Sections

Research

The 2007 Brown Center Report on American Education: How Well Are American Students Learning?

Tom Loveless
A headshot of Tom Loveless.
Tom Loveless Former Brookings Expert

December 11, 2007

Introduction

This is the seventh edition of the Brown Center Report on American Education. As in the past, the report consists of three sections. The first section examines the latest test score data on math and reading achievement.This year the analysis focuses primarily on results of the 2007 National Assessment of Educational Progress (NAEP), including a discussion of NAEP achievement levels.

The second section investigates a general theme or trend in education. This year the second section scrutinizes enrollment patterns in private and public schools. Section three looks at an issue of policy relevance. International test data are examined to see whether a relationship exists between national math scores and the amount of time students spend learning mathematics in different countries.

In past Brown Center Reports, no effort was made to tie the three sections together. The studies stood on their own. This year, however, the studies do have something in common: they investigate phenomena in education that—at least at first blush—do not make sense. Data on the percentage of students performing at “proficient” on the NAEP performance levels, for example, are routinely cited as evidence that U.S. schools are underperforming. But as a 2007 study by Gary Phillips of the American Institutes for Research shows, every nation in the world—including high flyers such as Singapore and Japan—would have significant numbers of students falling below proficient if they were administered the NAEP test. How can that be?

The second section’s study of public and private school enrollment is precipitated by another oxymoron. Public opinion polls consistently show that the public considers private schools superior to public schools. Yet private school enrollment peaked around 1960 and has declined since then. People express a belief in public opinion surveys that they apparently contradict when selecting schools. What is going on?

The puzzle that is featured in the third section involves previous research on time and learning. When researchers have attempted to find a correlation between national test scores in mathematics and the amount of time different nations devote to teaching mathematics, no relationship has been found. Very odd. The correlation of homework and national test scores is stranger yet, with a negative relationship being the usual finding—the more homework given in math, the lower a nation’s test score. We analyze these relationships using a different approach and come up with different findings.

Increasingly, education’s most important questions are researched, debated, and decided with data. This is surely a healthy development as the field moves toward embracing the scientific methods that have benefited the intellectual disciplines that inform public policy. That said, with an election year now looming, it is important to be on the lookout for oversimplifications of either education’s most pressing problems or its most promising solutions. Simply marshalling good statistics is not enough. The educational enterprise is exceedingly complex, and many cross-currents exist in educational data—evidence supporting one hypothesis, when looked at from another angle, might be seen as supporting an alternative. Even with sound data, many mysteries remain in American education. This issue of the Brown Center Report explores three of them.

Download the Full Report
Download the Press Release
Download Presentation by Tom Loveless
Read More about the Launch Event and View Clips
Read More about the Brown Center Report on American Education (2000-2007)