Monday, March 30, 2020

Thoughts on Bridging the Social Distance in Online Learning and Ideas Related to Doing That

Before getting to the substance of the post, I want to note that the premise is based on the assumption that universities will be offering totally online classes again this summer and on into the fall, possibly in spring 2021 as well.  Even if the curve does finally flatten, with the limited testing that is currently available and looks to be available in the near future, we literally will have no idea of the number of asymptomatic carriers of the coronavirus, and thus will have no way to determine the health risks involved in a return to face-to-face instruction until a vaccine is made widely available.  If anything, this piece by Fareed Zakaria in the Washington Post makes the forecast about staying with online instruction seem all the more likely.   Clearly, now what's being done in online instruction is purely a reaction to an emergency situation.  Starting with this summer, however, there will be some lead time in which to anticipate the instruction will be online.  So it makes sense to ask what improvements can be made, rather quickly, so the quality of instruction rises.

My focus here is on undergraduate instruction only and to keep matters simple I will only consider classes that are taught as lecture-discussion, in moderately sized classrooms, or where the lecture is in one section taught in a large auditorium and the discussion is in a different section, typically taught by a TA.  I want to begin with the following ideas, which you might call learning principles, or stylized facts about learning, or imperatives to take into account when making modifications to how a course is taught.

  • Every Learner Must Express Their Own Formative Thinking - How such expression occurs might very well vary from person to person.  One would expect that extroverts would do this in conversation with others while introverts might have purely an internal conversation, but then perhaps express the thinking through some creative activity.  For example, I write blog posts as "think alouds" to work through some issue.  If I've hit my mark, it helps others to think through those issues for themselves. Also note that sometimes people behave against type - extroverts may engage in deep reflection for expressing their thinking while introverts may engage in thoughtful conversation with others. 

  • The Shy Student Problem - Many students are reluctant to speak up in the face-to-face classroom.  They won't raise their hand even when they have a question to ask.  Likewise, they won't raise their hand when the instructor poses a question to the class.  Some of these students genuinely prefer to listen to their classmates and to the instructor, rather than contribute to the discussion themselves.  Many of these students are afraid of looking stupid in front of the instructor and their classmates, so sometimes the preference for listening is a dodge to conceal this fear.  This is not new.  I learned about it in spring-summer 1996 when Cheryl Bullock and I interviewed many SCALE faculty.  Addressing the shy student problem was the primary reason that most of these instructors were willing to try going online. Of course,  there have been many factors since then which impact how students now behave in the face-to-face classroom.  If my own recent teaching experience is any indicator, the issue is more acute now than it was in the mid to late 1990s. 

  • People Tend To Be More Open When Typing On A Screen - This too is not new. Absent the facial cues, it is easier to think of what to say because there won't be quick feedback about saying the wrong thing.  This is why one approach to the shy student problem is to have students do writing online of some sort. 

  • Students Tend To Be More Open With Others Their Own Age - These kids, on the path to adulthood, live in a hierarchical world.  Those in authority, mainly parents and teachers, are older than them, and learning to please the authority figure may mean denying oneself of true motivations and feelings.  It is easier to express those sometimes hidden thoughts with peers, who likely are thinking along similar lines.  Further, there are generational/cultural differences based on common experience (think about music or online games) that encourages communication within generations and blocks communication across generations.

  • The Most Basic Pedagogy Is Imitation Of Expert Performance - We refer to it as the master-apprentice model or, more colloquially, as monkey-see monkey-do.  Indeed, much of what caring instructors aspire to do is provide a good model for their students.  Yet sometimes instructors who are experts in their fields are blind to obstacles that novice learners face.  Those who are somewhat experienced yet still novice might be better teachers this way because they have a better sense of what someone new to the subject is going through. 

  • Is The Innovation Specific To Online Or Can It Also Be Deployed Face-To-Face? Nobody knows now when instruction will return to what it was or if that will happen.  Expectations about that will vary.  If an instructor is to make changes in how the course is taught, thinking of that as a kind of investment, the instructor would like to get good return on that effort.  It stands to reason that modifications which can be deployed either online or face-to-face are more likely to elicit greater instructor response.  
The core idea is to resurrect the notion of Inward Looking Service Learning (INSL), where more experienced students provide service to the university by helping other students who are less experienced, and conceiving of that as both a productive activity, meaning the inexperienced students really do benefit from receiving the help, and an educative activity for the experienced students, where it falls into the category - learning by teaching.  In the first year of this blog I became enamored with this idea and thought then that the university should embrace it fully.  Near the end of this post I will critique my own thinking then, so while I still think this is a good idea, it is no panacea, especially if not accompanied by other modifications to supplement it. But I first want to articulate the idea on how it might be implemented at the individual course level, which is described in the second of seven posts on INSL. The experienced students would serve as peer-mentors in a course they had already taken.

The idea included taking a very old concept, the study group, which typically would be formed by some of the current students themselves, with it being an opt in for those who became study group members, and making the following modification.  The instructor would break up the class into study groups where each student enrolled in the class would be assigned to some group. Each group would also be assigned a peer-mentor, who would serve as the group leader.  Each study group would meet online in synchronous fashion,  with the agenda for the group meeting being determined by the course work that needed to be covered during the session. The peer-mentor would encourage participation by all group members, address outside-the-group meeting, if a particular member seemed to be struggling more than a little bit, and try to keep the group on track so the meeting remains functional, although in the current circumstances it might be appropriate to allocate some time to other diversions as people who are abiding by the social distancing requirements might need some non-academic interactions with other people that are purely fun/social or are expressing their fears about whether society will rebound fully or not.

