Title: Problems facing qualitative researchers: Some examples.
Affiliation: Carnegie Research Institute, Leeds Metropolitan University
Related SIG: Physical Education and Sports Pedagogy
Paper presented at the British Educational Research Association New Researchers/Student Conference, Institute of Education, University of London, 5 September 2007
The research process has been likened to a journey littered with “stumbling blocks, changes of direction and major breakthroughs” (Brackenridge, 2001, p.148). This paper will draw on my experiences as a PhD candidate conducting qualitative research into good practice within competitive youth swimming to examine attempts (successful and not) made to mitigate against these stumbling blocks. It will use the five problem areas facing social researchers identified by Brewer (1993) as a framework to discuss how these problem areas shaped the study design and practice. It is hoped that discussing potential solutions to some of the problems encountered during my study will assist others in the preparation of future research.
Brewer’s five problems and ‘sensitive’ research
Following his ethnographic research on the Royal Ulster Constabulary in Northern Ireland, Brewer (1993) identified five types of problems facing social researchers – problems of methodology, technique, ethics, social context and personal security. Methodological problemsare the theoretical and epistemological issues raised by the technique and practice adopted; technical problemsare those associated with the researcher’s techniques and practices; ethical problems describe the moral dilemmas raised by the research; contextual problems are those that arise from the social, political and economic environment in which the research is set; and personal security problems refer to the researcher’s physical safety when conducting research.
While the first three problems are common to all research, the latter two are more commonly associated with ‘sensitive’ research. Lee & Renzetti (1993) suggest any topic, depending on the context, can be sensitive. However, they identify topics that pose a significant threat to the participants and/ or the researcher as the most likely to become sensitive. For example, a topic that:
“… potentially poses for those involved a substantial threat, the emergence of which renders problematic for the researcher and/ or the researched the collection, holding and/ or dissemination of research data.”
(Lee & Renzetti, 1993, p.5)
Threat may arise because the topic explores a taboo subject or an area of social deviance, impinges on powerful social interests or examines participants’ personal values. Examples of sensitive research includes that on drug use (Inciardi, 1993; Fischer et al., 2006), assisted dying (Bilsen et al., 2004), paramilitary activity (Brewer, 1991; Feenan, 2002), police activity (Westmarland, 2000), domestic violence (Bassuk, 2006; Martin et al., 2006) and sexual abuse (Finklehor & Williams, 1988; Burgess et al., 2005).
My research looks at good practice in youth swimming. Although this may not initially appear to be ‘sensitive,’ issues of child safety and child protection – both potentially controversial topics – are central to understandings of good practice when working with young people in any setting. In order to outline some of the main methodological, technical, ethical, contextual and security problems associated with my research and how these issues shaped the study design, an overview of my research and its context is warranted.
My research: Context
Although swimming is the most popular participation sport in the country (Office for National Statistics, 2002), there is a notorious lack of success among British elite swimmers at an international level; the British swim team have won only eight Olympic medals since the 1998 Seoul Games and have not had a gold medalist in almost 20 years (Green & Houlihan, 2006). To turn the team’s fortunes around, a new, tough-talking national performance director, Australian Bill Sweetenham, was employed in 2000. Branding British Swimming’s elite-development programme “a massive flop, an abysmal failure” (Lord, 2002), Sweetenham set about restructuring the elite-development programme in a way that mirrored the government’s new focus on developing medal-winning athletes. Until the mid-1990s, government policy emphasis was largely directed towards mass participation initiatives or what has been termed ‘Sport for All’ (Coghlan and Webb, 1990; Houlihan, 1991, 1997; Henry, 2001). This focus shifted in 1995 with the publication of the Conservative government’s sport policy statement Sport: Raising the Game (Department of National Heritage, 1995), which shifted the focus onto the development of elite performers (Houlihan, 1997). This focus on elite sport continued with the election of the New Labour government in 1997 and the publication of Game Plan (DCMS/ Strategy Unit, 2002).
As part of his plan, in 2003 Sweetenham introduced The Swimmer Pathway, a swimming-specific Long-Term Athlete Development (LTAD) plan as touted in Game Plan, and began withholding Lottery funding and GB team places from swimmers who refused to follow his regime (Lord, 2002). Under the LTAD plan, swimming is classified as an early specialization sport, requiring sport-specific training from the ages of 9-10 if medal success is to be achieved (Vorontsov, 2002, cited in ASA, 2003).
Sports scholars have expressed concern, however, that this increasing focus on developing elite athletes is incongruent with athlete welfare initiatives. As Malkin, Johnston & Brackenridge explain:
“Where there is stress on sports to produce ‘results,’ in the form of tangible medal successes ... coaches and organisations are required to put their energies into performance enhancement and this may undermine the humanitarian ethos or ‘spirit’ of codes that emphasize athlete development and welfare.”
(Malkin, Johnston & Brackenridge, 2000, p.152)
It is against this backdrop that my research into good practice in competitive youth swimming takes place. The aim is to explore good practice within the culture of competitive youth swimming. The principle objectives are:
To explore how good practice within the culture of competitive youth swimming is realised in three case-study clubs at different levels of the elite-development ladder.
To explore the perceptions and understandings of good practice among key adult personnel who deal directly with competitive junior swimmers.
An ethnographic approach was adopted. Observations took place at three North West swimming clubs at different levels of the elite-development ladder that were affiliated to the national governing body of the sport, the Amateur Swimming Association (ASA). This was followed by in-depth interviews with key figures in the clubs who work directly with competitive junior swimmers. To date, I have completed observations at one club, Central Seals, and have interviewed eight key adult personnel there. Observations at the second club, Northern Eels, are currently underway. All quotes used in this paper are from observations and interviews at Central Seals.
