Those are the same two decades over which the conventional wisdom has been university professors are underworked and overpaid. Why haven't U.S. students, who are quite good at discovering everything else easy, swarming into the graduate programs? (Whether social justice math and other deducationist fads are confounding effects will be left to the reader as an exercise.)
With a steep rise in the number of foreign graduate students in the last two decades, undergraduates at large research universities often find themselves in classes and laboratories run by graduate teaching assistants whose mastery of English is less than complete.
The issue is particularly acute in subjects like engineering, where 50 percent of graduate students are foreign born, and math and the physical sciences, where 41 percent of graduate students are, according to a survey by the Council of Graduate Schools, an association of 450 schools.
The article focuses on the steps being taken by universities, sometimes with the coaxing of trustees, or of legislators, to ensure that graduate assistants are sufficiently proficient in spoken English to be able to take questions and improvise, skills that native speakers often would do well to refine. Meep Meep .... has further thoughts along those lines.
There is, however, no investigation of the following situation, which I have encountered more than once. The student finds it easier to complain about difficulties understanding "take rrimit, goes to jhero" in quiz section than to make the effort to comprehend what "if abs(x-c) < ?(delta), then abs(f(x)-L) < ?(epsilon)" means in the first place, which might mean sitting down with pencil, paper, and book open for a while. (What would Kronecker make of my efforts to get a delta and an epsilon into the post here?)