Analyzing violent crime, workflow

Workflow using ArcGIS Pro

Nota:

The steps below are based on the 1.2 release of ArcGIS Pro, but they should work fine for later software releases as well. You may notice slight differences in the maps below if you are using ArcGIS Pro 1.3 or later, the result of an improvement in how points are positioned within the fishnet mesh (Optimized Hot Spot Analysis) or within the cube (Create Space Time Cube). You may also notice additional parameters added after the 1.2 release for some of the tools used in this workflow. With the 2.0 release of ArcGIS Pro, the name of the Create Space Time Cube tool was changed to Create Space Time Cube by Aggregating Points. To follow the steps below, you may download and unzip the data in the data package provided at the top of this workflow, or improvise using your own data.

Create a hot spot map of violent crime densities.

1. If you haven't done so already, download and unzip the data package provided at the top of this workflow.
2. Open ArcGIS Pro and browse to the BrokenBottlesPkg.ppkx project package.
3. Once the project opens, find and open the Optimized Hot Spot Analysis tool. If the Geoprocessing pane isn't open, click the Analysis menu tab, then click the Tools button.

Sugerencia:

Whenever possible and appropriate, create your workflow output in a geodatabase rather than as a shapefile. Field names in shapefile output may be truncated, and there are other advantages to using a geodatabase to store your data.
4. Run the Optimized Hot Spot Analysis tool with the following parameters. The Analysis Boundary layer defines the study area.
   • Input Features : Violent Crime 2014
   • Output Features : the name of your output feature class such as ViolentCrimeHotSpots
   • Incident Data Aggregation Method : Count incidents within fishnet grid
   • Bounding Polygons Defining Where Incidents Are Possible : Analysis Boundary



While the tool runs, it reports the cell size it used for aggregation and the distance it used for analysis (the scale of the analysis). To see this information, hover over the progress bar below the Geoprocessing pane and click the icon to pop out the progress messages. You may resize the message pane by pulling on the lower right corner of the pop out window.



Notice that for this analysis the cell size is 1,375 feet and the scale of analysis is 4,563 feet (4,554 Feet with the most current software).f you are comparing multiple hot spot maps, you will want to make sure that the study area, cell size, and scale of analysis all match.

The output map created by Optimized Hot Spot Analysis is shown below:



2014 violent crime hot spots.

Create a hot spot map of liquor vendor densities.

1. Use the Optimized Hot Spot Analysis tool again with the following parameter settings. You will use the output from the violent crime hot spot analysis to define the study area and cell size.
   • Input Features : Liquor Vendors
   • Output Features : the name of your output feature class such as LiquorVendorHotSpots
   • Incident Data Aggregation Method : Count incidents within aggregation polygons
   • Polygons For Aggregating Incidents Into Counts : ViolentCrimeHotSpots



Nota:

Because you used the exact same study area for both analyses, the scale of analysis should match exactly. Be sure to check it, though. Sometimes, when the distributions of points are vastly different, there will be a mismatch. If you do see a mismatch, run Hot Spot Analysis on the output from Optimized Hot Spot Analysis, setting the Distance Band or Threshold Distance parameter explicitly to match the hot spot map you want to compare.

Now you can compare the hot spot maps to see where their activity spaces overlap.



Hot spots for violent crime and for liquor vendors.

Create a hot spot map of poverty.

While you may use the Enrich Layer tool to get poverty data, to ensure your results match those below and to avoid consuming credits, use the data provided in the data package you downloaded. The Enrich Layer tool always gives you the most current data available. When this workflow was created, the ACS poverty data was for 2009-2013.
1. Navigate to Poverty.lpk included with the data package you downloaded. Drag it on the map.
2. Find and open the Optimized Hot Spot Analysis tool a third time.
3. Set the parameters as follows and run the analysis.
   • Input Features: Poverty
   • Output Features: the name of your output feature class such as PovertyHotSpots
   • Analysis Field: 2009-2013 ACS Households with Income Below Poverty Level




Hot spot map of the number of households with income below the poverty level.

Overlay the hot spot maps to determine areas of overlap.

1. Find and open the Select Layer By Attribute tool. You will run the tool on all three hot spot maps, each time selecting records where the Gi_Bin field is equal to 3 (a three for this field indicates a statistically significant hot spot at the 99 percent confidence level). The Gi_Bin field name will reflect the scale of analysis (4554 for the most current version of the software).



2. Next, find and open the Intersect tool. Add all three layers as Input Features, provide a name for the output, such as iCrimeLiquorPoverty, and run the analysis.



Licencia:

If you are working with a basic (ArcView) or Standard (ArcEditor) license, first use Intersect for the selected Violent Crime and Liquor cells, then intersect the result with the selected poverty cells.

