Finding the best teachers

You can’t tell good teachers from bad teachers (or good from bad quarterbacks) till you see them in action, writes Malcolm Gladwell in The New Yorker. So let lots of people try teaching at apprentice wages, but hire only the best, perhaps one out of four apprentices. Highly effective teachers are worth high wages.

. . . there is no point in raising standards if standards don’t track with what we care about. Teaching should be open to anyone with a pulse and a college degree — and teachers should be judged after they have started their jobs, not before.  

BaltoNorth contrasts Gladwell’s approach with McKinsey findings summarized in The Economist: High-performing school systems train only the best applicants. Salaries don’t have to be high if the profession’s prestige is high.

Singapore screens candidates with a fine mesh before teacher training and accepts only the number for which there are places. Once in, candidates are employed by the education ministry and more or less guaranteed a job. Finland also limits the supply of teacher-training places to demand. In both countries, teaching is a high-status profession (because it is fiercely competitive) and there are generous funds for each trainee teacher (because there are few of them).

 In response to Gladwell, Eduwonkette makes an excellent point: Teachers are good or bad in context. Like quarterbacks, teachers work as part of a team.

Even the most gifted quarterbacks end up with pretty crappy pass completion stats if their teammates consistently miss the ball. And a great quarterback doesn’t look so great if he’s a poor fit for the team he’s playing with. The same goes for teachers.

Teach for America, which combines rigorous screening with a two-year apprenticeship, combines both approaches. But many of the new teachers don’t stay in the classroom. I see The New Teachers Project, which works with TFA, is training effective teachers in Louisiana, which analyzes the value added by graduates of teacher education programs in the state.

Teachers, what do you think? Would you rather see apprenticeships for would-be teachers — and how would this avoid hurting the apprentices’ students? — or much higher requirements for teacher trainees?

About Joanne


  1. Teaching should be open to anyone with a pulse and a college degree — and teachers should be judged after they have started their jobs, not before.

    This indicates a very low opinion of the importance of good quality teaching. Would Malcolm Gladwell go to a new doctor if medicine was open to anyone with a pulse and a college degree, and doctors were only judged after they have started their jobs, not before?

    If a quarterback turns out to suck, okay, you’ve lost a couple of games. If an elementary school teacher turns out to suck, all their students have lost several months of education at least, making life much more difficult for their later teachers, and quite possibly harming those students for life.

    I agree with Eduwonkette’s point too.

  2. I wonder how Singapore and Finland are screening — it cannot simultaneously be true that, per Gladwell, we have no idea how to identify promising teachers in advance, and that, per Singapore and Finland, we do.

    Broadly speaking, I agree with Gladwell — I think Singapore and Finland are solving the problem not so much by knowing how to identify quality candidates as by ensuring the pool from which they are selecting is more generally competent, and I don’t see that happening in the US any time soon. However, I’d really rather see the apprenticeship take place as part of meaningful teacher education; rather than spending a few semesters in the classroom on theory and then throwing people in to student teaching full-time, spend time in the classroom on practice, and then have people teach part-time and surround them with opportunities for discussion and reflection, with both mentors and other new teachers (and bring in the theory here, if you like). That would both give you a sense of how good they are likely to be in the classroom *and* let them make effective use of their early teaching experiences.

    Gladwell’s system means that people with first-year teachers may be at greater risk of having a bad experience, but people with first-year teachers are likelier to have a poor experience in *any* system. If you couple that idea — I admit, a big if — with aggressive retention efforts, I don’t know that people *on average* end up with worse teaching. I don’t agree with Tracy that the system represents a low opinion of teaching — quite the opposite, if the good ones are actively retained; it’s just an admission that we haven’t got any better ideas, so let’s accept the risks of casting a wide net (more crappy first-year teachers) in favor of the benefits (more good experienced teachers).

  3. “Teaching should be open to anyone with a pulse and a college degree”

    Coupled with no way/no incentive to act on judgement about how good a teacher is, isn’t this about the system we have now?

