Last Friday, the Office of the Registrar erroneously sent out an email encouraging students to fill out their fall semester course evaluations on SCORE. Approximately three hours later, the system recognized its error, noting that there were “no ongoing evaluations” at the time.
Though this was just a hiccup in the system, the email was a reminder of the vast troves of data that the Office of the Registrar collects semiannually. Three months ago, the Office of the Registrar collected through SCORE 15,521 student responses rating the overall quality of courses on a scale of 1 (Poor) to 5 (Excellent). From this data emerges a complex picture of trends in ratings, with humanities, upper-level and smaller courses receiving the highest marks and quantitative and scientific, lower-level and larger courses the lowest.
What factors cause some courses to be higher-rated than others?
To determine these factors, The Daily Princetonian tallied the number of high- and low-rated courses within each department in the humanities, social science, engineering, interdisciplinary and natural science divisions on SCORE and then looked for commonalities between those courses.
Data for freshman seminars and writing seminars was not included because the courses were open only to one class. A low-rated class was defined as having an overall quality of course rating below 3.00 (Fair) while a high-rated class was defined as having an overall quality of course rating at or above 4.50.
The results showed that the humanities had the highest percentage of high-rated courses and lowest percentage of low-rated courses out of all the divisions. Of the 278 humanities courses listed on SCORE in the fall of 2011, 34.9 percent, or 97 courses, were high-rated and only 2.9 percent, or eight courses, were low-rated. The interdisciplinary division, which includes the cultural studies programs, also earned high ratings, with 14 out of its 55 courses being high-rated and only two being low-rated.
The social sciences had more variation. Although 36 out of the 140 courses listed on SCORE — 25.7 percent — were highly rated, 17 — or 12.1 percent — were low-rated. This was the highest percentage of low-rated courses across all the divisions for which data was available.
Meanwhile, engineering and the natural sciences had fewer high-rated courses than the humanities but fewer low-rated courses than the social sciences. While 21.2 percent of the 123 courses in engineering were high-rated, 9.8 percent were low-rated. The natural sciences followed a similar trend: 14.0 percent of the natural sciences classes earned a high rating while only 5.3 percent earned a low rating.
The Humanities’ Success
Professors interviewed for this article suggested that the high ratings for humanities courses could be explained by their flexible structure, small class sizes and the lack of competition to concentrate in humanities-based departments.
Gideon Rosen, a philosophy professor and the chair of the Council of the Humanities, speculated in an email that course structure plays a large role in the ratings, noting that humanities classes are “less rigidly structured.” Rosen explained that humanities classes generally require fewer prerequisites and textbook readings are rare, which he said improves the teaching because the course has been developed “from the ground up.”

Rosen also argued that the open-ended solutions may bolster interest in the humanities since they allow students to engage in scholarly disagreements at an early stage in their academic career. Rosen added that this separated the humanities from the natural sciences.
“That sort of experience is very different from the experience students have in the sciences, where the student has to master an enormous amount of established material before she's in a position to understand the unsettled bits,” he said.
Within the humanities, East Asian Studies is among the highest performing departments, with an overall quality of course rating of 4.48 for the entire department and nearly half of its 30 courses rated on SCORE above 4.50.
Professor Benjamin Elman, the chair of the department, attributed the success of EAS to its small class sizes, which he said typically ranged from eight to 14 students. He added that part of their high ratings can be traced to the textbooks authored by the department and the language credentials of the faculty.
Elman explained that the department does not hire graduate students for language classes and many instructors have advanced degrees in their languages and related fields.
Elman also explained that he thought humanities courses generally may be more highly ranked than other divisions because of the grading distribution of entry-level humanities courses.
“If you have very large student bodies, the departments have a very high threshold for those they give A’s and B’s to,” Elman said. “So if you have a couple hundred students, there is more likely to be a curve and more likely to be a testing of people into the majors.”
He said there is a “sense of inclusion” in the humanities, he added, that makes the smaller departments friendlier.
