Query results: October, 2020

Now that the semes­ter is over, I can take a few moments to reflect over the 2020 edi­tion of my flu­id dynam­ics course at the Uni­ver­si­ty Otto von Guer­icke of Magde­burg.

The semes­ter began right as Ger­many shut down due to the pan­dem­ic cri­sis. The course, with 153 stu­dents, took place entire­ly online. I spent near­ly half of the semes­ter work­ing from abroad, attend­ing to a close rel­a­tive. I was nev­er able to meet any stu­dent physically.

The course pro­gressed on a week­ly basis. I released one chap­ter every week, a sin­gle PDF doc­u­ment con­tain­ing lec­ture notes and prob­lems. In the doc­u­ment, I linked to YouTube videos (on aver­age, five per chap­ter) where I cov­ered impor­tant ele­ments of the the­o­ry, and solved prob­lems step-by-step. I also linked to inter­est­ing illus­tra­tive videos. In week­ly, two-hour Zoom ses­sions, I answered any and all ques­tions that were thrown my way.

In addi­tion, every stu­dent received indi­vid­u­al­ized home­work as part of the peer-grad­ed home­work pro­gram I devel­oped with my col­leagues Math­ias Mag­dows­ki, Ger­mán San­ta-Maria and Jochen König. This year, par­tic­i­pa­tion in the home­work was manda­to­ry and count­ed 40% towards the final course grade. I hoped to run four instances of the home­work, but in the end, because of per­son­al mat­ters, only three pieces of home­work were hand­ed out.

A series of week­ly four-ques­tion quizzes, run through the University’s Moo­dle instance, added up to 10% of the final grade. The remain­ing 50% of the final grade con­sist­ed of the final two-hour invig­i­lat­ed exam­i­na­tion, with the same struc­ture (choose three prob­lems among five) as in for­mer years.

A lot has hap­pened through the semes­ter, and I have many mixed feel­ings about this edition.

First, I am glad I learned to record videos, and I think I got it right. A good-qual­i­ty, sub­ti­tled record­ing of a prob­lem being solved by hand (I used a Wacom tablet, an exter­nal micro­phone and an exter­nal web­cam plugged into a lap­top) is a much bet­ter tool to learn with than a live ses­sion in a class­room with a chalk­board. I chose YouTube as a com­pro­mise, because it suits stu­dents well, but will try to migrate to a Peer­Tube instance in the future.

Admin­is­tra­tive man­age­ment was a night­mare and cost me hours of work week­ly through­out the semes­ter. My uni­ver­si­ty does not com­mu­ni­cate well with either teach­ers or stu­dents —the local TV sta­tion knew of the revised semes­ter dates before I did— and pre­dictably the com­mu­ni­ca­tion between teach­ers and stu­dents is chaot­ic. Addi­tion­al­ly, my uni­ver­si­ty has no effec­tive con­trol over the course reg­is­tra­tion process: for exam­ple, just get­ting an up-to-date list of stu­dent emails is a chal­lenge. I hope bet­ter man­age­ment will alle­vi­ate me of that load in the future.

Com­mu­ni­ca­tion chan­nels between stu­dents remained out of my reach for the com­plete semes­ter. I heard What­sApp groups are the thing; in any case the Moo­dle dis­cus­sion boards of my course failed to reach crit­i­cal mass. Too many stu­dents address me direct­ly. I tried to redi­rect requests to the inter­ac­tive Zoom ses­sions and Moo­dle forums, but in the end I count only 174 Moo­dle posts, and 860 emails sent and received over the semes­ter (I do not count the 1800 mails processed auto­mat­i­cal­ly as part of the peer-grad­ed home­work pro­gram). An aver­age of 4.5 mails received, 4.5 sent every day is just too much for one class. A few stu­dents are dis­re­spect­ful (although always rev­er­ent), and many have com­plete­ly unre­al­is­tic expec­ta­tions, not hav­ing tru­ly real­ized what it means to be a sin­gle instruc­tor with 153 stu­dents on the side of research-relat­ed duties.

Zoom ses­sions went very well. The tech­ni­cal robust­ness of Zoom impressed me, espe­cial­ly in view of the low lev­el of com­put­er lit­er­a­cy and equip­ment of my stu­dents. Atten­dance (share of exam-goers present for 45 min­utes or more) began at 65%, decreas­ing lin­ear­ly with time down to 17% in the last ses­sion. I regret that so few stu­dents chose to turn on their cam­era, even though I respect their deci­sion. As a teacher, I yearn for visu­al con­tact and body lan­guage dur­ing inter­ac­tive ses­sions, but this semes­ter, I end­ed up star­ing most­ly at a wall of usernames.

One nice suc­cess was intro­duc­ing illus­tra­tions of peo­ple in the lec­ture notes, to pass across impor­tant points in para­graphs titled “advice from an expert”. I had some illus­tra­tions made to order, and licensed oth­ers under a lim­it­ed-use, com­mer­cial license, all fol­low­ing the advice of my col­league Dr. Kristin Hecht.

