Managing Ethical Issues in Sexual Violence Research Using a Pilot Study

Curationis 32 (1): 52-58 Conducting research in the area o f sexual violence has complex ethical and practical challenges for the researcher. Managing ethical issues in sexual violence is important and can be achieved through the use o f pilot studies. The primary purpose o f the pilot study was to identify and manage potential ethical and practical problems that could jeopardise the main study or violate the ethical and human rights of participants in the main study on women's journey o f recovery from sexual assault. The secondary purpose was to collect preliminary data in order to determine the human, financial and time resources needed for a planned study. The methods and processes used in conducting the pilot study in the study on women's journey o f recovery are discussed according to each o f the objectives o f the pilot study, methods used to achieve the objective, observations or findings made during the pilot study, and implications for the main study. This article aims to demonstrate how a pilot study was used to manage identified potential ethical and practical research issues during the recruitment o f participants and data collection for the research that was conducted by the first author to inves­ tigate women's journey o f recovery from sexual assault trauma within the first week following sexual assault.

The primary purpose o f the pilot study was to identify and manage potential ethical and practical problems that could jeopardise the main study or violate the ethical and human rights of participants in the main study on women's journey o f recovery from sexual assault. The secondary purpose was to collect preliminary data in order to determine the human, financial and time resources needed for a planned study. The methods and processes used in conducting the pilot study in the study on women's journey o f recovery are discussed according to each o f the objectives o f the pilot study, methods used to achieve the objective, observations or findings made during the pilot study, and implications for the main study.
This article aims to demonstrate how a pilot study was used to manage identified potential ethical and practical research issues during the recruitment o f participants and data collection for the research that was conducted by the first author to inves tigate women's journey o f recovery from sexual assault trauma within the first week following sexual assault.

Introduction
Conducting research in the area o f sexual violence has complex ethical and practical challenges for the researcher. These challenges are over and above the ordinary ethical considerations posed by ordinary research, such as issues o f confidentiality, informed con sent and respect for autonomy and pro tection o f the vulnerable participants (Ellsberg & Heise, 2005:34). This is due to the fact that a woman participating in sexual violence research is exposed to more risks than anyone participat ing in any other general research. These risks are related to the woman's vul nerable status as a result o f the recent or old sexual violation itself, and the sensitive and intrusive nature o f re search questions in sexual violence, which may sometimes force the partici pant to relive the sexual violence situa tion in order to be able to answer the research er's questions accurately. Thus research may pose as a potential secondary v ictim iz atio n threat (Sullivan & Cain, 2004:603;WHO, 2001: 6;Jewkes, Watts, Abrahams, Penn-Kekana, Garcia-Moreno, 2000:93). Sec ondary victimisation often results when the woman is made to wait for longer than she normally would have if she was not participating in the sexual vio lence research, or if she is coerced to participate in research as part of her treatment or her sexual assault trauma management regimen.
Managing ethical issues in sexual vio lence is important and can be achieved through the use of pilot studies. This article aims to demonstrate how a pilot study was used to manage identified potential ethical and practical research concerns during the recruitment of par ticipants for research that was con ducted to investigate women's journey of recovery from sexual assault trauma within the first week following sexual assault.

Background and rationale for conducting the pilot study
The need for conducting a pilot study prior to the research on women's jour ney of recovery was identified at pro posal development stage. Women's journey o f recovery was a longitudinal qualitative study which investigated wom en's journey o f recovery from sexual assault during the first six months following the incident of sexual assault. It was anticipated that the na ture o f the research questions rendered the research intrusive and sensitive. The participants were bound to be vul nerable especially at that stage follow ing the incident of sexual assault. Some o f the proposed recruitment and re search procedures were new and had never been used before within the South African context. It was therefore important to conduct a pilot study.
In consultation with the study refer ence groups (who are co-authors of this article), the decision to conduct a pilot study prior to the main study was made and supported by the research super visors and the university ethics com mittee. The purpose was to safeguard potential participants in the main study in terms of possible violation of their ethical and human rights during the recruitment phase.

