The GeoMed/Epi Project:

Glossary of Course Terms

The Glossary which follows is an enhanced version of an appendix to the first reference cited below. We thank all our sources, but the authors of that one in particular.

Other glossaries.


Cartesian coordinate system

points, lines, and polygons are most commonly defined on maps using x, y Cartesian coordinates such as longitude/latitude based on principles of Euclidean geometry. This Cartesian coordinate system is the most commonly used tool for measuring and analysing various properties of spatial location.

Choropleth mapping

shows data value for a predefined area such as a census tract (or watershed) or statistically defined Thiessen polygon, where the entire area is shaded to a representative summary statistic. The variation of observations within a single boundary is assumed minimal, for example, there is no accommodation for differences in standard deviation, thus depicting a homogeneous surface. Any major change in attribute value is considered to occur at the boundary.

Census block

the smallest area for which the Census Bureau collects and tabulates decennial census data. A geographic area normally bounded on all sides by visible features such as streets, roads, streams and railroad tracks, and occasionally by non-visible boundaries such as city or county limits, property lines, and short, imaginary extensions of streets or roads.

Census block group

a combination of census blocks that is a subdivision of a census tract or census block numbering area (BNA). A block group consists of all census blocks whose numbers begin with the same digit within a census tract or BNA.

Census block numbering area (BNA)

a statistical subdivision of a county delineated for the purpose of grouping and numbering census blocks in areas where local communities have not established census tracts. BNAs were delineated by state officials and the Census Bureau for the 1990 Census.

Census tract

a small, relatively permanent statistical subdivision of a county in a metropolitan area or a selected non-metropolitan county, for presenting decennial census data. Census tracts are designed to be relatively homogeneous with respect to population characteristics, economic status, and living conditions, and to contain between 2500 and 8000 inhabitants.

Continuous Measurement Survey

a proposed system that would be facilitated by the Census Bureau's development of a Master Address File, to provide current long-form (sample) data on a continuous basis.

Dasymetric maps

a type of choropleth map in which natural boundaries are used in place of political or other official boundaries. Natural boundaries in an urban setting, for example, might be formed by parks, areas of multi-unit dwellings, or industrial zones.

Delauney triangles

the process of converting point data into contours that uses a triangulated network which joins neighbouring points together. Interpolation of values observed at point locations produces an approximation of the distribution.

Digital Line Graph (DLG) and Digital Elevation Model (DEM)

two major U.S. Geological Survey (USGS) digital data file structures. DLG is line map information in digital form. These files include information on planimetric base categories, such as transportation, hydrography, and boundaries. DEM files consist of a sample array of elevations for a number of ground positions that are at regular spaced intervals.

Digital orthophotos

are digital photographs or 'photo maps' that result from processing aerial photographs to remove image distortion and displacement due to perspective, camera tilt and terrain relief These have one scale (even in varying terrain) and show actual detail (not by lines and symbols) of all the information captured in the photograph. True distances, angles, and areas may be measured directly without making corrections for image displacement. Orthophotography from satellites should be possible in the near future.

Empirical Bayes (EB) mapping

a parametric statistical procedure to stabilize statistics, prior to mapping, by Bayesian modelling which 'shrinks' the statistics from areas with a small population toward an overall mean. EB-stabilized statistics may not provide a good estimate of the distribution of the actual underlying data unless the presumed parameter model is approximately correct.

Geocoding

digital procedure for finding map coordinates that correspond to data attributes of features. For example, address geocoding is the ability to reference a street address or street intersection to a location on the map. The Census Bureau's TIGER system provides a national computer-readable map database for geocoding operations.

Geodetic control

a network of surveyed and monumented points on the earth's surface whose locations are established in accordance with national accuracy standards. A state plane coordinate system comprises a system of x, y coordinates for each state and is commonly used in GIS.

Grid

provides graph theory relationship principles to mapping by use of a grid mesh to define a regular but arbitrary polygonal framework for 'holding' geographic data. The grid technique inherently involves association with a co- ordinate system, but it does not necessarily require precise association.

Global Positioning System (GPS)

hand-held or vehicle-mounted satellite transmission technology that allows for real-time digital locational positioning (longitudeAatitude/altitude) on the earth's surface, with high metric accuracy. Positional readings are entered into digital GIS mapping databases and displays.

Internet (and related terms)

A worldwide system for linking smaller computer networks together. Internet connected networks use a particular set of communication standards to communicate, known as TCP/IP. The World Wide Web (WWW) is a bypertext system which links images, sounds, and text, collectively known as hypermedia. A user navigates the WWW, travelling from page to page, and browses information via mouse clicks on text and images. Uniform Resource Locator (URL) is an address on the World Wide Web.

Isarithmic mapping

suited for mapping continuous data, the method examines the gradual change between observations by joining control points of equal value. These values may be interval-level measurements, ratios, or correlation coefficients. The most familiar of isarithmic techniques is the contouring approach where observations are considered to be exact representations of an attribute.

Kernel estimator

relates to a class of density estimators (that avoid dependence on essentially arbitrary spatial units) called 'kernel methods' from which to derive a smoothing estimator or parameter; 'adaptive' kernel estimation provides that the bandwidth parameter should be spatially variable to account for variations in density in the study region.