  4. I tend to come down on the “wide net” side of this issue, but I’m not sure it’s a good dichotomy. I do agree that you can’t tell much about a teacher until you see him or her in action. But I would add that even when you do see him or her in action, you can still be very misled.

    I have come to the conclusion over the years that terms such as “good teacher” or “bad teacher” miss a very important point. A teacher is good or bad only in a given context. Perhaps this might be most obvious when thinking about teaching in inner cities versus small town or suburban schools. A teacher who does very well in one situation might do poorly in the other, and vice versa. But it’s not just these situational extremes that are relevant. I first became aware of this connection of teaching to context when our children were young. One particular sixth grade teacher seemed to be a poor teacher when our daughter had him, but two years later we felt he did very well with our son. Their differing personalities made quite a difference. So is he a good or a bad teacher? I think he’s both.

    So I don’t know if a wide net or selective screening is the better way to go. But I have long been convinced that a lack of understanding of teaching and learning is the real problem. Without an understanding of what’s going on we can’t do a good job of matching teachers to the situations in which they can do best.

  5. Teachers, what do you think? Would you rather see apprenticeships for would-be teachers — and how would this avoid hurting the apprentices’ students? — or much higher requirements for teacher trainees?

    Students, parents, real classroom teachers, and taxpayers would benefit from the repeal of teacher credential requirements, beyond a demonstration of subjet-matter competence: for prospective high school teachers, a degree in the subject or a related field or a sufficiently high score on the subject-area GRE, for prospective elementary school teachers, sufficiently high Verbal and Math SAT scores. How to avoid harm? Let apprentice teachers serve two years as paid teachers’ aides, department gofers, and in-house substitutes.

    Colleges of Education add nothing to teacher competence, and the College of Education degree requirement serves only to artificially create scarcity and elevate wages, to screen prospective teachers for politically correct views, and to guarantee employment for $80,000 per year Professors of Education.

  6. Lightly Seasoned says:

    I’d rather see tighter screening. As I observe the efficacy of teachers I work with, I see that the best teachers are consistently those who are very bright and succeeded in other areas before coming to teaching. I think this is perhaps because they developed an independent work ethic. They have an inner drive that keeps them striving to find what works best instead of relying on the lesson plans they wrote 7 years ago. The “gentleman C” types whom everyone likes (especially the kids) are usually not that effective.

    Once the screening is done, then put in apprenticeships (not as gophers and subs — good lord — but as real assistants who teach lessons, help with planning and grading, and work with the kids on academics).

    So, I want it all. I want both. And some common sense. And ed classes that address the real needs of teachers entering the field. Christmas is coming, isn’t it?

  7. It seems to me that the suggested lowly-paid apprenticeship is already present in the form of student teaching. I attended a rather selective grad program to earn my MAT that required a year of student teaching and that experience, more so than any of the other classes we took, determined our success in the program. Mentor teachers were expected to provide frequent feedback on our performance and our supervisors were in to observe every two weeks or so.
    Add to that the state-required year of mentoring by an experienced teacher after hiring (in NY) and there are already plenty of filters in place to eliminate substandard teachers.
    That being said, I’ve seen plenty of other schools (especially undergrad) that only ask for 10 weeks of time spent awake in the classroom to count towards a student teaching experience.

    The problem is not to design more hoops for teachers to jump through, but to actually utilize the existing systems. Too many teachers pass through certification programs without frequent evaluation and receieve tenure simply because they’ve participated in enough extracurricular activities even though they are substandard in the classroom.

  8. Good post. My experience has shown me that in matters of education there always seem to be an imbalance, either one has competent teachers and a low pay-scale, thus the teachers leave the district and/or sometimes the whole State, or a high tax base to pay the salaries of marginal or burnt-out teachers without a full return on the overall education of the children. Balance is key and finding one standard for all should be paramount. Just my 2 cents.

  9. From Gladwell’s article,

    “According to Hanushek, the U.S. could close that gap simply by replacing the bottom six per cent to ten per cent of public-school teachers with teachers of average quality.”

