Doyle2016

661230
research-article2016
QRJ0010.1177/1468794116661230Qualitative ResearchDoyle and Buckley
Article Q
Embracing qualitative research: R
a visual model for nuanced Qualitative Research
research ethics oversight 1–23

© The Author(s) 2016
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1468794116661230
qrj.sagepub.com
Elaine Doyle
Kemmy Business School, University of Limerick, Ireland
Patrick Buckley
Kemmy Business School, University of Limerick, Ireland
Abstract
The research ethics review systems within universities evolved from the positivist biomedical
model but have expanded to include all non-clinical research involving human subjects. However,
the application of the biomedical paradigm to qualitative research often creates significant problems.
This article highlights the fundamental differences between biomedical and humanities and social
science (HSS) research, illustrating that one size does not fit all when it comes to research ethics
review. Recognising the resource constraints faced by many higher level education institutions, we
develop a model which encompasses the traditional research ethics concepts without requiring
separate oversight procedures. After its original construction based on extent research ethics
literature, the model was evolved based on findings from qualitative interviews carried out with
expert members of research ethics committees. The model can be adapted to multiple contexts
through the application of different levels of tolerance in each domain. Our contribution is twofold:
(1) to synthesise from the literature an explicit rationale for differentiating research contexts
when it comes to research ethics oversight; and (2) to provide research ethics committees with a
workable visual model that can be used to aid decision making in diverse research domains.
Keywords
anonymity, confidentiality, informed consent, qualitative research, research contribution,
research ethics committee, research ethics oversight, research participants
Introduction
The research ethics review processes within academic institutions have evolved from
a paradigm developed to suit positivist biomedical research. These principles were
Corresponding author:
Elaine Doyle, Kemmy Business School, University of Limerick, Ireland.
Email: elaine.doyle@ul.ie
Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

2 Qualitative Research
designed to prevent the kind of abuses inflicted by the Tuskegee Syphilis study and by
Nazi Scientists during World War II. In recent decades the remit of these processes has
extended to all non-clinical research involving human subjects, including qualitative
humanities and social science (HSS) research. It is acknowledged in the literature from
the US, Canada, the UK, New Zealand and Australia, that the application of biomedi-
cal research ethics paradigms to qualitative research often creates significant problems
for qualitative, social science research which can be subjective, messy and non-linear
(see, for example, Boden et al., 2009; Connolly and Reid, 2007; Cortada, 1998; De
Wet, 2010; Guillemin and Gillam, 2004; Haggerty, 2004; Hammersley, 2009; Holland,
2007; Israel and Hay, 2006; Kleinman and Vallas, 2001; Lincoln and Tierney, 2004;
Monaghan et al., 2012; Nind et al., 2013; Smith and Rust, 2011; Van Den Hoonaard,
2006). To quote Hurdley, ‘despite multiple debates in qualitative research concerning
ethical regulation, the fences just keep closing in’ (2010: 524). Dingwall remarks, ‘the
system of pre-emptive ethical regulation developed in the biomedical sciences has
become a major threat to research in the humanities and social sciences’ (2008: 1),
while Israel and Hay claim that ‘social scientists … believe their work is being con-
strained and distorted by regulators of ethical practice who do not understand social
science research’. (2006: 1).
It is essential that research participants be treated respectfully (Smith and Rust, 2011).
However, the principles of anonymity, free and fully informed consent, confidentiality
and withdrawal are becoming sanctified by research ethics committees as absolutes,
regardless of the situation (Buckley, 2011; Murray et al., 2011). It must be acknowledged
that not all ethical considerations apply in the same way in all methodologies, particu-
larly because of the two main differences between biomedical and HSS research, namely;
the nature of the research interventions and the relationship between researcher and
researched. These key differentiating factors are what drive the need for biomedical and
HSS research to be treated differently when it comes to ethics oversight.
There have been many calls for ‘tools to support ethical practice in participatory and
other non-postpositivistic research’ (Kleinman and Vallas, 2001: 1060). Dingwall sug-
gests the health of social science research in UK universities rests on a tipping point and
urges the academy to subject research ethics review processes to serious scrutiny (2008).
Despite these calls, the extant literature largely falls into two categories: (1) descriptions
of ethical dilemmas encountered by researchers; and (2) critiques delineating the unsuit-
ability of biomedical research ethics procedures for HSS research.
Halse and Honey suggest mutiny against the current mode of review has a magnetic
attraction but report that ‘if there ever was a glorious golden age of unfettered freedom
for research, it is unlikely to be resurrected in a neoliberal world of legislative controls,
legal responsibilities, and institutional audit and accountability’ (2007: 349). Absenting
HSS research from any kind of ethics oversight is neither feasible nor desirable. Instead,
social science academics need to move beyond demonstrating the failings of current
practice and build a consensus around more suitable review processes.
Not only are legislative controls and institutional regulations growing, but many uni-
versities across the globe are also facing financial pressures (Altbach et al., 2009), result-
ing in increased resource constraints. In this context, an entirely separate system of
research ethics oversight of HSS research is not a financially realistic option.