Literature review
Review of existing literature on ethical and practical issues o f sexual violence research revealed various potential and actual ethical and practical challenges as well as specific recommendations for researchers in handling such potential ethical, legal and practical problems (Ellsberg & Heise, 2005:35;Cottingham & Jansen, 2005;no pages Sullivan & Cain, 2004:603).
Further review of the literature on the usefulness o f conducting pilot studies in qualitative research, and specifically sexual violence research, revealed vary ing views and problems cited regard ing the use of pilot studies, which are broadly outlined below.

Ethical, legal and practical issues in sexual violence research
No researcher would intentionally set out to revictimise his or her study par ticipants. However, if adequate precau tions are not taken to minimise the risks during the research process, both physical safety and psychological well being can be affected, which may re sult in some ethical misfortunes. Every sexual violence researcher has an ethi cal obligation to be vigilant and to pre vent violation of human rights during their research (Sullivan & Cain 2004: 603;WHO, 2001:7).
The common ethical principles for re search in human subjects, including respect for persons, protection of vul nerable persons, confidentiality, be neficence and distributive justice, are reported to be the cornerstone for re search in sexual violence. Disclosure and participants' safety during research are specific ethical issues for serious consideration in sexual violence re search (Ellsberg & Heise, 2005: 35). A researcher would have to provide a thorough description o f the research process, associated risks and benefits, and voluntary participation in order to safeguard the p artic ip a n ts' safety (Cottngham & Jansen, 2005 no page).
The issue of anonymity in sexual vio lence research is crucial. This refers to an additional protection o f one's iden tity, even from other participants in the same study, e.g. in focus groups. Care ful consideration of the data collection methods should ensure that both con fidentiality and anonymity are main tained at all times (Sullivan & Cain, 2004:603).

Ensuring protection of women's safety
The protection o f both participants and the research team or interviewers is of paramount importance in sexual vio lence research. There is always the threat o f potential retaliatory violence from the perpetrator, his friends or fam ily. Participation in sexual violence re search may also be misinterpreted by the community, friends or family mem bers as involvement in criminal inves tigative procedures (Sullivan & Cain, 2004: 603;WHO, 2001: 10). Incidents of physical abuse from controlling part ners are also common among women participants who participate in sexual violence research without their part ners' knowledge (Ellsberg & Heise, 2005:35 More precautions should be consid ered when making telephonic contact for the recruitment or other purposes during the study in order to ensure participant's safety. Such precautions include devising protocol with the par ticipant to be used when the perpetra tor or anyone other than the partici pant answers the telephone. Sullivan and Cain (2004:603) recommend pre tending that the researcher has dialed a wrong number, pretending that the research is about women's health in general, and asking for the participant by her specific name. The participant should ensure that the persons whose telephone numbers are given as alter natives are aware that they may be con tacted by the researcher at some point. D esigning and m asterin g such protocols can be done during pilot study.

Access to participants for recruitment purposes
Researchers in sexual violence research often face challenges when considera tion is made of recruitment sites to ac cess participants. The researcher would require measures to beat the bureau cratic system before access to police stations for recruitment purposes can be allowed (Cottingham & Jansen, 2005, no page). Sharing of any of the patients' or clients' information with the re searcher contravenes some of the core values for most o f the rape crises centers or health services for sexually abused women which include privacy, co n fid en tiality and protection o f women. The researcher should be able to convince the authorities that the re search will not jeopardise the safety of the participants before he or she is given access to the facility for recruit ment purposes (Sullivan & Cain, 2004: 603).

Compensation for research versus coercion
The issue o f com pensation for re search participation, the amount to be compensated and the form o f the com p en satio n has receiv ed atten tio n among sexual violence researchers be cause it has both ethical and practical implications. Sullivan and Cain (2004: 603) propose that women should be compensated adequately and in cash (not vouchers or cheques because of the difficulties in exchanging these for cash). However, they warn that the amount of compensation should not be so much that the participants are indi rectly 'coerced' to participate in the study because o f the money.
This information from literature review formed the basis for the decision to conduct the pilot study. While most of these recommendations from literature made sense, most had never been used by the researcher nor the research ref erence groups in the proposed study o f women's journey o f recovery from sexual assault. Hence it was found im perative for the first author to conduct a pilot study prior to the main study in order to put these into practice with out jeorpadising the participants' ethi cal and human rights.

