The Teach Like a Champion Paradigm

At the middle school where I currently serve as Math Instructional Coach, part of our professional development consists of reading Doug Lemov’s Teach Like a Champion.  This book provides a list of “49 techniques that [ostensibly] put students on the path to college” and is a Teach For America (TFA) favorite.  I have nearly finished the book and many of the techniques are great – I believe it can provide considerable value to teachers.  The book has a few major flaws worth discussing, however.  Lemov makes two assumptions that are representative of what I believe to be some of the biggest problems with TFA and the broader education reform movement.

First, Lemov assumes “the highest levels of student performance” are the goal of education and measures performance with standardized test scores.  He briefly acknowledges that “state test results…are not sufficient” but then buries that comment in a bunch of prose defending the value of what test results measure.  As one of my best former teachers argues in a recent article, reformers far too often concentrate on academic concerns “to the complete exclusion of the characteristics that make a difference in kids’ lives.”

The vast majority of TFA staff and corps members I have encountered care deeply about their students, and I imagine Lemov does as well.  But I’d like to see educators focus at least as much time and energy discussing how to inspire kids to be better people as Lemov does explaining how “a walk to the bathroom is a perfect time for a vocabulary review.”  The singular focus on academic achievement and test scores in the education reform debate has also contributed to the nationwide decline in art and physical education classes since No Child Left Behind.

Second, and much more dangerously, Lemov uses his charter network’s test scores to assert: “Great teaching, [our] teachers have proven, is strong enough to close the achievement gap.”  As evidence for this claim, Lemov cites the scatterplot below, which documents the scores of every public school in New York state on a 2009 English Language Arts assessment:

FRPL stands for free and reduced-price lunch, an indicator of student poverty levels. The circled data point is Rochester Prep, a school from Lemov’s charter network, Uncommon Schools.

Lemov’s declaration – that this chart and others like it demonstrate that teaching effectiveness closes the achievement gap – is the fuel for much of the national conversation blaming the achievement gap on ineffective teachers and schools.  Reformers argue that dismantling teachers unions, linking teacher evaluation to student test score data, shutting down “failing” public schools in favor of charters, and a whole host of other changes would drastically improve educational outcomes in this country.  TFA and books like Teach Like a Champion, both of which suggest educator skill causes the results in the graph shown above, empower these reformers.  Sadly, not only is a causal link between teacher effectiveness and student performance data unestablished and the data itself unreliable, but data presented on charter school success is also incredibly misleading.

In-school factors have never been shown to explain more than one-third of the achievement gap and recent research suggests income inequality is the predominant cause of the achievement gap.  Additionally, outliers are expected in any dataset, Uncommon Schools is one of several that fall well above the trendline in the scatterplot above, and it’s possible (though I believe it unlikely) that the school just got lucky.  But let’s leave those considerations aside for a second and ask whether the student populations at Uncommon Schools are truly representative of the populations the schools draw from.  Lemov states,

Our students are selected at random from the districts where we work, have a higher poverty rate than the districts from which we draw, and, contrary to myth, are often the least, rather than the best, prepared students in those districts (one of the major reasons parents exercise school choice is that their students are struggling and increasingly at risk in their original schools; they are moving from as much as to).

His claim that his students are selected at random is just plain untrue and contradicted by his description of school choice at the end of the sentence.  While students who apply for admission are admitted via a random lottery, any introductory statistics student can tell you that a random sample from a group of volunteers in an area is most definitely not a random sample from the entire area.  Lemov provides no evidence to support his claim that his students are less prepared than their counterparts who don’t apply because there is no evidence to support this assertion; in fact, families who have the knowledge and commitment to seek new schools for their students are more likely to find success in any environment.  Lemov’s statement, if not willfully deceptive, reflects a poor understanding of statistics.

Gary Rubinstein, a 1991 TFA corps member and seasoned educator, has extensively studied data presented by “miracle” schools, schools like those in the Uncommon Schools network that achieve seemingly unbelievable results.  Rubinstein has found that in most cases those results are unbelievable because they are, well, not worthy of being believed.  Whether charters lose students by expelling them, by not-so-subtly suggesting to their families that kids won’t be successful at the school (a practice a friend of mine who worked at a KIPP school called “counseling out”), or through some other method (perhaps unintentional), the student attrition rates at these charter schools are very high and the percentages presented in most cases don’t account for this fact.  When a “miracle” school’s change in the student body is taken into account, the school’s results are hardly noteworthy.  These schools often help students, to be sure, but the students they help are generally small, unrepresentative subsets of students in poverty.

Schools in the Uncommon Schools network are quite possibly awesome schools for the students they serve, and the teachers Lemov observed and wrote about are most likely excellent teachers.  Their results, however, despite Lemov’s and reformers’ claims, do not indicate that great teaching can level the playing field for children in poverty.  Teachers can and should work hard to change students’ lives and there are undoubtedly cases where we succeed.  But when reformers advance changes based on misleading data and/or false claims from people like Lemov, our students suffer and the goal of educational equity becomes harder to realize.