My name is Arthur Chiaravalli and I teach Mathematics and Language Arts at Haslett High School. In 2005, I had the opportunity to hear assessment expert and consultant Ken O’Connor speak at the Hannah Community Center in East Lansing. What stunned me about Ken’s presentation was how it challenged so many of the assessment, grading, and reporting practices that I—and just about every other teacher I knew at the time—had been using without question.

From my days as a student, through my pre-service training, all the way up to my hearing Ken O’Connor speak, I had spent little time considering these practices or how they might be improved. I suppose that, on some level, I knew that teachers employed a vast, idiosyncratic array of grading scales, weights, late policies, dropped scores, and curves in the calculation of a student’s letter grade. I did not, however, realize the adverse effect these and other common practices, as well as their inconsistency across teachers, had on learning in my classroom. Ken opened my eyes to the fact that the primary purpose of grades is to communicate information about students’ achievement of learning goals. Anything else is just muddying the waters.

At the time, I was teaching at our district’s alternative school and it was all too evident the ways in which the grades I assigned my struggling students were not so much a reflection of their mastery of the material as it was a manifestation of their chaotic lives. Many of these students were commuting from other districts or raising a child, or struggling with addiction, poverty, abuse, or homelessness. When I assigned an “E,” it was almost always a reflection of how these other, non-academic factors were wreaking havoc in their lives. My colleagues and I usually wavered somewhere between so-called compassion on the one hand (lowering expectations for students we knew were struggling personally) and rigidity on the other (giving them an “E” because, darn it, someone’s got show these kids that actions have consequences!). In light of Ken’s presentation, however, I saw that neither of these alternatives resulted in any real knowledge about whether students had achieved or were capable of achieving the intended academic goals. There simply wasn’t enough information to make that determination.

As time went on, I began to employ many of the “fixes” Ken had prescribed for “broken grades” (I strongly recommend Ken’s 15 Fixes for Broken Grades as a way to learn more about the topic). In particular, I began to assess the student’s current level of achievement by standard, rather than just scoring assignments and averaging them into one meaningless conglomerate grade. If a student failed an exam—whether for reasons of family turmoil, lack of mastery, or just laziness—I would keep teaching and keep assessing until the level of achievement was where I wanted it to be. In other words, failure was not an option. Students did not receive credit until they had mastered the stated power standards of the class. Contrast this with the more common way credit is granted—if your average is 59.5% (buffered by good behavior, some extra credit, and an unflagging ability to bring your pencil to class), we give you the stamp of approval and send you on to the next level. When I awarded credit for a class, I felt confident that the students were going forth with the skills needed to succeed at the next level. I doubt teachers sending on a student who passed with a D- can express a similar level of confidence.

Since that time, our district’s alternative school has been closed and I have now been a teacher at the “mainstream” high school for three years. The culture of this school is much more traditional in its orientation toward topics of assessment, grading, and reporting. I have, with varied results, attempted to pilot some mastery-based programs within the context of this more conventional point-accumulation paradigm. My after-school Skill/Point Recovery program allowed students the opportunity to (1) identify the learning objectives for which they lost the most points on the most recent chapter test, (2) spend additional time learning the skills or concepts associated with those objectives, and (3) take 4-to-7-question assessments reassessing each objective. Once a student achieved an 80% or better on these assessments, all the points associated with that objective were added back to the original test. I felt justified in adding back all the points because the questions on the assessments were often more far ranging and diverse than they were on the original test, which probably featured only one or two questions of that type. Likewise, all earlier assignments were either “no counted” or raised to the level of this improved summative assessment grade. Thus, the conglomerate grade reported in the automatically-averaging online grade book at least reflected the most recent level of attainment in each chapter. I allowed students to keep attempting these assessments—all of which featured different questions of the same type—until the end of the following chapter at which point the summative grade became fixed.

The link in the paragraph above (and here) goes to a complete description of the Skill/Point Recovery program, including a link to an article I wrote with Ken O'Connor about this.