This poll is pretty meaningless, because it doesn't ask how many people have without problems.
For example, if out of 1000 from company A, 10 had problems, but out of 30 from company B, 5 had problems, it would appear that the company A is worse, even though it has a much lower problem rate.
For example, if out of 1000 from company A, 10 had problems, but out of 30 from company B, 5 had problems, it would appear that the company A is worse, even though it has a much lower problem rate.