Thompson quotes the Uber study showing that ride quality/service does affect tip rates (after other stronger factors) – but then claims the study shows the oppo

Thompson quotes the Uber study showing that ride quality/service does affect tip rates (after other stronger factors) – but then claims the study shows the opposite! For tipping to be a useful feedback mechanism for the quality/service dimension, quality/service doesn't have to be the strongest, or among the strongest, factors determining the tip size. Just a factor. And given that all of the other factors mentioned (like demographics of driver/rider, rider habits, location/duration of ride, etc) are fully determined once a ride is matched, it appears that quality/service is the single influenceable contributor to tip size. So, contra Thompson, it is absolutely the case that, once all the non-controllable factors are fixed, the remaining differential in tip size is a reward for excellent service.