Razib Khan One-stop-shopping for all of my content

January 31, 2011

What do the people think?

With all the geopolitical tumult and news I was a bit curious to see what The World Values Survey could tell us about public opinion in Egypt and Tunisia. Unfortunately, Tunisia hasn’t been in any of their surveys, though Egypt has. So I thought it might be interesting to compare the USA, Sweden, Turkey, Egypt, and Iraq, for wave 5, which occurred in the mid-2000s. The main thing I took away from the exercise is to reflect that Americans are a more equivocal people than I had expected. Many of the questions have a 1 to 10 scale, and I’m providing the most extreme answers. So the low fractions for Americans for some questions point to a relative moderation on some topics…which is kind of weird when you are asking whether “People choosing their leaders is an essential characteristic of democracy.” Since that’s the definition of democracy broadly construed anything below a 10 out of 10 seems strange to me.

(Control + should increase font-size if it is too small)



USA Sweden Turkey Egypt Iraq Religion “very important” 47 9 75 95 96 Politics “very important” 11 16 13 9 37 Family life “very important” 95 92 99 98 96 Most people can be trusted 39 68 5 19 41 Satisfied with life (10 out of 10) 7 12 21 11 3 Great deal of control of life (10 out of 10) 17 16 24 14 9 Men have more ...

September 21, 2010

Swedes are not sexist or nativist

A party, the Sweden Democrats, is about to enter the Swedish parliamanent which is described in this way in Wikipedia:

The party has its origins in the nationalist movement Bevara Sverige Svenskt (”Keep Sweden Swedish”)…During the mid 1990s, the party leader Mikael Jansson strove to make the party more respectable, modelling it after other “euronationalist” parties, most prominently the French National Front. This policy continues to be followed by the present leader Jimmie Åkesson. This effort included ousting openly extremist members.

Yes. More respectable by modeling itself on the National Front. Here’s a bit about the organization which eventually grew into the Sweden Democrats:

Bevara Sverige Svenskt (”Keep Sweden Swedish”) was a Swedish nationalist movement based in Stockholm and is a slogan used by various Swedish nationalist parties. The stated objective of the BSS movement, and the aim of the slogan, was to initiate a debate in order to reduce immigration from non-European countries and repatriate non-ethnic Swedes.

The Swedes, and the world, are shocked. Should they be? From what I can tell the Social Democratic Party of Sweden no longer has a hegemonic grip on Sweden’s politics. But the core working class base of such coalitions is shrinking because of economic restructuring throughout the developed world, with the remnants often defecting to Right-populism. Today Gunnar Mydral would have to look to writing a book about his own nation, which has about the same foreign born proportion as the USA (though that is a touch deceptive as many of these are other Scandinavians or Finns).

This prompted me to look in the World Values Survey. Specifically, the last wave which started around 2005. One thing you notice in the survey is that Swedes are very politically correct, even compared to their Nordic neighbors. I have read that the ecological awareness imputed to Native Americans in part because of the Noble Savage idea has actually resulted in a real shift and striving by many Native Americans to actually implement those ideals. Sometimes I wonder if the Swedes are so “progressive” and “forward thinking” in surveys because everyone always pats them on the back for being progressive and forward thinking. Sweden sure is the least sexist and nativist nation in the WVS.


There are two questions which ask about job preference in times of scarcity. First, “Employers should give priority to (nation) people than immigrants,” and second, “Men should have more right to a job than women.” There are three responses: agree, disagree, neither. Let’s code agree = 1, disagree = -1, and neither as 0. Weight by proportion and get an index of “nativism” and “sexism” within the population. If you get a score of -1, that would mean everyone was nativist or sexist. If you get 0, that would indicate perfect balance. 0.5, a touch on the nativist or sexist side. The plot below has sexism on the x-axis, and nativism on the y-axis.

natsex

Though I think racism is more taboo than sexism internationally (if Saudis explicitly treated blacks in their nation as they do women there would be a natural boycott. One of the reasons the Saudis banned slavery in 1960 had to do with protests which they kept encountering in the civilized world). But sexism is more taboo than nativism (I think there are important reasons for the rank order, but that’s not a matter for this post). The correlation between nativism and sexism is ~0.76, so variation in sexism explains 58% of the variation in nativism. As you can see Sweden is a definite outlier.