    So the current teacher selection processes are batting somewhere between .900 and .940? If so, I wouldn’t say that this average reflects processes that aren’t good at selecting teachers.

    A lot of companies follow Gladwell’s general approach and set a goal of firing some percent of employees every year. But let’s not forget the obvious, this firing happens every year! So the purpose is not to reach 100% accuracy of matching employees to jobs but to keep the percent of underperforming employees from growing. The percentage that companies use can go as high as 10%, but is more typically closer to 2%. Which would mean that schools are already in the ballpark, but might be able to improve. I’d like to know if there are enough people wanting to be teachers to bring the school percentages down to the 2% range? Maybe this can be addressed at the margin? Improve quality standards at schools of education and open up teaching to non-teacher trained college graduates with longer probation periods for these teachers.

  10. The countries that McKinsey cites select teaching school candidates from the top of their secondary school cohorts. If you look at any of the data about entering teachers SAT scores in the US, they are, as a group about average. If you look closer, and break out by verbal and math scores, you see that, fortunately, high school math teachers are above average in math and English teachers above average in verbal. But if you go down the line to elementary certified teachers, they are below the mean. Lowest of the group are teachers certified in phys ed and special ed. This has improved somewhat since the highly qualified requirements specify that teachers have to demonstrate content knowledge.

    Those are the facts.

    Here are the anecdotes–I get worried when I get notes from school with obvious grammatical errors (of the non-typographical type), or realize that the testing a hypothesis or understanding the applications of simple statistics, are beyond some teachers.

  11. Just as Margo/Mom said, I too become upset when I see schools and school districts sending out materials with errors. It is especially heinous when they are simple grammar errors that any teacher worth her salt would chastise a student for making.

    As for finding the best teachers, my question would be: best teacher for what? An elementary teacher is different from a secondary teacher, and in secondary schools, and English teacher will be different than the business teacher in a computer lab.

    The vice principal who did my evaluation this year had this insight, “I wish core teachers could see how elective teachers do it. They are so much more linear, and you are multitasking.” I think the question should be: do students learn from the teacher? If so, then that is a good teacher.

  12. There’s no assurance that the teachers who supervised apprentices would themselves be much good. There’s little assurance that the supervising principals would recognize good teaching or know much about it. There’s considerable certainty that many education professors know little about either teaching or about students.

    The new Dark Ages have begun and the barbarians are in charge of many of our institutions, especially our education institutions. The challenge right now may be to figure out how to pass real knowledge on to the few who are looking for it.

  13. Kudos to eduWonkette for making use of the football metaphor without allowing even the implication of the “C” word – competition – to creep into the post. Similarly, and neatly, excluded are the associated concepts of accountability and goal-orientation.

    Jinkies, what would football be like if no one counted, or even noticed, goals and the good players contributions to those goals weren’t noted?

    Oh yeah, it’d be very much like public education.

  14. Can someone tell me how accurately teacher seniority is defined? To the year? To the day? I’m thinking about teacher layoffs and guessing that seniority is the determining factor. But I’m wondering if administrators have much flexibility in choosing poorly performing teachers at layoff time.

  15. So, Andromeda, you’d go to see a new doctor if medicine operated by Malcolm Gladwell’s system?

  16. Just a week or two ago, there was an article on this site about a current research project about the brains of children and how they varied. So, how is teacher acceptance going to go in the light of this information?
    I had some students expelled for drug dealing. How does that rate me as a teacher?
    The student who, with rich daddy/mommy giving him/her everything doesn’t care a bit about his/her life and will not and willfully refuses to improve?
    What about the terrible educational research that get’s passed off in this country-how to we avoid the latest fads?
    I suggest that we think long, hard and deeply about teacher training as well as the future of students in public schools.
    Personally, I think we need to be more selective in all areas, not just one.