Doyle and Buckley 3
This article answers the calls for tools to support ethical practice in qualitative
research. Drawing on our collective experience as members and chairs of social science
research ethics committees in a university in Ireland, we develop a research ethics model
encompassing the variables relevant to research involving human subjects. We show
how this model can be applied in a practical way to both biomedical and non-clinical
contexts through the application of different levels of tolerance in each domain. While a
precise, quantitative formulation of this increased tolerance is impossible to achieve, our
model seeks to clearly identify specific areas where principles taken from the bio-
medical review processes are too stringent for some HSS research. After its original
construction based on the research ethics literature, the model was further developed on
the basis of findings from a series of qualitative interviews carried out with members of
various types of research ethics committees spanning different disciplines across two
jurisdictions (the UK and Ireland). The perceived usefulness of the model was also
examined with interviewees. Our aim is to provide a visual tool/model which will assist
research ethics committees in evaluating qualitative research, without creating a dual
system of research ethics oversight for different research domains. We are aware of no
other visual model being currently used in research ethics oversight.
The remainder of this article is structured as follows. The section ‘Background con-
text’ examines the key differences between biomedical and HSS research and identifies
how current research ethics review processes are failing HSS research, providing a
rationale for contextual sensitivity in research ethics oversight. The ‘Initial model’ sec-
tion of the article sets out the key research ethics variables commonly discussed in the
research ethics literature and discusses how these are considered in the context of bio-
medical and HSS research. The integration of these variables into a draft visual model is
then described. The research method is outlined before findings from interviews with
research ethics committee members about the constitution and potential utility of the
model are set out in the ‘Findings’ section of the article. These findings informed the
further evolution of the model from its initial iteration and the finalised version of the
model is outlined at the end of this section. The ‘Exemplar’ section illustrates the appli-
cation of the model in varying contexts. We conclude in the final section of the article by
indicating where further research could be usefully directed.
Background context
In very broad terms the academic research ethics oversight process in Ireland, the UK,
Australia, New Zealand, Canada and the US requires employees, students or anyone
associated with a higher level educational institution, who is conducting research involv-
ing human subjects, to obtain ethical approval for their work. This involves the researcher
submitting a research ethics application to a committee within the relevant institution.
The committee considers the ethical probity of the application and gives approval for the
research to proceed (or not, as the case may be). These review processes evolved from a
paradigm developed to suit positivist biomedical research. Their initial aim was to pre-
vent medical abuses, with medical schools/institutions/departments being the first to
implement such processes. The positivist, biomedical paradigm strongly favours full
anonymity of research participants and free and fully informed written consent. Recent

decades have seen the remit of these processes extended to all non-clinical research
involving human subjects.
Differences between biomedical and HSS research

