# GIS and Spatial Analysis

History I · Geography · https://tryals.app/learn/history-i/gis-and-spatial-analysis

## Snow's Map and GIS

In 1854, John Snow mapped London cholera deaths and had the Broad Street pump handle removed. The outbreak was already declining, and Snow already believed cholera was waterborne; the map did not generate his hypothesis.

**GIS** does not answer questions. **It makes spatial arguments visible, checkable, and falsifiable**. Its true power lay in the anomalies—such as surviving brewery workers and a widow who drank the water from afar—which made miasma theory untenable.

GIS operates by stacking **layers** over shared coordinates. Each layer is ordinary, but their overlays reveal historical relationships.

## Spatial Operations and Traps

| Operation | The question it answers |
| --- | --- |
| **Overlay** | What coincides with what? |
| **Buffer** | What is within X of this? |
| **Least-cost path** | Where would movement go given terrain? |
| **Viewshed** | What can be seen from here? |
| **Density** | Is the pattern clustered? |

**Least-cost path** computes routes across a cost surface of slopes and rivers. **Viewshed** determines line-of-sight visibility for towers and castles.

**The central danger:** GIS renders guesses and sketches with sub-metre precision. Avoid the **ecological fallacy**, assuming aggregate district data applies to individuals.

## Practice questions

10 of this lesson's 13 practice questions, with answers. The full set is in the app.

### 1. What is actually true about Snow’s 1854 cholera map?

A. Snow already believed cholera was waterborne before drawing it
B. The map generated the hypothesis that cholera was waterborne
C. Its power lay in anomalies, the brewery workers, the widow across town
D. The outbreak was already declining when the pump handle was removed

**Answer:** A. Snow already believed cholera was waterborne before drawing it; C. Its power lay in anomalies, the brewery workers, the widow across town; D. The outbreak was already declining when the pump handle was removed

**Why:** The map was a rhetorical instrument in an argument Snow was already having. It did not find the answer, it made a spatial argument checkable, which is a different and more useful thing.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/what-is-actually-true-about-snows-1854-cholera-map

### 2. A least-cost model is most useful to a historian when the route it predicts disagrees with the route actually taken.

**Answer:** True

**Why:** True. Agreement merely confirms that people avoided hills. Disagreement identifies where something other than cost governed — politics, ritual, ownership, a decision taken for reasons the terrain cannot express — and that residual is the historically interesting part.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/a-least-cost-model-is-most-useful-to-a-historian-when-the-route-it

### 3. A GIS analysis finds a strong spatial correlation between two historical variables at parish level. What is the standing objection?

A. Spatial autocorrelation between neighbouring places
B. That parish boundaries changed over time, making the units incomparable
C. That correlation cannot establish causation in any analysis
D. That GIS software introduces projection errors into the coordinates

**Answer:** A. Spatial autocorrelation between neighbouring places

**Why:** Nearby places share soil, markets, dialect and lords, so treating parishes as independent samples inflates significance. Tobler's first law made into a statistical problem. Option 3 is true of every analysis and so distinguishes nothing; this objection is specific to spatial data and has specific remedies.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/a-gis-analysis-finds-a-strong-spatial-correlation-between-two

### 4. A GIS layer labelled 'as they were in the eighteenth century' can misrepresent the past even when every feature on it is accurate.

**Answer:** True

**Why:** True. Features drawn from across a century may never have coexisted, so the map shows a moment that never happened. Each element can be individually correct while the composite is false, which is why historical GIS layers carry a date range and are queried by date rather than displayed whole.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/a-gis-layer-labelled-as-they-were-in-the-eighteenth-century-can

### 5. A map showing that two aggregate variables correlate across districts supports a claim about the individuals in them.

**Answer:** False

**Why:** False. That is the ecological fallacy: districts with more of X also having more of Y says nothing about whether the same *people* have both. The classic case is Robinson's 1950 finding that literacy correlated with immigration by state while the individual correlation ran the opposite way.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/a-map-showing-that-two-aggregate-variables-correlate-across-districts

### 6. Sort each claim: **supported by GIS** or **an artefact of its rendering**?

**Answer:**

- Supported by GIS: This signal tower could see the next one, The cheapest route over this terrain follows the valley, These deaths cluster more tightly than chance predicts
- Artefact of rendering: This parish boundary ran exactly along this line, This disputed frontier was precise to the metre, Individuals in this district shared the district’s average trait

**Why:** GIS renders your assumptions beautifully. The right-hand column is what happens when crisp output is mistaken for precise evidence, including the ecological fallacy at the bottom.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/sort-each-claim-supported-by-gis-or-an-artefact-of-its-rendering

### 7. Two GIS layers clustering in the same district establishes a link between them.

**Answer:** False

**Why:** False. Both may cluster on a third thing, population density, so anything counted rather than rated will appear to coincide. Snow's cholera map worked because he compared *rates* against a specific water supply and had a mechanism, not because the dots overlapped.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/two-gis-layers-clustering-in-the-same-district-establishes-a-link

### 8. Why has viewshed analysis been uncomfortable for writing about castles in "commanding positions"?

A. It confirmed that castles were sited to dominate their surroundings
B. It turns assertion into a testable claim
C. It showed that visibility was irrelevant to medieval siting
D. It proved signal towers were decorative rather than functional

**Answer:** B. It turns assertion into a testable claim

**Why:** Could this tower see the next? Was the castle sited to be seen, or to see? These were assertions, and now they are tests, which a good deal of confident writing has failed.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/why-has-viewshed-analysis-been-uncomfortable-for-writing-about

### 9. Parish boundaries digitised from an eighteenth-century estate sketch render crisply on a modern map. Why is that appearance of precision misleading?

A. Because precision of display is not accuracy of source
B. Because eighteenth-century surveyors did not use standard units
C. Because parish boundaries were legally undefined before the nineteenth century
D. Because georeferencing always introduces errors of several metres

**Answer:** A. Because precision of display is not accuracy of source

**Why:** A crisp line asserts a confidence the source cannot support. This is the standard GIS trap: the display resolution is set by the software, and the accuracy by a hand sketch.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/parish-boundaries-digitised-from-an-eighteenth-century-estate-sketch

### 10. Complete the principle.

**Answer:** GIS does not answer questions, it makes spatial arguments **checkable**. Its value is in the overlay: a **least-cost** path shows where movement would actually go given terrain, and where it disagrees with a real road you have found a reason that was not about cost. But georeferencing imports old errors and renders them as **precision**, so a crisp line is a claim about evidence you may not have.

**Why:** Checkable, least-cost, precision. Snow’s map is the model for all of it: a rhetorical instrument that made an argument falsifiable, in a dispute he was already having.

Page: https://tryals.app/practice/history-i/gis-and-spatial-analysis/complete-the-principle