Types and functions of pilot studies
Literature on conducting pilot studies revealed different types and functions of pilot studies, and opposing views and problems regarding the use of pi lot studies. The first type of pilot studies is often referred to as a preliminary exploratory study or feasibility study. This refers to the pilot study as the process of test ing the prospective research design. Its purpose is to identify logistical prob lems which might occur when using the proposed methods. This type o f pilot study is also used to collect prelimi nary data, determining the human, fi nancial and time resources needed for a planned study. It can also assist the researcher in developing the research question and research plan. This is more common in qualitative research where the researcher is allowed to em bark on research without a broad re search question. It is also useful in training the novice researcher in han dling as many elements of research as possible before embarking on the main study (Van Teijlingen & Hundley, 2001: 295;Strydom, 2001:180;Polit&Hungler, 1991:62;Lindquist, 1991:91).
The second type o f pilot study, often referred to as 'pretests', indicates the testing of one or more aspect o f the research, such as the data collection instrument or the data analysis proce dure. This type is common in quantita tive methodology and is used in the development and testing of research instruments and establishing whether the sampling frame and technique are effective (Van Teijlingen & Hundley, 2001:295;Strydom, 2001:180;Polit & Hungler, 1991:62,91;Lindquist, 1991: 91;Barbie, 1990:220;).
Pilot studies add value and credibility to the researcher's work. The pilot study results can be used to convince funding bodies and other stakeholders o f the researcher's or research team's competency and knowledge regarding the feasib ility o f the m ain study (Strydom, 2001: 178;Lindquist, 1991: 91).
Good pilot studies allow for evaluation o f success and failures in implementa tion o f proposed research methods, through identification and management of elements that could jeopardise par ticipants' safety or violate their ethical and human rights before exposing the major study participants to similar prob lems (Strydom, 2001: 178;Lindquist, 1991:91).
Both types of pilot studies were used by the researcher to pretest the recruit ment methods proposed for the main study; to safeguard the safety of the participants against violation o f their ethical and human rights and to add value to the credibility of the research er's work as well as to collect prelimi nary data, determining the human, fi nancial and time resources needed for a planned study.

Opposing views on use of pilot studies
The pilot study is a prerequisite step to a successful research process.

Problems associated with pilot studies
Conducting the pilot study does not guarantee that the main research study will be a success. It also does not solve all of the problems that the researcher may wish to rectify before embarking on a major study. The researcher may fail to make accurate predictions based on the pilot study data or findings.
Conducting the pilot study may also require a significant amount of human and financial resources. This may im pinge on the funding for the main study (VanTeiljingen & Hundley, 2001:295;Lindquist, 1991:91).
A common concern among research ers is the inclusion of data from the pi lot study or inclusion o f pilot study participants in the main study sample. Some state that this could lead to the 'contamination' of the findings of the main study. However, according to VanTeiljingen and Hundley (2001:295) such inclusion has little or no negative effect in qualitative studies.
The major problem associated with pi lot studies is the scarcity o f published pilot study reports and researchers' experiences in conducting pilot stud ies. The lack of published literature on researchers' experiences on pilot stud ies often lead to repetition of avoid able ethical and practical problems and further exposure of potential research participants to potential ethical and human rights violation risks (Van Teijlingen & Hundley, 2001:295).
The purpose o f publishing the first author's experience through this arti cle is to inform future researchers of the ethical and practical pitfalls and management o f the ethical issues in sexual violence.