The remainder of this paper will discuss some of the main methodological, technical, ethical, contextual and security problems associated with my research (see Table 1) and consider how these shaped the study design. I recognise, however, that many of the problems discussed here are integrally linked; technical problems identified in the planning stages of the study impacted on the methodological decisions made, and problems of personal security are a direct consequence of the context in which the study takes places, for example. As such, the distinctions made by way of subheadings are artificial; each problem does not necessarily occur independently.
The first methodological problem encountered is common to all research: Which method is best suited to achieving the research aims and objectives? My research is concerned with understanding competitive swimming culture and how key stakeholders at clubs understand good practice in relation to youth swimming. As such, ethnography was considered the most suitable data-collection method. Ethnography involves a range of techniques, including direct observation and formal and informal interviewing. As Hammersley & Atkinson (1995, p.1) explain, it involves the researcher “participating overtly or covertly in people’s daily lives for an extended period of time, watching what happens, listening to what is said, asking questions.” The strength of ethnography, then, lies in its ability to document in detail the world being observed in terms of the meanings and behaviours of the people in it. Consequently, a period of observation plus in-depth interviews were considered the most appropriate ways of gaining an understanding of swimming culture and how key stakeholders at clubs understand and carry out good practice.
The solution to the above problem created further methodological challenges, including deciding which clubs to observe and which individuals to interview. With around 1,600 competitive swimming clubs affiliated to the ASA (ASA, 2004), a method of identifying those clubs at different levels of the elite-development ladder was needed to select clubs from each level to take part. The National Speedo Swimming League competition was identified as a useful tool to categorize clubs at different developmental levels. More than 500 clubs, representing 12,000 swimmers, compete annually in The National Speedo Swimming League (National Speedo Swimming League North West, 2005). The league is divided into three levels of competition:
Premier Division– featuring elite clubs
Division One – featuring ‘sub-elite’ clubs
Division Two – featuring ‘lower-level’ clubs
One club from each division in the National Speedo Swimming League North West region, where the researcher has an identity presence as a former British international swimmer, was selected to take part in the research. Central Seals, the first club observed, is an elite club from the Premier Division. Northern Eels, where I am currently observing, is a ‘lower-level’ club from Division Two. The third club to be observed will compete in Division One of the league.
The decision on who to interview was linked to the aim and objectives of the study. Since the research was concerned with good practice when working with competitive junior swimmers, the adult personnel who deal directly with these swimmers were identified and invited for interviews. These were:
The Club Welfare Officer (CWO), who is in charge of child welfare issues within the club
Once decisions had been made on the methods to be used and the sample set, problems that are both methodological and technical in nature became central to the study’s success. These included how to gain access to the field and to interview participants and how to gain the trust of those involved. A discussion of the key technical problems encountered follows.
A key technical problem was how to present the research to best encourage access to the field. I had initially wanted to focus the research on child protection issues within swimming because the ASA in 1995 became the first NGB to introduce a child protection policy, following a high-profile case of child abuse involving Olympic swimming coach Paul Hickson (Myers & Barrett, 2002). The flaws in this approach were recognized during the planning stages, namely that this focus treated child protection as distinct from the more general principles of good practice rather than part of them, and that such a controversial focus would inhibit access to clubs and interview participants.
Ultimately, focusing the research on good practice as a whole, a topic with which all those involved in swimming should be concerned (ASA, 2004), was key to facilitating access to the research field and interview participants as head coach Andrew noted at our first meeting:
“I’d like to learn more about good practice, you know, finding out how the club is doing, and whether it could improve on anything.”
It also was anticipated that the appeal of giving coaches and others involved in running swimming clubs the opportunity to express their views on the sport to which they dedicate so much time would encourage participation. Both these points were emphasized on the information sheets distributed to all potential participants. The decision to observe the sessions of more than one coach at each club also was important. This allowed the researcher to get a broader understanding of swimming culture and to avoid singling out one individual, which could inhibit trustful relations. At Central Seals, this approach worked well with all six of the coaches agreeing to take part in the study although, as I will explain shortly, the head coach Andrew’s approval and involvement also proved crucial.
Deciding when to observe also was crucial to facilitating access. The swimming season runs from September to July. For elite clubs with national-level swimmers, July is the busiest time in the swimming calendar as national age-group, youth and senior championships are held throughout this month. August, meanwhile, is the quietest month; many elite clubs shut down to allow swimmers and coaches time off. Observing an elite club in July and August, then, would be difficult because coaches and swimmers would either be frequently away at competitions or on holiday. Consequently, it was decided that the ethnography at Central Seals would be completed before the July peak when officials are less likely to be distracted by preparing for competitions and, thus, more accommodating to my role.
Being granted access to the field for observation purposes presented another technical problem. Sullivan & Cornfield (1982, cited in Lee, 1993) argue that when negotiating access within an organisation, it is important to speak directly with those in the most senior positions in order to give the research legitimacy. In swimming terms, the club’s head coach represents the most senior position. However, while physical access to a fieldwork setting may be secured in advance by those in senior positions, actual access can remain problematic (Brewer, 1993). This is because where settings are formerly organised, such as in a swimming club, there exists what Dingwall (1980) calls a “hierarchy of consent” and often it is simply assumed that superiors have the right to permit their subordinates to be studied. This can result in mistrust from those being observed, leading to difficulties in gaining actual access to the setting (Lee, 1993). Gaining access to the field and winning the trust of those involved, then, is a continual process that is renegotiated and consolidated, particularly during long periods of fieldwork.
