Week 5

Contents lists available at ScienceDirect

Applied Nursing Research

journal homepage: www.elsevier.com/locate/apnr

Original article

“It's just more personal”: Using multiple methods of qualitative data collection to facilitate participation in research focusing on sensitive subjects☆

Jennifer Heatha,⁎, Heidi Williamsona, Lisa Williamsb, Diana Harcourta

a Centre for Appearance Research, Faculty of Health and Applied Sciences, University of the West of England, Bristol, BS16 1QY, UK b Chelsea and Westminster Burns Service, Chelsea and Westminster Hospital, London SW10 9NH, UK

A R T I C L E I N F O

Keywords: Interview Data collection Participant choice Response preference Engagement Sensitive research

A B S T R A C T

Background: Often researchers use only one interview method within a single study. However, it is increasingly common for a variety of interview methods to be employed within a project. Providing choice to participants may facilitate recruitment, particularly when research focuses on a sensitive subject. Aim: This paper aims to explore participants’ reasons for their choice of interview method in qualitative research focusing on a sensitive subject. Methods: Qualitative data was collected from 12 participants regarding their choice of interview method (face- to-face, Skype, telephone or email) in a wider study investigating parents’ experiences following their child’s burn-injury. Interview data was transcribed verbatim and thematic analysis was used to analyse responses. Results: Participants most frequently chose to take part in an interview by telephone (n = 5), followed by email (n = 4), then Skype (n = 2), and face-to-face (n = 1). Four themes emerged suggesting that participants’ deci- sions to participate via a particular method were determined by personal convenience, their belief in their ability to be open with the researcher despite potential upset caused by the topic, their ability to get a “feel” for the researcher, and concern about giving adequate depth in responses. Conclusions: Flexibility regarding the ways in which participants can take part in qualitative research may improve participant access to research, recruitment, and response-rate. Although, the depth and quantity of data gathered using different methods can vary. It is important that researchers consider the procedures and ap- propriateness of using different methods of interviewing and what impact such methods might have upon themselves and their participants.

1. Introduction

When research focuses on a sensitive topic – topics which poten- tially pose a threat to those who are, or have been, involved in them (Lee, 1993) - the method through which information is collected can be particularly important to participants. Personal data is most likely to be disclosed when assurances of privacy, confidentiality and a non-con- demnatory attitude are provided (Wellings, Branigan, & Mitchell, 2000). Issues of privacy and personal choice regarding how to disclose information may be particularly relevant when the focus of research is considered private, stressful, or sacred, where disclosure might cause stigmatisation or fear, or where there is the presence of a political threat (Lee, 1993). Such issues might be particularly relevant to nurse researchers and other healthcare professionals who play vital roles in

recruiting people into studies (Bartlett, Milne, & Croucher, 2018). A key underpinning of qualitative research into sensitive topics is

the establishment of rapport (Dickson-Swift, James, Kippen, & Liamputtong, 2009), requiring researchers to take steps to make par- ticipants feel relaxed and comfortable enough to share their experiences (Liamputtong, 2007). Clark (2008) suggested that a lack of under- standing about the research methods used might promote indifference or resistance in participants. Therefore, transparency throughout the research process is important, ensuring that participants can appreciate the true purpose of the research and understand why the research is being conducted in the way that it is (McQuaid, Barton, & Campbell, 2003).

Qualitative research typically utilises one interview method in iso- lation. However, there are growing numbers of options available for

https://doi.org/10.1016/j.apnr.2018.06.015 Received 4 April 2018; Received in revised form 18 June 2018; Accepted 23 June 2018

☆ Funding source: This research was part of a program of PhD studies funded by the University of the West of England, Bristol, UK. ⁎ Corresponding author at: Centre for Appearance Research, Faculty of Health and Applied Sciences, University of the West of England, Frenchay Campus, Bristol BS16 1QY, UK. E-mail address: jennifer3.heath@uwe.ac.uk (J. Heath).

Applied Nursing Research 43 (2018) 30–35

0897-1897/ © 2018 Elsevier Inc. All rights reserved.

T

researchers wishing to conduct interviews. Face-to-face interviews are seen as the gold standard (Deakin & Wakefield, 2014). They have several key strengths; flexibility, spontaneous personal and observable interaction, and more control over the interview environment than would be possible during remote methods of interviewing. There are also disadvantages with face-to-face interviews, such as the high cost per participant, geographical and time constraints associated with tra- velling.