Note: don’t attach too much normative baggage to my use of the terms “sexist” and “nativist.” They seemed compact and communicated the underlying sentiments.

Here are the raw values:

Men should get preference in jobs over women
Country Agree Disagree Neither
Sweden 2.10% 94.10% 3.80%
Andorra 4.40% 89.90% 5.70%
Ethiopia 6.00% 85.60% 8.40%
Norway 6.50% 88.60% 5.00%
United States 6.80% 66.40% 26.80%
New Zealand 8.00% 72.60% 19.40%
Finland 9.60% 81.50% 8.80%
Netherlands 12.50% 81.40% 6.20%
Serbia 12.50% 63.10% 24.30%
Slovenia 13.60% 73.50% 13.00%
Australia 13.90% 64.70% 21.40%
Canada 14.30% 77.90% 7.80%
Great Britain 16.20% 76.10% 7.70%
Spain 17.40% 76.00% 6.60%
Peru 17.70% 72.80% 9.50%
Germany 17.80% 66.80% 15.40%
France 18.10% 73.80% 8.10%
Guatemala 19.10% 72.30% 8.60%
Hong Kong 21.60% 44.20% 34.30%
Uruguay 21.90% 69.30% 8.90%
Italy 22.00% 59.20% 18.80%
Switzerland 22.10% 62.90% 15.00%
Brazil 22.30% 64.10% 13.60%
Bulgaria 24.20% 52.60% 23.20%
Mexico 25.30% 67.60% 7.00%
Trinidad 25.30% 65.70% 8.90%
Rwanda 25.30% 64.20% 10.50%
Japan 27.10% 17.90% 55.00%
Argentina 27.70% 60.00% 12.30%
Chile 30.20% 46.30% 23.50%
Poland 30.80% 51.00% 18.20%
Thailand 32.30% 40.60% 27.20%
Ukraine 32.50% 44.70% 22.80%
Zambia 33.60% 51.50% 15.00%
Romania 35.20% 40.90% 23.90%
South Korea 36.50% 26.40% 37.10%
Cyprus 36.50% 46.40% 17.10%
Russia 36.60% 43.70% 19.70%
South Africa 37.10% 49.50% 13.40%
Moldova 38.10% 39.00% 22.90%
Viet Nam 40.80% 37.70% 21.50%
China 42.30% 32.70% 25.10%
Taiwan 43.60% 36.00% 20.40%
Malaysia 49.00% 15.20% 35.70%
Morocco 50.80% 33.20% 16.00%
India 51.40% 20.50% 28.10%
Burkina Faso 52.30% 34.80% 12.90%
Georgia 52.50% 26.10% 21.40%
Turkey 53.30% 29.80% 16.90%
Ghana 53.60% 37.40% 8.90%
Indonesia 55.40% 36.20% 8.40%
Mali 62.40% 22.80% 14.80%
Iran 69.40% 16.50% 14.10%
Jordan 88.20% 7.90% 3.90%
Egypt 89.10% 4.30% 6.60%