To avoid difficulties in gaining consent and actual access to the setting and to generate trust, all coaches – head coaches first, followed by assistant(s) – were approached individually about the study and access negotiated with each directly. In practice, the assistant coaches at Central Seals were given little choice over their involvement, however. The head coach, Andrew, had been contacted first through a gatekeeper. When Andrew agreed to participate, he introduced me to the assistant coaches individually so I could outline the study to them. However, Andrew told the assistant coaches before they met me that they were obliged to take part in the study. As Andrew explained to me: “They’ll take part because I’ll tell them to” (fieldnotes, 24/4/07). Although I stressed to the other coaches that involvement was voluntary and that Andrew would not be informed who had consented and who had not, it was no surprise that all agreed to take part.
The process of gaining access was not as simple as just getting all the coaches on board, however. Although coaches occupy the most senior positions within swimming clubs, other official posts also exist. The club chair oversees the general running of the club and, as such, represents the club’s symbolic head. To ensure the club as a whole was aware of my presence and remit, when access with coaches had been secured, the club chair was informed about the study and agreement sought that I could observe those coaches who had consented to take part. Equally, to ensure parents of athletes knew of my presence, an information sheet was posted on the club notice board containing my contact details and I wore a Leeds-Met branded T-shirt and name tag at the pool so I was easily identifiable. Andrew had initially been reluctant for me to post the information sheet but on explaining that it was required by my university ethics committee, he agreed. However, the sheet was removed twice in two weeks by someone unknown so on the third attempt, to add credibility, I printed it out on Leeds Met-headed paper and asked Andrew to attach a note asking that it remain on the board until further notice. This was more successful and it was not removed until I left the club.
Another technical difficulty arose from the context of conducting observations at a swimming pool. Pools are notoriously hot and elite swimming clubs tend to train twice a day, early in the morning and late in the evening, when pools are otherwise unused and to fit training around school. Two of the six squads at Central Aquatics trained twice a day and two other squads also had morning sessions. These typically began at 5am while the evening sessions finished at 9.30pm. In the first week of observations, I observed each of the six squads five times, totalling 30 hours. The hot atmosphere coupled with travelling 1 ½ hours each way meant I quickly found observing both morning and evening sessions too punishing a schedule as this extract from my research diary at the end of week one attests:
“After only a few days, I am tired, irritable and my eczema has flared up through the hours at a time spent on the humid poolside. The 5am starts and 9pm finishes make it a really long day and I’m not necessarily being productive.”
Consequently, I was forced to limit my observations to once daily, prompting Central Seals head coach Andrew to joke I was “whimping out” by not committing to early morning starts (fieldnotes, 30/4/07).
In terms of the technical difficulty associated with gaining access to interview participants, the decision on when to conduct interviews was crucial. Approaching individuals to be interviewed is easier once trust has been established (Robson, 2002). For this reason, potential interview participants were contacted towards the end of the observation period, after I had been a regular presence at the club. Over the weeks, all the coaches grew progressively accepting of my presence; they began to call me over to sit with them and even asked for help with the sessions. Consequently, gaining their consent to a formal interview was relatively straightforward. Rapport building was more difficult with the CWO and team manager, however. Parents were discouraged by several of the coaches from spectating at the sessions because, as Andrew explained it, “they can get too involved, timing their kids” (fieldnotes, 25/6/07). As the roles of CWO and team manager are invariably held by parents of swimmers, regular sightings of these individuals was rare so building a rapport with them was more difficult than anticipated.
The question of how to record data represents another technical problem. This is particularly true during fieldwork, where the ethnographer’s conventional notepad can be obtrusive and create mistrust (see for example, Brewer, 1991; Flintoff, 1993). To reduce the visibility of note taking and establish trustful relations, I decided to forego a notepad and record notes after an event by memory. In practice, at Central Aquatics this was initially unnecessary as observations were carried out from an elevated spectator area where I could write unnoticed by the coaches. As the weeks progressed, however, and I was called to sit on the poolside, unobtrusive note taking became impossible. Instead, I kept a voice recorder in my shorts and recorded reminders of key events whenever I was able to leave the poolside. Mostly, these recordings took place in the toilets and were expanded upon on the drive home then written up on a PC on arrival. I found using a voice recorder to take ‘notes’ particularly useful as it was both easier and faster to articulate verbally events I had witnessed than jotting partial notes.
In terms of interviews, eliciting frank and lengthy responses, especially when discussing a sensitive topic, can be difficult (Bringer, 2002). This represents another technical problem. In such cases, Brackenridge (1997) recommends giving participants as much control as possible over the interview to facilitate discussion, while Feenen (2002) argues that stressing the research is independent encourages participants to talk openly. Distributing the interview questions in advance also can help establish trust and promote open dialogue (Robertson, 2000; Olivier & Olivier, 2001; Feenen, 2002). Others have suggested that potential participants are more likely to agree to participate in studies when strong assurances of anonymity and confidentiality are given (Singer, Von Thurn & Miller, 1995).
To establish trust and transparency, potential participants were informed on information sheets and during initial meetings that the research was independent of the ASA and examples of how confidentiality and anonymity would be assured were given. To further reassure participants, the interview questions were circulated to participants before consent was given and the details of a university member who could be contacted if a participant wanted to complain about the research process were distributed. In addition, to give participants some control over the interview process, the choice of when and where the interview would take place was left to each individual. Luckily, the pool was based at a large leisure complex and finding a suitable space was easy, although loud speaker staff announcements interrupted interviews on several occasions. In terms of scheduling interviews, several participants preferred to be interviewed at the end of their first training session, meaning 7am. In such cases, practical matters such as ensuring I had sufficient petrol in my car and organising interview scripts and recorders the night before were essential.
