What Do Teachers “Produce”?

by Guest Blogger
April 12th, 2011

by Diana Senechal

In a recent head-scratcher of an article in Education Next, economist Eric A. Hanushek puts forth the argument that effective teachers produce higher salaries in their students.

The logic? Well, according to labor data, students whose high school test performance is one standard deviation above average (that is, students at the 84th percentile) can expect to earn 10 to 15 percent more per year than the student of average achievement. Hanushek assumes, apparently, that this high performance was the result of large gains over the years (as measured by test scores). We’ll get to that in a moment.

Now, according to Hanushek, if we consider that a “high-performing” teacher (at or above the 84th percentile) produces achievement gains of 0.2 standard deviations above those of students with an “average” teacher, and if one takes into account attenuation over time, one finds that such teachers will boost their students’ collective earnings by hundreds of thousands of dollars. The figure offered for a teacher in the 84th percentile with a class of 20 students is $400,000 per year; even a teacher in the 60th percentile will raise students’ earnings by $106,000.

With all due respect to Hanushek, I find that his argument oscillates between the silly and the scary. The silly part is this: there is no evidence (as far as I know) that students in the highest percentiles in high school are those who made the greatest gains on their standardized tests over the years. In fact, I suspect that most of them did pretty well on those tests all along. The top level on many of these tests is not very high; once you reach a certain level of proficiency, your gains don’t show. Unless it can be demonstrated that these top-percentile students did indeed have the greatest gains—and that their teachers had the highest value-added scores—the argument flops.

Also, there’s no reason to assume that “high-performing” teachers—those whose students make the greatest gains—bring their students to the 84th percentile or higher. It is quite possible that the larger gains occur at lower levels. For many reasons, I suspect that they tend to cluster around the average—but whether or not that is the case, there is no indication that they continue in linear fashion up to the top.

As for the scary part, let us take the argument to its logical conclusion. Suppose teachers could “produce” higher salaries in students, and suppose the “highest-performing” teachers produced the highest salaries, on average. Wow—then you’d have a cadre of test score virtuosi churning out lawyers, CEOs, social network inventors, surgeons, and change readiness consultants by the thousands. Now, some people enjoy those professions, but not all do.

Who, then, “produces” the foresters, violinists, English professors, marine biologists, simultaneous translators, teachers, firefighters, museum guides, electricians, editors, and cabinet makers? Does this fall to the not-quite-so-high-performing teachers? If so, maybe the ultimate “effectiveness” is not entirely desirable. This does not mean, of course, that anyone should settle for so-so teaching and learning. Yet we cannot assume, across the board, that more or higher equals better.

Now, most people want a good salary, up to a certain threshold. Very few want to live in poverty, to depend on others, or to be left without choices. But beyond that threshold, many may choose a profession or job that doesn’t pay spectacularly but is otherwise rewarding. Many want to keep their job low-key so that they can do things outside of work.

I realize that that isn’t quite the point—that we are talking about the difference between those who reach a certain level of achievement in school and those who don’t—and the consequences of such a gap. But even there, many ambiguities remain. Although high achievers tend to earn higher salaries, not all do or wish to do so, nor do lower achievers (within a certain range) necessarily end up lost and impoverished.

What do teachers “produce”? If there is free will, they produce nothing. They teach, inspire, and encourage their students; they demand the best of their students; and they point to many possibilities, through the subject matter and their own examples. They help students reach a point where they can support themselves and do something they enjoy. But it is the student who takes off and does it—often making choices that confound the teachers and parents. That is how it should be. Otherwise, for all our fanfare over the Future, we would be trapped in an eternal Industrial Age, with teachers turning out remote-control dolls.

Diana Senechal’s book, Republic of Noise: The Loss of Solitude in Schools and Culture, will be published by Rowman & Littlefield Education in November 2011.


  1. Bravo, Diana. Brilliant. And revelatory: This is what economists assume about the purpose of schooling in America: it’s accessing credentialing and scores for personal financial gain. Winning the, umm, race to the top.

    What’s missing? Plenty. Economic benefits to society. Citizenship. Communal quality of life.

    Will share this piece widely. Well done.

