I’m curious how the method of collecting data is biased away from dogs who are less likely to be talked/bragged about by their owners. Also, that suggests it is

I’m curious how the method of collecting data is biased away from dogs who are less likely to be talked/bragged about by their owners. Also, that suggests it is likely more likely to represent a person’s/family’s _current_ dog and to ignore previous dogs. Dogs which are given back, thrown away, currently homeless, in a rescue organization, or were put down are likely to not be represented in this survey. Those are likely to have less breed-predictable temperaments, so I would assume the results here are a ceiling, not necessarily accurate.