How well does music predict your politics? | Hacker News Reader

How well does music predict your politics?

notes.variogr.am

42 pointsmarcelcor45 comments

I usually don't like stories along the lines of "scientists correlate brain size with political party" but I liked this one. I'm just not sure there's anything meaningful here.

Most folks in the states know that Democrats are by and large young, city folks. Republicans are older people who live in rural areas (fly-over country). So it's not very surprising that Republicans like country music and Democrats like younger pop music.

Two things stuck out. One, that Pink Floyd listeners trend Replubican! Who knew? Second, that Democrats have a much more diverse taste in music than Republicans (not news, probably, but interesting).

Seriously, Pink Floyd? Maybe it's a case of age outweighing the lyrical content and the overall cultural bent of the music?

I'd have accepted the beatles-ish even split, since everyone likes Pink Floyd and the Beatles even though both are very counter-culture. But leaning Republican is just plain weird for the band that made "Us and Them", "Another brick in the wall", "Have a cigar", etc.

I was very upset when I saw that result but a good scientist never shows his bias :(

I do really think that once we get better at "artist evolvement" (separating Pink Floyd into two or three separate artists as they had very distinct phases) the PF signifier will drop. I only listen to Syd and "It Would Be So Nice" era PF and definitely am not a Romney booster.

Even if you divide it, what do you get, the relativist, existentialist, anti-war and anti-corporate mainstream era contrasted with the full-on acid freak hippie Syd era? Both seem pretty culturally liberal to me.

Everybody becomes The Man over time. The CEOs of today were getting stoned in their dorm rooms in the 70s and 80s.

I don't know - distrust of and feelings of alienation from elites are pretty common conservative themes. I'm both conservative and a big Pink Floyd fan, and never felt that that was weird.

Of course, any band with enough albums is likely to have both conservative and liberal-leaning songs. Even the Beatles - a song like 'Taxman' isn't exactly liberal.

Agree. I think people assume Republican means social conservative values only (thus they relate to country music). But Libertarian types are frequently Republicans. I see Pink Floyd themes being conservative. Conservatives are anti all powerful controlling government (freedom from tyranny).

> Conservatives are anti all powerful controlling government (freedom from tyranny).

No. People who support democratic (note the small D) ideals are against an all powerful controlling government and tyranny. Note that this can include anarchists, in addition to those favoring a more traditional states. Neither conservatives nor liberals have shown themselves adverse to using the machinery of the state to give force to their views.

> Artists whose fans are most correlated to Republican

> Artists whose fans are most correlated to Democrat

I'm French. I have never heard of the first eight artists listed as Republican friendly; I know and am familiar with most artists in the Democrat list (all but two).

Do "Republican artists" target something very American that doesn't appeal to an international audience?

Maybe, would be cool to try it on location data! As I stated in there, we only used TPs that had US set as location, so the data will be very US-focused.

Most of those "republican" artists are country music artists, while most of the "democrat" artists are pop/rap (although it's easy to confuse the two these days). I would expect the rest of the world would not find country music as appealing as pop.

Honestly, modern "country" is "pop". They're distinguished demographically (country artists and fans skew more white and rural than the general US population) and geographically (Nashville vs. LA as the centers of the business), not so much musically. Other than a handful of tells (the accent in which the music is sung, the occasional use of slide guitar or fiddle) there's very little difference.

All those artists are part of a genre we in America call "country music". It is basically the modern, more pop-like evolution of the music of artists like Johnny Cash, Patsy Cline, Hank Williams (senior), and Willie Nelson. And if you haven't heard of those artists, you should check them out. If you're interested, the best modern country artist (in my opinion at least) is someone not appearing on the list: Brad Paisley.

Oh, ok. I know and love Johnny Cash, Tammy Wynette et al. I don't know existing country artists.

I would guess this "music analysis" is in fact a proxy for other predictors such as location (inland / coastal regions) and age, no?

age

The first thing I thought was that I'd love to see this with age accounted for as well

Brad Paisley is a great recommendation. I'd also recommend George Strait (#2 on this list).

Country music / nationalism / "America! Fuck yeah!"

Several others have identified country music. As an American, this a totally boring, expected result. Country music stars headline right leaning benefit concerts, conventions, radio shows, and TV shows.

And there's a natural alignment between country music and conservatism: both celebrate certain attitudes epitomized by the cowboy persona (strong, independent, hard working, etc.) while simultaneously trying to forget that those pesky intellectuals, liberals, foreigners, and minorities even exist. In case you think that last comment is unfair take a quick peek at the white/black ratio of those top 10 artists: 10/0 for Republican, 4/6 for Democrat.