3. Clear the selections, and turn off all other layers in order to see the output showing the overlapping locations. These locations will be your proposed areas for a liquor moratorium.





Areas where violent crime, liquor vendor, and poverty hot spots overlap.

Create a space-time cube and analyze the crime trends within it.

1. Find and open the Create Space Time Cube tool.
2. For version 1.4 or later of ArcGIS Pro, to ensure the cube output aligns with the hot spot output, you must set the Processing Extent to match the Analysis Boundary layer. Click the Environments tab and change the Extent to Analysis Boundary.



3. Set the Create Space Time Cube Parameters as follows and run the analysis. The cube is a netCDF file, so it must be created in a folder rather than inside a file geodatabase. Setting the Distance Interval to match the hot spot map cell size will allow you to overlay the crime trends with hot spot maps later. For version 1.4 of ArcGIS Pro only, you must convert 1375 US Feet to 1375 International Feet (1375.00275).
   • Input Features: Violent Crime 2014
   • Output Space Time Cube: the name of your output cube such as ViolentCrimeCube.nc
   • Time Field: Date
   • Time Step Interval: 4 Weeks
   • Time Step Alignment: End time
   • Distance Interval: 1375 Feet; for version 1.4 and 2.0 of ArcGIS Pro only, use 1375.00275 instead (see note below).



The tool will report it completed successfully but will not add new layers to the map.

Nota:

With version 1.4 and 2.0 of ArcGIS Pro, the space time cube will not align with the output from Optimized Hot Spot Analysis unless the cell size is entered in US Feet. US Feet are slightly larger and have more precision that International Feet. With later versions of the software, Feet parameters will automatically be interpreted as US Feet.

Precaución:

If you want a valid space-time analysis of incident data (like crime events), make sure each bin in the space-time cube is exactly the same size. Selecting Months for the Time Step Interval, for example, will result in some bins having more days than others (31 days for March, but only 30 days for April). This will bias your analysis because months with more days will likely have more incidents. Use Days, Weeks, or Years instead of Months.

Sugerencia:

Use the End time Time Step Alignment. If the incident range of dates doesn't divide evenly into your Time Step Intervals, you want the bias to be associated with the first (oldest) time period rather than the last (most recent) time period. For example, suppose your data covers fifteen and a half weeks and you select 1 week for your Time Step Interval. If you select Start time for the Time Step Alignment, all of the bins will have a full week, except the bins for the last (most recent) time period; it will only have half a week.

4. Find and open the Emerging Hot Spot Analysis tool.
5. Set the following parameters and run the analysis.
   • Input Space Time Cube: ViolentCrimeCube.nc
   • Analysis Variable: COUNT
   • Output Features: the name of your output feature class such as ViolentCrimeTrends
   • Neighborhood Distance: 0.5 Miles
   • Neighborhood Time Step: 1
   • Polygon Analysis Mask: Analysis Boundary



6. Examine the results. The trend categories are defined as follows.

Pattern Type	Definition
New Hot Spot	A location that is a statistically significant hot spot for the final time step and has never been a statistically significant hot spot before.
Consecutive Hot Spot	A location with a single uninterrupted run of statistically significant hot spot bins in the final time-step intervals. The location has never been a statistically significant hot spot prior to the final hot spot run and less than ninety percent of all bins are statistically significant hot spots.
Intensifying Hot Spot	A location that has been a statistically significant hot spot for ninety percent of the time-step intervals, including the final time step. In addition, the intensity of clustering of high counts in each time step is increasing overall and that increase is statistically significant.
Persistent Hot Spot	A location that has been a statistically significant hot spot for ninety percent of the time-step intervals with no discernible trend indicating an increase or decrease in the intensity of clustering over time.
Diminishing Hot Spot	A location that has been a statistically significant hot spot for ninety percent of the time-step intervals, including the final time step. In addition, the intensity of clustering in each time step is decreasing overall and that decrease is statistically significant.
Sporadic Hot Spot	A location that is an on-again-off-again hot spot. Less than ninety percent of the time-step intervals have been statistically significant hot spots and none of the time-step intervals have been statistically significant cold spots.
Oscillating Hot Spot	A statistically significant hot spot for the final time-step interval that has a history of also being a statistically significant cold spot during a prior time step. Less than ninety percent of the time-step intervals have been statistically significant hot spots.
Historical Hot Spot	The most recent time period is not hot, but at least ninety percent of the time-step intervals have been statistically significant hot spots.
New Cold Spot	A location that is a statistically significant cold spot for the final time step and has never been a statistically significant cold spot before.
Consecutive Cold Spot	A location with a single uninterrupted run of statistically significant cold spot bins in the final time-step intervals. The location has never been a statistically significant cold spot prior to the final cold spot run and less than ninety percent of all bins are statistically significant cold spots.
Intensifying Cold Spot	A location that has been a statistically significant cold spot for ninety percent of the time-step intervals, including the final time step. In addition, the intensity of clustering of low counts in each time step is increasing overall and that increase is statistically significant.
Persistent Cold Spot	A location that has been a statistically significant cold spot for ninety percent of the time-step intervals with no discernible trend, indicating an increase or decrease in the intensity of clustering of counts over time.
Diminishing Cold Spot	A location that has been a statistically significant cold spot for ninety percent of the time-step intervals, including the final time step. In addition, the intensity of clustering of low counts in each time step is decreasing overall and that decrease is statistically significant.
Sporadic Cold Spot	A location that is an on-again-off-again cold spot. Less than ninety percent of the time-step intervals have been statistically significant cold spots and none of the time-step intervals have been statistically significant hot spots.
Oscillating Cold Spot	A statistically significant cold spot for the final time-step interval that has a history of also being a statistically significant hot spot during a prior time step. Less than ninety percent of the time-step intervals have been statistically significant cold spots.
Historical Cold Spot	The most recent time period is not cold, but at least ninety percent of the time-step intervals have been statistically significant cold spots.
No Trend Detected	Does not fall into any of the hot or cold spot patterns defined above.