The instructor, in turn, would serve as a mentor of the peer-mentors, by having an online session with them where they could share their experiences and where the suggested agenda for the the next group meeting would be proposed.  The instructor might also arrange to visit with individual groups, perhaps for a fraction of the full group meeting, just to get a sense of how things are going and to demonstrate to the students that the instructor has an interest in how students are learning with this approach.  (In large lecture classes, it is the graduate student TA who would perform this function.)  Otherwise, instructor produced content would be delivered to students asynchronously and students would review that content as well as complete any assignments that were due between group meetings.  So those meetings would partially be for taking a reflective look at the work the students had already done and partially for getting the students ready for what would come next.

One would want to evaluate whether the approach works for students in the class.  Will they find it a boon, have true personal contact with a peer-mentor who can help them learn?  Or will they see it as a way for the instructor to become even more distant than happens in the usual mode of instruction?  It only makes sense to perpetuate the practice if most of the students are in the former situation.  One also has to consider how the peer-mentors benefit from the arrangement. As I mentioned previously, there is real learning for them from doing this work, which might be categorized as learning communication skills and learning leadership skills, as well as possibly learning the subject matter of the course in a deeper way than they had during the time they were students in the class.  There is an issue about getting college credit for this learning.  If there were an institutional embrace of the practice, that would have be to addressed in a systematic fashion.  In the current situation credit might be given in an ad hoc manner, letting the peer-mentors receive independent study credit, possibly using the peer-mentoring as a gateway for the instructor to actually supervise independent study projects that the peer-mentors do.  But if too much of that is deemed necessary, it will be self-defeating, because it will up the instructor workload beyond what is reasonable.

What is clear is that the peer-mentors need to be paid, perhaps on an hourly basis, or maybe via tuition reductions that are granted to such students.  So, one needs to ask where the money will come from to make such payments.  In the INSL posts from 15 years ago, I went through some effort to show how an institutional embrace could be made self-financing. Here we don't have that, so something else is needed.   And now I will make reference to my previous post, where a worry was expressed about graduating seniors entering the job market at this impossible time.  If the university had the wherewithal to do this, should it encourage these students to defer graduation till after the health risk has passed?   My sense of things is that this is a sufficient emergency situation to warrant spending down the university endowment substantially and to really lean on major donors (the ones who have buildings named after them) to provide additional funding for hiring undergraduate students.  One might also lean on well-paid faculty and staff to make contributions to the university. Without this, the idea is a non-starter.  With it, you can then ask, just how much of this sort of thing can we do?

I promised a critique of the INSL approach and I want to offer that, tie it in with what came previously, and then close.  I came up with INSL because in the late 1990s and early 2000s I used undergraduate TAs in my intermediate microeconomics course (they held online office hours) and got reasonably good results from that practice  In turn, I adopted that approach from some instructors in Engineering who had been using undergraduate peer-mentors well before I started to do it.  This leads to the first critique.
  • Innovators and Early Adopters among the instructors make the innovation shine.  The same innovation when deployed by majority instructors can seem rather dull.  This is my conclusion from the main innovation I oversaw as the Director of the Center for Educational Technologies, when the innovation was getting a traditional class and giving it an online component, so the class became Web enhanced.  The traditional way of overcoming this issue is by having the majority instructor receive professional development support from campus, college, or departmental providers.  But that doesn't scale well and/or is a slow way to get interesting adaptations of the innovation.  The issue is whether in the current tough situation we find ourselves whether there is some way to quick start the learning of the majority instructors, so they behave more like early adopters in this instance. I don't have an answer for that here, but it is worth thinking about. 
A second issue concerns the prior attitudes towards group work.  The majority of students are not neutral on the matter.
  • Diligent students will report on prior group projects that there are other students who free ride on the efforts of the group and don't do their fair share of the work.   I have heard this sufficiently often when I query students about this that I take it to be the norm on campus.  Particularly galling here is when the group shares a common grade for the project.  If there were individualized grading then the free riding wouldn't seem as much as an issue to the diligent students.  In our context, however, it would be the peer-mentors who assigned the individualized grades.  Do we want to advocate for undergraduates to do that?  A flip issue here would happen if there is no direct grading for the study group at all.  Would the diligent students see the group as productive regarding their own performance in class on the work they do that does get assessed?   If so, they would participate vigorously.  If not, however, the study group might degenerate entirely.   Holding the study group together so all members participate and the discussions are valued by group members might be putting a very large burden on the peer-mentors.  What can help them be up to that burden?
A third issue concerns group meetings held online and that students may be more distracted when online than in face-to-face meetings.
  • Students are in the habit of multiprocessing when online.  The multiprocessing counteracts student engagement and thus lessens the quality of the group meeting.  Among the related issues that need to be worked through here are: (a) should group meetings be by video chat, audio chat, or text chat?  (b) should a group member be designated to take minutes of the meeting and thereby be absolved from much participation in the discussion?  and (c) does it matter what device students use for these group meetings, smartphone or laptop?  I do want to note that meetings can be recorded and recordings can be shared with the instructor, if necessary.  So if a student is under performing during the meeting, the peer-mentor can note this, share the recording of the meeting with the instructor and leave it to the instructor to deal with the student.  This is an example of the typical monitor/punish approach to labor shirking.  One wonders if there is an alternative approach that emphasizes carrots rather than sticks, so taps into student intrinsic motivation.  If this alternative does exist is that due to the subject matter of the course or to the pedagogic method?  There needs to be some unpacking in answering those questions. 
Now let me wrap up.  The Webinar approach to online classes is one that leaves most students as anonymous members of the audience, even when text chat is enabled and questions can be posed there.  Most instructors can't manage their own presentation and the text chat flow at the same time.  They will generally favor their own presentation and the online session will then not be very interactive.  So this post considers an alternative where students are in study groups and interact there.  The student groups are led by peer-mentors who have taken the course previously.  This has a lot of upside potential to engage students in the class.  Whether that upside will be realized, however, remains an issue.  In ordinary circumstances, one would do pilot offerings with the approach to learn what does encourage the upside to become the reality and how to avoid some of the pitfalls that can be anticipated in advance.  At present I'd like to see larger than a pilot offering, driven by those instructors who understand creative things need to be done very soon to address the quality of instruction issue.   Trying anything new entails taking on some risks.  If the risks are understood, mitigations can be taken to lessen them, if not to eliminate them outright.