Natives should get preference in jobs over immigrants
Country
Sweden 11.80% 79.90% 8.30%
Andorra 29.80% 53.20% 17.00%
Norway 34.70% 57.30% 8.00%
Netherlands 40.10% 49.80% 10.20%
Canada 40.90% 46.10% 13.10%
Australia 41.60% 36.40% 21.90%
France 42.10% 46.40% 11.50%
Serbia 44.70% 28.80% 26.40%
Switzerland 48.00% 35.50% 16.50%
New Zealand 51.90% 29.30% 18.80%
Great Britain 52.90% 36.40% 10.60%
Ethiopia 54.70% 29.30% 16.00%
Finland 54.90% 30.80% 14.30%
United States 55.40% 20.00% 24.60%
Germany 55.70% 27.90% 16.40%
Spain 57.70% 34.20% 8.10%
Thailand 61.20% 16.80% 22.10%
Japan 62.70% 6.10% 31.20%
Italy 63.90% 19.10% 17.00%
Turkey 64.40% 23.20% 12.40%
Romania 65.10% 14.60% 20.30%
China 66.00% 13.70% 20.40%
Ukraine 69.90% 16.20% 13.90%
Burkina Faso 71.70% 18.80% 9.50%
Argentina 71.90% 17.40% 10.70%
Hong Kong 72.30% 3.80% 23.90%
Uruguay 72.50% 21.30% 6.30%
Rwanda 72.60% 18.00% 9.40%
Slovenia 73.70% 15.00% 11.30%
Viet Nam 74.30% 10.80% 14.90%
Mexico 74.80% 19.60% 5.60%
India 75.20% 6.10% 18.70%
Moldova 75.50% 8.50% 15.90%
Bulgaria 76.60% 14.70% 8.70%
Zambia 77.00% 11.40% 11.60%
South Africa 78.30% 11.00% 10.70%
Cyprus 78.60% 12.20% 9.20%
South Korea 78.90% 2.40% 18.70%
Guatemala 79.60% 10.20% 10.30%
Chile 79.80% 7.50% 12.70%
Brazil 81.40% 9.50% 9.10%
Russia 81.40% 9.00% 9.60%
Poland 81.60% 8.40% 10.00%
Peru 82.20% 12.50% 5.30%
Mali 83.80% 7.10% 9.10%
Trinidad 84.00% 10.80% 5.30%
Morocco 84.90% 5.60% 9.50%
Ghana 85.20% 8.60% 6.10%
Malaysia 86.10% 2.10% 11.80%
Georgia 87.00% 4.50% 8.60%
Indonesia 87.40% 5.50% 7.10%
Iran 89.00% 5.40% 5.60%
Taiwan 91.00% 3.80% 5.20%
Egypt 97.90% 0.20% 1.90%
Jordan 98.50% 0.80% 0.70%

August 28, 2010

Nigerians agree despite religious differences

Filed under: Data Analysis,The Tenth Parallel,World Values Survey — Razib Khan @ 11:13 am

I am currently reading Eliza Griswold’s The Tenth Parallel: Dispatches from the Fault Line Between Christianity and Islam. The first half of the book is about Africa, and much of that is given to religious conflict in Nigeria. Africa’s most populous nation happens to be split down the middle religiously, with a Muslim north and a Christian south, meeting in the “Middle Belt” to contest. Griswold describes a very competitive religious marketplace.

One thing I was curious about though: are the religious conflicts in Nigeria simply due to coalitional fissures, or deep substantive divergences which track the religious differences? To illustrate, if Muslims and Christians share a village, then Christians who slaughter pigs in public places because pork is their primary protein source will likely have tensions with Muslims, who as a matter of substance object to pig slaughter which might pollute the landscape (this is a problem in parts of Southeast Asia where Muslims live downstream from Christians). In contrast, if you have economic difficulties in a region, and it is fractured ethnically or religiously, trivial tensions may quickly exploded into violence. In other words, in the second case religion is just a “quick & dirty” coalitional marker around which inevitable conflicts are going to swirl (in Mauritius Muslim Indo-Mauritians play a “wild card” role between Christian Creoles and Hindu Indo-Mauritians, despite greater substantive religious affinity with the Christians and greater cultural and racial affinity with the Hindus).

To answer this question I looked at the World Values Survey. For Nigeria there was data from 1995 and 2000, so I combined them to increase my sample size. Additionally, I wanted to focus on the Yoruba ethnic group, which is religiously divided between Muslims and Christians. In the WVS the religious categories actually break down further among the Christians, and I selected Pentecostals and Protestants for the Yoruba because of the large N for these groups, along with Muslims. Additionally, I selected Hausa Muslims as a comparison. The Hausa are an overwhelmingly Muslim northern ethnic group, while the Yoruba are a religiously pluralistic southern group (the Igbo of the southeast are as Christian as the Hausa are Muslim).

Please note that the survey was taken during a period of military rule by Hausa strongmen. I included only a subset of questions. You can follow to link to do your own queries.

Mus = Muslim, Pent = Pentecostal, Prot = Protestant. Some cells for Pentecostals are missing because for some questions all Protestants were aggregated together.

