Are Swedes and Norwegians lighter than Estonians?

[princeton90]

Eye, hair, and skin tones.

[Tooting Carmen]

Yes but the difference isn't big.

[princeton90]

bump

[Komintasavalta]

I used a shell command like this to make a list of the IDs of all players in the Estonian men's top-level soccer league (https://jalgpall.ee/voistlused/52/premium-liiga):

d https://jalgpall.ee/voistlused/52/premium-liiga|@ href|g /team/|u|aa https://jalgpall.ee|d|pup '.team-detail .table-01 a attr{href}'|g player|u>abf/eeplay

I used a shell command like this to make text files for the photo, name, and nationality of each player:

zr /3 abf/eeplayer|while ir p;do s=`d https://jalgpall.ee/voistlused/player/$p`;p "$s"|g -A1 Kodakondsus|ta|z \>2|z \<1|aa $p\ >>abf/eenation;p "$s"|pup '.player-gallery img attr{src}'|aa $p\ >>abf/eeimage;p "$s"|pup '.player text{}'|aa $p\ >>abf/eename;done

Excluded players

There were a total of 257 players. I excluded 48 players because their nationality was not listed as Estonian. I further excluced 68 players because I estimated them to have a foreign background, mostly because they had a Slavic-sounding name, but I also excluded two players who had a Finnish name, one player who had a Danish or Norwegian name, and one player who was clearly non-white based on their photo.

I didn't bother spending much time manually excluding players with a foreign background. If I wasn't sure if a player's name was Estonian, I did a quick Google search about the player in a few cases. If I was more familiar with Estonian names, I could've probably excluded a few more foreign players.

Last year I calculated the frequency of light eyes among soccer players who I estimated to be native Swedes on two different occasions. On the second time, my result for the frequency of light eyes increased about 2 percentage points, partially because I spent more time excluding foreign players and I did a quick Google search about the ethnic background of every player who I had not excluded previously.

Results

Here are the IDs of the players I assigned to each eye color class:

- Blue (119/141≈84.4%): 1114 5728 5768 5784 6053 7315 7584 7667 7806 8369 9182 9543 9782 10340 10627 10695 10747 11311 12292 13175 13279 13755 13757 14687 14688 14992 15018 15035 15202 15207 15592 15626 15633 16199 16231 16307 16316 16530 16799 17442 17927 18211 18463 18915 18979 18995 19147 19153 19353 19378 20292 20781 20908 21197 21662 22612 22729 22732 22822 22858 23040 23047 23252 23889 24257 24358 24402 24507 24734 24762 24767 24872 24931 25065 25371 25586 25592 25642 25643 25658 25746 25749 26042 26123 26174 26178 26324 26528 26812 27090 27592 28031 28148 28296 28335 30572 32515 35215.
- Green (3/141≈2.1%): 6622 10192 28940.
- Hazel (4/141≈2.8%): 6737 16314 22971 23789.
- Brown (15/141≈10.6%): 8090 9190 10413 15554 15560 16176 19260 20928 21055 21782 22636 24400 24827 24916 24920.

I excluded these 116 players because I estimated them to have a foreign background: 1577 2190 3623 4044 4375 5644 5674 5675 5709 5746 5776 7381 7465 7514 8280 8490 8552 8673 8725 9020 9026 9727 9952 9953 10349 10777 11389 13392 13663 13804 13974 15393 15426 15620 16071 16994 17256 17454 18184 18331 18441 19256 20020 20224 20336 20483 21512 22251 22673 22766 22781 22875 22878 22883 23019 23022 23445 24256 24362 24458 24474 25518 25603 26067 26089 26215 26453 26458 26524 26590 26800 26909 27557 27872 27887 28202 28606 34942 35661 35757 38289 38301 38384 40785 44106 44146 44167 46626 46938 47008 47009 49595 49604 49614 49623 49632 49729 49916 49926 50014 50029 50082 50127 50128 50130 50131 50146 50179 50181 50182 50225 50229 50286 52072 52244 52332.

Classification of eye colors

I classified eye colors using a simple scale with four classes:

- "Blue" also includes gray and blue-green eyes, and it includes blue, gray, and blue-green eyes even if they have clear yellow or brown central heterochromia.
- "Green" includes some of the lightest shades of hazel eyes, or light green eyes with yellow or brown central heterochromia. Eyes which I consider "light hazel" are however rare, and I didn't classify any of the Estonian soccer players as having light hazel eyes.
- "Hazel" also includes dark green eyes.
- "Brown" also includes yellow and amber eyes. They are however rare, and I didn't classify any of the Estonian soccer players as having yellow or amber eyes.

