How Much Do They Really Teach?

A lot of dollars are riding on how many courses professors in the University of North Carolina system teach (or how many they are perceived to teach). Roughly half of the UNC budget consists of professors’ salaries.

UNC faculty supposedly taught an average of 3.7 classes in the Fall semester of 2012, according to the Delaware Study of Instructional Costs and Productivity. The Delaware Study is used to determine the official workload statistics at UNC schools and many other colleges and universities across the country.

That 3.7 figure seems to be too high, considering that most of the university system’s faculty work at large research institutions where they are expected to teach an average of 2.0 or 2.5 courses per semester. A closer look suggests that it may be not only inaccurate but perhaps deliberately so.

One glaring reason why the Delaware Study’s figures for the UNC system are likely to be an inaccurate measure of actual teaching loads is the way part-time professors are handled. The 3.7 average courses taught is for “full-time equivalent” (FTE) professors. This includes not just the tenure-track professors (used as a term that includes tenured professors as well), but full-time non-tenured lecturers, part-time teachers known as adjunct professors, and graduate students who also teach part-time. It defines an FTE professor as any combination of part-time teachers whose course loads add up to four. The teaching load of these FTE professors are then added to those of full-time professors and lecturers to arrive at an average workload of the department, school, or system.  That is what the 3.7 figure represents.

But this process is circular—and therefore provides no worthwhile information. If, by definition, the workloads of all adjunct professors and graduate students will always be converted to an FTE load of four courses, then the workload of one FTE professor will never deviate from four. Adding the FTE teachers to tenure-track professors will almost always inflate the average course-load of a department or university, as most tenure-track professors average fewer than four courses per semester.

Furthermore, it is more important for taxpayers and legislators to know the teaching loads of highly paid tenure-track professors who also have other duties, rather than the teaching loads of those who are strictly hired to teach. While the UNC system has not yet revealed the 2012 figures for tenure-track professors, the Pope Center has obtained the official tenure-track teaching load average for Appalachian State University for the Fall of 2011: it is 3.2 courses per semester. But is it correct?

Appalachian State is a particularly good representative example: while not a large research institute, it is the sixth largest school in the system, the largest of the non-research intensive schools. And its official FTE average for the Fall of 2011 was 3.6, just .1 over the system average of 3.5, while its 3.7 course average for 2012 FTE average was exactly the system average of 2012.

The Pope Center is conducting its own faculty teaching load study, and we were provided access to some official enrollment data at Appalachian sent to Delaware Study researchers for the Fall of 2011. This data had gone through a “grooming” process from the “raw” data found on the registrar’s website. The grooming consisted of adjustments made jointly by an ASU staff member and the department heads for use in the Delaware Study.

The Pope Center computed teaching load averages in strict accordance with the Delaware Study guidelines for both groomed and raw sets of data for a three-department sample at Appalachian State. We found anomalies in the university’s  groomed data for the Fall of 2011, such as courses mislabeled as lectures, which count toward the workload average, rather than as independent study classes, which do not. These anomalies inflated the Biology department’s official average for tenure-track professors from 3.0 courses to 3.8 courses.

It also drove up the average for our three-department sample (History and Economics were the other two) from 2.8 to 3.3—just above the school’s official 3.2 Delaware Study average for that semester. The difference between the official figure and our findings is not just a rounding error—a 0.4 increase in tenure-track faculty workloads strictly applied throughout the entire UNC system would mean savings in excess of $100 million.

The problems seen in ASU’s data and its excessively high average may not be exceptional within the UNC system. Although Appalachian State is the only school for which we were able to compare the official Delaware Study average tenure-track teaching loads with the Pope Center findings, it is somewhat middle-of-the pack or lower when it comes to the difference between its official FTE average and its legislated FTE average teaching load standard, as can be seen in the chart for the Fall of 2012:

UNC System Claims Its Teaching Loads (FTE) Are Much Higher than Required

Category Campuses Legislated Standard Courses per Semester Official DE Study* Courses per Semester (2013) Difference
Research I NC State 2.0 3.2 1.2
  Chapel Hill 2.0 3.0 1.0
Research II & East Carolina 2.5 3.6 1.1
Doctoral A&T 2.5 3.4 0.9
  Charlotte 2.5 2.9 0.4
  Greensboro 2.5 4.2 1.7
Master’s Appalachian 3.0 3.7 0.7
  Fayetteville 3.0 4.0 1.0
  Central 3.0 4.2 1.2
  Pembroke 3.0 3.9 0.9
  Wilmington 3.0 3.7 0.7
  Western 3.0 3.1 0.1
  Winston-Salem 3.0 3.4 0.4
Baccalaureate Asheville 4.0 4.3 0.3
  Elizabeth City 4.0 4.4 0.4

