EDUCATION: Math performance by 15 year olds

Hank Levin comments on an international study comparing math performance of 15 year olds, in which US children did not do very well: This is an important finding, but can be easily misinterpreted.  Let me make just two points:
1-       Performance on these tests is heavily weighted by family background (parents present, number of siblings, parental education, income, and so on).  Despite all of the emphasis by the present administration on “evidence-based school decisions” for spending the funding, the body of scientific evidence on how to improve schools and spend more efficiently is amazingly devoid.  At one time it was believed that educational researchers were not up to the task, so for the last twenty-five years the research funds have been spread among economists, statisticians, sociologists, mathematicians, and other more reputable research communities.  Still, today, when educational research concludes that one method or approach is “statistically superior” to another, the actual achievement results are quite small, typically less than one-fifth of a standard deviation.  The average African-American achievement score is about one standard deviation (17th percentile) below the average for Anglos (50th  percentile).  Only about one-fifth of the gap would be closed by those interventions supported by strong evidence, and even that is exaggerated because the evidence is usually based upon small-scale results rather than national replication.  It is well-known that replication of results is far from routine as in the case of the evidence on reduction in class size.
2-       Public Expenditure data are highly inadequate for assessing the total resources going into education.  My colleague Professor Mun Tsang (an MIT- and Stanford-trained economist) and his students have done numerous studies around the world. For example, Korea appears to provide a bargain with a top score and relatively low educational expenditure when one looks at public budgets.  However, a careful study of Korea found that for every dollar of public expenditure, families spend more than 50 cents from their private resources on education, on average.  Of course, some families spend more and some spend less.  For the poor it is an enormous sacrifice.  Most of this private money is spent on tutoring and cram courses for the exams, so it is not clear whether it is the test preparation and repetition that is successful in raising test scores or actual learning (or some combination).
3-       Political Correctness in the U.S. refuses to acknowledge the fact that a large portion of minorities and immigrants are in family situations that do not support school achievement.  For example, an overwhelming majority of black children are born out of wedlock.  About one-quarter of black men in their twenties who have dropped out of school (half of black males do not graduate from high school according to the best data sources, Urban Institute?Swanson study) are incarcerated.  Many live in sub-standard housing in violent neighborhoods. Single parents with poor education predominate among many of these families.  Immigrant families from poor regions of Latin American (e.g. Mexico) arrive with a poor primary school education in which they are not literate in their first language.  These are the socioeconomic conditions that forty years of research tell us will produce poor educational results?and that evidence is the most consistent and solid that we have in educational research.  The achievement gap between such “at-risk” children and the mainstream child is already large when children enter school, and the differences are reinforced by the out-of-school situations.  Bear in mind that children spend about 91 percent of waking hours from age 0-18 out-of-school and about 9 percent in school.  We need to be honest as a society and recognize that unless we can intervene with the demographics and living conditions, schools are highly limited in their effectiveness.  Pat Moynihan got in trouble for writing this basic truth in 1965 in a report on The Black Family when he was Assistant Secretary of Labor.  Only recently have we seen persons like Bill Cosby and Spike Lee raise the issue again, and they have been castigated for doing so.

Your comments are invited. Read the home page of the World Association of International Studies (WAIS) by simply double-clicking on: Mail to Ronald Hilton, Hoover Institution, Stanford, CA 94305-6010. Please inform us of any change of e-mail address.

Ronald Hilton 2004


last updated: January 16, 2005