Cumulative from Midterm

Question 1

Types of data found in the American Community Survey, CDC Wonder

Answer 1

Am Comm Survey- similar to Census tract. 5 year demographic variables, 10 year summaries of full population at census tract- national level

CDC Wonder- Mortality, Morbidity, and Environmental data layers at county level

Question 2

Shapefiles and Geodatabases

Answer 2

Shapefile- group of files containing data for drawing boundaries, knowing which projection map file is in, and any data associated with each polygon boundary

Geodatabases - a way to store GIS info in one large file, can contain multiple point, polygon, or polyline layers

Question 3

Landscape Epidemiology

Answer 3

combined study of disease ecology with heterogeneous landscape

Question 4

SatScan

Answer 4

-Moving bandwidth
-Monte Carlo
-Maximum Likelihood Ratio using 999 observations
-Looking for high and low clusters at the bottom 5%
and top 95%

-Types
-Bernoulli: Cases and non-cases
-Discrete Poisson: cases per population, incidence
-Multinomial: All belong to different non-ranked
categories i.e. diabetes
-Ordinal: Categorized and ranked i.e. cancer stages
-Exponential: Used for continuous data such as
survival times
-Normal: Case data continuous, birth weight per
person
-Continuous Poisson: Looks for global clustering, as in
spatial autocorrelation

Question 5

Spatial regression

Answer 5

Finding the relationship between exposure and outcome, with or without covariates. Works best when all data is at the same level i.e. county-level polygon

Question 6

Land Use Regression vs. IDW

Answer 6

Land Use Regression: an algorithm often used for analyzing pollution, particularly in densely populated areas. The model is based on predictable pollution patterns to estimate concentrations in a particular area.

IDW: Simple interpolation using distance (weighted average)