What Ails Harvard?

"A new scholar who is doing interesting work in Jason Richwine, a recent PhD graduate from the Kennedy School of Government at Harvard and is a research fellow at the American Enterprise Institute. ... One of Richwine's achievements has been to calculate the average IQ of various immigrant groups--he has provided far more detail on this matter than any other research I am aware of. ... The results are, for the most part, predictable, as the IQs of most immigrant groups are approximately the same as those of their home countries: Mexican immigrants have an average IQ of only 82, which is below that of American blacks, and Northeast Asians' IQ is 106. ... The average Indian IQ is 81, but Indian immigrants to the US have an average IQ of 112, leading them to become the nation's 'new model minority', as Richwine wrote in a Forbes article. ... Richwine works with Robert Putnam's concept of 'social capital,' which designates social networks based on turst and reciprocity. ... Where social capital is greater, Putnam says, 'children grow up healthier, safer, and better educated; people live longer, happier lives; and democracy and the economy work better.' ... Richwine brings together scattered research findings proving that IQ and the capacity to build social networks are related. He refers to the Kohlbergian tradition of research in moral development that finds that high IQ people are more likely to reason about moral issues in a principled manner. ... Richwine suggest that the link between IQ and social capital ought to inform our immigration policy: the negative effects of diversity on trust would be mitigated by restricting entry to high-IQ immigrants", Ian Jobling at White America, 19 August 2009, link: http://whiteamerica.us/index.php/blog/print/jason_richwine_on_iq_immigration_and_social_capital.

"High-school students' performance last year on the SAT college-entrance exam fell slightly, and the score gap generally widened between lower-performing minority groups and white and Asian-American students, raising questions about the effectiveness of national education reform efforts. ... The reading scores are the worst since 1994. Many observers Tuesday viewed the flat results of recent years as discouraging in light of a more than 25-year effort to improve US education. 'This is a nearly unrelenting tale of woe,' said Chester E. Finn Jr., president of the Thomas B. Fordham Institute., a Washington, DC think tank. ... The results come a week after the disclosure that only a quarter of 2009 high-school graduates who took the ACT, the other main college entrance exam, had the skills to suceed in college. ... A record 1.53 million students took the [SAT] in 2009. About 40% were minority students, up from 29% in 1999. Education analysts said scores would be expected to drop as more students take the test, so College Board officials interpreted the stability in scores as encouraging. ... Jack Jennings, president of the Center on Education Policy, a nonpartisan research organization in Washington [said] 'The bottom line is the country is changing dramatically. Unless minority kids are educated better, we are going to be in trouble because pretty soon they are going to be the majority'," my emphasis, John Hechinger at the WSJ, 26 August 2009, link: http://online.wsj.com/article/SB125121641858657345.html.

Gregory Mankiw (GM), Harvard professor has a 28 August 2009 post which mentions "omitted variable bias" in attacking a NYT piece, link: http://gregmankiw.blogspot.com/2009/08/least-surprising-correlation-of-all.html. Way to go GM!

GM notes his "previous blog post on SAT scores and income generated a surprising amount of blogosphere pushback", 29 August 2009, link: http://gregmankiw.blogspot.com/2009/08/and-i-thought-i-was-being-boring.html. GM, you're naive. Don't you remember the 2005 flap over Larry Summer's (LS) statements about women and mathematics ability? You and LS are in the same department. Talk to him.

What gives? Richwine has a Harvard PhD and Putnam is a Harvard Professor. I suggested auctioning immigration slots decades ago.

Suppose someone tried to teach me to play basketball. He would have had a "tale of woe" that could have lasted for decades too. So? Are our "education analysts" missing something? Why should average SAT scores fall as more students take the test? If you have a representative sample of high schoolers in your "standardization sample" all other things being equal, if your sample size increases, there is no reason average scores should drop. If you encourage more high schoolers with bad grades to take the SAT, the average score will drop as high school grades and SAT scores are positively correlated. Are these "analysts" afraid to identify this possible explanation: as the percentage of Non-Asian Minority (NAM) test takers increases, average SAT scores will drop and there is nothing anyone can do about it. It doesn't matter if "minority kids are educated better", if we even know what that means. As America's demographics change, we will come to resemble the native lands of our new immigrants. That is the bottom line.