Myron Dueck, Paul Thomas, and Melissa Tomlinson offer their ideas of testing utopia
Assessment remains one of the most contentious aspects of education policy for a variety of reasons, from the way results are used to evaluate teachers to accusations that bias gives affluent, white children an advantage to the effect of high-stakes testing on the whole teaching process. Despite the polemics, few can argue against the importance of gauging student learning and progress, so what should testing look like?
Language Magazine asked for feedback about what assessment could look like in a perfect world — that is, a world without budget restrictions, where school districts have access to the most cutting-edge materials and technology, and where educators enjoy unprecedented autonomy because they are confident, prepared, and have access to unlimited professional development, resources, and training.
If I were to stretch into the science-fiction world and dream up the ultimate future assessment system — let’s call it “Assessor 2100” — it might include the following variables:
• Constantly updated, real-time indicators would be automatically sent from each student to a central assessment-data collector.
• On a one-to-ten scale, a monitor near each student would display a holographic, constantly changing number reflecting understanding.
• Student understanding of smaller, individual topics would be automatically monitored, as well as understanding of larger, comprehensive concepts.
• The teacher would receive real-time suggestions on what to revisit, stress, or highlight depending on student understanding in the moment.
• The teacher would receive an easy-to-read analysis of the exact moment and activity when each student “got it,” along with classwide data showing optimum moments in learning.
• All data would be funneled into a reporting system indicating understanding on all required learning outcomes and on personalized learning modules that may have cropped up.
Essentially, Assessor 2100 would both measure and promote understanding for teachers and students. It would provide immediate and specific feedback and enable students and teachers to react to the data. At times, this feedback would facilitate learning, and at other times, it would verify it.
While we don’t have the technology now for Assessor 2100, we do have strategies for assessment and grading that can help us achieve some of these ideals. To engage in the conversation, we have to think differently about assessment and grading — especially in regarding what purposes we want them to serve.
A few years ago, I stopped using some of the antiquated grading practices to which I had become accustomed, and the results were staggering. I stopped using zeros for work not completed, and instead took steps to address behaviors that caused the missing work and offered supports for getting it done at school. I found my data was more reflective of learning, and the disposition of my students improved. If students received an INCOMPLETE on a required assignment, I applied the same status to their course standing. One student pleaded to “just receive a zero” on a missing assignment in order to avoid actually doing it. When I insisted that she complete it in order for me to grade it, she delivered a solid assignment.
When I jettisoned the nebulous use of deductions for late work, I saw students respond very favorably. Instead of using grading penalties completely unrelated to the measure of understanding, I adopted a system that enacted supports after an assignment was late by 48 hours. Students facing work requirements after school or suffering from dysfunctional living arrangements made my classroom completion center a “second home” during lunch and after-school sessions.
I stopped grading uniform homework as though it were a reflection of understanding. When I moved to quizzes based on suggested homework, students became more accountable for their own learning and less dependent on capable friends. At-risk students, and their parents, applauded a system that no longer levied penalties for incomplete homework that, given their circumstances, was sometimes virtually impossible to complete.
In most of my assessment decisions, I shifted to recognizing that IF something was learned was more important than WHEN it was learned. The path taken to arrive at the point of understanding should not play a role in determining whether or not I got there. If I am supposed to arrive in Boston by the end of summer holidays, and I do, who cares what route I took? I now strive to make mastery learning the constant and time the variable.
Perhaps this is what I would most want from Assessor 2100. In the meantime, while I wait for technology to catch up, the more I can make my assessment system based on timely, specific, and formative feedback, the closer I will get to my dream.
Myron Dueck is a vice principal, teacher, and author from School District 67 in British Columbia, Canada. Myron has presented in Los Angeles and Chicago at the ASCD Annual Conference and has twice been published in Educational Leadership magazine. His first book, Grading Smarter, Not Harder: Assessment Strategies that Motivate Kids and Help Them Learn, was released by ASCD in July 2014.
“Addressing teaching writing in 1957, English educator and former National Council of Teachers of English president Lou LaBrant explained, “Knowing about writing and its parts does not bring it about, just as owning a blueprint does not give you a house.”
LaBrant succinctly confronts several failures of traditional approaches to both instruction and assessment. First, her analogy exposes how we often value representation (blueprint) over the authentic thing (house); notably in her essay, she confronts that too often we allow grammar tests to represent a student’s ability to compose original texts, and consequently, students rarely draft original compositions.