Online interviews, such as those conducted over Skype, are often presented as a second choice or alternative when face-to-face inter- viewing is not possible (Deakin & Wakefield, 2014). They allow face-to- face communication with the opportunity to appreciate some body language and other non-verbal communication, which telephone in- terviews cannot. However, they also rely on the participants having good internet access and some participants might not be comfortable “on-camera”, not presenting as they would in a person-to-person si- tuation (Bertrand & Bourdeau, 2010).

Another alternative, email interviews, are being increasingly used by nurse researchers (Hershberger & Kavanaugh, 2017). Email inter- views have a number of strengths. They offer low cost interaction without the need to travel, do not require researcher or participant presence at the same pre-specified time, and are potentially more ac- ceptable to those who might decline or be unable to participate in spoken interviews but willing to answer questions posted on their computer screens (Duffy, Smith, Terhanian, & Bremer, 2005). Online data collection has been found to improve access to hard-to-reach groups and participation from ethnic minorities (Joseph et al., 2013), particularly those who are geographically diverse (Wilkerson, Iantaffi, Grey, Bockting, & Rosser, 2014). Duffy et al. (2005) also found that online respondents might be less susceptible to social desirability bias because of the lack of researcher presence.

Knapp and Kirk (2003) argue that different modes of data collection will result in equivalent results among technologically savvy groups, although this assumes no connectivity problems or transmission delays that can hinder online data collection, whether it be via Skype or email (Fox, Morris, & Rumsey, 2007). Couper (2011) also noted that relying on online data collection methods risks selective sampling, leading to nonresponse bias from those without access to the internet, for example those of low socioeconomic status or older participants.

The telephone interview is another option, but this is often viewed as a less attractive alternative to the face-to-face interview (Novick, 2008). Notable strengths include good geographical coverage, personal interaction, and low cost compared to face-to-face interviewing. How- ever, disadvantages include a lower response rate compared to face-to- face communication, and the inability to observe the participant (Groves, 1990). Groves (1979) found that respondents expressed more discomfort about discussing sensitive topics over the telephone than face-to-face, with most reporting that they would have preferred to be interviewed face-to-face rather than by telephone. More recent studies have disagreed with Groves' claim, finding that interviewees feel comfortable with telephone interviews when discussing intimate, sen- sitive, and personal issues in an open and honest manner, and are less concerned about humiliation than when speaking face-to-face, perhaps due to increased familiarity with telephones (Chapple, 1999).

With multiple options for interviewing available to researchers, when planning a study, it is important that public involvement (PI) is utilised. PI can help to inform researchers how best to engage potential participants, advising on factors such as the appropriateness of research questions, study information, and the data collection format (Brett et al., 2014). Clark (2008) highlighted the importance of this, de- scribing how research engagement is not necessarily interesting for everyone, and that different methodological techniques will appeal differently to those who do engage. Therefore, researchers themselves have an important role in ensuring that participants have an optimal research experience (Bartlett et al., 2018).

As different techniques of data collection with appeal to different

people, it is increasingly likely that more than one type of interview will be employed in a single study; such as Skype and face-to-face (Deakin & Wakefield, 2014), or a combination of face-to-face, telephone, and email interviews (Dures, Morris, Gleeson, & Rumsey, 2011). This flex- ibility may improve participant access to research and is discussed in this paper in relation to research on a sensitive topic: parents' experi- ences of having a child suffer a burn-injury.

Burns research in general can be sensitive, focusing on an event that is often very stressful. When parents are the focus of research following their child's injury, participation might be perceived as threatening due to feelings of guilt, perceptions of failure, and the upset caused by re- calling events (Heath, Williamson, Williams, & Harcourt, 2018). There may also be fear of judgment or stigmatisation from the researcher, and/or concerns about the anonymity of data (Braun & Clarke, 2013; Pyer & Campbell, 2012). This issue is particularly pertinent in this population; McQuaid et al. (2003) found that parents of burn-injured children can be wary of research of their or their child's experiences, sometimes suspecting an alliance between researchers and social ser- vices. As such, this is an under-researched area despite the fact that 58,000 children attend Accident and Emergency departments due to burn-injuries every year in the UK (National Burn Care Review, 2001).

Centralisation of healthcare means that patients can live long dis- tances from hospitals and some may find it difficult to travel (Jo, 2007). For this reason, multiple methods of interviewing are increasingly re- quired to access the necessary participant group, as important data may be gleaned from those based in a range of geographical locations, with diverse socioeconomic statuses, ages, and/or backgrounds (Deakin & Wakefield, 2014). Recruitment can also be difficult for other reasons. For example, burn injuries can have a significant impact on appearance (Lawrence, Mason, Schomer, & Klein, 2012) and the experience of living with an unusual appearance is another sensitive area of research. When studies focus on appearance or disability, some options for par- ticipation may be more practical than others. The option to participate without having to physically meet strangers or attend an unfamiliar location may be particularly appealing to those with a visible difference (Fox et al., 2007). Therefore, researchers need to be mindful not to exclude certain groups, for example, by only including those with in- ternet access or within reasonable travelling distance.