The primary differences between biomedical and HSS research lie in the nature of the
research interventions and in the relationship between the researcher and the participant
(Cortada, 1998). In biomedical research participants may risk death or serious injury.
HSS research generally presents no equivalent risk (Haggerty, 2004; Hoecht, 2011).
There is no doubt that HSS research participants may be emotionally upset, have their
dignity undermined, or experience very real invasions and violations of their privacy
(Lincoln and Tierney, 2004; Owen, 2006). There are examples of high profile social sci-
ence research studies which would now be considered unethical, such as Stanley
Milgram’s (1974) research on obedience or Philip Zimbardo’s (1973) prison experi-
ments. Nevertheless, the harms that HSS research is capable of are generally of consider-
ably lower magnitude than biomedical research (Haggerty, 2004). In the words of
Dingwall ‘whether we recall the Nazi medical experiments or the abuses of the 1950s
and 1960s biomedical science, no evidence of comparable harm has ever been adduced
for HSS research. We do not ever seem to have killed, maimed or caused permanent
mental disability to anyone in the name of science’ (2008: 3). Despite this, in many juris-
dictions all research is subjected to the same ethics review process without an adequate
assessment of the potential for harm to research participants (Lincoln and Tierney, 2004).
Secondly, the nature of biomedical research results in a large asymmetry of knowledge
between researcher and participant (Cortada, 1998). In most HSS research, however, par-
ticipants are not individuals lacking in agency who are being experimented on. They are
informants, domain experts giving their knowledge of a situation to the relatively unin-
formed researcher. Much HSS research, particularly participatory research methodologies
such as ethnographies, involves participants whose lives are affected by the issue at hand.
Participants are often involved in shaping research questions to address issues of concern
to them (Kleinman and Vallas, 2001). This kind of approach significantly reduces the
researcher/participant distinction typically evident in the biomedical sciences. It results in
a more complex relationship than that assumed by the research ethics review process
(Kleinman and Vallas, 2001; Stacey, 1988). The relationship between participant and
researcher in HSS research often evolves as the research progresses, making it difficult to
outline comprehensively before the research commences (as demanded by review pro-
cesses). It is, therefore, more appropriate in a HSS research context to undertake a much
more nuanced assessment of the relationships between all the stakeholders within the
research context rather than casting the researcher in a role which attributes all the respon-
sibility and power inherent in the research context to him/her, without any consideration
of the potential for human agency among participants (Kleinman and Vallas, 2001).
Dangers for HSS research

Due to the difference between the biomedical and HSS research contexts, current
research ethics processes are accused of failing HSS research in several ways. First,

Doyle and Buckley 5
review committees in the US, Canada, the UK, Ireland, New Zealand and Australia are
increasingly cautious about granting approval for research on what are perceived as sen-
sitive topics or involving vulnerable populations, resulting in potentially valuable
research being stymied. Particular topics, for example research on school children, have
become so difficult to secure approval for that they are waning (Van Den Hoonaard,
2006). The practice of discouraging edgework serves to preserve the status quo, causing
stagnation within disciplines (Johnson, 2008: 229). Review committees in some jurisdic-
tions stand accused of ‘limiting, controlling and even stopping particular research from
being done and particular areas of enquiry being explored in the interests of avoiding
controversy and litigation with the inevitable consequences for the exercise of academic
freedom’ (Davenport, 2005: 207). It is argued that the uncritical application of restric-
tions imposed on some HSS research, causing particular topics or participant categories
to be inaccessible, is unethical and may result in a huge cost to society (Cortada, 1998;
Lewis, 2008; Murray et al., 2011). In the words of Dingwall, ‘When we give up doing
participant observation with vulnerable or socially marginal groups because of the regu-
latory obstacles, then a society becomes less well-informed about the condition of who it
excludes and more susceptible to their explosions of discontent’ (2008: 10).
Second, the inclusion of qualitative research in research ethics review has resulted in
a substantial drop in immersive methodologies in some countries, for example Canada
and the US, in favour of the ‘safer’ qualitative methodology of the interview (Johnson,
2008; Van Den Hoonaard, 2006). Qualitative research is a valuable way of understanding
truth through the process of studying something holistically. It contributes theories, mod-
els and descriptions of human experiences and perceptions within particular contexts.
(Borland, 2001). If immersive qualitative methods are declining due to restrictions
imposed by ethics oversight, this represents a significant danger for society in terms of
our understanding of certain cultures and communities.
Third, successfully navigating the application process by learning to complete appli-
cation forms correctly, does not mean that the underlying research will be ethical and
may actually distract HSS researchers from the real issues that emerge as the research
progresses but after approval has been granted (Davenport, 2005; Halse and Honey,
2007; Hammersley, 2009; Hammersley, 2014; Smith and Rust, 2011; Stark, 2012).
Qualitative researchers must conform with research ethics processes, not because they
accept the moral authority of the processes but because their reputations and careers may
be damaged otherwise, making the authority of the process coercive rather than moral
(Haggerty, 2004).
The fundamental differences between biomedical and HSS research, and the damage
being done to HSS research as a result of the imposition of the biomedical model of
review, requires the introduction of a more nuanced system involving higher levels of
tolerance for some HSS research proposals. We now move to illustrate how the required
flexibility can be introduced into research ethics oversight by outlining the original draft
of our visual model based on the research ethics literature (discussed below). After the
model was originally constructed, a series of semi-structured interviews were carried out
in order to inform its further development. Findings from these interviews and their
impact on the development of the model are discussed in the ‘Research method and find-
ings’ section of the article where the final iteration of the model is also outlined.

