Student Responses to Changes in Introductory Physics Learning due to COVID-19 Pandemic
Matthew Dew, Jonathan Perry, Lewis Ford, Dawson Nodurft, Tatiana Erukhimova
SStudent Responses to Changes in Introductory Physics Learning due to COVID-19Pandemic
Matthew Dew, Jonathan Perry, Lewis Ford, Dawson Nodurft, and Tatiana Erukhimova Department of Physics and Astronomy, Texas A&M University, College Station, TX 77843 ∗ Department of Physics, University of Texas, Austin, TX 78712 (Dated: September 25, 2020)
As a result of the spread of COVID-19 during spring2020, many colleges and universities across the US, andbeyond, were compelled to move entirely to remote, on-line instruction, or shut down [1]. Due to the rapid-ity of this transition, instructors had to significantly – ifnot completely – change their instructional style on veryshort notice [2]. Our purpose with this paper is to re-port on student experiences and reactions to the switchto emergency remote learning at two large, land-grant,research intensive universities. We aimed to explore howstudents have received and dealt with the shift to re-mote learning that began in March 2020, specifically inintroductory physics and astronomy courses. By provid-ing timely student feedback, we hope to help instructorstune their efforts to build a more effective remote learn-ing environment.While online courses have been around for severaldecades, most major universities do not offer more than ahandful of their classes online [3]. Existing literature onthe principles of design and evaluation of online coursessuggest that high-quality online courses require severalmonths of preparation as well as infrastructure and sup-port for both students and instructors [2, 4]. Even withthe careful preparation of an online course, studies priorto the 2020 pandemic have often revealed a deficiencyof collaborative aspects, in both student-student andstudent-instructor interactions [5–7]. Literature focusedon online introductory physics, or courses with onlinecomponents, show these interactions being linked to stu-dent success [8, 9]. While the educational community putin extraordinary effort in spring 2020 to keep teaching,the timeline of the shift to emergency remote learning didnot allow for the thoughtful planning of a typical onlinecourse.Given how fast and total the switch to remote learn-ing has been, we do not fully understand how studentshave been affected – both in general and for specificstudent populations. This transition can exacerbate in-equity and disadvantage students from lower income fam-ilies [10]. Pandemic-induced isolation and socio-economichardships may also affect students mental health [11].To explore how students responded to the shift toemergency remote learning, we developed and adminis-tered a questionnaire gauging the impacts on studentsmotivation and interactions with their courses, their ∗ send correspondence to: [email protected] peers, and instructors. We also examined how student re-sponses depend on demographic factors which are listedbelow. METHODS
The questionnaire was distributed to 2,320 studentsenrolled in summer 2020 courses, which were delivered re-motely, at both participating institutions. At InstitutionA, the questionnaire was distributed via email by depart-mental administration. Additionally, a member of theresearch team attended a lecture for each class encourag-ing students to take the questionnaire. Here, there werea total of 13 lecture sections from 6 courses taught by 8instructors – covering both algebra-based and calculus-based introductory physics, an introductory astronomycourse, and a sophomore-level modern physics course. AtInstitution B, the questionnaire was distributed by fac-ulty via the Learning Management System. Here, therewere 6 lecture sections from 6 courses taught by 5 in-structors – covering three, two semester tracks of algebra-based and calculus-based introductory physics.The questionnaire was entirely anonymous. Studentparticipation was voluntary with no positive or negativeinducements. Distribution occurred during the secondhalf of courses, which were primarily 5-week courses (onecourse was 10 weeks). A total of 708 responses werereceived, 508 from Institution A (37.5% response rate),and 200 from Institution B (20.7% response rate).The questionnaire consisted of 8 demographics ques-tions and 27 questions on student preferences and be-haviors. The 8 demographics questions asked for stu-dent gender, race, first-generation status, classification,weekly work hours, internet access, and devices used forclasswork. Questionnaire items were built with three dif-ferent response types: 5-point Likert scale, multiple an-swer, and ranking.Each of the response types mentioned above was ex-amined in a distinct manner. Likert-scale items are rep-resented by the fraction of students who chose each an-swer, summing to 1. Multiple answer items are repre-sented by the fraction of responses for each individualchoice. Ranking items assigned a point value to each an-swer, where each response’s score was normalized by thetotal number of points distributed. The 95% confidenceintervals were calculated and are included on each figure. a r X i v : . [ phy s i c s . e d - ph ] S e p RESULTS & DISCUSSIONDemographics:
Students responding to the ques-tionnaire were more likely to be female (59.8%) thanmale (38.9%). The remaining percentage of respondentsidentified as non-binary or preferred not to say. Nearlythree-quarters of responses came from juniors and sopho-mores. Students primarily identified as white (58.6%),with Asian and Hispanic/Latinx/Spanish origin identify-ing students responding in roughly equal numbers (23.7%and 22.7% respectively). Students were allowed to selectmore than one race or ethnicity and were counted in allcategories selected. Around one in five students identifiedas a first-generation college student.
