Why They Fight

Why does the education establishment so hate Teach For America? Turf and politics aside, one reason might also be the incredible amount of misinformation floating around. As 10K alumni descend on Washington to celebrate the organization’s 20th anniversary it’s a good time to look at some persistent myths – for instance that the research on Teach For America’s effectiveness is “mixed” or that the organization shows that just anyone can teach.  So that’s what this week’s School of Thought column at TIME is about:

In 1989, when Wendy Kopp proposed the idea in her senior thesis at Princeton of quickly training outstanding college graduates to teach in high-poverty schools for at least two years, her adviser told her she was “quite evidently deranged.” The comment has become legend since Kopp, unfazed, went on to launch Teach For America after she graduated, and on Saturday more than 10,000 of the nonprofit’s alumni will gather in Washington to celebrate its 20th anniversary.

Kopp ultimately earned an A on her thesis, but when it comes to learning from her organization’s experience, the education field deserves a big, fat F. Over the past two decades, Teach For America (TFA) has grown from a scrappy start-up to a national corps with an annual budget of $212 million and a staff of 1,400. Along the way, it has generated a great deal of research about how to improve the teacher training and selection strategies that are commonly used today. Yet the reaction from the education establishment remains one of intense hostility, which echoes through state capitals, Washington and even the courts, where lawsuits have been filed to curtail the use of TFA teachers. (See 11 education activists to watch during 2011.)

TFA’s 20th birthday seems like as good a time as any to unpack the misconceptions put forward by its critics (and by some of its proponents). Here are five of the most common:

Read the entire thing here.

Update: Question in the comment section below about TFA-producing schools.  The list cited in the comments below is based on a profile of 2009 TFA Corps members.  In the TIME column I was referring to cumulative data over the organization’s history.  Cumulatively, the top 10 TFA-producing schools are (in order): University of Michigan – Ann Arbor, University of California – Berkeley, University of Texas at Austin, University of Wisconsin – Madison, Cornell University, University of North Carolina at Chapel Hill, University of California – Los Angeles, Northwestern University, University of Illinois at Urbana-Champaign, and Harvard University.

17 thoughts on “Why They Fight

  1. Alex Trenton

    Why? For the same reason journalists hated bloggers for a long time: no one likes being embarrassed by someone who does it better and cheaper, thereby making their expensive efforts look superfluous.

  2. Roger M.

    The TIME article should be titled Five Loads of Baloney to boost TFA.

    Having had a close up look at TFA, I would have to say that there’s enough truth to these so-called myths to pay closer attention to the assertions; rather than these glib attempts to debunk them.

    TFA is still “for Ivy League (and elite school) dilettantes” how about some numbers; I noticed that only baloney number three, about retention, cited some specific percentages.

    Scholastic achievment is dependant on three groups: students; teachers; and parents; until all three are held equally accountable, any effort to make real changes work are “quite evidently deranged.”

    @Alex: bloggers do not do a better and cheaper job than actual journalists; with some exceptions, they put out assertions that are long on opinion and thin on research.

  3. phillipmarlowe

    Some of them turn out to be total duds. One recent example: when then-schools chancellor Michelle Rhee (herself a TFA alumna) told principals in Washington to get rid of low-performers as part of a budget reduction measure, …

    Hello History, get me a rewrite.

  4. Chris Smyr

    Roger, what numbers and data are you referring to when you label the article “baloney”?

    Phillip, the floor is yours to try and have an honest discussion, yet every time the option is made available to you, you pull off weird stunts like this. What are you vaguely referring to with your comment?

  5. Alan Vanneman

    “Again and again, the most rigorous studies show that TFA’s selection process and boot-camp training produce teachers who are as good, and sometimes better, than non-TFA teachers, including those who have been trained in traditional education schools and those who have been teaching for decades.”

    But I thought the idea is that they would be better than regular teachers, who are supposed to be terrible. If TFA simply turns out teachers who are indistinguishable from “regular teachers,” why do we need it? When Wendy Kopp founded TFA, was her idea to show that “outstanding” college graduates placed in high-poverty schools would produce the same results as existing teachers?

  6. Chris Smyr


    Some of the studies Andy cited *do* suggest they perform better than “regular” teachers. Check out TFA’s research page for more info on these:


    Since TFA places in schools and districts where the choice is often not between TFA and other well-qualified teachers, but between TFA and other new recruits or underperforming veterans, even equal performance to “regular teachers” is an improvement for these schools. Remember that the mission of TFA is *not* to put TFA teachers into every school, but to make sure that every classroom is led by a good teacher. If every school in the country had access to highly talented and high-performing teachers, TFA wouldn’t exist in the form it does today.

    Finally, the research findings that college graduates trained by TFA perform equally or better than those through more traditional avenues ought to say a lot about the supposed importance of the traditional model.

