Face-to-face interviews can feel intrusive and intimidating, whether the participant has to enter the researcher’s space, or vice versa (Braun et al, 2017). Participants may not want the researcher to see their living circumstances (source: interview with researcher; consensus conference in relation to underserved populations), or feel that having a researcher come to their home means they have to tidy up and ‘host’ them (e.g. provide food and drink), which can be burdensome, particularly for those with poor health or disabilities.

Remote data collection can balance this out somewhat, as the participant is in a private place of their own choosing, inaccessible to the researcher, with autonomy to reveal or conceal as much of that space as they want to (Hanna and Mwale, 2017). As such, remote interviews have the potential to tip the balance of power in favour of the participant by affording them more decision-making opportunities (Piacentini et al., 2022). Participants are free to ‘take control of the conditions’ of the data collection in various ways, for example, in the case of videoconferencing, by deciding where to locate a camera, how to appear within the frame, whether/ when to turn it on, as well as whether or not to use blurred or virtual backgrounds (Sipes et al., 2022; Prior and Lachover, 2023, Ślęzak, 2023; Carter et al., 2021c). As Engward et al. (2022) note, ‘What is viewed and accessed by the researcher on screen is determined by the participant, while the wider inhabited domain remains concealed.’ (Engward et al, 2021: 2)

With cameras off, the perceived burden of ‘selfpresentation’ may be entirely removed (Szulc, 2022). Whilst potentially disconcerting for the researcher, (Valdez and Gubrium, 2020) this absent, incomplete and/or heavily curated view of the participant can be empowering for them as they have the control to manage the way they appear, which can be particularly empowering for those with stigmatised or marginalised identities (Ślęzak, 2023). Disabled people, for example, may be able to conceal their disability in remote interviews and therefore limit how much information the interviewer has about it. Similarly, these options can also be utilised by disabled researchers (Brown and Boardman, 2011), or researchers who wish to conceal other aspects of their identity, such as age or ethnicity, as one researcher commented regarding interviews with young people:

had not given participants any information about myself beyond the formal information required for the research information sheets, and for the most part, the absence of verbal and visual cues seemed to minimise the age difference between me and my participants... [...]... One participant finally asked me if I minded if he asked me what ‘age range’ I was in. I found myself hesitating before I texted back that I was in my fifties, suddenly conscious that he might believe me to be younger than this. The awkwardness of this moment underlined the potential of this interview method to obscure differences between the participant and researcher in ways that can be considered helpful as well as deceptive(Gibson, 2020: 622-623)

Whilst some researchers argue that finding shared characteristics between participants can help build rapport, especially in remote contexts where this is harder won (O’Conner and Madge, 2017; Brown & Boardman, 2011), being an ‘outsider’ in remote data collection has also been found to conversely invite more detailed responses and reduce bias (Asselin, 2003), as more explanation is deemed necessary by the participant (Dahlin, 2021) and the researcher is more likely to ask clarifying questions (Dwyer & Buckle, 2009). However, as noted by one of our consensus conference participants, this detailed explanation can be experienced as burdensome, and this burden disproportionately falls on people from marginalised communities or backgrounds:

…I personally would probably assume that the researcher’s culture was not the same as mine (statistics show there are less African Caribbean heritage researchers accessing research funding), and…this may alter communication accordingly. There may be cultural nuances that I would feel wouldn’t be understood, and [I] would have to go through a process of explaining and often this feels like an extra burden.. […]…I may just decide not to share..(Consensus conference participant)

This demonstrates the salience of bodily signifiers in qualitative research, and a decision to remove them through use of an audio/text only technology on either side of the research exchange- researcher or participant- needs to be considered in terms of its impacts on inclusivity and representation.

Ensuring the privacy, and (where appropriate) anonymity of research participants, is an important aspect of qualitative research. Indeed, it has been suggested that capacity for privacy is more important for participation and disclosure in qualitative research than whether the method is remote or not. (source: consensus conference) MORE INFORMATION Due to the richness and depth of data usually aspired to within qualitative research, both privacy and anonymity can be challenging. Remote methods, however, offer opportunities to enhance privacy and anonymity not possible in face-to-face research, as well as introducing additional risks.