Initial model
The principles governing ethical research involving human subjects in the US, Canada, New
Zealand, Australia, the UK and Ireland, are broadly derived from the Nuremberg Code and
the Helsinki Declaration. These are autonomy, nonmaleficence, beneficence, and justice
(Halse and Honey, 2007). These principles are typically operationalized in the research eth-
ics oversight process using a number of core concepts or practices. Different scholars priori-
tise and categorise these concepts differently (see, for example, De Wet, 2010; Tilley and
Gormley, 2007; Vainio, 2013; Hammersley, 2014). We originally based our model on the
four key research ethics concepts of: 1) informed consent; 2) anonymity; 3) risk of potential
harm; and 4) reciprocity. These were chosen on the basis that they are the variables cited
most often in the literature on research ethics review. We add the experience of the researcher
to our draft model as a moderating influence. Typically, research ethics oversight commit-
tees in the jurisdictions mentioned above address the issues raised by these concepts when
considering applications. Difficulties arise for HSS applications when committees impose
the biomedical concepts by default and fail to contextualise them within their relevant disci-
pline or methodology. Each of these concepts is discussed below in the context of both
biomedical and HSS research to illustrate the importance of contextualisation.
Informed consent
Informed consent is one of the cornerstones of research ethics. It requires the researcher
to explain comprehensively the objectives and implications of the research so potential
participants can make fully informed decisions about their involvement (Kitchener and
Kitchener, 2009). Because of the potential for physical harm and the information asym-
metry between researchers and participants, written informed consent is essential for
biomedical research.
Fully informed, written consent is contentious in HSS research for several reasons.
First, in particular contexts (such as ethnography, participant observation and explora-
tory interviews) consent forms may be inappropriately formal and/or unworkable (see
for example, Wiles et al., 2006). Written consent may be impossible in cultures where
literacy, language, culture and politics are inhibitors (Whittaker, 2005). Second, securing
informed consent may prevent or impede research that is otherwise innocuous, such as
content analysis of anonymous on-line discussions or an analysis of video footage of
people interacting in public places (Haggerty, 2004). Thirdly, informed consent provi-
sions make it virtually impossible to conduct research that involves even minimal decep-
tion, such as not telling participants about the true aims of the research when doing so
would influence behaviour (Haggerty, 2004).
Securing informed, voluntary consent from participants is always optimal. However,
there are HSS research contexts in which a lesser standard than fully informed, volun-
tary, written consent may be tolerable. If we represent this on a continuum, it might look
like Figure 1:
Research ethics committees considering a biomedical application would typically
insist that fully informed, voluntary, written consent be secured from all participants (far
left of the continuum). However, a committee considering HSS applications may allow

Doyle and Buckley 7
Figure 1. Informed Consent Continuum.
for a less stringent stance depending on the circumstances, taking a position towards the
centre or right of the continuum.
Anonymity
Another core concept underlying ethical research is anonymity, regarded as the means
through which the privacy and confidentiality of participants is maintained. While ano-
nymity is feasible in experimental, quantitative studies, it is contested and criticised in
HSS contexts (Vainio, 2013). It may be wholly inappropriate to insist on anonymity in
interpretive, qualitative research (Moore, 2012; Tilley and Gormley, 2007). Interpretive
methodologies frequently demand familiarity between researcher and participant in
order to be effective. Participants may want to speak on the record and have their version
of events documented. This may be their motive for co-operation (Cortada, 1998; Wiles
et al., 2006). Commenting on the difficulties of anonymity where participants are a small
group, Van Den Hoonaard observes, ‘research ethics codes insist on anonymity while the
practice of research makes it virtually impossible’ (2003: 147).
Similar to the case of informed consent, scenarios can be envisaged where anonymi-
sation of data as a means of maintaining confidentiality is inappropriate in a HSS con-
text. Committees considering both biomedical and HSS applications might tolerate a
lesser level of anonymity in the context of HSS research – or none where participants
want to be identified (right side of Figure 2) than in the case of biomedical applications
(left side of Figure 2).
Risk of potential harm

