Skip to main content

Remote Learning

Teaching Assistant Responses to COVID-19

Investigating Relationships Between Stress, Self-Efficacy, and Approaches to Teaching

Journal of College Science Teaching—January/February 2023 (Volume 52, Issue 3)

By Cody Smith, Deepika Menon, Annette Wierzbicki, and Jenny Dauer

Undergraduate and graduate teaching assistants (TAs) are tasked with instructing undergraduate courses with little to no professional development (PD). To better develop PD opportunities, it is important to understand the benefits of improving TAs’ self-efficacy and the stressors associated with their roles. This study investigated how stress related to the COVID-19 disruption of the spring 2020 semester affected TAs’ self-efficacy. Pre-, mid-, and postsemester surveys of self-efficacy and perceived stress were analyzed along with a structured postsemester interview addressing TAs’ stress related to the disruption. Although no relationships were found between self-efficacy and stress, qualitative results indicated that stressors related to changes in and uncertainty around their responsibilities influenced TAs’ confidence in their teaching. The results of this study are discussed in relation to PD programs providing experiences with multiple teaching modalities, strategies for time management, and mental health resources.

Undergraduate and graduate teaching assistants (TAs) play key roles in the instruction and learning of undergraduate science, technology, engineering, and mathematics (STEM) education (DeChenne et al., 2015). Duties that TAs commonly take on include leading laboratory and recitation sections (Fagen & Wells, 2000), and meeting with students for office hours or one-on-one supplemental instruction. Other duties may include assisting instructors during lecture and grading. Given their proximity to undergraduate students and the potential to influence student learning, TAs need to have high self-efficacy beliefs for teaching. Self-efficacy beliefs are the strongest predictors of motivation and performance (Bandura, 1997). Literature suggests that TAs often lack necessary training or experience, leading to stress that affects their teaching performance (Shum et al., 2020). Although the literature shows that TA pedagogical training is universally insufficient (Reeves et al., 2018), the teaching and course leadership experiences that TAs gain contribute to their development for careers both inside and outside academia. Given their importance, TAs’ lived experience within their roles and the associated benefits and stressors are important to understand for TA professional development and retention in teaching careers.

Self-efficacy is the judgment of one’s ability to handle challenging situations, and it has important implications for performance in the classroom (Bandura, 1997). Prior studies in teacher education suggest that teachers with higher perceived self-efficacy are better equipped to overcome challenges and remain in the profession (e.g., Chesnut & Cullen, 2014; Klassen & Chiu, 2011; Menon & Azam, 2021). TA duties come with a variety of challenges, which are augmented by other factors associated with the TAs’ status as students and researchers. In addition to the stressors associated with balancing multiple roles and responsibilities, TAs often lack the experience and development as instructors to confidently perform their tasks (Holmes et al., 2013). Many institutions or departments that do offer TA development opportunities only address the basic orientation to policies associated with leading a group of students, such as time management and how to conduct oneself when working with students (Gardner & Jones, 2011; Wyse et al., 2014). Seldom do TA trainings delve into more nuanced aspects of teaching, like teaching philosophies and approaches or theories on how students learn, nor do they give feedback on teaching practices. Moreover, TAs often have little or no training in handling unanticipated situations that may arise in the classrooms, and they may lack persistence in working within challenging environments. Understanding TAs’ self-efficacy beliefs is especially important for addressing higher rates of anxiety and depression among graduate student populations in general (Evans et al., 2018; Woolston, 2019).

The outbreak of COVID-19 caused disruption to the spring 2020 semester and compounded existing issues even further, as TAs had to abruptly move online to learn themselves and to instruct courses. Furthermore, the pandemic caused personal anxiety and uncertainty about health, safety, financial security, and myriad other issues. One study reported a 13% increase in anxiety disorders and a 19% increase in major depressive disorders among approximately 4,000 STEM PhD students between May and July 2020 (Chirikov et al., 2020). Studying the impact of COVID-19 and the toll it takes on TAs can be important for understanding responses to other unanticipated situations and supporting TAs’ self-efficacy so they can better cope in the future.