  17. Lightly Seasoned says:

    By the year usually — our contracts all start and end on the same day. So, two other teachers were hired the same year I was in my department, and I was in actuality the last hire, but for seniority purposes, we all started the same exact day.

    Administrators have far more flexibility in hiring and firing than people are led to believe.

  18. To improve the quality of teachers, one way to start is to improve the quality of the teaching programs, for one thing. There are too many “get your license quick” programs which claim to make it easier for already qualified folks to get out there and teach–but is that the market they’re truly playing to? Not in my experience. I see few people retiring from corporate management going to those programs, and see many more folks seeking to upgrade their current working income. Those quick-certification programs are usually less rigorous as far as demanding actual work product in the form of developing lesson plans, managing students, and developing course materials than the standard programs may be. YMMV, though, as you’ll get differing levels of rigor even within the different schools–I know that the general education program at my university was less challenging in many ways than my special ed training was.

    What I do know is that the student teachers I’ve subsequently seen going through the school I teach at differ in quality based on the programs they come from. The quick certification teachers not only don’t seem to be required to produce the same level of work sample (my sped program required 100 page plus work samples; many of these programs get by with 25 pages or less) but their student teaching experience and time is shorter and less thorough. They also seem to attract some real doozies for the supervising professors (one really bad, horrible, clueless prof last year provided a lot of fodder for the staff room sacred cow skewer patrol as we reassured the student teacher that said prof really was an idiot who didn’t have a clue about classroom management).

    Also, there’s just no darned way to get around the actual experience of the first few years of teaching. Even teachers who came into teaching from being EAs struggle with the difference between being an aide and being a teacher. If anything, I think it’s harder for a former aide to make that transition because they know what good teaching looks like, and they are harder on themselves when they can’t quite replicate it yet. Good teaching requires the ability to develop instincts which can be backed up by data. The instincts and abilities a good aide has don’t quite translate completely over because quite often they don’t have the training to develop or explore different teaching techniques. They can identify that something does or doesn’t work, but they can’t always identify why or pick up on the data that matches–a training and experience issue. They also struggle with the hidden side of teaching–the paperwork and accountability pieces they’ve not worked with before.

    My idea? It’s expensive, but I think that there should be a median, apprentice class between classroom aide and teacher. Sort of like a student teacher, but a paid apprenticeship higher paid than an aide but lower than a teacher. Perform as a team teacher with the cooperating teacher, plan instruction with the cooperating teacher, and report exclusively to the cooperating teacher and the school administration instead of the teacher training program. The apprentice teacher is an employee of the school district so is privy to everything that an employee is. Right now, student teachers are torn between what a teaching certification program requires them to do which may work at counter-purposes with what works for a particular building or district program. They have no real accountability to the building or the district, but to their licensing program. That can be good or bad.

    Benefit of this type of program? Improves student/teacher ratios, allows for greater differentiation of instruction within the classroom, allows for more effective data gathering and student monitoring. Also allows more time for the sort of instruction which improves student outcomes with quicker feedback–i.e., daily writing, daily math practices, science labs and write ups, better supervision of research processes…and so on. I think a year of this sort of program is likely to be a better preparation overall than simple student teaching, especially if the district is already committed to someone by hiring them as an apprentice–then the district and the principal is committed to making sure that this training works. Right now the only accountability in student teaching is to the training program–and that can be very political and ideology-driven. This also gives the apprentice a potential protection (as an employee) against a poor fit in training placement–I’ve seen good potential teachers wash out of training because of a witchy cooperating teacher.

    Biggest problem? Won’t be cheap, and not every teacher is amenable to sharing classroom control. Others won’t do a lot of training but just dump all their work on the apprentice.

  19. Tracy: “So, Andromeda, you’d go to see a new doctor if medicine operated by Malcolm Gladwell’s system?

    I would. Apprenticeship was the route to most professions and skilled trades until about 200 years ago. Today, in some States, one can become a lawyer through apprenticeship and the State bar exam.

