Charged with leading a committee that would revise his school's grading and reporting system, Warren Middleton described his work this way:The Committee On Grading was called upon to study grading procedures. At first, the task of investigating the literature seemed to be a rather hopeless one. What a mass and a mess it all was! Could order be brought out of such chaos? Could points of agreement among American educators concerning the perplexing grading problem actually be discovered? It was with considerable misgiving and trepidation that the work was finally begun.
Few educators today would consider the difficulties encountered by Middleton and his colleagues to be particularly surprising. In fact, most probably would sympathize with his lament. What they might find surprising, however, is that this report from the Committee on Grading was published in 1933!
The issues of grading and reporting on student learning have perplexed educators for the better part of this century. Yet despite all the debate and the multitude of studies, coming up with prescriptions for best practice seems as challenging today as it was for Middleton and his colleagues more than 60 years ago.
Points of Agreement
- Grading and reporting aren't essential to instruction. Teachers don't need grades or reporting forms to teach well. Further, students don't need them to learn (Frisbie and Waltman 1992).Teachers do need to check regularly on how students are doing, what they've learned, and what problems or difficulties they've experienced. But grading and reporting are different from checking; they involve judging the adequacy of students' performance at a specific time. Typically, teachers use checking to diagnose and prescribe and use grading to evaluate and describe (Bloom et al. 1981).When teachers do both checking and grading, they become advocates as well as judges—roles that aren't necessarily compatible (Bishop 1992). Finding a meaningful compromise between these dual roles makes many teachers uncomfortable, especially those with a child-centered orientation (Barnes 1985).
- No one method of grading and reporting serves all purposes well. Grading enables teachers to communicate the achievements of students to parents and others, provide incentives to learn, and provide information that students can use for self-evaluation. In addition, schools use grades to identify or group students for particular educational paths or programs and to evaluate a program's effectiveness (Feldmesser 1971, Frisbie and Waltman 1992). Unfortunately, many schools attempt to address all of these purposes with a single method and end up achieving none very well (Austin and McCann 1992).Letter grades, for example, briefly describe learning progress and give some idea of its adequacy (Payne 1974). Their use, however, requires abstracting a great deal of information into a single symbol (Stiggins 1994). In addition, the cut-off between grade categories is always arbitrary and difficult to justify. If scores for a grade of B range from 80 to 89, students at both ends of that range receive the same grade, even though their scores differ by nine points. But the student with a score of 79—a one-point difference—receives a grade of C.The more detailed methods also have their drawbacks. Narratives and checklists of learning outcomes offer specific information for documenting progress, but good narratives take time to prepare, and—not surprisingly—as teachers complete more narratives, their comments become increasingly standardized. From the parents' standpoint, checklists of learning outcomes often appear too complicated to understand. In addition, checklists seldom communicate the appropriateness of students' progress in relation to expectations for their level (Afflerbach and Sammons 1991).Because one method won't adequately serve all purposes, schools must identify their primary purpose for grading and select or develop the most appropriate approach (Cangelosi 1990). This process often involves the difficult task of seeking consensus among several constituencies.
- Regardless of the method used, grading and reporting remain inherently subjective. In fact, the more detailed the reporting method and the more analytic the process, the more likely subjectivity will influence results (Ornstein 1994). That's why, for example, holistic scoring procedures tend to have greater reliability than analytic procedures.Subjectivity in this process, however, isn't always bad. Because teachers know their students, understand various dimensions of students' work, and have clear notions of the progress made, their subjective perceptions may yield very accurate descriptions of what students have learned (Brookhart 1993, O'Donnell and Woolfolk 1991).When subjectivity translates into bias, however, negative consequences can result. Teachers' perceptions of students' behavior can significantly influence their judgments of scholastic performance (Hills 1991). Students with behavior problems often have no chance to receive a high grade because their infractions overshadow their performance. These effects are especially pronounced in judgments of boys (Bennett et al. 1993). Even the neatness of students' handwriting can significantly affect a teacher's judgment (Sweedler-Brown 1992).Training programs can help teachers identify and reduce these negative effects and lead to greater consistency in judgments (Afflerbach and Sammons 1991). Unfortunately, few teachers receive adequate training in grading or reporting as part of their preservice experiences (Boothroyd and McMorris 1992). Also, few school districts provide adequate guidance to ensure consistency in teachers' grading or reporting practices (Austin and McCann 1992).
- Grades have some value as rewards, but no value as punishments. Although educators would undoubtedly prefer that motivation to learn be entirely intrinsic, the existence of grades and other reporting methods are important factors in determining how much effort students put forth (Chastain 1990, Ebel 1979). Most students view high grades as positive recognition of their success, and some work hard to avoid the consequences of low grades (Feldmesser 1971).At the same time, no studies support the use of low grades as punishments. Instead of prompting greater effort, low grades usually cause students to withdraw from learning. To protect their self-image, many students regard the low grade as irrelevant and meaningless. Other students may blame themselves for the low mark, but feel helpless to improve (Selby and Murphy 1992).Sadly, some teachers consider grades or reporting forms their “weapon of last resort.” In their view, students who don't comply with requests suffer the consequences of the greatest punishment a teacher can bestow: a failing grade. Such practices have no educational value and, in the long run, adversely affect students, teachers, and the relationship they share. Rather than attempting to punish students with a low mark, teachers can better motivate students by regarding their work as incomplete and requiring additional effort.