As the use of remote methods often involve the researcher handing some degree of control over the management of the research space to participants, factors that the researcher would ordinarily attempt to minimise in face-to-face research, such as background noise and distractions (which can affect audio quality) as well as disruptions caused by the data collecting technology itself e.g. poor internet connectivity, become beyond the reach of the researcher (Engward et al., 2022, Oliffe et al., 2021, Deakin and Wakefield, 2014, Goh and Binte Rafie, 2023). Furthermore, the presence of third parties in the research space (known or unknown) - sometimes referred to as ‘Zoom bombers’ (Boland et al., 2022a) - can sometimes ‘hijack’ or derail data collection (Prior and Lachover, 2023; Chiumento et al., 2018). These interruptions can lead to researchers misinterpreting visual cues, such as smiles or turning of heads, which could be non-verbal cues relating to the conversation, or a response to the presence of others in the room (Seitz, 2016). Anticipating, and planning for, distractions when creating the topic guide and time allowed has been suggested by Gray et al (2020) as a means of managing these scenarios. When Zoom bombers are children, however, there can be both privacy and distress concerns When Zoom bombers are children, there can be both privacy and distress concerns, particularly if the content of the data collection is inappropriate for a child to hear (Newman et al., 2021). Offering funding for childcare during the research encounter can somewhat reduce these situations, however it is not always a feasible option (e.g. child with separation anxiety, data collection occurring at an unsociable time). Indeed, researchers may feel disempowered to police the boundaries of the research space and request that these ‘interlopers’ leave (even when/if they notice the data quality is being altered by their presence). This is particularly difficult when the ‘research space’ is the interloper’s own home. Those in shared accommodation in particular may struggle to find a suitably private space to participate in research (Singh et al., 2021, Hernández, 2014, Marhefka et al., 2020). Whilst remote research conducted with children has been found to work best when parents are not in the same room as the child (source: interview with researcher), this might not be possible in all scenarios, and researchers may not even be aware that additional people are present, due to their heavily restricted view of the participant. For sensitive research topics, this lack of control of who overhears the participant, or who sees their screen (Gibson, 2020) can be particularly challenging (source: researcher interview).

For participants in conflict-affected locations where there is a high degree of political surveillance (Douedari et al, 2021), in precarious home situations, or those where coercive control, surveillance or domestic violence are a concern (Little 2017), the methods for remote data collection and maintenance of privacy and confidentiality have to be carefully considered. The remote nature of research may actually heighten participant concerns about surveillance due to the potential for infiltration or interception of data through hacking (Oltmann, 2016). Indeed, email is often considered to be particularly insecure (Fritz and Vandermause, 2018). Methods that require use of a stationed technology might also mean participants having to participate within an unsafe environment, bringing risks to personal safety, reducing the chances that they will agree to take part at all, or restricting how much information they can safely share (Newman et al., 2021; Little, 2017). Such participants may prefer telephone interviews to online interviews due to the difficulty in finding a private space in the home (Pedersen et al., 2023) and in these scenarios researchers should not mail consent forms and information sheets to the home, but instead provide them digitally. Participants may also need specific instructions about how to stop/pause the interview if they are interrupted

There are reported instances, however, of when the presence of ‘interlopers’ can enhance data interpretation; who the interlopers are, and how they affect the data may actually provide insight into the subject being researched (Prior and Lachover, 2023).

Researchers should consider the privacy and anonymity of participants at all stages of the research design. Whilst the privacy of the research space may not be within the researcher’s power to maintain, there are factors that can be considered to allow participants the opportunity to maintain privacy and anonymity:

1. Supporting participants to access technologies for data collection that afford them as much privacy and anonymity as required. Video conferencing platforms may be considered more private than telephone because although we are used to giving out our email addresses, a telephone interview usually means the participant giving out their personal phone number (Sipes et al., 2019). However, participants sometimes prefer to use technologies that may have lower privacy standards than those approved by ethics committees or institutions. Researchers should, where possible, avoid paternalism by insisting on use of particular technologies, although it may be appropriate to alert participants to any threats to anonymity and privacy associated with their chosen technology, as well as consider the requirements of their funder, institution and ethics committee.
2. Supporting participants to choose a time or date for their data collection when they will have greatest privacy in the location the data collection will occur and providing them with the means to end data collection abruptly if needed (Hanna, 2012), this might include having a ‘code word’ to signal that privacy has been breached (Peterman et al, 2020).This also needs to be considered for asynchronous methods such as email and instant messaging, as the participant’s screen may be visible to others (Gibson, 2020).
3. Whilst the flexibility and choice made possible by remote research may offer new possibilities for inclusivity and access, it is important to consider how privacy concerns can be increased for participants for whom engagement in research is always mediated by others, e.g. carers or interpreters (particularly if this person is a family member, friend or community member known to the participant outside of the research encounter). It also affects those who do not have independent access to technology, or do not own their own devices. Indeed, privacy will likely look very different for these groups as Mikulak et al (2022) note:
   

   While the digital gap might be narrowing for people with learning disabilities, for many access to technology remains mediated through family and support workers. This was true for many of our participants who were reliant on support staff to facilitate access to technology and/or a particular platform. Some participants did not have their own devices and needed a staff member with access to a laptop to be on shift. Others had their own device but were not comfortable using it independently, especially when using an unfamiliar platform such as Zoom.(Mikulak et al., 2022: 275)

   Where a participant’s inclusion in the research hinges on the involvement of another person, researchers should be alert to any signs of coercive control (e.g. speaking on behalf of participant, controlling the topics discussed) as there might be implications for safeguarding (source: consensus conference). Researchers should also be mindful of different cultural interpretations of ‘privacy’, as for some social groups, having a family member present would still be regarded as a private space (source: consensus conference).

   Communicating these known limitations to privacy in a way that is accessible for the social groups to which they apply, as well as working with participants and intermediaries to codevelop methods to enhance privacy as far as possible is key to inclusive research practice.

4. Anticipating, and informing, participants of any known risks to their privacy, anonymity and data security, particularly those they may not have considered, as well as being alert to any emerging risks, is of critical importance. For example, in the context of group data collection, informing participants of any contact details, profile pictures, bios or linked accounts that will become visible to others in a group setting (Douglas et al, 2021). Newman et al (2021) suggest that participants in remote focus groups should be encouraged to adopt nicknames or pseudonyms and, for videoconferencing contexts, to use virtual backgrounds to heighten their privacy. Similarly, if the research is to be conducted by people who are known to the participant or have a duty of care towards them (e.g. a health care professional, social worker, teacher) participants need to be made aware of this before recruitment into the study, and any implications for them explicitly highlighted.
5. Those unfamiliar with the technology are at greatest risk of inadvertently sharing more information than they intended, including identifiable data or sensitive viewpoints/ experiences. Scenarios can include: participants not realising, or forgetting, when a microphone or camera is on/off, assuming another participant in a focus group cannot see them if that participant’s camera is switched off but theirs is on, posting a comment to the whole focus group when the message was intended for the researcher alone). Researchers should be wary of ‘offloading’ the work and ethical imperative of data security onto their participants. Indeed, access to the digital world should not be conflated with digital skills to safely and securely participate in remote research (Engward et al., 2022). Young people, for example, have been found to be more ‘savvy’ about protecting their online privacy than older generations (Bolin et al., 2023). Risks can be managed by determining what personal information could be visible to others through data collection, and supporting participants to remove, reduce or disguise their personal information where appropriate, for example, by using an alias, avatars (Gadalla et al, 2015), masks or blurred photos (rather than live images), or an account set up specifically for the research (Cook, 2012). This can make the research more appealing and safer for participants who belong to ‘hidden’ populations (e.g. those engaged in illegal activity), but can also reduce trust on both sides of the research exchange (Teubner and Flath, 2019; source: consensus conference; interviews). To avoid paternalism, decisions to use these strategies should not be imposed onto participants, but instead should be facilitated by discussion with the researcher. Researchers may face scenarios where data management plans and ethical approvals are in conflict with the wishes of participants (e.g. a participant wanting to use a personal account for data collection rather than setting up an anonymous one), and researchers should anticipate where these tensions are likely to occur and how data management strategies can flex around participant preferences, and situations where they cannot.
6. Videoconferencing platforms, when used for voice only, or text-based methods using nonpersonal accounts, can afford participants the greatest levels of anonymity and privacy in remote qualitative research because they can be undertaken without links to email accounts or profiles, and don’t involve giving a personal phone number out to a researcher or other participants (Sipes et al., 2022; Ślęzak, 2023). Furthermore, certain videoconferencing platforms, as well as instant messaging and email platforms allow the participant to block the researcher following data collection to ensure that there is no further contact (Sipes et al, 2019). As noted by Carter et al. (2021c), there are particular topics and populations where anonymity and privacy are essential for participation, for example marginalised, stigmatised or underserved groups (Piacentini et al., 2022; Lathen and Laestadius, 2021), or where the researcher has a pre-existing relationship or duty of care in relation to their own participants, for example, health care professionals researching their own clinical practice (source: consensus conference).
7. Where levels of privacy and anonymity are greater, participants may feel more able to report more sensitive or stigmatised experiences and discuss less socially acceptable feelings, as noted by Jenner & Myers (2019):
   

