The Wrong Conversation About Teacher Education

This week, Mathematica Policy Research released another experimental design study on the effectiveness of Teach for America (TFA).  (A brief summary is available here.)  Ostensibly a program evaluation linked to the $50 million i3 grant TFA received from the U.S. Department of Education to scale up its model, the study randomly assigned more than 2,000 pre-K to fifth grade students across 13 districts to TFA corps members and non-TFA teachers and then compared students’ performance at the end of the 2012-2013 school year.  The headline findings from the study were (1) that students of TFA corps members (who had an average of 1.7 years of experience) performed equally well as students assigned to non-TFA teachers (who had an average of 13.6 years of experience) and (2) that TFA corps members teaching pre-K to second grade students had a positive and statistically significant effect on students’ reading achievement (equivalent to 1.3 additional months of learning), relative to their non-TFA colleagues.

These are important findings and ones that ought to prompt a good conversation among researchers and policymakers interested in improving teacher education and overall teacher quality.  For example, are there teaching techniques that are being used by the newly trained and intensively supported TFA corps members that could also be employed by the rest of their colleagues?  In a blog post about the study, co-CEOs Matt Kramer and Elisa Villanueva Beard alluded to the intensive supports and mentorship available to corps members seeking to improve their practice.  Related to this, maybe there are cohort effects experienced by TFA corps members that could inform the way that colleagueship and school communities are nurtured.  Or maybe TFA corps members spend inordinately more time preparing lessons that could lead us to redesign the school day to allow more prep and collaboration among teachers.  In short, there are many possible explanations for the effects seen in the Mathematica study, any one of which could have potentially far-reaching and innovative implications on the teaching profession. 

But these are not the kinds of nuanced conversations we are likely to have.  Instead, I fear that the conversations will be in response to provocative headlines like this one:

Technically, as noted above and when accompanied by all the usual research caveats, there is truth in this statement.  But without a good grasp on research methodology, someone reading this tweet could easily infer some pretty sweeping claims about the relative effectiveness of first- and second-year corps members compared to experienced career teachers. 

I certainly have known my share of decidedly average (even some below average) career teachers, but I also know many teachers who have been teaching over a decade and who far surpass their novice colleagues in effectiveness.  Then again, there are exceptions to every rule, and my anecdotal experience proves nothing and counts very little in the grand scheme of things.  That is why I am not going to take issue with the methodology of this study.  I am way too much of a novice researcher to do that competently. 

Besides which, to me the methodology – and the caveats that accompany any methodology – are a distraction from my more fundamental objection to this study and the way it will be perceived by the sector at large.  My objection is that the perceived implications from this study far overshadow its modest intentions and thus distort the very real conversation we ought to be having about teacher preparation and teacher quality.

TFA enlisted Mathematica to conduct a program evaluation, but the public conversation we are likely to have about this study is not going to be about the program.  Rather, we are going to be drawn into making defiant proclamations about (or feeble defenses of) the woeful state of teacher quality, as evidenced by these 14-year veterans who cannot outperform new college grads.  The universe of teachers who are not TFA corps members or alumni is enormous and enormously diverse, so a blanket statement that claims first- and second-year teachers who have been recruited, trained, and supported by TFA are at least as good and maybe better than any alternative is pretty bold and frankly shocking. 

But that’s the headline, so that’s where the conversation will start.  After reading that headline, we will either be left with great admiration for TFA or despair about the legions of teachers who cannot manage to do better or possibly both.  To be clear:  a conversation about the state of the teaching profession and how we can improve it is a conversation that I – and many scholars, practitioners, and policymakers, in the U.S. and abroad – desperately want to have.  I only wish that we had more of this conversation before this latest TFA study was released.  Standing in shadow of the study, a conversation about the teaching profession is constrained, because we are left with the false impression that the best alternative to what we have now – a rather schizophrenic mosaic of university-based and alternative certification programs – is Teach For America.  After all, a rigorous “gold standard” experimental design study found that TFA corps members outperformed the rest of the field, so why shouldn’t they be the best alternative? 

The reason is because the ills that plague the teaching profession are more far reaching and more complicated than TFA alone can address.  Indeed, when asked by the New York Times this week how we could improve teacher quality, not one of five leading scholars and practitioners mentioned alternative certification programs like TFA.  Teacher preparation and teacher quality are functions of complex interactions between many stakeholders.  To overhaul preparation and improve quality, we should instead be looking to systems that have done just that. 

I'll end by reiterating that I do think there are important things we can learn from TFA.  They have managed to make teaching a much sought after job for college graduates.  They are committed to recruiting a diverse teaching corps and have succeeded at doing so in ways that many teacher prep programs have not.  They are changing in ways that encourage and support corps members who want to make teaching their career.  In many ways, TFA is a learning organization that could be a model for schools and districts.  Such a reciprocal exchange, however, will only be possible if studies like the one released this week do not (even mistakenly or unintentionally) reinforce the impression that there are two distinct and competing types of teachers out there:  TFA and non-TFA, effective and not effective. 

To be a profession worthy of the distinction, teachers must see themselves – their successes and their failures – in all of their colleagues, and they must work together to improve their individual practices and their collective standing.  Now, let’s have the conversation I would most like to have: the one about how teaching becomes just such a profession