Animated Mean Center of Population for the United States: 1790 to 2010 | Hacker News Reader

Animated Mean Center of Population for the United States: 1790 to 2010

census.gov

78 pointsmaxerickson39 comments

If you interpolate between the 1970 and 1980 points, mid-1972 would be fairly close to Belleville, Illinois - where, in mid-1972, I was born.

...thus confirming my long-held belief that I am the center of the universe.

<joke>

The fact that you so naturally extrapolate "being the center of the United States" to "being the center of the Universe" also confirms a long-held belief of mine about the average American.

</joke>

Or that you're completely average :)

Another notable result of the 1980 census was that it was the first one where the majority of Americans lived west of the Mississippi River.

Not true[1]. Perhaps the "population center of mass" is west of the Mississippi, but that is a different statistic.

[1] http://wiki.answers.com/Q/What_it_the_population_west_of_the...

As an example of how this could be possible, imagine the population of the USA were 3: two people living in East St. Louis, and one person living in Hawaii.

It would be fun to actually compute the center of mass, knowing that people are not uniformly massive.

A good starting point might be average weight statistics on a per-state level. I doubt you could get any finer-grained information though.

Distance, as well as population, is a factor in the mean center of population. Even today there isn't close to a majority of the population living west of the Mississippi.

Good point, I only wish I'd seen it in time to fix my comment.

Another animation from the University of Kentucky Appalachian Center, "US Population Growth from 1790 to 1990" tells the story better: http://www.outragegis.com/animations/population-growth.htm

I wonder if the accelerating trend towards higher Southern populations in the 1900s was brought on by the invention and development of home air conditioning. I know I would have never moved to Texas without it.

Indeed, it is will known that air conditioning enabled the southern population to rise:

Discussion for the US south (and some discussion of the social consequences of the technology): http://www.independentmail.com/news/2007/jun/24/air-conditio...

Discussion more generally for the world, and also some side effects: http://www.washingtonpost.com/blogs/wonkblog/wp/2012/07/07/h...

... just as a side note, the use of the word "Animated" is being a little liberal.

A beautiful illustration of manifest destiny.

It's amazing how fast the US expanded. In 60 years, within one life time, the US changed from a small new England colony to an empire larger than the size of Europe.

However, the growth of the British Empire was probably faster during this time period.

I'm probably missing something, but the points c. 1850 seem way too south for me. The Confederates had around half of of the population of the Union (http://www.mrnussbaum.com/civil_war/unionconfederacy.htm) and yet the population center is just south from the Mason-Dixon line (but north of the initial front).

Was the population concentrated around the border between North and South? The West didn't matter - California had less than a million inhabitants in 1860.

Edit: All southern states on the border with the exception of Virginia (except West Virginia, of course) fought for the Union, so they are probably included in the statistics for the North. Still, the North was more populous even if the border states had joined the South.

Settlement in the west skewed farther south than north during that period. Minnesota and Wisconsin were very sparsely populated when they were federal territories. To this day, most of the population concentration of Minnesota is almost four degrees of latitude from Minnesota's northern border.

Exactly.

For center of population, as for center of mass, position matters. The farther you are from the center of population, the greater your weight. One Floridian might contribute the same southward pull as ten Virginians.

The bottom line is that America was still an agricultural nation in 1850. The population of the northern states was biased towards the south, where farming was more productive. Meanwhile, the Deep South had a large slave population, because that was where the cotton grew.

Thus, Texas and Louisiana had far more people than Minnesota Territory. Mississippi had more people than Wisconsin. Alabama had more people than Michigan. And Georgia had more people than ... well, Ontario is not part of the United States.

Oh sure, there were a lot of people in Illinois, Ohio, Indiana, and Pennsylvania. But these states were just too far south to exert much of a northward pull.

I hadn't thought of the slave population initially. Not to bring up a sore topic for no reason but I wonder if these calculations use the Three-Fifths compromise.

Looks like a big jump between 1940-1950 due to the gold rush, I imagine a similar jump will probably be seen for 2010-2020

You probably meant to write "1840-1850" if you are referring to the Gold Rush. Another big change that happened in the same decade was a change in the borders of the United States.

