That also assumes there is no response bias. The linked article doesn't seem to go too deep on methods but the source is a content marketing piece for Hospital IQ, so I'd take it with a grain of salt.
lol.. what kind of calculation is this?
n is 1 in this case, the survey was not repeated 200 times from which you derived 90% as the mean number of nurses quitting.