Are low-income students the reason why the U.S. does poorly on international tests? No.

Editor’s note: This is the second installment in a year-long series in which Founder and American Experiment Senior Fellow, Mitch Pearlstein, focuses on student achievement and how Minnesota students compare with others around the country, as well as how American students compare with other young people around the world. 

More than most people who write about academic achievement and reform, I focus on what’s going on in the lives of children and adolescents outside of school. This is certainly not to say I’m oblivious to often immense differences in quality among schools.  It is to say that debates and efforts about how to improve learning in Minnesota and the rest of the United States routinely downplay (as I’ve been known to note) immense rates of family fragmentation and other social and cultural impediments to learning.

In this spirit, I fully recognize (no great insight here) how poverty holds back many students, a truism with which virtually all educators, scholars, and commentators agree, needless to say.  Nevertheless, too few of them adequately acknowledge that a big reason why large numbers of young people are poor is precisely because nonmarital birth rates are higher in the United States, often far higher, than just about anywhere else in the industrialized world.  Or more accurately, they may well agree that family fragmentation is a direct cause of poverty; they just don’t like to say so publicly, for whatever their P.C. or other reasons may be.

But what if the proportion of families who are poor, whatever the causes, is in fact substantially higher in the United States than in other places?  Wouldn’t that give credence to the view that American kids in the aggregate are not doing well, less because of the weaknesses of thousands of schools and more because of the economic fragility of millions of families?  Wouldn’t that at least suggest that many teachers and administrators have been overly criticized for decades and continue to be?

Yet what if the proportion of American students growing up in poor families really isn’t higher than is the case elsewhere?  Wouldn’t that suggest that poor performance in math, science, and reading is attributable, in more than passing ways, to too many inadequate schools, driven by too many questionable policies, directed and employed by not enough strong administrators and teachers?  It would seem so.

So what do national and international data instruct in these matters?  Let me turn to an acute analysis by a friend of mine, Michael J. Petrilli, president of the Thomas B. Fordham Institute in Washington, and Brandon Wright, the editorial director there.  Their article – “America’s Mediocre Test Scores: Education Crisis or Poverty Crisis?” – appeared originally in the Winter 2016 issue of Education Next, a very good journal.  Excerpts and citations below are from an abridged version of their paper, dated December 16, 2015, disseminated by the Fordham Institute.

Rank Rankings

Petrilli and Wright argue that in order to prove poverty is a “major factor” in causing America’s “meager academic achievement,” at least one of two claims need to be demonstrated.

  • Either poor students in the United States do worse than do poor students elsewhere.
  • Or the child poverty rate in our country is “substantially” higher than child poverty rates in countries to which the United States is compared.

Please note that the term “poor student” in this context has to do with their families’ socioeconomic status; not the quality of their academic performance.

Petrilli and Wright correctly point out that, “Low-income students everywhere do worse on tests of academic achievement than their more affluent peers,” and that this is “as true in the United States as in every other country ever studied.”  The key question, they write, is not whether poor kids here (always on average) do worse than richer kids, as they do, of course.  Rather, it’s whether our poor kids do less well than poor kids in other countries.  How to find this out?

As with many scholars, they look to test data collected under the aegis of the Organization for Economic Co-Operation and Development (OECD) by the Program for International Student Assessment (PISA).  More specifically, PISA developed an index that takes into account parental occupations, education, home educational resources, family wealth, and possessions connected to “classical” culture.  This information is then used to stratify student populations of various countries into quartiles.

Without my delving into the methodological weeds, Petrilli and Wright describe how this approach enables researchers to compare how similarly situated students in those countries do on tests administered by PISA.  So we learn, for example, that Belgium and France do relatively better at teaching their “higher-status students” than their “lower-status students,” while Canada and Finland do the exact opposite.  (For deeper weeds, go here and here.)

How does the United States fare?  It seems we do equally well (or equally poorly) at teaching our least well-off young people as we do in teaching those from families in the in the index’s top quartile.  Meaning more specifically?  The authors cite another study by three exceptional scholars – Eric A. Hanushek of Stanford, Paul E. Peterson of Harvard, and Ludger Woessmann of the University of Munich – which argues:

Overall, the U.S. proficiency rate in math places the country at the 27th rank among the 34 OECD countries that participated in the Program for International Student Assessment (PISA).  That ranking is somewhat lower for students from advantaged backgrounds (28th) than for those from disadvantaged backgrounds (20th).

Beyond taking the opposite of comfort in such rankings, what do these two findings mean? Petrilli and Wright conclude this portion of their paper by noting how, rather than demonstrating how disadvantaged American students do less well when compared to disadvantaged students in other nations, it’s actually advantaged American students “who are falling short in international comparisons.”

Think about that for more than a second.  Consider, for example, how large proportions of Minnesotans and others across the country, in lamenting huge achievement gaps, surely assume there is far less to worry about – if anything to worry about at all – when it comes to our best students.  Isn’t competition intense, after all, reaching for the Ivies and other highly selective colleges and universities?  Also consider how parents, after spending big money moving to affluent suburbs because of presumably first-tier schools there, are not necessarily inclined to think of what goes on in them as second tier.  But that often is the case across the country, Minnesota included.  If it were not, “advantaged” American students would not be “falling short in international comparisons.”

This is the segue to the question posed in the second bullet above, about whether child poverty rates in the United States are substantially higher than elsewhere?

Demonstrated so far is that the performance of disadvantaged students is less a cause of America’s bad international rankings than is the performance of our most advantaged students.  But what if we simply proportionately have a lot more kids in poverty than do other nations?  If that is the case, it’s still possible, Petrilli and Wright acknowledge, that “students from low-income families are dragging down our national averages.”  And that if that is true, “poverty could still be the explanation for our mediocre test scores.”

Using 2010 data from what’s known in the trade as the Luxembourg Income Study, Petrilli and Wright calculated poverty rates in various countries, taking into account all types of income, including social welfare benefits.  By so doing, the U.S. poverty rate turns out to be lower than that of the United Kingdom, just about the same as Germany’s, and “barely” higher than Finland’s.

It is true, as Petrilli and Wright point out, that some scholars – particularly those who believe the United States has more of a “poverty” crisis than an “education” crisis – define poverty much more broadly than in the way just mentioned.  For them poverty is defined in “relative” terms, as in the proportion of families “with less than half the median income” in a particular country.  By this conception, the United States really does have more children living in poverty than is the case in most other advanced nations.

But as Petrilli and Wright again correctly point out, such a definition and indictment have less to do with poverty than with the much more encouraging fact that median income in the United States is high.  We can be proud of that while simultaneously appreciating how many people are angered by what they see as too much income inequality – which, getting to the nub, is what they see as an inevitable and unfair result of the skewed economic successes of recent years.  Still, it’s absurd to suggest that great numbers of Americans with incomes less than half the median live in “poverty.”  If this were the case, Minnesota could be seen as particularly and paradoxically impoverished precisely because we are a relatively wealthy state with an above-average median income.

Petrilli and Wright conclude by arguing that “poverty can’t explain away America’s lackluster academic performance” no matter how “soothing” it might be for many educators and politicians among others to use it as an excuse.  They should stop, the two write, and instead “start getting serious about improving the achievement of all the nation’s children.”

That last line reads too much as a cliché for my taste.  But persuasively explained and fundamentally important is how I would describe everything else they argue.

Dr. Pearlstein’s most recent publication is “Can America’s Religious Traditions Strengthen Marriage?  Minnesota Leaders Say ‘Yes’ and Propose How.”