When the results come in from the first test in my introductory survey course, I am not surprised by the large number of D's and F's. I've come rather to expect that many, if not most, of the students will fail.

The subject matter is not rocket science, and the homework load for each session of the course is modest: 15 pages of textbook reading, or 40 to 50 pages in supplementary nonfiction books written for a general audience. I average students' final grades from four objective tests and three subjective papers.

My homework expectations are near those found to produce the best results by Richard Arum and Josipa Roksa in Academically Adrift: Limited Learning on College Campuses, a stinging indictment of low expectations in higher education. I require a little less than the 20 pages of writing that Arum and Roksa deem optimal for an intro course, but I assign a little more than the 40 pages of reading a week that they recommend.

That amount of work, it turns out, is way more than many of my students have bargained for. Why that is the case can be variously explained, but the fact itself cannot be denied. It's something college instructors have to deal with. Here's what I've noticed about how we do that.