So they studied ONE company. 10,000 people, but effectively a sample size of 1 since everyone would be affected in a similar way by that ONE company's policies

So they studied ONE company. 10,000 people, but effectively a sample size of 1 since everyone would be affected in a similar way by that ONE company's policies and experience (plus it's an Asian big company, and they tend to be the most conservative bum-in-seat kinds). They used the enforced remote work period caused by COVID in a company with little-to-no experience in remote work, then took the results of the company's quickly devised approach to remote working policy and used that to draw conclusions. Basically, this study is worthless. ... Except that someone at the Economist figured (correctly) that this would be a good way to get traffic.