- Grading and reporting should always be done in reference to learning criteria, never on the curve. Using the normal probability curve as a basis for assigning grades typically yields greater consistency in grade distributions from one teacher to the next. The practice, however, is detrimental to teaching and learning.Grading on the curve pits students against one another in a competition for the few rewards (high grades) distributed by the teacher. Under these conditions, students readily see that helping others will threaten their own chances for success (Johnson et al. 1979, Johnson et al. 1980). Learning becomes a game of winners and losers—with most students falling into the latter category (Johnson and Johnson 1989). In addition, modern research has shown that the seemingly direct relationship between aptitude or intelligence and school achievement depends upon instructional conditions, not a probability curve.When the instructional quality is high and well matched to students' learning needs, the magnitude of this relationship diminishes drastically and approaches zero (Bloom 1976). Moreover, the fairness and equity of grading on the curve is a myth.
Learning Criteria
- Product criteria are favored by advocates of performance-based approaches to teaching and learning. These educators believe grading and reporting should communicate a summative evaluation of student achievement (Cangelosi 1990). In other words, they focus on what students know and are able to do at that time. Teachers who use product criteria often base their grades or reports exclusively on final examination scores, overall assessments, or other culminating demonstrations of learning.
- Process criteria are emphasized by educators who believe product criteria don't provide a complete picture of student learning. From their perspective, grading and reporting should reflect not just the final results but also how students got there. Teachers who consider effort or work habits when reporting on student learning are using process criteria. So are teachers who take into consideration classroom quizzes, homework, class participation, or attendance.
- Progress criteria, often referred to as “improvement scoring” and “learning gain,” consider how much students have gained from their learning experiences. Teachers who use progress criteria look at how far students have come rather than where they are. As a result, scoring criteria may become highly individualized.
Teachers who base their grading and reporting procedures on learning criteria typically use some combination of the three types (Frary et al. 1993; Nava and Loyd 1992; Stiggins et al. 1989). Most researchers and measurement specialists, on the other hand, recommend using product criteria exclusively. They point out that the more process and progress criteria come into play, the more subjective and biased grades become (Ornstein 1994). How can a teacher know, for example, how difficult a task was for students or how hard they worked to complete it? If these criteria are included at all, most experts recommend they be reported separately (Stiggins 1994).
Practical Guidelines
- Provide accurate and understandable descriptions of learning. Regardless of the method or form used, grading and reporting should communicate effectively what students have learned, what they can do, and whether their learning status is in line with expectations for that level. More than an exercise in quantifying achievement, grading and reporting must be seen as a challenge in clear thinking and effective communication (Stiggins 1994).
- Use grading and reporting methods to enhance, not hinder, teaching and learning. A clear, easily understood reporting form facilitates communication between teachers and parents. When both parties speak the same language, joint efforts to help students are likely to succeed. But developing such an equitable and understandable system will require the elimination of long-time practices such as averaging and assigning a zero to work that's late, missed, or neglected.
Averaging falls far short of providing an accurate description of what students have learned. For example, students often say, “I have to get a B on the final to pass this course.” Such a comment illustrates the inappropriateness of averaging. If a final examination is truly comprehensive and students' scores accurately reflect what they've learned, why should a B level of performance translate to a D for the course grade?
Any single measure of learning can be unreliable. Consequently, most researchers recommend using several indicators in determining students' grades or marks—and most teachers concur (Natriello 1987). Nevertheless, the key question remains, “What information provides the most accurate depiction of students' learning at this time?” In nearly all cases, the answer is “the most current information.” If students demonstrate that past assessment information doesn't accurately reflect their learning, new information must take its place. By continuing to rely on past assessment data, the grades can be misleading about a student's learning (Stiggins 1994).
Similarly, assigning a score of zero to work that is late, missed, or neglected doesn't accurately depict learning. Is the teacher certain the student has learned absolutely nothing, or is the zero assigned to punish students for not displaying appropriate responsibility (Canady and Hotchkiss 1989, Stiggins and Duke 1991)?
Further, a zero has a profound effect when combined with the practice of averaging. Students who receive a single zero have little chance of success because such an extreme score skews the average. That is why, for example, Olympic events such as gymnastics and ice skating eliminate the highest and lowest scores; otherwise, one judge could control the entire competition simply by giving extreme scores. An alternative is to use the median score rather than the average (Wright 1994), but use of the most current information remains the most defensible option.
Meeting the Challenge
The issues of grading and reporting on student learning continue to challenge educators today, just as they challenged Middleton and his colleagues in 1933. But today we know more than ever before about the complexities involved and how certain practices can influence teaching and learning.
What do educators need to develop grading and reporting practices that provide quality information about student learning? Nothing less than clear thinking, careful planning, excellent communication skills, and an overriding concern for the well being of students. Combining these skills with our current knowledge on effective practice will surely result in more efficient and more effective reporting.
