For many decades, the U.S. education system has operated under a standard grading system that most stakeholders (principals, parents, teachers, students, and colleges) accepted as the standard. Now, the education community is beginning to look at alternatives to traditional grading practices. To explore that concept further, we convened a group of educators, including current and former principals, to chat about the upsides and downsides of alternative grading systems. Roundtable participants included Tom Dodd, principal of Lesher Middle School in Fort Collins, CO, and the 2017 NASSP National Principal of the Year; Andy Greene, retired principal of Candlewood Middle School in Dix Hills, NY, and an adjunct professor at Stony Brook University; and Jay McTighe, an educational writer, author, and consultant with more than four decades of experience in education at the school, district, and state levels. Principal Leadership Senior Editor Michael Levin-Epstein moderated the discussion.
Levin-Epstein: Why should we even be looking at other kinds of approaches to grading?
Dodd: It’s hard to talk about alternative assessment approaches without criticizing current widely accepted grading practices. Most teachers grade the way they were graded. Most parents expect their children to be graded the way they were graded. But ask those same teachers and parents how we arrived at a 100-point scale, A to F grades with 9–10-point ranges, or the threshold between a satisfactory/unsatisfactory or pass/fail mark, and you’ll be hard pressed to get a rational explanation. A score in the range of 0–59 is generally recognized as an F, 60–69 a D, 70–79 a C, and so on. How did that come to be? What about the 4.0 grade scale at the university level? Where’d that come from, and how does it convert to a 100-point, A to F scale in classrooms? When you challenge the historical way we’ve always done things, which people have come to accept, you risk pushback and confusion. There’s increasing research challenging the status quo of traditional grade reporting and touting the merits of a more performance-based, proficiency-driven system tied to local, state, or provincial standards, or other criteria like the International Baccalaureate’s (IB) aims and objectives.
If you look across subject areas, grade levels, and schools, teachers are assessing students with different priorities (participation or performance), at different intervals (quarter, trimester, semester, or year), based on different tasks (attendance, homework, quizzes, labs, tests, papers, etc.). Elementary schools might assess students as unsatisfactory, satisfactory, or excellent from kindergarten through second grade, then begin the traditional grading practice of averaging a collection of points in third grade. Why? They may also assess desired behavior using categories like “some of the time,” “most of the time,” or “all the time.” High schools, with their focus on Carnegie units/credit hours, graduation, and college scholarships, might weight grades in Advanced Placement and International Baccalaureate courses, but not the prerequisite advanced track courses that got students to those courses. Why not? Educators talk a lot about instruction and curriculum, and a lot less about developmentally appropriate assessment that defines and measures mastery. What should a student know and be able to demonstrate after a given period of instruction? Can you think of a class from your school experience in which the perception was everyone got an A just for showing up and participating? What about the opposite—can you think of a teacher who would not award an A even if the student had mastered all that was explicitly taught?
McTighe: I’ll begin with an essential question: What is the purpose of grades and associated reports? Arguably, the primary purpose is to communicate clearly and honestly to parents, students, teachers, administrators, and other organizations, including perhaps colleges or employers, the extent to which students have learned material that they are studying or the extent to which they’ve achieved targeted learning goals. But the reporting system can also communicate other factors, such as students’ work habits, behavior, attendance, and their learning progress over time. If we start with a clear sense of purpose, we can then evaluate the effectiveness of different approaches to grading and reporting.
Greene: I think that the purpose for grades has to take a leap into how scores are being reported based on the standards. A grade of a C doesn’t really cut it anymore, and I do believe that parents and students need more specific information related to how well students are doing against identified standards if they are to progress.
The Devil Is in the Details
Levin-Epstein: Can you give us an example of an alternative grading system that’s now being implemented in secondary schools that you think is working well?