Like methodological and technical problems, ethical problems are common to all forms of research. When the topic is sensitive, however, researchers need to be more aware of the ethical issues it poses than when the research is more innocuous (Lee, 1993). As noted earlier, my research focuses on good practice when working with junior competitive swimmers, and the potentially sensitive topics of child safety and child protection are central to this. Importantly, however, I did not ask about specific incidences of abuse, bullying or unprofessional behaviour. Rather, I focused on individuals’ understandings and opinions of good practice in competitive youth swimming and how these are incorporated in swimming culture. Nevertheless, the following potential risks to interview participants were identified:
Participants may have personal experience of unprofessional behaviour, bullying or abuse within or outside of swimming and raising the subject in this study may trigger a negative reaction.
Participants may come to believe during the course of the study that a child they know is in danger of bullying or abuse from someone in swimming or outside it.
Participants may say too much and later regret their comments (Finch, 1993).
Participants may lose confidence in their own coaching/ working practice.
Advice on how best to mitigate against these potential risks was taken from experienced researchers working in sensitive fields. For example, to familiarise potential participants with the research topic, information sheets were distributed to potential participants and those interested were invited to meet with me individually to discuss the research before agreeing to participation (Robertson, 2000; Feenen, 2002). Participants also were repeatedly informed that they may, without penalty, refuse to answer a specific question or rescind consent at any time (Brackenridge, 2001; Oliver & Oliver, 2001). Participants also were reminded of my decision to halt any interview if a participant appeared to be in difficulty.
To prevent participants from losing confidence in their own working practice, I followed the advice of Brackenridge (2001) and distributed debriefing packs to interview participants containing ASA and sports coach UK definitions of good practice and abuse, information on relevant legislation and reporting routes within and outside swimming, and the contact details of organisations from which they may access further information and training and counselling. In addition, to ensure appropriate backup was available in case a participant revealed or suspected abuse, bullying or poor practice, I took the contact details of counselling services in the area to all interviews so as to avoid being drawn into counselling individuals themselves when I am untrained to do so.
Particular ethical difficulties surround the possibility of participants discussing knowledge or suspicions of poor practice, bullying or abuse. For example, should researchers report suspicions of illegal behaviour or poor practice? The design of the study influences the level of information available to report suspicions. By their nature, for example, anonymous questionnaires leave the researcher with limited information to report if poor practice, bullying or abuse is revealed. The level of information available to the researcher is greater when face-to-face interviews are conducted, however, as in my study. Deciding how to deal with this ‘guilty knowledge’ (Fetterman, 1984) is complex. Researchers can obtain guilty knowledge if participants name and accuse, openly or accidentally, an individual of unprofessional behaviour, bullying or abuse, or are witness to poor practice, bullying or abuse. All adults, including researchers, are obliged by law to report child abuse if they are aware of it or if a child is in danger of it (DfES, 1989). Indeed, Sagarin & Moneymaker (1979, cited in Feenan, 2002) argue that claiming to protect confidentiality in light of a criminal offence amounts to aiding and abetting that crime. Crucially, hearsay statements do not carry the same weight as directly witnessed accounts (Feenan, 2002) and the veracity (or lack thereof) of any allegation must be decided by professionals (Brackenridge, 2001). Similarly, the ASA mandates that allegations of poor practice and bullying should be reported to its legal department or to the anonymous Swimline facility, which is modelled on the NSPCC’s Child Line.
The consequences of any unfounded allegation also must be considered: An allegation of abuse can destroy the career of the accused, even if they are later proved innocent (Brackenridge, 2001; Bringer, 2002), and allegations of poor practice can cause resentment within swimming clubs and result in the coach leaving the club. It is vital for researchers to consider which actions they will not voluntarily disclose and which actions they will automatically report to the relevant authorities. These decisions must be discussed with participants within the informed consent process (Steinberg et al., 1999). Steinberg et al. (1999) recommend researchers report suspicions of illegal behaviour only if there is sufficient evidence to support a reasonable suspicion. On the other hand, Kotch (2000) notes that reporting suspicions of abuse can breech researcher-participant confidentiality and argues that maintaining confidentiality is more important than reporting suspected illegal behaviours. Van Maanen (1983), meanwhile, argues the decision of whether or not to report/ break confidentiality depends on the nature of the relationship the researcher has with the participants and the seriousness of the crime. Past researchers studying child protection and abuse have reported suspicions of illegal behaviour and poor practice if there is sufficient evidence of it (see for example Brackenridge, 2001; Bringer, 2002). Indeed, there is evidence to suggest that reporting or informing participants that the researcher intends to report suspicions of illegal behaviour has little impact on participants’ decision to take part (Dawes Knight et al., 2006).
Given these issues and following the protocol for similar studies of a sensitive nature (see for example, Brackenridge, 1997, 2001; Feenan, 2002), participants were asked to use aliases during interviews if discussing unprofessional behaviour to prevent me from acquiring ‘guilty knowledge.’ In practice, although participants were reminded not to use real names before their interview, several did although not in the context of discussing unprofessional behaviour. Other researchers have encountered similar situations (see for example, Feenan, 2002; Brackenridge et al., 2007), suggesting asking participants to use aliases is insufficient protection against acquiring ‘guilty knowledge’ and further safeguards should be taken.