Challenges in the quantitative-based courses
The success of the humanities can perhaps be viewed in a larger framework: Of the 45 low-rated courses listed on SCORE from this past fall, 29 were quantitative and scientific in nature, including many engineering, natural science and math-based social science courses.
The economics department had seven classes with a rating below 3.00. Six of those seven were large introductory or intermediate-level courses. By comparison, the department had 14 high-rated courses, all of which were small, specialized 400- and 500-level classes.
Economics lecturer Swati Bhatt, who taught ECO 300: Microeconomic Theory, said that she thought the academic adjustment of underclassmen to intermediate classes partially explained the low ratings of non-advanced courses.
“My sense is that a lot of the students in the intro and intermediate courses tend to be freshmen and sophomores,” Bhatt said. “While they may have had some preparation in economics, the level at which it is taught here in our department is quite high. So the difficulty of course tends to impact their evaluations.”
Economics professor Henry Farber, who taught ECO 100: Introduction to Microeconomics last fall, said he believes that teaching introductory courses is often more challenging than teaching upper-level courses because of the diversity of students’ backgrounds in economics.
“Sometimes you strike that balance well and sometimes you don’t,” Farber said. “When you teach a more heterogeneous group, it is harder to please everyone.”
He added that higher-level courses attract students who are likely to be more interested in the material.
“Let’s face it: In a big intro class, some students are more motivated than others, whereas in a small, upper-level class, the students are self-selected to be interested in the material,” he said. “That’s a huge difference — when the students come in predisposed to like the material, the professor is going to get better ratings.”
Assistant Professor of Economics and Public Affairs Emilia Simeonova, a visiting lecturer from Tufts University, said she thought students are more motivated to take higher-level courses because they are offered less frequently, which improves the ratings of those courses.
Like Farber, Simeonova said she believes that introductory-level economics courses present unique challenges for instructors — a reason, she said, that they are often taught by the “best and most experienced teachers” in each department.
“You cannot afford to lose people halfway, or have people get disillusioned or thinking that ‘Economics is just terrible, and we never want to touch it again,’ ” Simeonova said. “That’s something that no economics professor would ever want to hear.”
Science and Engineering Courses
The challenges of teaching quantitative and scientific courses extend beyond the realm of the social sciences to the E-Quad.
Among the engineering and natural science departments, the mechanical and aerospace engineering department had the highest number of low-rated courses, with three classes rated below 3.00 in overall course quality.
Professor Michael Littman, the MAE undergraduate departmental representative, said he is not overly concerned by the low ratings, noting that the department has strong enrollment.
“If you want to look at how a department’s doing, one measure is the evaluation, but a big one is numbers,” Littman said. “In my 30-plus years at Princeton, classes are bigger now than they’ve ever been. We’re on a growth phase, and frankly I don’t see that changing.”
MAE has adapted its curriculum to compensate for its increased size. In some areas, the department has increased the numbers of courses students can take to fulfill requirements. For example, students are now allowed to take design courses in other departments — such as electrical engineering and civil and environmental engineering — to satisfy the MAE mechanical design requirement. Littman said that this has decreased the class size of the previous prerequisite.
“The problem is, we have a limited size faculty,” Littman said. “If the department maintains its increased size over another several years, there may be a case for increasing faculty size. For now we have to live within our means.”
Littman also said that the low ratings for quantitative and scientific classes may be triggered more by the presentation of the material rather than the material itself.
“Some of the most difficult courses have material that’s the most opaque, and I really respect that students often will give high rankings to rough courses,” Littman said.
Littman added that quantitative classes may also score lower because students who perform objectively poorly in the class tend to give the course a lower rating.
“Quantitative courses have numerical ways to judge if students get it, so scores tend to be lower under those circumstances,” he said.
Cynthia Menkes, the undergraduate program coordinator in the electrical engineering department, shared Littman’s perspective that differences in course structure impact course ratings.
