It looks like you can't trust at least this time of the radio ratings that they cut, as they shortchanged WMVP ESPN Radio 1000 of its online listeners, and Robert Feder posted a couple article of that, including Jim Pastor, the vice president and general manger the station that blasted Nielsen for the undercount, here is his report on that https://www.robertfeder.com/2019/05/24/espn-1000-blasts-nielsen-ratings-undercount/ , and here is his original post on the undercount as his lead story in his Robservations column https://www.robertfeder.com/2019/05/23/robservations-nielsen-bungles-ratings-espn-1000/ . This is making me wonder what they are going to do in the future, to make sure that this does not happen again.
The PPM is not an adequate device for measuring most online listening which is generally done by means of earphones. While Nielsen has a device that plugs in between the device and the earphone, next to nobody uses it.
Online listening is best captured by other systems such as counting the stream log-ons. But Nielsen does not combine these at present, although there is a future product that combines different platforms and methodologies. However, radio would have to pay more for this kind of measurement and there is resistance to a price increase in a no-growth industry.
Apparently WMVP switches frequently from total line reporting to separate stream reporting. This is confusing for Nielsen, and they reissued the report once they verified that WMVP was eligible for TLR in April.
What this sounds like is a typical station with a down book blaming the ratings provider rather than their own programming. But when the numbers go up, it's never due to the ratings service but, instead, to their own programming genius.