PI was utilised in the study described within this paper, aiming to ensure the appropriateness of all aspects of the research, including the interview schedule and data collection methods, to allow parents to discuss fully their experiences of their child's injury and subsequent support. As a result, potential participants were given a choice about how they took part in a semi-structured interview (either face-to-face, by Skype, email, or by telephone). This paper describes reasons parti- cipants gave for their chosen method of participation in a qualitative study focusing on their experiences of having a child suffer a burn- injury (Heath et al., 2018). The aim of this being to explore the reasons behind participants' preference for particular interview methods to help future researchers consider how offering choice might facilitate greater participation in research, particularly in that which focuses on a sen- sitive subject.

2. Methods

2.1. Ethical approval

Ethical approval was granted by the Research Ethics Committee of the Faculty of Health and Applied Sciences at the University of the West of England. For inclusion in the study, parents had to be English speaking, aged at least 18 years, with a child who suffered but survived a burn injury before the age of 18 years requiring hospital treatment. Exclusion criteria were non-accidental injury and when the child had died because of their injury.

J. Heath et al. Applied Nursing Research 43 (2018) 30–35

31

2.2. Recruitment

Participants were recruited via advertisements on burn charity websites, social media, a radio broadcast, university press release, and emails to eligible parents who had opted into a university research mailing list. These avenues were utilised as relationships with relevant, known, organizations is thought to improve participants' perceptions of the research as credible and acceptable (Altpeter, Houenou, Martin, Schoster, & Callahan, 2011). The advertisements directed parents to a website hosting the study information, consent form, and space to add their contact details should they wish to take part. The first author used these details to arrange an interview with consenting parents.

2.3. Participants

A purposive sample of 13 parents/carers (11 mothers, 1 father and 1 grandfather) whose child had experienced an accidental burn-injury was recruited. Informed consent was obtained from all participants via the online survey, and where audio recording was used, verbal consent was also obtained. Twelve parents/carers (10 mothers, 1 father and 1 grandfather) answered a question about why they had chosen their interview method.

2.4. Interviews

Interviews were carried out over the telephone, using Skype, in person or via email by the first author (a Clinical Psychologist and PhD Researcher) using the same semi-structured interview schedule for each method. Interviews covered: parents' experience of the injury and treatment, which models of support were available to them, and whe- ther they thought peer support would have been valuable (Heath et al., 2018). Interview mode was participant-led. During the post-interview debrief, participants were asked, “Can you tell me why you chose this method of participation?” Recorded interviews were transcribed ver- batim and email interviews were stored as word documents for analysis. All data files were anonymised and stored securely.

2.5. Analysis

The data was organised and analysed using a semantic inductive thematic analysis, following the six steps described by Braun and Clarke (2006). Direct quotations have been used from participants to ensure that interpretation of the data directly links to their words. The first author, with input from the supervisory team, conducted all data ana- lysis. Member checking then validated the interpretation and adequacy of the data.

3. Results

The mean age of the 12 participants whose data contributed to this paper was 42.7 years (range = 29.0–63.0 years). Nine identified as ‘White British’, two as ‘British’ and one as ‘British Indian’. At the time of interview, seven participants were employed, two were homemakers, one was a carer and another a student. The grandfather was retired. Nine participants were married, one was single, one was separated, and one chose not to disclose this information. The mean age of their child at the time of injury was 3.7 years (range = 0.6–13.0 years), and the mean time since the injury event was 3.1 years (range = 0.6–15.3 years). Nine children had experienced scalds, one from bath water and eight from hot drinks, two had flame burns, and one had injuries from flash flames. Interview characteristics are shown in Table 1.

Following thematic analysis of the responses to the question re- garding participation method, four themes emerged: convenience, openness despite upset, getting a “feel” for the researcher, and depth of response (Table 2). These themes are described below and illustrated

with quotations. Participants have been given a pseudonym to protect their anonymity.

3.1. Convenience

Participants' choice of method was often “purely for convenience”, giving them the opportunity to take part without putting additional demands on their time.

“I'm a very busy mum of three; I study as well, so it was just con- venience for me.” [Sue: Email].

For some, it was key that participation was via a method that was familiar or habitual to them, not requiring access to something that was not already in routine use.

“By the time it's taken me to write out all the answers I can't be bothered [with email]…[Skype is] just easier. I guess I'm used to Skyping a lot for work… so it was just the most convenient form for me, for us to talk. [Mary: Skype].

