Spatial Data Analysis Flashcards
What is the goal of spatial data analysis?
The goal of spatial data analysis is to explore and express dependencies between observations in space
Use examples to explain the various components of data quality: Accuracy, precision, consistency and completeness.
Accuracy measures the discrepancy between the encoded value and the actual value of a particular attribute. Accuracy is the inverse of error. There are spatial, temporal and thematic accuracy.
A databased dated 1992 depicts a two-lane road. However, the road was converted to a four-lane road in 1991.
Precision refers to the amount of detail that can be discerned. It is also known as granularity or resolution. Spatial resolution is the minimum size of objects on the ground that can be discerned.
A satellite image may show a vehicle (high precision/resolution) or a less clear image of the same thing (low precision/resolution)
Consistency refers to the absence of contradictions in data. Consistent data is not necessarily accurate or precise.
Spatial r: Only one point may exist at a given location. Lines must intersect at nodes. Polygons are bounded by lines.
Temporal consistency: Only one event can occur at one time at a given location. An inconsistency exists if a different entity appears at the same location on two maps of the same date.
Completeness is the relationship between objects in the database and those in the real world. It is defined in terms of “errors of omission over space, time or theme”.
Model completeness is the agreement required between the database specification and the abstract universe for a particular application. It is an aspect of usability.
What is the difference between resolution and sampling rate?
Resolution is the fineness of detail. It refers to the time interval required to obtain spectral reflectance data for one pixel.
Sampling rate is the frequency of repeated coverage.
For example, motion pictures have a resolution of a thousandth of a second (per frame) but a sampling rate of 24 frames per second.
What does it mean by spatial autocorrelation?
Spatial autocorrelation measures the degree of dependency among observations in a study area. This can be done using Moran’s Index.
What is the Modifiable Areal Unit Problem?
The Modifiable areal unit problem (MAUP) is a source of statistical bias that can affect the results of statistical hypothesis tests.
It affects results when point-based measures of spatial phenomena (e.g. household income) are aggregated into districts (e.g. city boundaries, neighbourhood boundaries etc.). This causes a smoothing effect.
The size and forms of these boundaries are modifiable, which can generate different results
What is the ecological fallacy?
Ecological fallacy is an error that can occur when a researcher makes an inference about an individual based on aggregate data for a group.
The fallacy assumes that all members of a group exhibit characteristics of the studied group.
For example if it is assumed that lower-income people are more likely to commit a crime, based on a correlation between income and crime, it is an ecological fallacy.
What is the scale?
It deals with the relationship between ecological fallacy and modifiable areal unit problem.
- The geographical scale can affect the observations and
interpretation
- Empirical knowledge about the extensions of spatial
entities / phenomena is important for finding the optimal
scale.
- It is a difficult data mining issue for unknown things.
What is the non-uniformity?
Spatial information is distributed with different concentrations, which can cause redundant or missing samples in different places. The hypothesis that any of the events could have occurred anywhere in the study area does not hold.
What is the edge effect?
Edge effects refer to situations where patterns of interaction or interdependency across the borders of a bounded region are ignored or distorted. It happens when artificial boundaries are imposed on a study in order to keep the study manageable.
It is possible to reduce the edge effect by allowing tiles to overlap each other to some extent at their borders.