In addition, to empower participants by giving them control over what happens to the information they have, it was decided that if a participant claimed to have knowledge of or concerns about unprofessional behaviour or abuse, they would be informed of the appropriate reporting mechanisms and of requirements to report such behaviours. I also decided that if at any point I developed a reasonable suspicion that illegal behaviours were taking place, I would report my suspicions to social services and Swimline, and that if I witnessed poor practice, I would report it to the ASA. To minimise the impact of this decision on researcher-participant confidentiality, participants were informed of my intentions at several stages of the research process. This appeared to have no significant impact on participants’ decisions about whether or not to take part as all those approached to participate agreed.
To limit the potential of participants saying too much and later regretting their comments but feeling powerless to rectify the situation (Finch, 1993), I gave participants control over the digital voice recorder used to record each interview; participants were shown before the interview how to start and stop the recorder and were given the device to place nearby during interviews in case they chose to pause or halt the interview at any point. None did, however. In addition, copies of anonymised transcripts were made available to each participant and they were invited to change or delete anything they were unhappy. This system of member checking was encouraged rather than enforced by the researcher, however, out of recognition that some participants may find it difficult to read their experiences in print and others may not want the additional workload (Robertson, 2000). In practice, three of the eight participants at Central Seals returned their transcripts unchanged while the remainder did not request to see their transcript.
In all research, the protection of research participants is paramount (BSA, 2002; SRA, 2003; Dench, Iphofen & Huws, 2004). So far, this section has considered the potential risks to participants arising from my research. Importantly, however, there also are risks for researchers, particularly when working on research of a sensitive nature (Brannen, 1988; Lee & Renzetti, 1993). The following were identified as potential risks to the researcher:
Coping with potentially harrowing accounts of past or present unprofessional behaviour, bullying or abuse inside or outside swimming.
Deciding how to deal with ‘guilty knowledge’ if disclosed.
Managing participants who attempt to recruit the researcher into personal causes.
Being psychologically affected by one’s research is not uncommon (see for example, Hearn, 1998; Kirkwood, 1999; Robertson, 2000; Brackenridge, 2001). Researchers should be aware of the possibility of emotional effects on them caused by the research and make arrangements for handling this (Hearn, 1998). Brackenridge (2001) worked with a social worker before interviewing sexual abuse survivors and Robertson (2000) received counselling to help her recover from her research into bulimia. Loftland & Loftland (1995), meanwhile, suggest keeping a personal diary can offer researchers an outlet for their feelings when conducting sensitive research.
As Brackenridge (2001) found, participants occasionally wish to discuss harrowing accounts of past or present unprofessional behaviour, bullying or abuse with researchers they consider sympathetic to their cause. This is true even when the researcher does not directly ask about such incidences. It was recognised that this could occur during interviews and, consequently, the potential existed for me to be psychologically affected by the research. This risk was discussed at length with my supervisors and I spoke with a counselor about my concerns before undertaking the research. Although none of the research participants have so far disclosed anything that has had a lasting negative impact, I have found writing a personal diary a useful avenue through which to express my frustration when interviews did not go as well as expected or anger at some participants’ interest not stretching beyond swimmers with talent.
A consequence of participants disclosing past or present unprofessional behaviour, bullying or abuse to researchers is deciding how to deal with ‘guilty knowledge’ (Fetterman, 1984). As noted earlier, deciding what to do with this information is a complex task. Importantly, research data given in confidence do not enjoy legal privilege and may be liable to subpoena by a court (BSA, 2002). Researchers have a responsibility to inform participants of this and to consider in advance which actions they will not voluntarily disclose and which actions they will report to the relevant authorities. In my study, the advice of experienced researchers was followed and I decided to report suspicions of illegal behaviour or poor practice to the ASA and/or Swimline if there was sufficient evidence to support a reasonable suspicion. However, out of an awareness that false accusations are damaging, this decision related specifically to events that were directly witnessed rather than hearsay statements made by participants. Thankfully, I have had no reason to do so to date.
Attempts by participants to recruit the researcher into personal causes have occurred in previous research (see for example, Finkelhor & Williams, 1988; Brackenridge, 2001). In a swimming context, it could occur if a participant is unhappy with how the club is being run, with their child’s treatment or if they feel their child has been/ is being abused or bullied and asks for help in reporting instances of abuse/ poor practice. In order to maintain independence, researchers must resist the temptation to be drawn into individual cases (Brackenridge, 2001). I decided never to agree to report a grievance on behalf of a participant. Instead, as noted earlier, I decided to encourage participants to report any such grievances to the appropriate authorities by referring them to the reporting mechanisms in swimming provided in the debriefing packs. In practice, this issue never arose. This could have been because parents of athletes, who are most likely to be aggrieved by their child’s treatment, were not part of my sample.
Having outlined some of the main methodological, technical and ethical problems encountered in my research, the following section will discuss problems that are particularly relevant to ‘sensitive’ research topics, firstly problems of context.
No research takes place in a vacuum; rather, the context of the research has a bearing on it. Problems related to the context in which a study takes place are more severe when the research is sensitive in nature and have to be continually borne in mind by the researcher at all stages of the research process (Brewer, 1993).
A central contextual problem in relation to my study is that public and sports-sector interest in good practice and, more specifically, the related issues of child protection and child abuse, is high following several cases of child abuse within and outside of sport (Brackenridge, 2001). This leads to the potential problem that research of this nature could enter the public, not just the academic, domain. This could lead to problems of personal security such as those experienced by Brackenridge (2001); in the mid-1980s, before child protection in sport became a government objective, Brackenridge’s research into child sexual abuse in sport was regarded by many as drawing undue attention to a ‘minor’ problem and she suffered personal insults and ostracism from coaches and sports organizations and received crank telephone calls and hate mail.
