Technically, that says nothing about whiteness having anything to do with unionizing risk vs. not. That one states low ethnic diversity of the employee base cor

Technically, that says nothing about whiteness having anything to do with unionizing risk vs. not. That one states low ethnic diversity of the employee base correlates with organizing risk. That just means they're targeting locations where there is an obvious thread tying people together, an easy marker of which that a surprising number of people voluntarily give you is self identification of ethnicity. Not that it's white, black, asian, etc... What the real comment there is, is that there is a tendency for places where you have a pronounced homogenous workforce that culturally identify or share some other commonality with one another are considered unionization risks. Solidarity amongst the ranks breeds contempt of leadership, especially at a distance. It's harder to divide, and they're more likely to forge an outgroup of corporate. This is why you fill in "I'd rather not answer" on those forms, or leave it blank. There is no data point you are asked to give that some data analyst doesn't intend to mine for value in one way or another, and if you think "no one would ever do that!", I would advise you to eschew offers of bridge sales at questionable locations. Also, it's damn easy to get even diverse groups to organize once everyone understands how the game is played. The thing is, most people don't talk about that sort of thing or haven't seen it with their own eyes/done it enough to understand just how much gets done in an "under the radar", "I never said that", fashion. If ya don't believe me, start asking for things in writing. Ask HR what they think Market Rate is. Whole different side of people comes out. Workplaces in particular rely on the fact that workers are by and large, just out to make a living, self-interested, and conflict averse. When unions seem to be starting to form, it's easier to paint them as a threatening outgroup, threatening other worker's jobs because "paying more would impact margins/break the bank", or that the union isn't ultimately out to help anyone but the union; or is out to hold the high performers back. This predates on that self-interest motive, and tries to project an in-group association between the employer and the non-union employee. Psychology works, sadly. Workplaces fear labor organization more than anything else. It's why the Taft-Hartley Act would never have gone anywhere without illegalizing secondary striking (whereby unions would shut down an entire industrial sector to show solidarity for one another). Capital allocators absolutely could not tolerate that kind of threat from the labor force. It's said that's balanced by illegalizing the practice of blacklisting by the Capital class; but you'll find that as capital gets more and more centralized in the hands of fewer and fewer investors, it's much easier to practice that type of collusion without paper trail, or make it so obfuscated as to not be worth looking for. See the Silicon Valley no-poaching agreement.