“[Quantitative and scientific classes] present different challenges to students — lab work, projects — challenges not normally encountered in straightforward lecture classes,” Menkes said.
Acting Vice Dean of the School of Engineering and Applied Science Claire Gmachl noted that she thought the material is harder in these types of classes, which causes students to give them lower ratings.
“Part of it might be — a lot of these quantitative courses are required courses, and required courses tend to score lower than electives. But unless you take them, you can’t become a good engineer,” Gmachl said. “There’s a sort of general understanding that these are really hard courses to teach. It’s harder to teach, it’s harder to understand, so the satisfaction is lower.”
The disparity between required and elective courses is reflected in the trend that upper-level courses often rank higher than lower-level ones. Upper-level classes are more likely to be electives than required classes.
“Upper-level courses are taken by people who really want to be there. In the lower level, a lot of people are taking them for distribution requirements. Not everybody wants to be in that class. It makes people unhappy,” said professor Mark Rose, the director of undergraduate studies in the molecular biology department.
Rose said it is common to find disaffected students in lower-level courses.
“People come into [biology] with certain expectations, and I think this is true of all the science departments,” Rose said. “People try to figure out what to do in the long term, discover an interest in high school, then take the college-level course and lose interest, as it’s more challenging and frustrating.”
Menkes also explained that part of the high ratings for upper-level classes is attributable to the presence of graduate students in upper-level classes, who tend to rate the courses highly, she said.
Gmachl said that there are particular challenges to teaching introductory engineering classes and that the teaching methods are different from those in the humanities.
“You aren’t taught engineering in high school, so expectations might vary across a very large spectrum,” she said. “And then preparedness — how well your prior schooling prepared you — makes the degree of difficulty very different for different students.”
Gmachl also emphasized how the University aids faculty who struggle with their teaching, with everything from McGraw Center luncheons with distinguished educators to a program in which junior faculty in engineering are mentored by an experienced professor in SEAS.
“There are a lot of safeguards built, a lot of resources you can tap into to learn how to teach better,” she said.
Models for Success
Although quantitative and scientific courses were among the lowest-rated this past term, the computer science and electrical engineering departments bucked the trend and stood out with a number of highly rated courses.
Menkes, the electrical engineering department program coordinator, said that her department’s small size was one of the reasons why its courses are consistently highly rated. Of the 22 courses listed on SCORE last semester, half of the department’s courses were highly rated.
“Our students get a lot of one-on-one interaction. We’ve had reduced enrollment … the same as ELE departments across the country,” Menkes said. “We’ve always had a good ratio, even with higher enrollment, so for students here during lower enrollment it’s a great experience for them.”
Gmachl agreed that small courses have advantages.
“It’s easier to teach; you have a better understanding of where each student is,” she said. “You’ll also probably have less of a spread. There’s sort of a class size where you know your students by name, and that’s a different way of being taught than if you’re one of a big number.”
The computer science department performed well in its upper-level courses, with six out of its 19 courses earning a high rating. However, the lower-level computer science courses like COS 126: General Computer Science are well-known for their success and enroll hundreds of students each term.
David Walker, the COS department’s representative for undergraduate affairs, noted that in his department, the faculty usually teach a course at least three times in a row. The professors, he said, adapt to the feedback they receive.
“We have certainly put a lot of effort over the years into making our intro courses as interesting as we could,” Walker said. He explained that by making COS classes more relatable to non-majors, the introductory classes have become more appealing.
He said he considered computer science a special subject, in the sense that everyone uses computers to crunch their own data and analyze their own problems.
“In the past it was always — you like computers, you’re a geek. You want to crunch 0s and 1s,” Walker said. “Today, everybody uses Facebook, everybody goes to YouTube, everybody can see the kinds of non-geeky social applications that computer scientists are building.”
This is the first in a set of two articles about data from SCORE course evaluations and the role such feedback plays in academic life at Princeton. The second installment will be published on Wednesday, April 4.