Improving our understanding of how self-efficacy affects TAs can inform how to support them as instructors. Smith and Delgado (2021) proposed a model of TA teacher efficacy that indicates that those who possess low self-efficacy and view issues related to classroom teaching as challenging are more inward-focused (teacher-centered), whereas those who possess higher self-efficacy and view teaching issues as manageable are outward-focused (student-centered) regarding their impact on student learning. Stress derived from uncertainty in one’s position may compound a TA’s belief that teaching issues are challenging, especially if the uncertainty is linked to changes in the delivery of the course or other constraints outside the TA’s control.

This study will investigate the self-efficacy of undergraduate and graduate TAs associated with changes in instruction after the COVID-19 disruption to the spring 2020 semester. Research questions that guided this study include the following: (i) What are the relationships among TA teacher efficacy and stress resulting from the COVID-19 disruption? (ii) What experiences do TAs describe that contribute to their stress and teacher efficacy? Implications of the results for supporting TAs during similarly stressful events and circumstances will be discussed.


Research context and data collection procedures

This study was conducted at a large midwestern university in the United States during the spring 2020 semester. Table 1 includes demographic information of the 13 TAs in undergraduate STEM courses who participated in this study. Data-collection procedures included surveys, open-response questions, and interviews at three time points: pre-semester, the week before courses moved online (midsemester), and postsemester. To track TA teacher efficacy development over time, the Graduate Teaching Assistant Teacher Self-Efficacy Scale (GTA-TSES; DeChenne et al., 2012; see also Online Appendix A) was administered electronically at all three data-collection points. To determine stress levels due to abrupt change to online instruction during the pandemic, the Perceived Stress Scale (PSS; Cohen et al., 1983; see Online Appendix B) was administered electronically when the announcement to move online was made. Items on the PSS were modified to fit the setting and purpose of this study. For example, “How often have you felt nervous and stressed?” was reworded as “Since the mode of instruction in courses has changed due to the COVID-19 disruption, how often have you felt nervous and stressed?”

Along with the PSS, two items were added for participants to indicate, using a 5-point Likert scale, the degree to which their stress was related to or not related to stressors within their TA duties (with 1 representing “not at all stressful” and 5 representing “extremely stressful”). Six additional open-response items were added to the PSS to ask for descriptions of stress related to TA duties relative to other current stressors, changes in responsibilities for the remainder of the semester, preparation and adjustment to conducting online classes for the remainder of the semester, concerns regarding conducting online classes, how to address these concerns, and how confidence in carrying out TA duties has changed. A postsemester structured interview (Online Appendix C) was also conducted online via remote videoconference. Transcripts of the interviews were generated for analysis.

Data analysis

Quantitative data were analyzed using SPSS Statistics for Macintosh version 27.0. To examine changes in teacher efficacy throughout the semester, a one-way repeated measures analysis of variance (ANOVA) was used to determine any statistically significant changes in GTA-TSES across all three data-collection time points. A Pearson correlation coefficient was used to determine relationships between PSS results and pre- to midsemester change, midsemester, postsemester, and mid- to postsemester change in GTA-TSES results. An independent samples t test was used to compare means between the two Likert-scale items added to the PSS, measuring the degree to which stress was related or not related to TA duties.

To describe TAs’ experiences contributing to stress and teacher efficacy, we used a case study approach (Tellis, 1997) to analyze the open-ended questions that were included along with the PSS. Our analysis of the open-ended items centered on the two questions regarding the comparison of teaching responsibilities that TAs had before and after moving their courses online and whether they perceived any changes in their confidence with regard to their ability to carry out their TA duties over the same time frame. The constant comparison method was used to categorize responses into thematic codes agreed upon by two coders (Corbin & Strauss, 2008). Interview data were used to further describe how the COVID-19 disruption might have influenced TAs’ stress levels and teaching experiences. Interview transcripts were broken down into utterances of independent thoughts, which were coded by two researchers with a moderate level of agreement (k = 0.76; McHugh, 2012) using the framework by Mazzola and colleagues (2011) that details stressors among graduate students. Categories of stressors pertinent to the setting of this study and thus used in our analysis of the interview data included work overload, role ambiguity, lack of control, constraints, unproductive time and effort, and interpersonal conflict. Due to the nature of this study being set in a pandemic possibly bringing about stressors novel to those found by Mazzola, emergent categories outside the framework were also considered by the coders as they arose.