  7. Jack

    The problem is not the performance of TFA, which is indeed commendable, but rather TFA turnover. Namely, TFA corp members are often not surpassing their two year commitment. TFA is not a sustainable, systemic solution to a problem that is widespread in urban locales. TFA also damages the profession by implicitly suggesting that intellectual training concerning pedagogy and content knowledge is irrelevant to teaching, which is simply not true. Indeed, many traditional COE need to reform, while others are doing a wonderful job. The continual need to tie performance to test scores as validation is also misplaced. Examining student outcomes might be more useful, but that would require much more study. Furthermore, discovered which components of both university prep and TFA are tied to student achievement.

    Finally, the research cited on TFAs website is not necessarily definitive. Critiques from academics abound, concerning methodology and subject choice (see Cochran-Smith or Zeichner for a balanced assessment). Some of the myths likely have truth within them, and a more nuanced analysis is needed. Remember, TFA is an organization that wants to sustain itself…thus, they are only going to cite studies that support that end.

  8. Tiffany

    Andy, I’d like to hear your thoughts on how other alternative certification programs (TNTP especially) compare to TFA

  9. Chris Smyr


    1) We’ve discussed TFA retention rates before, as I will link you to below:



    Your claim that TFA teachers “are often not surpassing their two year commitment” is false.

    2) “TFA also damages the profession by implicitly suggesting that intellectual training concerning pedagogy and content knowledge is irrelevant to teaching”

    (Alright, I’m sorry, but my sarcasm levels kicked into overdrive after reading this plea for amnesty…)

    Oh, boohoo! The impact of TFA teachers rightfully brings up the question of whether or not these traditional routes that rely heavily on pedagogical theory ought to be considered as important as some folks presume them to be. If the TFA model is able to produce equal or better results compared to other more traditional routes to teaching, there’s no reason it should be vilified as you’ve done here. The teaching profession would do well to acquire and utilize the best practices possible to train their teachers, even if that means (*gasp*) revamping some of its more traditional ways.

    If you must make the pained analogy of this as an attack on the teacher profession, at least rightfully point out that it is an attack on the more stubborn, traditional ways of the profession– an ambush with data as the bullets.

    3) “The continual need to tie performance to test scores as validation is also misplaced. Examining student outcomes might be more useful”


    If you’d like to offer a better metric of performance that includes the positives of focusing on test scores (standardized results, objective analysis, easy to implement), please do. But realize that TFA does *NOT* only focus on test scores. TFA puts focus on many different forms of evidence of student achievement. Thing is, test scores are usually the only ones from which we can make any valid statements regarding the overall effectiveness of TFA teachers.

    4) “Finally, the research cited on TFAs website is not necessarily definitive.”

    The research cited there and discussed on this blog are not definitive because of some variant of “TFA said so!” They’re definitive because they’ve used adequate controls and solid experimental designs, something that is sorely lacking in other studies. As Andy says, this is not a debate that should revolve around quantity, but on quality, and there are some high-quality studies that suggest positive results from TFA.

  10. dick

    I’m weary of crossing you chris but something is off in this statement:

    “Since TFA places in schools and districts where the choice is often not between TFA and other well-qualified teachers, but between TFA and other new recruits or underperforming veterans, even equal performance to “regular teachers” is an improvement for these schools. Remember that the mission of TFA is *not* to put TFA teachers into every school, but to make sure that every classroom is led by a good teacher.”

    The most rigorous studies compare TFA teachers to a similar cohort of teachers that work or would have worked in their school. So when they find TFA teachers are as good or better than others, it means that they are as good or better than the “new recruits or underperforming veterans” whose spot they took. This is still an accomplishment for the other reasons you cite above, but it doesn’t mean filling a school with an “as good” TFA-er will necessarily improve teacher quality in the school.

  11. Chris Smyr


    I guess it is important that one defines what the set of “regular teachers” includes, and specifically whether TFA teachers score equally or better than this group. For example:

    1) My reasoning was that even equal performance to new recruits would be better than the folks that schools *aren’t* hiring as a result of TFA. TFA teachers are generally compared to the effects seen of teachers who get the jobs within these schools. I can’t imagine these schools are hiring, or favor hiring, *every* warm body that comes along and turns in a resume. In this case, equal performance to those that are hired and have student achievement data to analyze would be more desirable than the poor performance a school might see from another ignored applicant.

    For example, Mathematica found that there was a sizable group of novice teachers in their control group that had zero student teaching experience, which is evidence for what I’m arguing above.

    2) When carrying out these comparisons, many results show that TFA is producing *better* results than the control groups of other new hires, which more obviously suggests that TFA is helping improve teacher quality at the school.

