Beginning in fall 2017, some students and educators at the University of Michigan may be getting help on writing assignments from computers.
Campus Technology reports that a team of educators developed a writing-to-learn tool called M-Write, which uses automated text analysis (ATA) to identify the strengths of a writing submission.
Developed by two professors, the tool was initially meant to help students grow their conceptual learning skills in large courses and to help streamline the grading process, reports a UMich article.
ATA works by “using a variety of text analysis techniques, such as vocabulary matching or topic matching, which the algorithm detects.” Using M-Write also lets educators identify the students who are going to need help.
UMich reports that M-Write’s ATA capabilities will be used in a statistics class in fall 2017. Writing fellows will verify the tool’s grading before sending the assessment to the student with a computer-generated revisions list.
“The verification to us is very interesting because it not only gives the human graders a moment to pause and reconsider their assessments but more importantly it provides a direct feedback loop to the algorithm development and allows us to create a better one,” says Chris Teplovs, the lead developer at the Digital Innovation Greenhouse that helped develop ATA.
UMich reports that professors who piloted the tool, like economics lecturer Mitchell Dudley, appreciated the ease in grading writing assignments and the ability to assign more ways for students to demonstrate what they’ve learned.
Elements of machine learning like ATA have also been incorporated into other college courses to create adaptive learning programs.
General Psychology is a general education course with many sections that can often be taught by adjuncts. The professor saw adaptive learning as a way to provide more-uniform content across all sections. Finding a way to maintain quality and personalize the experience for large online classes increases access, allowing more students to be accommodated.
ECAR reports that overall students appreciated how the system was able to determine their knowledge and comfort level and then distribute course content at the right speed and format for the psychology course and other pilot courses at UCF.
“As with any new tool, adaptive learning provides a new set of capabilities and insights — and a lot of very useful data — that can be used to explore ways to increase students learning and success,” ECAR reports.