Monday, November 15, 2010

The Gini Index and Educational Achievement By Walt Gardner

With reformers relentlessly demanding that schools produce measurable outcomes, it's curious that the Gini Index is rarely mentioned. I say that because what Italian statistician Corrado Gini wrote in 1912 has direct relevance to today's debate.

Sometimes referred to as the Gini coefficient, it measures the range of income inequality in a society from 0 (no inequality) to 1(total inequality). Sweden, for example, has an index of .23, while Namibia has .7. The U.S. has one of the world's worst Ginis for an industrialized country at .468 in 2009. This is not surprising since wealth is being reconcentrated in the upper one percent of the population in a way not seen since the Gilded Age.

The change has not escaped the attention of commentators. In his Nov. 7 New York Times column, Nicholas D. Kristof wrote that the U.S. "now arguably has a more unequal distribution of wealth than traditional banana republics like Nicaragua, Venezuela and Guyana" ("Our Banana Republic"). Echoing this view, on Nov. 8, the Los Angeles Times published an op-ed by Michael I. Norton and Dan Ariely based on a sample of 5,000 people, including young and old, men and women, rich and poor, liberal and conservative. They found, among other things, that "Americans reported wanting to live in a country more like Sweden than the United States" ("Spreading the wealth").

The implications for schools are inescapable. Researchers have repeatedly emphasized the effects that poverty has on performance. According to UNICEF, the U.S. already had the highest rate of childhood poverty in the industrialized world long before the latest Census Bureau report showed that one in five children are now living in poverty. Overall, the share of Americans in poverty climbed to 14.3 percent in 2009, the highest level since 1994.

If Gini were alive today, he would find great material for his index. When any society is characterized by a high index, it is bound to exhibit the kind of socioeconomic differences that impact schools. Students can overcome their backgrounds, but they tend to constitute a small percentage of the overall population. That's why it makes little sense to compare test scores of one country with the test scores of another. Just as Zip codes serve as a reliable predictor of scores on standardized tests, so too does a country's Gini serve an equally valuable purpose.

This does not mean that inspired teachers can't help students from chaotic backgrounds. As Richard Rothstein recently explained in his keynote at ASCD's 2010 Conference on Teaching and Learning ("One-Third Agenda Won't Close Gaps"), teachers are the most important in-school factor in achievement. But teachers are not miracle workers. By themselves, they cannot compensate for the deficits that students bring to the classroom. These out-of school factors play a disproportionately large role in academic performance. The Harlem Children's Zone recognized the distinction. That's why it provides its students with wraparound services, which are underwritten by wealthy philanthropists.

In light of these facts, I have a proposal. Require that whenever test scores are published, the Gini index must be published with them. After all, federal law has long mandated that stocks may not be sold to the public without a prospectus. Why should test scores be any different? They are evidence of investment in public education. The more relevant information taxpayers have, the better able they will be to make judgments about the performance of schools.

http://blogs.edweek.org/edweek/walt_gardners_reality_check/2010/11/the_gini_index_and_educational_achievement.html

No comments:

Post a Comment