Robo-Readers
Although
much advancement has occurred in the field of automatic grading of student
essays, there are simply too many issues with the program to use it on a wide
scale. With our current technology, a robot can’t think. They can’t feel the emotion
that a paper evokes as a part of an argument, nor appreciate the truth.
Eliminating
the human element in grading papers has come an extraordinary way. In A Win for the Robo-Readers, the author
quotes a study by the Hewlett Foundation. They said that in a study of 22,000
short essays, the computer scores were comparable to that of human scores. This
was an incredible accomplishment; however, it is important to realize that
these were essays written to be graded by a human reader, not a machine.
According to Michael Winerip in Facing a
Robo-Grader? Just Keep Obfruscating Mellifluously, the system has too many
flaws. Les Perelman, a director of writing at MIT says that test prep can
easily fool the system. When teachers and students learn the preferences of the
machine, writing will be transformed into something mechanical. Instead of
improving the writing of students everywhere, the Robo-Reader has the potential
to seriously harm the English Language.
One of
the most troublesome flaws is the inability of a machine to “be human.” Scoring
no longer is based on a solid argument, but according to the official website
of the inventing company, it relies on a “content analysis based on vocabulary
measures” as well as a series of grammatical test. This all sounds great in
theory, but when dealing with rhetorical strategies that stray at all from the “proper
use of the English Language,” the Robo-Reader utterly fails. Some of the greatest
authors of all time utilize fragments and short paragraphs to solidify and
emphasize their points, but with these new computer programs, these are
automatically wrong. According to Michael Winerip in an interview by Melissa
Block, the substance of the argument is not important as long as the computer
thinks you are arguing well. He says “You could say that the war of 1812
started in 1945” and as long as this “fact” is incorporated into a well-structured
argument, “there are all kinds of things you could say that have little or
nothing to do with reality that could receive a high score.”
This in
itself is a major flaw in the program, but by viewing the so called “feedback”
given by the computer, the hole is ripped even wider. The computer is looking
for a specified length and a few specific terms to give a grade, but in
reality, this is just a small part of writing. A solid argument can be
incorporated in a much shorter or longer paper than the machine deems
satisfactory. By trying to grade papers
this way, machines are pushing us not to think even more. Standardized tests
and grading are no way to learn, instead, we must push creativity and outside
the box thought: two things that machines are incapable of understanding.
Although
the thought of letting machines grade papers is tempting, it is simply
idealistic. There are too many flaws in the way that they grade and could hurt
the English Language as a whole. Nothing can replace humans in a job that
requires high level, creative thought.