Table 2 provides individual results for the GTA-TSES, PSS, stress unrelated to TA duties, and stress related to TA duties, as well as mean and standard deviation for each. Results from the repeated measures ANOVA indicated there were no statistically significant differences in TA teacher efficacy as measured by the GTA-TSES across all three time points (F = 0.514, p = 0.60). The average level of stress among participants was moderately elevated from the validated norms associated with a similar age range of the normative population (M = 14.2 ± 6.2; Cohen et al., 1983). PSS scores 15 and below were considered relatively low, PSS scores from 16 to 20 were considered relatively moderate, and PSS scores 21 and above were considered relatively high. The Pearson correlation tests showed no statistically significant relationships between PSS and pre- to midsemester change in GTA-TSES (-0.04 ± 0.39; r = 0.10, p = 0.74), midsemester GTA-TSES (r = -0.11, p = 0.72), mid- to postsemester change in GTA-TSES (0.10 ± 0.36; r = 0.12, p = 0.70), or postsemester GTA-TSES (r = 0.02, p = 0.94). The degree to which stress was related to stressors outside of TA duties was found to be significantly greater than the stress related to TA duties (p < 0.001).

Two major categories emerged from the TAs’ responses to six open-ended items placed at the end of the PSS survey to further understand their sources of stress and its impact. Four TAs reported that they had reduced duties during the move to online courses, whereas nine TAs maintained the same duties. Six TAs expressed less confidence in carrying out their TA duties, and seven had no change in their confidence. TAs were grouped by two factors (stress level and how much TA duties changed), which produced three groups (high stress with reduced duties, high stress due to change in mode of instruction, and low stress but no change in regular duties), which we describe in the following sections. Within the description of each of the six groups, we focus on our emergent themes related to self-efficacy and stress (a total of 10, including four themes more than the six themes identified by Mazzola and colleagues [2011]). Table 3 describes these 10 categories with participant quotes, and Table 4 provides a summary of the frequency of the categories that occurred in participants’ responses.

High stress, reduced duties

Four categories were prominent in this group, including change in nature of tasks or environment, distant or removed from responsibilities, student engagement, and student success. Three TAs (pseudonyms: Julietta, Valentina, and Hector) reported high levels of stress even though their duties were reduced to grading and administering exams as opposed to being responsible for teaching recitation or problem-solving aspects of the course prior to moving online. Julietta’s primary source of anxiety and stress prior to the disruption was supporting students’ learning and preparing them for weekly quizzes when they could not attend class. For instance, she wrote, “Some students that had some personal problems and couldn’t attend classes were not responding to emails and I didn’t know what to do in that situation.” She further mentioned that she missed the opportunity to teach firsthand the insect identification and outdoor insect collection activity, which could have led to an increase in her confidence to teach in the future. The reduction of TA responsibilities did, however, make her more available to answer students’ questions about their grades on assignments.

Valentina taught the same course as Julietta and also expressed having missed out on teaching. She spoke about the challenge of being motivated to work when she had so much free time at home. However, she was still responsible in her reduced duties; as she said, “If it was related to classes, to stuff that I really had to do like grade...I will do it and it didn’t really affect me that way as they connect via email or meeting with them via remote videoconference.” On the contrary, when courses were moved online, Hector’s recitation was canceled, and he was relegated to meeting with students virtually one-on-one as needed and grading their assignments. He mentioned the stress he felt as an instructor and that the stress was largely due to being worried about the students’ learning and performance in the course. Specifically, he said, “Students who were not performing as well on assignments were also not utilizing my feedback and guidance in order to submit higher-quality work.” This issue was exacerbated after the course was moved online and he figured that students were less responsive than before. His circumstances as a student and dealing with being far from family who were also struggling with losing income and putting their education on hold added to the overall stress he was feeling at this time.

