There are five basic problems here with these rating systems in general: 1. Not everyone leaves ratings so there's a selection bias; 2. Everyone uses a differ

There are five basic problems here with these rating systems in general: 1. Not everyone leaves ratings so there's a selection bias; 2. Everyone uses a different standard. The number of times I've seen a restaurant 1-star review that basically said "the food was great but we have to wait 15 minutes for a table when turning up unannounced with a party of 12 on a Friday night" is astounding; 3. Psychology makes people disinlined to leave bad reviews. The article mentions this with the woman trapped in a room. To me that's a safety issue and should be an automatic 1-star and a complaint to AirBnB and possibly to authorities. She didn't want to hurt the owner's business? The owner is a parasite; 4. Ratings inflation is a real problem; and 5. The platform is incentivized to having high ratings. The last one is a big one and not talked about as much. Think about it: if you, as a user of AirBnB (for example) leave low ratings, that's bad for the platform. Other users feel less confident about using the platform. The side effect of this (intentional or unintentional) is that you want to make it harder for these people to use the platform. If you're Uber, you might not match them to a driver. If you're AirBnB, you might hide certain properties or make it more likely that the owner won't accept their reservation.