The classes "blue" and "green" are always considered light-colored, and the classes "hazel" and "brown" are always considered dark-colored. Therefore light hazel eyes are classified as green in order to count them as light-colored, and dark green eyes are classified as hazel in order to count them as dark-colored.

Even if the eye color of a player was not clear from the photos which I found, I still attempted to guess their eye color. Otherwise I might have ended up excluding a disproportionate number of players with certain ambiguous eye colors.

There was no player in any league who I estimated to have clear enough heterochromia that I would have classified their left and right eyes as having different colors. There was no player who had pink eyes because of albinism.

Results compared to earlier studies

Here are my results, along with the results of studies I did last year that followed the mostly the same methodology (https://www.theapricity.com/forum/sh...=1#post6424961):

- Players estimated to be ethnic Estonians in Estonian men's top-level soccer league (n=141): 86.5% light-colored (84.4% blue, 2.1% green), 2.8% hazel, 10.6% brown.
- Players estimated to be ethnic Finns or Finland-Swedes in Finnish men's top-level hockey league (n=370): 86.2% light-colored (80.5% blue, 5.7% green), 3.2% hazel, 10.5% brown.
- Players estimated to be ethnic Swedes in Swedish men's top-level soccer league (n=147): 83.7% light-colored (76.9% blue, 6.8% green), 7.5% hazel, 8.8% brown.
- Players estimated to be ethnic Swedes in Swedish men's second-level soccer league (n=189): 85.7% light-colored (80.4% blue, 5.3% green), 3.7% hazel, 10.6% brown.
- Players estimated to be ethnic Swedes in Swedish women's top-level soccer league (n=148): 80.4% light-colored (73.6% blue, 6.8% green), 5.4% hazel, 14.2% brown.
- Players estimated to be ethnic Swedes in Swedish women's second-level soccer league (n=194): 83.1% light-colored (78.3% blue, 4.8% green), 5.3% hazel, 14.3% brown.
- Players estimated to be ethnic Norwegians in Norwegian men's top-level handball league (n=172): 86.0% light-colored (blue and green), 24.0% dark-colored (hazel and brown).
- Players estimated to be ethnic Norwegians in Norwegian women's top-level handball league (n=172): 84.0% light-colored (blue and green), 26.0% dark-colored (hazel and brown).
- Players estimated to be ethnic Icelandics in Icelandic women's top-level soccer league (n=206): 85.9% light-colored (blue and green), 14.1% dark-colored (hazel and brown).

Males have on average lighter eye color than females. Among males, the frequency of light-colored eyes was the highest in Estonians (86.5%), followed by Finns (86.2%), Norwegians (86.0%), and Swedes (84.8% (123+162)/(147+189)). Among females, Icelandics ranked first (85.9%), followed by Norwegians (84.0%) and then Swedes (83.1%).

Differences in methodology for excluding players

When I excluded players who I estimated to have a foreign background, I used slightly different methodology in the case of different leagues:

- I'm the most familiar with Finnish names, which might have given Finland an advantage in terms of how many foreign players I was able to exclude.
- The websites of the Finnish and Estonian leagues listed the nationality of each player, which might have given Finns and Estonians an advantage in light-eyedness, because I was able to automatically exclude players with a foreign nationality.
- I spent most time manually excluding foreign players in the case of the Swedish soccer leagues, which might have given an advantage to Sweden.
- In the case of Icelandic women's socer league, it was easy to exclude players with a foreign name, because most native Icelandic female surnames end in "-dóttir".
- I didn't attempt to exclude Finland-Swedes in the case of the Finnish hockey league, which might have given a slight disadvantage to Finland in terms of light-eyedness, because Karin Mark found Finland-Swedes to have darker eyes than Finnish-speaking Finns.
- In the case of the Swedish soccer leagues, I probably failed to exclude some players with Finnish ancestry, which possibly gave Sweden an advantage in light-eyedness, especially since many Finns who moved to Sweden were Eastern Finns, and Karin Mark found Eastern Finns to be more light-eyed than Western Finns.
- I might have failed to exclude a fairly large number of Estonians with Russian, Ukrainian, or Belarusian ancestry, which might have given Estonia a major disadvantage in light-eyedness.