Nevertheless, there was perhaps a preference for face-to-face in- teraction where possible, although this could be sacrificed in favour of convenience.

“If we'd had a Skype account and I'd sorted it all out, then I might have done Skype but I think the phone call's just as quick personally.” [Carol: Telephone].

3.2. Openness despite upset

It was crucial for participants to feel that they could communicate openly with the researcher despite any upset they felt recalling the events and their experiences of their child's injury. For some, it was easier to do this in writing.

“I would find it hard to talk openly over the phone or Skype, and also struggle even when typing this, I feel a swell of emotion. Last night when I was typing my responses, I cried plus experienced the noises associated with the accident so I think doing it this way, for me anyway, has been better.” [Colin: Email].

Remote methods of communication, in which neither the partici- pant nor the researcher are actually visible, might be preferable for some participants. This may be the case for those concerned that the presence of a researcher, who might in some way implicitly judge their situation, could hamper their answers to potentially challenging or sensitive questions.

3.3. Getting a “feel” for the researcher

A sense of rapport, and sometimes a preference to see the re- searcher, was important to participants in order for them to experience a “more personal” connection during their participation in the research.

“I think email's impersonal. If I hadn't have spoken to you on the phone I wouldn't have got the feel I've got, I think you get a feel for a person on the phone don't you?” [Jim: Telephone].

Enabling the researcher to connect more personally with the parti- cipant, ensuring that the researcher was getting the information that they needed, particularly as the focus was on such an important and emotive issue, was also significant for participants.

“By talking to someone you can build a better picture up… I feel it's important about research like this and I feel that if you're typing the questions that you might not get everything you need… I thought this was the best way to get a feeling for, I suppose you, as well you get a feeling for the family.” [Beth: Telephone].

In addition to this, it was vital that the researcher used appropriate interpersonal skills to provide the participants with a safe, non-judge- mental space to communicate their experience.

“I've felt more emotional at times, and still able to say the right answer, because I can see your face as opposed to it feeling like there's someone judging the way you behaved, or behave. I would have probably been more cagey [on the phone] and more like yeah

J. Heath et al. Applied Nursing Research 43 (2018) 30–35

32

everything was fine in a lot of answers.” [Trish: Skype]. Due to the significance of the topic for these parents, it was also

important that data collection did not feel like a “normal anonymous survey”.

3.4. Depth of response

Participants were keen to support the research and wanted to take part in a way they thought would best ensure their responses provided in-depth data.

“With email, I don't feel you can give full enough answers, you could ask a question and I could give a three line answer whereas, when we're talking about it on the phone, I think you get more depth of informa- tion.” [Fran: Telephone].

Some participants felt that full interpersonal interaction was sig- nificant to the research process, valuing the verbal and nonverbal communication achieved during a face-to-face interview, and re- cognising the potential relevance of such information to the researcher.

“When there's a silence, on the phone you don't know what I'm thinking but you can see my face now, whether I'm recalling stuff... I think if I was on the phone or Skype I'd feel much more pressured to come up with an answer straightaway, whereas I feel face-to-face I can think about it and you can understand why I'm thinking about it… and with email it's too easy to just say something, whereas I think you'll know by my voice whether it's genuine or not, whereas I don't think you can do that on email.” [Theresa: Face-to-face].

4. Discussion

This paper has presented the preferences of 12 participants re- garding their choice of interview method when participating in re- search that focused on their experiences of having a child suffer a burn injury (Heath et al., 2018). Results indicated that their preferences were determined by personal convenience, their belief in their ability to be open with the researcher despite potential upset caused by the topic, their ability to get a “feel” for the researcher, and concern about giving adequate depth in their responses.

It is worth noting that, due to the preventable nature of burn in- juries in young children and the inherent probability of parent/carer

guilt and self-blame, this population is particularly sensitive to judg- ment from others. Therefore, although some participants benefit from seeing or hearing a researcher who demonstrates a non-judgemental attitude, it might be preferable for others to participate remotely to facilitate their openness during research so that fear of judgment does not hamper communication. It can also be the case that participants' self-disclosure is facilitated when the research is conducted outside of the clinical setting due to the informal nature of alternative environ- ments (Borbasi, Chapman, Gassner, Dunn, & Read, 2002). In cases where the clinical setting may restrict disclosure (for example, when a participant is concerned that a hospital-based interview might arouse painful memories) then remote methods of participation, or researcher flexibility, may be preferable. Such issues are likely to be relevant to patients within a number of other areas of healthcare.

