Our next goal: we need to pack more learning into fewer years, for fewer dollars. Or else. (Viewpoint).

I'VE BEGUN TO DREAD LISTENING TO THE DEBATE ABOUT America's schools. We create new tests to raise the bar on students, then wonder why the scores are low. We declare that more students must go to college, then complain that SAT scores are dropping, and the pool of college-bound seniors doesn't seem as bright. And when we have ourselves thoroughly confused by numbers, we give up and resort to anecdotal evidence.

For a more useful approach to analyzing education, I recommend a new report from Educational Testing Service, The Twin Challenges of Mediocrity and Inequality: Literacy in the U.S. from an International Perspective, a readable, compelling, 42-page study of how the U.S. compares to its peer nations in teaching basic literacy. Written by Andrew M. Sum, professor of economics at Northeastern University; Irwin S. Kirsch, senior research at Educational Testing Service; and Robert Taggart, president of the Remediation and Training Institute, the report has important messages for higher education.

The report analyzes data from the International Adult Literacy Survey (1994-1998), a 23-nation assessment of proficiency in reading prose, handling quantitative material, and working with documents such as application forms. Sum, Kirsch, and Taggart worked with the IALS data to compare the educational achievement of 17 high-income countries--the U.S. and most of its developed-nation peers. (The survey included mostly Western European nations, excluding a few U.S. peers such as Japan.)

On average, we didn't do so badly in the analysis. U.S. mean proficiency scores are almost identical to the mean scores for the whole group. In country rankings, we don't do as well: 9th out of 20 in the prose scale, 14th on the document scale, 13th on the quantitative, and 12th on the composite. In short, on average, we're not the world leaders we'd like to be, but we're keeping up.

The bad news is in the details: The IALS breaks scores into five levels. For instance, on the prose score, level one means you can find a simple piece of information in a newspaper article. At level five, you're analyzing to extract a fairly complex argument. It's thought that you need to be at level three to succeed in the contemporary labor market. When Sum et alia analyze the distribution of scores internationally, they notice some interesting points:

* The U.S. has a large number of people at the highest proficiency levels. (Only Sweden has more citizens with level-five prose scores.)

* We also have a lot of low-scorers. About 21 percent of adults aged 16 to 65 scored at level one in prose--and only five peer nations did worse. Overall, about 45 percent the U.S. population scored in levels one and two on the composite scale.

* Our highest scores tend to be higher than other countries' and our lowest scores lower. "The U.S. is the world leader in literacy inequality among young adults," say Sum and company.

It's impossible to do Twin Challenges justice in this space; download it at www.ets.org/research/pic, and you'll find your own points of interest. Here's what leaped out at me:

The U.S. leads the world in the number of years of schooling it provides its citizens. We're number one in the world in the percentage of people with 12 years of education, and number one in the percentage with 16 years. To the extent that our scores are satisfactory at all, those extra years of school are a major factor. But we pay for them. We tend to pay more per year of schooling than our peer nations, and we pay for more years; Sum, Kirsch, and Taggart estimate that, including factors like lost wages and opportunity costs, education accounts for about an eighth of the U.S. GNP.

The problem is this: At every level of educational attainment, U.S. citizens do less well in measures of basic skills than their counterparts in peer nations. Our high school graduates have lower average scores than the graduates of other nations' high schools. So do our college students. The system, despite its many desirable aspects, isn't very efficient in terms of output per dollar or output per year.

In an increasingly competitive world, that's a situation that can't last indefinitely. We've already seen the kinds of actions our society has taken to attack the deficiencies, real and imagined, of K-12 education-regulation, politically driven curriculum reform, incentive systems that punish the needy. We don't want that sort of intrusion into higher ed. But it's probably headed our way if we can't solve the essential problem of efficiency. Tuition hikes can make higher ed unpopular. Issues of access can trouble our consciences. But inefficiency, I fear, can permanently damage a precious institution.

More learning in fewer years, for fewer dollars. It sounds crass, but education will not be able to control its own fate unless it accepts this as a goal. Is efficiency on your agenda? It should be.

Patrick Clinton is a consultant to leading trade/consumer publications, and author of Guide to Writing for the Business Press. He has been an assistant professor at Medill School of Journalism, Northwestern University, and editor of this publication.