While the issue of risk or harm within HSS research contexts is not inconsequential, dif-
ferent kinds of risks and benefits are associated with different kinds of research.
Biomedical research may carry the risk of death or serious injury while social science
research presents no equivalent risk (Haggerty, 2004). However, there is a suggestion
that ‘zero risk’ has now become the new gold standard for research ethics review boards,
coupled with a recognition that such a norm is unrealistic and impossible to obtain given
the interpersonal nature of HSS research (Kennedy, 2005). It is observed that the ‘range
of potential research related harms envisaged by REBs [Research Ethics Boards] at times
seems to be limited only by the imagination of different reviewers. Any change in a
research participant’s condition or disruption of their routine can be conceived of as a
potential harm’ (Haggerty, 2004: 400–403).

Figure 2. Anonymity Continuum.
Figure 3. Risk of Harm Continuum.
HSS research may risk participants being upset, embarrassed or humiliated, feeling
that their dignity has been undermined, or that they have experienced invasions and vio-
lations of their privacy (Lincoln and Tierney, 2004). However, it is argued that these are
risks that participants can understand and assess for themselves (Cortada, 1998). The
methodologies employed in HSS research are not alien to ordinary people. Observing
others, asking them questions, engaging in group discussion, looking at pictures, reading
documents – these are all activities that ordinary people perform daily and pose no
greater risk than when encountered in everyday life (Holland, 2007). Research partici-
pants can use this knowledge to manage their own risk (Cortada, 1998). In general, par-
ticipants can withdraw if they feel uncomfortable or distressed. Furthermore, it is argued
that the researcher typically has a more realistic view of the likelihood of harm actually
occurring to the participant than the research ethics committee, which may focus on the
worst case scenario (Connolly and Reid, 2007).
We would expect research ethics committees would have limited tolerance for risk in
biomedical applications, given the nature of the relevant harms (far left of the continuum
in Figure 3), unless these were balanced against very large potential research contribu-
tions. However, some level of risk of harm might be tolerated in the case of HSS research
applications (middle point of the Figure 3 continuum).
Reciprocity
Academic research should serve society and advance knowledge (Christians, 2005;
Lincoln and Tierney, 2004). Gregory describes research as ‘driven by the desire to
advance knowledge and understanding [on our own part, and, if we are lucky, on behalf
of others]’ (2003: 14). In collaborative research projects research topics may reflect the
interests of participants and may be of direct or indirect benefit to participants and/or
their communities. Reciprocity occurs when the research gives back to participants and/
or to their communities (Lincoln and Tierney, 2004). Research participants may gain

Doyle and Buckley 9
Figure 4. Reciprocity Continuum.
directly through increased self-understanding, satisfaction in assisting, learning about

research methods, as well as the more general benefits to theory and knowledge of human
behaviour. Some research projects contribute very little to knowledge or to the partici-
pants and/or their communities, but may have the benefit of educating the researcher
about research methods (for example, undergraduate research projects or a thesis under-
taken as part of a taught master’s degree).
All research studies should make some contribution, regardless of the research context.
The contribution may be direct or indirect and may benefit participants, their communities,
or society in general in terms of knowledge expansion. While predominantly benefiting the
student, the development of research skills also contributes to the creation of an informed
and literate citizenry which is the bedrock of modern society. Research ethics committees
considering both biomedical and HSS applications should expect the research to be posi-
tioned from the left hand side to the mid-point of the continuum set out in Figure 4.
Researcher
Our final variable is the researcher. In the era preceding formal research ethics review a
researcher’s training and experience were presumed sufficient to ensure ethical probity.
Academic researchers were assumed to possess integrity and trustworthiness (Hoecht,
2011). The pursuit of knowledge and the sacredness of academic freedom were dominant
paradigms and justified research with minimal monitoring or interference (Whittaker,
2005). In the modern world, while the rights/entitlements of research participants have
gained power and legitimacy, the rights/entitlements of the researcher have arguably
diminished (Whittaker, 2005). There seems to be a lack of trust in the integrity of
researchers (Hoecht, 2011) and growing distrust between researchers and research ethics
committees (Haggerty, 2004).
Research ethics committees are perceived to focus on the worst possible outcomes,
assuming researchers to be irresponsible and even dangerous (Davenport, 2005;
Haggerty, 2004; Stark, 2012). Not only may this perception cause de-motivation and de-
professionalization (Davenport, 2005), it also suggests a significant lack of respect for,
and trust in, hardworking, well-meaning researchers (Halse and Honey, 2007). In gen-
eral, there is no evidence to support the notion that social scientists routinely engage in
unethical research behaviour (Davenport, 2005; Lincoln and Tierney, 2004).
That said, researchers may be blinded by proximity to their work and unable to view
their behaviour with sufficient objectivity to identify potential for harm (Davenport,
2005). Hedgecoe observes ‘researchers are not the best people to decide on the risks and
benefits of their work, having, as they do, a tendency to underestimate the former and