Overall I estimate that these differences in methodology gave the largest disadvantage to Estonia or Norway.

[Komintasavalta]

I now noticed that I had failed to exclude one player who had a Russian name (or some other Slavic name) from the set of players in the Estonian men's soccer league. He had dark eyes, so after I excluded him, the frequency of light eyes among players who I estimated to be native Estonians increased from about 86.5% (122/141) to about 87.1% (122/140).

One reason why I don't place much value on amateur "studies" like this is that it's difficult to reliably exclude people with a foreign background.

I guess I'll have to repost this quote from Heapost (2008) (Pigmentation variation in Finno-Ugric peoples):

Among the peoples studied, Baltic Finns have the lightest hair. Generally, brown hair is predominant among them, but light hair is found more frequently than black. The average point is usually below 3.0. The distribution territory of particularly light hair (M<2.8) covers the greater part of Estonia, eastern and central part of Finland and extends to northern Finland. There are even groups with the frequency of blond hair over 40% and dark hair only up to 12% (M=2.4..2.5). Those are found in North Estonia (Rakvere and Kohtla-Järve) and in Savo province in Finland (Kiuruvesi). So strongly depigmented groups have not been found among the other peoples studied, not even among Finnish Swedes, who are usually similar to most Baltic Finns by their fair hair.

Finnish Swedes (excluding Åland Island, where the average point is 2.28) are most similar by hair colour to the relatively darker-haired groups of Baltic Finns (M=3.0..3.1). These include Finns in West Finland (Kokemäki, Kurikka), and, in the east, Vepsians, Karelians from the Lydic dialect territory (Girva) and Izhorians of Soikino. It could be added that, according to Juhan Aul, the Swedes who used to live in Estonia before World War II were also somewhat darker-coloured than Estonians [1, 2]. The relatively darker-haired groups mentioned differ significantly from the groups with the fairest hair (M<2.8).

In the table below from Heapost (2008), the average eye color (M = mean) is the lightest in Finns, followed by Izhorians, Estonians, and Finland Swedes. The average hair color is the lightest in Estonians, followed by Finns, Karelians, and Finland-Swedes.



I copied the data below from Heapost (2008). The division between Estonian and Finnish subpopulations was added by me, and it's based on dialect areas and not geography.

Sample size	Percentage of light blonde, blonde, and dark blonde hair	Percentage of blue and gray eyes (9-12 on Bunak scale)	Average hair color	Average eye color	Location	Population
100	36.5	54.0	2.80	0.49	Haapsalu	Northern Estonians
98	38.4	71.9	2.62	0.28	Lihula	Northern Estonians
98	31.9	66.8	2.67	0.37	Audru	Northern Estonians
100	36.3	68.0	2.75	0.34	Rapla	Northern Estonians
101	31.9	74.2	2.71	0.28	Põltsamaa	Northern Estonians
100	43.5	73.0	2.46	0.28	Rakvere	Northern Estonians
104	49.5	69.2	2.42	0.34	Kohtla-Järve	Northern Estonians
75	22.4	66.7	2.95	0.35	Iisaku	Northern Estonians
104	22.9	60.6	2.86	0.43	Kilingi-Nõmme	Northern Estonians
100	33.7	55.0	2.68	0.52	Karksi	Northern Estonians
100	39.6	72.0	2.58	0.34	Otepää	Southern Estonians
102	25.0	68.6	2.81	0.34	Põlva	Southern Estonians
100	29.4	58.0	2.92	0.44	Võru	Southern Estonians
105	26.2	67.6	2.86	0.34	Askola	Western Finns
92	25.0	62.0	2.87	0.42	Mynämäki	Western Finns
85	20.8	62.4	3.01	0.41	Kokemäki	Western Finns
112	25.0	63.4	3.06	0.38	Kurikka	Western Finns
96	24.2	70.8	2.97	0.30	Hauho	Western Finns
140	38.9	80.0	2.64	0.24	Keuruu	Eastern Finns
146	27.5	63.7	2.76	0.38	Ristiina	Eastern Finns
101	41.5	70.3	2.54	0.30	Kiuruvesi	Eastern Finns
112	32.0	71.4	2.84	0.29	Kesälahti	Western Finns
116	34.5	74.2	2.65	0.27	Ylitornio	Northern Finns
143	30.0	76.2	2.79	0.28	Kuusamo	Eastern Finns
145	23.3	67.6	2.95	0.37	Salla	Northern Finns
82	23.3	70.7	2.81	0.30	Savukoski	Northern Finns
124	27.0	72.6	2.84	0.29	Åland	Finland-Swedes
146	17.2	61.7	3.08	0.42	Närpes	Finland-Swedes
140	21.1	61.4	3.02	0.44	Lilyendal	Finland-Swedes