Tuesday, March 24, 2020

Should students slated to graduate this spring delay that indefinitely?

I'm not in the know here.  I'm just thinking aloud.  Are companies that have made job offers to students already still honoring those?  Are students who have yet to get an offer in jeopardy that no offers will be forthcoming, indeed no interviews for jobs will happen as well.  What should students do in this circumstance?

And what should the university do to help such students in this very frightening time?

We might want to appropriate the acronym ABD and redefine it as All But Degree.  In the current circumstance, which I would term an emergency, students who are ABD might then become partly grad students and partly undergrad help employees - NOT SUBSTITUTES FOR GRAD TAS - but rather, in addition, to help make the online instruction more personal.  This might happen in the summer or fall semesters.  If the crisis has largely passed by the spring, they would graduate then.   The year between when they were scheduled to graduate and when they actually graduate would be something of a twilight zone.

If there's a better answer here, I'm all ears. 

Sunday, March 22, 2020

Reconsidering the Notion of Acceptable Risk

I found this opinion piece by David L. Katz provocative.  As I understand, the data so far show that mortality risk for people who test positive for Covid19 is low for people under 60, and then it rises, so that very senior citizens are at extreme risk, while those in their mid 60s like me are at risk, but it is still very likely that such people will survive the illness.  The piece argues that we should should target the at risk population for social distance practices.  The rest of the population should interact, let the disease spread, but acquire herd immunity as a consequence, which would ultimately defeat the disease.  This was a new argument for me and I'm still trying to process it, which is why I'm writing this post.

So, first, there is the question about the news we are hearing and other possible news we are not hearing. We seem to be overwhelmed with stories about those just tested who show infection.  When it is a public persona, that might be enough information.  But otherwise, is age of the person reported as well?  What of the prior health of the person?  Does that get reported too?  Conversely, we don't seem to be getting any news about people who were tested positive a month ago or even earlier than that who got sick, but then recovered and are more or less normal now.  Nor have we heard about when such a person is no longer contagious.  My conclusion, based on the availability heuristic, is that we are overplaying the risk from being infected, so most of us won't be able to entertain the argument in the Times opinion piece, as it goes against too much of what we believe to be true.

Then there is, how we as individuals as well as how we as members of organizations deal with risk, when the downside can be measured in money terms rather than in illness which might lead to death.  Here the perception of liability matters and that liability is tied to whether due precaution has been taken.  The decision making apparatus often drives to a solution that minimizes liability without caring about the potential upside risk, which conceivably could provide substantial benefit in the event it is realized. In other words, a full cost-benefit analysis might sometimes be consistent with absorbing liability, but organizations don't go that route out of fear of the downside risk.  In this, I'm thinking of my days as the main ed tech guy at Illinois, where the learning management system (we called it Illinois Compass) offered up the solution to risk of violations of copyright (an instructor using somebody else's materials without getting their permission first) and student privacy (a student's membership in a particular class is protected information that should not be made available to outsiders without the student's permission).  So closed container systems, which the lms exemplifies, became the vogue, irrespective of whether they encouraged deep learning by the students or not.  I have since come to view that what the lms does should be much more limited and that much of the online instruction should be done out in the open.  That better promotes learning.  But I feel like a voice in the wilderness on that one.

Finally, there is the issue of whether it is better to endure the downside risk rather than to try to avoid it, so as to lessen the fear of future such risks and make the person more resilient.  Here I'm thinking of pieces like What Straight-A Students Get Wrong by Adam Grant and The Overprotected Kid by Hanna Rosin. Getting bumps and bruises along the way is part of being a robust child.  Once in a while it will be worse than that.  Should parents protect the child against that downside risk?  Nowadays that seems the norm.  Yet we have a large fraction of college students who are having serious mental health issues.  Surely there is some linkage between the two.

Right now it seems to be younger people who are violating the restrictions of social distancing.  If they are living with older parent or their grandparents, that indeed might be problematic.  But if they are living on their own or with roommates their own age, maybe these violations will in the end be our salvation, as a species.

I'm no epidemiologist.  And I'm in the at-risk group.  But as an economist, seeing all these stories about how medical masks and virus testing are in very limited supply, I can make a reasonable case that in addition to hoping for dramatic increases in supply, we almost surely should be doing things to limit demand.  Yet I'm wondering whether our trying to be responsible precludes that.  It's something that needs to be thought through.

Friday, March 20, 2020

Student Created Content and Open Courseware

This is a kind of followup to my previous post, where I argued that instructors will need to exploit division of labor of some sort in getting their courses online.  But I want to take a different angle here - using students as producers of instructional content.

To make this a sensible solution there needs to be some forecast for how long online instruction will remain an across-the-board activity.  If we are to rely only on official pronouncements, those go only as far as the current spring semester.  What about in the summer and then the fall and perhaps even next spring?  I have not read about student internships/jobs for this summer but I'm guessing that a good number of them have been cancelled or will be cancelled and the prospect of students finding other work will be nil.  Indeed, this may be true even for those who graduate after the spring semester.  I want to repeat that I'm simply guessing here.  I have no information on this point.  But if the speculation is on the mark, then the demand for courses may be much larger than usual, as taking them would be a way to accelerate time to graduation or adding additional qualifications post graduation, and the students' opportunity cost of time will be remarkably low in this instance.  Further, if a vaccine is only first available during the spring 2021 semester, and its date of availability won't be known in advance, then this situation will persist into that semester.