  12. Jack

    I think you are missing my point slightly. I’m not arguing that TFA is wholly a bad thing. In fact, it clearly has produced some positive results. However, it has problematic elements, just as traditional programs do. To suggest otherwise, is to merely be a cheerleader…but that is not usually a problem the current reform movement recognizes. My problem with the current reform movement, beyond TFA, is that it is driven more by ideology than evidence…as are blogs like this one. Charter schools, pay-for-performance, even school choice have all showed considerably mixed, and often, negative results. Eventually, holding the company line becomes irrational. This countervailing evidence, however, is brushed aside rather brusquely, at least in political and policy circles. Should someone, god forbid, suggest that more systemic problems underlie educational inequity (poverty, institutional racism, etc.), they are branded as a whiner and favorable to the status quo. This sort of binary thinking is predicated in your response and most of the entries on this blog. Systemic solutions would be far more radical, but they don’t fit the preconceived logic of the current reform movement.

    Let me briefly respond:
    1. They are not being retained as teachers at high rates…given that teachers have the most impact on student learning, this is problematic. Sure, they might stay in administration or get into policy work…but they are not having a direct effect.

    2. If one wants to improve the candidates going into education as a career, it would behoove one to boost the status of the profession, not weaken it. Few high status professions (lawyers, doctors, engineers) would ever accept such little training. If one wants to systemically improve the candidate pool, then one should make education attractive. Unfortunately, not all choose careers based on intrinsic motivations or social concern. Status often matters.

    3. The debate about test scores being valid continues; it is not closed, as you seem to assume. Furthermore, the focus on test scores drains education of its creative capacities and crafts a purpose that is entirely aligned with economics, rather than civic or democratic concerns. Scores tell a small portion of the story, and may actually be used to create more inequity. Qualitative data matters, no matter what Arne Duncan spouts.

    4. The data is mixed. The Mathmatica report compared TFA members to under-credentialed, woefully under-prepared candidates. This data was then used, as so common in this current debate, as a way to prop up TFA. Other reports have been positive, I don’t dispute that.

    Ultimately, TFA is not a panacea. It might do some good, but it has problematic aspects as well, as does much of the current reform game. But no one on these blogs or in other arenas would dare suggest such a notion…yet, they are quick to blame traditional programs, all of which are certainly not bad. In the end, I have a feeling much of the current reform will have little systemic impact, as all of it fails to deal with underlying problems. But, I’ll probably be labeled as supporting the status quo for saying as much, which only shows the poverty of thinking saturating the reform movement.

  13. Chris Smyr


    I’m not “cheerleading” for TFA. I am addressing the inaccuracies of the critiques that you have raised. Please clarify which point of yours I am missing.

    1) The research that we have suggests similar retention rates to other entering teachers in similar performing schools. I can link these to you if you’d like. There are numbers in the links I’ve already provided.

    2) You are begging the question of how TFA is “weakening” the profession. You are also begging the question of why professional development for teachers should be similar to that of other professional positions such as in science and law. There are huge differences there, and I can speak to the science side of the debate more anecdotally, but basically it is untrue that to be a great educator requires intensive study of pedagogy and years of preparation. TFA is already showing that that understanding is faulty, and logically it ought to make sense as well. Why?

    Education degrees and more traditional approaches to teacher training may provide useful theory, but it is the on-going practice of those theories that makes a better teacher. Learning how to teach requires the *students* to test your ability to manage a classroom and direct instruction, not a university teacher. Just by this nature of this profession, an education degree/traditional credential means a completely different thing than other specialized degrees, like a PhD would mean to a scientist or an MD would mean to a doctor. Both of those latter degrees are directly related to successful theoretical AND practical work those professions do, which is not the case in education. It’s likely only possible to learn how to teach when we’re surrounded by 20-30 kids in an authentic classroom who are naturally assessing our abilities every single minute of class. The students themselves are a tougher assessment process than any other test or term paper imaginable.

    3) You are not really giving any support for your claims here. For example, if there are other valid ways we can measure achievement equally well as tests, what metric do you then propose assessing students with that would give valid, standardized conclusions about different teachers with different students in different schools?

    How does the existence of testing “drain the creative capacities” of education?

    Why exactly does the goal that a student should know X, Y, and Z content standards dictate how that material is taught to them?

    How is this only “aligned with economics, rather than civic or democratic concerns”?

    4) You claim the data is mixed, yet you don’t offer a supporting example of a well-written paper with a strong experimental design. Instead, you here critique the Mathematica report, but I’d like some evidence for that claim. The report has this to say about their controls:

    “For our analysis, we defined “control teachers” to include any teacher in the study who was not a TFA corps member either at the time of the study or at any time in the past. “Control teachers” therefore included traditionally certified, alternatively certified, and uncertified teachers—any active teacher who came from any source other than TFA.


