Reading: Schuurman ch 2 & 3

February 18, 2014


On no snow for Christmas in Finland

Schuurman: GIS: A Short Introduction

Ch. 2: GIS, Human Geography, and the Intellectual Territory Between Them

GISystems & GIScience based on assumptions that privilege certain approaches to understanding the world (natural and human).

Geography: diverse, undisciplined discipline, origins in 1800s

GIS since the late 1960s, parts of cartography & quantitative methods

  • sometimes rocky relationship between these and Geography in general

Mind the Gap: The Distance Between Human Geography and GIS

Little overlap between GIS and Human Geographers until the late 1980s

Geographers critique: GIS is mere technique, no intellectual component

  • GIS processes facts, but can’t generate meaningful understanding
  • GIS based on positivism and/or naïve empiricism: neither well respected approaches/theories in Geography

Positivism/empiricism: experiment/test/trial: sense perceptions are the only admissible basis of human knowledge and precise thought; natural and social processes can be understood (via hypothesis testing and data analysis) and follow strict laws; designed to supersede theology and metaphysics.

  • ex) central place theory in Geography
  • ex) much of science and some social science.
  • ex) less comfort with qualitative methods
  • ex) less comfort with theories that use empirical data but don’t see laws governing human behavior and activity (feminism: role of gender in shaping society, but these are not laws – they can be overcome and changed for the better of all)

1980s: lots of debates

  • GIS people with a more positivistic, scientific approach vs human geographers with more qualitative, social theory approaches
  • GIS very limited view of the world, requires very specific, empirical data, can ask very specific questions, and get very specific results.
  • GIS driven by corporate and military needs
  • GIS expensive and exclusive; elitist

Brian Harley (The Nature of Maps), Denis Wood (The Power of Maps)

  • critique of maps: social constructions for creating and maintaining power
  • selectively show certain things, not others; create and enforce social status quo
  • maps create a space of political territories (broader scale) and privately owned property (detailed scale) and make those human conceptions real in the landscape
  • ex) the “nation” / “states” – relatively new concept; problem in Mid East
  • ex) property ownership: relatively new human concept
  • ex) zoning

Maps created by elites to shape and enforce geographic reality to suit their needs

John Pickles: Ground Truth: apply same critique to GIS

  • GIS is for maintaining order, just like paper maps before them
  • Friday Harbor meeting: beginning of a dialog
  • alternatives: Particapatory GIS, counter-mapping, qualitative methods “GIS2”

Epistemology and Ontology in GIS

  • epistemology: the methods we use to study the world; each has assumptions and perspectives that shape the questions, analysis, and interpretation of results
  • ontology: what things really are (how the world must be to make sense of it)
  • ontology (computer science)

ex) we have extensive GIS technology for determining the fastest route for an ambulance to get a sick person to the hospital, but we don’t ask why so many people get sick.

ex) GIS is used extensively to plan new developments and roads but is very much less able to help understand the extensive negative impact of such development on the environment

ex) Forest in India

Use quantitative methods to test IR energy reflectance from various types of vegetation and different kinds of land cover in an area; gather data, test hypothesis, generate specific measurable value that differentiates a forest from other areas.

  • Use remote sensing to define areas that have a certain % reflection of IR energy; any less than that % is not a forest, any more than that % is.
  • then create a map of forests (and not forests; can do this all in GIS from afar)
  • Empirical epistemology, we can “sense” reflected energy and use that do define and distinguish forest from other areas.
  • Empirical ontology: the world consists of measurable objects, some of which reflect energy and specific kinds of energy reflectance lead us to understand and locate real forests.

Use qualitative methods such as interviews and mental mapping to have different people in the same area of India show where forest is on a map, and describe what forest is to them

  • state foresters: will claim much more territory is in forest as it is their job to preserve and create forested areas
  • farmers: will identify tree covered territory as wasteland or unproductive land, not forest
  • forest dwelling, hunters/gatherers: will focus on areas that are diverse and provide them with food and resources; not “forests” planted by the foresters (not diverse, not a good source of food and resources)
  • a qualitative epistemology: assumes that the reality of “forest” is shaped by human social factors; collect data (mental maps) but interpretation leads to ideas of how a forest is a human construct, even “untouched” forest
  • a qualitative ontology that suggests that sensory measurements in the world are incapable of measuring and helping to understand the social construction of forest; “reality” is shaped and made via social processes.  Social theory explains social processes, but these are not “laws” or unchangeable.
  • Counter Mapping: Peluso: Whose Woods Are These?

