Hacker News new | past | comments | ask | show | jobs | submit login

The actual paper https://www.nature.com/articles/s41586-023-05772-8#Sec2

It does not appear they controlled for ethnicity, race or any other measure of genetic stock. Doing so is important in my view. For example, the oft-cited average US male height 5'9 or 5'10", but for white American men it's more like 5'11"




They don't control for ethnicity, but they study for 200 countries, not just the US. But more importantly -- this has taken place from 1990 to 2020. Have there been big demographic shifts between rural/urban in that time?


>But more importantly -- this has taken place from 1990 to 2020. Have there been big demographic shifts between rural/urban in that time?

In Western countries, yes.

  Country        Year  Urban Ethnic Makeup                   Rural Ethnic Makeup           
  -------------- ----- -----------------------------         ------------------------------
  United States  1990  73% White, 14% Black, 8% Hispanic     | 86% White, 11% Black, 3% Hispanic
                 2020  57% White, 12% Black, 18% Hispanic    | 84% White, 8% Black, 9% Hispanic 
  United Kingdom 1991  90% White, 4% South Asian, 2% Black   | 97% White, <1% South Asian, <1% Black 
                 2011  75% White, 12% South Asian, 6% Black  | 97% White, <1% South Asian, <1% Black 
  South Africa   1991  14% White, 77% Black, 9% Coloured     | 15% White, 68% Black, 17% Coloured 
                 2011  14% White, 79% Black, 8% Coloured     | 11% White, 77% Black, 11% Coloured 
  India          1991  69% Hindu, 14% Muslim, 11% Other      | 69% Hindu, 14% Muslim, 13% Other 
                 2011  63% Hindu, 17% Muslim, 11% Other      | 68% Hindu, 15% Muslim, 13% Other 
  China          1990  91% Han, 2% Mongolian, 2% Hui         | 89% Han, 5% Zhuang, 1% Miao    
                 2020  92% Han, 1% Zhuang, 1% Hui            | 88% Han, 6% Zhuang, 1% Miao    
  Peru           1993  53% Mestizo, 31% Indigenous, 16% White| 73% Indigenous, 23% Mestizo, 4% White 
                 2017  60% Mestizo, 16% Indigenous, 24% White| 74% Indigenous, 21% Mestizo, 4% White


Crazy to see how the level of white in anglo countries has fallen like a rock.


Cities*


In the US, the top 100 cities alone make up over 50% of the total population.


Where'd you find that stat?

> The total 2020 enumerated population of all [331] cities over 100,000 is 96,598,047, representing 29.14% of the United States population (excluding territories)

From: https://en.m.wikipedia.org/wiki/List_of_United_States_cities...


The 50% stat is talking about metro areas, not city boundaries. That is, what people on the ground would use to describe in a city vs. the legal borders.

So, for instance, Greater Los Angeles has more than 18 million people by itself, where your source lists it as under 4 million.

https://en.wikipedia.org/wiki/Greater_Los_Angeles


free market hard at work


[flagged]


With a globalized world we're (all humans, not us specifically) all gonna end up a nice tan color.


If there's no more Japanese, Nigerian, Irish, etc. anymore, then that doesn't sound like diversity to me. It sounds like homogeneity.


there is more to diversity than the colour of our skin, but yes I would expect the more we're all crowding in on this mudball with lots of free travel between boarders or bluring of borders the more we're going to all end up looking much like each other.

Space may change that. we may end up with some very strange looking sections of humanity once we get into space.


Thanks for compiling this data. Where did you get it from?


Send your thanks to chatgpt :)

Here are the sources-

  US: US Census Bureau
  UK: Office for National Statistics
  South Africa: Statistics South Africa
  India: Census of India
  China: National Bureau of Statistics of China
  Peru: National Institute of Statistics and Informatics


Wait what


?