    Comment by Nancy Flanagan — April 12, 2011 @ 11:46 am

  2. Totally agreed that Hanushek is silly.
    But I am not sure we have to swing in the entire other direction, saying that students have free will and teachers are just inspiring them and pointing the way.
    Students are affected by genes, parents, culture, schools, and teachers.
    Hanushek’s model of “achievement” is impoverished, to be sure, but I am more in favor of showing how his vision is limited, using his own methods (like Rothstein, etc) than saying that students make choices that confound teachers and parents. In general, I am not so sure about that. Some students may confound their parents, but mostly, kids who grow up in an environment rich in capital (cultural and financial) become professionals or otherwise well remunerated themselves. Kids who don’t, generally don’t.
    We should reject Hanushek, but not his assertion that kids are incredibly affected by their environments. His problem is a definition of the environment so narrow as to be laughably silly.
    Finally, I’ll agree that it is terrible that somehow the most cited expert on the nature of learning is a labor economist. It seems like those who actually study learning don’t have the same sort of breezy confidence that leads to such nice pithy conclusions.

    Comment by Cedar Riener — April 12, 2011 @ 1:35 pm

  3. Thank you, Nancy!

    And thank you, Cedar, for your good point. I do not mean to swing in the opposite direction–of course children are incredibly affected by their environments. But, as you point out, Hanushek’s definition of environment is far too narrow.

    One could argue that Hanushek is wrong to consider salary alone–that one must also consider the type of work; its compatibility with one’s own interests, abilities, and priorities; its demands and rewards; and many other factors. But it is impossible to assign a numerical value to these things. And that’s why it isn’t sufficient to argue from within his framework.

    One can say, with some confidence, that better education increases a person’s options (with a few caveats). But it does not follow that everyone will select the option with the highest pay. That is where free will comes into play–or some combination of free will, environment, and personality.

    Comment by Diana Senechal — April 12, 2011 @ 2:21 pm

  4. Terrific piece! This was exactly what bugged me about Hanushek’s piece when I first read it, but I couldn’t quite put my finger on it. I like your addition in comments above, too — that education increases options (generally). But some outcomes and some realities simply cannot be quantified.

    Comment by KT — April 12, 2011 @ 7:32 pm

  5. Diana, I’ve responded directly in the comment section of Hanushek’s article. It is mind boggling how he assumes that our schools would be just as good as those in Canada and Finland ‘if only’ teachers did a better job.

    I’ve pasted in for him a link to the Common Core paper, “Why we’re behind – what top nations teach their students but we don’t”


    Talking about good teachers, today I came from my parent conference with my 2nd grader’s teacher. He’s the best I picture for this profession. Everything he said was perfectly timed, perfectly applied – about the progress my daughter is making in math, in science, reading and writing. About her social life, her friends. About the way she learns.

    She does not like to read much. Should we insist with her that she should read more? No, we should not force her, because she already reads difficult material, a year above grade level. And if we do force things, she might resist.

    She turned in an assignment the other day about such and such famous character. I think he got the idea to read about that character from one of her class mates, he said. I pointed out that she also had a book at home about the subject. “Ah”, he said, “that explains why she write this other assignment in this particular way”! At what grade, I ask him, are students expected to stop learning the mechanics of reading, and start relying more on knowledge-of-facts strategies? About now, he answered, in 2nd grade.

    My 2nd grader is very outgoing socially. Will that interfere with her learning, I asked? No, he says, it is normal for her to be at that. We should not compare her with her older sister, who is more of an introvert.

    At one point he said: “Your daughter is right where she should be: she’s learning a great deal socially from discussing with her friends.”

    Now I have my own opinion about the uses and misuses of social learning. I think I developed a 6th sense for some of these things. But my teacher had it completely wrapped, without catching a breath: “It’s true”, he said with a chuckle, “that at the beginning of the year there was a lot of wheel spinning in the discussions among kids, but now they are genuinely learning from each other.”

    It was 1/2 hour where I could not wait to hear him to tell more. You may have seen me quite often critical about schools, here or on other blogs. But when I look at him, I think perhaps we are not doing all that bad.

