"so a building with high rent will filter them out"
This is probably true. I mostly rented houses, but the few apartments were nightmares (and cheap!). In my experience, small to medium landlords were the worst. 1- or 2-unit owners are easier to deal with but (sometimes) slower to fix problems as they arise. "Mom and pop" rentals were by far the best experience all around.
I now own a house in a small city. I've noticed each street varies in quality and safety. Streets with densely packed multifamily homes have littered streets and more crime. Single family neighborhoods have less trash (none on mine) but are more likely to fall into disrepair than income producing property. I've also lived in the bad area of my city. I stayed there longer than anywhere. My landlord and neighbors were great. That street was special despite being densely packed multifamily homes.
There are many variables affecting livability. Though, I do wonder about the efficacy of modeling apartment/condo/street satisfaction. Pricing seems less like a filter, and more like a KPI representing maximization of some hidden variables (those affecting housing satisfaction).
Also, I wonder if a dollar value could be placed on "good neighbors".