Explore, engage, empower: methodological insights into a transformative mixed methods study tackling the COVID-19 lockdown

The study addresses how a transformative mixed methods research design can simultaneously achieve epistemic (knowledge production) and transformative (action-oriented) objectives under urgent and uncertain conditions, such as the first COVID-19 lockdown in Switzerland. Positioned within traditions of action research, participatory action research, and transdisciplinary and transformative research, the authors argue for relational, interactive research that co-produces knowledge and change. They note the need to reflect on methodological implications, integration across components, and power dynamics. The research questions are: (1) How do different population groups experience and cope with the COVID-19 lockdown in Switzerland, and what role do housing conditions play? (2) To what extent can a transformative mixed methods design leverage research to achieve epistemic and transformative objectives in times of crisis? The project, Swiss Corona Citizen Science, intentionally sought to explore, engage, and empower during the crisis by combining quantitative, qualitative, and collaborative methods and by providing spaces for mutual learning.

The paper situates itself within several strands of scholarship: action research (e.g., Lewin) and participatory action research (e.g., Fals Borda), which seek to produce knowledge while inducing social transformation; transdisciplinary and transformative research addressing socio-ecological crises, emphasizing co-production of knowledge, inclusion of diverse perspectives, and outcomes such as awareness and capacity building, empowerment, and network creation. Mixed methods literature emphasizes planned integration across components and attention to social processes and power dynamics in implementation. The authors highlight critiques about conflating knowledge generation and societal change, and the need to explicitly manage tensions and conflicts when pursuing multiple objectives concurrently, especially in crises.

Design: A transformative mixed methods, sequential multi-stage design with increasing intensity of citizen involvement and decreasing participant numbers across components. Components and timeline (all online due to lockdown): - Country-wide online survey (Apr 8–May 10, 2020; disseminated via EPFL press release, website, social media, researcher networks; languages: German, French, Italian, English). Aim: explore impacts of lockdown on housing and well-being; provided supportive resources (e.g., counseling hotlines, domestic violence reporting, solidarity networks); respondents could opt-in for later components. - Mobile crowdsourcing app (Challenges): Challenge 1 (Apr 23–May 18, 2020), Challenge 2 (Apr 30–May 25, 2020). Tasks guided reflection and sharing (close-ended items, open narratives, images) about daily life during lockdown and visions for post-crisis futures. Recruitment from survey opt-ins and public channels. - Semi-structured interviews (Apr 20–May 20, 2020): 60 self-selected survey respondents; explored adjustments in living spaces, resources (material, temporal, social, emotional, physical), temporal evolution, and expectations. - Citizen Think Tanks (CTTs): Two rounds of online sessions per topic (May 27–28 and Jun 10–11, 2020) co-run by citizens and scientists using formative scenario development. Five topics: housing, mobility, local economy, digital governance, tourism. Built on survey and app results; reports co-authored by citizens. Integration strategy: - Design-level integration: sequentially linked questions/tasks; sampling integration by inviting participants onward; agenda-setting for CTTs from survey/app findings; merging datasets and creating narratives in analysis. - Citizen perspective: scaffolded individual and social learning through progressive engagement; citizens became researchers of their own situation and engaged in mutual learning. - Collective perspective: integrated diverse life-world knowledge to address an emergent phenomenon, culminating in dialogical formats to co-create post-crisis scenarios. Analysis: - Quantitative: Descriptive analyses (R) on survey (n=6919; 6269 residing in Switzerland) and close-ended app items (sentiments, challenges). - Qualitative (app): Inductive content analysis of open-ended responses on changes/challenges/recommendations during lockdown, capturing manifest content (what) and latent meaning (how: description, assessment, uncertainty, recommendation); one coder with iterative team discussion. - CTTs: Analysis of audio/whiteboard notes focusing on interests/expectations, participation forms, interaction directions, discussion characteristics; short evaluation survey on perceived participation. - Images: Inductive synthesis and curation of image themes (privacy preserved) into mosaics for changes during lockdown and preferred futures. Ethics and implementation context: - Ethics: Approved by the Human Research Ethics Committee of EPFL. Some sensitive health/economic questions were removed due to expedited timelines for additional approvals. Anonymization prevented linking respondents across components. - Crisis-driven constraints: Rapid virtual team formation over 3 weeks; all-digital data collection; adaptations to evolving lockdown measures; balance between methodological rigor (e.g., sampling, pre-testing) and timeliness/societal relevance.

