Level 2: Personal Experiences
The first level, personal beliefs, is based on sincerely held views that are sometimes supported by examples. Some teachers, administrators, and parents—and indeed some students—believe that grading as punishment is effective, even though a century of evidence undermines the veracity of that belief. The second level, personal experience, extends beyond belief statements and relies instead on what many educators regard as the most compelling evidence of all—their individual encounters with students. Beliefs do not spring from a void; they are often based on one’s own learning and experiences. Many people who did well in school attend college and become educators or administrators. They formulate their beliefs about grading and discipline policies based on a system that is clearly effective for them. “It worked for me!” the opponents of grading and discipline reform will say. They are, of course, correct. Grading systems practices that are, according to the evidence, damaging (such as grading as punishment, the use of zeros on the hundred-point scale, and the use of the average; Reeves, 2011a) do, in fact, motivate some students to higher levels of performance. In my conversations with educators and administrators, I have asked, “For whom were these grading practices effective?” The most frequent answers are, “It worked for me!” and “It worked for my own children!” This is not surprising, coming from the viewpoint of college-educated professionals who are now public school educators. They loved school. They were good at school. They figured out the grading and discipline systems and responded well to rewards and punishments based on grades. The question is, What percentage of today’s students do we expect to become public school educators? If the answer is less than 100 percent, then perhaps we should reconsider attributing to all students the same motivational scheme—grades as rewards and punishments—that were effective for many teachers. Personal anecdotal evidence may be powerful when forming our belief systems, but we must be careful about generalizing our personal experiences to all students. It is preferable to adopt an attitude of “There is always something to learn,” and, indeed, I have seen teachers who, even in the final months of their careers, continue to improve their practices, learn from research, and change their techniques in order to better serve the cause of student learning.
Level 3: Collective Experience
Irving L. Janis (1982) coined the term groupthink to describe the tendency of people to acquiesce to the group and submerge their own better judgment. Sometimes the consequences are trivial, such as in psychological experiments in which subjects are shown two lines, clearly of different lengths, but other observers (confederates of the researchers) claim the lines are of equal length. If there is only one other person in the room, the research subjects hold their own, maintaining the lines are clearly different lengths. But as more people in the room claim the lines are of equal length, the research subjects cave to the inaccurate observations of the group (Heath & Heath, 2013). The consequences are less benign when the conversation participants are debating issues of greater importance.
Although an individual teacher may seek to change his or her practice—perhaps by calling on students randomly or by simply rearranging desks—professional isolation can be terribly lonely. Group experience provides the illusion of certainty, of unanimity, of proof: “It’s not just me who believes this—it’s the entire mathematics department! It’s the entire third-grade team! Our faculty voted, and we are unanimous.” Thus, popularity supplants reason. It is critical to remember, however, that even multiple instances of personal experiences or beliefs are still just personal experiences or beliefs. They remain anecdotal, and as such, are fallible in the absence of more credible research.
Level 4: Systematic Comparisons
Systematic comparisons produce some of the most effective and persuasive research that can directly influence professional practices. Consider the example of a teacher who makes an alteration to her teaching practice in the second semester. She can subsequently make three comparisons to form a very strong basis for research conclusions. The first is a comparison of the same group of students before and after the intervention. For example, the teacher can compare student performance in the first semester, when there was no change in professional practices, with student performance in the second semester, when the teacher altered a particular aspect of her practice. Perhaps it was a change in when students do homework—from at-home practice to in-school practice. Perhaps it was a change in the way the students revise and respond to teacher feedback. Perhaps it was the use of student self-assessment before submitting work to the teacher. The beauty of a systematic comparison is all other variables are held constant; the only change is the change in teacher professional practices. The students are the same, as are the schedule, curriculum, assessment, and teacher. If there is a change in results, it is very likely due to the change in teacher practice. In the second case, the teacher might make year-to-year comparisons, comparing the attendance, behavior, or academic performance in the first semester of one year to the first semester of the previous year. In these situations, the students are different, but the curriculum, assessments, time allocation, schedule, and teacher are the same. Finally, in the third case, the teacher might compare her results with the results of colleagues with similar students who did not change professional practices. Such studies allow for minimal variables and, hence, enable the teacher to draw powerful, generalizable conclusions about the effects of the professional practice she changed.
Every school and system, no matter how small or large, can conduct systematic comparisons like these. Teachers do not require a federal research grant, university evaluators, or any special expertise. Reframing Teacher Leadership to Improve Your School (Reeves, 2008b) provides a number of systematic comparisons examples at the elementary, middle, and high school levels.
Level 5: Preponderance of the Evidence
The apex of the research-type levels is when different researchers, operating independently using different research methods and working with subjects in different parts of the world, come to strikingly similar conclusions. For example, professionl learning facilitator Jenni Donohoo’s (2017) synthesis of research finds that teacher efficacy is strongly related to gains in student achievement. John Hattie and colleagues also cite collective teacher efficacy as a powerful variable related to student results (Waack, n.d.). My own quantitative analysis of more than two thousand U.S. and Canadian schools places teacher efficacy at the top of the influences of student achievement over the course of three years (Reeves, 2011b). Qualitative researchers who engage in deep observations and case studies have come to remarkably similar conclusions (Hargreaves & Fullan, 2012). When focusing on level 5 research, we move away from the claims of dueling experts. Some of the reasons teachers may be cynical about education research are because “You can always find an expert to say anything,” and “Today’s claim may be discredited tomorrow,” so they don’t know what to believe. But if several different researchers using different methods in different places substantiate a claim, we are no longer looking at anecdotes or isolated claims, but rather at the preponderance of the evidence. Such research provides compelling evidence for why educators should try implementing a particular change in their schools.
The original 90/90/90 research was a hybrid of what this chapter describes as Level 3: Collective Experience and Level 4: Systematic Comparisons research. But in the second decade of the 21st century, the research on success in high-poverty schools is firmly rooted in Level 5: Preponderance of the Evidence. Different researchers using different methods operating independently have come to very similar conclusions about the elements of success in these schools and, most importantly, the replicability of those findings.
Summary
Advocates of improved teaching and leadership practices are caught in a quandary. When they propound good ideas without evidence, they are guilty of vacuous rhetoric. But even when they have evidence from a variety of sources and methods, they meet the wall of opposition labeled, “But it doesn’t apply to me.” The same argument could perhaps be made against any variety of medical interventions, as the research on pharmaceutical and surgical interventions is all performed on someone aside from the patient for