A study of unit workload comparisons commissioned by the Faculty Senate showed a steady increase in the average workload per unit across majors since 2009. But two associate vice provosts who helped conduct the study were quick to stress the preliminary status of the study, noting some of the methods were “not very precise at all” and based on “only very imperfect tools.”

The results of the study are unlikely to surprise many current Stanford students. Presented at last Thursday’s Faculty Senate meeting, the study found that the value of a unit varies greatly between departments. Departments in the School of Engineering reported the highest hours of work per unit: both mechanical engineering and computer science reported more than three hours of work per unit, as compared to the 1.5 to 2.5 hours averaged by other departments.

After concluding a two-year study on undergraduate education last January, the Faculty Senate decided to investigate one area of student life that was not covered in the report: the undergraduate major. Faculty found, however, that a discussion about majors would be impossible without establishing a firm concept of the unit of credit.

As a result, the Office of the Vice Provost of Undergraduate Education (VPUE) launched a study last year of student workload per unit, using results from course evaluations. VPUE drew its data from nine different majors, which were chosen to represent a variety of academic disciplines.

In addition to the disparities between majors, the study indicated that workload per unit may be increasing across the board. Since data collection began in 2009, four of the nine majors studied reported approximately 15 minutes more being spent on coursework per unit each year. Workload also increased for the other five majors, although less steadily.

No departments reported decreasing workloads.

Those involved with the study were quick to note its preliminary nature and the subsequent difficulty of drawing any concrete conclusions from the data.

“One of our main findings was how difficult it is to get reliable data,” said Martha Cyert, senior associate vice provost of undergraduate education. “This report is quite preliminary, and was meant to inspire further conversations about some of these issues rather than provide definitive answers.”

VPUE compiled the study using data from the course evaluations students are asked to complete at the end of every quarter. On these evaluations, students report how much work they spent on work outside of class in five-hour intervals. The study averaged each interval response; for instance, a student who selected one to five hours would be marked as having spent 2.5 hours on work outside of class. It also focused on a few departments to represent course workload across a variety of fields.

According to Cvert, this method of averaging is “not at all precise.”

In addition, the evaluations do not provide information about how many units a student was enrolled in if a course was offered in variable units. This makes it impossible to know, for example, if a student reported 10 hours of work for a three-unit class or a five-unit class.

Another concern is the fact that results are average values for all courses within a department. If a department has one particularly large enrollment course, for example, CS106A, responses from that one course may have skewed the data for the entire department.

Variations amongst courses and professors may have affected the data as well.

While the data is rough, Shari Palmer, associate vice provost of undergraduate education, said the study could indicate the “relative quantity of time” spent on each unit in the representative departments. Cyert agreed, stating that the data “should be thought of only as a benchmark,” one that at most offers “a very rough comparison between courses.”

The University has largely attributed the higher number of hours spent on coursework per unit in engineering majors to accreditation requirements.

The engineering departments are much more akin to professional schools than other majors. Therefore, like medical schools and law schools, many engineering departments must be accredited to award state-licensed engineering degrees. The Accreditation Board for Engineering and Technology requires that strict standards in coursework be met–for example, a year of math or a full year of science. In addition, students must take courses within their specific field of interest.

As a result, units in higher-level engineering courses are “devalued,” according to Eric Roberts, professor of computer science and Faculty Senate veteran. Roberts said this was necessary so students would be able to take all the courses they need to graduate on time.

He also added that the data for engineering majors might be slightly skewed.

“In problem set courses, there are people who can easily take ten times as long to complete the assignment,” Roberts said. Those outliers raise the overall hours of coursework per unit ratio.

This suggests that variations in the value of a unit are inherent in the system.

According to Palmer, it is “never going to be possible–nor desirable–for the university to attempt to enforce a strict unit of credit.” Roberts agreed, calling this variation “a necessary feature of a system.”

“I don’t think we should aspire to having the same numbers of units for each major or the same number of hours of work per unit,” Cyert said. “I think we should aspire to offering the best possible education we can both within and outside of the majors.”

Still, she added that part of offering the best education possible is to understand what students are experiencing.

To that end, the University plans to continue studying undergraduate education. There is talk of changing the course evaluation to better assess how much time students spend on a course per week and the Senate’s Committee on Undergraduate Standards and Policies (CUSP) will discuss this report in November to decide how best to proceed in light of the study’s findings.