Maps for storytelling: Best practices

From DevSummit
Jump to navigation Jump to search

Types of maps:

--Physical maps

--Mind maps

--Maps for people to access voting services

--Maps we use every day to get from A to B

--Mapping Haiti

--Community mapping projects � using them for advocacy

When is it a bad idea to use maps?

-- When location is not the most important dimension of the data

(example: mapping petty corruption where sector, institution are more important)

--When a mapping project would out communities that don�t want to be found;

--Ethical considerations: check the responsibledata.io

--Sometimes have 2 versions of a map: one internal and one external

--Aggregate data if you want to maintain anonymity of people you map (from point data to density or shapes)

Data collection map tools:

--JS Cloud for mobile data collection

--Esri

--Mapbox

--Google maps

-- QGIS- you can do both visualisation and analysis (useful if you need to transform the data)

--Colour brewer -- for colour schemes

--Fusion tables for Google � useful for geo-locating data (transforming addresses into latitude, longitude)

--Batch geo-coder

--Arc-GIS

Organisations that uses mapping well:

--The hate map-- 50 state maps that has icons for hate groups showing where they are; makes people realise how widespread the problem is (splcenter.org/hate-map) � Esri or Mapbox

---Side box, when you zoom in you get more information about the area you are looking at

--Racial inequality in schools - school districts

--Appalachian voices - type in a zip code and it tells you which mountain has been destroyed for power; great story mapping project

Projections:

--Mercator, Peterson � ways to express spherical data into a flat map

--Initial projections were useful for navigation on sea but not distorted countries and continents

--Important to choose the right projection for your data � search for map projections for the project you are working on

Think of the map as a message - what is that you want to communicate? Make sure that what the one thing you are showing is prominent (add other elements like roads, etc. only if the reinforce the main point) eliminate noise, better to have a more simplistic map that communicates a message

How to make sure you are not lying/ being deceitful with maps:

--Choropleth map -elections results

-Books: Storytelling with data; How to lie with maps

--Ensure data is correcting for population density (per capita instead of numbers)

--Think what the unit of analysis is and how you aggregate data (market, county, zip code, etc.)

Digitise physical maps - Esri has a tool to upload PDF maps, where you can trace; every time you trace, it creates locations and point that you can add notes to; Illustrator is also used for cartography projects;