Average values weighed by sample size:

Percentage of light blonde, blonde, and dark blonde hair	Percentage of blue and gray eyes (9-12 on Bunak scale)	Average hair color	Average eye color	Population
35.02	65.91	2.69	0.37	Northern Estonians
31.29	66.22	2.77	0.37	Southern Estonians
25.79	66.45	2.93	0.35	Western Finns
33.85	72.64	2.69	0.30	Eastern Finns
27.09	70.57	2.82	0.32	Northern Finns
21.50	64.89	2.99	0.39	Finland-Swedes

[Not a Cop]

I now noticed that I had failed to exclude one player who had a Russian name (or some other Slavic name) from the set of players in the Estonian men's soccer league. He had dark eyes, so after I excluded him, the frequency of light eyes among players who I estimated to be native Estonians increased from about 86.5% (122/141) to about 87.1% (122/140).

You can add this to your table: https://www.theapricity.com/forum/sh...thern-Russians

[Komintasavalta]

I now finished classifying the hair color of Swedes and Estonians, with these results:

- Players estimated to be native Swedes in Swedish men's top-level soccer league (season 2019-2020): 38.1% (56/147) light, 61.9% (91/147) medium and dark.
- Players estimated to be native Estonians in Estonian men's top-level soccer league (season 2020-2021): 41.3% (59/143) light, 58.7% (84/143) medium and dark.

I used a simple scale that consisted of two classes:

1. Light. Blonde and light brown hair. Includes most cases of red hair.
2. Medium and dark. Medium brown hair and darker. Includes some cases of slightly reddish hair with a medium brown base color, even in cases where the beard is clearly reddish.

I kept changing my mind about where I should draw the boundary between the two classes, but I ended up choosing strict criteria for what I counted as light hair. Many cases that I counted as "light brown" would probably be considered dark blonde by others. In many cases, hair which I classified as medium brown still looks more light than dark overall, but it doesn't necessarily look light enough to look more light brown rather than medium brown.

As before, I used slightly different methodology for excluding players who I estimated to have a foreign background:

- Unlike the website of the Swedish league, the website of the Estonian league listed the nationality of each player, so I was automatically able to exclude players who were listed as having a foreign nationality.
- In the case of the Swedish league, I had earlier excluded many players who I estimated to have foreign ancestry by doing a quick Google search about every player who was part of my current cohort, but it was so time-consuming that in the case of the Estonian league, I spent less time manually excluding players. The Estonian league is a smaller-scale league, so there was less information about the Estonian players online.
- Many Estonians with an Estonian-sounding name have recent Russian ancestry, and many Swedes with a Swedish-sounding name have recent Finnish ancestry, but I did not exclude either type of players (apart from maybe one or a few Swedes, but it was about a year ago so I don't remember anymore). It potentially resulted in a major disadvantage to Estonia, but it might have resulted in a slight advantage to Sweden. In Karin Mark's dataset, the Finnish-speaking population of Finland was listed as being more light-haired than the Swedish-speaking population of Finland, and Eastern Finns were listed as being more light-haired than Western Finns, but many Swedes with Finnish ancestry have Eastern Finnish ancestry.

If I would've been able to accurately exclude all players with recent foreign ancestry, my guess is that it would've increased the gap in light-hairedness in favor of Estonia.

Below are players who I marked as potential corner cases. I classified each of them based on multiple photos at least at twice.

[Gota_type_]

When I visited Tallin I saw the rarest looking people in all of Europe. Yes, many of them were blonde (although a rare type of blonde, like bronze or something like that), but the biggest impact was that some (a relatively high number of them) had rare facial features, like a different type of people (not easily seen in western Europe), like more "primitive" (just those that I am talking about and not the the majority of the population) when compared with Swedes for example.

I don´t know why is this but it was my experience. Not even in Helsinki I saw these rare specimenes. Not that they were extremely different, but with rare aspect in their faces, not really western, not really baltic, not sure how to define it.

[Gota_type_]

Whe