The Case Against Robo-Graders

I wrote before about competing visions for online learning. On one hand, you can have teachers and professors put teacher-centered instructional activities (like lectures) online so that they can use their time in class with students to work on student-focused activities (writing skills, real-world application of math and science concepts, etc.). This movement continues to gain support. 

On the other hand, you could have a teacher monitoring a central computer while a large class of students completes a series of lessons on individual computers, followed by computer-based assessments. No one has to speak to one another.

I think the first method of online learning is the better one. Unfortunately, I think we’re probably more likely to go in the direction of the second because of budgetary considerations.

There is a new development in online learning: essay-grading software. This software aims to learn how a teacher grades a given written assignment, then grades the rest of the student essays automatically. The potential for cost-cutting is high: For one thing, the College Board would no longer have to pay for travel and lodging for hundreds of teachers and professors to grade each AP exam every year.

In order to be an effective tool for teachers, this software would need to do all of the following things in a history class: assess whether a student’s thesis addresses a history prompt in its entirety, whether each paragraph’s topic sentence is consistent with the thesis and is also historically accurate, whether each topic sentence is supported by accurate factual evidence and provide commentary as to why each piece of evidence is relevant, while ensuring that students are using an academic tone and are not plagiarizing.

For a history paper, for example, the program would need to provide specific feedback on each student’s strengths and weaknesses, including essay-writing skills and content recollection. Writing for English would have an entire different set of criteria.

Even if this were possible, essays have other benefits aside from assessing students. Teachers use the feedback from grading papers in order to assess where they need to focus their lessons. Teachers might find that students misunderstand factual content, or need extra work on their essay-writing skills.

One of my best professional development experiences has been as an AP exam reader. After six days of grading AP essays last year in Kansas City, I returned to San Diego with a much clearer understanding of how to improve my writing skills lessons. In particular, I learned how much the quality of students’ thesis statements can directly impact their entire essay scores on the free-response questions.  

My concern is that, even if software like this isn’t able to assess those basics, I imagine there will still be a lot of support for it because of its potential to cut costs.

In order to counter the calls to make cost-cutting reforms, teachers and leaders need to make the case for a rigorous, relevant curriculum and justify its potentially higher costs.

Oscar Ramos is a contributor to Voice of San Diego. Follow him on Twitter @OscarRamosSD or email oscar.ramos.blog@gmail.com.