With wolfram alpha, or another plugin, I hope? You posted insanely detailed statistics and cited an LLM solo


Yeah i literally just said "give me the specific stats about ethnic makeup of urban/rural and how theyve changed in a few different countries from 1990-2020. what is your source for this data. put the data in a text-based table format"


I've used chatgpt in a research context numerous times and I always ask for sources and double check the sources. When asked for numbers most of the time it makes them up, when citing sources it also makes up titles of papers and DOIs. It's highly likely the information you got is entirely false


Yes, rural to urban migration is a huge factor across vast swathes of the world, and not just in poor countries. Many wealthy countries like Japan and South Korea are seeing the rural countryside rapidly depopulate, as the young decamp to cities in search of education and jobs.


And tinder matches


> Have there been big demographic shifts between rural/urban in that time?

The best one that I can think of is the fact that rural children probably spend more time outside engaging in physical activity. Even in the US, I know of high schools where a substantial number of the students are working full-time jobs outside during the summer, in addition to doing outdoors sports and other activities during the school year.


The average for non-Hispanic white men in the US is roughly 5'10" https://en.wikipedia.org/wiki/Average_human_height_by_countr...


I also dont see any mention of parent height.

Perhaps tall people are more likely to find work in rural areas and short people's skills are more likely to align with urban environments.

I still wonder if there is something else going on. But feel like instead of a survey of 200 countries it would be more interesting to just focus on one country with a more homogenous genetic population and measure lots of different possibel variables


> I still wonder if there is something else going on. But feel like instead of a survey of 200 countries it would be more interesting to just focus on one country with a more homogenous genetic population and measure lots of different possibel variables

I think the point is that looking at 200 countries makes the possibility of confounding factors much lower


Why are global population differences not sufficient? Is there an a priori reason to expect a different distribution of genetic factors in cities versus rural areas that feed city migrations?


Absolutely. Pick any city of your choice. New York, Los Angeles, London, Joburg, Bangalore, Hong Kong, Lima... Are the ethnic demographics of the city not substantially different from those of its rural surroundings?


> Absolutely. Pick any city of your choice. New York, Los Angeles, London, Joburg, Bangalore, Hong Kong, Lima... Are the ethnic demographics of the city not substantially different from those of its rural surroundings?

Yes, but are they all ethnically skewed in the same way? If the concern is genetic factors, it doesn't really matter if cities have different ethnic demographics than rural areas in all two hundred countries as long as the rural/urban racial skew is different in the different countries.

For example, in the US rural areas may tend to be whiter than urban areas, but in countries that aren't predominantly caucasian, the opposite may be true.


Absolutely, without any doubt in my mind. DC [1] and Baltimore when compared to the rural areas are significantly different ethnic demographics. This might be one of the most dramatic, but I have seen it in nearly every city I have ever visited. LA, Seattle, DC, Baltimore, Philadelphia, Miami, SF, Portland, London, Lisbon, Amsterdam, Frankfurt, Munich, Vienna, Geneva and dozens others that I have traveled to have significantly different demographics in the city than their rural areas surrounding them.

The only places I feel have been homogeneous with their rural areas surrounding them, have been some of nordic places like Copenhagen, Oslo, Stockholm and Tallinn.

1 - https://ggwash.org/view/6497/maps-show-racial-divides-in-gre...


Megacities, sure.

But Tulsa, OK vs the surrounding non-metro counties? I'm less certain without diving into the data.

The median US city size is Naperville, IL at 149,104 people.[0]

[0] https://en.wikipedia.org/wiki/List_of_United_States_cities_b...


Still pretty different. Compare Tulsa city to neighboring counties: https://www.census.gov/quickfacts/fact/table/osagecountyokla...

Naperville is a suburb of Chicago, and without looking I can confidently tell you that it's more white than Chicago, but less white than a nearby rural area. I will look though: https://www.census.gov/quickfacts/fact/table/chicagocityilli...


Great insight -- thanks!


Anecdotally it really seems like this should be true.