High stress due to change in the mode of instruction

This group had three prominent categories, including work burden, constraints, and change in nature of tasks or environment. Gretchen and Oakleigh (pseudonyms) both reported high levels of stress, maintained the same duties, and felt they had less confidence, yet both discussed different stressors that brought about these results. Gretchen was equally as affected by stressors related to her TA duties as those outside of teaching. She reported having the same general teaching responsibilities after the move to online instruction, though she expressed lower confidence in TA duties given the formatting changes in the course she taught. Her largest concern was the difference in presenting information online instead of in person. She explained, “I had not used [an online teaching platform] pretty much at all before. ... I didn’t know how to create a meeting. ... I didn’t really know how to do any of it.” Earlier she engaged students in using whiteboards, and the immediate verbal and nonverbal feedback from students was important for her to know if they understood what was being covered in class. However, when teaching online, Gretchen was concerned about student learning, especially because some students did not turn on their cameras. Although she felt less confident with online teaching overall, she was hopeful that her confidence would improve after she gained experience teaching in this new setting. She expressed that she did gain that confidence when she said, “Now I feel really comfortable using [the online teaching platform].” But she also mentioned about feeling stress that was compounded by having to figure things out on her own while also transitioning to her own online classes as a student. She said, “It made me more nervous as a student ... especially with my premedical course load, I kind of was wondering how they would teach us those topics online.”

Oakleigh was more stressed by things outside of her TA responsibilities and mentioned specific personal issues as well as general concern for society, which was likely felt by many during this time. She was “very concerned with her teaching at the moment” but understood that she still needed to “grade papers so students can have feedback.” She was concerned with her students’ performance going forward and said she planned on “not grading as intensely but providing as much feedback as possible and useful information for the remainder of assignments.” The largest influence the disruption had on her stress levels was the overwhelming number of emails, reminders, and changes in her courses as a student. She said, “I am somebody who keeps up with the majority of the grading, but then also having all the expectations in other classes did make it almost less of a priority to get those papers graded that early.” She was disappointed with how she was able to handle her TA responsibilities, and the added stressor of dealing with personal issues complicated matters even further.

Low stress but no change in regular duties

Four categories were prominent in this group, including distant or removed from responsibilities, change in nature of tasks or environment, student engagement, and student success. Of the five TAs who reported low levels of stress, four of them (pseudonyms: Ava, Elizabeth, Amare, and Naomi) had the same responsibilities when classes moved online and had no change in their confidence levels. Ava said, “I think the biggest issue for me will be trying to develop a productive routine when I am used to having a mental separation between campus and home.” To cope with these issues, she created a schedule for each day and sought information on how to use online teaching platforms. Having these things in place motivated her to stay on top of her own school and research work so that she would be more equipped to support her students. Aside from being concerned that her students would be more distracted learning online, Elizabeth believed working remotely as a TA would go well but was concerned with the classes she was taking as well as her certified nursing assistant certification training. Furthermore, she was “more concerned about other people in my community and especially my parents,” who own a small business and work in health care.

We found that some TAs commended their professors, who were the instructors of the course, for organizing the online instruction in an effective way, which caused low stress to them in their TA roles. For instance, Daniella mentioned that her supervising professor had organized the course well, which gave her an added sense of support. Additionally, she felt prepared to move the course online after she took time to learn the online teaching platform she would be using and its functionality for teaching. Similarly, Amare felt low stress and thought moving things online would be straightforward and streamlined because “adequate information was disseminated by the coordinating professor” and “an adjustment time was given, which helped manage the stress.” He did express concern over students’ dedication and attention to class but believed that by asking students questions, he could “ensure they are really attentive.” Naomi reported low stress with the same responsibilities when moved online, but unlike Ava, Elizabeth, and Amare, she felt slightly lower confidence in her TA duties, saying, “It is going to be difficult not having the students in front of me to provide immediate feedback to how they are understanding the material and how I am teaching it.” Her main concern regarding shifting to online was how she would connect with students while not being able to get immediate feedback from them and that students would be more easily distracted. This concern made her less confident that teaching through an online platform would be as effective as in person.