Admittedly, the situation might be different for incoming first-year students, who opt for community college alternatives as a way to avoid paying higher tuition.  (That the university might itself offer tuition discounts while in this totally online phase is an interesting possibility, but I won't consider it here.  Perhaps I will write a subsequent post on that.)  There is complexity in this because if a student doesn't enroll at the university in the fall does the student retain a place at the university to be exercised when face-to face instruction resumes?  I haven't seen this matter discussed yet.  I'm sure it will come to the fore soon enough.

While many instructors will be scrambling this spring to get their courses online, the forward thinking among them will start to consider the possibility of division of labor.  If other alternatives don't avail themselves then these instructors might very well consider redesigning course assignments and course projects so that what students produce could potentially be deployed as instructional materials in the subsequent semester.  Pedagogically, this would be an instance of - you really learn something well when you have to teach it to others.  So the teaching idea would be that students would learn some aspect of the course deeply, enough so they can teach fellow students on that topic.  Indeed, 20 years ago, I had thought this sort of thing would have become a big deal and was fully integrated into the instructional process.

Such integration has not yet happened and, to be blunt, if students are producing course materials that are productive, in that the instructor does use them in a subsequent offering of the class, then this is work and the students should get paid for doing such work.   But the instructor won't have funds to pay the students and, indeed, if the approach required student payments it would not work. So I'd like to ask what it would take for the students to be willing to donate their creations as gifts to other students who might use them in the future.  That is the issue addressed here.

My belief is that this requires creating an ethos where all students donate such creations willingly. Such an ethos would be greatly facilitated if the instructor also donated course materials that the instructor created.  If this actually were to happen, then the creators would be interested in how the materials were subsequently utilized.  One of the motives that would drive the willingness to share is a sense that others will utilize the materials and get something out of them.

Based on my own experience, there can be two different types of sharing.  One you might call planned sharing, which would be with students who take the same course from the instructor in the future.  It's the planned sharing that would drive changing the class assignments in this way to produce useful instructional materials. But there is another type of sharing that might very well happen, which I would call serendipitous sharing.  Others, students and instructors elsewhere, find the content available online and access it.  Then the normal social media rating scheme of likes and dislikes directs these users to content they'd likely be interested in.

For the serendipitous use to happen two things must be done.  First, the content must be publicly available.  If it is in a closed container that precludes such use a priori.  Some years back I experimented with putting some of my content at (This included an experiment where I read my blog posts aloud, so they could be podcast.)  There was some serendipitous viewing, but only a modest amount.   I also have a profarvan channel at YouTube.   While some of the videos there have hits of the same order as the stuff at, there are other videos with 2 orders of magnitude greater hits than the stuff.  Admittedly, this is a small sample on which to draw conclusions.  But it is my belief that if one wants to promote serendipitous use of instructional content, then put it where users are likely to find it.  That matters, a lot.

Should serendipitous use count much (or at all) in making these decisions?   I will speculate here as I don't believe my past experience speaks to the question well for a few reasons.  I have not tried to share student created content publicly.  So I don't know whether there would be much serendipitous use of such content.  At present the use of this sort that I get is from students who are taking the course elsewhere, where they are finding their own course materials inadequate for some reason, so are looking to supplement them online.   I have only a little experience of other instructors adopting my content for their course.  The question then is whether instructors would do that if the material was available at scale.  Roughly 15 years ago there was intense interest in the Merlot Project, which aimed at promoting such reuse, via peer review of the instructional materials.  I wonder who uses Merlot nowadays.  I surely don't.

What I might do now, instead, is have my students screen some externally produced content and let them rate it.  The stuff that gets a rating of at least as good as the stuff in our class, with this coming from several students in the course, I would then look at and consider deploying the next time I taught.  If instructors at other institutions did likewise, this would give a mechanism through which quality content, whether produced by instructors or by students, gets redeployed for what I've been terming serendipitous use. Were such a process in place, then serendipitous content should count.

There is then a chicken and egg problem that needs to be solved.  There must be a sufficient volume of publicly available content already out there for this sort of serendipitous use to manifest.  My view is to be idealistic at present.  Make the stuff publicly available and findable now, in the hope of jump starting this process down the road.  This makes sense even if we return to on ground instruction eventually, as long as such instruction has a substantial online component thereafter. At the individual instructor level, I buy this argument.

At the institution level, it's a different matter, of which there are two main concerns, copyright ownership and the possibility of diluting the university brand.   If the university puts in real and substantial resources in producing online content, the copyright is likely held by the university.  The examples I've been giving above don't feature university contribution of that sort.  Then there is the question of why students would pay big bucks tuition at a well known university when they can get all the same content online for free.  In response, I'm reminded of the line from Good Will Hunting:

“You wasted $150,000 on an education you coulda got for $1.50 in late fees at the public library.”

Nevertheless, this sentiment might not be very reassuring to institutions that have been struggling to identify recurring sources of revenue.  Surely, they don't want to shoot themselves in the foot, if they can avoid it.  So they are very unlikely to go all in on sharing of online instructional content.  Yet, if you asked whether such sharing would be publicly beneficial,  I think a strong argument can be made that it would be, at least if the serendipitous sharing I discussed above would become a reality in this case.  Institutions need to come to grips with this argument, maybe not immediately, but soon.  And they need to work through as well, how the benefits from instruction are not the same as the value of the online materials used in instruction.  I'm not saying this will be easy, but it is necessary to consider this and come up with a reasonable response.

Thursday, March 12, 2020

The Next Bottleneck as Instruction Quickly Moves Online (Out of Necessity)

Some years ago I wrote a post called The Traffic Helicopter Theory of Management, which was aimed at more normal times, yet where a variety of blockages would emerge on campus in IT, so an appropriate response was to find a way to "re-route traffic" in order that more or less ordinary function could be resumed.  This was admittedly a tactical approach, so not as dignified as its more strategic alternative that takes a longer term view.  Yet it was nonetheless necessary, because one blockage after another becomes extremely demoralizing and once demoralized it is hard to accomplish anything, near term or long term.