    We conducted two types of comparisons of TFA and control teachers. First, we compared classes taught by TFA teachers with classes taught by all control teachers, which could include both novice and veteran teachers. In this case, the average years of teaching experience was far higher for the control teachers than for the TFA teachers. To control for differences in teaching experience, we conducted a second type of comparison based on classes taught by novice TFA teachers with novice control teachers. We defined “novice teachers” as including teachers in one of their first three years of teaching during the study year.”

    5) I don’t recall ever hearing someone say that TFA is a panacea. Also,

    “But no one on these blogs or in other arenas would dare suggest such a notion…yet, they are quick to blame traditional programs, all of which are certainly not bad”

    Well, lots of people have suggested that TFA is harming children, and other variants of the sort. What I like about this blog is it gives a outlet for those people to actually reason out why they think that way, and then we can discuss it further, as we’re doing right now.

  14. Jack

    Ok, final response…I doubt we will agree here, and I am spending too much time on this blog, when I should be working:) So, I’ll let you have the last word.

    I think we have very different conceptions about what it means to be a teacher; perhaps this is a philosophical difference. I, for one, believe they need both rigorous pedagogical and theoretical training as well as sustained practicums interacting with children. Teaching is an intellectual job, one that requires an enormous amount of knowledge and reflection in order to be successful.

    In terms of testing, I’m not proposing another metric…it will continue, I have no doubt about that. But qualitative studies have shown (see Darling-Hammond…via EBSCOHOST) that teachers do often “teach” to the test, which does, in some ways, limit options.

    The retention rates are indeed the same as other teachers who teach in urban classrooms…this is still very problematic, for both sectors (traditional and TFA)…and also somewhat understandable given the stressors of teaching in said urban classrooms.

    Here is a quote from the Mathmatica study (p. 14).

    “Most of the novice control teachers did not have substantial teaching-related training.
    Table IV.1 shows that only one-third of the novice control teachers possessed a bachelor’s
    degree in education, and none possessed a master’s degree in education at the time of the survey.
    Prior to entering teaching, only 31 percent had spent 10 or more weeks student teaching, and 53
    percent had no student teaching experience at all. Less than 38 percent of the novice control
    teachers reported having a regular or initial teacher certification, and more than 50 percent
    reported having a temporary or emergency certification. Only one of the novice non-TFA
    teachers attended a college classified as “most competitive,” “highly competitive,” or “very
    competitive,” according to Barron’s Profile of American Colleges.”

    So, at least in this study, the control group was sort of the worst of the worst…

    Anyways, the debate will continue. I think TFA does some good things, and I think it is problematic in some ways as well. I’m off!

  15. Chris Smyr


    But if you stop replying, I’ll also have to do work myself!

    I agree with you that teaching requires knowledge and reflection (TFA is also big on reflection and continuing improvement, as well), but I disagree as to how extensive the knowledge must be to be successful, and think that additional coursework is not the most useful “knowledge” for being successful in the classroom. If the goal is to find the best ways of training/hiring more good teachers, we’d absolutely want to test these different hypotheses. Considering that the research suggests TFA training is as good as, or better than, other routes to teaching (if student achievement is the goal, and I hope we can agree that it is), it’s a shame that it keeps getting labeled by some as hurtful to the profession.

    I would also agree that qualitative evidence would be beneficial to assessing students and providing further evidence to help with teacher evaluations, for example. I don’t see a practical way to incorporate it *instead* of testing, though.

    Also, what you’re saying here, that teachers tend to teach to the test, is something that needs to be elaborated on. If teachers are focused on teaching toward defined content standards, this can rightfully be seen as limiting their options. The alternative, however, would be not having the standards, and this would make it almost impossible to have any sort of standardized accountability for students or teachers. I think of these limits as absolutely necessary in that case.

    Furthermore, your tone on TFA retention rates here has greatly changed from your initial comment. I do agree with you that retention of teachers in urban classrooms is problematic, though I don’t blame TFA for it.

    Lastly, in reply to the Mathematica quote, those are the demographics of the novice control teachers, but they do *not* indicate those teachers were hand-selected for their lack of experience. It was a small sample size, but I don’t know if the numbers would actually be that much better given a larger sample size… This is also notwithstanding the fact that the comparisons made between TFA and the entire control group of teachers (veterans included) also showed equal or better performance.

  16. Jack

    Cheers to a good debate… I am not one of those who completely disparages TFA…in fact, I do think it is useful…and I did go the traditional route via teacher training, and am now in a graduate program within a “traditional” college of education…I will say, not everyone dislikes TFA…in fact, our dean is a former alum:)

    In any case, great conversations, and I appreciate the civil debate…this is rare.

Leave a Reply

Your email address will not be published. Required fields are marked *

× one = 8

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>