Despite references to telephone interviews not being well suited to qualitative interviewing due to the lack of face-to-face contact re- stricting the development of rapport (Irvine, 2011), telephone inter- views were the most frequently chosen option for participation in the study (Heath et al., 2018). Hershberger and Kavanaugh (2017) also found telephone interviews more frequently chosen when offering a choice between email and telephone interviews in research within an- other sensitive area. According to participants in this study, a telephone interview offered a convenient method of participation whilst also al- lowing them to get a “feel” for the researcher. The interaction with the researcher over the telephone could also provide participants with a sense that the depth of their response was appropriate.

The theme ‘getting a “feel” for the researcher’ reflects the estab- lishment of rapport. As described earlier, this requires researchers to manage emotions expressed within researcher-participant interactions and make participants feel comfortable (Liamputtong, 2007), providing a safe opportunity or environment for them to speak openly with the researcher (Farooq, 2015; Pyer & Campbell, 2012). The method of in- terviewing used is likely to influence participants' responses, affecting the extent to which they feel comfortable enough to answer openly and honestly about a sensitive or personal topic (Dures et al., 2011). A more informal or relaxed interviewing style is likely to facilitate the dis- closure of material that might be withheld in more formal settings (Hart & Crawford-Wright, 1999).

When reasons for choosing Skype were explored, convenience and the ability to get a “feel” for the researcher emerged as themes. Participants' decisions to choose a method of participation that posi- tioned them remotely from the researcher was perhaps a strategy em- ployed by them to manage an emotional situation at a safe distance (Groves, 1979), whilst also being ‘face-to-face’ with the researcher. Reasons why telephone interviews might be preferable to participants, particularly when discussing sensitive topics, have been proposed pre- viously; comparatively, telephone encounters afford greater anonymity and reduced intensity (Chapple, 1999).

Indeed, the sharing of experiences via telephone and email may have been facilitated for some participants by the greater ‘social dis- tance’ between them and the researcher than is afforded in a face-to-

Table 1 Interview frequency, characteristics and emerging themes.

Method N Participants Mean (range) length of interview (minutes) Emerging themes

Telephone 5 4 mothers 1 grandfather

58 (50–75) Convenience Depth of response Getting a “feel” for the researcher

Email 4 3 mothers 1 father

2 emails: First email contained all questions. Second email contained individualised questions to clarify and expand on answers provided.

Convenience Openness despite upset

Skype 2 2 mothers 67 (61–73) Convenience Getting a “feel” for the researcher

Face-to-face 1 1 mother 57 Depth of response Openness despite upset

Table 2 Themes and sub-themes.

Themes Sub-themes

Convenience a. Demands on time b. Familiarity

Openness despite upset Getting a “feel” for the researcher a. Personal connection

b. (Non)Judgment Depth of response a. Reassurance

b. Interpersonal communication

J. Heath et al. Applied Nursing Research 43 (2018) 30–35

33

face encounter (Cook, 2012). This has been related to hypotheses about the role of social desirability, with the avoidance of revealing person- ally-perceived negative characteristics being heightened in the physical presence of another. This affect is reduced somewhat in telephone communication compared to face-to-face interview (Groves, 1990). Whilst participants might acknowledge the value of sharing their ex- periences, communicating them can be challenging due to concerns about how they might be perceived. Therefore, it could also be hy- pothesised that those participants who are concerned that researcher presence would cause them to shy away from particularly challenging or sensitive questions might feel more comfortable or safer about self- disclosure when using a method in which they are not physically visible. This factor may well have influenced the choice of some to participate via email (Cook, 2012; Hershberger & Kavanaugh, 2017; Meho, 2006).

However, these theories were not consistently supported in this study. One participant commented that not being face-to-face with the researcher might make participants present as “braver”, which could make the communication “very contrived”. In this case, seeing the re- searcher, as opposed to being on the telephone, was reassuring as facial expressions communicated a non-judgemental attitude that facilitated communication that was more candid. This finding supports Groves (1979), who reported that most respondents would have preferred to be interviewed face-to-face rather than by telephone regarding sensitive topics.

The third interview method also chosen for its convenience was email. Telephone calls could be made easily to participants at home or work; however, one could argue that email was even more convenient, potentially allowing responses to be composed intermittently, from any location, and with more privacy. Parents choosing to participate by email also commented that email allowed them to communicate their feelings and experiences despite the significant upset they experienced whilst recalling them. All other methods would require the researcher to be witness to this, which is perhaps too exposing for some partici- pants.