    Professors of Education regularly extol real-world experience over classroom lectures. If they see the irony, they do not let on. The argument undermines the case for school generally and the case for Ed school in particular.

    More than anything else, the US education industry needs a large, competitive, for-profit sector to winnow out bad practice.

  20. Tracy: “So, Andromeda, you’d go to see a new doctor if medicine operated by Malcolm Gladwell’s system?”

    I would. Apprenticeship was the route to most professions and skilled trades until about 200 years ago.”

    Malcolm, I hope someone has your medical power of attorney right now.

  21. I believe the medical profession is already using both approaches, training and apprenticeship (internship). If your primary source of medical care is an inner-city hospital emergency room I believe your chances of being under the care of an apprentice is pretty high.

  22. Mike,

    What inspires the elevation of academic instruction over on the job training? Becker defines “school” as an institution dedicated principally to education. “Principally” clearly implies a continuous variable. One can imagine, say, teaching hospitals which accept 14-year-olds (the age at which Admiral Richard Howe and Captain Robert FitzRoy went to sea) and raise people through candy-striper, nurse’s aide, EMT, etc, and selects for further training those who will enhance the institution’s reputation.

    I suggest that many instructional structures which we currently accept originated in historical accidents, and are in no sense necessary.

  23. Malcolm, you seem to privilege apprenticeship over education; I would argue that education, coupled with apprenticeship, is better, and that apprenticeship works better and more efficiently if significant education comes first. Praxis, in other words, must accompany education, but praxis without education is incompetence. I would also argue that saying “Apprenticeship was the route to most professions and skilled trades until about 200 years ago” is irrelevant, unless you want also to argue that bleeding to let out bad humours is the height of current medical practice.

  24. Not (“schooling” = “education”). Apprenticeship is a form of education. Practical considerations usually motivate on the job training. Classroom instruction too often becomes an end in itself. Classroom instruction communicates some information more efficiently than on the job training. Consider, as a third way, self-paced instruction or independent study, in the fashion of some British universities. Students often do not attend lectures, but instead read transcripts of lectures, study on their own time, and discuss their material with tutors. Why couldn’t teacher apprenticeship work like this? Professional apprenticeship could incorporate the bookish part of instruction by this strategy. If that’s not “education” then the LSE abd some Oxbridge colleges aren’t educational institutions.

  25. I understand from my friends who did medicine that before new doctors are let loose on patients, they go through a lot of learning (eg dissecting dead bodies, giving each other injections, learning how to take medical histories) and are tested on that learning. The doctor in the emergency department may be an apprentice, but they’re not just anyone with a pulse and a college degree.

  26. Physics Teacher says:

    I think we need better training and apprenticeship.

    Regarding training:

    Look, for example, at music. One could argue that all good musicians have certain things in common, like keeping time. Yet you take guitar lessons from guitarists, bass lessons from bassists, and drum lessons from drummers. If musicians were trained the way teachers are we’d have students taking trombone lessons from guitarists. When the trombone student would ask “how do I get a sound of this thing?” the guitarist would reply “I dunno. You’ll just have to find a strategy that works. That’ll be $75 for today.”

    Training should be age and subject specific. We need to get rid of this ridiculous “generic” teaching.

    Apprenticeship should be at least a full year and the teaching student shouldn’t be paying the teacher’s college for nothing.

  27. I think we need better training and apprenticeship.

    What for? It’s not like there’s any professional distinction made between good teachers and bad teachers.

  28. Physics Teacher says:

    What for? It’s not like there’s any professional distinction made between good teachers and bad teachers

    Of course not. That’s because the current crop of supervisory personnel receive the same poor training brought on about by the same crazy philosophy.

    Imagine guitarists supervising trombone players, and vice-versa. Imagine drummers supervising opera singers. Imagine garage band singers supervising violinists playing in the philharmonic. All under the assumption that all good musicians have something in common and that this commonality is observable by anyone who’s ever picked up an instrument.


  1. […] Joanne Jacobs presents Finding the Best Teachers posted at Joanne Jacobs. […]