Unsurprisingly, country music finds most of its popularity in the US, Canada, and Australia. Interestingly, however, there has been a country music festival in France for over 20 years, Country Rendez-Vous. http://en.wikipedia.org/wiki/Country_music#Other_internation...

both celebrate certain attitudes epitomized by the cowboy persona (strong, independent, hard working, etc.) while simultaneously trying to forget that those pesky intellectuals, liberals, foreigners, and minorities even exist.

Meh... as much as I don't care for a lot of what passes for "conservatism" these days, one has to be fair and say that being conservative doesn't preclude you from being an intellectual, a foreigner, or a minority. Yes, the stereotype is that intellectuals lean left, and that minorities lean left, but it's hardly that black and white (no pun intended).

I agree with you, but there does seem to be a correlation there. And if you disagree, then ask yourself how many Republicans would be vote for a gay, atheist, or Muslim president. Bill Maher put it best:

>Let me defend myself. I would never say and I have never said, because it’s not true that Republicans, all Republicans are racists. That would be silly and wrong. But nowadays, if you are racist, you’re probably a Republican.

Those words could only be spoken by someone who doesn't get out much.

Is that right? Here are some recent polls from Alabama/Mississipi in which 52% of Republicans believe Obama is secretly Muslim and 29% believe interracial marriage should be illegal.

http://articles.latimes.com/2012/mar/12/news/la-pn-poll-obam...

You're suggesting... what? Democrats have equal hatred for non-whites and non-christians??

He's probably suggesting that there are many types of bigotry and no one political party has a lock on them.

Yes, in America if you're bigoted against blacks, you probably vote Republican. If you're bigoted against whites, you probably vote Democratic. If you're bigoted against Muslims, you probably vote Republican. If you're bigoted against all religious people, well, you probably vote Democratic.

It's probably a function of the area I live in, but every antisemite I've run into in the past few years has seemed very left-wing. (I say 'seemed' because I like to keep a healthy distance from antisemites. I'm only going by shouting and signage.)

While we're on the topic, the political leanings of bigots are fluid. A generation or two ago, if you were bigoted against blacks, you probably voted Democratic.

The other aspect of this is that a Mississippi Democrat is different from a Massachusetts Democrat.

In that particular case I strongly suspect that if you polled white Democrats in those states you'd get similar answers on the interracial marriage question.

Probably not similar on the "Is Obama secretly a Muslim?" question, but as Democrats they're less likely to believe negative things about their party head.

My point is that there are many, many racist Democrats in the country, and if you don't think so you just haven't gotten around very much.

> But nowadays, if you are racist, you’re probably a Republican.

For definitions of "racist" that are "interesting" to say the least.

> Bill Maher put it best.

Maher is a tool who makes Limbaugh look smart.

For example, it was "racist" for McCain to not address the NAACP convention in 2008.

Likewise, it was "racist" for Romney to mention points of disagreement in his NAACP convention speech earlier this week, as proved by their boos. (Any time a white person is bood by a mostly black audience, the speaker is racist. And no, their applause doesn't erase that.)

That said, it would have been "racist" for Romney to only mention points of agreement.

Speaking of boos, if a mostly white audience boos a black speaker, that's because they're "racists".

Coming from Texas, I didn't find the country/republican correlation at all surprising. But here's my question: has it always been this way, or would this be another example of "the Big Sort,"[1] where cultural and political affiliations are becoming increasingly clustered and polarized?

[1] http://www.amazon.com/The-Big-Sort-Clustering-Like-Minded/dp...

Yet another example of a fallacy of representing the political spectrum in one dimension. People are being conditioned to thinking in terms of left-right, good-bad, democrats-republicans; instead of in terms of real issues. While entertaining, this article adds nothing of value to the political discussion.

Rather, its yet another example of people using a single, readily identifiable feature to classify a data set.

It's not an article about politics. It's an article about binary classifiers.

Grooveshark's beluga project (http://beluga.grooveshark.com) reveals a lot of the same patterns. Looking up Mormon tabernacle choir, for instance, shows a high listenership of politically active, religious conservatives, while people that listen to system of a down tend to be moderate liberals with less political involvement.

Music demographics data is fascinating!

This is far more interesting than the linked article, mostly because it actually shows data for the music that I listen to.

For example, Paul van Dyk, Three Drives, and Ferry Corsten all have "IT Specialist / Programmer" as highly over-represented (wow!).

