Students who accumulated zeros on the one-hundred-point scale quit trying because they knew no matter how hard they worked, they would fail the class.
• Professional practices: Teachers changed from the one-hundred-point scale to a simple A–F grading system, where A = 4, B = 3, C = 2, D = 1, and F = 0. Teachers also switched from calculating the final semester grade based on the average of all work to giving a final grade based on the teacher’s judgment of student proficiency from the latest and best evidence of student learning.
• Results: D and F grades decreased 38 percent in social studies, 45 percent in English, and 62 percent in mathematics.
The consistent element of the science fair approach is that it uses local research with local students in local schools—your students in your school or district. Instead of considering a different sample of students from a different school, no matter how representative, the science fair approach allows teachers to compare the same students from the same neighborhoods with the same demographics with the same teachers within the same year. This compelling before-and-after approach allows teachers to conclude that all the other factors influencing student achievement are consistent—the only change is the teachers’ changes in professional practices.
Argument 2: “Anecdotes Are Not Evidence”
The assertion, “Anecdotes do not equal evidence,” is a very fair criticism, as many articles and books in education are best described as journey stories—the experience of a single teacher or administrator. However informative these experiences may be, they are anecdotes, not research. This is why we must all be critical consumers of educational research when a speaker or writer blithely claims, “Studies show …” or “Research says …,” when the studies or research may only be a sample size of one. This does not eliminate the value of case studies, but it is much more helpful to draw inferences when researchers accumulate a large number of cases. Certain educational researchers are leading the way in studying the successful practices of multiple school districts and condensing the results into a format educational leaders and teachers can implement in their schools. For example, Heather Zavadsky (2009), director of research and implementation at the Texas High School Project, analyzes the workings of several well-run school districts in her book Bringing School Reform to Scale: Five Award-Winning Urban Districts. Reflection on a single high-performing urban system is not nearly as helpful as Zavadsky’s (2009) synthesis of a variety of school systems and her ability to find common elements despite differences in geography, governance systems, funding, and student populations.
Using the science fair approach discussed previously, teachers may think their own work is “just an anecdote” and therefore not worthy of being shared with colleagues. But when that individual experience is grouped with dozens or hundreds of colleagues’ experiences, then patterns can emerge that no longer rely on anecdotal evidence. Moreover, a focus on teaching practices allows educators to distinguish between programs and practices. Vendors would like to claim a particular program, curriculum, or technology application leads to gains in student achievement, but programs alone accomplish nothing. The overwhelming conclusion of our review of more than two thousand school plans (Reeves, 2011a) is that it is practices, not programs, that hold the key for improvement in student results. In addition, a focus on practices turns the analytical lens where it belongs—on practices that are replicable rather than the mystical qualities of the individual teacher. Kim Marshall (personal communication, September 16, 2019), educational researcher and author of The Marshall Memo website (https://marshallmemo.com), thoughtfully distinguishes between teachers and teaching by noting that when the focus is on the teacher, great practice is relegated to the ethereal realm (“She’s just a gifted teacher!”) or calumny (“He’s just a terrible teacher!”). Neither of those observations is particularly insightful guidance for any professional beyond the superficial “Be good and don’t be bad.” But, as Marshall suggests, when we focus on teaching—the actual practices in which teachers engage—then we can address specific practices in discipline, feedback, curriculum, lesson planning, and professional responsibilities to perform effectively. The guidance is not, “Be more like Ms. Smith,” but rather like the following.
• “It’s important that you move around and are close to the students, not remain behind your desk.”
• “It’s important that you give students immediate feedback during class, not just on their papers several days after they did the work.”
• “It’s important that you engage every student by using whiteboards, cold calling, or similar techniques, and not merely recognize students who raise their hands.”
These concrete practices focus on what teachers actually do, not who they are as people. Moreover, these are practices coaches and administrators can model in real time, showing their colleagues in the classroom that no matter one’s responsibility in the school, they are all teachers and must be willing to show they still have the ability to demonstrate to colleagues and students the importance of effective teaching practices.
Argument 3: “The Research Depends on Heroic Teachers and Administrators Whose Efforts Are Unsustainable”
Movies about high-poverty-turned-high-performance schools often feature a particularly resourceful, motivated, or heroic teacher who enters the classroom and leaves masterful change in his or her wake. Moviegoers see this in films like Stand and Deliver (Musca & Menéndez, 1988) or Freedom Writers (DeVito, Shamberg, Sher, & LaGravenese, 2007). Although the storylines are compelling, the teachers in these moving tales do not represent the average teachers who, thrust into a high-poverty school often with inadequate preparation and support, are fighting to maintain a modicum of discipline, stay one page ahead of the students in the curriculum, and learn the craft of teaching.
I acknowledge there are exceptional teachers in successful high-poverty schools, but the equity and excellence research deliberately avoids these out-of-the-ordinary cases. The schools we learn the most from have the same teacher assignment policy, same union contract, same per-pupil funding, and in many cases, same school and classroom architecture as their less-successful counterparts. Great teachers make for compelling stories, but greater teaching is the only credible source of replicable and sustainable practice.
Argument 4: “Publishers’ Research Is Commercially Tainted”
The history of commercially tainted research in education is a long one. Just as purveyors of sugar- and processed-food-funded research purport to show their products are healthy—research that led to a multigenerational increase in obesity—so also do the sellers of video games attempt to show these are indispensable tools for student learning (Egenfeldt-Nielsen, 2006). This challenge is certainly not limited to the field of education. Although one would think that medical research is at the apex of credibility, the fact is, commercially funded research is more likely to be published than studies of higher quality and free of commercial bias (Hogan, Sellar, & Lingard, 2015; Lynch et al., 2007).
There is a significant burden on school administrators and other decision makers to be critical consumers. They must understand the nature of the research to influence purchasing decisions and the degree to which that research is applicable to the conditions of the schools and districts of the purchasers. One of the most frequently misunderstood terms is significance, which, in the context of research, almost always means statistical significance. In simple terms, the differences between two groups are considered statistically significant if an analysis of those differences shows they are unlikely (less than a 5 percent chance) to be different due to random variation. For example, if a group of students who participate in a particular instructional reading intervention score 79 percent on a test, and another group of students who did not participate in that reading intervention score 75 percent, then researchers can compare those two groups and, based on the number of students in the groups and the variation in their scores, determine that the difference between 79 percent and 75 percent is unlikely due to randomness. But that is a very different proposition than saying the reading program caused the students in the first group to score higher. Medical researchers sometimes