Problem Statement
Based on literature review and some of the experiences from consultants and senior researchers, ethical and practi cal problems in relation to recruitment of participants for the proposed longi tudinal study to investigate women's journey of recovery from sexual assault trauma during the first six months fol lowing the incident of sexual assault were anticipated. The main ethical is sues o f concern were the identification o f appropriate and ethically sound method for locating, accessing and re cruiting women who have been sexu ally assaulted within the first week of the assault for participation in a sixmonth longitudinal study on the jour ney of recovery. While literature review provided some of the recommendations, none of these were reported to have been used within the South African context in sexual vio lence research, some had been used in dom estic violence in general only (Jew kes, Watts, A braham s, Penn-Kekana, Garcia-Moreno, 2000:93) and therefore their usefulness in sexual as sault could not be guaranteed fully. It was therefore imperative for the re searcher to conduct a pilot study to identify and manage the potential ethi cal and practical problems that could jeopardise the main study or violate the ethical and human rights of participants in the main study on women's journey of recovery from sexual assault.

Purpose of the pilot study
The primary purpose of the pilot study was to identify and manage potential ethical and practical problems that could jeopardise the main study or vio late the ethical and human rights of participants in the main study on wom en's journey o f recovery from sexual assault. The secondary purpose was to collect preliminary data in order to determine the human, financial and time resources needed for a planned study. While ethical considerations were care fully considered and highlighted in the research proposal accepted by the University Ethics Committee, it was also important to put some o f the proposed methods to test before embarking on the main study. Some o f the proposed methods had never been used before within the South African context.
The objectives o f the pilot study were as follows: • To identify the most appropri ate and ethically sound method for locating, accessing and re cruiting women who have been sexually assaulted within the first week of the assault for par ticipation in a six-month longi tudinal study on their journey o f recovery. • To identify all practical prob lems that could occur when us ing the proposed recruitment and data collection methods without compromising the par ticipants' safety. • To determine the feasibility of the main study in terms o f the financial and human resources.

The methods used
This was preliminary exploratory study or feasibility study that also used pre testing o f the proposed recruitment strategies within a longitudinal quali tative research study design. Partici pants were selected using the open sampling method which is the selec tion of participants 'on site'. In open sampling the researcher recruits partici pants who meet the inclusion criteria, but is not influenced by evidence of any research-related concepts (Strauss & Corbin, 1998:215). Open sampling is similar to convenience sampling, in which the researcher selects and re cruits participants who are readily available (Polit & Hungler, 1991:642). The problems identified during the sampling, recruitment and retention of participants is discussed further below. Data collection methods for the pilot included intensive interview sessions with participants, series of meetings with "gate keepers' and key informants as well as consultation with the re search supervisors and research refer ence group. The idea of research refer ence group was also new to the re searcher, but it was found to be very useful.
Preliminary data analysis, using quali tative data analysis methods was con ducted concurrently with data collec tion. It also involved extensive discus sions with research supervisor and re search reference group on all identified potential ethical and practical issues the main focus on the aims o f the pilot study.
The methods and processes used dur ing the pilot study are discussed ac cording to each o f the objectives o f the pilot study, methods used to achieve the objective, observations or findings made during the pilot study, and impli cations for the main study. Strategies for addressing the potential problems identified are also discussed.