Synchronous vs. Asynchronous:
The majorityof the opinions expressed on this item are in the stronglyprefer categories, indicating a strong polarization amongrespondents. Overall, students at Institution A exhib-ited a slight preference for synchronous classes over asyn-chronous classes; more students at Institution B showeda stronger preference (around 2-to-1), Figure 1. Perhapsrelated to the preference for synchronous learning, stu-dents had a strong agreement ( ∼ FIG. 1. Student responses to “Between synchronous classesand asynchronous classes, what do you prefer for lecture?”
Remote Learning Capabilities:
About 40% ofstudents do not have their own study space to work undis-turbed at least most of the time. Shared spaces may
FIG. 2. “Which part(s) of online classes do you want to keepwhen classes return to normal?” mean that these students do not have privacy to neces-sarily feel comfortable using a microphone or webcameraduring every component of a course. Though all respond-ing students have access to at least a desktop or laptopcomputer, less than half reported access to a printer, andeven fewer reported access to a scanner (22.5%) or a doc-ument camera (10.5%). It should be noted that bothinstitutions made specific efforts to provide technology(including purchase of laptops) for students during thespring semester.Resources such as textbooks, recitations, andinstructor-provided online resources have become moreimportant to student learning since the pandemic. Stu-dents report that they are most likely to ask a friend, orreference online problem-solving resources ( e.g.
Chegg)when stuck on a problem. This behavior is not surpris-ing, and may not be significantly different than studentbehavior during a typical semester. However, it is im-portant context when considering study habits, whichare discussed below.
Community & Engagement:
On the whole, stu-dents did not feel connected or engaged with their class-mates during their 2020 summer physics courses. Overhalf of the students reported feeling “rarely” or “never”connected or engaged with classmates during lecture,Figure 3. These responses are similar for all componentsof the course that students were asked about. This likelystrongly contributed to more than 50% of students re-porting far fewer social connections being made in thiscourse compared to their average face-to-face classes.Student study habits changed drastically in the socialcomponents of their courses. Before the pandemic, abouttwo thirds of the students met at least once a week tostudy. During the pandemic, however, nearly the samefraction of students reported studying only alone, Figure
FIG. 3. Student responses to “I felt engaged/connected withmy classmates during this course’s lecture.”
4. While virtual meetings allow students to communicateand potentially study together, the ability to collaborateand share work in real time has become more difficult.
FIG. 4. Student responses to “On average, for each class,how many times a week did you meet with classmates to studybefore/during the pandemic?” for (a) Institution A and (b)Institution B.
Despite the decrease in group studying, one fourth ofstudents never contacted their instructor outside of class.Many students who attended office hours multiple timesduring the course were neutral on the visits usefulness.About 35% found them helpful.One method of engaging students during lecture,in-class demonstrations, received significant agreement( ∼ Stress & Motivation:
Students reported quality ofeducation as the most common cause of stress at bothinstitutions (63%), Figure 5. This is followed by an ex-pected feeling of stress about personal and family health,both around 50%. It is surprising that in a time of pan-demic more students reported stress relating to their ed-ucation than their own health.
FIG. 5. Student responses to “Which of these issues havebeen a cause for stress for you since the pandemic? Markall that are applicable.” This is plotted by the fraction ofstudents reporting this cause of stress.
Perhaps related to student stress over education, re-ported motivation for being able to work on assignmentschanged drastically after shifting to remote learning. Fig-ure 6 shows a striking decline in the ease of students beingable to work on their course assignments from both in-stitutions. One should keep in mind this is not a directcomparison between two physics courses taken face-to-face and remotely. The “before pandemic” question in-cludes students who have never been enrolled in a physicscourse.Both Institution A and Institution B experienced a sig-nificant increase in enrollment during summer courses,more than doubling from prior years. While many re-sponding students stated they would have taken thecourses at these institutions regardless of circumstances(48%), other students changed their plans; either not en-rolling at an alternative institution (19%), or taking thecourse because it was offered online (18%).
