A recent article in the Washington Post showed the results of an informal survey conducted through Facebook. See the graph below for my reproduction.
This graph is the result of a non-scientific survey, but it illustrates a strong connection between the percentage of people in a state who report wearing a mask when they are away from their homes and the number of people who know someone who is currently exhibiting COVID-19 symptoms. The correlation between these two statistics is -83%. Correlation ranges from -100% to +100%, so -83% is a very strong negative relationship between the number of people who wear masks and the spread of COVID-19. The R^2 value is the square of this number. It is 69%, which indicates that mask wearing explains approximately 70% of the variation in the spread of COVID-19 from state to state. While correlation is not causation, this provides evidence that the more people who wear masks, the lower the spread of COVID-19. It also indicates that everyone needs to wear masks to slow the spread of COVID-19. While not perfect, this is one of the most compelling pieces of evidence I have seen so far that we all need to mask up.