The situation is anything but normal now.  Everyone is operating under a lot of stress for a a variety of reasons, not the least of which is the fear of the evident health risk that they themselves face. At the U of I, where spring break is next week and face-to-face classes have been cancelled after that and will instead be taught online, the official announcement of this went out yesterday evening, there have been readying activities done at the administrative level and in support units, but whether instructors themselves are ready is anyone's guess.  One might anticipate pushback from them along the following lines.

Luddites among the group are apt to claim the learning curve is too steep for them, especially within the tight time frame.  Those who already have engaged in Web enhanced (in the current course), or blended or totally online instruction (in some other course) likely won't be opposed to moving online in an emergency situation as there is now.  But they might still say there isn't enough time to get this done and they might be resentful that their spring break will have to be used to get up to speed for this purpose.  In other words, there very well might be two strikes against us before this online instruction even begins.  What things might be done to make all of this more do-able?

There are two general suggestions offered here.  The first is that in terms of credibility with the instructors, other instructors within the same department typically have more credibility than do support providers.  (This makes sense if how one does things online depends on the subject matter.  Somebody in the discipline can tailor the approach for that.)  If there are such instructors with substantial online teaching experience, it makes sense for one of them to assume a leadership role. On a provisional basis, the department Executive Officer could appoint such a person to an administrative position to coordinate the online efforts in the department.  The usual caveats apply about incentivizing such an appointment - other obligations would need to be reduced, some cash reward might be needed to induce the person to play this role, etc.  The person holding such a position would serve as mentor/friend with the instructors in the department who have to move their courses online.  They might do this in part by being a conduit regarding effective practice from those instructors who are making good progress to others who are struggling.  And they might serve as a broker with the support providers, so they have a point person in the department to interact with, rather than getting a flood of requests from each individual instructor.

The second suggestion is an old economics idea - division of labor.  It is why in the late eighteenth and early nineteenth centuries the factory system more or less eliminated home production.   In many cases today instruction can be likened to home production.  Each instructor does their own thing.  In a division of labor approach to an online version of the course, each instructor would do their bit only and what the instructors produced together would constitute the whole course, the production of which might be likened to the factory system.  It might be that the whole course is cloned several times, so each instructor can have the online course reside within the learning management system used on their own campus.  Or, possibly, there is a common host for all the instructors.  Let's consider a few examples to illustrate how division of labor in online instruction might work.

Perhaps the easiest case to envision is a large class that has a lecture section and then discussion sections run by TAs.  In the face-to-face version, each TA is a resource for the students in the TA's section(s).  Students in other sections get no benefit from this particular TA.  In the online version, the TAs become a common pool resource.  Some online activity is needed to enable that.  Some course restructuring might have to be done as well.   Moving the lecture online is pretty straightforward and won't be considered here.  We'll focus on the TA role, rather than the instructor role.

First, each TA should make a rather brief talking head video as a means of introduction to all the students in the course.  (There are tips on this and other suggestions that follow in a previous blog post on getting a class online in a pinch.). These introductions need to be posted in a commons area on the class Website so students can find them easily.  Next, if the weekly meeting between the instructor and the TAs is now conducted online, then after the first few minutes, where connection-to-the-the conferencing-system issues can be addressed and any confidential information can be shared, the rest of the meeting should be recorded with the recording posted on the class Website, so all students could have access to it.  This would be a way where online is better than face-to-face.  Students get to see how their teachers regard class issues and how the teachers aim to address those issues.  Third, if the discussion sections typically included a presentation by the TA at the beginning, there then needs to be common presentations to be delivered online to all the students in the course.  Each week one or two TAs would be assigned to make the online presentation for that week.  It probably makes sense for the more experienced TAs to do this earlier and the less experienced TAs to do this later, but it should be the instructor that develops this schedule and gets the TAs to buy into it.  Fourth, if another part of the discussion section is Q&A, that part must be handled differently when going online.

Imagine instead there are online office hours.  If the flow of students to those office hours were uniform regardless of the time, then the staffing could be determined to meet that flow as could the method by which students would interact with the TA.  In that previous blog post I was kind of down on using text chat in Zoom for an all class meeting, but it might be the right tool for office hours.  Further, chats can be copied and pasted into posts on the course Website.  So students who don't participate directly in the chat might be able to get their questions answered by reviewing the chats of other students.  With pooled online office hours, the total number of hours available to students should greatly increase.  Further, as it is well known that students are nocturnal, if (some of) the TAs are willing there could be office hours in the evening.  That likely would be popular with the students.

If TAs wrote their own spot quizzes that were administered in their section(s), that would need to be replaced with common online quizzes.  Let me leave off the issue of cheating till the next point and here only worry about logistics.  Online homework done in the learning management system has been around for quite a while.  Some instructors allow multiple attempts for homework and have a large time window from when it is made available to when it is due.  For a quiz to be considered different from homework, normally one would think of a single attempt and a limited time of availability.  I would opt for the former, but for the latter instead retain the large time window.  The possible logistics issues (e.g., the student might lose their network connection while the quiz is being conducted) favor the longer time window.  And that also lessens the stress on students, which under the circumstances would be a good thing.