Explore (epistemic insights): - Mental state during confinement (survey, n=6269): ~40% felt the same as before; 37% more anxious; 28% more depressed; 46% lacked physical interactions; 59% missed loved ones (20% moderately agreed); 11.5% reported improved state of mind. - Daily structure and time (app): Predominant sentiment was loss of structure/sense of time; routine activities (e.g., shopping) became anchors; ~two-thirds struggled to separate work from leisure/care/family time. - Home transformations (survey; n answers=3110; 38% made at least one change): 36% reorganized rooms during the day; 30% moved furniture; 28% changed room uses. App images showed plants, deliveries, hygiene materials, and improvised work/education/cooking setups, indicating creative adaptations to telework and space constraints. - Gendered distribution and emotions: Among parents (n=1830), 5/10 women vs 1/10 men reported being exclusively in charge of homeschooling. Emotions: most common was “calm” for both men (23%) and women (20%); second most for men “confident” (16%), for women “tired” (19%), suggesting disproportionate burden on women. Engage (transformative engagement breadth/depth): - Participation numbers: Survey 6292 total (6269 CH residents analyzed). App: 216 (Challenge 1), 163 (Challenge 2). Interviews: 60. CTTs: 28 and 27 participants in two rounds respectively. Sequential design retained a subset across components. - Sample composition: Overrepresentation of French-speaking residents, women (64%), and higher education (51% with higher diploma vs 44% in Swiss population). Underrepresentation of working-class/industrial sectors; sessions mainly in French limited DE/IT-speaking participation. Notable sub-groups reached: many health and social care professionals; 195 respondents who lost their jobs. - Quality of engagement: App prompted intimate, creative, proactive contributions (texts, images, drawings), e.g., coping with limited access to nature via home gardening. CTTs fostered dialogical engagement, debates, expanded agendas, and empathic exchanges; in the housing CTT, 71% reported the sessions allowed them to express their ideas. Empower (abilities to act): - App recommendations reflected reflexivity and agency: establishing routines (sleep/wake, dress for work-from-home), mental health time, new leisure (home gardening), quality time/remote socializing, deceleration of life’s pace. - Future visions (survey/app/CTTs): Immediate desires post-lockdown included seeing family/friends (~30%; women 34% vs men 23%) and travel (men 16% vs women 11%). Preferred futures emphasized environmental protection, sustainability, solidarity, harmony with nature, reduced consumption, and slower pace. CTT evaluations indicated scenario work stimulated reflection on desirable futures and understanding of personal contributions to change; fewer felt like “being a researcher,” suggesting limited role-blurring within the short timeframe.

The transformative mixed methods design enabled simultaneous pursuit of exploration, engagement, and empowerment during a crisis, addressing the research question of leveraging research for epistemic and transformative aims. However, three key tensions emerged: - Unequal engagement opportunities: Despite wide survey reach, participation skewed toward French-speaking, higher-educated women, with underrepresentation of working-class groups. Sequential narrowing deepened initial biases and raised ethical questions about time/emotional demands on crisis-burdened citizens. - Navigating social and epistemic control: Balancing objectives required continual negotiation within the interdisciplinary team and with citizens. Design choices (e.g., creative vs standardized prompts) reflected competing priorities. Urgency limited conceptual/theoretical integration; divergent epistemologies (individual vs systemic notions of housing) surfaced, prompting reframing to “domestic living space.” In dialogical formats citizens contested/modulated agendas, reducing researchers’ control over data generation and outcomes, underscoring distinctions between epistemic quality across methods and the entanglement of social and epistemic control. - Situatedness and positionality: Shared crisis context shaped roles; researchers’ reflections on privilege and gender influenced design aims. Attempts to blur roles (citizens as co-researchers; scientists as citizens) were partial; researchers still pre-structured the space, yielding mostly “invited” rather than “claimed” participation. Nonetheless, sequential engagement supported appropriation of the research space and mutual learning. Overall, the findings indicate that transformative mixed methods can generate insights and support action, provided integration and power dynamics are explicitly addressed and crisis-driven constraints are considered.

The study demonstrates that a transformative mixed methods approach—combining survey, mobile crowdsourcing, interviews, and Citizen Think Tanks—can simultaneously deepen understanding of crisis effects on domestic living spaces and support engagement and empowerment through reflexive and dialogical processes. Methodological contributions include: articulating integration from scientific, citizen, and collective perspectives; operationalizing sequential engagement to scaffold learning; and exposing tensions in social/epistemic control and positionality under crisis conditions. For future crises, the authors recommend: explicitly distinguishing and negotiating multiple objectives; designing integration across perspectives while attending to crisis-induced needs; addressing social/epistemic hegemonies to build transformative capacity in heterogeneous alliances; adapting institutional and organizational conditions for agile, cross-sector research collaborations; and fostering researchers’ preparedness for evolving roles. The paper calls for mainstreaming action-oriented, transformative mixed methods research and cultivating a research culture that supports societal transformations.

- Sampling and representativeness: Overrepresentation of French-speaking, higher-educated women; underrepresentation of working-class/industrial sectors; language constraints (predominantly French CTTs/interviews) limited national inclusivity. - Sequential attrition: Intentional reduction in participant numbers across stages amplified initial sample biases and risked engaging mainly the “usual suspects.” - Measurement and linkage: Anonymization and ethics constraints prevented linking individuals across components, limiting analyses of longitudinal/participation trajectories; some sensitive questions were removed due to expedited approvals. - Crisis-driven trade-offs: Limited time for pre-testing, sampling rigor, and theoretical integration; all-digital formats may have shaped participation and interaction quality; easing of measures altered timelines and agendas. - Impact assessment: No systematic ex-post impact study; uptake of provided support resources and degree of sustained empowerment remain unknown; not all data/components (e.g., interviews in depth) could be included in the methodological reflection within space limits.