Qualitative research has gained legitimacy in a wide range of scientific fields because it enables the description and interpretation of in-depth data (Liao et al., 2017), generating grounded theories of the social world (Glaser and Strauss, 2017). It operates in the world and in real situations; it is contextualized; it studies social processes; and it is holistic, inductive, flexible, analytical and systematic (Gayou-Jungerson, 2003). Qualitative research, which occupies a relevant position in the field of educational research (Hennink et al., 2020; Johnson and Christensen, 2014), through its assumptions, allows for the development of reasoned proposals, explanations of phenomena and informed decision-making for educational action, contributing to both theory and practice (McMillan and Schumacher, 2005).

Qualitative research can be conducted using different epistemological approaches (Doherty et al., 2015), and data collection can take place in different ways. The common feature of all approaches is the close relationship between the researcher and participants (Sousa et al., 2020). This element is called into question in times of crisis, be it the Covid-19 pandemic or natural disasters, conflicts and wars.

In January 2020, the Coronavirus disease was declared a public health emergency of international impact. On 5 May 2023, three years after its announcement, the World Health Organization declared the end of the Covid-19 pandemic. The pandemic had several negative consequences worldwide, not only restricting global economic growth, but also affecting various dimensions of human social life (Jain and Brockova, 2022). These three years also brought profound changes in research practices. In the context of qualitative research, which presumes the proximity between the researcher and the participants, data collection has been compromised due to the regulations and restrictions imposed as a result of the pandemic, with conducting face-to-face interaction with individuals and field studies being restricted indefinitely.

In this scenario, collecting qualitative data through digital tools has been seen as a promising solution (Dodds and Hess, 2021; Vindrola-Padros et al., 2020) in terms of overcoming physical, psychological and ethical challenges (Santana et al., 2021). Videoconferencing platforms, online audio/video calls, telephone or mobile phone calls, emails, online surveys and text chatting are some examples of digital tools used to collect qualitative data (Deslandes and Coutinho, 2020; Jain and Brockova, 2022; Sah et al., 2020).

While digital tools have been used since the 1990s, and the debate about their advantages and limitations was already present in the scientific literature (Thomas Dotta et al., 2019), with the Covid-19 pandemic, and the restrictive measures imposed on face-to-face interactions, their use was expanded (Nind et al., 2023). Despite the advantages, notably because of the possibility of ‘making research happen’ (Nind et al., 2023: 5) during the pandemic, several limitations were identified, especially when transposing traditional methods to virtual environments. In the context of the relationship between researcher and participant(s), crucial elements of qualitative data collection constraints include difficulties in reading body language and facial expressions (Deslandes and Coutinho, 2020; Pocock et al., 2021), barriers to building trust (Kobakhidze et al., 2021; Walker et al., 2021; Webber-Ritchey et al., 2021) and much shorter and less contextual responses from participants (Davies et al., 2020). Concerning access to and from participants, the use of a videoconference platform could exclude those with little or no access to technology (Chatha and Bretz, 2020; Dodds and Hess, 2021; Melis et al., 2021), exacerbating issues such as representational justice (Kobakhidze et al., 2021; Otto and Haase, 2022; Wa-Mbaleka and Costa, 2020) and the under-inclusion (Gilmore-Bykovskyi et al., 2021) of vulnerable people and groups. In the field of ethical issues, weaknesses have also been identified, including the difficulty of protecting confidentiality through virtual connections (Kobakhidze et al., 2021; Rahman et al., 2021), obtaining the free and informed consent of vulnerable populations, and knowing whether compensation is merely an incentive or potentially coercive (Santana et al., 2021).

The constraints on data collection with digital tools mentioned in studies undertaken during the Covid-19 pandemic are the same as those identified in studies carried out before the pandemic (Aborisade, 2013; Bampton and Cowton, 2002; Hamilton and Bowers, 2006; James, 2007; Thomas Dotta et al., 2019). The increased risks in times of crisis and emergency are linked to the adoption of quick fixes, which are not subject to a process of evaluation, critical discussion and grounding in studies that have already been carried out. For Nind et al. (2023), it is not sustainable for a researcher to find their way into research without a foundation in evidence based on existing knowledge. The authors consider that it is essential to consult analyses of emerging and tested methodological adaptations, even though they know that the method must be adapted to each situation in all cases. Therefore, the influence of the data collection method and its possible constraints must be taken into account in the analysis and interpretation process. However, researchers must communicate their findings transparently and prove how data collection mediated by digital tools has impacted the analysis, interpretation and dissemination of their results (Pocock et al., 2021).