   Exceptional disclosures’…only ever occurred in private, most commonly via Skype, and covered such sensitive topics as unprotected sex, miscarriages, abortions, postpartum depression, and arrests.

   (Jenner & Myers, 2019: 173)
   

   Researchers should consider the tensions between a potential increase in ‘exceptional disclosures’ when greater anonymity is possible for participants, as well as the ways that heightened anonymity can preclude the effective implementation of safeguarding measures. A lack of safeguarding procedures can have serious implications for participant and researcher not only in terms of safety, but also psychological distress.

8. It is important that privacy and anonymity are sufficiently considered through to the post data collection stage of the research. The researcher will be powerless to prevent participants using details gathered through the data collection (e.g. phone numbers) for example in the case of WhatsApp group chats to send unsolicited messages to other participants unless this is managed before data collection (Neo et al., 2022). There is also an additional risk of unauthorised recordings being made that are beyond the control of the researcher (source: consensus conference). These can be more easily generated in remote contexts: for example, in instances where the researcher does not have full sight of the participants (so would not see recordings being made) (Boland et al., 2022a), where it is possible to take screenshots of information or images (Neo et al., 2022), and if recording links are accessible on videoconferencing platforms after the data collection event has ended. Whilst some risks to privacy and data security are more easily managed than others, the importance of privacy and respect for data security should be emphasised to participants in group data collection scenarios. However, the implications for trust (both between participants themselves as well as between participants and the researcher) and disclosure need to be weighed against data security and privacy concerns (Teubner and Flath, 2019).
9. Researchers should consider providing participants with a password-protected electronic invitation to the research space. This both prevents digital interlopers and reinforces the notion that the data collection is private (Oliffe et al., 2021). Re-iterating the need for privacy at the start of data collection (particularly group data collection) can also help to set expectations for the handling of participants’ data. Researchers using videoconferencing may also consider ‘locking’ the session, and/or using a virtual ‘waiting room’ or ‘lobby’ where identity can be established before entry is permitted to avoid meeting spaces being ‘hacked’ (Hern, 2020).
10. Researchers need to be aware of the changing privacy features of the different technologies and platforms they intend to use, and consider these in relation to participants’ geographical location, social and political context, age and digital literacy. Privacy features, and the ways these are regulated, are constantly subject to change. As noted by Salmons, ‘Researchers must be fully aware of the features – and hazards – of selected interview technologies’ (Salmons, 2022). Being aware of the ‘digital traces’ that might be left on the technology platform being used is critical to maintaining the privacy of participants (Marlowe and Allen, 2022). Where platforms offer a choice of local or cloud storage, local storage will be more secure, or data can be stored on a secure device and deleted from video-conferencing applications (Dube et al., 2023).
11. Researchers should consider the level of data they need to collect in order to answer their research questions, and the risks it poses to participant privacy. For example, conducting data collection via videoconferencing with cameras on involves an additional dimension of highly identifiable data (visual) that would not typically be memorialised (through recording) in the case of face-to-face interviews. This can pose additional risks to participant privacy. Kahn and MacEachen (2022) used a dictaphone to capture data generated through videoconferencing, rather than the in-built record function, in order to protect participants, particularly as the topic of their research was sensitive. Through doing so, the interpersonal benefits of visual data could be retained, without some of the increased risks associated with reduced anonymity. Data protection legislation requires justification for any identifiable data collected, so the decision to include this type of data needs to be weighed against particip
12. Special care needs to be taken with data security and storage. Researchers should be careful to delete text chats from devices (Chen and Neo, 2022). Privacy extends to data gathered through online means (social media), and how the data is stored and later used or misused (mismanaged, stolen, hacked, sold etc). For instance, when remote interviews are stored in ‘the cloud’, researchers need to consider where in the world the server is, and what kind of data protection laws are in force (source: consensus conference). Confidentiality agreements with transcribers, and secure means to transfer audio data to them are also critical.
13. Researchers should also consider their own privacy when conducting remote research, by not using personal accounts or technologies (e.g. phones) for data collection (Humphries et al., 2022), and being conscious of how much personal information is accessible to participants (e.g. profile pictures, backgrounds on videoconferencing platforms, desktop wallpaper when sharing screens). Professional research boundaries can become particularly difficult to maintain as researchers typically have public profiles and/or social media presence (source: consensus conference). At the extreme, this may present a safety risk for the researcher, e.g. through persistent messages and threat (Mattheis and Kingdon, 2021). Building in ways to protect researcher privacy, for example, by use of dedicated research devices and profiles, separate to those used by researchers in their personal lives, and use of ‘backgrounds’ (e.g. images of interiors) on videoconferencing platforms. Funders and institutions can support researchers to protect their privacy through the provision of devices (or the funding to acquire these) and secure environments (e.g. single occupancy offices) in which data collection can occur.
14. It is important to consider the online/ social media profile or footprint of both researchers and research participants. This can shape recruitment, rapport, expectations, boundaries and safety before, within, and beyond the data collection event (source: consensus conference) as well as potentially affecting the privacy of participants (Marlowe and Allen, 2018).

