Sunday, March 29, 2020

Grading in Times of Pandemic

The stress level for both students and faculty is higher than ever, as we resume classes after Spring Break with one month left in the semester. Hardly any of us has any experience at all with remote teaching. We have been forewarned that technological glitches inevitably will occur and, even if they do not, the dynamics of a virtual classroom are certain to be very different from those of an actual classroom. However confident we might or might not be about our classroom teaching, few of us have any reason to possess confidence in our online teaching abilities, especially in circumstances where a few or many students might not be able to participate live in the class session, but must watch a recording of it.

Regardless, remote teaching will be simple when compared to remote examining and grading, which raise more difficult policy problems for university administrators and operational problems for faculty. The biggest problem, of course, is that there is no reliable method of examining students virtually that does not carry a heightened risk of academic misconduct. It simply is not possible to ensure that students are not communicating with one another during an exam. Requiring students to write papers, rather than examining them, could be an effective substitute (so long as each student is assigned a distinct topic). The problem of plagiarism would be no greater than usual, and reliable plagiarism detection software is available. But requiring all students to write papers for all their courses would be extremely burdensome for them, especially given the time and effort it takes to write one decent paper, let alone several, as well as for faculty members, especially those of us teaching large-enrollment courses. In addition, a paper on a single topic really is not a substitute for an exam because it cannot cover all the topics and issues covered in a course. Meanwhile, inherently unreliable, "virtual" exams provide no basis for assessing grades in the normal fashion.*

Given the uncertainties of teaching effectiveness using new technologies, problems relating to examining and grading, and the high levels of stress already felt by both students and faculty members, it seems sensible to make one other change in ordinary procedures: Instead of issuing letter grades this semester, courses should be graded pass/fail. Overall, such a change would lower the "stakes" of exams, thereby ameliorating stress for both students and faculty. More specifically, it would relieve stress on students, who worry overly about grades even in the best of times; and it reduce pressure on faculty to draw nuanced grading distinctions based on exams that are likely to be even less reliable than they usually are as indicators of student learning. Unlike letter grades, a grade of "pass" has no effect on student grade point averages, which arguably should be left as they are during this semester of upheaval and displacement.

So far, my two academic homes at Indiana University have adopted different approaches to grading this semester. The IU Maurer School of Law has, wisely in my opinion, opted for a pass/fail approach, with strong support from the faculty and a poll of law students. That a super-majority of students preferred pass/fail this semester is somewhat surprising. Law students ordinarily are more competitive than most other students in large part because of the extreme grade-consciousness of legal employers. That they overwhelmingly preferred pass/fail this semester is a clear indication of the stress they are experiencing, as well as their good sense. They must believe, correctly it is to be hoped, that legal employers will recognize this semester as an aberration.

My other half-employer, the O'Neill School of Public & Environmental Affairs, without asking faculty and students about their preferences, has chosen to follow the Provost's recommendations to stick with letter-grading, with exceptions for cases where faculty members find it difficult, with respect to specific students, to assign a letter grade. A faculty member can appeal for permission to grade an entire course pass/fail, but they are adopting a wait-and-see approach of not granting any requests before exams have been administered. I consider this not only unfortunate but downright wrongheaded. It certainly is not in the interest of SPEA students, particularly undergrads, who have less reason to obsess about grades than do law students, for example. Nor is in the interest of faculty members who must attempt to derive nuanced letter-grade distinctions from remotely administered exams that are exceptionally fallible. If it is not in the interests of faculty or students, it must be somehow in the larger interest of the university or the State of Indiana. It is difficult to imagine what that interest might be, and no explanation has been provided. It is true that some students require letter grades for scholarships or other reasons, but it would be much less trouble to provide a letter grade as an exception than as a rule.

In the absence of any well-reasoned** explanation for sticking with letter grades in this unique semester, and the presence of good reasons for assessing student performance on a pass/fail basis, I can only conclude that the law school is responding more sensibly under the circumstances than SPEA or the Provost's Office. If they really were as concerned as they claim to be with student well-being in these parlous times, not to mention faculty well-being, they should change their preference from ordinary letter grades to pass/fail. Keeping to the usual letter-grade requirements this semester strikes me not only as foolish but even cruel, particularly to students who already are under a great deal of stress. Extraordinary times are no time for ordinary academic practices.

     *The grading problem is compounded by certain averaging and curving requirements in different departments.
     **I do not take arguments from tradition, or "because other schools are doing it," or the need to keep things "as normal as possible" under the circumstances to be "well-reasoned." Nothing is normal in the university right now. Not only are students away; they are not allowed to come back. All nonessential personnel are likewise barred from campus, a category which includes most faculty members for most purposes. All faculty-student interactions are now "virtual," including classes and exams. To keep a single element - the grading system - constant while everything else changes is not just abnormal but perverse. 

