Yearly Average Calls Per Neighborhood (2013-2019)

Peer Feedback

I received suggestions to use average calls per year instead of totals as it would give a better look at how neighborhoods differ each year. Also, I received some feedback on what kind of interactivity could help provide better insight. So I implemented sorting (descending) based on call type group to sort the heatmap cells in order for users to see which neighborhoods have the highest average calls per year in each category. I also implemented details on demand when hovering over a datapoint to see the actual number of calls for that call type group.

Data Wrangling

Data was wrangled first prior to download, through the editor provided on the data provider's website. There, we filtered the data for calls since 2013. Then in Tableau, we filtered out any rows that did not have a valid Call Type Group value;

Visualizations

This visualization shows the number of calls for each Call Type Group in each neighborhood. The four valid Call Type Groups are:

1. Alarm
2. Fire
3. Non Life-Threatening
4. Potentially Life-Threatening

The x-axis indicates the Call Type Group, while the y-axis indicates the neighborhood names. The number of calls per Call Type Group for each neighborhood is encoded by color in a heatmap. The provided legend shows the full range of total numbers of calls. The top five neighborhoods with the highest call activity are in order:

1. Tenderloin
2. South of Market
3. Mission
4. Financial District/South Beach
5. Bayview Hunter's Point

This may be partly due to the high population density of San Francisco, especially in the downtown neighborhoods. Note, this dataset does not include population data so this visualization is not meant to be an accurate reading of if neighborhoods with high amount of calls are more dangerous than the others. Also, there seems to be a trend for areas with a high amount of calls to also have a high amount of potentially life-threatening calls.

To give more context, another visualization is included to show the total number of calls and the Final Priorities of the calls from each neighborhood. The bars of each stack are encoded by color for Final Priority. The Final Priority is determined by the Fire Departments to say whether or not the call was actually an emergency or not. While people may report an emergency when calling in it is up to the Fire Departments to determine if the call is deemed an emergency or not.

Future Interactivity

Future interactivity for these visualizations as they stand today could be brushing for each each neighborhood to order to highlight the interesting points of interest. Visualizations are subject to change.