Suppose a college dean wanted to predict which first-year students would remain continuously enrolled at her institution for at least three years. She might look at the students' standardized-test scores, their study habits, or whether they live on campus. Those are all factors that are known to be associated with retention rates.
But she might also try asking first-year students a simple question: Do you like it here?
In a paper presented here on Wednesday at the annual meeting of the Association for Institutional Research, two graduate students at the University of Maryland at College Park said that students' enrollment patterns at their institution were strongly predicted by how they answered a survey question in the eighth week of their first semester.
That question, which is part of a Beginning Student Survey that is regularly administered at Maryland, reads as follows: "At present, your general attitude toward the University of Maryland is ..." followed by a five-point scale that ranges from "strongly negative" to "strongly positive."
Fledgling students' answers to that simple, banal question turned out to be strongly associated with their odds of dropping out or transferring away from Maryland over a six-semester period, according to the study that was presented here. If their attitude toward the university at that early date was positive, they tended to stay; if it was strongly negative, they tended to leave.
The question had stronger predictive power than more-familiar variables like students' self-reported study skills or their involvement with student organizations.
"The simple message here is, Attitude matters," said Jessica Mislevy, who wrote the paper with Corbin M. Campbell. Both are doctoral students in education at Maryland.
"The general attitude toward the campus plays a clear role," Ms. Mislevy said. "That suggests that students are able to detect very early whether a campus is a good fit for them."
Ms. Mislevy and Ms. Campbell looked at the experiences of more than 2,000 people who enrolled as first-time, full-time students at College Park in the fall semester of 2002.
Six semesters later, in the fall of 2005, 76 percent of those students had been continuously enrolled at Maryland; 12 percent were enrolled at Maryland but had "stopped out" for at least one semester along the way; 8 percent had transferred to other colleges; and 5 percent had dropped out of college entirely. (The authors identified students as dropouts if there were no current records for them in the National Student Clearinghouse database.)
Ms. Mislevy and Ms. Campbell studied dozens of different items from the Beginning Student Survey to see which ones predicted enrollment behavior. For both women and men, the single most powerful predictor was the "attitude toward the University of Maryland" question.
In response to that finding, Ms. Mislevy said, a campus committee on retention plans to encourage faculty members and dormitory resident advisers to have conversations with first-year students about their perceptions about the university. In cases in which students strongly dislike Maryland, the committee would like to know exactly why.
The authors also identified several other patterns.
Among women, several factors seemed to be associated with their tendency to stop out, drop out, or transfer. They were more likely, for example, to stay continuously enrolled if they lived on the campus.
Interestingly, women were more likely to "stop out" for a semester or more if they had higher scores on first-year survey questions such as "I am earning the grades I want" or "I've stayed motivated."
"In many cases, these are women who also reported that they lacked a clear future direction," Ms. Mislevy said. "Maybe these are women who are motivated, who are going full steam ahead, but who need to pause at some point during college to decide where to go with their lives."