This article was originally published for Education Week (April 5, 2011). It is posted here with the permission of the author.
Written by Eric A. Hanushek
The teachers’ unions have put themselves in a difficult position, with Wisconsin, Indiana, and Ohio demonstrating that the traditional labor stance is untenable. So far, media attention to the union story has focused on the fiscal side—state deficits, teacher-benefit packages, and the like. Without question, these are important issues, but they are dwarfed by the implications for teacher effectiveness and improved student achievement. Now is the time to go beyond the rhetoric and to show that all of us—including the unions—are truly behind ensuring effective teachers in all classrooms.
Central to this issue are the conflicting strands of the unions’ position. Union leaders want to say that teachers are very important for public schools. But then again, not too important—because teachers should not be expected to make up for poverty and for uninvolved families. They want to highlight the exceptional teachers. But then again, not too much—because they do not want attention drawn to completely ineffective teachers. They want to argue that the pay is insufficient to attract and retain good teachers. But then again they do not want to draw attention to the salary schedule that refuses to acknowledge differences in teacher effectiveness.
Interestingly, recent research into teacher quality strongly reinforces the “buts” in the sentences above.
Studies examining data from a wide range of states and school districts have found extraordinarily consistent results about the importance of differences in teacher effectiveness. The research has focused on how much learning goes on in different classrooms. The results would not surprise any parent. The teacher matters a lot, and there are big differences among teachers.
What would surprise many parents is the magnitude of the impact of a good or a bad teacher. My analysis indicates that a year with a teacher in the top 15 percent for performance (based on student achievement) can move an average student from the middle of the distribution (the 50th percentile) to the 58th percentile or more. But that implies that a year with a teacher in the bottom 15 percent can push the same child below the 42nd percentile. With a teacher in the bottom 5 percent, a student may plummet from the middle of the distribution to the bottom third at the end of the school year. Obviously, a string of good teachers, or a string of bad teachers, can dramatically change the schooling path of a child.
This is not just a problem of inner-city schools or of minority students. The research adjusts for student backgrounds and for what each child knows at the beginning of the year. The results apply to suburban schools and rural schools, as well as schools serving our disadvantaged population.
Why doesn’t this issue reach the level of a national scandal? First, it is likely that ineffective teachers are generally hidden, in the sense that few kids get a string of bad teachers. Principals know very well who the ineffective teachers are, so they can balance a bad teacher one year with a good teacher the next. This implicit averaging process also means that it does not look like schools can do much to alter family background and what the child brings to school.
Second, parents do not quite know how to interpret results on achievement tests. The teachers’ unions have, since the advent of the No Child Left Behind Act in 2002, conducted a campaign to convince people that these scores do not really matter very much. Here they are flatly contradicted by the evidence.
People who know more as measured by standard math, reading, and science scores earn more throughout their lifetimes. It is indeed instructive to look at the difference implied by higher achievement. Somebody who graduates at the 85th percentile on the achievement distribution can be expected to earn 13 percent to 20 percent more than the average student. This applies every year throughout a person’s working life, yielding a difference in present value of earnings of $150,000 to $230,000 on average.
With this information, we also have the ingredients to calculate the economic impact of a good (or bad) teacher. By conservative estimates, the teacher in the top 15 percent of quality can, in one year, add more than $20,000 to a student’s lifetime earnings, my research found. And that impact holds for each student in the class. For a class of 20 students, we see that this very good teacher is adding some $400,000 in value to the economy each year.
There is also the other side. A teacher in the bottom 15 percent of the distribution is subtracting at least $400,000 from the economy each year.
These differences are clearly large enough that it is worth discussing policies that reward and retain the good teachers but do not keep the poor ones.
There is a national aspect to this issue. In its report “How the World’s Best-Performing School Systems Come Out on Top,” the consulting firm of McKinsey & Co., reviewing why students in many nations achieve more than those in the United States, pointed out that high-achieving nations do not let bad teachers stay in the classroom for long. The same analyses of teacher effectiveness cited above indicate that the United States could climb to the top of the international rankings if we followed the McKinsey advice: Replace the bottom 5 percent to 10 percent of teachers with average teachers.
The quality of the labor force, as measured by math and science achievement, has been shown to have dramatic effects on the growth of national income. In fact, moving to the top of the international distribution would mean a present value of some $100 trillion to the United States if past growth patterns hold in the future.
These numbers all seem sufficient to return to the “buts” in the opening. All parties, including the teachers’ unions, need to consider seriously how we can reward the vast majority of teachers who are doing an excellent job while not retaining that small portion that is doing serious harm.
The unions’ slowness in responding to reality has stung them with surprising quickness and force as newly strengthened governors have labeled the unions as the problem. If the unions want to continue, they must now act to get in step with the new reality. Specifically, they must get out in front on the teacher-quality issue.
Interestingly, Randi Weingarten, the president of the American Federation of Teachers, recently gave a major policy speech in which she addressed the issue of establishing an evaluation system that could identify what teachers need to do to improve and that could also form the basis of dismissal actions if an ineffective teacher did not improve. While the improvement process would take a year, the dismissal process would take less than 100 days—a noticeable improvement over today.
Even if all players agree to the structure, however, the central issue is what happens on the ground. Schools and unions have been saying “we need better evaluations” for decades, but there has been little pressure or incentive to produce a superior system, and less to use the system for personnel decisions. Not surprisingly, it generally has not happened. Perhaps with current pressure on the unions, or perhaps with the increased likelihood that newspapers will publish value-added scores in the absence of a comprehensive evaluation system, something will happen.
ERIC A. HANUSHEK is a senior fellow at Stanford University’s Hoover Institution and a member of the Koret Task Force on K-12 Education.