I have a very serious problem with what some psychology researchers are doing lately, which is what you mention. They're just using Amazon Mechanical Turk as a

I have a very serious problem with what some psychology researchers are doing lately, which is what you mention. They're just using Amazon Mechanical Turk as a way to get a sample population, but it isn't a representative population of anything besides people who work for AMT. It's lazy and makes for shitty research