Even the best scientific surveys usually produce "findings" rather than "facts"; that is, they take a picture of a flock of geese whose formation wasn't the same just before and won't be just after.
The picture may also be taken through a blurry or distorted lens so as to make it appear that the number of geese is larger or smaller than it actually is.
I'm thinking of how Gallup indirectly sustained a fiction about America's average church attendance for many decades. From the 1960s onwward the figure hovered in the 40-45 percent range, despite everyday impressions that the figure was too high.
Then an enterprising team at Purdue University tackled the question by measuring actual attendance within a region and extrapolating from that. Their results were around 25 percent which seemed to make greater sense. Their research has since been corroborated.
So with both appreciation for what competent surveying can helfpfully deliver and a degree of wariness about becoming too fixed on particular outcomes, I call attention two two findings that related to the working class.