Cities in mostly ethnically homogeneous states tend to still have things of value that might draw people of diverse ethnic backgrounds. I'm in Nashville Tennessee and there are far more people of Asian or middle Eastern descent here in the city than there are in the surrounding rural areas, due to the universities and the job opportunities which draw people from out of state.


That's the median city in the set of cities with population greater than 100k, not median of all cities.


I thought that due to redlining and white flight, the answer is that yes, there are more white people as a percentage of the total in rural areas compared to cities.


White flight? More like, most immigrants don't move to rural areas to start a farm. White flight is to suburban areas, not to rural areas.


I saw this play out during COVID when some people who entertained the idea of WFH in a rural area had to contend with lack of internet.


Depends where you live, though it could be a big problem.

I've done splicing work for several farming co-ops out west that ran fiber to the homestead. Small town of like 200 has GPON to most houses.


I’m Australia/New Zealand the primary driver would be international migration for urban job/education opportunities


Of course. Cities have been transportation hubs for centuries, and have been the arrival destinations for inter-country travel for centuries. Cities are a natural hub for diversity.

But if by “a priori” you mean something innate in humans or mathematical, I think you’ll be hard pressed to find that. Or what that really means in this context.


> But if by “a priori” you mean something innate in humans or mathematical, I think you’ll be hard pressed to find that. Or what that really even means.

A priori = Beforehand, or before analysis performed. Standard phrasing in research, defines reasonable hypotheses to test.


Of course. But everyone has different knowledge of human history so it’s hard to say what counts as a priori for you versus me.

For example, most americans have a-priori knowledge that rural americans are demographically whiter than urban americans, and could hypothesize global trends.

If that counts as an a priori reason - I’ll take it. In summary: “A priori” is ambiguous here / not a very useful modifier afaict.


Made sense to me when I used it. Have a good one.


Note that there are no biological differences between races, as that would be considered racist. Therefore, it was not necessary to control for that during the study.


There is no meaningful genetic variance in height between cultures, so there’s no need to control for it.

Or are you suggesting that non-genetic cultural factors could be at play here too? Possibly!


>between cultures

I don't see cultures mentioned anywhere in the article or the comment you replied to.

If mean that there are no differences in the genes that determine height across populations, that is incorrect. For example, the average male height in Pygmy populations is 5 feet, or about 7 inches lower than Cameroon's Bantu populations who live alongside Pygmies. We know there are genetic differences between Pygmy and Bantu populations that are specifically associated with height, and that average Pygmy height increases along with the percentage of Bantu ancestry (https://journals.plos.org/plosgenetics/article/info%3Adoi%2F...). Similarly, differences in height across European populations have been demonstrated to be partly due to differences in the hundreds of genes associated with height (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3480734/).


Fair, Pygmy genetics are minorly distinct. No other group is, and unless you're suggesting the urban Pygmy population exploded to account for these changes, it's irrelevant.


> Pygmy genetics are minorly distinct. No other group is

One group in all of humanity happened to have measurable differences in genes affecting height?

How does that explain the many other distinct "Pygmy" groups entirely separate from those in West Africa, such as those in the Philippines, who have genetic differences in growth hormones that cause them to be shorter (https://doi.org/10.1515/JPEM.2002.15.3.269)? "Pygmy" simply refers to one of these many groups, they are genetically and geographically distinct.

Pygmy groups show how absurd the claim that there are no differences in height is, but the same kind of variation has been shown within European groups on a smaller scale (consistent with the greater genetic diversity in Africa than Europe). We know that similar differences in genes that influence height exist, to a less extreme degree, across all populations. For example, males in both the Tutsi and Dinka ethnic groups average nearly 6' tall, despite living in areas where malnutrition is/was widespread and alongside other groups that are much shorter on average.

I'm genuinely curious where you got the idea that there are no genetic differences affecting height between ethnic groups/countries. Researchers have identified genetic markers responsible for the majority of variation in human height among Europeans (https://www.biorxiv.org/content/10.1101/588020v1), and shown that those markers vary significantly across European groups just as they do in Africa (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3480734/).