This study seeks to provide an understanding of how self-efficacy of TAs within science courses were affected by COVID-19 and how to best support them in their roles as instructors. Although the study found no statistical relationships between stress and teacher self-efficacy among participating TAs during the spring 2020 semester, the qualitative results show that the changes in the needs and demands of the instruction and TAs’ duties and responsibilities affected their stress levels. Qualitative data further show a more nuanced perspective than the quantitative data, as some TAs indicated that increased stress was influenced by the nature of how their TA duties changed during the pandemic. We found that the TAs who were moderately to highly stressed due to the COVID-19 disruption and had the same responsibilities when classes moved online reported lower confidence when teaching online, whereas those who had reduced duties when moving online reported no change in their confidence.

Relating these findings to the model of TA teacher efficacy by Smith and Delgado (2021), we see that stressors can influence TAs’ confidence in their ability to carry out teaching tasks. Although we did not measure teaching performance as part of this study, the model suggests that stress may further contribute to whether TAs focus more inwardly or outwardly in their teaching, thus impacting their self-efficacy and teaching performance. Past studies have reported that classroom teachers with low confidence preferred traditional science teaching approaches, such as reading and worksheets, and avoided hands-on science (Appleton & Kindt, 2002). Having stress and anxiety about one’s responsibilities could lead to lower self-efficacy and increased teacher-centered instruction. In this study, the stressors that TAs experienced ranged from personal issues and general concern regarding the pandemic to concern for engaging students with newer technology while at the same time being overloaded with their own classes as students and learning a newer platform for their own coursework. The model of TA teacher efficacy emphasizes one’s own experience as a large source of confidence that leads to student-centered teaching practice (Smith & Delgado, 2021). As TAs gain more experience, they learn to repeat what they perceive to have worked well and to not repeat what has not worked well, which helps them develop more realistic expectations of how students will respond to certain teaching strategies. Although many of the TAs in this study had multiple semesters of teaching experience, and they all had a half-semester of experience prior to the disruption, the lack of experience in online teaching may have had an impact on those TAs who felt less confident when moving their classes online.

The current situation of education during the pandemic has given us a lot to consider about how to structure higher education. There is, however, something to be said for other stressful events in TAs’ experiences as students, researchers, instructors, and with any other hats they wear, as well as how these events impact their teaching performance. This study used a unique experience to expose the stressors that TAs face on a day-to-day basis (such as work overload, time and resource constraints, and feelings of being underprepared) to understand how stress relates to teacher efficacy. Although their feelings were perhaps heightened by the pandemic, TAs reported stressors that may have impacted their confidence in their teaching abilities, which influences one’s persistence and resilience to overcome obstacles to teaching effectively (Tschannen-Moran & Johnson, 2011).

There is a greater need for more professional development opportunities for TAs that include knowledge and use of technological teaching tools. Additionally, programs should address virtual instruction and how to use teaching technology to effectively deliver material electronically in instances when similar, if only brief, disruptions occur due to inclement weather, natural disasters, or other events outside of TAs’ control that prevent class from meeting face-to-face. In addition to learning about virtual teaching technology and techniques, TAs would benefit from learning strategies for time management and efficient grading and planning techniques to help mitigate the stress associated with being overloaded with work. That said, TAs can be more prepared to balance their teaching duties by more efficiently managing the time and energy they put into preparing for and leading their courses. Outside of preparing TAs for their roles, those educating TAs should offer or provide referrals to mental health resources and strategies for managing stress to support TAs as they balance the responsibilities of their many roles. Future studies exploring the role of stress on TAs should include protocols for measuring teaching performance to better understand the role that stress and its interaction with teacher efficacy plays on whether TAs carry out student- versus teacher-centered teaching practices.

Cody Smith ( is an assistant professor in the Department of Kinesiology at Missouri State University in Springfield, Missouri. Deepika Menon is an assistant professor in the Department of Teaching, Learning, and Teacher Education; Jenny Dauer is an associate professor; and Annette Wierzbicki is a research technician in the School of Natural Resources, all at the University of Nebraska-Lincoln in Lincoln, Nebraska.


Appleton, K., & Kindt, I. (2002). Beginning elementary teachers’ development as teachers of science. Journal of Science Teacher Education, 13(1), 43–61.

