Teaching
Perelman taught in and directed writing programs at Tulane University and the University of Southern California. At MIT, he taught writing and composition and served as the director of Writing Across the Curriculum and an Associate Dean in the Office of Undergraduate Education.Criticism of essay scoring
SAT
Following his 2005 study of essay samples as well as graded essays provided by theAutomatic scoring
In 2012, Perelman demonstrated that long, pretentious essays could achieve higher scores from the ETS scoring engine e-Rater as opposed to well written essays. In 2014, Perelman collaborated with students at MIT and Harvard to develop BABEL, the "Basic Automatic B.S. Essay Language" Generator. The nonsense essays generated by BABEL are claimed to perform well when graded by AES systems. Automated graders, Perelman argues, "cannot read meaning, and they cannot check facts. More to the point, they cannot tell gibberish from lucid writing." Perelman's work is cited by the NCTE in their Position Statement on Machine Scoring, which expresses similar concerns about the limitations of AES:Computer scoring systems can be "gamed" because they are poor at working with human language, further weakening the validity of their assessments and separating students not on the basis of writing ability but on whether they know and can use machine-tricking strategies.
Influence on Australian Educational Testing
During 2017-2018, Perelman was commissioned by the New South Wales Teachers Federation to write three reports to assist in efforts to reform Australia's national primary and secondary school assessments, the National Assessment Program – Literacy and Numeracy (NAPLAN). His work was a major factor in the decision of the National Education Council to overrule the Federal Education Minister and prevent the use of Automated Essay Scoring for the writing portion of the NAPLAN tests.References
External links