During the early stages of the COVID-19 pandemic, we conducted a small exploratory study focused on how US high school and college students navigated the many disruptions to their lives by collecting journals written by students themselves, which captured everyday conversations about COVID-19 taking place within their social networks. The COVID-19 pandemic posed challenges to research and data collection that are similar to those documented in the study of disasters.Reference Fine and Abramson 1 , Reference Tierney 2 In this brief report, we discuss our approach to using journaling techniques to address 3 key challenges to collecting data during a disaster like a pandemic— 1) accessing respondents when preparation time and resources are limited; 2) ensuring protection for both participants and researchers in a context when human interaction is severely constrained; and 3) needing both rapid response and flexibility in research design and data collection.
Methods
We recruited 8 students (4 high school and 4 college, henceforth “journalers”) to record field notes (henceforth “journals”) 3 times a week of anonymized conversations (in person or virtually) and reflections about COVID-19. Journalers were asked not to change anything about their daily lives for the project, thereby protecting their safety. Our data set included over 460 entries. Both groups were compensated. Entries were anonymized in 2 phases, once during data collection and again during analysis. Human subjects’ approval was obtained from the University of Maryland, College Park, and American University.
The data were collected in the early months of the COVID-19 pandemic, starting in April 2020 and continuing until August of that year. During this dynamic period, stay at home orders were implemeted and revoked, education went online, and the economy closed and later re-opened.
Results
Accessing respondents
During the early stages of a disaster, it is generally difficult, unsafe and/or unethical to access respondents for in-person data collection. This was certainly the case with COVID-19. Due to time and financial constraints, we utilized a small convenience sample recruited through personal networks. While the participants represent a highly select group, the journal method expands the reach of the data beyond the individual and into the network, reporting and reflecting on the experiences of others they knew, heard, saw online, or interacted with, as shown in the following example. In this entry, one of the high school journalers shares a conversation that they had over a social media app with friends who were starting college later that year, bringing levity to a stressful and unpredictable reality about temporary classrooms being built by their university to reduce COVID transmission:
Person A: “on [University X online portal] it says [University X] is building “TC” [temporary classrooms] for some of the classes. …
Person B: those look like fun
Person C: that’s crazy
Person D: outdoor classes
Person E: but will they be… air conditioned?… a fan at least?
Person D: so fun
Person F: we can only hope… apparently my colloquium is in a parking garage? So I might be needing one of those fans Iamo [sic]
Person G: This is going to be a freshman year for the history books
(HS3, 12 July 2020).
This conversation shows how a group of young people discuss and cope with uncertainty as the start of their first year of college approached. It gives direct insight into practices of resilience– while they were still anxious and unsure of what the coming year would look like, they were able to offer some relief through their exchange. While that student could have discussed this conversation with a researcher during a one-on-one interview over videoconference, there is additional value in this data insofar as it captures digital interactions as they occur. Within the dataset, over 20% of the recounted conversations occurred virtually – such as through FaceTime, Instagram, Facebook messenger, or other online video chat and instant messaging platforms.Footnote a Journaling thus provides an opportunity to study human interaction as it is occurring.
Ensuring protection for both participants and researchers
A second challenge to data collection during a disaster is concern about human subjects’ protection. By taking place at a distance, journaling leaves the physical health and safety of both participants and researchers unchanged, as well as affords a high amount of autonomy to the journaler. An unsurprising and consistent theme we observed over the 4 months of data collection was concern for personal and familial health, and the sometimes ensuing tension with others over different precautions and activities. This was particularly salient for the college journalers because moving back to one’s family home was a common experience as residence halls closed and classes transitioned online; their lives were also not as powerfully circumscribed by parents and guardians as the high school student group. The entry below from a college student journaler discusses an argument with a sibling living in their household, who had planned to see her boyfriend who was exhibiting COVID-related symptoms:
I was in the living room watching some TV when my sister … said how her boyfriend thinks that he has COVID. … She elaborated that he feels exhausted, has a fever, and headache but no respiratory issues thus far, and that he is going to go get tested in a couple days. I said it’s good he’s getting tested and I mentioned that she should probably get tested too. She said she wasn’t going to go get tested because she hasn’t seen him since before the weekend, and it didn’t make sense unless his test came back positive. I said, “Alright, just don’t go over to his house until you know that he’s negative.” She was sort of quiet after I said this, which made me somewhat suspicious …She said that she was planning on bringing him food later and that it’s not a big deal since he probably doesn’t have COVID… I blew up on her saying that what she was doing is selfish, seeing someone who is clearly sick, and potentially exposing our family to the virus… (COLLEGE1, 13 July 2020).
In the initial stages of the pandemic, coronavirus was known as an illness that disproportionately affected older populations. Journal entries show that college journalers, in particular, were concerned about their older parents and household members they had moved back in with, and were sometimes upset with those they felt were not being vigilant enough to protect others. Journaling, then, allowed us to be privy to these sorts of conflicts in real time rather than recalled after the fact.