Physical and psychological safety for both research participants and researchers is of paramount importance (Boynton, 2017). Aspects of safety and safeguarding relevant to face-to-face research may be amplified, reduced or eliminated through use of remote methods (Santhosh et al., 2021); remote methods can also introduce unique safety and safeguarding factors. Furthermore, certain social groups are more vulnerable to safety risks in remote spaces than others (Stone et al., 2020).

Location

When locations are suggested for face-to-face data collection, there is a risk that researchers may suggest a place where one or more participants have experienced trauma. Participants may also suggest locations that involve risk to themselves and/or the researcher and would potentially not be supported as a suitable site of data collection by an ethics committee and/or a risk assessment. Indeed, face-to-face data collection can generate risks for people who are endangered by having researchers in their home environment. This might include those engaged in sex work or living in an abusive relationship. The use of remote methods can be seen as circumventing this difficulty with locations by allowing participants the freedom to select the setting for their participation. (Source: researcher interview)

While choice around the conditions of the data collection may be empowering to participants, they may nevertheless select locations that have the potential to cause harm. Indeed, there are examples in the literature of participants taking part in emotionally challenging remote data collection (e.g. relationship breakdown and suicide) whilst in charge of vehicles (Oliffe et al., 2021; Epp et al., 2022). While researchers can prepare participants for data collection by exploring appropriate locations they may have access to, and times and dates when distractions can be minimised, the researcher may not ultimately know the participant’s location and context until the data collection encounter. Whilst rearranging data collection can be suggested, this may not be the participant’s preference, meaning that the responsibility of ultimately deciding whether to proceed will fall to the researcher themselves.

