Technology-assisted Grading and the Real Three C’s of Composition.
Abstract: With the increase of technology comes a predisposed inclination to rely on technologies to assist in grading. Rich Barlow (2019) explains this move, “In 18 of the 21[states], only a fraction of essays have human readers double-checking the AI [Artificial Intelligence] assessment, the latter not only spitting out test results in mere minutes but being cheaper than flesh-and-blood back-up” (para.5). Studies have shown the usage of technologies to assess Composition can be complicated and error ridden. The construct validity of E-raters and grammar checkers were questioned and found to be dependent upon length, mechanics, and verbiage. This article will discuss the usage of technology-assisted grading and the metrics of reliability; define the real Three C’s of Composition complexity, context, and conversation with text; examine how human understanding is necessary for assessment; share specific examples of assessment errors; explore a path forward that utilizes technology but reinforces the need for human understanding.