Tag Archives: testing

In The Mismeasure of Teaching Time, Sam Abrams exposes the myth that teachers in the United States spend nearly twice as much time leading classes as teachers in many other OECD countries. Abrams, Director of the National Center for the Study of Privatization in Education at Teachers College, Columbia University, details basic contradictions in the U.S. figures reported by the OECD and repeated by many journalists and scholars. His analysis suggests that U.S. teachers still do spend more time teaching than their counterparts in other OECD countries, but only about 15 percent more. This is still an important difference, but, as Abrams argues, one that has overshadowed more significant differences: in particular, teacher pay and the structure of the school day. Soon after the study was published last week by the Center for Benefit-Cost Studies of Education (CBCSE), we had a chance to ask Abrams when he first noticed the problems with U.S. figures, how he determined what was going on, and what the implications are for educators and the general public (see also recent EdWeek coverage). We share his response below:  

After repeatedly seeing this misinformation about teaching hours in books and articles, I wrote to the OECD in January 2012 to inform them that the U.S. hours were way off and provided as evidence terms of teacher contracts from several major school districts.  I moreover explained that I had been the scheduler of a public high school in New York City for seven years as well as a teacher for many more. The people I contacted at the OECD conceded the U.S. hours appeared inflated and relayed this information to the U.S. representative to the OECD, who, in turn, I was told, stood by the numbers.  I decided at that point to save my argument for a book I was writing on educational privatization.  But as that book was taking longer than expected and as this myth was getting repeated on a regular basis in op-ed pieces, think tank studies, and books, I decided in October of 2014, after compiling more evidence, to address the problem directly with the National Center for Educational Statistics (NCES), the source of the U.S. figures.

Tom Snyder, the NCES Program Director for Annual Reports and Information, could not have been more helpful. When presented with my case, as I wrote in my study, Snyder undertook with his staff a review of their data and a week later reported that the United States had indeed been inadvertently overstating teaching time. Snyder’s openness and cooperation was a lesson to me that researchers can get a lot more assistance from authorities than I had thought.

I should add that if I had not been a teacher, I would not have been in much of a position to question the data.  This implicitly points to the divide between policy and practice. As a teacher, I had never heard of Education at a Glance and I didn’t read many books about education policy.  I read books about history and economics, as I was a teacher of history and economics, and I struggled to keep up with reading and grading my students’ papers.  It was only after I became a researcher and began to work on my book on educational privatization in 2008 that I started reading a lot of books about education policy.  With my background as a teacher, I could quickly see that some policy analysis, like this argument about teaching time, didn’t hold up.

The salient implication of this finding is that we’ve been tilting at windmills. Teaching time is a phantom problem.  U.S. teachers do teach more than their OECD counterparts but, as I explain in my study, only marginally more.  The real and telling differences between teaching in the United States and other OECD nations concern relative pay and the structure of the school day.  These problems have been obscured by the difference in teaching time because the alleged difference has been so dramatic.  For journalists and scholars, that dramatic difference has been impossible to ignore.  And they’ve understandably focused on it at the expense of these two other real and telling differences.

As I noted in my study, U.S. primary teachers, according to Education at a Glance, earn 67 percent as much as their college classmates while their OECD counterparts earn 85 percent; U.S. lower-secondary teachers earn 68 percent compared to 88 percent for their OECD counterparts; and U.S. upper-secondary teachers earn 70 percent in contrast to 92 percent for their OECD counterparts.  The data on pay appear quite reliable, as I explain in my study, because the method of collecting data on pay differs substantially from the method of collecting data on teaching time.

In absolute terms, U.S. teachers may make as much as their OECD counterparts but not in relative terms, because in other OECD nations, on average, bankers, lawyers, doctors, engineers, and management consultants make much less.  So, this is a social contract issue.  It isn’t so much that U.S. teachers don’t earn much. It’s that their college classmates who went into banking, law, medicine, engineering, and consulting make a lot more.  Teachers accordingly get priced out.  It’s thus hard to attract people to the profession and hard to retain them if they come aboard.  Fixing this problem constitutes a steep challenge.  It would necessitate raising not only teacher pay but also marginal income tax rates as well as tax rates on long-term capital gains.  But we at least need to look at this problem with our eyes wide open.

What’s not hard to fix is the structure of the school day.  In this regard, as I explain in my study, U.S. practice differs significantly from that of many other OECD nations. The architects of No Child Left Behind (NCLB) and Race to the Top (RTTT) wanted to close the achievement gap by identifying academic deficiencies through regular testing and by holding school administrators and teachers accountable for their students’ results. The goal of closing the achievement gap is clearly noble. But the high-stakes testing that defines NCLB and RTTT has had perverse consequences. In particular, it has reduced, as I documented in my study, time for play, art, music, and drama because school administrators have felt great pressure to pack more academic instruction and test prep into the school day in order to boost test results.

School administrators have had little choice.  Test results are ultimately relative.  If school administrators in one district reduce time for play, art, music, and drama to make more time for academic instruction and test prep, then school administrators in neighboring districts are hard pressed not to do the same.  We have a stadium effect.  If spectators in the first row at a basketball game rise to see a big play, then spectators in the next row must do the same. In a heartbeat, everyone in the arena is on their feet. The same holds for test prep and schools.  The result is a narrowed curriculum, intense pressure on students, teachers, and principals alike, and a tight day, lacking the breaks between classes necessary for students, teachers, and principals alike to regroup, reflect, and get some fresh air.

There has long been an assembly-line pace to the school day in the United States, with short breaks between classes and brief lunch periods. Raymond Callahan made that clear more than fifty years ago in his incisive book Education and the Cult of Efficiency, published in 1962. But with the high-stakes testing introduced by NCLB and RTTT, the pace has intensified.

As a coach for several years with the Ice Hockey in Harlem program in Central Park, I’ve come to learn that kids of all ages, from five to seventeen, crave and need play. We can’t get the kids off the ice at the end of practice. Kids need to improvise with their bodies and experience the joy of playing with their peers. At a certain point, as I write in my study, more academic time becomes counterproductive. Leaders of such major companies as SAS and Google, in this light, have understood that more work at the expense of relaxation and recreation likewise becomes counterproductive and have accordingly encouraged their employees to take breaks and designed their offices with relaxation and recreation in mind.

We could fix this, as I explain in my study, by getting rid of high-stakes testing.  We have had the National Assessment of Educational Progress (NAEP) since 1969 to test samples of students in fourth, eighth, and twelfth grades.  We need little more than that. If we could move on from high-stakes testing, then our schools would be a lot more like schools in many other OECD nations, and the lives of U.S. teachers would be a lot more manageable.