McTighe: There’s been a push in North America and elsewhere to implement what some are calling “standards-based” grading. The logic of standards-based grading is sound. It simply purports that we start with established standards that specify what students should know and be able to do in various subjects. We then use appropriate assessments of that knowledge and those skills as a basis for grading and reporting the extent to which students have learned them. This makes sense as an approach to grading. One of the key principles in standards-based grading contends that grades should focus initially on the extent to which students have learned standards, and [then] separately grading and reporting on other factors, such as progress, work habits, behavior, and attendance.
Consider a situation in which three students each receive a B for a marking period for three very different reasons. Student No. 1 has shown solid mastery of basic knowledge and skills, but not at an advanced level. Student No. 2 is very capable and displays advanced knowledge and skills, but doesn’t do homework and often acts as a “class clown.” While this student’s performance may be advanced, the teacher refuses to award an A given the child’s minimal effort and annoying behavior. The third student has an IEP, but works very hard and has made significant progress, even though their performance level is well below the standards. However, the teacher wants to reward this student’s work habits and growth, especially given the learning challenges the student faces. Given these examples, the common grade of B is inherently misleading, since it reflects three fundamentally different levels of achievement, progress, work habits, and behavior. Standards-based grading seeks to make letter grades more meaningful and honest by separating qualitatively different factors and reporting them separately.
Greene: I would agree. I think that report cards must be more specific, and they have to take out the other factors that sometimes go into how people configure grades. It’s really vital that parents and students get accurate information of where they stand against certain criteria without other nonacademic factors that are occasionally considered. Behavior must be separate from grades. Finally, for me, and hopefully we’ll get into this a bit, it’s very important for youngsters with special needs to have grades that are fair and also communicate to parents where students stand in relation to their IEP goals and where they stand in relation to grade-level standards.
Levin-Epstein: Tom, in your middle school you have instituted a competency-based assessment and reporting program. Tell us about that.
Dodd: We instituted criterion-referenced assessment and reporting using the IB Assessment Criteria and Approaches to Learning Skills three years ago. We wanted to give students and parents more detailed information about academic performance (mastery) and work habits (process). Bob Marzano’s notion of a guaranteed and viable curriculum—which is giving students, regardless of teacher, access to the same content, skills, and concepts—was another motivator. So, we eliminated percentage grades, averaging, and the 100-point, A to F scale. Our teachers embed the Colorado Academic Standards into their unit plans, and as an IB Middle Years Program, create eight-point, task-specific IB rubrics across eight subject areas. Each subject area has four criteria that are averaged (because the four criteria are weighted equally) for an overall 0–8 course grade. All five of these scores per course (the four criteria scores and overall score) are communicated via our online gradebook and quarterly report cards.
In the past, a student might get a B in eighth-grade English, but as Jay alluded to, what’s a B in eighth-grade English? Did he/she correctly answer 80–89 percent of whatever the teacher asked? Did he/she perform better than 80–89 percent of his/her peers in the class? At Lesher, that student’s quarterly report card might show a 6 in Analyzing, a 5 in Organizing, a 5 in Producing Text, and a 4 in Using Language, for an overall English course grade of 5, based on the IB rubric descriptors. A 5 on an 8-point scale is not 63 percent, or a D. Per the IB rubric, a 5 means that the student can competently, substantially meet the criteria with a considerable degree of accuracy. We also separate behavior from the academic grade using the IB’s Approaches to Learning Skills as work habits-1) Self-management, 2) Communication, 3) Research, 4) Social, and 5) Thinking. We measure them using a rubric with an alpha scale- N = Novice, L = Learner, P = Practitioner, E = Expert-and communicate these scores for each course via our online gradebook and quarterly report cards as well. The IB’s Approaches to Learning Skills align very closely with Colorado 21st-century standards.
Trend or Fad?
Levin-Epstein: What percentage of secondary schools do you think are seriously looking at the standards-based approach or other approaches to traditional grading?
Greene: I will tell you from my vantage point, being a principal for 26 years, people are starting to talk about it. However, it’s a big change for parents. Parents went to school, they saw numeric grades, or they saw A’s, B’s, and C’s, and for them, you really have to clearly communicate the purpose of grades. Although schools are looking at standards-based grading, it’s slow to get off the ground generally. Parents need to be comfortable with the concept, and districts cannot rush the process.
