Blog Plan

[axiong23]

Julius Tyson & Austin Xiong Professor Correia 12 April 2023

Blog Plan

1. Do you plan for your final project to be an extension of the mid-semester project?

No. Our new dataset contains information on date, time, location, and results for Amherst College Emergency Medical Service (ACEMS) calls from Fall 2016 to Spring 2023 containing approximately 1500 entries. The data is provided by Austin and the ACEMS team, who have gotten written permission from admin to use the dataset for our analysis. Our goals are threefold. First, we will analyze when and where campus emergencies typically occur. Second, we will observe the effect of COVID restrictions on the number of medical emergencies. Third, we will look into which medical calls tend to require more advanced backup and which medical calls are resolved within the ACEMS scope of practice.

2. Describe what you hope to deliver as a final product:

We will include four displays in our final product. To satisfy the new material requirement, one of the displays will incorporate spatial analysis.

The first display will map the counts and types of calls in specific locations across campus. If feasible, we hope to use a heatmap to indicate the distribution of ACEMS responses. We will embed an interactive Leaflet map with highlighted sections of the campus, allowing users to navigate through campus and see how the call volume and type varies with campus locations.

The second display will be a double bar graph with the x axis being the type of medical emergency (musculoskeletal, cardiovascular, respiratory distress, etc) and the y axis will be a count. For each category, one bar graph will show the number of those types of calls that require further support (ambulance, hospital, health clinic, etc) and the other bar graph will show the number of those types of calls that were resolved with only ACEMS care.

The third display will show a scatterplot of cases over time. The x axis would be time labeled by quarters from Q4 2016 to Q1 2023. It will draw two lines — one indicating the total cases and the other showing the number of alcohol-related cases in those quarters. The graph will indicate the deployment of new college COVID policies with vertical gray dotted lines. We will observe how COVID policy affected campus life, using alcohol-related calls as a metric for student socialization.

The final display will be a bar graph of average number of calls during the day (categorized as before 8 AM, 8 AM - Noon, Noon - 4 PM, 4 PM - 8 PM, 8 PM - midnight) as well as average number of calls during each day of the week.

We will also appropriately explain our displays, methodology, and work process in text.

3. Outline a schedule for your group’s progress:

By 4/20: Data wrangling complete, began work on graphical displays. Status update 1 submitted.

By 4/27: Substantial work on displays/displays completed, began work on text and additional blog components. Status update 2 submitted.

By 5/4: Project essentially completed besides minor tweaks and adjustments.

By 5/9: Blog completed and submitted, reflection submitted.

[katcorr]

So many excellent ideas here! This sounds really interesting and you have a very solid, detailed plan. (If you end up needing to cut back and can't create all that you propose here -- particularly since there are only 2 of you now -- that is fine; just make sure you get the spatial data component done.) I look forward to reading the results!

Proposal: 10/10

[katcorr]

Also: regarding Leaflet resources (aside from the Leaflet section of the lab on spatial data), this Leaflet for R guide could be useful, in particular the Markers page and the Lines and Shapes page (scroll down to Circles section). Remember that I think it would be impossible to get shape files for the Amherst campus, so any code doing that would not be relevant to your project. But adding markers, circles, pop-ups etc. can be.

[axiong23]

Status update #1:

Most of the data wrangling for the visuals are done. The remaining data wrangling requires a little more recategorization of the variables as well as dealing with the variables that have few values in them. Two preliminary graphs for our second visualization have been made to see what the data looks like. We are only slightly behind schedule. The rest of the data wrangling will get done by this weekend and we will begin making visuals starting Monday and will be back on track.

[katcorr]

Ok! Update 1: 5/5

[jtyson25]

Status update #2:

Or wrangling is 100% done and most of the visuals are done. The only real obstacle is implementing an interactive leaflet map of campus. We have made good progress implementing the leaflet in Shiny and allowing the user to select polygons. The visual will be complete upon fixing one bug with how the data are displayed after interacting with the polygons.

We've also begun writing the actual content of the blog and thinking about how to organize visualizations for a cohesive data narrative. We are on track to finish well!

[katcorr]

Great! Update 2: 5/5

Also -- I forgot I had told @axiong23 that I would take a look at the leaflet issue. I will do that today! Please send me a reminder if you don't hear back by this afternoon :)