In terms of my research, it was decided that the likelihood of my current study entering the public domain was low given that it would not be formerly published. The high public profile of good practice and, more specifically, the related issues of child protection and child abuse, however, raises another contextual problem: the sensitivity of those involved to questions concerning child protection. This contextual problem is closely linked to the earlier technical problem of how best to facilitate access to the research field. The purpose of one’s research must be explained in a meaningful and open way but also in a way that does not jeopardize the chances of having research access denied and that will enable effective data collection. The way in which the research is presented is crucial. As noted earlier, it was recognised that child protection forms an integral part of good practice. A study that focused solely on child protection issues fails to make this link and the sensitivity of adopting a focus solely on child protection issues could deter potential participants. Consequently, the research adopted a wider focus of good practice generally in youth swimming. Consequently, questions concerning child protection formed only a small part of the interview schedule and were asked towards the end so I had time to establish rapport with participants. As noted earlier, to further assuage participants’ fears, the interview schedule was distributed to participants in advance. This was particularly helpful in persuading the head coach at Central Seals, Andrew, to be interviewed and to agree I could approach the other coaches about participating. As noted in this research diary extract, when I first approached Andrew for an interview, he expressed concern about the focus of the interview:
“Andrew said he wants to know what the interview will be about. He explained that he feels all coaches are ‘under suspicion’ in the current climate, saying: ‘My bugbear is that the way it is now, we’re all seen as under suspicion when in fact the number of cases of abuse in swimming is tiny. I hate that and I don’t want my staff to feel like they’re being put under suspicion.’ ”
Two days later, I gave Andrew the interview questions to look over and, once he had scanned the pages, he immediately agreed to an interview. The other coaches also agreed and I later learned Andrew had told them the interview “was OK, not too hard” (fieldnotes, 12/7/07).
Another contextual problem arose from the researcher’s personal characteristics, which inevitably affect one’s research practice (Warren, 1988). In this instance, being a woman ‘insider’ (Seidman, 1998) conducting research in a male-dominated environment raised specific contextual challenges. Women researchers have reported being subject to sexual harassment and ‘gender joking’ (Cunnison, 1989) and paternalistic attitudes from male participants (see for example, Easterday, Papademas, Shorr & Valentine, 1977, cited in Brewer, 1993; Brewer, 1991; Flintoff; 1993). Conversely, there may be advantages to being a woman researcher; young female researchers often are regarded as “acceptable incompetents” (Loftland, 1971, cited in Brewer, 1993), or seen as non-threatening which can lead to increased infiltration into the field and easier rapport building (Easterday, Papademas, Shorr & Valentine, 1977, cited in Brewer, 1993).
At Central Seals although I experienced no harassment or ‘gender joking,’ I occasionally felt participants did not take my research seriously and considered me an “acceptable incompetent” (Loftland, 1971), particularly head coach Andrew. This was most noticeable during Andrew’s interview: He began by saying I had only 50 minutes to conduct the interview; although I had told him at least an hour would be necessary; he frequently answered his mobile phone during the interview; he guffawed at my questions and ended by suggesting I research a different topic entirely and should redraft my questions appropriately. I decided a smile was the best way to respond to these challenges in order to continue the research.
In contrast, my ‘insider’ status (Seidman, 1998) as a former elite swimmer had a positive impact on the research, placing me as non-threatening and facilitated rapport building with participants. This was evidenced on several occasions. For example, head coach Andrew said he is often approached about participating in research studies but previously has always refused because ““they [researchers] don’t know what it’s like. You’ve been a swimmer, you’ve been there so you know” (fieldnotes, 24/4/07). On other occasions, Andrew and assistant coach Steven both asked me to help time swimmers, and Andrew even left me to lead the session while he ran an errand. Through these events I felt a trusted and ‘credible’ member of swimming culture (Adelman, 1985), which I am certain contributed to Andrew and the other coaches agreeing to participate.
I feel being an ‘insider’ to swimming culture also facilitated open discussion among myself and the coaches. On several occasions, Andrew, Steven and John discussed the tactics they use to discipline swimmers, from throwing out swimmers who arrive late and proscribing push ups for misbehavior to Andrew’s admittance that:
“I have locked [the swimmers] in the land room and thrown things at them because they’ve been pissing about. I know it’s not good practice but we’re in a competitive environment here.”
The coaches, I feel, assumed my complicity in their training methods because, as a former elite swimmer, I would understand that their behavior was ‘necessary’ in order to produce elite athletes. It is unlikely they would have been so open to an outsider to swimming culture.
Personal safety problems
The physical safety of a researcher can be threatened during any research, for example, through the everyday activities required to conduct the research, such as travelling to unfamiliar locations and interviewing unfamiliar people (Brewer, 1993). Attempts were made to build aspects of researcher safety into the study design. For example, as noted earlier, participants were given a choice over their interview location. However, for personal safety reasons interviews were not conducted at participant’s homes. Some have suggested that allowing participants to be interviewed at home is more likely to put them at ease and result in an improved rapport with the researcher (Ruane, 2005). However, previous experience of interviewing participants at their homes created additional problems, such as frequent disruptions from ringing phones, visits from friends and even interruptions from pets! (see Lang, 2005). Instead, a quiet area in the leisure centre where the participant worked was suggested as an appropriate venue. To further ensure my safety without compromising participant confidentiality, a friend was informed of the location and time of each interview and I arranged to call to confirm my safety on leaving the venue. A sealed envelope with the interviewees’ personal details was left with this friend in case I failed to contact them within a stated period, although this was never needed.