Kriging

this mapping method represents the variable under study as a continuous process, unconstrained by the borders of geographic units and where sudden transitions between levels of two neighbouring areas are avoided. It provides the variance of the estimated values from the spatial variability of the actual data, for example, a standard error map, and these error maps can be useful to introduce new sample values for analysis.

Master Address File (MAF)

The Census Bureau's permanent list' of addresses for individual living quarters that is linked to the TIGER database and will be continuously maintained through partnerships with the U.S. Postal Service, and Federal, State, regional and local agencies, and private sector. (The MAF will eventually include addresses for business establishments.)

Non-parametric spatial smoothers

mathematical methods of smoothing data spatially that are not dependent on parametric statistical distributions, for example, nearest-neighbour estimators, Markov point processes.

Polygon

a closed, two-dimensional figure with three or more sides and intersections. For example, a polygon could be represented as an enclosed geographic area such as a land parcel or political jurisdiction.

Relational database model

database structure where each record is a 'row' and each field is a 'column'. A set of rows stored under columns comprise a 'table'. For graphical objects, the graphical information and corresponding coordinates are attached to each row in the relational database table.

Remote sensing

the analysis and interpretation of the earth's landscape and resources using aerial photography or satellite imagery. This is especially useful for public health analysis in the study of disease host and vector habitats, extent and magnitude of disease events and natural disasters, and changes in observations over time.

Slope, aspect

the term slope addresses the steepness of an area while aspect is related to the direction in which an area is oriented. The calculation of the angle of the topographic slope is generally based on a digital frame file with its grid of polygons and the classification of that surface of slope variation into ranges. The calculation of topographic slope can be displayed with colour or shading,

Spatial autocorrelation

tests for spatial autocorrelation are designed to quantify the extent of clustering and are measures of similarity between association in value (covariance, correlation, or difference) and association in space (contiguity). A listing of spatial autocorrelation statistics for increasing orders of contiguity is referred to as a spatial correlogram.

Spatial stochastic process

formalizes the way in which spatial association is generated or generally expresses how observations at each location depend on values at neighbouring locations, that is, on the spatial lags. As with time series analysis, spatial stochastic processes can be classified as spatial autoregressive (SAR) or spatial moving average (SMA) processes.

Suitability analysis

a variant of spatial analysis in which GIS map layers are integrated to form a composite choropleth map for decision making. Often used in site selection, it results from an application of one or more mathematical relations (functions, transformations) to the attributes of one or more maps. Examples are maps derived from weighted intersection overlay and weighted multidimensional scaling.

Theissen Polygons

See Voronoi.

TIGER database

A computer file that contains geographic information representing: the position of roads, rivers, railroads, and other census-required map features; attributes associated with each feature, such as feature name, address ranges, and class codes; position of the boundaries for those geographic areas that the Census Bureau uses in its data collection, processing, and tabulation operations; and attributes associated with those areas, such as their names and codes. This file is stored in multiple partitions, for example, counties or groups of counties, although it represents all U.S. space (including Puerto Rico, and the Outlying Areas) as a single seamless data inventory.

TIGER/Line files

public extracts of selected geographic and cartographic information from the Census Bureau's TIGER (Topologically Integrated Geographic Encoding and Referencing) database.

Topography

the collective features on the surface of the earth, including relief, hydrography and cultural features. Topographic maps of the U.S. are produced by the U.S. Geological Survey for GIS use. In public health, topographic maps can reveal disease associations with elevation, surface water, wind direction, solar exposure, time and other factors.

Topology

the relationship between points, lines, and geometric forms that remain consistent throughout spatial operations in a digital mapping environment.

Trend surface analysis

a regression model that has as its explanatory variables the elements of a polynomial in the coordinates of the observations. It is useful to show broad features of the data but can also be used to remove those broad features to allow some other technique to work on the residuals or as part of the computation for stochastic process prediction.

Triangulated Irregular Network (TIN)

a contouring method of linear interpolation that bases its predicted surface on the flat plane that can be fitted to any three non-collinear points. It assumes that point observations represent exact values and that the sampling has been carried out at sufficient detail to fit a continuous surface.

Voronoi or Theissen Polygons

method of mathematically transforming point data into thematic maps based not on predetermined reporting units but rather on proximity of the distribution of points. One of the points which define the edges of a Thiessen polygon is always the nearest neighbour to the point in the centre of the polygon.


REFERENCES
  1. Croner et al. 1996. Geographic Information Systems (GIS): New Perspectives in Understanding Human Health and Environmental Relationships. Statistics in Medicine 15:1961-1977.
  2. Snow, J. On the Mode of Communication of Cholera, 2nd edn, Churchill, London, 1855. Reproduced in Snow on Cholera, Commonwealth Fund, New York, 1936. Reprinted by Hafner, New York, 1965.
  3. Croner, C. M., Pickle, L. W., Wolf, D. R. and White, A. A. A GIS approach to hypothesis generation in epidemiology, Technical Papers, ASPRS/ACSM/RT92, Proceedings: GIS and Cartography, 3,275--283 (1992).
OTHER GLOSSARIES

Website maintained by Andy Long. Comments appreciated.
longa@nku.edu