Random AsidesSkepticism

Bad Chart Thursday: How Hip Hop, Heavy Metal, and My Little Pony Can Kill You

University of Sydney researcher Dianna Theadora Kenny recently wrote an article titled “Music to die for: how genre affects popular musicians’ life expectancy,” about her research into early mortality and causes of death among popular US musicians by genre, complete with head-banging charts.*

The first chart, according to Kenny, “plots genres over time (oldest to youngest genres), showing the average age of death of popular musicians by genre and gender against life expectancy (LE) for US males and females born in the same year.” (The chart subtitle says “genre and sex” not “genre and gender,” but women mysteriously disappear from the analysis anyway, as we’ll see with the second chart.)

Age of death and musical genre

At first glance, it seems like newer genres of music actually increase average life expectancy for everyone except the musicians themselves, for whom the genres have the opposite effect. I’m not sure whether you need to listen to these genres or whether the very existence of, say, rap and hip hop automatically extend your lifespan. I’m also a little uncomfortable with the idea that most of us are basically musical vampires draining the life out of musicians to extend our own.

Skeleton singerFortunately, after rereading the article and the comments, I discovered that the life expectancy lines aren’t flat, as we might expect, because they represent the life expectancy of a person born in the average year of birth for each genre. So we’re not vampires (at least, not according to this chart). The increase in life expectancy we see from left to right on this chart simply reflects the progression of time represented by the genres, although not by actual units of time. I think you have to play the chart backwards to get that information.

Clear labels would have helped make this chart a little more comprehensible, but ultimately, a line chart is simply the wrong instrument for what she’s trying to express. It implies meaning in the left to right progression of the lines when her point is to make individual comparisons within each genre, the life expectancy of the musicians versus that of the general population. Her data are a series of solos that the line chart treats as harmonies in the same song. The results are predictably cacophonous.

Even with a bar chart, however, the comparison with generic life expectancy doesn’t really tell us much. The life expectancy data are not age adjusted, which would have helped minimize things like infant mortality rates that affect the average life expectancy of the general population but aren’t relevant to what we might expect from popular musicians who are not actually infants.

More specific comparisons to relevant subsets of the population would have been more useful in examining whether musical genre is correlated in any substantial way with early mortality, much less whether it affects life expectancy, as the article title claims. Comparing the musicians to those with similar socioeconomic backgrounds, for example, would be more meaningful, as would comparing musicians to those in the same genre who were popular but who stopped performing (at all or at the same rate).

Ultimately, though, the biggest flaw with the comparison among genres is that the popular musicians in newer genres are going to be younger overall, so any deaths in this group are going to be deaths at a relatively young age simply because the bulk of the population is still young. Kenny notes this in her article, but she doesn’t believe this fully explains the differences among genres in age of death. She then goes on to present a section on causes of death, which seems to suggest that these are additional explanations for the early mortality in newer genres. She provides this chart, which apparently (from the text that follows) focuses only on men, although she doesn’t explain why.

Cause of death by genre

In this chart, musicians in the newer genres are much more likely to die from unnatural causes: accidents, suicide, and homicide. But again, if you are looking at deaths among a young population, you’re not going to find as many deaths by natural causes, so the deaths you find are likely to be by unnatural causes. No genre-specific speculation needed.

MLPskullIf, for example, we compared the cause of death among people who played with the original metal-tipped lawn jarts when they were kids and people who played with My Little Pony toys when they were kids, we would no doubt find a higher number of deaths from unnatural causes among the latter because much of that population would be too young to have died from natural causes. Are My Little Pony dolls more dangerous than lawn jarts? Do they affect life expectancy? (OK, so Applejack probably does. I’ll grant that. But Pinkie Pie? Twilight Sparkle?).

I hate to think of the tragedy that lies ahead for Brony rap metal musicians.

Finally, if you look at the table, you can see how the stats might change based on where the researcher decides to place a particular musician. How many of the metal musicians who committed suicide could have been categorized under rock just as validly? And what do you do with the artists who work in multiple genres or the fusions? Where do the country folk artists go? The jazz blues musicians? The electronic hip hop? Rap metal? Pop rock? If genres are themselves so fluid and difficult to pin down, how is it possible to pin down a correlation of any significance between a genre and mortality?

I actually don’t think it’s far fetched that popular musicians might be at higher risk of death by unnatural causes, especially those who are touring a lot and exposed to an endless stream of drugs and alcohol. The bands, other musicians, and roadies on a tour might experience similar risks. Kenny mentions a study that found adverse childhood experiences to be a common factor in mortality among famous pop and rock musicians, which I think points to the ultimate issue here. Of all the many factors in a musician’s background and experiences that could play a role in early mortality, music genre seems to be one of the least likely suspects, and this research seems to support that more than anything.

On the other hand, maybe the surgeon general really should put a cancer warning on accordions.

*Anthemic air guitar solo to @Samuel_Erkison for bringing this article to my attention.

Rainbow Dash image modified from the original by Ben Scholzen.
Skeleton image by Thomas Hawk.

Melanie Mallon

Melanie is a freelance editor and writer living in a small town outside Minneapolis with her husband, two kids, dog, and two cats. When not making fun of bad charts or running the Uncensorship Project, she spends her time wrangling commas, making colon jokes, and putting out random dumpster fires. You can find her on Twitter as @MelMall, on Facebook, and on Instagram.

Related Articles

9 Comments

  1. I think you covered my two (fundamental) problems with this study I put in the original comments a while ago. I’m very interested to see the final article, it would be very difficult to do this study based on decadent cases only as getting actuarial (life) tables for musicians (by genre) sounds difficult.

    1. I saw your comment and her response. In general, the comments pointed out several problems that apparently will be addressed in the actuarial analysis, but yeah, I don’t have high expectations for that, in part because the existing analysis for what is there seems flawed and in part because, as you say, the actuarial analysis definitely sounds difficult. I’m guessing there will be gaps in information that will need to accounted for, and this accounting is likely to be done with educated guesses in the same vein as the guesses in her original article–lots of confirmation bias. But maybe we’ll be pleasantly surprised.

  2. Your second-to-last paragraph is spot on. I work professionally in the recorded music industry and have for over a decade (mostly in metal music). I can’t tell you how much difficulty there is categorizing genres on a professional level, let alone the fan argument level.

    I once worked for a music startup and I think we spent three months trying to work out what genres to use and how to classify them. A decade later and the music services out there still don’t have a standard (and I believe that’s the case because as you point out genres are fluid).

    1. Book publishing has a similar problem with categorizing, although the categories are not as important in reaching the audience as in music, I don’t think.

      A single song can mix genres! Such a bizarre hypothesis to test in the first place.

  3. Notably absent are classical musicians. Almost all of them, especially those born in the 18th and 19th centuries, are dead (except the vampires, of course.)

  4. I agree with almost all you said… but Applejack affecting life-expectancy? What? Pinkie-Pie is conclusively the most dangerous of all the ponies, (perhaps comparable to Rainbow Dash). Yes, arguably her fourth-wall-breaking powers mean that she may be capable of surviving in situations where the others wouldn’t … but seriously, Applejack?

  5. Wait a second, is this chart seriously only considering those musicians of a genre that have already died, while ignoring those in the respective genres which are still alive? That’s the dumbest thing I’ve seen all week.

    That would mean that if in a hypothetical new genre, everyone lives to be a 100, but one guy died at the age of 16, this chart would show 16 as the life expectancy. That has got to be what they do – no way is the average life expectancy for all metal musicians that low. Barely any of them have died young.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to top button