Automated Writing Feedback Benefits English Learners

Automated Writing Feedback Benefits English Learners

“Writing is labor intensive exercise for both the student and the teacher and requires a lot of one-on-one,” said Wink Swain of Measurement Inc.

A 2012 case study agreed, "One reason that teachers may assign essays infrequently is that grading them is both time consuming and increasingly difficult. If a typical teacher is now instructing a single class of 20–30 students, and—in some cases—those same teachers are carrying 2-4 of those classes, a single long-form essay assignment of 1,000 words (4-5 pages) can generate between 100 and 600 pages of written material. Asking teachers to read, grade and provide substantive feedback on each one represents a formidable commitment of time and attention."

For 15 years a handful of companies have offered automated essay scoring. Studies have demonstrated scoring software can replicate trained graders, making it useful for testing situations. Perhaps even more importantly, these same scoring engines can provide useful formative six trait writing feedback.  

Jeff Pense, a Canton Georgia English teacher, assigns 28 essays each year to his 130 middle school students. He notes that “Each one of those was submitted for assessment an average of six times, so that’s well over 20,000 drafts that were assessed and graded, just for my classroom. There’s no teacher who could do that.” Pense uses Pearson’s WriteToLearn to provide formative feedback to students.  

Swain reports that teachers using the PEG Writing feedback system report that they don’t have to spend as much time on the mechanics of writing (spelling, capitalization, punctuation, etc.) which allows them to focus on more substantive skills (organization, elaboration, development of ideas, etc.).

Writing programs like PEG “meet students where they are, keep them focused and motivated on the task at hand without having to involve the teacher at every point,” according to Swain.   

Automated feedback, doesn’t replace the teacher, “It does give them more control over instruction and feedback,” said Swain, “It’s like having a tireless teaching assistant who can quickly assess each and every draft.” Feedback tools also provide teachers a summary of each student’s progress.

Writing assistants provide immediate feedback to students which helps internalize the encouragement, something “especially important for anyone coming to a task where they are notably behind,” according to Swain.

PEG is licensed statewide in North Carolina as NC Write and used extensively in Durham Public Schools. Like NC Write, Utah Compose is a writing solution for the state of Utah. Last year, it scored more than three million essays. PEG is also used by more than 136,000 secondary student through a Texas Education Agency pilot program (representing all 20 regions, 100 districts, 250 schools) most with large percentages of English language learners.

Dr. Joshua Wilson, University of Delaware, has found significant benefit in the use of writing feedback systems with students with disabilities finding results “encouraging, showing that students with disabilities grow at an [more] accelerated rate than their typically-developing peers and actually close the gap.” Wilson will be evaluating PEG use by Texas English learners.  

Pearson’s literary tool, WriteToLearn, supports English language learners with writing prompts and feedback that focus on the fundamentals of writing including language usage, sentence structure, mechanics and ordering of ideas. The NextGen version has specialized prompts for ELL students, a scoring rubric focused on writing fundamentals that corresponds more to the kinds of errors ELL students make, vocabulary training which focuses on teaching students about words they would need before reading texts they would summarize, and improved guided feedback and writing tips which can also be displayed in Spanish and Chinese.

Pearson recently launched TELL (Test of English Language Learning), a tablet based assessment with an interactive, fun interface for screening, diagnosing and monitoring ELL student progress.  It integrates automated scoring of writing and speaking technologies and uses tablet features to support interactive activities including reading short texts out loud, listening to instructions,  interacting with pictures, watching and describing videos and retelling stories.

More than half of the 24,500 students in Compton USD are English learners. The district uses WriteToLearn to encourage more and better writing (watch this YouTube).

A 2012 study evaluated Criterion Online Writing Evaluation Service by ETS in a college-level psychology course and a significant reduction in the number of article errors in the final essays of the non-native speakers.

“We use Criterion as an integral part of a comprehensive and coordinated plan to improve student writing across campus,” said CSU Fresno faculty. “This online program gives students the opportunity to rewrite and reduce the distractions in their papers so that professors can focus on the quality of content and evidence in written assignments. Many students, especially English Language Learners, find that the immediate feedback helps them gain a better understanding of grammatical features in writing.”

With recent enhancements for English learners, efficacy studies of writing feedback systems like Criterion, PEG, and WriteToLearn would be valuable to the sector.

For more see:

Dee Kanejiya

Founder and CEO at Cognii

9 年

Tom, this corroborates well with the usage we are seeing for Cognii as well. Previously, a community college or a large university would rarely assign open-response essay questions and students would rarely write answers. But after they started using Cognii, students are writing heavily and they are loving the AI-generated instant formative feedback that tutors them towards mastery. Great point Farzad H. Eskafi about the need for more advanced NLP approaches for assessment of writing and creativity.

回复
Farzad H. Eskafi

Product Lead @ Scale AI | Angel Investor | Ex LinkedIn, Ex-founder w/ 2 exits

9 年

Great post Tom. Personally, I'm very interested in any types of automation for essay grading and any types of open-ended exercises. The industry has definitely made quite a lot of progress in the field and a long way to go. Most focus on using LSA techniques - but some have even ventured further into sentiment analysis and motivation and creative "quantification."

回复
Troy Arnel Crayton, Ph.D.

Social Policy Analysis, Race Analysis, Social Research, and Strategy

9 年

Thank you for this post, Tom Vander Ark! The products reflected upon in your post and the suggested post "Next-Gen Personalized Learning for ELL Students" seems to be previewing the need for our "personalized learning" tools to take the next-step in learning facilitation. Eager to see products such as those mentioned to incorporate capacity to "learn," as found by folks like Kurt Fischer/Harvard Graduate School of Education and even as far back as John Dewey. It will be great to see the evolved versions by which the product learns a student's specific "lived experiences" as basis for "teaching" material and concepts. Exciting potential for learning equity foreseen!

回复

要查看或添加评论,请登录

Tom Vander Ark的更多文章

社区洞察

其他会员也浏览了