September 4, 2013

Wedding Crunchers

The website Wedding Crunchers displays keyword trends from 1981 through 2013 in New York Times wedding announcements in Google nGram style. For example, back during the first Reagan administration, more NYT wedding notifications mentioned "Republican" than "Democrat" or "Democratic." Recently, however, Democrats have opened up a 3 to 1 lead.

RapGenius has an interesting analysis with lots of graphs.


sunbeam said...

From the article about the page you linked to:

"This seems consistent with demographic trends, namely that the city’s Indian population increased by 118% from 1990 to 2000, while the city’s overall population grew by 9%.

That might not be the whole story, though. The NYT and other media outlets have reported on South Asians’ growing presence on Wall Street, and if you believe that the wedding announcements largely reflect people on Wall Street (more on that later), and there are more Indians on Wall Street, then you would expect to see more Indians in the weddings section."

I'm not sure how to state this exactly. I don't really think about Jews that often. I know they exist, but it's nothing that particularly concerns or interests me in most ways.

Well except for the whole immigration thing, which leads me to what I want to say.

I'm a cynic. I get the whole IQ argument. Fine, that's established.

But as important as IQ is, consider a hypothetical situation where no one is willing to play "fair" in a business transaction. In other words if you are the same ethnicity as I am, you get a better price than if you are not. And if you are some opposed ethnicity you don't get any price at all.

Now you are going to be able to use intelligence to beat me in some things, no doubt. But at the end of the day, no matter what I'm going to screw you over if I can. Unless you happen to be my ethnicity.

Now if there are a lot of people like me seems like you have a problem. And if the IQ differential isn't a total mismatch, you have a real serious problem.

Once upon a time there was hatred and resentment of Wasps having their hands on the reins of money and power. Now we have a more varied group holding those reins. Usually the criticism was they didn't deserve it, they cheated, etc.

But that is the way things are done. What made the Wasps of earlier times special was that they were located and in control of the place where the reins of finance in this case were located.

I'm saying that a good bit of the recent (since about 1970) increase in the wealth of Jews is a side effect of the city where a large fraction of them live. If for some reason Jews had gone to Baltimore or something instead of NY, I kind of think that the financial hub wouldn't have shifted to Baltimore, and Goldman Sachs might not exist.

So what happens when another group of clannish people of close enough intelligence to make it a game want the same thing? And they want to do it the same way?

All I can say is this is the situation they wanted. They shouldn't blame anyone else if it doesn't turn out the way they want.

josh said...

Surely the influx of gay weddings plays a part.

countenance said...

Maybe the NYT doesn't much care anymore about Republicans who get married. I don't know if the NYT covers wedding announcements on its own volition as if it was news, or if someone buys space in the NYT to announce the wedding. Either way, partisan selection bias applies, because the NYT won't much care about Republicans that get married, and Republicans that want to get married won't buy space in the NYT to announce it.

Anti-Democracy Activist said...

Occam's Razor says: Republicans have become (rightly) convinced that the NYT is hopelessly biased against them and, now that a wide range of high-quality alternatives are available via the web, tend not to read (much less place wedding announcements in) the Times anymore.

This is the simplest, least exciting, least socially significant, and most probably true explanation. It's also a warning against reading too much into a single data point.

Anonymous said...

My favorite graph is the one that shows the number of Chinese, Indian and Hispanic names versus the name Cohen. Very illuminating.

agnostic said...

Something not quite right with how they're measuring things here. Doesn't mean it's giving you a systematically wrong or misleading picture, but I don't quite trust their results.

The key is -- they're not counting by articles. You would think the denominator would be the total number of articles in a given year, and the numerator would be the number of them that contained a certain phrase.

But their "About" section says that they're not showing the fraction or percent of ARTICLES, but the fraction or percent of PHRASES OF THE SAME LENGTH.

So, if you search for "from New York," the denominator is the number of 3-word phrases across all articles in the year, not however many articles were in that year.

Like if you search "the," it should be 100% -- every article would have that word. But instead it's around 6% (the percent of all 1-word phrases that are "the"). Also it fluctuates over time, when it should be constant. We would read some fluctuation and perhaps even seek a cause for something that is not of interest (i.e., how many 1-word phrases are this word or that word).

This also means that if you combine phrases of different length, they no longer have a meaningful common denominator. Like "Harvard + University of Pennsylvania" will refer to all 1-word phrases for the Harvard, and all 3-word phrases for U of P. They should refer to the same number, i.e. the total number of articles from that year.

Some of the patterns over time look plausible, but only when we already knew what to expect (e.g., decline of Lehman Brothers). If we want to discover something we don't already know, we have nothing to check the results against to see if it makes sense. That's why we wanted to do the search in the first place. Not see it reflect what we already knew.

I'd take this thing with a grain of salt until it switches over to counting by articles rather than "phrases of the same length as the target."

Bob Loblaw said...

It may just be that Democrats are more willing to inject politics into all of life's endeavors.

Anonymous said...

Reaganism freed the urban liberal elites from old obligations to big labor. Since Reagan did it, the blame falls on the GOP, but since urban liberals are more creative and enterprising, they were bound to reap the most rewards.

So, GOP gets the blame for greed, but rich Democrats, unloosed from big labor, reaped the rewards.
And since Dems and leftist control the elite schools, even rich kids of rich GOP parents turned Liberal.

Looks like idiocracy hit the conservatives first. Just look at the party leaders of American conservatism. A bunch of bozos.

anonomy said...

David Brooks Bobos in Paradise goes into the difference between pre 1960s wedding announcments and contemporary ones. He determines 'whatever happened' (like he doesn't know)it's been good for Jews.

Prior to that you got to the top of the announcements (like my parents) if you had old bloodlines - my mother's side was far from rich.

Today its based on wealth -and how you fit into the global elite heirarchy.

one key component is missing
Before the 1960s the REAL social column was in the wasp run herald-tribune - if it wasn't there, people knew the NYT's one was BS - this kept the NYT from doing what it is trying to do now - socially engineer society to advance jewish interests and marginalize white christians.

Anonymous said...

Carlos slim took an ownership stake in NYT so that he could help it socially engineer society to advance Jewish interests and marginalize white Christians? Either he's been outfoxed or you are wrong.