The advent of computer-scored essays on the State of Texas Assessments of Academic Readiness (STAAR) has stirred a mix of reactions, with some educators expressing skepticism about the new system's effectiveness. According to the San Antonio Report, starting this year, a large portion of STAAR tests will be graded by computers, a pivot that has left district leaders and teachers questioning the accuracy of virtual grading over human evaluation.
The Texas Education Agency (TEA) turned to computers for grading STAAR essays primarily due to a surge in open-ended questions district leaders are gearing up to review tests scored as zero as they continue to scrutinize the reliability of machine grading versus human assessment. A "hybrid scoring system" described by Kelly Kovacs, the chief academic officer for the Schertz-Cibolo-Universal City Independent School District, begins with a computer initially scoring the essays, with a quarter subsequently reviewed by a person, as per information shared during a district board meeting covered by the San Antonio Report.
While TEA officials argue that the computer system is trained and overseen by humans and has yielded results akin to human scoring, school leaders like Jennifer Price, chief academic officer for the Keller Independent School District, expressed reservations: "I think anytime a computer program is going to take on grading of something of this magnitude, I think it is concerning," she told the Star-Telegram. Concerns have been raised about whether the machine's focus on specific keywords and phrases might prompt educators to change their teaching methods, potentially prioritizing formulaic writing over more nuanced, complex expression.
At the heart of this issue, computers have been entrusted to handle a task once solely the purview of human judgment, parsing through students’ writing and assigning a score that predicates their academic progression, a responsibility lying heavy on concern for TEA's gravitation towards AI complements despite assuring human intervention. TEA's Chris Rozunik informed the Star-Telegram that the computer program uses the same rubric as human graders and is trained on actual scored essays. However, districts like San Antonio ISD are wary, with spokesperson Laura Short saying, "We worry that nuanced, complex writing will not receive a valid score because the AI engines may be looking for specific words, phrases, and formulaic construction," in a statement obtained by the San Antonio Report.
Despite these sweeping changes in STAAR test scoring, educators like Tony Perez from Northside ISD consider stalwart teaching methods crucial to student success. "I think good writing instruction is just good writing instruction," Perez told the San Antonio Report, a sentiment reflecting a determination among educators to sustain quality education amid evolving assessment landscapes.
