It's just as likely of a conclusion (as pointed out elsewhere) that people are informed because of unionization - i.e. that union workers are better empowered t

It's just as likely of a conclusion (as pointed out elsewhere) that people are informed because of unionization - i.e. that union workers are better empowered to point out design flaws or other recall-worthy issues (without fear of reprisal over having embarassed the management) than non-union workers. That is: using recalls to measure product quality is highly misleading for reasons that take all of five seconds to identify, and the fact that the paper chose such a metric despite such an obvious flaw reeks of bad faith.