As discussed above, my research presented challenges that are similar to those faced by all social researchers. Indeed, while the solutions presented here may not be applicable to all research, it is hoped they are good starting point for others working in this and other areas.
Adelman, C. (1985) ‘Who are you?’ Some problems of ethnographic culture shock. In: Burgess, R. G. (ed.) Field methods in the study of education. Lewes, Falmer.
ASA (2003) The swimmer pathway: Long-Term Athlete Development. Loughborogh, ASA.
ASA (2004)Wavepower: Child welfare in swimming – Procedures and guidelines. Loughborogh, ASA.
Bassuk, E., Dawson, R. & Huntington, N. (2006) Intimate partner violence in extremely poor women: Longitudinal patterns and risk markers. Journal of family violence. 21 (6), August, pp.387-399.
Bilsen, J. R., Vander Stichele, R. H., Mortier, F. & Deliens, L. (2004) Involvement of nurses in physician-assisted dying. Journal of advanced nursing. 47 (6), September, pp.583-591.
Brackenridge, C. H. (1997) Sexual harassment and abuse in sport. In: Clarke, G. & Humberstone, B. (eds.) Researching women and sport. London, MacMillan. pp.126-141.
Brackenridge, C. H. (2001) Spoilsports: Understanding and preventing sexual exploitation in sport. London, Routledge.
Brackenridge, C., Pitchford, A., Russell, K. & Nutt, G. (2007) Child welfare in football: An exploration of children’s welfare in the modern game. London, Routledge.
Brannen, J. (1988) The study of sensitive subjects. Sociological review. 36 (3) pp.522-563.
Brewer, J. D. (1991) Inside the RUC: Routine policing in a divided society. Oxford, Oxford University Press.
Brewer, J. D. (1993) Sensitivity as a problem in field research: A study of routine policing in Northern Ireland. In: Renzetti, C. M. & Lee, R. M. (eds.) Researching sensitive topics. London, Sage. pp.125-145.
Bringer, J. (2002) Swimming coaches’ perceptions and the development of role conflict and role ambiguity. Unpublished Ph.D. thesis, Leisure and Sport Research Unit, Cheltenham and Gloucester College of Higher Education
British Sociological Society (2002) Statement of ethical practice for the British Sociological Association. Available from: <http://www.britsoc.co.uk/equality/63.htm>. [Accessed: 7 December 2006].
Burgess, A. W., Brown, K., Bell, K., Ledray, L. E. & Poarch, J. C. (2005) Sexual abuse of older adults. American journal of nursing. 105 (10) October, pp.66-71.
Coghlan, J. & Webb, I.M. (1990) Sport and British politics since 1960. Falmer Press, London.
Cunnison, S. (1989) Gender joking in the staff room. In: Acker, S. (ed.) Teachers, gender and careers. London, Falmer. pp.151-167.
Dawes Knight, E., Smith, J. B., Dubowitz, H. P., Litrownik, A. J., Kotch, J. B., English, D. & Everson, M. D. (2006) Reporting participants in research studies to child protective services: Limited risk to attrition. Child maltreatment. 11 (3) August, pp.257-262.
DCMS/ Strategy Unit (2002) Game Plan: A strategy for delivering the government’s sport and physical activity objectives. London, Cabinet Office.
Dench, S., Iphofen, R. & Huws, U. (2004) An EU code of ethics for socio-economic research (RESPECT). Brighton, The Institute for Employment Studies.
Department of National Heritage (1995) Sport: Raising the game. Department of National Heritage, London)
DfES (1989) The Children Act 1989. London, HMSO.
Dingwall, R. G. (1980) Ethics and ethnography. Sociologial review. 28, pp.871-891.
Easterday, L., Papademas, D., Shorr, L. & Valentine, L. (1977) The making of a female researcher. Urban life. 6, pp.333-348. Cited in: Brewer, J. D. (1993) Sensitivity as a problem in field research: A study of routine policing in Northern Ireland. p.133.
Feenan, D. (2002) Researching paramilitary violence in Northern Ireland. International journal of social research methodology. 5 (2) pp.147-163.
Fetterman, D. M. (ed.) (1984) Ethnography in educational evaluation. Beverley Hills, CA., Sage.
Finch, J. (1993) ‘It’s great to have someone to talk to’: Ethics and politics of interviewing women. In: Hammersley, M. (ed.) Social research: Philosophy, politics and practice. London, Sage. pp.167-180. Cited in: Robertson, J. (2000) Ethical issues and researching sensitive topics: Mature women and ‘bulimia.’ Feminism and psychology. 10 (4), p.532.
Finkelhor, D. & Williams, L. M. (1988) Nursery crimes: Sexual abuse in day care. London, Sage.
Fischer, B., Rehm, J., Patra, J. & Cruz, M. F. (2006) Changes in illicit opioid use across Canada. Canadian Medical Association journal. 175 (11) pp.1385-1387.
Flintoff, A. (1993) One of the boys? An ethnographic study of gender relations, co-education and initial teacher education in physical education. Unpublished PhD thesis, Open University.
Green, M. & Houlihan, B. (2006) Elite sport development: Policy learning and political development. London, Routledge.
Hammersley, M. & Atkinson, P. (1995) Ethnography: Principles in practice (2nd ed.) London, Routledge.
Hearn, J. (1998) The violences of men: How men talk about and how agencies respond to men’s violence to women. London, Sage.
Henry, I.P. (2001) The politics of leisure policy (2nd ed.). Palgrave, Houndmills.
Houlihan, B. (1991) The government and politics of sport. London, Routledge.
Houlihan, B. (1997) Sport, policy and politics: A comparative analysis. Routledge, London, New York.
