I actually had a bit more time and read back...
I don't think Jim123 was denigrating academic analysis, or indeed saying that he doesn't analyse, formally or informally... more that for him, "liking" feels like a gut reaction to the song.
However, I too would bet that an academic analysis of good songs would be likely to produce a better hit rate of songs that Jim123 would have a good gut reaction to.
The fact is we all analyse, all the time. We judge creativity by a great number of measures. Exactly what appeals to our taste may appear to be fairly individual, but that doesn't mean that it isn't predictable.The simple fact is that if you can work out what appeals to you about the songs that you like, and you can distill that down into set factors... you will be able to analyse those factors and then use that knowledge within your writing process to increase the likeability of your songs.
I guess it also depends on what we consider to be "academic". I would say that anything that uses knowledge, study and learning falls into this category.
I mean, simple grammar can affect likeability. We may well suspect that an element of poor grammar works within certain songs, but even then there are limits.There are rules about language, rhyme, rhythm... all of those can be measured.
We can also know that hooks placed in certain places tend to be more effective. Similarly, we know using the title within the song can have a positive effect on the song being remembered.... and so on.
Analysis doesn't stop there in terms of usefulness.
Does all that correlate to a successful song? No.
In mathematics you can learn how to manipulate numbers, you can learn techniques, formulae etc. However to then use those mathematical skills to come up with a theory of the universe, takes a combination of knowledge, skills, analysis AND creativity.
To use creativity without the others can condemn you to wandering around going nowhere fast.
The same is true about using reaction without a framework or understanding the context of the reaction.
If you use what you learn, you can guide your future writing so that you are more likely to create a song with broad and strong appeal.
My sister is a senior lecturer in Psychology. Her expertise is language and how it is used and understood. Marketers use their knowledge of language all the time, in order to manipulate us into liking and buying. Advertising is founded on this knowledge. As is brand creation.
The point is, if you think that knowledge, understanding, analysis has nothing to do with likeability or predicting likeability, or improving your ability to create more likeable songs... you are kidding yourself.
If you don't think that you do that as a listener, even informally, you are really, really kidding yourself. Otherwise, how on earth do you assess anything? Likeability is just as measurable and almost as predictable as anything else.Of course it is not absolutely dead set, because it still relies upon your creativity in applying your knowledge (just like the mathematics)
Cheers
John