It's sneaky because it's taken out of context. Yes, pro-union people were targeted. This may or may not be statistically significant. And what is "pro-union?

It's sneaky because it's taken out of context. Yes, pro-union people were targeted. This may or may not be statistically significant. And what is "pro-union?" If they have 30k employees, and 3% of them are pro-union, and they laid off 1% - what are the odds that some of the folks laid off are pro-union but weren't targeted? What what are the odds that someone that wasn't targeted assumes they were? I have no idea what the facts are, but blind assertions like those in the article are foolish. Imagine reading an article about you that says "Some neighbors felt he was holding more guns than needed and enough ammunition to kill dozens of people." Or they said that your Internet service provider had noticed large bandwidth which is consistent with someone watching perverted pornography. These are baseless assertions - and therefore shallow.