    Comment by andrei radulescu-banu — April 12, 2011 @ 8:02 pm

  6. It amazes me that a highly regarded economist would have so little understanding of basic economic principles and the nature of percentile. First of all it should surprise no one that the top 84th percentile would tend to have the highest paying jobs. Did you think that the bottom quartile would be the highest paid? The problem is that he fails to understand how percentile works. 84th percentile means that these students scored better than 84 out of 100 of their peers. If however all the students taking the test moved to the raw score of the first group, this would not change the percentile. At the 50th percentile half the group scored above that score and half the group scored below that score. If the entire group raises their score by 100 or 1000 points it does not change the percentile. Half are still above a certain score and half are below it.

    Hanushek also shows a lack of understanding of supply and demand. If everyone is qualified to be a doctor, lawyer, or CEO and no one wants to be a janitor or garbage collector then the excess supply of the former decreases their value while the shortage of the latter increases their value. This is why the argument that everyone should go to college because college graduates earn a larger income lacks logic. If everyone has a college degree then the extra value because you have something that only a minority of people possess, ceases to exist. Instead we have college educated janitors and store clerks. I will admit that there is nothing wrong with highly educated janitors and store clerks. Education has many intrinsic rewards in and of itself. I’m simply speaking from a purely economic standpoint.

    I will save for others the impact that an individual classroom teacher can have in all this. I am only speaking to the economic fallacy of Hanushek’s arguments. To put exact dollar figures in future earnings tied to the raising of test scores ignores the basic principles of economics.

    Comment by Mary S. — April 13, 2011 @ 1:31 pm

  7. [...] Diana Senechal on what teachers ‘produce’: “What do teachers ‘produce’? If there is free will, they produce nothing. They teach, inspire, and encourage their students; they demand the best of their students; and they point to many possibilities, through the subject matter and their own examples. They help students reach a point where they can support themselves and do something they enjoy. But it is the student who takes off and does it—often making choices that confound the teachers and parents. That is how it should be. Otherwise, for all our fanfare over the Future, we would be trapped in an eternal Industrial Age, with teachers turning out remote-control dolls. ” (The Core Knowledge Blog) [...]

    Pingback by Quick Hits (4.13.2011) — April 13, 2011 @ 4:02 pm

  8. [...] What Do Teachers “Produce”? is by Diana Senechal and appeared in the Core Knowledge Blog. [...]

    Pingback by The Best Posts & Articles Explaining Why Schools Should Not Be Run Like Businesses | Larry Ferlazzo's Websites of the Day... — April 13, 2011 @ 11:05 pm

  9. So to “justify” my existence as a teacher or prove my worth,should I ask all of my previous students how much money they are currently making? If these past students of mine are making more money than me, have I done a good job? If they are employed or are in college have I done a good job? If they have graduated from college and are now in a good paying career have I done a good job? Is it only on these standardized tests that I am measured? When the parents aren’t parenting I get blamed for not doing enough, but when a student is resilient and graduates despite their “parents” have I done a good job or is it only to the student’s credit? Do I get any credit for the good things or only for the bad?

    Comment by Tracy Galligan — April 14, 2011 @ 1:09 am

  10. [...] What Do Teachers “Produce”? is by Diana Senechal and appeared in the Core Knowledge Blog. I’m adding it to The Best Posts & Articles Explaining Why Schools Should Not Be Run Like Businesses. [...]

    Pingback by Today’s Collection Of Good School Reform Posts & Articles | Larry Ferlazzo's Websites of the Day... — April 14, 2011 @ 1:46 am

  11. [...] Diana Senechal ponders the question, “What do teachers produce?” (Core Knowledge Blog) [...]

    Pingback by Remainders: Walcott leaps first waiver hurdle in unanimous vote | GothamSchools — April 14, 2011 @ 5:22 am

  12. I would object to Hanushek’s argument on two grounds: (1) High-performing students don’t necessarily pursue the most lucrative jobs. While it’s unlikely that a straight-A student would end up working at McDonald’s (or not working at all), it’s perfectly plausible that the student may choose to pursue a nursing degree or become a forest ranger (rather than work as an investment banker), simply based on his or her interests. (2) I disagree that teachers should be given all the credit for student success (or failure): Studies show that BY LARGE MARGINS the most important factors influencing scholastic performance are those outside the schoolroom: Namely, SES, parental education levels, cultural attitudes toward education, and similar factors. The teachers who have high-performing students are very likely to be simply those teachers who work in environments where you find academic high-achievers, such as schools in upper middle class suburbs.

