University President Lawrence H. Summers won his job in part on the strength of his commitment to issues of undergraduate education. And while Summers will not reveal in-depth plans, when he is asked about improving the undergraduate experience, he invariably cites Harvard Professor of Education Richard J. Light and his latest book, Making the Most of College: Students Speak Their Minds.

Light, who holds a joint appointment at the Kennedy School of Government and the Graduate School of Education, is gaining attention at campuses across the country for his detailed look at the factors which influence student satisfaction with the undergraduate experience. But nowhere are his findings more relevant than at Harvard.

Light bases his book on the 1,600 interviews he and his associates conducted with recent Harvard undergraduates. His research was encouraged by two previous Harvard presidents, and his findings have been embraced by numerous other Harvard administrators and faculty. He has met once with Summers for a one-on-one discussion on the College, and earlier this morning helped herald Summers’ installation by moderating a panel on the future of undergraduate education.

With little else to go on, Light is a good clue as to where undergraduate education under Summers might be heading.

Light says that if he were czar for the day, with unlimited resources and absolute authority, the single issue he would address would be that of student-faculty interaction.

“If the resources were somehow gettable, the best thing would be making it so a student could take one small class a semester,” Light said in an interview last week.

Indeed, a good portion of his book is dedicated to a discussion of class size, mentoring and the informal opportunities for contact between the teachers and the taught.

Light’s conclusions here, as in many other areas touched on in the course of his discussion, are deceptively simple. Students say they crave seminar-style classes where they have the chance to deal directly with distinguished professors in small settings. Schools then should focus resources on hiring the faculty to make these classes a reality for more students. Students need interpersonal contact with their professors. Schools then should encourage more faculty to advise. And discussion sections should be scheduled close to dinner so that conversation can spill over into the less formal setting.

On other topics, Light’s findings come as similar sets of simple observations on what students want, coupled with suggestions for boosting satisfaction and performance: Extracurricular activities are surprisingly important to satisfaction and do not significantly take away from academic performance. Time management skills are closely related to academic success. Diversity adds to the quality of the educational experience.

These observations in turn mold educational philosophy—and more importantly for Harvard, probably Summers’ philosophy.

Light readily admits that despite the apparent simplicity of his recommendations, in practice they often are far more complicated. But he maintains that his type of research can be useful even when schools are operating under constraints.

Limited resources mean that not every class can be the intimate seminar that students indicate is so satisfying, he says. In fact, more small classes and a fixed faculty size mean more gigantic lecture-style classes, as the middle-sized classes are forced out. Light remembers a dean from a small liberal arts college being appalled at the size of some of Harvard’s bigger classes that weigh in at more than 800 students.

Light answers this criticism with his data, which show that when students were asked to describe the size of their most substantive class, the one in which they learned the most, the majority of the students fell on the two ends of the spectrum.

“Roughly half of the students described a class that was very small, and roughly half a class that was very large. Almost no one described a class that was in the range of 50 to 60 students,” Light said.

The small classes succeed because of the one-on-one contact, and the large classes are good enough to succeed without it. Thus, Light says, even with medium-size classes ballooning into gigantic ones, it is possible to strike an acceptable balance.