Ethically sound methods for locating, accessing and recruiting participants
The first objective of the pilot study was to identify the most appropriate and ethically sound method for locat ing, accessing and recruiting women who have been sexually assaulted within the first week of the assault for participation in a six-month longitudi nal study on the journey o f recovery.
The methods used to achieve that ob jective included a series of meetings held with the 'gatekeepers' (doctors, nurse m anagers and professional nurses) to negotiate access to the hos pital identified as a recruitment site, and to convince them that the study would not be detrimental to potential partici pants as patients of the hospital.
From these meetings, a key informant was identified among the registered nurses in the Trauma Unit. Her role was to identify potential participants; those who were interested in participation in the study were told more about the study through completion o f a 'con sent to be contacted' form designed by the researcher. This was meant to be signed by the potential participant in agreeing to be contacted by the re searcher for research purposes. These forms were to be collected by the re searcher daily. She would use the in formation on the form to call the poten tial participants for recruitment to par ticipate in the study. The researcher then had to obtain informed consent and start data collection within the first week following sexual assault trauma.
The 'consent to be contacted' form is a new concept, which to our knowledge had never been used in South African research before. It was therefore impor tant to pilot test it for usefulness as a recruitm ent strategy before it was adopted in the main study. It was also important to identify any potential harm related to its use during the pilot phase, so that this could be eliminated in the main study.
Following the agreed upon protocol, the researcher collected the 'consent ing to be contacted' forms and con tacted all the women who had signed them.
The observations and findings from this objective were many and included the following: • The first observation was that when women were contacted telephonically, they could not recall anything about having given consent to be contacted by the researcher. • The women were either sur prised or angry that the re searcher was calling them for research purposes. The re searcher offered apologies and refrained from asking for volun tary participation if the poten tial participants were upset. • This reduced the number of potential participants for re cruitment. As a result, only three women were recruited over the pilot period of five months. • The key informant was always too busy and the unit too shortstaffed to participate fully in re cruitment of potential partici pants. Reflecting on the recruitment process; the above observations and findings resulted in the following possible ex planations: (i) there was a possibility of potential coercion of participants into signing the 'consent to be contacted' form by the researcher; or (ii) of get ting their information to the researcher without their permission. Any o f these possibilities would have meant that the women's privacy and voluntary partici pation rights were violated.
Consultation with the reference groups and research supervisors and further literature review resulted in the adop tion of the following strategy to resolve the problem: • The observations were dis cussed with the key informant. This revealed the difficulties ex perienced by the staff in partici pating in recruitment of partici pants on behalf of the re searcher, including staff short ages and separation o f clinical from research roles of the nurse. • A need to review the recruit ment protocol for the main study was identified and communi cated in writing to and sup ported by the university ethics committee.

Using proposed recruitment and data collection methods
The second objective of the pilot study was to identify ethical and practical problems that could occur when using the proposed recruitment and data col lection methods without compromising the participants' safety.
The methods used to achieve this ob jective were as follows: • Those who agreed to partici pate in the study were asked to select the most secure and safe venue for conducting the inter views. • All three pilot study partici pants selected the researcher's office for safety and conven ience. Interviews were held ac cording to the proposed meth ods, including completion of the demographic questionnaire at the beginning of the first inter view session. • Participants had to be called telephonically to remind them of or to confirm the next visit. It was found that two of the three participants had no reliable/ working telephones or mobile phones. One had to be con tacted through the telephone of her mother's employer in order to remind her about the inter view dates. • The second participant relo cated twice during the period of the study. Her mobile phone got lost before the second interview was held. She would send the researcher a 'please call m e' message from her friends' mo bile phones whenever she needed to communicate. How ever, there was always the po tential o f losing confidentiality when the researcher responded to these messages. The re searcher had to learn new and cautious ways of responding to the 'please call m e' messages. She also had to learn how to introduce herself on the phone without divulging the purpose of her call, while being able to identify whether the receiver of the call was a participant or not. A crucial observation discovered at this point was the need for strategies to retain participants in the study through effective communication. The researcher and each participant agreed on what the researcher could say to other people when she wanted to leave a message for the participant, or when she was responding to a 'please call me' message from an unfamiliar mobile phone number. This was based on the recom m endation o f designing tel ephone protocol with participants as pro p o sed by S ullivan and Cain (2004:603). Another interesting observation at this point was the women's need to partici pate in sexual violence research. This is often neglected or undermined in lit erature. The second participant was a very complicated case. Her baby was stolen and killed by the rapist on the same day that he raped her. On hearing about this during the second interview session (at the end of four weeks fol lowing the incident), the participant was asked if she would prefer to terminate participation in the study. She insisted on her need for participation in the re search, stating that "even if it does not help me now, as long as it will help others in the future Her insistence to participate in the study is supported by Sullivan and Cain (2004:603), who stated that if women believe in the pur pose of the study, they will go to greater lengths to participate. The challenge for the researcher was to balance the respect for the participant's need and insistence to engage in this type of re search without jeorpadising her human rights.
The third participant had an active mobile phone all the time. This made it easy to locate and communicate with her, even when she had moved to an other province.
The main implications regarding this objective were the difficulties and ex penses incurred in retaining the study participants for the period of 6 months. This assisted the researcher in antici pating and budgeting appropriately for the main study.
The greatest potential ethical issue identified during the pilot study was the potential of divulging information to other people if the phone number or mobile phone did not belong to the participant. The researcher had to learn to deal with it and hence the develop ment of the telephone protocol men tioned above.