Only one participant volunteered for a face-to-face interview. For convenience, the first author offered to conduct the interview at a lo- cation of the participant's choice, but she chose to be interviewed at the researcher's place of work after describing feeling drawn to the inter- view location as it shared the name with, and was geographically close to, the hospital in which her child was treated. Although the parent did not describe this as convenient, it was preferred due to its familiarity. In addition to familiar methods of communication feeling convenient, perceived familiarity in other aspects of the research might be another factor that influences participants' openness and depth in responding despite the emotive situation.

Having the ability to expand on answers, time to think and, in the case of face-to-face interviews, the ability to use non-verbal methods of communication to demonstrate honesty were all important. Irvine (2011) found that telephone interviews were on average shorter than those conducted face-to-face because of the participant speaking for less time. The current study found that the mean length of interviews was longest for those conducted via Skype. There was more variation in the length of the telephone interviews, and the length of the face-to-face interview did not differ from the mean length of telephone interviews. Perhaps face-to-face interaction, with the physical separation afforded by Skype, allows participants to talk more openly than via telephone or in a ‘real world’ face-to-face setting.

It has previously been suggested that when there are no visual cues during an interview, the researcher must work hard to maintain the participant's motivation and stimulate their interest (Irvine, Drew, & Sainsbury, 2013). Therefore, telephone interviews may require greater concentration and energy on the part of the researcher to keep the participant engaged whilst maintaining the natural flow of the con- versation. An inability to do this successfully could result in shorter interviews, providing less data for analysis (Farooq, 2015). The length

of interviews in the current study suggests that the lack of visual data in the interview process did not create an insurmountable challenge, al- though it was also noted that the data gleaned from email responses was less rich than that offered during spoken interviews.

Email interviews also deprive researchers of data regarding facial expressions and body language, eye contact, and tone of voice (Meho, 2006). It is possible that some important visual or nonverbal cues that would be observed and could potentially act as prompts for further discussion in face-to-face or even telephone interviews are missed during email interviews. Although, whilst there are some consequences, it is also acknowledged that email interviews reduce, if not eliminate, some of the challenges associated with telephone, Skype, or face-to-face interviews. Such challenges can include perceived participant-re- searcher characteristics that might, consciously or unconsciously, in- fluence communication; for example, assumed status differences, race, gender, age, voice tones, dress, shyness, gestures, and disabilities. Therefore, if participants are able and willing to engage in research via email, then accommodating individual preferences might facilitate their participation and can generate rich insightful data (Hershberger & Kavanaugh, 2017).

The topics explored within the interviews for the wider study (Heath et al., 2018) were clearly important to the participants. Despite any upset experienced, the participants may have felt that they had a stake in providing all of the information they could in order to facilitate the development of parental support for others who faced similar cir- cumstances. Studies have supported this idea showing that many people, particularly those who identify as vulnerable, are keen to par- ticipate in research altruistically where sharing their experiences may improve the lives of others (Alexander, 2010). For some participants, the opportunity to talk about their experiences might have been a therapeutic process (Hart & Crawford-Wright, 1999), but it is also im- portant to consider the impact on the researcher.

After focusing on the participant experience, we would be remiss at this point not to discuss the potential impact on the researcher. The revelations made by participants during interviews can be disturbing for qualitative researchers (Hart & Crawford-Wright, 1999). Johnson and Clarke (2003) found that researchers using telephone and face-to- face interviewing methods to collect sensitive data experienced a number of difficulties. These difficulties centred on issues of a lack of training and inexperience, concerns regarding confidentiality, role conflict, costs to the participants, the desire for reciprocity, and feelings of isolation. When considering the impact of participating in research on sensitive subjects, it is therefore important to acknowledge the po- tential impact on all those involved, including the researchers (Dickson- Swift et al., 2009).

Frequent emotionally intense encounters can be associated with high levels of emotional stress (Rager, 2005). Collecting interview data in a face-to-face capacity, or via Skype, allows researchers to see par- ticipants' pain. During telephone interviews it is heard and, with in- creasing use of email interviews, it is now frequently read. It is neces- sary to be mindful that hearing a participant's distress, or even reading about it asynchronously, might be particularly difficult for researchers in the absence of non-verbal cues as to the participants general well- being. It is therefore imperative that, in addition to signposting parti- cipants to appropriate support, researchers are also adequately sup- ported with time between interviews to debrief and recover.

This study has several limitations. The findings are based on re- sponses to one methodological question in a wider interview schedule (Heath et al., 2018). The sample is predominately female and there may be gender differences in male and female preferences and responses to different interview methods, particularly those using technology (e.g. email, skype, instant messaging) (Hershberger & Kavanaugh, 2017). However, despite parents of burn-injured children often experiencing high degrees of shame and guilt about what happened, which is likely to affect their willingness to engage in research about their experience, an adequate sample and data set was achieved to meet the research

J. Heath et al. Applied Nursing Research 43 (2018) 30–35

34

aims and provide valuable insights into the methodological preferences of participants in sensitive research (Guest, Bunce, & Johnson, 2006). Researcher preferences and perceptions of these interview methods are not explored in this paper but have been published separately (Heath, 2018).

