GDJ Fall 2020

When I first heard about the Covid Tracking Project, I was very impressed by the work they do to collect the numbers and bring them to the public. At the beginning of my volunteer experience, I chose to do 3 states – Iowa, Pennsylvania and Nebraska. Even though the data can be found on each state’s official website, sometimes it does take longer to find the correct number and put it onto the sheet. This is definitely a job that requires patience and care.

Among the states, I spent the most time on Iowa. It was the first state I did and It took me a while to look around the web page and locate the right number. Through the process, I also learned about new terms such as Antigen test, PCR. I had to carefully read the instructions. First I had trouble locating the hospitalization numbers, eventually I found it under a different section within the dashboard. I could imagine how much work it might require a person to find all these numbers from different pages.

I haven’t got access to actually do a shift yet. But I think the practice rounds really prepared me a lot. People from the team are also very helpful along the way. The use of Slack was super helpful and effective as well for communication. I really look forward to learning more about the project and contribute to the data entry team.

Going into this assignment, I was anxious. Despite our training, I felt unprepared for the experience of live group-data entry, so I spectated a core data entry shift the day before I volunteered. Even though I did not contribute to the session, the sense of community that permeates the Covid Tracking Project was apparent. Lighthearted introductory questions posed by the shift lead in the #data-entry Slack channel took my mind off the seriousness of the data this project collects, easing the tension I felt. I watched data entry veterans joke with one another, adding to the expansive collection of custom Slack reaction emojis. Hannah Hoffman reached out to me, and another shift spectator to let us know that we could ask her any questions throughout the shift. I did not have any questions at the time, but I found her assistance during my data entry shift incredibly helpful.

My data entry shift was not without its hiccups, but with the guidance of the double-checking crew and Hannah Hoffman I was afforded deeper insight into the efficacy of the tracking project’s checking processes. First, I was tagged in Slack by a double checker who reminded me to update the Local Time column for my first state (MD). Until I had been tagged by a double checker, I did not understand why in the #data-entry channel checkers conferred in state-specific posts, particularly when the channel’s Shift Bot provided a forum for general inquiry. I appreciated the opportunity to discuss my error within a state-specific post, because the feedback I received was easy to follow in Slack’s Thread sidebar. I should also note that my double checker was incredibly supportive, and she gave me positive affirmation when I confirmed my correction.

An interesting mistake that I made might be an opportunity to clarify the project’s data entry instructions. I was given feedback that, for a cell that required calculation, the expectation was that I enter the equation that made up the cell’s value. Instead, I had been using my computer’s calculator and writing in my calculated values into their cells. I can appreciate why the Covid Tracking Project follows these protocols however, it is not clear from the section of the data entry instructions that discusses “calculated” cells that this is the expectation. I would suggest in section 8 of the checker instructions, either edit sub-section d or add another sub-section, that makes clear that calculated values should show the values that make up the calculated value, not just to “type all data by hand”.

I completed my Covid Tracking shift on October 26th, 2020. I completed a total of five states, including New York, South Carolina, Indiana, New Mexico, and Massachusetts, after observing for half an hour or so.

The instructions for the shift were extremely straight forward, and the people in the team were also very helpful, consistently checking in to see if we needed help and making sure we understood the tasks.

I liked how much attention was given to the accuracy of the data from the entire team. Initially, I was a little confused about the date of data published, but that was quickly resolved as I began to chat with some of the checkers.

Slack was a very effective tool for communication, and I liked how responsive the entire team was. Overall, a very fun experience.

My volunteer core data entry shift was more challenging than I expected; there was definitely an adrenaline rush. I signed up for a shift two days after observing and being actually in the spreadsheet, gathering data, while making my way through the nuances and state specific caveats was profoundly different than simply watching the chats and checkers in the sheet. The shift started with intro check-in via ice breaker, general announcements and standard operational slackbot messaging. I was quite impressed with the logistics of the project, it is pretty systematic at this point with the slackbot sending its daily threads, separate state conversation threads, and team leads and mentors on the shift who know the work in and out and able to provide guidance and support with some of the quickest turn around I have ever seen.