McTighe: I agree. Changes in grading practices do not come easily. I have observed that there are at least two other factors that tend to perpetuate traditional grading approaches. One of those is the report card itself. If a school or a district has a single category (e.g., reading, science, or English) and the school uses an A to F grading system, teachers are essentially forced to generate a single grade for the category. In other words, the report card format itself can perpetuate ineffective grading because it suggests that teachers need to mix various factors (i.e., achievement scores, growth, progress, work habits, factor in kids who may have learning challenges) to provide a single grade.
A second problem is manifest in some computer-based grading programs. While it certainly makes sense to use software to record evidence of student achievement, some grading programs require teachers to enter scores (for example, scores on tests or quizzes) and then, once certain numbers are entered during the marking period, the program automatically averages these to generate a final grade. This is a problem if we are committed to a standards-based grading system, because averaging may not, in fact, be the most appropriate measure of achievement at the end of a marking period.
For example, a student may not have learned material early on, and their assessments show it (with a D or 62 percent). But over time, as they develop greater understanding and proficiency, they may, in fact, eventually master the material such that their later grades should be B or A. However, if their lower early scores are averaged with their later high scores, the resulting grade may not be an honest reflection of their actual level of attainment. In other words, later scores should count more than the early ones. Advocates of standards-based grading also challenge teachers who grade everything, including formative assessments, and average those with scores on summative assessments that may occur toward the end of a unit or a marking period. Formative assessments are meant to inform (i.e., to provide feedback to teachers and students), not to evaluate. Accordingly, formative assessment results should not be included as part of an achievement grade.
Greene: I think that’s very critical, because you shouldn’t grade students’ learning while in progress; you want to see if they ultimately met the standard. Students should be judged more heavily if they mastered the standard, whatever the standard is, whatever the outcomes are, versus just averaging all of their grades during the course of a semester. That doesn’t seem to be fair for any aspect of grading. By letting students know that they have a chance to demonstrate mastery by the end of the marking period, it will help them stay positive and motivated for a longer period of time.
Levin-Epstein: How do you go about implementing an alternative grading system?
Dodd: Our school piloted IB criterion-referenced grading for two years before the other nine middle schools in our district joined us. We worked through a lot of technology issues at the district level in trying to get the district report card to communicate what we wanted it to communicate. We did a book study with our staff, brought in a consultant, and sent teachers to conferences and IB workshops. We sent parent letters home and hosted an IB assessment information night, but the best communication was course-specific and came directly from teachers’ classrooms. Students also helped get the message out through their fall and spring student-led conferences. Parents felt much more reassured when their children could show them artifacts and explain to them at the beginning of a unit what they were expected to know and do, and how they would be assessed.
Levin-Epstein: Does the high school have a traditional grading system?
Dodd: Our high schools are still deeply rooted in traditional grading, and I don’t see that changing anytime soon. We created a conversion chart for high school graduation credit-bearing courses taught in our middle school, like algebra; geometry; Algebra 2; and Spanish, French, and German 1 and 2. Students have been matriculating from elementary to middle to high school forever, and I would challenge the assumption that elementary or high school teachers in the same grade, or even the same department, are grading the same way. Maybe the low graduation rate and high dropout rate in some schools could be correlated to traditional grading practices? We are a team-based middle school. Our teachers co-create 0–8 leveled tasks and assessments using the IB assessment criteria, and table-grade using high, medium, and low examples without students’ names on them (because if you put students’ names on them, you might see a wider disparity in teacher judgment).
We have the latitude to be a little more experimental at the middle level, and I think that’s a good thing, because we don’t yet have credits, transcripts, graduation, college admissions, and scholarships tied to seat time, grade-point average, class rank, etc. We also focus a little more on the social-emotional side. Students arrive at our school from about 20 different elementary school options, and they have success here. We promote students to all four of our large, comprehensive high schools, and they experience success there; some even become Boettcher and Daniels scholars, National Merit finalists, valedictorians, etc.