Data Models and Ontology

Vector data model: point, line, area (necessary to encode data into the GIS)

  • People: US Census blocks
  • define an area as a particular block, count the number of people
  • all of space is filled with blocks
  • what is the real nature of humans and where they live?

Raster data model: grid of cells

  • land use: each cell (can be very fine) assigned a type of land use based on energy reflected from it
  • complex mosaic of land uses often generalized into agricultural, commercial, etc
  • again, all space is some kind of land use
  • what is the real nature of land use?

Object oriented data models

  • see all geographic features as objects; location as one attribute
  • group together similar objects (roads) and have subclasses (federal, state, local)
  • hierarchical: each “parent” object has attributes common to its subclasses
  • discrete, separate entities in a neutral space; can fill space or not
  • what is the real nature of any geographic feature?  “Forest” as an object?

All data models are reductionist: they simplify complex reality

Need to know how that is occurring and how it shapes understanding when using
GIS or any method

Looking for the Social in GIS

Social aspects of science, technology, GIS

  • ex) funding and research on health biased towards white males

Science as autonomous vs social

Kuhn: The Structure of Scientific Revolutions

  • paradigms: accepted practices and belief systems in science; structure how science is done until enough doubt is cast to accept a new system.
  • what is to be observed and scrutinized, the kind of questions that are supposed to be asked and probed for answers in relation to this subject, how these questions are to be put, how the results of scientific investigations should be interpreted.
  • ex) Copernicus proposed a cosmology with the Sun at the center and the Earth as one of the planets revolving around the Sun.

GIS technology: how have applications developed for the military and environmental science come to shape studies using GIS for non-military and non-environmental science applications?

What is the point ?

GIS is growing rapidly as a method used by diverse people and for diverse applications use growing faster than an understanding of the assumptions and limits of GIS particularly more conceptual, theoretical, even philosophical issues

Important to approach and use GIS with a few things in mind

  • it does tend to privilege one of many approaches to understanding the world: empirical, positivist, scientific; it is not the only way to address particular issues and is not “neutral” or “objective” or necessarily better than other approaches with different assumptions.
  • it is connected to social context: it is a powerful, persuasive tool that has been developed for military and government applications; it is used by experts with training and organizations with big budgets and power; it is in many ways a very elitist method for understanding the world.
  • GIS is always in flux: GIS is not set in stone: development of internet GIS: GoogleMaps and GoogleEarth and slew of similar applications: how will this more “populist” GIS open the door to different kinds of GIS and GIS analysis?  GIS technology will always evolve within a social context.

Schuurman GIS: A Short Introduction

Ch. 3: The Devil is in the Data: Collection, Representation, and Standardization

“Data are not the transparent manifestation of reality in digital terms.  They are the expression of particular points of view and agendas that begin as observations, and are transformed into numbers in data tables that provide the basis for spatial analysis.”

“Data are an artifact that reflects people, policy, and agendas.”

The Politics and Practicalities of Data Collection

Human data collection: often by area (Census block, zip code)

  • U.S. Census: Politics of counting people
  • Census count leads to allocation of money, voting districts
  • undercount of homeless, poor, minorities (3.3 million in 2000)
  • Pima Co. Az: 15,000 undercount, $30 million loss in funds
  • Delaware Co. OH: undercount of about 8000
  • statistical sampling can correct (opposed for Political reasons)

Environmental data collection: often by location

  • GPS: relatively to very accurate locations
  • Military origins; selective availability; competing system (Galileo, EU)
  • primary (collect yourself) vs secondary (use already collected) data

Organizing Data

  • table of data in GIS: like a spreadsheet: ArcGIS Demo
  • spatial data: location (in some coordinate system): where
  • attribute data: describe the spatial data: what
  • consistency: should not be gaps or missing data (although common)
  • scale: large vs small scale maps;
  • scale does not exist in computers; generalize to view: ex) Google Maps
  • scale at which data has been collected (detailed vs general): ex) DALIS data vs. ESRI data
  • aggregation: group of Census blocks > Block group > Census Tract: ex) Geog 222 Exercise 6
  • data interpolation: filling in missing values: terrain shading, temperature

Metadata: Data about Data

Sharing data leads to the need to know about data: when collected, at what scale, who…

  • ex) DALIS data

Sharing Data: Interoperability

  • “a common language for computational environments”
  • cross-platform and cross-software data compatability
  • like a text (.txt) file

Semantic interoperability: the practical problem associated with “philosophical” issues

  • ex) pond: means different things to different people/institutions, thus different in different data sets: how to integrate?
  • ex) wildlife biologists (forest classified in terms of habitat vs foresters (forest classified in terms of resource assessment)
  • ex) different ways of defining what a road isuse metadata to assess these differences
  • they will always exist: ignoring them can lead to problems

Moral of the story: data are not reality!


