40,000 and counting.
That’s how many qualitative data points we have collected in the study. It’s a lot of data! And it’s in 20 languages. Even with all of our teams, we still need to be able to analyze it quickly, accurately, and rigorously.
Dr. Phil Resnik is one of our research partners. He’s a Professor of Linguistics and Computer Science at the University of Maryland.
He and consortium director, Dr. Allison Squires are long-time colleagues. When the pandemic’s first wave hit, he reached out to her and wanted to know how he could help. That’s when she and one of the consortium's clinical experts, Dr. Maya Clark-Cutaia, were developing the first study.
He had a solution to the problem of text-based data analysis: Topic modeling. As it’s always best to hear an expert first hand, this video of Dr. Resnik explaining the method is really good. If you don’t want to hear about the nitty gritty details, skip to time point 13:24 (804 seconds).
Comments