Data Collection in Extraordinary Times: Thoughts on Universities Transitioning to Online Classes

ONOZUKA Yuki
Fellow, RIETI

In an attempt to slow the spread of COVID-19, many universities have canceled commencement and entrance ceremonies and postponed the start of the new school year. As no one really knows when the situation will return to normal, some institutions have begun conducting classes online. Most universities in Japan had previously never offered online classes, so some people may see our current predicament as an opportunity for increasing online class offerings for the future. I personally expect that the partial transition to online classes will be effective in both reducing the burden that classes place on university faculty and ensuring that they have time to conduct research.

Considering such circumstances, my initial intention for this column was to survey prior economic research relating to online education by universities. After establishment of the Coursera and edX platforms in 2012, Massively Open Online Courses (MOOCs) attracted considerable attention among economists, who have conducted studies on online classes and online education (see McPherson & Bacow [2015] for a survey). However, the current online classes being conducted to prevent the spread of COVID-19 involve different circumstances and purposes than the typical online classes mentioned above. It seems that this previous research may not necessarily be helpful in examining the current movement.

Asynchronous versus Synchronous Learning

For the MOOCs mentioned earlier and distance learning conducted by universities in other countries, university faculty have adopted a method whereby each student may study the material online when he or she prefers. For example, MOOCs allow students to learn by watching a video of a lesson that has been recorded in advance. This method of delivering classes is referred to as "asynchronous." Of course, in the current situation as well, Japanese universities are able to provide asynchronous classes. However, we often hear of cases where the teaching faculty are using Zoom or another teleconferencing software to offer "synchronous" classes, in which the class is streamed online in real time. Why have conventional online classes tended toward asynchronous delivery while universities are now offering synchronous as well as asynchronous classes in the current COVID-19 situation?

Let me here briefly categorize the advantages and disadvantages of asynchronous and synchronous classes. Some advantages of asynchronous classes are: (1) once a class has been recorded, there is very little additional cost necessary and (2) students may access the recording anytime they like. A disadvantage is that the instructor and students are unable to communicate in real time. Another disadvantage that might be mentioned is the cost involved in preparing and recording a class, including equipment and material preparation. Because students may not pose questions in real time during a recorded class, there is a high likelihood that students may be quite dissatisfied with the class, unless the recording is of a certain quality. In video creation, it is of course necessary not only to ensure that audio and visual quality (including written material that is displayed) is of sufficient quality, but also that key terms are well defined and that explanations are clear and comprehensible.

On the other hand, synchronous classes offer the advantage of allowing the instructor and students to interact with each other in real time. Nevertheless, they also have some disadvantages, such as (1) students have to attend the classes at a set time and (2) instructors must teach each class session, meaning that they have to even teach multiple sessions of the same class material. Synchronous classes also require considerable network bandwidth, meaning that if several instructors wish to livestream their classes at the same time from the same location, such as a school or company, the communications infrastructure may not be able to cope with the demand. Thus, even though "online classes" appear to be a single concept, the advantages and disadvantages are quite different depending upon whether the class is offered asynchronously or synchronously.

Difference from MOOCs

Traditional online classes are in an asynchronous format, while many online classes being conducted today are in a synchronous format. This is because that these two types of online classes involve quite different circumstances and purposes. The emphasis in the case of MOOCs and distance education has been on making efficient use of economies of scale. In other words, the desire has been to deliver classes to more people at an even lower cost. With asynchronous delivery of classes, universities are able to offer the class to each additional student at a very small increase in cost. Also, recorded classes may be accessed at any time, so people who work, live far away, or find it physically difficult to go to a university to attend classes for some other reason are able to take advantage of the asynchronous learning environment. With synchronous delivery of classes, it is difficult for people who work during the day or who live in different time zones to attend the classes. This has traditionally been the reason that most universities have made their classes available through asynchronous delivery.

However, in the current case, the students taking the online classes are the same students who would physically attend classes on the university campus. Those students are also all at home and have the leisure of attending classes as originally scheduled. In other words, there is no need to offer classes that are accessible at any time. As mentioned earlier, one of the advantages of an asynchronous class is its accessibility regardless of the time of day. In addition, universities do not really have the time to create high-quality recordings and most do not offer online classes as part of their usual curriculums, so they lack the necessary equipment in the first place. Furthermore, even if recordings are produced, universities are not sure whether they will be able to use these recorded classes even next year, to say nothing about the school years thereafter. Accordingly, in comparison to MOOCs and conventional online classes, the current situation offers few potential benefits for universities wishing to adopt an asynchronous style of delivery. For that reason, it seems that not a large percentage of universities have geared their energy toward conducting their classes in an asynchronous format.

Even More of a Reason to Collect Data Now

In making the transition to online classes now, there is concern about the type of influence this will have on student learning (credit acquisition, grades, etc.). In the case of asynchronous classes, prior research, such as the study by McPherson & Bacow [2015], could provide some frame of reference. However, there has been little or no prior research estimating the impact on student learning for synchronous classes at the university level. Previous studies may not really prove useful in predicting what sort of effects live synchronous classes have on student learning (Note 1).

Conversely, synchronous university class delivery has been made possible precisely because of our current situation, and this presents a magnificent opportunity to collect data for investigating the effects. Some may think that the previous dearth of synchronous class offerings indicates that there is little possibility of them being more widely available after the pandemic ends. Yet, only recently have synchronous classes become a relatively tangible presence, so it is my supposition that we do not possess sufficient knowledge of the potential which synchronous classes entail.

The comparison of synchronous and asynchronous class delivery might also be useful for developing an even better way of conducting classes online. For example, randomly separating students in a certain course into a group taking asynchronous lessons and a group taking synchronous lessons would enable us to examine the causal relationship of learning effectiveness for these two formats. One may surmise that both faculty and students are still in very much a process of trial and error trying to find the best learning format amidst much confusion. Nevertheless, so that our experience now may be put to good use later, it is important, precisely because we have the opportunity now, to collect data that we may analyze about how classes are being conducted.

Although the discussion has so far focused on university online classes, it may also be applied to other situations where data collection is important during extraordinary times. Since the first COVID-19 infections were confirmed in Japan, the number of people infected has continued to rise, dealing a serious blow to our economy and our way of life. But, just as the saying goes, "in the middle of chaos lies opportunity," and so we have the opportunity to take action now that we did not take or were unable to take during normal times. Innovative ideas will emerge as a result of focusing our resourcefulness on resolutely standing up to and taking on the crisis. Properly preserving experience in the form of data, analyzing it, and making use of it after a crisis will help us to improve our situation over the long run.

April 21, 2020
Footnote(s)
  1. ^ Research has been conducted in educational technology and other fields relating to online classes. However, to the extent that I have investigated, it appears that there is no research examining a causal relationship, not correlation, between online classes and student learning.
Reference(s)
  • McPherson, Michael S. and Lawrence S. Bacow (2015), "Online Higher Education: Beyond the Hype Cycle," Journal of Economic Perspectives, 29(4), pp. 135-154.

June 22, 2020