With the nomination of Betsy DeVos to the position of Secretary of Education for our nation, I am again reminded of the vitriolic partisan debate surrounding education reform. One side – the side to which DeVos subscribes – believes that our school system is thoroughly broken, standardized tests are good metrics to measure students and schools, teachers’ unions block any meaningful reform that policy makers try to implement, and that free-markets (in the form of school choice through extensive charter options and voucher programs) in education is the only way to fix the system.[i] This is appealing to many because it advocates for inexpensive solutions, where the markets monitor themselves and weed out poorly performing schools because parents don’t ultimately choose those for their students; it also identifies an enemy in the form of teachers’ unions, a force to unite against. At face value, business people also often find this side sensible because it resembles the way that consumers determine the success of companies and products, and this makes sense given their experience.
Meanwhile, the other side maintains that our public schools aren’t actually that bad, that standardized tests are biased and highly flawed measures of educational success and shouldn’t be used, that teachers unions are essential in supporting students’ and teachers’ rights in a system that otherwise seeks to strip these, and that the answer is to better fund schools, lower class sizes, and better support teachers (in knowledge, resources, and pay). The solutions suggested by this side are more expensive, as they require more funding to support schools, teachers, and students. Nonetheless, this argument is generally more appealing to teachers or others who have worked in schools, who have seen a lot of wonderful things happen in public schools and know that teachers work tirelessly with very little support for very little pay.
I understand why a person would latch on to either argument. There is a lot of misinformation out there and people often operate on an incomplete picture of the situation – especially if it fits into their worldview – because they don’t have the time to better inform themselves about all the complex facets of schooling in this country. So I want to offer my somewhat informed opinion for those who have not had the same luxury I have had to extensively research and ruminate on education.
The first thing that any informed voter or consumer of education should know is that teachers are the most influential school-level factor in student achievement and advancement.[ii] Basically, this means that the individual teacher in a classroom has way more influence over a student’s progress than the school, the district, or the broader educational policies that affect all of this. As such, any reform that supports a teacher to grow and improve, and retains already good teachers, is a reform that can make an appreciable difference.
Some people respond to this fact by supporting policies that seek to increase teacher evaluation to weed out the ineffective teachers. At one point, I admittedly fell into this trap. I had just been RIF’d[iii] for the second time in my four years teaching in LAUSD and I was mad (a fact that drove me to leave the classroom and pursue doctoral studies). So I thought, why can’t we just determine these layoffs by teacher quality instead of seniority? But as I began to study education and learn more about the system, I realized my folly. Few districts have installed an even remotely reliable system of teacher evaluation, and LAUSD was certainly not one of them. And those who claim that teachers can be judged by test scores alone – through a system like Value Added Models (VAMs), which attempt to account for a students’ growth over a year – fail to understand the statistical and/or practical constraints of this approach. These include: 1) many teachers teach subjects that are not tested, 2) standardized tests often don’t adequately measure student learning, let alone critical thinking, test for arbitrary skills (I was an English teacher who had to look up “participle clause” to correctly answer a question in one test I proctored), and include “Idiotic, hair-splitting questions pertaining to nothing“, 3) “teaching to the test” can create gains that are not actually meaningful, 4) the effect of student demographics are not fully accounted for in VAMs[iv], and 5) statistically speaking, VAMS cannot be used to determine any causal judgments upon teachers because students are not randomly assigned to their classrooms. Low-income schools that serve a disproportionately high percentage of English Language Learners and students with disabilities, where class sizes are often higher, teachers are more overworked, and resources are fewer, will often have lower test scores and lower growth on these exams. Basically, there is no way to reasonably evaluate a teacher based on a students’ test score.[v] We cannot just slap a performance evaluation onto the back-end of a teacher’s year and expect this to change teacher quality. Instead, we must find ways to help the teaching force improve before and after they enter the profession.
Any reform that does not seek to improve teachers – the key input to student’s education – will have little impact on student progress. As such, market based models of education reform are flawed from the get-go. This is why you see dismal results in DeVos’ charter school experiments in Michigan. Similarly, the most recent longitudinal large-scale study of charter schools found that on average, charter schools do not improve test scores, and in fact attending a charter school seems to have a negative impact on a graduate’s wages. The only type of charter school that did advance test scores – no excuses charter schools, which are largely geared around standardized tests – did not improve students’ life outcomes, making the increase in test scores a specious advancement.[vi] Studies of voucher programs produce comparably lackluster outcomes. And in fact, the expansion of charter schools and voucher programs drains money from public schools (because money follows the individual students) and often results in fewer resources and support for teachers and students that need it most. Finally, students are human beings, and as noted above, they cannot be quantifiably measured like products; as a result, parents don’t often receive enough meaningful information about the quality of a school (especially if it is a relatively new school and does not have many years of longitudinal data on students’ life outcomes) to determine its potential efficacy for their child. “Free markets” – in which there is often very little regulation (e.g. enforcement of policies protecting students with disabilities) – is simply not the way forward if we want to improve the education of ALL students in this country. It is not the way forward if we want a better future.
In another blog, I will go into more depth regarding some of the things we can do to improve our system. Education is not some black box; we actually do have a sense of what works and what does not. Briefly, we need to recruit more excellent candidates into teaching, which means the profession has to be more desirable; we need to train teachers better before they enter the classroom, which means teacher education has to become more rigorous and thoughtful across the board; and we need to provide teachers with more and better support and professional development once they start teaching, because teaching is hard and it’s easy to burn out.[vii] All of this will take considerable time and investment, which is likely why it’s not the most politically popular position. But when it comes to our future, it’s worth it.
[i] Chubb & Moe (who are both scholars of politics, not education) champion this argument in their 1990 book Politics, Markets, and America’s Schools, which has become quite and influential text for education reformers.
[ii] Sanders & Rivers, 1994. This finding has also been reproduced and verified by a number of other scholars and is now commonly accepted as truth.
[iii] Reduction in Force layoff, given in reverse seniority order in accordance with many union rules, or in CA, state bylaws. These are disseminated irrespective of teacher quality.
[iv] A student’s socioeconomic status has more impact on his/her test score than any other factor. And slew of complicated variables that come along with this are hard to measure in a statistical formula.
[v] Meira Levinson and I tackle this issue in our paper No Justice, No Teachers (2015). Other scholars that exam this include Koretz (2009) and Braun (2005).
[vi] Dobbie & Fryer, 2016. http://www.nber.org/papers/w22502