There is no learning without assessment. According to Dylan Wiliam, whose research focuses on teacher development, assessment is the bridge between teaching and learning
. Wiliam states, “No matter how carefully we design and implement the instruction, what our students learn cannot be predicted with any certainty. It is only through assessment that we can discover whether the instructional activities in which we engaged our students resulted in the intended learning” (Wiliam, 2013
Assessment (assignments, exams, essays, homework, quizzes, in-class writing, projects, and other work that students use to demonstrate understanding) is part of learning. When grading assessments, instructors learn what it is students do or do not know and can make adjustments to the curriculum. Optimally, it’s also when students understand how they can bridge any learning gaps via provided feedback.
While many educators agree that assessment is a critical part of the student learning journey and an informative component of teaching, the ensuing act of grading assessments is an entirely different matter. Marking assessments can be an onerous task, often undertaken into the wee hours of the night, regardless of subject matter. Marking rarely takes place during instructional hours and is akin to “homework for teachers.” Furthermore, providing feedback to students while doing so can add to time spent grading and marking—again, regardless of subject matter. Whether grading an essay or an exam on organic chemistry or mathematics, upholding feedback loops and accurately measuring student learning can be stressful for instructors. As a result, multiple-choice exams have become popular; while they can, when well designed, measure student learning accurately, feedback is absent by nature.What is grading at scale?
Grading has become a pragmatic as well as a pedagogical endeavor for many educators.
Pedagogical best practices encourage educators to offer frequent, low-stakes assessments , which provide opportunity for feedback loops and for students to be supported in their learning journey before the pressure of high-stakes exams . While robust in theory, this can be challenging to implement and execute even within a small classroom setting, let alone a large introductory higher education course with hundreds, sometimes thousands, of students enrolled.
Even so, with minimal high-stakes assessments, such as the typical two-midterm/essays and one final essay/exam format that many higher education institutions employ, the grading load can be heavy. When there are, for example, one hundred students in a course, this means one such course produces three hundred assessments to mark; additionally, the best practice is to return them to students with written feedback before their next assessment, so the turnaround time must be expedient.
Large class sizes are an increasing global trend and unlikely to go away because of increased student enrollment, budget restraints, and in the case of higher education, student demand for introductory requirement courses. Despite deeply researched recommendations for class-size reduction, large classrooms are the norm for many instructors. (Brookings Institute, 2011 ). As a result, grading at scale is a unique pressure for educators today.
In fact, the pressure to grade work and its time-consuming nature has forced educators to enact some creative solutions; one such example is an emerging phenomenon of ghost grading , a term Dr. Sarah Eaton originated. In ghost grading, commercial third parties outside of an educational institution and who do not participate in the course curriculum are hired to mark and grade assessments for a course. While research on this topic is still ongoing, this shortcut solution likely has a negative impact on student learning as students do not receive feedback from the person they expect , let alone with subject area expertise.
In higher education, particularly in large introductory lecture courses, student work can be so numerous that collaborative grading is the primary solution. Collaborative grading–a group of TAs and instructors together teaching sections of the course or supporting the course grading assessments as a group task— is the most popular solution (and frankly speaking, the only way in many cases) to tackle the volume of work. In these situations, instructors (the teaching assistants and professor together) for one course will traditionally convene and grade student exams together, optimally using consistent rubrics . During the pandemic and emergency remote learning, many instructors did so with equitable grading software like Gradescope .How to engage fast and fair grading in teams
Grading in teams is effective, particularly when accompanied by clear expectations and tools to ensure fair and unbiased outcomes. What are some ways to enable this scenario?Collaborative grading tip #1: TAs or GSIs as a de facto grading team
This is the first step towards influencing fair grading and marking outcomes. The teaching assistant (TA) job is usually filled by an upper-level or graduate student at the university. By proxy, their first priority is to be a student, focusing on their own academic responsibilities and requirements. When they are a TA, they then add assistant and teacher to the list. Additionally, in some cases, there may be a different category to add to your grading team; “graders” or “grading TAs” are people who focus on grading and do not teach sections or hold office hours, but attend lectures so that they understand the content of the course.
Whether or not you choose your TAs or are assigned one, TAs often don’t come with prior teaching or grading experience. So it is important to support them in their new role and to help them understand course preparation. Having clear priorities and standards for grading and feedback is key to readying TAs for their responsibilities in your classroom.
Clarity around grading expectations can further be supported by asking a TA or grader to grade a single student’s work. The instructor can then more closely examine whether or not the TA understands guidelines and applies them with consistency (Dennis, 2019 ). Ensuring that the work of a TA is in alignment with a teaching and grading team is critical to fair outcomes.
When instructors are confident in their grading team and undertake the work to align expectations, then collaborative grading can be both fast and fair.Collaborative grading tip #2: Answer keys
Create and furnish an answer key or rubrics that offer guidelines to expected student responses and definition of performance level. Providing rubrics to graders allows them to understand and coordinate feedback on student work that furthers student learning and fosters consistency.
Grading and assessment tools like Gradescope enable personalized feedback that upholds rubrics. Additionally, Gradescope can store all your prior feedback to a certain question, so graders can opt to write new feedback or select one that is assigned previously. Again, this upholds fair and consistent grading.
On the other hand, some instructors choose to allow their teaching assistants to create their own rubrics per section; however, it’s important then to ensure they are in alignment with the rest of the team and the instructor’s expectations. If left unchecked, outcomes may vary section to section (for example, “the TA that grades easy” vs “the TA who is a hard grader and fails half the class”). Providing rubrics, too, can reduce the number of regrade requests, which then saves time for instructors.Collaborative grading tip #3: Grade by question
In addition to name-blind grading, grade by question, not by student, in order to reduce grading bias. Grading one question at a time—that is, if your exam has three questions, you grade all of question one first, then all of question two, and so on and so forth—ensures that you are using the same standard throughout your grading and are not biased by previous answers on the same test. (Aldrich, 2017 ).
Grading by question decreases bias and inconsistencies between members of a collaborative grading team; it is a function that is supported by many grading tools like Gradescope. Additionally, grading by question also benefits grading team workflows. A grader can decide to grade questions 1 and 2 one day, and 3 and 4 another day. Also, questions can be divided among graders; Grader A grades questions 1 and 4, Grader B grades questions 2 and 3, and so on (Ricci & Dogucu, 2021 ).
Name-blind grading can be enabled in grading software; it can also be executed by turning the covers of blue books back or dog-earing pages with names on them. Regardless of how name-blinding is done, it is a key component to unbiased grading.Overview: Best collaborative grading methods
Grading is an important part of teaching and learning ; it is an intersection at which instructors measure depth of student knowledge on particular concepts and where students can understand their learning gaps. Furthermore, grading and accompanying item analysis foster a robust curriculum and feedback provides next steps for the student’s learning journey.
Shortcutting grading is tempting. In fact, many educators feel they have no choice but to do so, in the face of immense workloads. These same educators believe that assessment and providing feedback to students are immensely important components of pedagogy. How can educators enable best practices of pedagogy while saving themselves time?
Collaborative grading in higher education is one way to manage workloads and spread it out amongst a team of graders. When paired with action to ensure that grading is fast and fair, grading can transform into learning for the instructor and the student.