It has also been suggested that a move to remote methods reduces researcher access to contextual data. Whilst the sight, sounds and atmosphere of the data collection setting may assist researchers in interpreting their participant’s world, particularly in the context of geospatial methods, there is some evidence that this can be replicated in remote contexts, e.g. participants choosing to show objects in their environment, or giving the researcher a ‘tour’ of their homes. Even ‘go-along’ interviews may have remote counterparts, as researchers have reported using virtual reality or Google Maps to remotely explore their participants’ vicinities (Kostakos et al, 2019), or exploring regional literature and media sources to gain access to a participant’s world (Keen et al., 2022).

While this situated data may add richness and may aid with the development of rapport, these techniques also raise ethical issues when researchers are virtually ‘taken to’ places where they would not go in person (source: interview with researcher). One example is in research with children, where audio-visual interviews via smart phones allow children to carry the phone around the house into usually private spaces. Researchers may feel less able to adequately supervise child participants when they are not co-located (source: interview with researcher). Equally, intervening to protect childrens’ space to participate in the research is harder when that researcher and participant are not co-located (source: interview with researcher). For example, adults co-located with children sometimes intrude on the data collection. However, there are instances where this dynamic can be productive (Mannay et al., 2023).

Indeed, in the context of remote interview methods, participants have to take greater responsibility for their own environment and its implications for wellbeing. For people experiencing intimate partner violence, ensuring the data collection content will not endanger them (source: consensus conference), safety checking at the start of data collection (asking who else is in the vicinity) and having a ‘safe phrase’ to exit the data collection quickly e.g. ‘I think you have the wrong number’ can be employed and signal danger to the researcher (Alderson et al., 2022). Indeed, people who are experiencing violence, particularly in closely knit environments, are more prone to surveillance, and they might be monitored when interacting with researchers (Little, 2017). Gatekeepers (e.g. advocacy groups) may be an important resource in managing safety (Alderson et al., 2022), and risk assessments may need to be carried out (source: consensus conference).

For researchers, the use of remote methods can move research into their own homes or other private spaces, potentially blurring the lines between public, work and private spaces (Lobe et al., 2020, Jenner and Myers, 2019). For example, during an interview with a researcher, they commented that it could be difficult to move directly from hearing difficult stories to ‘making tea for the family’ (source: interview with researcher).

While remote methods remove the risks of lone working when conducting face-to-face data collection and provides flexible opportunities to access participants who may otherwise be unreachable, the permeation of the research into the researcher’s physical space (especially if participants can gain a lot of information about the researcher from seeing their home environment) can be associated with psychological or even physical risks. These risks can occur before, during or after data collection events (e.g. receiving unwanted messages on work devices kept at home). Researchers should consider when/where they conduct their remote data collection, how much information about themselves is contained in their background, whilst also ensuring they have access to appropriate support available for both them and their participants, whilst retaining professional boundaries. Universities, research organisations, funders and ethics committees all have a role to play in supporting researchers to find safe physical and remote environments for qualitative data collection through risk assessments and safety protocols (Mattheis and Kingdon, 2021).

Psychological Distress

For any research topic, participant distress is a possibility (Braun and Clarke, 2013: 89). Indeed, participants are often motivated to take part in research despite knowing it might be upsetting. This can be attributed to an altruistic desire to improve the lives of others (Heath et al., 2018), or the data collection can provide an opportunity for participants to vent or process uncomfortable feelings (source: consensus conference; Humphries et al, 2022). However, it may be harder for researchers to identify signs of participant distress during remote data collection compared to face-to-face (Thunberg and Arnell, 2022; Epp et al., 2022; Humphries et al., 2022), which can make managing this hard. While rapport and human connection are integral to qualitative research, it is important to recognise the emotional labour of conducting qualitative data collection, and the psychological risks, including the possibility of ‘vicarious trauma’ for researchers (Isobel, 2021). Being able to offer flexible opportunities to take part in research is a significant benefit of remote data collection but may also entail risks for researchers.

As noted by Humphries et al. (2022), asynchronous remote data collection, in particular, can involve long periods of silence between contacts, which can leave the researcher feeling anxious about how the participant is coping. Existing guidance recommends that data collection is not conducted when a researcher is already tired from a long day or when support is not available (e.g. buddy or debriefing) (Silverio et al., 2022). In practice, a positive research culture and trust are essential for ethical debriefing to occur (whilst protecting participant confidentiality). Indeed, it involves varying degrees of personal and professional vulnerability, which may have different levels of impact on researchers depending on their career stage and how established their professional relationships are (in the context of fixed-term contracts for early career researchers, these relationships may not be as well developed).