Although this paper reports a small amount of data from a relatively low number of participants, this does not diminish its contribution to an important line of enquiry. The key message is that participants' pre- ferences can vary and should be considered to optimize data collection. However, whilst using multiple methods of data collection can increase the number of participants recruited into research, researchers should be aware that the different techniques used can produce varying depths and quantities of data which may affect data analysis (Heath, 2018). Further research should explore the procedures involved in carrying out email interviews and compare this method with other more established methods of qualitative data collection. For instance, the process of asking follow-up questions or probing for more information is quite different between email and spoken interviews, most notably because in email interviews this is not synchronous. In addition to the factors in- fluencing participant choice in interview method presented here, gender and age may also play a role and warrant future investigation. Future researchers may also wish to explore this topic further and in other areas of healthcare.

5. Conclusions

This study has highlighted rarely explored issues considered by participants when choosing how to participate in a qualitative study of a sensitive subject. The issues considered were convenience, their ability to be open and honest with the researcher despite any potential upset caused by the subject matter, their ability to get a “feel” for the researcher, and concern about providing adequate depth in their re- sponses. Findings suggest that, in order to maximise recruitment and the quality of interview data available for analysis, researcher flexibility is important and, when possible, participants should be offered a choice of participation method. Researchers should be aware of the procedures and appropriateness of using different methods of interviewing, and how such methods might affect them emotionally. When researching sensitive topics, researchers should ensure that, as well as safeguarding participant distress, they also consider the risks to themselves.

Declarations of interest

None.

References

Alexander, S. J. (2010). 'As long as it helps somebody': Why vulnerable people participate in research. International Journal of Palliative Nursing, 16(4).

Altpeter, M., Houenou, L. O., Martin, K. R., Schoster, B., & Callahan, L. F. (2011). Recruiting and retaining hard-to-reach populations: Lessons learned and targeted strategies from arthritis physical activity intervention studies. Arthritis Care & Research, 63(7), 927–928.

Bartlett, R., Milne, R., & Croucher, R. (2018). Strategies to improve recruitment of people with dementia to research studies. Dementia, 1–11. http://dx.doi.org/10.1177/ 1471301217748503. (1471301217748503) http://journals.sagepub.com/doi/pdf/ 10.1177/1471301217748503.

Bertrand, C., & Bourdeau, L. (2010). Research interviews by Skype: A new data collection method.

Borbasi, S., Chapman, Y., Gassner, L.-A., Dunn, S., & Read, K. (2002). Perceptions of the researcher: In-depth interviewing in the home. Contemporary Nurse, 14(1), 24–37.

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101.

Braun, V., & Clarke, V. (2013). Successful qualitative research: A practical guide for be- ginners. Sage.

Brett, J., Staniszewska, S., Mockford, C., Herron-Marx, S., Hughes, J., Tysall, C., & Suleman, R. (2014). Mapping the impact of patient and public involvement on health and social care research: A systematic review. Health Expectations, 17(5), 637–650.

Chapple, A. (1999). The use of telephone interviewing for qualitiative research. Nurse

Researcher, 6(3), 85 (through 2013). Clark, T. (2008). We're over-researched here!' exploring accounts of research fatigue

within qualitative research engagements. Sociology, 42(5), 953–970. Cook, C. (2012). Email interviewing: Generating data with a vulnerable population.

Journal of Advanced Nursing, 68(6), 1330–1339. Couper, M. P. (2011). The future of modes of data collection. Public Opinion Quarterly,

75(5), 889–908. Deakin, H., & Wakefield, K. (2014). Skype interviewing: Reflections of two PhD re-

searchers. Qualitative Research, 14(5), 603–616. Dickson-Swift, V., James, E. L., Kippen, S., & Liamputtong, P. (2009). Researching sen-

sitive topics: Qualitative research as emotion work. Qualitative Research, 9(1), 61–79. Duffy, B., Smith, K., Terhanian, G., & Bremer, J. (2005). Comparing data from online and

face-to-face surveys. International Journal of Market Research, 47(6), 615. Dures, E., Morris, M., Gleeson, K., & Rumsey, N. (2011). The psychosocial impact of

epidermolysis bullosa. Qualitative Health Research, 21(6), 771–782. Farooq, M. B. (2015). Qualitative telephone interviews: Strategies for success. Fox, F. E., Morris, M., & Rumsey, N. (2007). Doing synchronous online focus groups with

young people: Methodological reflections. Qualitative Health Research, 17(4), 539–547.

