Teaching assistants, adjunct faculty and online learning

March 7, 2014 Tony Bates

Lecture to hybridI am struggling these days with the issue of who should teach online courses, in terms of qualifications and status, and in particular, the issue of how to scale up credit-based online courses while maintaining or improving quality.

These questions are coming to the forefront because, through blended learning, practices that are common in face-to-face teaching come head to head with quite different practices in online learning.

What has made this an issue for me

Recently I’ve been involved in assessing proposals for funding for large-enrollment online credit courses. Most of the proposals have focused on using several/many teaching assistants working under a professor to provide the learning support. I’m also finding this model being increasingly used where institutions are moving to a hybrid model, combining both online and face-to-face components, especially where a former very large lecture-based course is being redesigned for hybrid learning. Even including the TAs, the instructor/student ratio is often 1:100 or higher for these large enrollment courses (in other words, the same ratio more or less as when the course was delivered solely through large lectures.) In the proposals, and in the reports I am receiving, there is usually no additional training for TAs about how to teach online, although in many – but by no means all – cases, they do get some kind of training in teaching face-to-face.

This is a problem for me, because I have always worked with a model for online courses where the instructor: student ratio has been under 40 for undergraduate courses, and under 30 for graduate courses. Scaling up has been handled by hiring on contract additional part-time adjunct or associate professors, either with a doctoral degree in the subject area, or with strongly related work experience. The adjuncts would be paid to take a short online briefing course on teaching online which sets out the expectations for online teaching. This was an affordable model because the additional student tuition fees would more than cover the cost of hiring additional contract instructors, once the course was developed.

However, this has been possible because most of the online courses I have been responsible for have been aimed mainly at higher level undergraduate students or graduate students. With both blended and online courses now being targeted at large first and second year classes, new models are being developed that I fear will not have the same level of quality as the ‘best practice’ online courses I have been working with.

Why this is not an easy issue for me

This is a particularly difficult issue for me to discuss for several reasons:

  • most of my experience is with fully online courses; when I have taught face-to-face, it’s usually been me on my own, and generally with relatively small groups of between 25 to 200 maximum
  • practices both for dealing with large face-to-face classes and with online classes vary considerably within each form of delivery, and from one institution to another, so making generalizations is fraught with danger
  • decisions about whether to use teaching assistants or part-time, contract instructors, are driven more by financial considerations than by best pedagogical practice, although institutions do their best to make it as effective educationally as possible once a model for TAs and/or adjuncts has been decided on
  • there are other factors at work besides money and pedagogy in the use of teaching assistants and adjunct faculty, such as the desire to provide financial support to international and graduate students, the idea of apprenticeship in teaching, and the supply and demand effects on the employment of doctoral graduates seeking a career in university teaching and research
  • there is no golden mean for instructor/student ratios in either blended or online learning. In the mainly quantitative/STEM subjects, much higher ratios are sustainable without the loss of quality, through the use of automated marking and feedback
  • MOOCs (rightly or wrongly) are giving the impression that it is possible to scale up even credit-based online learning at lower cost.

What follows then is tentative, and I’m ready to change my views especially on the evidence of others who have grappled with this issue.

My concern

My real concern is that the over-reliance on teaching assistants for online and blended courses will have three negative consequences for both students and online learning in general:

  • As with the large face-to-face classes, the pedagogy for online or blended courses will resort more to information transmission.
  • however, for the online or hybrid courses, student drop-out and dissatisfaction will increase because, especially in first and second year teaching, they will not get the learning support they need when studying online.  As a result, faculty and students will claim that online learning is inferior to classroom-based instruction
  • faculty will see online learning and blended learning being used by administrations to cut costs and over time to reduce the employment of tenured faculty, and will therefore try to block its implementation.

Why can’t TAs provide the support needed online if they can do this for face-to-face classes? First, I’m not sure they do provide adequate support for students in large first year classes, but I’m not in a position to judge. But in online courses in subject domains where discussion is important, where qualitative judgements and decisions have to be made by students and instructors, where knowledge needs to be developed and structured, in other words in any field where the learning requires more than the transmission and repetition of information, then students need to be able to interact with an instructor that has a deep understanding of the subject area. For this reason, I am more than happy to hire adjunct faculty to teach online, but not TAs in general (although there will always be exceptions). Furthermore this kind of teaching and learning (‘the learning that matters most’) is very difficult to do with a very large instructor/student ratio, although with good design and faculty training, we could possibly push numbers higher than 1/40.

