Yesterday, I presented a poster on this blog series at the Association for Education Finance and Policy, in Denver, and I had a challenge: how does one present a broad-brush historical argument in this format? So I hacked the idea of a poster, which is to present a limited amount of information as an entree to a discussion with people who decide they’re interested in the topic… and created a chart thematically tied to this series, rescaled to show proportionate changes (with the bottom of the chart representing the greatest proportionate loss 2020-2023 from the 2019 baseline), but without labeling the data:
I used the majority of the poster space for this figure, added the poster title, a QR code tied to the first entry in this series, and added some whimsical and clearly false labels to the series on sticky notes: bear spray sales, time spent not on Twitter, and good hair days.
I invited fellow AEFP attendees to make guesses and label the data. I answered questions without giving away the store, and here is how the poster looked at the end of the session:
And of course I promised to reveal what the actual sources were. Here’s the labeled figure:
Sources:
- OpenTable’s State of Industry data
- St. Louis Federal Reserve Bank FRED data server (women’s labor force participation chart)
- NWEA pandemic-era student achievement data dashboard
When I downloaded the OpenTable reservation data and arbitrarily picked the 20th of each month, the Mother’s Day spike of 2021 became obvious, and I decided to leave it as a tantalizing hint. The Illinois MAP data has a large gap for 2020 when no periodic testing happened, and I thought that along with the persistence of the drop, that was going to be obvious to someone who looked at the figure. But no one guessed that. I forget who guessed that the orange line was about labor-force participation, but Chris Marsicano (Davidson College) correctly guessed that the blue line was about restaurants, even if he narrowed it to NYC.
There are a few questions I hope this figure raises, from questions about the basic data (what does the OpenTable data tell us, exactly, or the MAP data?) to the different cadences within a society during and after major disruption, and what else (along with K-12 achievement and women’s labor force participation, to a lesser extent) has the multi-year pandemic effects so evident here?