This doesn't seem to take into account the non-European population^1...? I guess that's fine if you state it -- just people the census was counting (not even all white people in US territories all the time).

So this annoyingly helps sustain the myth that North America was an unpopulated forest until being 'settled' by 'settlers'.

1 (http://www.census.gov/history/www/genealogy/decennial_census...):

"Prior to 1900, few Indians are included in the decennial federal census. Indians are not identified in the 1790-1840 censuses. In 1860, Indians living in the general population are identified for the first time. Nearly all of the 1890 census schedules were destroyed as a result of the fire at the Department of Commerce in 1921.

Beginning with the 1900 census, Indians are enumerated on reservations as well as in the general population."

- http://www.census.gov/history/www/genealogy/decennial_census...

Hm: is "non-European" the word you wanted?

I'm not saying you're wrong, I'm really not sure. Even quickly skimming http://www.census.gov/geo/reference/pdfs/cenpop2010/COP2010_..., it's not clear to me whether slaves were excluded from this tally.

You're right, wasn't thinking of Africans. But I think you knew what I meant.

Insofar as that myth exists, it's certainly unwise to sustain it. However, it's my impression that Native American populations in "settled" areas have always been a small fraction of European-descended population, at least since smallpox and other diseases swept across the continent (which happened before 1790). That is to say, including the native populations or not including them would have little effect on this particular population statistic. In confirmation of this, one finds no obvious discontinuity between 1840 and 1860 on this map.

According to this paper* there were estimates of ~600k remaining in 1800. 3.9million citizens in 1790 USA -> nontrivial number and effect on the mean centroid, especially considering they are not uniformly spread.

As for no discontinuity between 1840-1860, it does say 'living in the general population'. So maybe that excludes the remainder.

The 'settled' areas being only a small fraction native of course depends on the time you measure it relative to the disease onset... which often comes after the first settler. I think at first the ratio is the other way.

*: AMERICAN INDIAN MORTALITY IN THE LATE NINETEENTH CENTURY: THE IMPACT OF FEDERAL ASSIMILATION POLICIES ON A VULNERABLE POPULATION. http://cid.bcrp.gob.pe/biblio/Papers/NBER/2006/Octubre/w1257...

So this annoyingly helps sustain the myth that North America was an unpopulated forest until being 'settled' by 'settlers'.

Who believes in this myth?

A staggering number of people, and a vast corpus of literature both modern and classic. Erasure is a fairly serious problem when discussing native populations.

There is a small map that shows the "approximate Area Enumerated" in each date. This doesn't undercount only Native Americans and they weren't numerous in that period. The west was sparsely populated by the standards of Europe and even the incorporated states where the many of the new immigrants came from.

The data and presentation taken literally are not misleading I suppose -- US population centroid.

It is a graph of manifest destiny, like another comment said - the spread of the US population and not talking about how it spread or what receded to make way.

You shouldn't tell yourself this stuff about there not being many other people there, and how even if you bothered to add them in it wouldn't change. It would - it would change quite a bit. This is not a median centroid.

I see the 'area enumerated' indicator. I see how it includes the west coast as of 1850.

We could go find some population estimates but there isn't good agreement before certain dates.

I'd like to see this kind of map, but with a more complete data set about people. This is just a political map and you could make it look a lot of ways with political choices...

Exactly. Apache was not always a web server.

It seems like it moves more in decades of greater economic growth. There are probably a lot of factors to account for that, but it's interesting nonetheless.

I've never seen a less surprising info-graphic. Nicely done, though.

Interesting. Here is a similar animated graph of the population of our solar system over the last 10K years or so.

http://www.nationsonline.org/gallery/Earth/Solar_System.jpg

I guess 'population' never included the original natives.

Go west, young man!

Ok, now I see. But - what is it telling me?

One ought to watch this to the sound of the Pet Shop Boys' "Go West" (https://www.youtube.com/watch?v=flWTWG-gTT4).

Reply on news.ycombinator.com