Pitfalls of spatial data Flashcards
Spatial autocorrelation diagnostic measures
joins count statistics
moran’s I
geary’s C
variogram cloud
Spatial autocorrelation structure
spatial variation across a study area
First order spatial variation
occurs when observations across a study region vary from space due to changes in the underlying properties of the local environment
Second order spatial autocorrelation
due to local interaction effects between observations
MAUP
the same basic data yield different results when aggregated in different ways
MAUP components
scale effect
zoning effect
MAUP scale effect
consequence of spatial aggregation
MAUP zoning effect
results from the multitude of zoning schemes that could be constructed and used at any given scale
Boundary issues in statistics
simpson’s paradox
Boundary issues in geostatistics
change of support problem
Boundary issues in geography
MAUP
Boundary issues in sociology
ecological fallacy
Ecological fallacy
when its inferred that data for areas under study can be applied to individuals in that area
relationships at one geographic scale are not necessarily true at other geographic scales
Four common statistical ecological fallacies
confusion between ecological correlations and individual correlations
Confusion between group average and total average
Simpson’s Paradox
Confusion between higher average and higher likelihood
Confusion between ecological correlations and individual correlations
the correlation between individual variables is deduced from the correlation of the variables collected for the group to which those individuals belong
Confusion between group average and total average
the average for a group is approximated by the average in the total population divided by the group size
Simpson’s Paradox
when comparing two populations divided in
groups of different sizes, the average of
some variable in the first population can be
higher in every group and yet lower in the
total population
Confusion between higher average and higher likelihood
Higher average implies higher likelihood, e.g. the ecological fallacy results from incorrectly assuming that individuals in wealthier states are more likely to be
wealthy
Preventing ecological fallacy
Be aware of the process of aggregating or disaggregating data
may conceal the variations that are not visible at the larger
aggregate level
Scale implications
anticipate influence of geography
assume not all locations will respond in the same way and responses will show spatial pattern
sampling should be dispersed
Non-uniformity of space
spatial features are not evenly distributed
Edge effects occurrence
where an artificial boundary is imposed on a study
Edge effects
features near boundary have fewer neighbors and are less reliable
Handling edge effects
use core study area
buffer study area
vary weights for features near boundary