Saturday, March 28, 2020

Plagues, Then and Now

If COVID-19 had appeared a few hundred years ago, it wouldn't have been called a "pandemic" but a "plague," such as the Great Bubonic Plague of the 14th century (caused by the bacterium Yersinia pestis) that, like COVID-19, originated in China and killed between 30% and 60% of Western Europe's population (its effects were less pronounced in less densely populated Eastern European countries). The historian, Harvard Professor Spencer Strub has an interesting essay in NYR Daily, in which he recounts how European societies attempted to deal with the plague, based on recommendations from scholars, who studied previous outbreaks. Under the prevailing miasma theory of disease, which predominated from ancient civilizations to the late 19th century, the chief causes of the pestilence were thought to be "putrified air," resulting from sinful behavior. []

Strub focuses on commonalities between plagues of old and modern, concentrating on irrational human behavioral responses based on uncertainty and fear. Hoarding, including of items unrelated to a particular virus such as COVID-19, such as toilet paper, seems to be almost instinctual among humans, tracing back to times when any kind of pestilence quickly led to famine. Those who could - the rich and powerful - always have socially and geographically distanced themselves from others, fleeing the densely populated cities for their country estates. And, of course, allegations that disease pandemics are god's punishment for human sins persist.

Perhaps more instructive are the important differences in social responses to great plague's of old and modern scourges like SARS, MERS and influenza, which Strub does not analyze. First and foremost, the probability that any pandemic might cause the death of 30% to 60% of even a relatively poor country's population is vanishingly small because of the very scientific advances that finally replaced miasma theory with the germ theory of disease. It often is observed that improvements in transportation technologies, especially air travel, have made it easier for infectious diseases to travel more rapidly from one place to another. It is less often acknowledged that scientific and technological advances have given humanity a huge advantage in containing and controlling pestilence. Two hundred years ago, the possibility of developing effective treatments and vaccines for a particular virus or bacterium in a matter of nine to eighteen months was zero. People did not possess anything close to a correct understanding of the diseases or their sources. Today, we take it almost on faith that within a reasonably short period of time following an outbreak, scientists (who most of the time are ignored) will have come up with safe and effective treatment and protection measures. And even before those precise remedies become available, we have other generally useful technologies, such as respirators, that can ameliorate disease symptoms, in most cases allowing patients to outlive the disease. The fact that respirators are in short supply is a political-economic problem, not a scientific or technological problem.

Instead of asking why we are so unfortunate to be experiencing the COVID-19 pandemic, we should be thankful that we are experiencing this pandemic now, in an age when it will do much less harm thanks to science.

How can we know that for sure? Consider again the bubonic plague, which wiped out half or more of the population of Western Europe in the 14th century, and is still with us today. According to the CDC [] as recently as the first half of the 20th century (1900-1941), the death rate of plague was 66%, a much higher death rate (ratio of infected persons who die) than COVID-19. Cases of bubonic plague still arise today, mostly in rural areas where vectors of the bacterium, including ground squirrels, chipmunks, wood rats and other rodents, are less controlled. But even in those locations, we don't worry much about it because of (1) the use of prophylactic measures limiting rodent infestations of humans and their homes; (2) lower population-density in areas where infection is more likely naturally limits person-to-person transmission; (3) science-based technologies such as blood testing and antibiotics allow for more rapid diagnoses and more effective treatment that serves to both control the spread of the disease and its mortality rate. The graph below (from the CDC) shows a total of 70 plague cases in the US in the fourteen year period from 2001 to 2013, an average of five cases per year. Of those cases, eight were fatal, yielding a death rate of 11.4%, six times lower than in the first half of the 20th century, but still approximately ten times higher than the death rate from COVID-19 and 100 times higher than the death rate from influenza. In 2015, the death rate actually spiked to 25%. So, bubonic plague remains a far more deadly disease than COVID-19 or influenza. And still there are no vaccines against the plague.

Graph showing human plague cases and deaths in the United States, 2000 to 2018.  There were 6 cases in 2000, 2 in 2001, 2 in 2002, 1 in 2003, 3 in 2004 with 1 death, 17 in 2006 with 2 deaths, 7 in 2007 with 2 deaths, 3 in 2008, 8 in 2009 with 2 deaths, 2 in 2010, 3 in 2011, 4 in 2012, 4 in 2013 with 1 death, 10 in 2014, 16 with 4 deaths in 2015, 4 cases in 2016, and 5 cases in 2017.That doesn't mean we can let our guard down and breath a sigh of relief about COVID-19. It is a novel coronavirus (not a bacterium like Bubonic Plague, and therefore not treatable with antibiotics) about which we still have a lot to learn. What we know already is that it is approximately ten to fifteenth times more deadly than influenza, which kills tens of thousands in the US each year, and that it can pretty easily be spread from person to person either through direct or indirect (e.g., surface) contact. Until we know more -- perhaps until some useful treatment(s), already in the pipeline, become available -- the importance of social-distancing can hardly be overstated. But unlike people living just a century ago, we have every reason to expect that effective treatments, and at least partially effective vaccines, will become available within the next year or so. It remains deplorable, of course, that countries prevaricated for ideological, political and economic reasons over the outbreak -- China withheld information and, in the US, the Trump Administration ignored scientific information and recommendations for a month or more before acting -- costing potentially tens of thousands of lives. And for religious and business leaders who maintain the medieval view that the cause of COVID-19 is human sinfulness and that the cure lies in a return to one god or another, I trust that, if you become infected, you will eschew scientific remedies and treatments, including ventilators. Oh, and you should probably avoid driving cars, flying on airplanes, and using computers.