Bandura, A. (1997). Self-efficacy: The exercise of control. Freeman.

Chesnut, S. R., & Cullen, T. A. (2014). Effects of self-efficacy, emotional intelligence, and perceptions of future work environment on preservice teacher commitment. The Teacher Educator, 49(2), 116–132.

Chirikov, I., Soria, K. M, Horgos, B., & Jones-White, D. (2020). Undergraduate and graduate students’ mental health during the COVID-19 pandemic. UC Berkeley: Center for Studies in Higher Education.

Cohen, S., Kamarck, T., & Mermeistein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 386–396.

Corbin, J., & Strauss, A. (2008). Basics of qualitative research: Techniques and procedures for developing grounded theory (3rd ed.). Sage.

DeChenne, S. E., Enochs, L. G., & Needham, M. (2012). Science, technology, engineering, and mathematics graduate teaching assistants teaching self-efficacy. Journal of the Scholarship of Teaching and Learning, 12(4), 102–123.

DeChenne, S. E., Koziol, N., Needham, M., & Enochs, L. (2015). Modeling sources of teaching self-efficacy for science, technology, engineering, and mathematics graduate teaching assistants. CBE—Life Sciences Education, 14(3), Article 32.

Evans, T. M., Bira, L., Gastelum, J. B., Weiss, L. T., & Vanderford, N. L. (2018). Evidence for a mental health crisis in graduate education. Nature Biotechnology, 36(3), 282–284.

Fagen, A., & Wells, K. S. (2000). National Doctoral Program Survey. National Association of Graduate-Professional Students (NAGPS).

Gardner, G., & Jones, G. (2011). Pedagogical preparation of the science graduate teaching assistant: Challenges and implications. Science Educator, 20(2), 31–41.

Holmes, N. G., Martinuk, M., Ives, J., & Warren, M. (2013). Teaching assistant professional development by and for TAs. The Physics Teacher, 51(4), 218–219.

Klassen, R. M., & Chiu, M. M. (2011). The occupational commitment and intention to quit of practicing and preservice teachers: Influence of self-efficacy, job stress, and teaching context. Contemporary Educational Psychology, 36(2), 114–129.

Mazzola, J. J., Walker, E. J., Shockley, K. M., & Spector, P. (2011). Examining stress in graduate assistants: Combining qualitative and quantitative survey methods. Journal of Mixed Methods Research, 5(3), 198–211.

McHugh, M. L. (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 22(3), 276–282.

Menon, D., & Azam, S. (2021). Preservice elementary teachers’ identity development in learning to teach science: A multi-site case study. Journal of Science Teacher Education, 32(5), 558–577.

Reeves, T. D., Hake, L. E., Chen, X., Frederick, J., Rudenga, K., Ludlow, L. H., & O’Connor, C. M. (2018). Does context matter? Convergent and divergent findings in the cross-institutional evaluation of graduate teaching assistant professional development programs. CBE—Life Sciences Education, 17(8), 1–13.

Shum, A., Lau, P., & Fryer, L. (2020). From learner to teacher: (Re)training graduate teaching assistants’ teaching approaches and developing self-efficacy for and interest in teaching. Higher Education Research & Development, 40(7), 1–18.

Smith, C. R., & Delgado, C. (2021). Developing a model of graduate teaching assistant teacher efficacy: How do high and low teacher efficacy teaching assistants compare? CBE—Life Sciences Education, 20(1), Article 2.

Tellis, W. (1997). Application of a case study methodology. The Qualitative Report, 3(3), 1–19.

Tschannen-Moran, M., & Johnson, D. (2011). Exploring literacy teachers’ self-efficacy beliefs: Potential sources at play. Teaching and Teacher Education, 274(4), 751–761.

Woolston, C. (2019). Ph.D. poll reveals fear and joy, contentment and anguish. Nature, 575, 403–406.

Wyse, S. A., Long, T. M., & Ebert-May, D. (2014). Teaching assistant professional development in biology: Designed for and driven by multidimensional data. CBE—Life Sciences Education, 13, 212–223.

Administration New Science Teachers Pedagogy Professional Learning STEM

Asset 2