Showing posts with label Social Science. Show all posts
Showing posts with label Social Science. Show all posts

Friday, February 05, 2010

Ethnicity among social scientists

The General Social Survey (GSS) sampled 118 American social scientists (economists, psychologists, sociologists, urban planners, and others). Here are the percentages by ethnicity:

The ratios are the share of all social scientists of a particular ethnicity compared to their share of the total population.  Americans of English, Italian and especially Jewish descent are overrepresented. Those of German, black, and especially Mexican heritage are underrepresented.  (Keep in mind that a GSS sample includes anyone who self-identifes as a social scientist, so it will include a lot of schmucks like me who barely count).

Saturday, October 31, 2009

Focusing on aggregates

Americans are taught to judge people individually, and in situations where you can get to know someone, this makes sense. But the truth is that you can predict aggregates better than individuals.

Using General Social Survey data, I calculated the correlation between a measure of IQ (WORDSUM) and income (REALINC) for almost 17,000 people born in the U.S.  It turned out to be .28. If you square that number, you get .08 which is called R-squared. It is interpreted as the proportion of variation in income that can be explained by your IQ. In other words, if I know one thing about a person--his IQ square--I am not going to be able to predict his income level with any accuracy at all.

But the situation changes dramatically if I calculate mean IQs and mean incomes for the 29 ethnic groups which have at least 30 respondents in each group. Now the correlation jumps all the way up to .77. If we square that, we get .59, which means that 59 percent of the variation in mean income is explained by the variation in mean IQ scores. So if I've got a random group of, say, Americans of Russian descent, chances are their average IQ is high, and I can make a pretty good bet that the group will earn an above-average income as well.

This is why Richard Lynn and Tatu Vanhanen's approach is so effective in IQ and the Wealth of Nations.  You grab a random guy in Japan, he might be smart or dumb; he might be rich or poor. But tell me the mean IQ of the country is 106, and I'm putting my money on it being a wealthy place.

HBD-ers are criticized for focusing on groups, but reality is most predictable at that level, and being scientific is being concerned with prediction.  

Thursday, June 12, 2008

Seven of the ten most violent countries in the world are Latin American: The World Heath Organization provides data on homicide rates in more than 70 countries. Here are the most recent estimates:

Male homicide victimizations per 100,000--most recent year available

1. Colombia 117.0
2. El Salvador 70.0
3. Sri Lanka 50.8
4. Russia 50.2
5. Brazil 49.0
6. Venezuela 48.7
7. Puerto Rico 33.2
8. Guatemala 32.1
9. Ecuador 30.8
10. Philippines 28.3

11. Paraguay 22.2
12. Kazakhstan 20.0
13. Estonia 19.6
14. Mexico 18.0
15. Ukraine 17.8
16. Panama 17.5
17. Latvia 16.8
18. Belarus 15.7
19. Moldova 14.7
20. Trinidad and Tobago 13.2

21. Argentina 12.1
22. Turkmenistan 11.9
23. Albania 11.8
24. Lithuania 11.7
25. Costa Rica 9.8
26. Chile 9.8
27. Kyrgyzstan 9.6
28. Thailand 9.5
29. USA 9.2
30. Uruguay 8.6

31. Cuba 7.9
32. Guyana 7.7
33. Georgia 6.9
34. Romania 5.2
35. Bulgaria 4.6
36. Uzbekistan 4.4
37. Azerbaijan 4.4
38. Armenia 3.6
39. Israel 3.5
40. Finland 3.5

41. Slovakia 3.4
42. Mauritius 3.2
43. Hungary 2.8
44. Poland 2.6
45. Portugal 2.5
46. Canada 2.2
47. Australia 2.1
48. Belgium 2.0
49. Slovenia 1.9
50. Korea 1.8

51. Ireland 1.8
52. Croatia 1.8
53. Bahrain 1.8
54. New Zealand 1.7
55. Spain 1.6
56. Netherlands 1.6
57. Greece 1.6
58. Czech Republic 1.6
59. Italy 1.5
60. Sweden 1.3

61. UK 1.2
62. Kuwait 1.1
63. France 1.1
64. Denmark 1.1
65. Norway 0.9
66. Hong Kong 0.9
67. Germany 0.8
68. Singapore 0.8
69. Switzerland 0.7
70. Austria 0.7

71. Japan 0.7
72. Egypt 0.1

The United States has a homicide problem, and one way to look at these numbers is in terms of how immigrants are likely to improve or worsen the level of violence here. Immigrants from East Asian and Western European regions should make things better, while people from Latin America should worsen the problem. Eastern Europe is a mixed bag. Of course, sub-Saharan Africa is also violent, but the countries don't have their act together enough to collect the necessary data.

Let's compare Colombia and Japan: the rate is 167 times higher in Cokumbia!

Now, of course, it's more complicated than this since immigrants are not usually representative of their mother countries, and conditions are different here than at home. But there is certainly a correlation: Asians and Western Europeans immigrants have proven to be law-abiding here, while Latin Americans have above-average rates of criminal involvement.

These patterns also undermine the arguments that social conditions in the U.S. generate criminality among immigrants: people who come here from violent countries tend to be violent here. They bring the proclivity with them.

One other point. An academic explanation of crime which has gone from being new and unknown to dominant in the past decade is called institutional anomie theory. It claims that societies have a crime problem to the extent that the economy dominates over other important institutions: families, schools, churches, government, etc. The free market generates the motivation to break the law, while all these other segments of society constrain behavior.

The bottom line here is that the freer the economy, the more crime that will be observed. Now, I can imagine an honest theoretician a century ago suggesting such an explanation, but even an amateur now knows that the data contradict this idea. Just look at the list. Many of the freest societies have little homicide, while many heavily regulated societies have a real problem.

International data is widely available, so Steven Messner and Richard Rosenfeld (the theorists) have no excuse. This is an obvious case of responding to the political demands of the discipline, rather than the realities of crime.

I'll say it again: mainstream social science is a sham.

Tuesday, June 10, 2008

Proof positive that mainstream social science is pure superstition: I just remembered two things I learned in grad school, and was too stupid to put them side by side and see their absurdity. I was taught by Ivy League faculty that: 1) (hetero)sexual attraction is socially constructed; and 2) homosexuality is inherited. So, heterosexuality is not from nature, but homosexuality is!