Question Hausa Mus Yoruba Mus Yoruba Pent Yoruba Prot
% Lower Class 51 34 22 23
% Completed University 6.5 12 36 23
% No Children 35 38 40 47
% Married 59 55 63 48
% Male 54 54 45 50
% Politics Very Important 39 20 12 21
% Work Very Important 89 94 96 91
% Religion Very Important 97 90 90 90
% Would not like people of difference race as neighbors 28 25 13 30
% Would not like immigrants as neighbors 28 20 12 15
% Would not like Muslims as neighbors 5.2 4 10 15
% Would not like homosexuals as neighbors 81 90 90 84
% Would not like people of a different religion as neighbors 36 14 20
% Totally satisfied with life (1-10 scale) 22 8 14 8
% Totally satisfied with financial situation (1-10 scale) 16 6 6 5
%Men should have more right to job than women 74 60 47 54
% Natives should have more right to job than immigrants 87 87 87 84
Mean, Ideal # of children 5.8 4.5 4 4
% A woman needs children to be fulfilled 81 93 96 93
% Disapprove of woman has single parent 89 76 92 77
% One should enjoy sexual freedom 23 16 9 11
% Marriage is outdated 18 19 11 13
% Agree strongly that men make better political leaders 63 56 44 51
% Agree strongly that university is more important for boy 46 22 12 21
% Very important that a woman is educated 80 76 82
% We need radical change in society 27 32 28 38
% Need larger income differences as incentives (1-10 scale) 21 25 41 15
% Gov. ownership of business should be increased (1-10 scale) 31 21 21 9
% The gov. should take more responsible (1-10 scale) 31 40 44 27
% Competition is harmful (1-10 scale) 7.2 8.2 1.8 14
% success is matter of luck and connections (1-10 scale) 12 13 4 13
% Wealth grows so there’s enough for everyone (1-10 scale) 28 17 29 28
% A great deal of confidence in armed forces 47 14 8 12
% A great deal of confidence in police 40 10 4 10
% A great deal of confidence in government 41 6 2 12
% A great deal of confidence in justice system 36 11 10 8
% A great deal of confidence in women’s movement 30 17 18 18
Mean, self position Left-Right (1-10) 6 4.9 5.5 4.7
Mean, rating of political system (1-10) 3.8 2.8 2.3 2.3
% Very satisfied with how democracy develops 28 13 15
% Favors open borders for immigrants 19 33 14 32
% Willing to fight for country 83 60 55 38
% There is a lot of respect for individual human rights 25 14 12
% Scientific advances will help 84 85 79 88
% A religious person 97 97 98 95
% God is very important in life (1-10) 88 92 93 86
% Get comfort and strength from religion 99 99 98
% Attend religious services more than once a week 72 85 58 60
% Raised religiously 97 97 93 90
% Believe in God 100 99 100 100
% Believe in Life After Death 87 80 86 86
% Believe in Hell 91 93 98 95
% Believe in Sin 85 97 98 95
% Believe in Devil 96 98 99 100
% Religious institutions give moral answers 85 69 71
% Agree strongly people atheists are unfit for public office 65 63 58
% Homosexuality never justifiable (1-10) 81 84 91 81
% Cheating on taxes never justifiable (1-10) 66 69 79 60
% Prostitution never justifiable (1-10) 77 85 90 79
% Abortion never justifiable (1-10) 78 83 82 72
% Divorce never justifiable (1-10) 55 64 74 56
% Very proud of nationality 78 66 57 60

And here’s a correlation matrix:

Hausa Mus Yoruba Mus Yoruba Pent Yoruba Prot
Hausa Mus * 0.94 0.9 0.93
Yoruba Mus * * 0.97 0.98
Yoruba Pent * * * 0.97

The Yoruba and Hausa have a high degree of agreement irrespective of religion, but there does seem to be a tendency for the Yoruba to agree across the religious divide. On the political questions I think the historical context is important. Additionally, it seems that the Pentecostals are the most religiously conservative and enthusiastic of these groups. Because the Hausa tend to be culturally Muslim (though there are a large minority of Hausa Christians in the sample) as a matter of course I was not totally surprised that they were not as zealous as one presumes Muslims to be.

Powered by WordPress