For your analysis there are no cold spots.



2014 Violent Crime Trends

To see a 3D visualization of the crime trend analysis data, you will create a new scene.

7. Click the Insert menu tab. Expand the New Map button and select New Scene.



8. Find and open the Visualize Space Time Cube in 3D tool and run it with the following parameters.
   • Input Space Time Cube : the path to space time cube you created above, ViolentCrimeCube.nc
   • Cube Variable : COUNT
   • Display Theme : Hot and cold spot results
   • Output Features : CrimeTrends3D



Zoom in and tilt the 3D scene so you can see the individual bins in relation to the base map. If you look closely, you may notice that some of the bins at the bottom of each time series are below the base map. This is because the base map is using elevation while the bins are not. To see the full time series at each location you just need to shut down the elevation service. Don't worry, though, the next time you insert a new scene, it will have the appropriate elevation services running.

9. Right click on Scene in the Contents pane and select Properties.



10. Select Elevation Surface, expand Ground, and remove all active elevation surfaces by clicking the red X next to each one.



11. Explore the 3D scene.



3D visualization of violent crime trends.

Create a hot spot map of unemployment rates.

1. If you are still in Scene, activate the Map view to work again with your 2D data.
2. Navigate to Unemployment.lpk included with the data package you downloaded. Drag it onto the map.

Nota:

The Enrich Layer tool always gives you the most current data available. When this workflow was created, the unemployment rate data was for 2015.

3. Open the Optimized Hot Spot Analysis tool.
4. Set the parameters as follows and run the analysis.

   • Input Features: Unemployment
   • Output Features: the name of your output feature class such as UnemploymentRateHotSpots
   • Analysis Field: 2015 Unemployment Rate



2015 unemployment rate hot spots.

Overlay the violent crime trend map with the unemployment rate hot spot map to determine areas of overlap.

1. Find and open the Select Layer By Attribute tool. You will use it once to select intensifying, persistent, and consecutive hot spots (Pattern Type COUNT is Equal to Consecutive Hot Spot Or Pattern Type COUNT is Equal to Intensifying Hot Spot Or Pattern Type COUNT is Equal to Persistent Hot Spot) and a second time to select the most intense unemployment rate hot spots (Gi_Bin Fixed 4556_FDR is equal to 3).



Precaución:

Be sure to click the Add button after creating the expression, otherwise all the features in the layer will be selected.

2. Next, find and open the Intersect tool. Add the violent crime trends and unemployment rate hot spot maps with their active selections, provide a name for the output results such as iCrimeUnemp, and run the analysis.





Areas where unemployment rate hot spots overlap persistent, consecutive, or intensifying violent crime trends.

Finally, select the public high schools within a quarter mile of the overlapping areas.

1. Find and open the Select Layer By Location tool.
2. Set the parameters as follows:
   • Input Feature Layer: Public High Schools
   • Relationship: Within a distance
   • Selecting Features: iCrimeUnemp
   • Search Distance: 0.25 Miles



3. Use the Copy Features tool to copy the selected high schools to a new feature class (this is optional, but it makes mapping and creating reports a bit easier).
   • Input Features: Public High Schools
   • Output Feature Class: the name of your output feature class such as SelectedHighSchools

You are now ready to make your final recommendations.

• Analyzing violent crime - analysis overview
• References and resources for learning more