / area of block group = housing density of block group Vehicles available in a census Vehicles available / occupied housing units = average tract vehicles per occupied housing unit Data Classification Classification is key to producing understandable maps that people can interpret readily Classification can strongly affect the apparent results Guidelines for classification: o Policy relevant (e.g., incorporating poverty line in household income map) o Scientifically meaningful (e.g., carrying capacity) o Informative How many classes? o From 2 to 5 ranges work best o The most common number is 5 o Related to symbolization (e.g., color vs. graytone) Classification Methods Used by ArcView Tip: For ArcView's explanations of its classifications methods, search for "natural breaks" in the topics index in the online help. The page that appears shows ALL the classification methods. (1) Quantile (Equal Count) assigns (roughly) equal number of cases to all categories provides a balanced image puts very different values together -- covers outliers and thresholds the break points may not be policy relevant or scientifically meaningful (2) Equal Interval Classification classifies according to data values with equal interval is easy for map user to understand tends to produce unbalanced map image (can produce empty categories) break points may not be policy relevant or scientifically meaningful (3) Natural Break Points look for "naturally" occurring groupings in the data internal homogeneity for each class (4) Standard Deviation distance from the mean statistically sound (6) Equal Area break points determined by polygon areas attempts to have equal total area of polygons in each group (5) User-Defined classification gives freedom to determine the break points possible to make the break points meaningful difficult for the reader interpret the map harder to justify Symbolization Three geometric categories of map symbols: o points (zero-dimensional objects) o lines (one-dimensional objects) o areas (two-dimensional objects) Complexity (scale factor) o e.g. cities as points in small scale maps Visual Variables Described by Mark Monmonier in How to Lie with Maps 1st ed.: Chicago: University of Chicago Press, 1991 2nd ed.: Chicago: University of Chicago Press, 1991 Size o o o o Shape: radius of circle width of line area of a shape best for describing magnitude data effective for showing qualitative differences (e.g., a school vs. a church; forest vs. water) Graytone value: o effective for describing intensity data (percentage of low income; population density) Texture: o effective for showing both qualitative and intensity differences o effective for producing easily reproduced black-and-white maps, but are often hard to read and interpret Orientation: o effective for showing flows o (e.g., migration flows; traffic flows) Hue (color): o effective for showing both qualitative and intensity differences o General Rules for Symbolization Selection of symbols should be based on: o logic (order/sequence in size, color, graytone) o common perception/convention Example: blue for water, green for plants o visual clarification Important to remember: o use graduated-point symbols indicate magnitude o use graytones or colors indicate intensity Data aggregation Different areal aggregations may yield very different patterns Basic rule: o start with disaggregated data whenever possible Other elements of a map Title and labels: o informative title: precise and concise o label: naming and signifying features Scale: o o o the choice of scales (e.g. regional planning vs. site design) different type of scales (graphic scale as strategy against distortion) drawing a scale bar in MapInfo North Arrow: o orientation of the map Legend: o o show class breaks clearly and accurately corresponding with symbolization Map projection: o Equal-area vs. conformal (angle-preserving) o Projection pages by Peter H. Dana What should we remember? Keep in mind the basic rules of cartography Take advantage of the computer technology to explore data and refine maps Verify your conclusions or outcome with good knowledge of the place and subject matter As map maker, don't lie with maps As map user, don't get misled by maps that lie Be creative but careful