Class 4 | Methods 3

Welcome to Methods 3, Lecture 4

This is a web page that can be viewed as slides.

→ to move forward

← to go back

Louisiana Loses Its Boot

"Using publicly available data, [we] created a map on which areas that commonly appear as land on government issued maps—woody wetlands, emergent herbaceous wetlands and barren land—were re-categorized to appear as water."

the modifiable areal unit problem

source

"One-third of all homes with a Flint ZIP code lie outside the city. Thus, the state’s numbers for Flint were watered down by an additional 50 percent of addresses that weren’t in the city and weren’t using Flint water."

source

Traditional Color Coding for Land Uses

Land-Based Classification Standards

opendata.cityofnewyork.us

bytes of the big apple

databank.worldbank.org

USGS earthquake data

you can copy and paste styles

you can save a layer with its style

Start thinking about methodology

you should almost never make choropleths with just counts

think about ways to normalize your data

this make the data comparable across regions

a GIS will let you:

visualize map data
create and edit map data
overlay map data
analyze map data

buffers

make a "buffer" around features

you can make negative buffers

Symmetric difference

usually you'll use geoprocessing functions in a chain

keep track of these as you work!

lab

in-class exercises, part 1

joins relate two datasets to each other

joins create new hybrid datasets

source

"I have this spreadsheet I want to map..."

you need a file with relatable geographies

and a column in each file that relates the two

lab

in-class exercises, part 2

American Community Survey

margin of error: higher with 1-year estimates and smaller geometries

NYC Census FactFinder

no matter which data you download, you'll need appropriate geographies to go with it

census blocks are the smallest area data is aggregated to

source

tracts have population between 1,200 and 8,000, optimally 4,000

source

NTAs are NYC-specific aggregations of census tracts

source

downloading census data

make sure you have the right boundary files

source

Methods 3

(review)

Choropleth?

Louisiana Loses Its Boot

the modifiable areal unit problem

Finding data

duplicating layers

you can copy and paste styles

you can save a layer with its style

basic statistics

Start thinking about methodology

you should almost never make choropleths with just counts

think about ways to normalize your data

this make the data comparable across regions

a GIS will let you:

Geoprocessing

buffers

buffers

you can make negative buffers

clipping

overlay operations

Intersect

Union

Symmetric difference

usually you'll use geoprocessing functions in a chain

keep track of these as you work!

lab

joins

attribute joins

joins relate two datasets to each other

joins create new hybrid datasets

"I have this spreadsheet I want to map..."

you need a file with relatable geographies

and a column in each file that relates the two

lab

what is the census?

decennial census

American Community Survey

margin of error: higher with 1-year estimates and smaller geometries

no matter which data you download, you'll need appropriate geographies to go with it

census blocks are the smallest area data is aggregated to

blocks

block groups

tracts

tracts have population between 1,200 and 8,000, optimally 4,000

NTAs are NYC-specific aggregations of census tracts

downloading census data

joining census data

make sure you have the right boundary files

join on AFFGEOID and GEO.id

join on AFFGEOID and GEO.id

or GEOID and GEO.id2

lab

lab

join on `AFFGEOID` and `GEO.id`

join on `AFFGEOID` and `GEO.id`

or `GEOID` and `GEO.id2`