The last part in moving the course online, and it is a biggie, is what to do about high stakes testing, which many large classes favor as the real means of assessing student performance.  Connected to that is proctoring to deter student cheating (keep your eyes on your own paper).  High stakes testing in an online environment is its own issue.  Under normal circumstances there are reasonable solutions, at some cost.  Here, however, I'm going to take advantage of my being retired and in no official capacity in making the following recommendation.  This is where you bite your lip and hold your nose.  Then treat the midterm and possibly the final too as a longer quiz that offers more points for right answers.  Of course, there are tools in the learning management system to deter cheating - randomizing the order of the answers to a question, randomizing the order in which questions are presented, and possibly other means to make it harder to cheat.  But they are far from perfect and a determined group of students will figure out a way to work around them.   Perhaps the campus will come up with a better solution.  If that happens, of course go with it.  However, don't insist on it.   We're already into plan B, with classes being taught online.  The key thing is to keep things do-able, or if you prefer, it is the lesser of two weevils

Now I want to move to other situations, focusing on smaller classes, so there is only one instructor per class.  To achieve division of labor here needs to have instructors who teach the same class, but at different universities.  In this next example, suppose these instructors already know each other. Perhaps they went to grad school together, or they previously co-authored a paper,  or became acquainted at a conference and then followed that up with online interactions afterwards.  There will be some trust between the participants that is an important asset ahead of time.  In this case they will jointly coordinate the division of labor and who does what. The production difference between this case and the previous one is possibly much greater reliance on recorded video chats between the participants.  There may be no need for talking head video at all.  And it might be that video chats that are in discussion mode might very well replace some of the presentation content that would be done if the classes had remained face to face.  If I were participating in this I would try things and see if they have some success.  If not, I would try something else.  Based on my own teaching experience, students who are very grade oriented (the vast majority of the students I had) would prefer that presentation content serve as preparation for the assessments.  But circumstances are clearly different now and there is an underlying question - why does what we are doing in the course matter under the circumstances?  A real answer to that question might be better revealed through discussion than through presentation. And as I wrote in that earlier post, at the undergraduate level most students haven't seen faculty colleagues in discussion with one another.  That novelty might help students to maintain interest in the content.

I want to talk specifically about online office hours in this setting.  To achieve the gains from division of labor, there needs to be a common site for the online office hours.  My suggestion is to rely totally on asynchronous communication - a student poses a question online.  Later, one of the instructors answers that question so that all students in the combined classes can see it.  Here is an example of what such a site might look like.  I made this site about seven years ago, when doing an experiment about the online videos at my profarvan channel.   In the example, students submit their questions via a Google Form.  The form is embedded in a Blogger page, to make it looks as if it is part of the site.  Posts on the main page of the site have student questions and my responses, along with tags on the subject matter.  I have to copy the student questions from this Google Sheet, which has the submitted questions.  Note that the student name is part of the Form, but it is optional.  A first name or initials is sufficient.  There is no intent to go around FERPA here.  This is simply for trying to personalize the response in some way, so it seems like human conversation rather than some automated reply.  What I'm suggesting is something like this, but with the different instructors each possible respondents.  What isn't there in this example, but shouldn't be too difficult to manage, is that if one instructor responds to a student, the other instructors know not to respond in addition, unless the instructors have staked different positions in their online video chat, in which case the additional response furthers the conversation.

My last example is when the instructor of the course doesn't have friends or colleagues at other institutions who might serve as partners in such a division of labor effort.  How are these partners then identified?  I don't have a single good answer here.  Instead, I will offer several different possibilities about how such affinity may be found between instructors who teach the same course at different universities.  (1) The department head might be tasked with collaborating with department heads at peer institutions.  Courses will be aligned as best as possible.  Instructors teaching the same course will be given a list of the names and email address of peer instructors.  Then it is up to them to leverage this information.  (2). Something similar, but now the book publishers act to coordinate the instructors based on textbook adoptions.  (3) Something similar, but now it is the LMS vendors who act to coordinate institutions with the same LMS, and then somebody (perhaps the department heads) acts to coordinate the instructors.  (4) Instructors act independently but post their early video content to a common site, say YouTube.  The YouTube aggregator lists similar videos in the sidebar (this is aimed at watchers of the video).  The instructors use this to contact instructors they don't know who have produced similar content.

An issue with all of these is how long it takes for some real collaboration to take place.  Can something productive happen before the semester concludes?  I don't know.  But I think it is worth trying under the assumption that something good will happen.

Let me wrap up.  I think we should expect fracture here.  There are too many possible sources of failure to expect otherwise.  Given that, what counts as a win is important to consider.  Minor fractures should be acceptable.  Major fractures won't be.  The planning that has happened to date, which has not involved the instructors, will have to confront that regarding instructors who are the weakest link in the chain, while at the same time the most critical.  There are still things to be done beyond the current planning, as I've tried to articulate here.  But I don't want to maintain that doing those things is sufficient to avoid fracture altogether.  So as we worry about what we should do next, we need to also consider what's realistic to aspire toward.  Managing expectations is as important as managing function.

Thursday, March 05, 2020

A Backseat Driver Fully Aware That The Person At The Wheel Is Not Listening

While my title is meant metaphorically, I do recall those days when I'd take one of the boys out to the country, so they could practice their driving without much traffic around, and that one time I took the younger one to the Assembly Hall parking lot on a Sunday when there was no event there, so he could practice while at the wheel doing figure eights.  In these cases, I was sitting in the right passenger seat.  I can't say that I was calm then.  But I did try to coach each boy as best as I could.  There was one trip to Door County I recall where I was sitting in the backseat behind the driver, the older boy.  I let my anxiety get the better of me more than once on the ride up there.  He felt the brunt of that.  Nonetheless, the family did reach its destination and I lived to tell the tale.  If there is a lesson in that observation, I didn't learn it well at all.

What I do want to talk about might be called micromanaging, if in the work context,  or Monday morning quarterbacking, such as in blog posts like this, where others who express their opinions are taken to task.  Part of this is a long ingrained habit, described in a post called Getting Closure as follows:

I have this habit/arrogance about me that in order to let go of an idea and move onto something else I must comment about it in a way that provides some insight. Once I get my two cents in I can refocus on what is ahead of me. Until then, however, I can't let go. I wanted to say something about why we love Mickey so much, but I wasn't sure what I should say. So I let the thought simmer. Yesterday, I did a Google search on Mickey Mantle versus Bo Jackson, thinking perhaps that would help. I found several links to fascinating content - this piece on tape measure home runs, who actually hit them, and the misreporting of how long the home runs actually were; a different piece from Baseball Digest about Bo's amazing feats on the diamond; and a YouTube video clip of Mickey with a rather young David Letterman, compelling to view. 