There was a full volunteer list on my day, so I ended up working on 3 state/territories: Louisiana, American Samoa, and Vermont. Starting with Louisiana, I noticed an anomaly where the number of COVID patients currently hospitalized and number of covid patients currently on ventilator went down by just enough overnight to raise concerns. When flagged, I was assured by the team lead that current numbers often fluctuate even cumulatively as state’s look back. Separately, I was really challenged with presence and meticulousness. Despite being fully immersed in the spreadsheet, I still managed to miscalculate a formula cell by using the wrong “positive cases” number. The feedback was “the Negative number (cell AQ) should use the positive cases (PCR) num” which illuminated to me the importance of headers and really understanding what each cell/input is telling us about the virus.

I moved on to American Samoa, which was quite uneventful. The last date from the island was in May 2020, where a fourth amended declaration of continued health emergency was announced. This state’s notes advised the checker to ‘Be your best detective. Good luck. tip: this usually comes from press releases, news or official social media. but it is updated infrequently.’ After some google perusing online, I came across the NY Times article which examined AS and highlighted how “The territory moved swiftly to halt nearly all incoming flights, rapidly boosted testing ability and took advantage of social distancing strategies that had already been adopted in response to a measles outbreak at the end of last year.” Not to mention the territory has responded to the COVID-19 pandemic similarly to recent disease outbreaks such as zika, dengue, and the measles. They have an effective system for managing outbreaks.

Lastly, Vermont was straightforward, but I still managed to make some errors with the calculated cells and not calculating with the correct ‘positive cases’ number. Once again, headers here and expert precision meticulousness here is super crucial.

Greeted by a Slack message from the day’s lead checker, I started my data entry shift for the COVID-19 Tracking Project at 3:30 p.m. (ET) sharp on Monday, Oct. 26. I gathered information on four states: South Dakota, Idaho Connecticut and Washington. I kept the data entry instructions open in a tab, and read and reread the notes in the sheet. The job requires an eye for detail and patience. The language used in some categories is similar, and states use different words to categorize information. On one occasion, I almost put in incorrect information because I misread the note for the cell, which thankfully turned red and I was able to correct it before marking the row as done. The notes and the formulas and external scripts in the sheet make the work straightforward. For the most part, my shift went smoothly, going back and forth between the sheet and the project’s Slack channel. After I completed one row, I felt I had gotten into the swing of things. 

I was impressed by the magnitude of the project while working my data entry shift. Keeping track of data from all 50 states plus U.S. territories is a monumental task. But teamwork makes it possible. I was happily surprised to see how well the team communicated with each other via the Slack channels. With an organized system, it is easy to convey information and get the work done as quickly as possible. The shift ended at 5:30 p.m. (ET) sharp, and I think three hours is quite fast considering the amount of jurisdictions and categories. 

One of my biggest takeaways from the experience was how limited data collection can be. The COVID Tracking Project depends on data provided by the states, and although it may be more detailed and complete than federal information, it is still lacking in some areas. I had to leave several columns blank because the data was not available in the states’ websites and dashboards. In some cases, certain information was updated weekly, so I had no new data to put in. The experience made me appreciate even more the importance of the work done by the project, as it fills in a gap of necessary information on the coronavirus crisis. Overall, the shift taught me the value of working as a team, communicating with fellow colleagues, and reading instructions and notes carefully. 

During my shift volunteering for the COVID Tracking Project, I entered data for Oregon, Idaho, Oklahoma, and Texas. I was glad the focus was on quality, not speed, as it took me a while to enter each value super carefully and make sure I was pulling from the right place. Once I got started, however, I noticed that I was able to get “in the zone” and the I was surprised how quickly the time passed.

As I was completing each box, I was very impressed with the quality of the instructions. The CTP team really went all nine yards to make sure new volunteers have a clear path to finding data that is sometimes nearly impossible to find on complicated and sometimes hidden web pages. I can only imagine how much work it must have taken to get this project off the ground and find where each data point (for all 50 states and additional territories!) was located among hundreds of web pages, PDF’s, spreadsheets, and even social media accounts.

I was also glad to be part of a team that worked together really well. When I couldn’t find the value for a particular data point for Oregon, the team was very helpful in answering my questions in Slack.  Overall, I was impressed by the work put in to make sure this colossal project runs so smoothly!