2 ms·

A specific group of people, or a random sample of people?

• frongpik 6y ago
  The wording implies FB is running an A/B experiment where randomly selected 0.01% of their users get shown 10% less political content and a committee of highly paid statisticians will write a 50 pages summary of how this experiment would affect their ads revenue, should the experiment be increased to 0.05% of their users. A separate committee of highly paid execs will decide if they should proceed. If there's a hint of revenue drop, they'll quietly cancel the experiment and we won't hear about it.