Chaitrika, Priscilla, and Eve's Data Visualization Site
Response Times For Each Neighborhood in San Francisco from 2012 to 2019
For this visualization we measured the average amount of time it takes for the 911 Operator to dispatch an available unit and when this unit arrives at the location of the incident. We wanted to find any discrepancies or variation between the neighborhoods of San Francisco. Some neighborhoods have more incidents reported, are farther from stations and hospitals in distance, and have heavier traffic in their area. All of these definitely affect travel and transport but I wanted to learn if there were any clear patterns present for each year.
Data Encoding
This graph measures the difference between the recorded Dispatch DtTm and On Scene DtTm in minutes. Each different colored segment represents the Call Group Type, either Alarm, Fire, Life Threatening, or Non Life Threatening, and each segment's length represents the average amount of time it takes for a unit to arrive the the scene after the 911 operator dispatches them for that specific type of call. For example, in 2019, if you called in the Tenderloin about a fire related incident, the unit would be on the scene on average in around five mintues.
Average Difference Between Dispatch and On Scene Time in Minutes
To help visualize the data, this graph is interactive. This graph is able to look at different years recorded in this data set by utilizing the drop down option located at the bottom left side of the graph, underneath the citation. You can choose a year from 2012 to 2019 to display the averages for that year. This graph also displays annotations of the average time in minutes for each segment of the bar graph. However the annotation feature is a little buggy so to use this you must change the year with the drop down options before the annotations can appear. To see the annotation you can just hoover the mouse arrow on one of the bars and an annotation should appear from your mouse.
In Conclusion
After removing null values and negative data, I was able to notice some patterns in the data. First, most of the data has some variation but most of the differences in times are very similiar. Considering the vastly different neighborhoods, I was expecting denser areas to have longer periods of time because of traffic or shorter periods of time because of frequency of incidencies. Usually the larger discrepancies are from fire incidents and alarm related incidents tend to be handled the fastest in every neighborhood. One thing I've noticed when alternating between years is that the city as a whole has shown a slight increase in waiting time. I infer that this is a side effect of climate change and an increase in population, perhaps.