Inciardi, J. (1993) Some considerations on the methods, dangers and ethics of crack-house research. In: Inciardi, J. A., Lockwood, D. & Pottinger, A. E. (eds.) Women and crack cocaine. New York, Macmillan.
Kirkwood, J. (1999) ‘Investing ourselves’: Use of personal response in feminist methodology. In: deGroot, J. & Maynard, M. (eds.) Women’s studies in the 1990s: Doing things differently? Basingstoke, Macmillan.
Kotch, J. B. (2000) Ethical issues in longitudinal child maltreatment research. Journal of interpersonal violence. 15 (7) July, pp.696-709.
Lang, M. (2005) Coaching matters: Female swimming coaches’ and teachers’ perceptions of appropriate and inappropriate coach-athlete sexual behaviours and the implementation of child protection policies. Unpublished MA thesis, Leeds Metropolitan University.
Lee, R. M. (1993) Doing research on sensitive topics. London, Sage.
Lee, R. M. & Renzetti, C. M. (1993) The problems of researching sensitive topics: An overview and introduction. In: Renzetti, C. M. & Lee, R. M. (eds.) Researching sensitive topics. London, Sage. pp.3-13.
Loftland, J. & Loftland, L. (1995) Analysing social research settings: A guide to qualitative observation and analysis. Wadsworth, Belmont.
Loftland, J. (1971) Analysing social settings: A guide to qualitative observation. Belmont, CA., Wadsworth. Cited in: Brewer, J. D. (1993) Sensitivity as a problem in field research: A study of routine policing in Northern Ireland. London, Sage. p.133.
Lord, C. (2002) Sweetenham imposes tough new regime for GB swimmers. The Times online [Internet, 16/9/02]. Available from: Malkin, K., Johnston, L. & Brackenridge, C. (2000) A critical evaluation of training needs for child protection in UK sport. Managing leisure. 5, pp.151-160.
Martin, S. L., Ray, N., Sotres-Alvarez, D., Kupper, L. L., Moracco, K. E., Dickens, P. A., Scandlin, D. & Gizlice, Z. (2006) Physical and sexual assault of women with disabilities. Violence against women. 12 (9), September, pp.823-837.
Myers, J. & Barrett, B. (2002) In at the deep end: A new insight for all sports from analysis of child abuse within swimming. London, NSPCC.
National Speedo League North West (2005) National Speedo League North West handbook [Internet]. Available from: <www.speedoleague.org.uk> [Accessed 5 February 2007].
Olivier, S. & Olivier, A. (2001) Informed consent in sports science. Sports science. 5 (1) Jan/ April, pp.1-4.
Pershing, J. L. (2003) Why women don’t report sexual harassment: A case study of an elite military institution. Gender issues. 21 (4) pp.3-30.
Robertson, J. (2000) Ethical issues and researching sensitive topics: Mature women and ‘bulimia.’ Feminism and psychology. 10 (4), pp.531-537.
Robson, C. (2002) Real world research (2nd ed.). Oxford, Blackwell.
Ruane, J. M. (2005) Essentials of research methods: A guide to social science research. Oxford, Blackwell.
Sagarin, E. & Moneymaker, J. (1979) The dilemma of researcher immunity. In: Klockars, C. B. & O’Connor, F. W. (eds.) Deviance and decency: The ethics of research with human subjects. Beverly Hills, CA., Sage. pp.175-193. Cited in: Feenan, D. (2002) Researching paramilitary violence in Northern Ireland. International journal of social research methodology. 5 (2) p.159.
Seidman, I. (1998) Interviewing as qualitative research: A guide for researchers and teachers in education and social science (2nd ed.). New York, Teacher’s College Press.
Singer, E., Von Thurn, D. R., & Miller, E. R. (1995) Confidentiality assurances and response: A quantitative review of the experimental literature. Public opinion quarterly. 59 (1), Spring, pp.66-72.
Social Research Association (undated) Code of practice for the safety of social researchers [Internet]. Available from:<www.the-sra.org.uk/staying_safe.html> [Accessed: 8 December 2006].
Social Research Association (2003) Ethical guidelines [Internet]. Available from: <www.the-sra.org.uk/ethical> [Accessed: 8 December 2006].
Steinberg, A. M, Pynoos, R. S., Goenjian, A. K., Sossanabadi, H. & Sherr, L. (1999) Are researchers bound by child abuse reporting laws? Child abuse and neglect. 23 (8) August, pp.771-777.
Sullivan, T. A. & Cornfield, D. B. (1982) Gaining access to the large corporation for research. Paper presented at the Convention of the Society for the Study of Social Problems, San Fransisco. Cited in: Lee, R. M. (1993) Doing research on sensitive topics. London, Sage. p.130.
Vorontsov, A. R. (2002) Multi-year training of young athlete as potential modifier of growth and development: Analysis of some biological concepts. Paper presented at the FINA World Sport Medicine Congress. Cited in ASA (2003) The Swimmer Pathway. Loughborough, ASA. p.16.
Warren, C. A. B. (1988) Gender issues in field research. Newbury Park, CA., Sage.
Westmarland, L. (2001) Blowing the whistle on police violence: Gender, ethnography and ethics. British journal of criminology. 41, pp.523-535.
White, G. E. (2000) Sexual harassment during medical training: The perceptions of medical students at a university medical school in Australia. Medical education. 34 (12) December, pp.980-986.
Yorganci, I. (1993) Preliminary findings from a survey of gender relationships and sexual harassment in sport. In: Brackenridge, C. (ed.) Body matters: Leisure images and lifestyles. Brighton, Leisure Studies Association. pp.197-203.