Those para­graphs had two pur­pos­es. First and fore­most, I am tired of flu­id mechan­ics being nar­rat­ed most­ly by old white men, and want­ed to put dif­fer­ent-look­ing peo­ple in charge of giv­ing tech­ni­cal advice to stu­dents. I hope my stu­dents, almost none of whom look like me, can iden­ti­fy them­selves with some of the per­sona dis­played in the lec­ture notes now. Sec­ond, using this tech­nique helped me with the tone of the lec­ture notes, which I had long found to be slight­ly bor­ing and too seri­ous. Now, I can keep the tech­ni­cal con­tent short and pre­cise, while indulging in a more relaxed tone on occa­sion, address­ing the read­er direct­ly. For good mea­sure, I also sprin­kled links to a few of my favorite sci­ence web­comics (such as xkcd) across the chapters.

The largest pain point remains the grad­ing. While the dis­tri­b­u­tion of final grades remains large­ly the same as in the last five years, there is a large dis­crep­an­cy between course­work and exam grades.
In the peer-grad­ed home­work pro­gram, there is evi­dence that a siz­able share of stu­dents par­tic­i­pat­ed in bad faith, which result­ed in low-qual­i­ty grad­ing of oth­er stu­dents. This is an inher­ent weak­ness of the pro­gram: if too many stu­dents, even when hand­ed a com­plete step-by-step worked-out guide, are unwill­ing or unable to iden­ti­fy the source of an error in the papers they grade, then they can­not be eas­i­ly “sort­ed out” of the sys­tem, and glob­al stu­dent moti­va­tion and learn­ing will suf­fer.
There is also plen­ty of evi­dence that the sys­tem of week­ly quizzes was abused. I had thought that because the stakes were low (quizzes amount to 10% of the final grade, so any one ques­tion weighs only 0,25%), there would be no incen­tive for gam­ing, and I was wrong.
Because of those issues, I went into the exam with the impres­sion that course­work grades did not reflect the true lev­el of learn­ing of a sig­nif­i­cant num­ber of stu­dents. The exam proved me right: with 55% of par­tic­i­pants fail­ing the exam prop­er, and 33% not even clear­ing a quar­ter of points, it was the worst stu­dent per­for­mance in record. This nec­es­sar­i­ly brings into ques­tion whether deploy­ing course­work is actu­al­ly worth the effort, since this year, it objec­tive­ly cor­re­lates with low­er learning.

There are many fac­tors that come into play here, not all of which are under my con­trol. First, this year the Uni­ver­si­ty lift­ed restric­tions on the num­ber of attempts allowed for each exam, and so some stu­dents may have been temped to have a “free try” (passed exams can­not be repeat­ed how­ev­er, which makes for weird dynam­ics). Sec­ond, grades secured with course­work reduce the pres­sure exert­ed on stu­dents dur­ing the exam. I have long under­stood that grades act as a met­ric for the student’s time bud­get­ing, both with­in my course and in between dif­fer­ent cours­es. This is com­plete­ly nat­ur­al, and I wouldn’t blame stu­dents for putting focus on oth­er exams where the stakes were high­er. (Prop­er qual­i­ty mea­sure­ment and man­age­ment would sub­stan­ti­ate those state­ments and help with strate­giz­ing, but unfor­tu­nate­ly, my uni­ver­si­ty is not there yet either.) And final­ly, the pan­dem­ic dis­pro­por­tion­ate­ly affect­ed stu­dents in my course, most of whom come from abroad. I can’t imag­ine what it means to work on an exam when you are barred from fly­ing back home to take care of loved ones, or uncer­tain whether you can even keep study­ing through the com­ing semester.

Look­ing ahead, I need to make the grades bet­ter reflect the lev­el of learn­ing. Quizzes won’t be grad­ed any­more, and I will like­ly make home­work an option­al, low-weight pro­gram again.
Ulti­mate­ly, it is clear that there are stu­dents in my course who have not attend­ed a sin­gle Zoom ses­sion dur­ing the semes­ter. There are stu­dents who do not think it use­ful to invest 20 min­utes watch­ing the exam prep video I sent them before mak­ing an attempt. There are stu­dents who are not able to manip­u­late a sim­ple equa­tion that con­tains only sym­bols, and there are stu­dents who have no inhi­bi­tion against writ­ing unin­ter­rupt­ed non­sense for two hours on a paper. Sure­ly, any of those last two issues pre­clude any suc­cess in an engi­neer­ing flu­id dynam­ics course. Before I make course­work a large part of the course again, I must be able to pre­vent those stu­dents from inter­fer­ing with the suc­cess of others.

I can­not con­clude this with­out men­tion­ing the sheer joy that comes out of all this work. It is uplift­ing and hum­bling now to meet stu­dents on cam­pus who took part in the course and express feed­back. This com­ing year, with most of the course con­tent now solid­ly cov­ered with 15-minute videos, I will be in a much bet­ter posi­tion to fol­low the foot­steps of my col­league Dr. Stef­fi Knorn and flip the course around. My vision is to be spend­ing all of my class time help­ing stu­dents actu­al­ly solve real­is­tic prob­lems in flu­id dynam­ics, using mod­ern tools and tech­niques. I look for­ward to 2021.