In addition, Tiësto (had to look up Tiesto) shows that advertising does in fact, work! ... 92% of the listeners on Grooveshark uses Windows Live Spaces.

He was hired by Microsoft to sell Vista to the Dutch market: http://www.reuters.com/article/2007/04/15/music-tiesto-dc-id...

Good find! :)

There are lots of things that predict your politics to varying degrees: age, gender, sexual orientation, whether you have children, education, income, job, where you live, to name a few. These can be used together to fairly accurately predict at least whether you lean left or lean right.

In any given election in the US you have about 40% of the general population who votes Democrat always, 40% who vote Republican always and 20% who decide the election results. Even the massive landslides of Reagan in 84 and Nixon in 72 didn't go more than about 60-40.

Politics on a mass scale is so predictable that maps are produced of political leanings that are used for redistricting.

A lot of these same factors can be used to predict music tastes. Age and where you grew up are probably the two biggest. So don't make the mistake that many in the press often do and mistake this for causation for correlation. Correlation isn't surprising given the common factors.

Metal fans could save us all from a two-party system.

Hmm... I'm a huge metal fan myself, and very much an anarcho-capitalist / libertarian / voluntaryist / $whatever_term_you_prefer. But, to be honest, I haven't really noticed much correlation between a preference for metal, and 3rd party political affiliation, at least among my metal loving friends. If anything, I'd say that metalheads perhaps run towards "politically apathetic / don't care" more than other groups. But that's a pretty subjective and totally non-scientific observation.

It is something I have noticed too though. That's why I personally didn't find it very surprising that they are hardest to classify.

I would have liked a distinction between "this music could be either" and "this music didn't signify anything". Would help tease out the difference between political apathy and appeal to both parties

RTFA. Their conclusion was not that metal fans have a preference for 3rd party affiliation.

Not strictly speaking, but if they say:

"I found it neat these non-predictive artists were mainly metal. Perhaps the genre that can finally bring this divided country together or break the lock on the two party system."

then it's a small bit of inference to think that they are suggesting metal fans are more likely to go for something other than the two incumbent parties.

Anyway, I'm generalizing a bit here. It's not always necessary to reply only to the conclusions than the authors of an article explicitly drew.

should consider throwing in age distinctions. [old music-old people-conservative] vs [new-music-young people-liberal] is neither interesting nor exciting.

Another Texan checking in.

It's worth pointing out that the artists listed in the red box do actually have younger fans (Kenny Chesney, Jason Aldean, Blake Shelton).

Surprised to see no mention of Radiohead...

I think age is probably the most important. Second to that, I suspect a higher portion of the data set being on the Democrat side. Both factors are so strong that they have old sayings about both.

[old music-old people-conservative] vs [new-music-young people-liberal] is neither interesting nor exciting.

Indeed. The saying attributed to Winston Churchill (he didn't actually say it)

http://en.wikiquote.org/wiki/List_of_misquotations

http://en.wikiquote.org/wiki/Winston_Churchill#Misattributed

"Show me a young Conservative and I'll show you someone with no heart. Show me an old Liberal and I'll show you someone with no brains." is a familiar statement that people change their political opinions over the course of life. I might add that most observers of politics in the United States think neither "conservative" nor "liberal" are useful terms for describing where people fit in United States politics.

For the record, my radio presets in my car are to "classic rock" stations (of which there are usually at least two at any one time in my town) and most of the music I rip to my computer is similarly music that was top-airplay pop music for the Baby Boom. When the Baby Boomers were young, that was probably considered somewhat radical, antiestablishment music, but I've been surprised at how many of my contemporaries have grown up into middle-aged people who might be characterized as "conservative" politically active voters today, while keeping the same musical tastes.

Some non-normalized results might be nice. If the sample size of Republican-leaning vs. Democrat-leaning are orders of magnitude different, then the diversity of taste comparison is questionable.

It looks a bit like you're matching location to both traits: country/republican to rural/lower density and pop/hiphop/democrat to urban/coastal

Pretty predicible - modern country music is pretty much pandering to right-wing values. It's not really any different than rap - it's just a matter of who your demo is; inner city black youth with baseball hats or middle-American farm boys with cowboy hats. Either way, it's music marketed toward marginalized, uneducated people that need to cling to some sort of social identity to feel empowered.

...why do YOU know so many people that "like all music but rap and country"? We're well-to-do, educated, white boys.

Reply on news.ycombinator.com