Surveys seem like a reasonable methodology for products where you can get large user samples, like with appliances and cars, but I have often had reason to ques

Surveys seem like a reasonable methodology for products where you can get large user samples, like with appliances and cars, but I have often had reason to question how well they understand what they are doing. For example, in the days of pre-digital photography, I was a close follower of technical reports in magazines like Modern Photography. When CR would do testing of of photography equipment, they seemed to be relatively clueless.