Crafting an efficient schedule of college course offerings means solving a complex puzzle. And more colleges these days are turning to algorithms to help reduce the number of classes that are either overloaded or full of empty seats.
A study out last week of about 200 colleges found that many course schedules are “unbalanced,” with 45 percent of courses analyzed filled to less than 70 percent capacity and 23 percent of courses classified as “overloaded,” meaning more than 95 percent full.
That inefficiency is having an impact on retention, the study found. The greater the inefficiency of the course catalog, the lower the graduation rate at the institutions analyzed.
The study was done by a company that sells software to try to analyze the efficiency of course schedules, Ad Astra Information Systems, so it has an interest in showing the problem is large and that it matters. And the institutions included in the study were the company’s customers, and had already self-identified as struggling with the issue of schedule efficiency.
Others have pointed out the issue in recent years. As a report a few years ago from the Education Advisory Board noted:
“With undiminished desire to expand enrollment, launch distinctive programs, or climb research rankings, more institutions are recognizing low space utilization to be a meaningful impediment to growth. For a decade, it was easier for the university to erect a new building than to convince faculty to teach on Fridays, accept smaller offices, or cede underutilized lab space. The downturn has changed that equation.”
These days, interest in applying data to solving the problem appears to be growing. Other companies offering tools to improve schedules include CollegeNET and EMS.
But even if an AI system can show college leaders where they need to create more courses, there’s still a bigger problem: The college may not have the resources to hire additional faculty to create those sections.
Educated Guessing
There’s always a bit of educated guessing in crafting course schedules. Colleges increasingly turn to analytics to try to make better guesses, says John Calahan, director of institutional effectiveness at Stephen F. Austin State University. For instance, historical data may show that most business majors take a course in American Literature. So college leaders can look at how many business majors they have to help decide how many sections of American Literature to offer.
Stephen F. Austin is among institutions that have tried to better align their offerings. After crunching the numbers, officials decided to add about 15 new faculty members to fill gaps and avoid overload in some subjects. The change ended up bringing in enough new tuition dollars to cover the salaries of those professors plus more than a million in additional revenue in the next academic year.
Calahan said that it’s too early to tell whether tweaking the university’s offerings has improved retention rates. But he says it has had an impact on plenty of students they were able to accomodate in several crowded courses that are required for certain programs. “If you can’t take English 131 this semester, then you can’t take English 132 next semester, so you start to get behind in your degree plan,” he says.
John Barnshaw, associate vice president for research and statistics at Ad Astra, led the company’s recent study. He said they recommend an “enrollment ratio” of about 85 percent in courses right as the semester begins to leave room for late additions.
Not everyone thinks that efficiency is the best way to measure the success of a course catalog. After all, some professors offer courses in their discipline to test out ideas from their research, and colleges are committed to offering a full range of subject, even ones that are not as popular.
Some faculty members at Stephen F. Austin are also worried that the algorithms would play too great a role in academic planning. One professor asked, “Is this software going to tell us how to run our department?” according to a recent article in Inside Higher Education.