Groves, R. M. (1979). Actors and questions in telephone and personal interview surveys. Public Opinion Quarterly, 43(2), 190–205.

Groves, R. M. (1990). Theories and methods of telephone surveys. Annual Review of Sociology, 16(1), 221–240.

Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An ex- periment with data saturation and variability. Field Methods, 18(1), 59–82.

Hart, N., & Crawford-Wright, A. (1999). Research as therapy, therapy as research: Ethical dilemmas in new-paradigm research. British Journal of Guidance and Counselling, 27(2), 205–214.

Hershberger, P. E., & Kavanaugh, K. (2017). Comparing appropriateness and equivalence of email interviews to phone interviews in qualitative research on reproductive de- cisions. Applied Nursing Research, 37, 50–54.

Irvine, A. (2011). Duration, dominance and depth in telephone and face-to-face inter- views: A comparative exploration. International Journal of Qualitative Methods, 10(3), 202–220.

Irvine, A., Drew, P., & Sainsbury, R. (2013). ‘Am I not answering your questions properly?’Clarification, adequacy and responsiveness in semi-structured telephone and face-to-face interviews. Qualitative Research, 13(1), 87–106.

Jo, M. (2007). The reconfiguration of hospital services in England. London: Kings Fund. Available online at http://www.kingsfund.org.uk/publications/briefings/the_1.html, Accessed date: 29 February 2008.

Johnson, B., & Clarke, J. M. (2003). Collecting sensitive data: The impact on researchers. Qualitative Health Research, 13(3), 421–434.

Joseph, C. L., Ownby, D. R., Havstad, S. L., Saltzgaber, J., Considine, S., Johnson, D., ... Gibson-Scipio, W. (2013). Evaluation of a web-based asthma management inter- vention program for urban teenagers: Reaching the hard to reach. Journal of Adolescent Health, 52(4), 419–426.

Knapp, H., & Kirk, S. A. (2003). Using pencil and paper, internet and touch-tone phones for self-administered surveys: Does methodology matter? Computers in Human Behavior, 19(1), 117–134.

Lawrence, J. W., Mason, S. T., Schomer, K., & Klein, M. B. (2012). Epidemiology and impact of scarring after burn injury: A systematic review of the literature. Journal of Burn Care & Research, 33(1), 136–146.

Lee, R. M. (1993). Doing research on sensitive topics. Sage. Liamputtong, P. (2007). Researching the vulnerable: A guide to sensitive research methods.

Sage. Mcquaid, D., Barton, J., & Campbell, E. (2003). Researchers BEWARE! Attrition and

nonparticipation at large. Journal of Burn Care & Research, 24(4), 203–207. Meho, L. I. (2006). E-mail interviewing in qualitative research: A methodological dis-

cussion. Journal of the Association for Information Science and Technology, 57(10), 1284–1295.

National Burn Care Review (2001). Standards and strategy for burn care. A review of burn care in the British Isles. Retrieved from http://www.britishburnassociation.org/ downloads/NBCR2001.pdf.

Novick, G. (2008). Is there a bias against telephone interviews in qualitative research? Research in Nursing & Health, 31(4), 391–398.

Pyer, M., & Campbell, J. (2012). Qualitative researching with vulnerable groups. International Journal of Therapy and Rehabilitation, 19(6), 311–316.

Rager, K. B. (2005). Compassion stress and the qualitative researcher. Qualitative Health Research, 15(3), 423–430.

Wellings, K., Branigan, P., & Mitchell, K. (2000). Discomfort, discord and discontinuity as data: Using focus groups to research sensitive topics. Culture, Health & Sexuality, 2(3), 255–267.

Wilkerson, J. M., Iantaffi, A., Grey, J. A., Bockting, W. O., & Rosser, B. S. (2014). Recommendations for internet-based qualitative health research with hard-to-reach populations. Qualitative Health Research, 24(4), 561–574.

Heath, J. (2018). Reflections on researcher positionality when using multiple techniques of qualitative data collection to facilitate participation in research focusing on a sensitive subject. Bath, UK: Paper presented at the Qualitative Research Symposium.

Heath, J., Williamson, H., Williams, L., & Harcourt, D. (2018). Parent-perceived isolation and barriers to psychosocial support: A qualitative study to investigate how peer support might help parents of burn-injured children. Scars, Burns & Healing, 4 (2059513118763801).

J. Heath et al. Applied Nursing Research 43 (2018) 30–35

35