Rapid response and flexibility
A third challenge when studying disasters and other similar shocks, such as pandemics, is ensuring that valuable knowledge about the event is not lost to time or other imminent threats. Because it can be implemented quickly, journaling as a data collection method allowed us to gather data as the COVID pandemic unfolded, keeping pace with a fast and ever-evolving public health crisis. The method is flexible and does not involve a high barrier to start up, as a longer survey or even the development of an in-depth qualitative interview guide may take.
While we instructed the journalers to recount and reflect on conversations about COVID-19, by late May 2020, events which we could not have predicted at the time were causing large societal ripples – namely, the murder of George Floyd at the hands of the state and subsequent nationwide protests. Accordingly, many of the journalers’ entries began to center around the movement for Black lives. Our data organically captured this point in time by including these reflections, unprompted. One college student journaler took on the topic of attending a protest. She shared a conversation with a concerned family member, who she called before leaving:
… I tried to reassure her that the protest wouldn’t get violent and if it did I would leave. I also explained that I needed to support the cause and that it was bigger than just me. She reluctantly said okay and just reiterated that she wanted me to be safe, explaining that I needed to wear a mask, socially distance from others, and wear a mask when I am in the car with [my friend]. I reassured her again that I would do all of those things and make sure I stayed safe. We hung up the phone and I felt tired of having to rationalize myself to my family when I wanted to go somewhere or do something. … But I have to make the best of the situation I am in and try not to upset the people I am living with. (COLLEGE4, 29 May 2020).
For this journaler, we see grappling with the COVID pandemic as it intersected with other national crises of competing import and concern. The entry also shows the push and pull between the young person and their family members, whose COVID risk was also tied to the journaler’s actions.
If we had pursued more conventional data collection methods, such as structured interviews or surveys, we would not have included questions about the 2020 uprisings, as they had not yet occured when our data collection began. If we wanted to pivot to include more of those questions, we would have had to develop new study protocols and a new (or modified) IRB submission, taking additional time and resources. While it is possible that these issues would have come up in a well-designed interview, the journal method has an acute strength in its flexibility and openness in a rapid-response environment.
Discussion
In this brief report, we drew on our experience during the COVID pandemic to demonstrate the potential for journaling as a feasible, efficient and effective method for collecting data on the lived experience of a disaster. We highlight 3 key advantages: accessing respondents, ensuring protection of human subjects, and flexibility. The journaling method allowed us access to both their and their social networks’ experience and also mobilized the population of interest as participant researchers about their own lives. Doing so centers the narratives and voices of affected communities, which may be particularly important with populations considered “hard to reach.”Reference Elliott, Watson and Harries 3 , Reference Mutchler, McKay and McDavitt 4 The use of journaling also insulates from any perilous external conditions, like a virus, and thus ensures human subjects’ protection. Lastly, journaling is nimble and can easily adapt to the changing severity and timeline of the disaster.
Similar to how Peek and Fothergill noted the utility of focus groups as a method for studying vulnerable groups post-disaster, we found that journaling provided focused time for participants to reflect on the events of the week related to the stresses caused by the COVID pandemic.Reference Peek and Fothergill 5 Studies have shown that journaling can provide potential benefits for mental health of young people and others.Reference Utley and Garza 6 More recently, audio diaries by youth in Gaza document the trauma of conflict but also show how they can be a tool of empowerment and resilience.Reference Mellen and TerBush 7 Thus, the selection of methods and how they are used during the data collection process can sometimes confer unexpected positive outcomes for participants.
Limitations
A key limitation of our study is sample selectivity. By definition, a convenience sample tends to leave out certain population groups. Future research could be done in collaboration with multiple institutions that serve different student populations. Triangulation through larger, population-based data can also corroborate or challenge study findings.
Conclusion
Researchers have used journaling in studies of HIV in Southern Africa,Reference Watkins 8 the initial COVID-19 lockdowns in Wuhan, China,Reference Yang 9 and the long-term economic impact of the pandemic.Reference Barford, McCarthy and Mueller 10 As much qualitative social sciences disaster research is conducted after the fact,Reference Tierney 2 journaling warrants further experimentation in both public health and disaster studies given the opportunity to study unfolding events and provide real-time data. The method is promising for future study of climate disasters, emergencies and hazards, and in other community-oriented research, as an alternative to more conventional methods such as interviewing, or as mixed-methods complements. We thus close with a call for including journaling as part of the toolkit for data collection during disasters.
Author contribution
DJ: Analysis, writing, editing. NA: Conceptualization, analysis, writing, editing. SM: Conceptualization, analysis, writing, editing. AH: Analysis, editing. KH: Analysis, editing. CMG: Analysis, editing.
Funding statement
For funding support, we thank the College of Arts and Sciences at American University.
Competing interests
No conflicts of interest to report.
Open access
