We can read the workers at Bessemer's Amazon plant said no to a union by a wide margin. If we look at the numbers, we are told there were 5.805 workers who received ballots to be mailed back: 3215 were returned so 2590 workers did not vote: that's 44,6% of the total workforce.
There were 1798 NO ballots counted (not including the 505 that were contested) that's 55% of the votes but only about 31% of the total workforce of 5.805 who were eligible to vote. The fact that 505 ballots were contested, supposedly mostly by Amazon, not the union, does not seem to concern many commentators, because even if they were all in favor of the union, it would not change the final result. Maybe so but why were so many ballots 'contested'? 15% of the votes contested, that's substantial. Who is going to give us a detailed explanation? Are we going to know wether they were NO or YES ballots?
Considering all that, I dont think it's accurate to say the workers said no to a union by a wide margin: 31% of the total workforce voted no, less than the 44% that did not vote. Last but not least, for the vote to take place, the workers had to sign pledge cards: it's not clear how many such pledge cards were submitted by the union and certified by NLRB. I read 2000 or 3000. Both numbers are higher than the number of No votes. Why did those pledges not translate two months after they were signed into more votes in favor of a union? What caused such an 'erosion'? The habitual answer is the negative impact of the intense anti-union propaganda produced by the employer (linked with the threat/fear of losing the job with the plant moving somewhere else). What's next?