The seeming time efficiency of remote methods, and the pressures of research delivery, can also lead to researchers scheduling multiple data collection events in a single day- a practice that may preclude appropriate self-care, as well as the cross-pollination of research findings and question refinement that often develops following analysis of early data. For asynchronous text-based methods, data collection can occur over much longer periods of time than synchronous methods, during which psychological distress, on the part of both the researcher and participant, may be more insidious, harder to detect and can also be cumulative in nature (Fritz and Vandermause, 2018). It is therefore important that researchers (as well as institutions, ethics committees and funding bodies) consider the psychological impacts of remote qualitative fieldwork, and the ways that these can extend beyond those associated with face-to-face research. Pascoe Leahy (2021) introduces the notion of an ‘afterlife’ of qualitative research that can persist long after the data collection event. For remote research, which can be conducted over extended time periods, the impacts may be particularly pronounced (Humphries et al., 2022). The use of distress protocols for both researchers and participants, which may include providing access to external counselling services for researchers (Bhatia et al., 2022) and/or de-briefing (Whitney & Evered, 2022).

Methods to ‘contain’ the research both during data collection itself (e.g. ensuring that messages or emails do not ‘pop up’ on a screen, but instead go directly into a dedicated folder that the researcher can access when they are in a position to), and afterwards (e.g. ‘debriefing’ with another member of the research team) can help to manage some of these difficulties, and offset the implications of working in an ‘unbounded’ research space.

Safeguarding

Research with children and other vulnerable groups raises particular issues around safeguarding (Bhatia et al., 2022). When working remotely, it may be harder to intervene to protect participants from themselves or others, and researchers may not have an address or contact details for participants, e.g. in text only data collection methods where only an online alias is used (Anderson et al., 2021). Researchers may have a feeling of powerlessness to intervene (Gibson, 2020; source: consensus conference), and there needs to be an appropriate balance between protecting participants and respecting their agency (Nkosi et al., 2022).

Safeguarding protocols in instances where participants disclose risk of serious harm to self or others may need to be adapted for remote context. For example, a process of consultation and escalation within the research team for considering whether the risk is sufficiently high that researchers require information about participants location and a plan of action. In relation to children, this threshold needs to be set low. Safeguarding protocols should ideally be co-produced with relevant stakeholders (source: consensus conference) and in partnership with organisations and institutions where participants can be referred (Bhatia et al, 2022; Hawk et al., 2021). This is especially the case when the research directly explores topics that may lead to disclosures, such as those on violence, abuse and self-harm (Bhatia et al., 2022).

Remote data collection can allow participants more control over what data they provide and when, including information about themselves that are more apparent when face-to-face (e.g. age group, mobility disability). This can be empowering for participants. Remote data collection involves handing responsibility for privacy and safety during data collection to the participant. This may be problematic for the participant particularly where having individual space is not the norm, where there is political surveillance, in precarious home situations and where the participant requires the assistance of a second person. In advance of data collection, researchers need to plan what they will do if they are concerned about confidentiality or participant safety. Researchers need to ensure participants understand how to avoid inadvertently sharing more information than intended through digital modalities (e.g. phone number revealed to other participants). Digital modalities can be set up to provide the greatest level of anonymity and privacy available for qualitative data collection. Researchers need to plan what to do if they are concerned for their participants safety and well-being, and for their own.

1. What are the power differentials likely to be at play during data collection; how might remote collection change these?
2. What will you and your participants not know about each other when using digital modalities (unless specifically revealed); how does this impact the data and its analysis?
3. How well do you understand the potential challenges for participants in ensuring their own privacy and safety during data collection?
4. Are you up to date with the privacy features of all the digital modalities you are considering; how will you advise your participants about privacy?
5. What will you do if you are concerned for the privacy, safety or distress of your participant?
6. What are your plans for your own safety and wellbeing?