In addition to the potential and limitations of data collection mediated through digital tools that can be found in the literature, it is worth highlighting expressive and creative forms of research that have been adapted to the context of the Covid-19 pandemic and the resources available, including letter writing (Maycock, 2021), diary writing (Scott et al., 2021), digital narratives (Malachowska et al., 2021), photographs and drawings (Isaacs et al., 2021) and other arts-based methods (Green et al., 2022). These methods, as well as not requiring significant adjustments to the pandemic, have promoted expression in any form available to communicate lived experiences (Nind et al., 2023).

The context of formal education (particularly preschool and K–12 schools, as well as higher education) as a topic of study, as a context for scientific intervention or as a target population for research (teachers, principals, students and parents, for example) was strongly impacted by this period of crisis, mainly due to the characteristics of the established types of participation in the constitution of qualitative research in education (Matos et al., 2023). According to Matos et al. (2023), higher education, as the context mainly responsible for promoting and developing research, currently recognises the major challenges for research being carried out in preschool and K–12 schools, highlighting the need to reconsider the ongoing types of research culture that are being undertaken.

During the Covid-19 pandemic, as already mentioned, strict restrictions were imposed, especially in the methodological area, which required the implementation and adaptation of strategies to ensure the continuity of scientific research. Literature reviews aimed at consolidating and systematising the available evidence, specifically in relation to qualitative data collection, can serve as a basis for understanding how a variety of methods can be used and adapted to contexts of physical distancing, as well as providing explanations of the challenges associated with their use (Hall et al., 2021).

Cretu and Ho (2023) conducted a bibliometric analysis of the impact of the Covid-19 pandemic on educational research, focusing on publications during 2020 and 2021. Their findings revealed a concentration of studies in chemical and medical education, predominantly from the USA, emphasising online learning, curricular approaches and the psychological effects of the pandemic on education. Other reviews examined the effects of the pandemic on education (Altinsoy and Boyraz, 2022; Sánchez Amate et al., 2021) and of remote teaching in higher education (Sum and Oancea, 2022). Ruiz-Real et al. (2020) found that health-related studies dominated Covid-19 research, with social sciences, particularly education, focusing on online education. Roychowdhury et al. (2022) highlighted themes such as psychosocial issues and economic concerns, identifying the USA as a central hub. No systematic reviews on qualitative data collection in school education were found.

Considering this background, we sought to conduct a systematic mapping (James et al., 2016; Newman and Gough, 2020) of studies conducted in schools during the Covid-19 pandemic that used digital tools for qualitative data collection. Our findings identified key themes within these empirical studies and problematised the transition from traditional face-to-face to online data collection processes. We aimed to provide insights into the use of these tools, contributing to the advancement of knowledge in the methodological domain during periods of crisis affecting the traditional use of qualitative data collection methods.

Systematic mapping provides an overview of a particular research topic (Haddaway et al., 2016), identifies the gaps and knowledge clusters and characterises the methodological designs (James et al., 2016; Newman and Gough, 2020). In a systematic mapping study in social sciences such as this one, critical appraisal of individual sources of evidence (to judge the trustworthiness of the studies) is optional. Given the objectives of this study, which involved analysing broader topics within studies undertaken in K–12 school contexts, the inclusion of a critical appraisal of individual sources of evidence is not applicable (Haddaway et al., 2016). The coding scheme for systematising results was designed to characterise studies on the use of digital tools in the collection of qualitative data in studies conducted during the Covid-19 pandemic focusing on schools. It was based on the assumptions of thematic analysis (Vaismoradi et al., 2013).