At the time of writing that post, I was retired, but for under six months, and hadn't yet found what I would engage in next.   I clung to the old habit of doing pleasure reading over the winter break.  In this case it was Jane Leavy's book called The Last Boy.  I had previously read her book about Sandy Koufax and enjoyed that.  But Mickey had a hard life and Leavy emphasized that through much of the book.  Reading the book proved not to be a pleasure, though it remained a compelling read.  And it generated a need in me to write the post I linked to above.

I re-post my my blog writing to Facebook.  At the time, I must have allowed my posts to be Public (now they are only available to Friends).  A classmate I knew from way back in elementary school found the Getting Closure piece, liked it a lot, and we reconnected as a consequence.  The Blogger Dashboard says the piece got only 32 hits and no comments.  When I re-post in Facebook I have no way of knowing whether people have read the piece or not, except through likes or comments.  I'm guessing that most people will read the preview if it shows up in their news feed, but not read the full piece.  I don't know if that explains what I say next or not.  I really do like it when I can see that a reader has made a personal connection to something I wrote.

When I was working, I found this sort of thing fairly regularly.  Very early on, this was via an online conferencing system called FirstClass, that was run by the SCALE project.  Indeed, I believe my posts there were a big reason why Burks Oakley invited me to be part of SCALE.  Subsequently, I had similar interactions on a variety of listservs.  The audience would be different from one list to another and the reaction would often come as a sidebar email only to me.   Then it came with my blog posts, where the connections would show up on the posts of others, such as here and here.   It certainly didn't happen with everything I wrote.  But my batting average was sufficiently high that I came to expect it with some frequency.  I believe that my analyses helped readers who were already thinking about a topic to penetrate it further.  In some sense, my ability to do this was a consequence of my prior economics training, which encouraged me not just to model things in a somewhat abstract way, but then to push for the implications of that model. Others, who hadn't gone through such training, didn't do this as a matter of course.

Providing analysis is one thing.  Correcting others who act in ways contrary to the analysis is quite another.  It's the latter which is the backseat driving I refer to in my title.  I find I'm doing more and more of that.  Yet the corrections are largely unwelcome.  Indeed, my readership overall has dwindled in retirement, as if my analyses are no longer relevant because I'm no longer "in the game." The readership is not quite nil yet, but it is heading in that direction.  So, to tweak a proverbial expression, I'm the one pushing on the tree in the forest to cause it to fall.  Should I continue to do that if I know it will not make a sound?  How do I reconcile that  question with the habit of putting my two cents in?

I will give a few illustrations below to demonstrate how the backseat driving works. 

In a recent post called Lamentations about our Politics, I gave three different critiques.  The third one was about the coalition of people who typically vote for Democrats.  These voters have different interests.  What can hold them together?  In the analysis, the solution was described as a bargain, where each group in the coalition gets something but then also gives up something else in return.  The challenge is in identifying a bargain that will hold.   Alas, the politics of Left, Right, or Center doesn't enable thinking about a bargain.  It casts things as essentially zero sum.   Consider this recent column by Charles Blow, who contrasts the results from the Nevada Primary with the results from the South Carolina Primary. 

First, if the results in Nevada and South Carolina are harbingers for the rest of the nation, this primary season will further explode the people-of-color, intersectional interests argument. It is completely plausible that black and Hispanic voters could consistently and repeatedly pick different candidates, Biden for the former group and Bernie Sanders for the latter.

Blow doesn't consider in his piece what sort of bargain would preserve the people-of-color, intersectional interests argument.  As a result, he can't get at this question.  If your preferred candidate in the Primaries is ultimately not the winner, will you nonetheless participate in the general election?  Is the get-rid-of-Trump sentiment sufficient?  If not, why doesn't Blow contemplate how the winner can can garner the support of the group who didn't get their preferred choice, while not having that viewed as a betrayal by those who had been for the winner in the primaries.   We know that in 2016 Sanders supporters were lukewarm at best to the Clinton candidacy.  If something of a repeat of that now seems likely, the evident question is how to avoid it in 2020.

I'm seeing a similar issue with some of my Facebook friends, who supported a candidate who has since dropped out of the race.  (As I'm writing this, Elizabeth Warren announced today that she is suspending her campaign.)  While I have some friends who are Bernie boosters, most are looking for a reason to support the current front runner, Biden.  Indeed, I'm not enamored with Biden and had hoped we'd have a younger candidate as the nominee.  But I've found the justification that friends are using to make this pivot wrongheaded.  It is entirely backward looking, at what Democrats typically advocate for, without considering the circumstances of the moment, which seem dire indeed. So yesterday I wrote this post, Parallels with 2008.  That Mario Cuomo quote, about campaigning in poetry but governing in prose, still has salience.  But we voters need to be more forward thinking about the prose that lies ahead, after the campaign is over.  Otherwise we are bound for disillusionment.  Even if the get-rid-of-Trump fever is sufficient for the Democrats to sweep in the November election, subsequent disillusionment that allowed the Republicans to win in the 2022 midterm elections might be a disaster, one we should anticipate now. Forward thinking of this sort is a way to temper the idealism that we use to motivate us.  But my Facebook friends don't seem to want that now.  There is already disillusionment from my women friends, who bemoan that the remaining candidates are old white males.  Immediate disappointment that a favorite candidate has dropped out is understandable.  But it can't persist if the Democrats are to prevail long term.