*Delaware Study

The chart shows that some other schools have a much greater gap between their legislated averages and their official averages than ASU’s 0.7 for 2012. Particularly questionable is UNC-Greensboro’s gap of 1.7 courses per semester for 2012.

Based on our review of UNC-Greensboro teaching loads, its official average of 4.2 courses per semester defies credibility. The Pope Center included two Greensboro departments in its study; for the Spring semester in 2013, the sociology department averaged 3.1 courses, while the nursing school averaged 2.2 courses. Though the Pope Center calculations are for tenure-track professors only, it is mathematically impossible for even an infinite number of part-time professors—whose average is 4.0 by definition—to raise the average above 4.0 (teaching loads don’t change drastically from one semester to the next). There are also not enough full-time non-tenured lecturers to significantly alter the numbers. Something seems very amiss.

Another area we explored was how teaching loads changed over the past few years. This is exceedingly important, as large cuts in state appropriations that began in the Fall of 2011 caused an outcry among UNC officials about the potential harm to the academic mission. We used the same sample of 14 departments from 7 different campuses, in a wide variety of disciplines, from the Spring of 2011 (the last semester before the cuts) and the Spring of 2013. During that period, we found no change at all in tenure-track teaching loads—for 2011 it was 2.35 and for 2013 it was 2.34. This lack of change conflicts with the UNC system’s claim that teaching loads (FTE) increased from 3.5 to 3.7 in those same years (Fall semesters).

Furthermore, it does not appear that there was a wholesale flight to using more adjuncts for teaching to hold down costs. In fact, the number of tenure-track faculty in the Pope Center sample increased from 298 to 320 in the two-year period.

With both tenure-track teaching loads and the number of tenure-track professors roughly stable, it seems that there was quite a bit of money in the UNC budget that could be cut in 2011 without affecting faculty workloads. This absence of stress on the system also raises the question how much more of that “cushion”—including such non-essential spending as unfilled faculty positions and unproductive staff jobs—is still in existence. In one eyebrow-raising example, at UNC-Chapel Hill Chancellor Carol Folt hired six new Title IX compliance officials when the work was previously handled by one part-time administrator.

The issue of UNC teaching loads needs further exploration. The official figures claimed by UNC are considerably higher than both legislated expectations and our study—conducted with as strict adherence to the Delaware Study methods as possible—and the data sent to the Delaware Study by the only school we could observe has glaring inconsistencies. Both of those situations raise the question whether the UNC system is providing incorrect information to legislators and the public.

A study using independent outside researchers under the direction of the Board of Governors, the legislature, or the governor’s office would be best to ensure that the public is not being fed incorrect information. Doing so could save taxpayers tens of millions of dollars without affecting the quality of instruction.

Additionally, it may be time to add a new wrinkle to the legislated standards for faculty working loads by differentiating according to academic disciplines as well as according to the types of institutions. Certainly in some fields—the humanities, especially—research is less likely to have important repercussions for the rest of society, and in those fields the teaching loads could be higher. In the university system’s own funding formula, teaching a course in the humanities consumes fewer resources—largely the instructor’s time—than does teaching the physical sciences. There are great savings to be had by a small increase in teaching loads in the humanities and some other subjects at the six large research institutions—likely into the tens of millions of dollars.

Of course, with so much money riding on state officials’ perceptions of UNC faculty workloads, there is an incentive for UNC officials to inflate their averages. Those with authority over the system—whether legislators or Board of Governors members—need to make sure that doesn’t happen. The way to do it is to conduct a more objective, more transparent accounting of faculty teaching loads that will include a critical examination of the data.

(Editor’s note: We are still waiting on some public records requests we made in April from UNC-Charlotte and Fayetteville State University. Their response could make slight differences to our computed averages for the entire UNC system, which were stated in the article as 2.34 for the Spring of 2013 and 2.35 for the Spring of 2011.)