Levin-Epstein: Other thoughts about what a better grading and reporting system might look like?
McTighe: I am drawn to an idea that Dr. Thomas Guskey proposed some years ago, where he talked about the need for “three P’s” in grading and reporting. The first P is to report accurately and honestly on a student’s performance level. This is where criterion-referenced, standards-based grading would report the extent to which students have achieved a learning goal according to an agreed-upon scale (e.g., A to F, a 4-3-2-1 rubric, or some other categorical system). The second P reports separately on progress—the extent to which students have improved over time from where they started.
The third P reports on process factors. Now, this third category would have to be operationally defined by a school or a district. But generally, we’re talking about work habits—the extent to which a student consistently does classwork, does homework, pays attention, listens to teacher feedback, and so on. My contention is that we would have a more honest and effective communication system if we reported on these three factors separately.
Greene: Just to add to that, as I mentioned before, with special education youngsters, I think that student grades have to be reported against where they are in their IEP goals and where they are against the grade-level standards so parents have an accurate measure of their true level of progress.
Also, it has to be clearly defined. Again, talking about Thomas Guskey, he speaks to the difference between accommodations and modifications in grade reporting. When a student receives an accommodation, the student should be held accountable to the same standard as other youngsters. But a modification, that’s different. The grades have to reflect whatever the modified standard is.
So, if you are going to look at leveled knowledge, similar to the SAT that has five levels, you could report out to parents how well students did on simpler details on the standard and then more complex details as well. This would give parents accurate information that “my son or daughter can do some basic information, but they need a little more work to get to the standard.” Since standardized tests have leveled questions, I think that’s going to have to be a piece of standards-based grading down the line.
McTighe: Nearly every school or district that has tried to make any kind of substantive change in its grading practices or its reporting process almost universally receives pushback. It may be from teachers who see the new grading system as requiring more work from them, and/or it may be from parents who don’t understand why it’s different than what they knew when they were students in school. So, my suggestion is straightforward: School leaders need to be able to make the case for any and all proposed changes to grading practice or the reporting format.
One way to do this is to begin with an honest discussion about the inadequacies or the limitations of the current grading/reporting system. My experience is if you ask teachers, parents, and students, they will be pretty quick to recognize and articulate some of the weaknesses in the current system. In other words, there has to be a compelling reason to make a change—not just because a principal thinks it’s a good idea or a central office staffer heard about it at a conference or read about it in a journal. You’ve got to get the people involved to recognize that there are some fundamental weaknesses in the current system that need to be addressed. Awareness of the problem sets the stage for needed changes.
A second practical tip for initiating any grading/reporting reforms (for example, shifting from norm-referenced bell-curve grading to standards-based, or reporting on three P’s—not just a single grade) is offered by Tom Guskey. He suggests that any grading/reporting changes be tried initially as a pilot project. For example, for one or two marking periods, a school could provide two grading systems; one would be traditional, what we’ve always done, and the second would be the new standards-based or three P’s approach. Let teachers, parents, and even students weigh in on whether the new reporting system is clearer, more communicative, more honest, and more consistent. If it’s a better system, it should win out. But it doesn’t obligate us to making the switch if we cannot “sell” the new system as a better approach.
Greene: There’s an old saying, “I can endure any ‘what’ if I understand the ‘why,’ ” and I think the “toe in the water” approach with changing grading systems is the way to go. Michael, you mentioned the middle school. One way to do it is to get parents on one of the teams and meet with the PTA and go through the process exactly as Jay said. Do it as an action research pilot program versus a wholesale change. I think that you have to go for the early-win-in-Iowa kind of idea, and not do a massive change. There needs to be a lot of communication that has to take place with the various constituency groups. Start off small, being really clear on the purpose of standards-based grading, and get everyone involved upfront. Being transparent is the key.