My last example is not about national politics. It is about instruction in the wake of the coronavirus.  I'm aware that there is a lot of planning going on about all university operation in the current circumstance.  But this sort of planning tends to be on the administrative side.  Are individual instructors thinking about these issues as well and how they might have to alter their teaching as a consequence?  I don't know, but I'm afraid they aren't.  At least so far, the planning by the support units doesn't include communicating with instructors now on this topic (at least not at the U of I).  It seems to me it might be necessary for instructors to segue from on-ground teaching to online teaching, within the same course being offered now. So I wrote this post, Teaching a Class Session Online in a Pinch - Some Suggestions.  I can anticipate this happening only at the last moment for many instructors.  That will surely lead to panic and congestion of available support services.  Perhaps there would be some downside risk in communicating about it now.  But if more instructors can take precaution now, that definitely would be a plus.  I think the upside trumps the downside.  Apparently others in a position to do something about this disagree with that assessment or they haven't thought it through.  The real issue is that disagreeing with the assessment provides an excuse for not thinking it through. 

Let me wrap up things here.  The backseat driving emerges as a consequence of worrying about things.  I don't think I can stop worrying, though it's obviously the simplest answer if I could.  So I need a next best solution, one that helps with managing the worrying.  Over the years, writing these blog posts has been useful for that.  The question is whether it can remain useful, even if there is no audience for them.  I don't have any answers for that now. I haven't written blog posts two days in a row for quite a long time. Let's see whether tomorrow there still is an urge to produce another. 

Wednesday, March 04, 2020

Parallels with 2008

I use the Magic 8Ball for making my prognostications.  (My disclaimer as a forecaster.)  Nevertheless, it seems reasonable to me that the Democrats sweep into office in November, not so much because Trump was impeached, but rather because the caronavirus still won't be contained and there will be enormous tumult as a consequence.  The global situation will be characterized by chaos.  Bringing some sensible order to that will be job one.  At this level, the situation is remarkably similar to 2008, where the burst of the housing bubble and the ensuing financial crisis created a like sense of disorder and a real fear that we'd experience a second Great Depression.

Here is a brief characterization of that history, according to my recollection.  President Obama came into office wanting to be a conciliator and thinking (incorrectly) that he was above the political fray.  TARP, a large package to stabilize the large banks, had been passed while Bush was still president, but the Obama administration had to administer it.  Perhaps it wasn't designed this way, but it served as something of a poison pill.  It was incredibly unpopular with voters, as it seemed the ones responsible for the crisis were getting bailed out while ordinary people were suffering financially.

The Stimulus Package was not a second New Deal.  One read about the absence of "shovel-ready projects" so we weren't going to spend our way out of the slump via big construction projects, at least not immediately.  In my view, it was a significant error not to keep at it till they were ready - a major infrastructure renewal plan, which did not emerge.  Instead the Stimulus Package largely was block grants to the states, who were all in fiscal trouble as their tax revenues went down substantially.  So this served to keep them with the same level of services as before the financial crises.  I can't argue with this, but in a real sense it was not new spending.  Then there were substantial tax cuts, mainly to lure the Republicans in Congress to support the bill.  Yet there was no Republican support for it in the House and only 3 Republican Senators voted for it (but this was enough to have a filibuster-proof majority).  I've written elsewhere that tax cuts are not sufficiently stimulating in a deep recession, which is still my belief now.  There was a kind of second stimulus package passed in 2010 after the November election.  Let's be thankful for Lame Ducks who understood what needed to be done and then did it.

However, the composition of the first stimulus is only one issue.  The magnitude of the stimulus was another.  Christina Romer, then chair of the Council of Economic Advisers, argued for a larger stimulus.  Paul Krugman, writing in the New York Times, agreed with Romer.  The package that was approved was in excess of $700 Billion.  Romer wanted a package over $1 Trillion.   The fear was that the larger package couldn't get through Congress.  Political expediency won out over the economic necessity of the moment.  

The next big thing to mention is the Affordable Care Act.  Unlike the stimulus, which was dictated by the moment, the Affordable Care Act resulted because of campaign promises.  Health Care was a big deal issue during the campaign.  Yet, to get the package through, the Public Option was dropped.  Many very enthusiastic supporters of Obama became less enamored with him, particularly young people who put in a lot of legwork getting out the vote in 2008.  At the same time there was a reaction from the Right, the Tea Party.  Democrats lost control of the House in the midterm elections of 2010.

What lessons have Democrats learned from this recent history?

Now I want to talk about private sector parallels.  I'm going to focus on economics only, as I'm sufficiently ignorant about parallels on virus prevention and treatment to comment.  If you've watched The Big Short or read books about the Financial Crisis like How Markets Fail,  you'll know that the big banks were taking risks (think of credit default swaps) that were strongly correlated across them.  In other words, when one of them failed the others would be on the verge of failure as well.  The systematic risk was too great.  We're seeing something similar now with respect to global supply chains, which all of the big-name companies rely on.  When things are going swimmingly, then the global supply chain is a way to keep costs down so that prices to consumers are low.  But when there are disruptions of a major sort, and it appears that the caronavirus will cause that or has already caused that, then there can be a negative feedback loop that brings the system to its knees.  The stock market may not be a much better predictor than I am, but I think it fair to interpret the rather steep decline in stock prices recently as a portent of system meltdown due to supply chain disruption.

To my knowledge, no Democratic candidates are making this an issue during the campaign.

If we have the caronavirus still not under control in November and the supply chain disruption then a major issue, these two factors will make for job one when the new administration takes office.   The economics of inequality issues: health care for all, a living wage, affordable child care, etc., will remain important, but won't have the immediacy created by those two other factors.

How can the Democrats avoid the cycle of enthusiasm for the new regime that propels an electoral win in November from subsequent voter disillusionment with the party in the midterm elections?

I am somewhat optimistic that party elders might address these matters  I can only imagine the behind the scenes conversations that took place to get Buttigieg and Klobuchar to drop out of the race and support Biden.  If and when Democrats do sweep into office, there needs to be something similar about sticking with job one until the situation is under control.  At the same time, there needs to be strong messaging with ordinary voters that this is not a betrayal of the campaign, but an absolute necessity to put the disaster behind us.

Those who cannot learn from history are doomed to repeat it.
George Santayana