> I'm genuinely curious where you got the idea that there are no genetic differences affecting height between ethnic groups/countries.

https://www.nature.com/articles/d41586-021-03622-z

https://www.nature.com/articles/136736a0

https://www.imperial.ac.uk/news/193488/race-science-about-bi...

https://www.nature.com/articles/153604a0

https://www.hup.harvard.edu/catalog.php?isbn=9780674660038

https://sitn.hms.harvard.edu/flash/2017/science-genetics-res...

https://medlineplus.gov/genetics/understanding/traits/height...

https://www.jstor.org/stable/3180830

Every single variation in "genetic trends" amongst cultures is negligible and otherwise entirely attributable to environmental factors.

The source of the ideas you're asserting as fact are very dark and extremely thoroughly debunked at this point. Honestly it's kind of shameful to present these 1920s ideas as modern.

Individually yes, genetics contribute to height, but the full range of height genetics exist in every culture.


>Individually yes, genetics contribute to height, but the full range of height genetics exist in every culture.

"The full range of incomes exist in every country, therefore differences in average incomes across countries must be negligible."

You've linked a long list of articles talking about race in America and Brazil, none of which I can see claim that there are no differences in genetic height by ethnic group. One is a book from 1944 that rightly rejects racially discriminatory policies and argues "that experience rather than inborn traits determines standards of social conduct." What part of that supports the idea that the 6' Tutsis or Croats are genetically predisposed to the same height as the Taron who average 4'3?

The MedlinePlus article you linked explicitly says "in some cases, ethnicity plays a role in adult height."

"Race" categories as described by all of your links that I saw ("black"/"white") are crude aggregations of hundreds of ethnic groups. Tutsis and Pygmies are both "African," but at the opposite extremes of human height. Framing the genetics of height in terms of American racial politics makes no sense.

>The source of the ideas you're asserting as fact are very dark and extremely thoroughly debunked at this point. Honestly it's kind of shameful to present these 1920s ideas as modern.

The "source of the ideas" are the peer-reviewed journals of PLoS Genet and Nature Genetics based on landmark genetic research published in 2012 and 2020. As in hundreds of other peer reviewed studies that have identified genetic markers responsible for height and how those markers vary across populations, including articles you yourself linked without apparently reading.

>Every single variation in "genetic trends" amongst cultures is negligible and otherwise entirely attributable to environmental factors.

Is the vastly higher rate of sickle cell anemia among people of West African descent is either "negligible and otherwise entirely attributable to environmental factors"? Is it simply Europeans' "environment" that causes them to go bald and get skin cancer at rates dozens of times higher than people of East Asian descent growing up in the same countries?


There are zero credible peer-reviewed journals that publish the idea that height or any other genetic traits are attributable to a single "racial group", as the biological concept of "race" does not exist.

So no, you have not linked any peer-reviewed study saying what you claim.


Without attempting to control for it you're not going to know, right? As an example of one isolated mechanism that might make a difference, fish is probably one of the only foods nutritionists almost all agree is healthy, is known to prevent stunting of growth in children, and fish consumption is higher in some cultures than others.


We've already more or less entirely ruled out genetics as a factor in cultural diversity, so there's no need to, again, try to control for it here.


I have no idea what you're trying to say and don't think genetics as a factor in cultural diversity has anything to do with the point in question. If eating fish makes you taller, and there's a country called Hypotheticalia with a strong north/south cultural divide where North Hypotheticals eat more fish than South Hypotheticals despite being genetically identical, then controlling for culture may change your results.

If whatever you wanted to say agrees with this, I agree with it; if it disagrees, I disagree with it.


Yeah, we agree then. I just wanted to make it clear that the racial arguments in explanation of this study aren't going to hold water.


Do you mean within a culture? And what do you mean by culture?


Anthropologists define culture as a set of learned behaviors, beliefs, values, customs, and artifacts that are shared by members of a particular group or society.


Controlling for income makes more sense.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: