International Tests Confuse Taxpayers
The results of tests of international competition are increasingly scrutinized as prima facie evidence of educational quality. Whether valid inferences can be drawn from the data alone is itself debatable. But more important, researchers confuse taxpayers when they reach different conclusions at a time when so much is riding on the outcomes.
The latest example involves the duel between Paul Peterson and David Berliner, the lead researchers of two separate analyses of the 2012 Program for International Student Assessment.
Peterson, Eric Hanushek and Ludger Woessmann conclude that schools don't do any better teaching students from better educated families than they do teaching those from less well educated ones ("U.S. Students from Educated Families Lag in International Tests," Education Next, Fall). They go on to label David Berliner and Gene Glass as apologists for the lackluster performance of American students compared with their counterparts abroad.
Yet a closer look at what Berliner and Glass have written calls that characterization into question. Berliner explains that the study Peterson et al. produced is based on the connection between the "level of parental education" and student performance ("Criticism via Sleight of Hand," Diane Ravitch's blog, Jul. 29). Berliner and Glass's study is based on the connection between "level of parental poverty" and student performance.
Parental income and parental education are sometimes related, but they are not interchangeable. And that's where the confusion lies. For example, of the nearly three million recent college graduates, 152,000 are working in retail sales, nearly 100,000 are employed in food service positions, 80,000 are clerks or customer service representatives, and 60,000 work in construction or manual jobs.
They live in neighborhoods with other working class parents. Their children go to the same school as the children of parents without a college degree. So parental income, rather than parental education, determines where children live and who their classmates are. As the Coleman Report in 1966 explained, these factors trump the quality of schools in determining the difference in average achievement between ethnic groups.
In fact, in analyzing the results of the Program for International Student Assessment, the Trends in International Mathematics and Science Study, and the Progress in International Reading Literacy Study, Berliner and Glass found that American students posted impressive results where the poverty rates of families were less than ten percent, or even between 10 and 25 percent. Conversely, where the poverty rates were in the 75 percent or higher range, students performed abysmally.
The operative word in understanding the conflicting claims made by Peterson and Berliner is "advantaged." If it refers to parental education, it has a different meaning than if it refers to parental income. Unfortunately, taxpayers in this country who are entitled to know if children are being well educated are misled, and the media do not help.