Differences by Demographics:
We examined re-sponses to each questionnaire item separated by studentdemographics based on gender, first-generation status,classification, employment, course enrolled in, and pri-mary motivation. Only one identifier, first-generationstatus, led to noteworthy differences from what has beenmentioned earlier.
FIG. 6. Student responses to “Before the pandemic/For thisphysics class, I was able to get myself to work on assignmentswithout difficulty.” for (a) Institution A and (b) InstitutionB.FIG. 7. Student responses to “Which of these issues havebeen a cause for stress for you since the pandemic?” for (a)Institution A and (b) Institution B. These responses are sep-arated by student responses to “Are you a first generationcollege student?”
First-generation students reported more frequentlythan non-first-generation students every listed cause ofstress except for quality of education, Figure 7. First-generation students are also less likely to have access toa private study space, 52% compared to 74% of non-firstgeneration students.
Summary:
In response to the sudden and singularchanges during spring 2020 necessitated by the pandemic,we developed and administered a questionnaire with thegoal of finding ways to better serve our students. We col-lected student feedback from summer 2020 introductoryphysics and astronomy courses at two large, land-grantinstitutions. Analysis of responses shows that studentsare experiencing drastically reduced social interactionsand connections through their courses. Individual mo-tivation to complete coursework has decreased precipi-tously. Students are also experiencing multiple causes ofstress, with quality of education topping the list. Thesereactions are expressed across all demographic categoriesin a similar way. Noticeable differences are observed onlyfor first-generation students who report causes of stressmore often and are less likely to have a private studyspace. Post-pandemic, increased availability of courseresources, particularly recorded lectures and online ma-terials, resonates with students.Although a direct comparison cannot be made betweenemergency remote learning and online education, studentresponses about interactions within courses mirror pre-pandemic literature studying online learning [5, 9, 13].Readers should keep in mind that responses to the ques-tionnaire represent only a snapshot of student preferencesand experiences gathered during condensed versions ofphysics and astronomy courses during summer 2020. Inreporting this, we have aimed to provide a slice of timelyinformation as educators continue to make instructionalchoices during remote learning and eventually, for thereturn to face-to-face classes.
ACKNOWLEDGMENTS
We would like to thank the faculty from both institu-tions who allowed us to distribute the questionnaire totheir students during their summer courses. This studywas supported in part by the Texas A&M University De-partment of Physics & Astronomy. [1] UNESCO, “Covid-19 educational disrup-tion and response,” (2020), retrieved fromhttps://en.unesco.org/covid19/educationresponse.[2] Charles Hodges, Stephanie Moore, Barb Lockee,Torrey Trust, and Aaron Bond, (2020), retrievedfrom https://er.educause.edu/articles/2020/3/the-difference-between-emergency-remote-teaching-and-online-learning. [3] National Center for Education Statistics, “Number andpercentage of students enrolled in degree-granting post-secondary institutions, by distance education participa-tion, location of student, level of enrollment, and controland level of institution: Fall 2017 and fall 2018,” (2019).[4] Patricia McGee, Deborah Windes, and Maria Torres,“Experienced online instructors: beliefs and preferredsupports regarding online teaching,” Journal of Comput- ing in Higher Education , 331–352 (2017).[5] Justin Paulsen and Alexander C. McCormick, “Reassess-ing disparities in online learner student engagement inhigher education,” Educational Researcher , 20–29(2020).[6] Patrick Lowenthal and Chareen Snelson, “In search of abetter understanding of social presence: an investigationinto how researchers define social presence,” Distance Ed-ucation , 404–411 (2018).[9] Kelly Miller, Sacha Zyto, David Karger, Junehee Yoo,and Eric Mazur, “Analysis of student engagement in an online annotation system in the context of a flipped in-troductory physics class,” Phys. Rev. Phys. Educ. Res. , 1529 (2020).[12] Greg Kestin, Kelly Miller, Logan S McCarty, KristinaCallaghan, and Louis Deslauriers, “Comparing the ef-fectiveness of online versus live lecture demonstrations,”Physical Review Physics Education Research , 013101(2020).[13] Baruch Offir, Yossi Lev, and Rachel Bezalel, “Surfaceand deep learning processes in distance education: Syn-chronous versus asynchronous systems,” Computers &Education51