I think this is one of the cases where observing past and using it to predict future adds a bias that's no longer valid these days. Most people I know that are

I think this is one of the cases where observing past and using it to predict future adds a bias that's no longer valid these days. Most people I know that are alone seem to be happy, the ones in relationships seem a mixed bag (some oscillating between high highs and low lows). Again, there is surely bias among people I meet (I can't make a stratified dataset with perfectly balanced classes), though it might be an indicator that socioeconomic and health conditions changed dramatically recently, invalidating long-term trends.