One possible solution

I’m not sure there is an easy solution to this problem. Whether online or face-to-face, large numbers of students per instructor limits what is possible pedagogically.

Furthermore, in my view online learning works better for some kinds of students than others. Students in their first year of university or college are not the best target group. They are often young, have little experience of independent learning, lack confidence or discipline in their study habits, and indeed expect to be in a face-to-face teaching environment and want the social and cultural milieu that a campus provides. What we should be doing though in their first and second year is gradually introducing them to online components so that they slowly develop the discipline and skills required for successful online learning. This still doesn’t resolve the issue though of very large classes.

So here’s my suggestion for these large introductory courses of 1,000 students or more (this is not new – see the National Center for Academic Transformation‘s course redesign):

  • create a team to design, develop and deliver the course. The team will include a senior professor, several adjunct professors, and two or three TAs, plus an instructional designer and web/multimedia designer.
  • The senior professor acts as a teaching consultant, responsible for the overall design of the course, hiring and supervising the work of the adjuncts/TAs, and the assessment strategy/questions and rubrics. This though is done in consultation with the rest of the team.
  • Most content is provided online.
  • Students work in groups of 30, and each of the adjuncts is responsible for several student groups. Students do both individual and group work (e.g. projects, problem-solving),
  • Students participate in ongoing online discussion forums, under the moderation of an adjunct or TA
  • The senior professor meets for one hour a week three times face-to-face or synchronously with  a group of 30 students; this brings the professor in face-to-face contact with just over 1,000 students a semester; adjuncts where possible meet once a week with a group on campus or synchronously.
  • Adjuncts and TAs mark assignments, and the senior professor monitors/calibrates the marking between instructors
  • Now think of what could happen if this course was shared with other universities. Savings could be made on course development, but the delivery of the course would still need instructors at the other universities. So there would be some economies of scale from sharing, but not a very large saving, because the development cost is a small proportion of the overall cost. This does not mean that institutions shouldn’t co-operate and share resources, but this will not bring the large economies of scale that are often claimed for sharing online courses.

Whatever detailed design is done, these large courses should have a clear business model to work with, which basically provides an overall budget for the course, that includes the cost of tenure track and adjunct faculty and TAs, and takes account of the students numbers (more students, more budgeted money), but allowing the senior professor to build the team as best as possible within that budget.

The two elephants in the room

The above scenario works with the current system of allocating resources to different level of courses. But there are two factors that lead to the very large class sizes in first and second year that no-one really wants to talk about:

Elephant in room

  • the starvation of first and second year students of teaching resources; senior faculty concentrate more on upper level courses, and want to keep these class sizes smaller. As a consequence first and second year students suffer
  • teaching subsidizes research: too often tuition revenues get filtered off into supporting research activities. The most obvious case is that if teachers spent more time teaching and less doing research, there would be more faculty available for teaching. Teaching loads for experienced, tenured faculty are often quite light and as stated above, focused on small upper level classes.

Do a simple calculation: divide the total number of students by the number of tenure track instructors  in your institution, and that will give you an overall average instructor/student ratio for the university as a whole. So if you have 40,000 students and 2,000 full-time instructors , you have an overall instructor/student ratio of 1:20. However, then deduct 40% of their time for research, so that equals 1,200 full time equivalent, or a ratio of one instructor for 33.3 students. Then deduct another 20% of their time for administration and public service and that leaves 800 FTEs, or a ratio of one instructor for every 50 students. Even with this fairly generous allowance of 60% of their time for other activities, and WITHOUT adjuncts or TAs, in this large university there should be enough instructors to teach without having the absurdly large first and second year classes commonly found in such large universities. Add in adjuncts and TAs, and this ratio drops even further.

So don’t expect online learning to solve this problem on its own.

Your turn

I would particularly like to hear from the relatively rare instructors who have taught large classes both face-to-face and online. Do you share my concern about using TAs for distance or hybrid courses?

I’d also like to hear in general about experiences with TAs or adjunct/contract instructors as well on this topic.

Previous Article
Comment on Getting started in writing an open textbook by Clint Lalonde

Hi John, To answer your questions: 1) Yes it separates design from...

Next Article
CAID webinar: instructional design in Spain and the Netherlands
CAID webinar: instructional design in Spain and the Netherlands

 What:  The Canadian Association of Instructional Designers is...