Systematic mapping seeks to identify and classify primary studies in a specific subject area, including research questions that are answered by searching for, identifying and extracting data, coded into specific categories. Therefore, in this study we aimed to conduct a systematic mapping of studies conducted in schools during the Covid-19 pandemic, which use digital tools for collecting qualitative data. Associated with this central objective, we also sought to identify central themes, and the temporal and geographical trends, of the selected publications. These objectives are materialised in the following Research Questions (RQ), corresponding to the objectives of this study:

RQ1: What were the trends in the identified publications in terms of time and geography?

RQ2: What were the main themes of empirical studies carried out in schools using data collection methods mediated by digital tools?

RQ3: What digitally mediated data collection methods were used, who were the subjects/participants and what were the data analyses used in these studies?

RQ4: How is the change from face-to-face qualitative data collection to data collection mediated by digital tools discussed by the authors of the studies analysed?

To define the eligibility criteria, and to begin the identification and selection of articles, we worked independently (and afterwards as a group) in reading the titles and/or abstracts, and inductively considered for inclusion: (1) empirical studies; using (2) qualitative methods or (3) mixed methods; and (4) carried out in elementary and secondary schools.

The following first set of exclusion criteria were defined: #1: higher education studies; #2: preschool studies; #3: studies that focused on health, medicine and other fields/professions, or focused on a specific topic unrelated to education; #4: studies that used only quantitative methodologies, literature reviews or meta-analysis studies; #5: theoretical or methodological studies. The second set of exclusion criteria were defined: ##1: studies in which data were collected before the Covid-19 pandemic (before March 2020); ##2: studies in which the data collection period was not specified; ##3: studies with no clear indication of whether data were collected online or in person; ##4: studies with in-person data collection; ##5: articles that could not be accessed.

The identification of the studies was conducted using the Scopus and Web of Science (WoS) databases. Without disregarding possible limitations, with consequences such as the exclusion of other valuable articles, the choice of databases is justified because these are established and internationally indexed databases with peer-reviewed journal articles, and they are considered to be the most comprehensive data sources for various purposes (Pranckutė, 2021). In addition, these databases offer advanced search and filtering options, and they are considered suitable tools for synthesising evidence in the form of systematic reviews (Martín-Martín et al., 2021). These options make it possible to articulate and formulate a query in a systematic search context, paying particular attention to a high degree of coverage, recall, precision and reproducibility (Gusenbauer and Haddaway, 2020).

In March 2023, we defined which methodological approach to take, namely the definition of keywords and the search query in the title and/or abstract (Table 1). For keywords that would specifically lead to the methodological dimension intended to be explored, a larger number of tests were carried out. We did not want to risk excluding any type of data collection technique (for example, specific words such as ‘interviews’ and ‘focus groups’). Therefore, broader terms were chosen (qualitative methodologies OR qualitative research), and only studies that mediated their use with digital tools, regardless of the data collection techniques used, were included.

We looked for articles in English, Portuguese and Spanish, since we are proficient in these languages; however, the search was conducted using English keywords. In general, articles published in peer-reviewed journals have abstracts and keywords in English, regardless of the main language of the article, so this did not compromise the search, and guaranteed that we obtained to the same results.

The identification of the studies in the Scopus and WoS was conducted on 31 July 2023. The filters applied in the two databases were: (1) type of document: peer-reviewed scientific articles; (2) date of publication: 2020–3 (this time span is related to the period identified as the Covid-19 pandemic context; (3) language: English, Spanish and Portuguese (due to the substantial nature of Anglo-Saxon scientific production and the language skills of the researchers); (4) subject area: social sciences (in the Scopus database) and education (in the WoS database).

To mitigate the risk of bias in the realisation of this study, we divided the work among us during the different phases of designing the search strategy, selecting information sources, setting eligibility criteria, and in the data collection selection process. Subsequently, we analysed agreements and disagreements, and final decisions were reached through consensus. The intercoder reliability, calculated in terms of percentage agreement, exceeded 87 per cent. Two specific software tools, EndNote and NVivo, were critically employed. Notably, there were no technological errors recorded during this process. The use of these software programs was efficient, resulting in effective outcomes.

