4 ms·

MUCH more likely than the original shallow fluff piece: https://en.m.wikipedia.org/wiki/Clustering_illusion https://en.m.wikipedia.org/wiki/Clustering_illusion

onion2k 11y ago
Is that actually more likely? Sometimes statistical "streaks" are real. Given that "celebrity culture" really kicked off in the modern way we have it now in the 1960s, so the number of celebrities from that era is far higher than previous eras, and that those celebrities are now of the age when people typically die, I don't think there's anything illusory happening. It's likely to be a very real, and very easy to explain, phenomenon.
coldtea 11y ago
The link you sent is about an effect that occurs with random sequences. Neither the baby boom, nor the huge growth of the media industry (radio, TV, recorded media, etc) after the fifties are "random". And then there's the fact that the earlier real stars we have now (as opposed to merely people famous in the past but already forgotten) are in their 70s and 80s (e.g. the 60s and 70s musicians and actors etc, all still quite relevant for modern audiences thanks to vinyl, VHS, tape, TV, CD, DVD and cable distribution, and --something previous generations lacked until recently-- instant access to all their works and news from the streaming, YouTube and the web). This is no clustering illusion, it's the inevitable combination of life expectancy + tons of people becoming stars en masse with unprecedented global distribution at more or less the same time that sees them now in an advanced age.
- shalmanese 11y ago
  The magnitude of increase can't be explained by demographic factors. It's far more likely that this is mostly due to a statistical blip with only a minor contribution from everything else.
  - coldtea 11y ago
    The magnitude of the peak for 2016 isn't what's important. The steady yearly increase is what's important.

onion2k 11y ago

Is that actually more likely? Sometimes statistical "streaks" are real. Given that "celebrity culture" really kicked off in the modern way we have it now in the 1960s, so the number of celebrities from that era is far higher than previous eras, and that those celebrities are now of the age when people typically die, I don't think there's anything illusory happening. It's likely to be a very real, and very easy to explain, phenomenon.

coldtea 11y ago

The link you sent is about an effect that occurs with random sequences. Neither the baby boom, nor the huge growth of the media industry (radio, TV, recorded media, etc) after the fifties are "random". And then there's the fact that the earlier real stars we have now (as opposed to merely people famous in the past but already forgotten) are in their 70s and 80s (e.g. the 60s and 70s musicians and actors etc, all still quite relevant for modern audiences thanks to vinyl, VHS, tape, TV, CD, DVD and cable distribution, and --something previous generations lacked until recently-- instant access to all their works and news from the streaming, YouTube and the web). This is no clustering illusion, it's the inevitable combination of life expectancy + tons of people becoming stars en masse with unprecedented global distribution at more or less the same time that sees them now in an advanced age.

shalmanese 11y ago
The magnitude of increase can't be explained by demographic factors. It's far more likely that this is mostly due to a statistical blip with only a minor contribution from everything else.
- coldtea 11y ago
  The magnitude of the peak for 2016 isn't what's important. The steady yearly increase is what's important.