“Data are compiled with a particular purpose in mind, and they reflect the assumptions and preconceptions of both the data collectors and data users.  They are, in fact, stories about the world that change depending on the teller.”

  • data is the basis of all GIS analysis
  • not a matter of good or bad data
  • not a matter of more or less accurate data
  • but a matter of the appropriateness of the data to a particular task
  • metadata clarifies the story the data can tell: who collected it with what assumptions under what conditions and for what purpose
  • vital to be critical and understand your data, not just take it as a given

Reading: Schuurman ch 4 & 5

February 18, 2014


Maurizio Cattelan  Love Saves Life  1995

Schuurman: GIS: A Short Introduction

Ch. 4: Bringing it All Together: GIS Analysis

GIS is often used to store data; analysis greatly extends the functionality of  GIS by allowing us to learn more about the stored data

Cadastral systems: property and attribute information (Delaware DALIS project)

  • storing data vs analysis (how many residential properties within 1000’ of river)

Examples of analysis:

  • measurement & distance calculation (perimeters, areas, line lengths)
  • point in polygon queries: does a point lie in an area?
  • shape analysis: shape of a line to assess difficulty of driving on a road
  • edginess analysis: deer habitat (prefer ediginess, forest grass boundary)
  • slope calculation

Overlay Analysis, Set Theory, Map Algebra

  • query: a question (show all owl locations > 500’ from road)
  • buffer: an area around a pt, line, area (show residences within 500’ of liquor lic. Appl.)
  • overlay analysis (find all soils of a particular type within a floodzone)
  • difficulty of polygon overlay: extensive calculation
  • set theory & map algebra: mathematical basis of GIS analysis

Spatial Analysis in the Field: Environmental Modeling

  • ex) modeling industrial pollution
  • predict the impact of a new industrial development in a particular location; help
  • in decision-making
  • air emission, noise, risk
  • link environmental modeling to spatial data

Building Intuitive Models: Multi-Criteria Evaluation

  • location decision analysis
  • find the best location for a new industrial development, given multiple criteria (away
  • from people because of pollution; near necessary transportation corridors, labor).
  • ex) locating a dump
  • factors and criteria: p. 110
  • resulting map: worse and better locations: p. 111

The Power of the Eye: Visualization and the New Cartography

  • ex) TB example

From Data to Analysis: A Case Study of Population Health

  • ex) population health: relating housing to health
  • ecological fallacy: aggregation or scaling introduces bias (p. 120)

MCE and Analysis

  • example of health vs density of population

Calculation and the Rationalities of GIS

  • critical perspectives

Ch. 5: GIS Training and Research

  • GIS is slow
  • Evolving research in GISci
  • Not ontology again!
  • Feminism and GIS
  • Systems vs Science

Some Resources:

Delaware GIS Data Exercise

February 18, 2014


Each student will create an ArcMap map with the following Delaware GIS data layers & describe (sentence or two) all data layers except those marked ‘ignore.’ (the ignore folders have data that is not relevant, or newer versions of the data are in other folders)

In essence, you are creating a very brief set of metadata (data about data) for all the available layers of information. There may be several shape files (.shp) in these folders, make sure to review all of them.

Keep your brain engaged: how might some of these layers be used in your course projects?

Put your metadata information in a blog posting.

You will use some of this data for your take-home mid term exam.

DUE: Wednesday February 26.

Delaware GIS Data Metadata is here.

If any data folders are missing please talk to your instructor.

Delaware GIS Data Layers:

Delaware_2008 and 2010 Ponds and Lakes





Delaware_Building Outlines

Delaware_ Census_Biock

Delaware_ Census_BiockGroup

Delaware_ Census_ Tract

Delaware_Economic Development Layers











Delaware_Master Point Coverage


Delaware_Natural_Heritage_ ODNR

Delaware_ Orthophoto _Detailed_2010



Delaware_Places of Interest


Delaware_Public Land Survey System







Delaware_ TaxDist

Delaware_ Topography

Delaware_ Townships

Delaware_ Townships_Historical

Delaware_ Watersheds_ ODNR

Delaware_ Wetlands

Delaware_ Woodland_ ODNR



Ohio Wesleyan Parcels