The selected articles were exported to the NVivo software. The features of NVivo allowed for the automatic classification of articles by year of publication. Subsequently, they were classified according to the country in which the study was conducted. This procedure made it possible to respond to RQ1. Thematic analysis (Vaismoradi et al., 2013) was used to identify the main themes addressed in the articles. This process followed several steps: (1) engagement with the data; (2) initial coding; (3) themes identification; (4) review of themes; (5) themes definition; and (6) final data analysis. Through thematic analysis of the articles, it was possible to address RQ2. The in-depth reading of the methodological dimensions of the studies led to their classification in terms of methodological approach, participants, data collection instruments, type of technological tool used and data analysis. The classification of methodological dimensions referred to the explicit indications of the authors of the articles regarding the defined categories. This final step addresses RQ3 and RQ4.

The initial screening yielded 1,169 articles identified in the Scopus database (n = 117) and the WoS database (n = 1,052). The selected articles were exported to EndNote, where duplicate articles (n = 11) were excluded, and 1,158 articles were submitted to a new selection stage.

The application of the first set of inclusion/exclusion criteria resulted in the excluded articles: #1 (n = 516); #2 (n = 17); #3 (n = 182); #4 (n = 8); #5 (n = 115). At this stage, 838 articles were excluded. The final screening was based on the remaining 320 articles, which were reanalysed. These 320 articles were distributed among the authors and subjected to the second set of inclusion/exclusion criteria. The following were excluded from the analysis: ##1 (n = 100); ##2 (n = 92); ##3 (n = 26); ##4 (n = 42); ##5 (n = 21). In this process, 39 articles were kept for final analysis, which are shown in Table 2. A more in-depth reading of the 39 articles made it possible to code them according to the year of publication, the country of origin of the study, the central theme and the methodology used. The process of identification, screening and inclusion of the articles in the review is illustrated in Figure 1.

The main findings of this research are organised to provide answers to the four research questions. Accordingly, for a better understanding of the results, the following four sections are related to each of the research questions.

The time span of the articles is 2020–3. The highest number of publications is concentrated in the years 2022 and 2023. This time distribution is related to the development of the Covid-19 pandemic itself and to the constraints to which publication processes are subject.

In terms of context, the 39 studies come from 20 countries: Argentina (n = 1), Australia (n = 1), Brazil (n = 2), Canada (n = 1), China (n = 1), Estonia (n = 2), Greece (n = 1), Indonesia (n = 1), Ireland (n = 1), Israel (n = 2), Japan (n = 1), Lithuania (n = 1), Mexico (n = 1), Philippines (n = 1), Portugal (n = 1), South Africa (n = 3), South Korea (n = 1), Spain (n = 2), Türkiye (n = 5) and the USA (n = 10). Figure 2 shows the relationship between the year of publication and the countries where the studies were undertaken.

The Covid-19 pandemic disrupted the usual practices of scientific research, especially data collection. Qualitative research, which relies on methods involving interactions with groups of people, has been strongly affected too (APA, 2020). Nonetheless, the attributes of qualitative research offer the potential to thoroughly explore the implications of unprecedented situations such as the pandemic, as well as other crises that disrupt and influence entire societies (Santana et al., 2021). The need to continue research amid the threat of Covid-19, as well as the potential of qualitative research in this context, has required qualitative researchers to adapt traditional approaches and explore alternative data collection methods. However, the urgent nature of the situation, that is, the need to find quick solutions, poses the risk of compromising data reliability, academic integrity and credibility (Otto and Haase, 2022; Torrentira, 2020). In this sense, further reflection on adaptations and/or the development of new data collection methods within qualitative research resulting from the pandemic period could be extremely beneficial for consolidating and/or revising the procedures used and their application in similar crisis situations (Wa-Mbaleka and Costa, 2020). With the aim of identifying and characterising studies conducted during the pandemic within a school setting that used qualitative data collection methods mediated by digital tools, this mapping took into account four research questions that will guide subsequent discussion and analysis.

The findings of this study have to be seen in light of some potential limitations. For instance, the data collection process considered only two databases – Scopus and WoS – which may have led to relatively low coverage of studies in social sciences (Heck et al., 2023). This issue can be addressed in future studies by developing the search strategy in collaboration with an information scientist.