Questions: Geographic Information Systems and Spatial Analysis

Question 1

A GIS analyst overlays a layer of childhood asthma rates with a layer showing proximity to highways and observes a strong spatial correlation. What is the most appropriate next step?

• A. Publish the finding as evidence that highways cause asthma
• B. Dismiss the correlation because GIS cannot prove causation
• C. Develop a theoretical mechanism linking vehicle emissions to respiratory health and seek additional supporting evidence
• D. Reclassify the asthma data into fewer categories to reduce the visual correlation

Answer: (C)

Spatial correlation in GIS is a hypothesis generator, not a causal proof. The correct next step is to articulate a plausible causal mechanism (vehicle emissions → particulate matter → airway inflammation) and test it with additional data — for instance, air quality measurements, longitudinal health records, or natural experiments. Dismissing the correlation ignores real signal; publishing it as causal proof commits the ecological fallacy.

Question 2

Choosing the Mercator projection for a world map is a neutral technical decision with no political implications.

• T. True
• F. False

Answer: False

The Mercator projection severely distorts area at high latitudes, making Greenland appear as large as Africa (in reality, Africa is about 14 times larger). This distortion systematically exaggerates the visual size of Europe and North America relative to equatorial regions. The Peters projection — which preserves area — was explicitly developed as a political counter to the Mercator. Every projection choice privileges some properties over others; those choices have implications for how viewers perceive the relative importance of different parts of the world.

Question 3

What is the difference between vector and raster data formats in GIS, and give an example of when each would be appropriate?

Model answer: Vector data represents features as discrete points, lines, or polygons with associated attributes; raster data represents the world as a grid of cells, each with a value. Vector suits discrete features like roads or country borders; raster suits continuous phenomena like elevation or land surface temperature.

Vector format is efficient for discrete, bounded features — a city boundary is a polygon, a river is a line, a weather station is a point. Raster format is natural for phenomena that vary continuously across space — elevation models, satellite imagery, and temperature surfaces are all grids where each pixel carries a measurement. GIS analyses often require converting between formats or overlaying both types, which is why understanding the distinction matters for choosing the right tool and interpreting outputs correctly.