Second, the implication in her analogy challenges the ways in which efficient formats of assessment (standardized multiple-choice tests) tend to drive instructional practices as well as curriculum. If LaBrant were alive today, she’d be on the front lines rejecting Common Core and its connected high-stakes tests as digging that same hole deeper, with no hope of climbing out.
Instead of blueprints (representations), our students need to be building their own houses, and in real-world classrooms, we need to set aside measurable outcomes and choose instead rich feedback for students addressing authentic artifacts of their learning.
Rich feedback includes written responses to student work as well as conferencing. In both cases, feedback should ask students questions about what they have created and why so that teachers can balance their roles as expert mentors while honoring students’ emerging autonomy. One important distinction about feedback is that it is instructionally enriching, and not designed to label (grade) or sort students. Feedback is part of the learning process that is an ongoing experience that includes students revising and resubmitting their authentic artifacts of learning.
What, then, are artifacts of learning? They’re not tests, but instead look like and share qualities with the sorts of things adults create and produce in the real world. Authentic artifacts come from student choice negotiated with teachers during workshop environments where students have large blocks of time and ample feedback from teachers and peers.
K–12 classroom teachers should be guided by how professionals in the academic disciplines act in the world (behaviors that are typical of each discipline) and what artifacts represent those disciplines. For example, professional writers and scholars write mostly by choice and then shape their compositions within the conventions of their fields or specific writing situations. Journalists research and write in ways that are different (although not better or worse) than biologists. Inviting students to behave in those ways — while offering rich feedback and allowing them to produce authentic artifacts of their learning — is our best alternative to traditional standards- and test-driven practices that deform learning instead of informing teaching and learning.
This approach becomes more valid than testing if then translated into grades because teachers and students have developed multiple data points over long periods of time (ideally in portfolio format) to inform those grades, overcoming as well the shortcomings of one-sitting high-stakes tests that remain biased by socioeconomic status, race, and gender.
Paul L. Thomas, EdD, taught high school English in rural South Carolina before moving to teacher education. He is associate professor of education at Furman University. His work can be followed at the becoming radical (blog) and @plthomasEdD on Twitter.
When we posed the question to the BadAss Teachers Association — a group with 51,000 members — there was one common theme. An overwhelming majority, something not always easy to find in large groups, all agreed that assessments should be teacher-created and student-driven. Teacher-created comes into play when a teacher plans a type of assessment based upon the learning needs of the students in the class. Some of the suggestions ranged from student portfolios to teachers recording observations as classwork was being completed. But what if we really take this outside of what has already been developed and think about the best way that our students could be assessed?
A few paradigms may have to be broken, such as a teacher being alone in a classroom or having limited hours in the day to meet with people outside of the classroom. To create an assessment, we all know that twe must first define what we want our students to learn.
If students are individuals with unique personalities, then it stands to reason that each assessment completed by a student should be an individual product. Thinking in terms of a large assessment that would show the growth of the student through the course of a unit, a sample of what the student already knows should be completed at the beginning of the unit. As a team, the teachers would have the time to meet and discuss the creation of this pre- and final assessment. At this point, the same task could be given to all of the students. After all, wouldn’t it be exciting to realize exactly how much individuality a student could bring to an assessment product after learning has been achieved?
Formative assessment would be done throughout in the form of teacher observation with anecdotal notes. Teachers would meet on a weekly basis to compare notes, discuss student learning, make suggestions, and collaborate on progress of the unit, where it was currently, and where it could go from there. The final assessment piece, designed within a set of loose parameters already outlined by the teachers, would give the students a chance to show their understanding of the unit through the creation of a piece that they design. Again, the teachers would meet to review all of the completed assessments. Discussions would take place about how well the students demonstrated an acquisition of knowledge. Reflections would occur about the whole unit as a process for the teachers to determine what went well and what might need adjusting for the next year. Notes would be utilized in the development of the unit for the following year.
This entire scenario is based upon the premise that teachers are given the time to develop and nurture a deep understanding, as well as have the ability to constantly collaborate with other teachers. It is based upon the idea that a student, when allowed to tap into natural learning abilities, will be able to obtain a deeper understanding. It is based upon the idea that students and teachers are individuals, and as such, education and learning need to be treated as fluid, adaptable ideas, not things that are standardized or common.
Melissa Tomlinson, the assistant general manager of the BadAss Teachers Association, has a master’s degree in the teaching of students with disabilities. She is currently employed in New Jersey as a middle school special-education math teacher and is the mother of two teenage sons.