🎯 Learning Objectives
- Calculate Global Moran's I to determine if data is clustered, dispersed, or random.
- Perform Hot Spot Analysis (Getis-Ord Gi*) to map statistically significant clusters.
- Understand the concept of Z-Scores and P-Values in a spatial context.
- Compare raw density maps vs. statistical significance maps.
📂 Scenario: The Crime Wave
The Mayor claims that crime is "out of control everywhere." You, the GIS Analyst for the Police Department, know that crime is usually concentrated in specific hotspots. You need to prove statistically where the 911 calls are clustering to allocate patrol patrols effectively.
lab09_hotspots.zip
Contains: 911_Calls.shp, City_Boundaries.shp, Census_Blocks.shp
🛠️ Step-by-Step Instructions
Select your preferred GIS platform to view instructions:
Spatial Join (prep)
Hot Spot analysis works best on polygons (counts), not just points.
1. Spatial Join 911_Calls (Join Features) to Census_Blocks (Target Features).
2. Output: Blocks_with_Crime_Count.
Global Moran's I
1. Search for Spatial Autocorrelation (Moran's I).
2. Input: Blocks_with_Crime_Count.
3. Input Field: Join_Count.
4. Run. View the HTML Report. Is the Z-score high (> 2.58)? Then it is clustered.
Hot Spot Analysis (Gi*)
1. Search for Hot Spot Analysis (Getis-Ord Gi*).
2. Input: Blocks_with_Crime_Count.
3. Input Field: Join_Count.
4. Run. The map will colorize statistically significant Hot Spots (Red) and Cold Spots (Blue).
✅ Submission & Assessment
To complete this lab, you must submit:
- Report Screenshot: The Moran's I report showing a high Z-Score.
- Hot Spot Map: Clearly showing the Red/Blue clusters.
- Write-up: Explain the difference between a "Heat Map" (Density) and a "Hot Spot Map" (Significance). Why is the Hot Spot map better for the Mayor?