Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts

Tuesday, May 05, 2015

Thoughts on music - is this really true?

This isn't a rant but the premise is something that made me think.
It doesn't annoy me in the slightest but made me consider how I now perceive music the way I do. And why that might be so.
This post on The Line of Best Fit got me started because it doesn't fit my experience at all. I will soon turn turn fifty and yet now I listen to not only more music but a wider variety of music, both new and old, than I ever did  - as a teenager or any time in between - and I estimate by an order of magnitude, so statistically significant. In part that is because I can but that does not mean that that I would do so if I did not wish to do so. Indeed I am rather proactive about finding new music rather than simply waiting for it to be served up to me by the industry and established media. Again this is something I rarely if ever attempted back when I was younger. Is this largely peer group pressure and the fact that as we get older, and in general terms more self-assured, it influences us less?


Let's look more closely at what the primary research implies. I will start with this graph:


On the face of it the indication is that I am just an exception to the rule. Another is that I  and many others are listening to "cool" (contemporary) music but also a whole lot more other "uncool" music as well?  This chart does not tell us that, and conveniently stops at age 48. I assume that means that I am regarded as a "dead" listener now.
Strange then that I am writing this. The rest of the survey is interesting not least because it seems to indicate that females' taste in music ages much less quickly than that of males. Interesting though it is I suggest that it might be influenced by factors that are not the headlines - the data is based on US Spotify streaming. What if females are more inclined to use Spotify more often than males, and possibly at different ages, as just one example?
There could well be very legitimate reasons for that to be true; this study tells us absolutely nothing about them.

The certainty is that a lot of money depends on divining the real trends in music consumption and the demographics of it. Importantly, and this must be true given long-standing trends, that it will vary dramatically from country to country, which is something else that the above survey, quite transparently, makes no attempt to address.

Thursday, December 10, 2009

Lists, lists, lists...

It may seem as though I've been posting lists for weeks. That is true but not for so many weeks as I have been thinking about them. This is not a lament, at least on my part, and being the not-particularly-organized person that I am lists have a dual function: they are simple in presenting the information that they contain but also serve as a safe anchorage for a wandering mind with an appetite for trivia.
I think that the only list already mentioned here that I have not yet posted concerning 'Music in 2009' is that of singles: it was always going to be a bit of a random, almost fatuous, list and I'm well through working on it.
The latest started as a 'private list', although it largely stemmed from earlier comments and considerations. I started to wonder about all the albums and artists I should have bought or heard live in 2009 but had failed to do. The next thought was that music in 2009 does not stand in isolation, either from 2008 or, as occurred to me this week, what might happen in 2010. Take this approach too far and the list will be of Fibonaccic proportions but comfortingly, perhaps if only statistically, not infinite.

Not that this is a new idea because Fibonacci was a 13th century mathematician from Pisa, Italy. It is an exponential-logarithmic scale in which the numbers increase very rapidly at first but then the rate of increase slows and tends towards a maximum value. It is hard to define but empirically it has obvious natural application: rabbits do not increase in number exponentially, at least not for ever, even in Australia!

Statistics are a great talking point, and make good headlines, but then there is that truism: 'There are lies, damn lies and statistics.'
We are all an unwitting part of these things so we had better get used to it. I might just share some of my list of 'albums I wish I had' in the coming fortnight if only so you can tell me where I went wrong.