Elements of Grading. Douglas Reeves

common ways to evaluate students. We can compare their performance to other students, or we can compare their performance to an objective standard. The first method, comparing students to one another, makes sense when one must allocate a scarce resource, such as admission to highly selective colleges. However, this method makes no sense when the stakes are really high, such as landing an airliner or performing brain surgery. In these examples, it is not enough to merely be better than one’s peers. Rather, it is essential that the pilot and the surgeon meet a standard—our second way to evaluate students. The claim that an airline doesn’t crash as frequently as other airlines or that a hospital has fewer deaths than the hospital across town provides scant reassurance to passengers and patients. What the public cares about is not who beat whom but rather the degree of proficiency pilots and surgeons possess.

      The essentials of standards are the same for students as for pilots and surgeons. When we compare students to one another, we have the worst of all possible worlds. This system regards students who have mastered standards as inadequate if it has assessed another student in the class as better. Worse yet, it labels students who have failed to master a standard as excellent, as long as they are superior to their peers.

      Is the comparison of student performance to pilots and surgeons overwrought? Not if you consider the impact of student failure on health and social well-being. As Alliance for Excellent Education (2010) research indicates, students who fail in school have dramatically higher medical care costs, lower incomes, and disproportionate use of the criminal justice system. Failures in the cockpit and surgical suite not only affect the people directly involved but all of society. Similarly, failures in the classroom affect all of us.

      Standards are a delusion if students are not evaluated on their proficiency. Despite the public embrace of standards since the 1990s, standards-based student evaluation has been slow to develop. While researching this book, I interviewed some of the most prominent international leaders in education reform. Most claim that although there are occasional exceptions, very few examples exist in classrooms or schools in which an entire education system backs up its rhetoric of standards with the reality of standards-based student evaluation.

What Makes Standards-Based Grading Different

      Imagine that you are teaching your teenage son to drive. You have selected a deserted parking lot on a Sunday morning where very little could go wrong. You relinquish the driver’s seat and make sure that your seat belts are so tight that they almost constrict your breathing. As your teen driver lurches forward, you notice that most of the parking lot lampposts bear the scars of car paint. It appears that you are not the first person who thought that the parking lot would be a good training ground for new drivers.

      There are two possible questions that you can consider. The first is, “Will my kid make fewer dents in the lampposts (and the family car) than the last dozen teenage drivers who were here?” That sort of thinking sets the bar dangerously low. The second question is, “What does my teenager need to do in order to become a safe and proficient driver?” This is what Jay McTighe and Grant Wiggins (2013) might call an essential question. Essential questions are those that not only require merely a binary response but also a deep inquiry into the thought processes behind the question.

      What do teenage drivers and academic standards have in common? Consider the case of Princeton University. This is home to a lot of very smart people, including Nobel Prize–winning faculty members and a student body that includes some of the most talented and intelligent young people the world has to offer. These students already know how to earn great marks in school. The reason they came to Princeton was not to accumulate more As on their transcripts but to engage in life-changing intellectual challenges. But even very smart people can make critical mistakes. Persuaded that Princetonians were getting by too easily, the university decided to limit the number of A grades to 35 percent of a class (Ad Hoc Committee to Review Policies Regarding Assessment and Grading, 2014). The idea that 65 percent of Princeton students were incapable of earning an A is mind-boggling. Nevertheless, this example is at the heart of what makes standards-based grading different. Grading based on the bell curve assumes that some people meet standards, very few people exceed the standards, and half of people fail to meet the standards. Standards-based grading, by contrast, allows any student—indeed, the vast majority of students—to meet academic standards as long they demonstrate proficiency in those standards.

      The mirror image of the Princeton example is the case of students who receive honor roll grades but are unable to meet basic literacy requirements (Reeves, 2006a). These students are quiet and never speak out of turn or challenge a teacher and receive good grades as a reward, while teachers punish their more boisterous peers for challenging and engaging behaviors.

      The fundamental characteristic of standards-based grading is that students are evaluated based on an objective measure of performance—don’t hit the lamppost; don’t plagiarize your senior thesis; and provide work that is, after much trial and error, a significant intellectual achievement when compared to objective criteria. The alternative is a stark contrast—it’s OK to hit the lamppost, but don’t hit it as frequently as the last student driving in this parking lot. It’s OK to plagiarize, as long as you don’t do it as much as the last student who got caught. Finally, your intellectual contributions don’t need to be thoughtful and creative, as long as they are not as thickheaded as the last thousand papers the professor has endured. While critics of standards-based grading complain that it sets the bar too low because too many students can succeed, precisely the opposite is true. It is comparative grading, typically based on the normal distribution, or bell curve.

      Finally, to be fair to Princeton, it is noteworthy that the document cited previously explains that the university changed its grading policy back to allowing faculty determination of proficiency, not the bell curve, to determine student grades.

What the Critics Say

      Standards, along with standards-based grading, are not without their critics. There are critics from both ends of the political spectrum who, while disagreeing on nearly every other element of education policy, are united in their condemnation of standards and standards-based grading. Two of the leading critics of standards are Diane Ravitch of New York University (http://dianeravitch.com) and Yong Zhao of the University of Oregon (http://zhaolearning.com).

      Professor Ravitch is particularly critical of the relationship between standards and the misuse of standardized testing—the connection between corporate test vendors and standards-based accountability. She rightly contends that school reform has never been achieved through threats and intimidation and has offered devastating critiques of those who magnify the deficiencies of schools. But it is important to separate the argument from the implications. There is broad agreement among researchers that standardized testing—particularly when used in a high-stakes environment to threaten students and teachers—has been a grossly ineffective and an unreliable tool. In fact, it is counterproductive (Guskey, 2015).

      When teachers are threatened with termination on the basis of standardized test results, the result is not necessarily an improvement in teaching and learning but a migration of teachers from low-performing systems to high-performing systems. This phenomenon leaves low-performing schools with a dearth of the best teachers and provides students who most need excellent instruction with an annual merry-go-round of new and inexperienced teachers, along with veterans who were unable to get a job elsewhere.

      This does not deny the fact that there are excellent and dedicated teachers laboring in low-performing schools, but the overwhelming evidence is that these are the exceptions (Haycock & Crawford, 2008). The faculties and administrations of low-performing schools are largely populated by a revolving door of professionals who, no matter how dedicated, know that the performance of their school could doom their careers, and they seek to leave as soon as possible.

      In essence, Ravitch’s criticisms of standards are not so much an objection to the establishment of academic standards but rather their ill-advised use for standardized testing and the resulting enrichment of test vendors.

      A different criticism of standards comes from Zhao (2014), who contends that all standards