Feasibility in terms of resources
The third objective of the pilot was to determine the feasibility o f the main study regarding the financial and hu man resources using the proposed study methods. The methods used included monitor ing and recording the time taken for each interview sessions and compar ing that with the proposed sessions of about one and a half hour to two hours per participant for budgetary purposes. The participants described the difficul ties they experienced and sacrifices they made in order to come to the inter view sessions, e.g. getting a baby sit ter to look after children at home and borrowing money from neighbours or family members in order to be able to get there for the interviews. This was not previously anticipated nor budg eted for in the proposed study.
In recognition of these identified prob lems and taking into consideration what the literature suggests about compen sation of participants (Sullivan & Cain, 2004:603), the proposed R20.00 per in terview for reimbursement of travel ex penses was increased to R50.00. In creasing the amount to R50.00 meant a significant change in the budget for the main study. However, the participants were very appreciative of the increased amount. This reassured the researcher that this amount would also be appre ciated by the main study participants. The crucial observation was the dis covery to the need to increase the R20.00 travel costs to R50.00. The R50.00 compensation was reported by participants to be a good incentive for participating in the study. However it also meant additional expenses to be budgeted for in the main study. Since compensation had to be in cash as per Sullivan and Cain (2004:603)'s recom mendation, the researcher had to en sure that she had enough cash in her office at all times. The researcher's tel ephone and mobile phone bills also in creased dramatically. This highlighted the need for appropriate budgeting for the main study.
In terms o f human resource aspects, all data collection and research project management was the responsibility of the researcher. While it could have been easier to train and appoint field work ers, it was felt that only the researcher could remain responsible for the datá collection and other research-specific roles in order to adhere to the interpre tive paradigm o f the study.
All three objectives of the pilot study yielded relevant information in manage ment of ethical and practical issues in sexual violence research. These discov eries made during the pilot study and their relevance for the main study were formally communicated to and sup-ported by the university research eth ics committee, the research supervisors and the research reference groups.

Discussions
Conducting the pilot study assisted in identification and management o f po tential ethical and practical issues where violation o f participants' ethical and human rights could have occurred in the main study. That resulted in the elimination of risky methods from the main study and thus safeguarding the participants' safety. These findings highlighted the need and importance for pilot studies especially among nov ice researchers and postgraduate stu dents who often resent conducting pi lot studies (Van Teijlingen & Hundley, 2001:295;Strydom, 2001:178;De Vaus, 1993:54). The main lesson from the findings of the pilot study was the importance of the identification of all problems with ethical or practical implications for the main study. While this did not guaran tee the success o f the main study, at least the researcher was confident of the elimination of risks that could have jeopardized not only the safety o f the main study participants, but also the credibility of the researcher's work. It could have also impinged on the fund ing for the main study (Van Teilj ingen & Hundley, 2001:295;Lindquist, 1991: 91). According to Strydom, (2001:178); and Lindquist, (1991:91) the pilot study results can be used to convince fund ing bodies and other stakeholders of the researcher's or research team 's competency and knowledge regarding the feasibility o f the main study.
The pilot study findings revealed the need for constant review o f decisions taken prior to the pilot study, and pre paredness to go back to the drawing board if necessary before embarking on a main study. Consultation with the ref erence group and research supervisors was also found to be useful for the evaluation o f success and failures in the implementation o f proposed re search methods, and thus to manage ethical and practical issues before em barking on the main study. This is also recommended by Strydom (2001: 178) and Lindquist (1991:91).

Conclusion
We believe that pilot studies are im portant in managing ethical issues in sexual violence research and in research in general. We also believe that report ing on pilot study findings and shar ing the challenges and difficulties ex perienced during a pilot study in a spe cific nursing discipline is an ethical re sponsibility and an obligation of all re searchers. We conclude by calling upon all scien tific nursing journals to give manu scripts that report pilot study findings and researchers' experiences of con ducting pilot studies the same consid eration given to other research-based manuscripts for publication purposes. In that way, ethical and practical risks can be avoided in similar studies.