    Comment by Attorney DC — April 14, 2011 @ 9:38 am

  13. Can you clarify? I.e., your last paragraph takes issue with the word “produce.”

    Let me ask it differently.

    If Teacher A does your list better than Teacher B — inspire, encourage, point to possibilities, explain stuff, relate stuff — do you think it’s likely that, while we cannot predict any single student, their students as a cohort will do better on reading and math tests?

    I’m trying to reconcile your argument with Robert’s usual argument in this space. His is that effective teaching (how Willingham would define it) does, in fact, lead to test score increases (but often teachers aren’t effective for various reasons, especially bad curriculum).

    Comment by MG — April 14, 2011 @ 12:03 pm

  14. I’ll let Diana answer for herself, but what I find ruefully ironic is how we decry the “widget effect” in teachers yet Hanushek’s logic assumes a widget effect among students–they are all exactly the same but for the efforts of teachers on their behalf. Surely, this is no more true of students than teachers. As the New Teacher Project website explains, “Our school systems treat all teachers as interchangeable parts, not professionals. Excellence goes unrecognized and poor performance goes unaddressed. This indifference to performance disrespects teachers and gambles with students’ lives.” Change “teachers” to “students” and we have a new phenomenon: The Hanushek Effect?

    Comment by Robert Pondiscio — April 14, 2011 @ 1:43 pm

  15. MG,

    Yes, on the whole, if there is a good curriculum, good instruction, and other favorable conditions, one expects test scores to go up, provided the tests measure what students have actually learned. But that’s because (a) the students are able to do better under such conditions and (b) they generally want to do so.

    But just how much the scores go up depends on many factors. And the test is still in the student’s hands. While most students probably want to do as well as they can, you have some who give up in the middle, some who try to get it over with quickly, some who get terrible test anxiety, some who don’t think they understand it when they actually do, some who throw themselves into it, and some who have a knack for tests.

    And then, when the students go out into the world, even more is in their hands. They have to make choices that will affect how they lead their daily lives. There is no reason to assume that all will go for the big salaries or that all should. Nor is there any reason to believe that a string of “effective” teachers can catapult large numbers of students into top percentiles, which in turn catapult them into top earnings.

    Comment by Diana Senechal — April 14, 2011 @ 4:13 pm

  16. Diana’s terrific article and the lucid comments cover the topic very well, but let me add a small anecdote in support of Diana’s statement, “The top level on many of these tests is not very high; once you reach a certain level of proficiency, your gains don’t show”:

    I had a set of 11th grade students who took the AP English Language & Composition exam and the PA state standardized Reading test in Spring of the same year, and compared their scores. The correlation between AP score and the standardized test score was negative. In brief: the kids who earned AP 5s had a lower standardized Reading test score than those who earned AP 4s. (They are all great kids — you should hire them if they come to see you — but I admit, the sample size was small.)

    Seemed interesting to me! …as does Diana’s upcoming book. Sign me up for a copy, Diana!

    Comment by Carl Rosin — April 15, 2011 @ 4:47 pm

  17. Thanks for a thought-provoking article, Diana. I think at the end of the day, the answer to your question is simple. Teachers, produce choice and opportunity – the ability to decide to be a firefighter, a forensic scientist or a financial analyst. Those with a limited education have an equally limited portfolio of career options from which to choose.

    Comment by Chauncey Nartey — April 17, 2011 @ 10:33 am

  18. [...] which I reply: Hear, hear. If economic gain is the measure of our success, we have lost sight our goals in [...]

    Pingback by What is the Value in a High Value-Added Teacher? « The Core Knowledge Blog — March 5, 2013 @ 11:51 pm

RSS feed for comments on this post. TrackBack URL

Leave a comment

While the Core Knowledge Foundation wants to hear from readers of this blog, it reserves the right to not post comments online and to edit them for content and appropriateness.