And, beyond that, the lack of realization that research is intended to find the songs that "everyone" likes, not songs that some people like and others despise.
And even beyond that, that research is done very specifically with moderate to heavy listeners to a specific station (or format if there are multiple stations in the market in the format) and in the core age target of the station. The cost for a full library music test on a gold based station can be from the low $20 thousands up to $35,000 or so for a station with a library over 500 songs.
And going a step further, it's all about "how much would you like to hear that song today?" and not "was that song one of your favorites when you were a teenager?"