Experience as a teacher turns out to be worthless

In this study by the Urban Institute, punks freshly graduated from college and paid about $30,000 per year were found to be more effective as schoolteachers than experienced union members who draw $80,000 of taxpayer blood for their 9 months of work each year.

Can it be that experience is worthless? I have taught classes from a pre-existing syllabus using a proven textbook. The very first time that I did the class, I worked a bit harder than I would have in subsequent classes, but probably managed to be 85 percent as effective as I could possibly have been. I have taught classes where I was one of the textbook authors and a designer of the syllabus. In this case, experience teaching was hugely valuable and resulted in massive changes to the syllabus, course structure, and textbook (fortunately online in HTML format and thus easily changed).

If public school teachers don’t write or choose their textbooks, it might not make any sense to pay experienced teachers more than beginners. http://www.cpsd.us/Web/HR/2008CTA_UnitA_Salary.pdf shows the salaries for the Cambridge Public Schools, some of the least effective in Massachusetts as measured by student achievement tests. A 22-year-old with a bachelor’s earns $41,000 for 9 months. A teacher with 11 years in the system would earn $69,000. That 11-year veteran who picked up online master’s and Ph.D. degrees in education would earn $81,500 (presumably there is no evidence to support the theory that a degree from University of Phoenix makes someone a better teacher).

If we wanted to spend the same amount of money every year, maybe it would be smarter simply to pay all teachers the same salary, e.g., $60,000 for 9 months of work. We would thereby attract a more able class of young college graduates and if after 7 years they got sick of the job they wouldn’t be tempted to stay because their salary was due to go up so dramatically.


  1. Colin Summers

    May 17, 2008 @ 10:40 am


    That’s a brilliant idea.

    I have had recent experience with this effect. My older son had a fifth grade teacher who was on the verge of retiring each year for the past five but apparently her pension would get a little larger each year she stayed. But she was wildly burnt out and nearly useless.

    My younger son currently has a fourth grade teacher who is a ball of energy but who is studying for a degree in family therapy. In a year or so she’ll move on from teaching to a profession where the salary can climb into the six figures.

  2. Michael R. Bernstein

    May 17, 2008 @ 5:05 pm


    Interesting proposal, but strictly from an economic standpoint, I’d include an additional bonus for the best teachers (assuming you can find a metric that can’t be easily gamed), ie. some kind of pay-for-performance.

    Something like… within a school (once you control for the previous year’s grades), teachers whose students have the highest grades (across the board, not just in their class) get sliding scale bonuses.

    Another way of putting it is that the teachers whose students, on average, improve most, are rewarded.

    Also, if you’re anticipating an average 7-year career for teachers, you should track the student’s subsequent post-secondary academic achievement and add on pro-rated bonuses for for that as well.

  3. jjrs

    May 17, 2008 @ 8:12 pm


    One possibility is that the people being hired now are just generally more competent than the hires of yesteryear.

    Another idea- why not allow talented teachers to write and choose their own textbooks? Mandatory texts and curriculum might be okay for people new to the job, but after years of work, teachers just become bored and jaded by the system. If they were given more control over how to teach, maybe the profession would retain better people. The job would be a lot more work, but motivated people would actually like that and find it worth focusing on. And the deadbeats that just want to turn to the next page, phone it in and go home would be encouraged to get out.

  4. Erik

    May 17, 2008 @ 10:49 pm


    Philip, My name is linked to a private democratically-run school (The Sudbury Valley School) in nearby Framingham, MA (maybe you’re familiar with it?) that my kids will attend (one starting next fall) when they turn 4 and thru “high school”. It’s per student cost is ~$7000 (or less I suppose as there are discounts for siblings) because there is essentially no bureaucracy. Staff (teachers) do have yearly salary increases, but there is no tenure — the students elect them each spring. (Well, staff have votes too of course, but they are extremely out-numbered!) So it’s up to the students to decide who is relevant/useful and who is not. As it should be.

    The school opened in 1968 and have been correct that tuition has DECREASED relative to comparable public school costs in the area over the years despite the rising average staff salaries.

    At university I had wonderful TA-led and awful professor-led courses. AND vice versa.

  5. yischon

    May 17, 2008 @ 11:36 pm


    I looked through the published study.

    The biggest limitation of the study is that the only measure of “effectiveness” is “as measured by student exam performance,” specifically the End-of-Course exams for Math, English, and Science courses in the state of North Carolina.

    Also of note is that the “punk” teachers are high-scoring, high achieving graduates from “some of the most selective colleges and universities across the country”. This is how the Teach for America (TFA) program works.

    Although the score improvement is “statistically significant”, it seems practically insignificant. On average the students of the TFA teachers scored 0.07 standard deviations better than the non-TFA teachers.

    So the study shows that the students of a high achieving group of new graduates from selective colleges got slightly (but “statistically significant”) higher scores on the End-of-Course exams in North Carolina than the students of the other teachers in North Carolina.

    Is this study even practically significant?

  6. jjrs

    May 18, 2008 @ 12:45 am


    Here’s another idea- perhaps the new crop was just better because the best teachers in the young crop have yet to quit.

    It could be that the only people willing to stay on and put up with the system are the least talented of the bunch. And in that case, you’d be better off trying to retain the talent. Maybe its better to fix the system rather than tailor the payscale to match the existing conditions.

  7. Former Grad Student

    May 18, 2008 @ 3:04 pm


    Michael R. Bernstein wrote on May 17, 2008 @ 5:05 pm

    “Something like… within a school (once you control for the previous year’s grades), teachers whose students have the highest grades (across the board, not just in their class) get sliding scale bonuses.”

    Sure! Great Idea! Then we’ll just have more teachers pushing students through college classes with grades they don’t deserve. That’s a huge problem these days, and I don’t want any part of it. If a students fails, he fails. If I know he’s trying, I’ll try to help him, but I won’t just pass him.

    The problem is that most teachers don’t have ambition for what they teach, and it shows through lame presentation skills and terrible information delivery overall. You can’t assess a teacher’s performance based on student grades because as we all know the student is responsible for 95% of the class. It doesn’t matter how stellar you are as a professor if the student doesn’t try.

  8. Bob

    May 19, 2008 @ 5:37 pm


    > Another idea- why not allow talented teachers to write and choose their own
    > textbooks? Mandatory texts and curriculum might be okay for people new to the
    > job[….]

    Sometimes a decent textbook is the only ameliorating factor in a class taught badly by a crappy teacher–in my experience as a student, anyway.

    Then again, I’ve seen teachers who at the time I felt were not so good, who more or less adequately redeemed themselves through picking good reading materials. 🙂

    Maybe we need better textbook selection? Feynman wrote some pretty amusing (and sad) stuff based on his experiences when he tried to get involved in a state’s selection process for standard textbooks for science classes.

Log in