This visualization uses Google Trends to explore the top Covid-19 related search queries since Jan 20th 2020. The data are restricted to search queries that began with the questions "How to ___" or "What is/are ___."
Trending search queries were obtained using the Google Trends API. A separate API request was made for each unique combination across a parameter space of three variables: PROMPT ("how to", "what is", "what are"); STATE (each of the 50 United States) or COUNTRY ("United States", "Australia", "Canada", "Great Britain", "India", "Singapore"); DATE-INTERVAL; (a seven-day interval ending on each day between Jan 20th and May 8th, 2020). Country-level queries were restricted to English-language searches only. For each request, we stored the top related queries that were returned.
The queries were filtered to only include queries related to Covid-19, its worldwide outbreak, and/or its effects (direct or indirect).
The queries were further categorized along a number of dimensions including: Query Type ("How to___" vs "What is__"); Query Category (e.g. "Symptoms & Treatment"); and Query Keyword (e.g. "sanitizer").
This approach yielded over 23,000 Covid-19 related queries at the US state-level, and over 8,300 Covid-19 related queries at the international level.
For the US state-level data, the queries were grouped by date, and states that contained each unique query were tallied. Each bubble depicts a query that was among the top trending searches in one state (smallest bubbles) or more states (larger bubbles) on a given date.
The country-level plots were generated using a 7-day rolling window mean; plot values represent percentages out of total Covid-19 related "How to" and "What is/are" queries on each day in each country.