Figure 5. The Basic Model.
overestimate the latter’ (2008: 880). It is entirely appropriate for social scientists to be
included in research ethics oversight, regardless of differences between biomedical and
HSS research. However, we suggest oversight processes should consider the quality and
skill of the researcher(s).
Integrating the principles into a workable model

Our model is fabricated by taking each of the four continua outlined above and scaffold-
ing them into a single structure, as shown in Figure 5:
In each case, the continuum moves from the central point of the model (representing
strict application of each principle), towards the extremity, where there is more tolerance
in the application of each principle. All the key research ethics concepts relevant to
human subject research are included, however, the model allows for a degree of toler-
ance to be applied depending on the context. As an example, acceptable biomedical
research ethics applications would map onto the model towards the centre. Fully
informed, voluntary written consent would be sought from participants, their data would
be made anonymous, the risk of harm may stray into medium to high but the benefits of
the research should be high.
We could similarly map a HSS application onto the same model but may expect that
some HSS applications would be represented by a larger circle (or a bubbling out on the
non-risk dimensions), on the basis that a greater level of tolerance is allowed in the appli-
cation of some of the research ethics concepts. For example, in the case of research

Doyle and Buckley 11
involving deception where no consent can be secured, there should be very limited risk
to participants, full anonymisation of data, and the contribution of the research should be
significant in order to counteract the lack of informed consent. Research proposals that
map to the extremes of all four continua are extremely unlikely to be acceptable. In this
way, this model can be used to initially map an application. Committees can then use
their expertise and discretion to adjudicate on a particular case. This model implicitly
includes the concept of contextual sensitivity. In accommodating this concept, it pro-
vides a more nuanced framework that can aid evaluation of both biomedical and HSS
research proposals.
We include the experience and skill of the researcher as a moderating influence within
our model. For an experienced researcher or team, a greater level of tolerance may be
acceptable in the application of the four research ethics principles, while a stricter appli-
cation would be appropriate when the researcher lacks experience. In terms of our visual
representation, the effect of this moderating influence would be to allow projects with
larger circles (when mapped onto the model) to proceed if they are supervised by more
experienced researchers. (See the ‘Exemplars’ section below for illustrations of how the
model works).
Research method
In an evolutionary process, the model we describe above was initially designed and then
shown to a number of members and chairs of research ethics committees in the course of
qualitative interviews in order to investigate the relevance of the core concepts included,
as well as the perceived usefulness of this kind of visual model. The interviews informed
the maturation of the model into its final iteration. We describe this evolution below.
Using purposeful and snowball sampling, 10 potential interviewees from 7 universities
across 2 jurisdictions were identified and contacted by email, given information on the
broad nature of the research and asked to contribute. All 10 agreed. Six were women and
four were men. Interviewee number, jurisdiction, committee type and position are set out
in Table 1.
The interviews were semi-structured with open-ended questions and probes used to
elicit each interviewee’s views. Nine of the interviewees consented to their interview
being recorded. The narrative data (audio files) were converted into verbatim transcripts
before coding and analysis. Data were coded using typical template analysis procedures
(Cassell and Symon, 2004: 256–270). Interviews commenced with questions about the
structure of the research ethics oversight procedures within the relevant institution and
the role the interviewee played within the process, before questions about the research
ethics concepts/principles typically used to make a judgement on research ethics applica-
tions were posed. After these discussions, the interviewee was shown the draft model
outlined above and given a brief explanation as how it might operate. Interviewees were
asked whether the variables included in the model were appropriate, whether there was
anything missing from the model and whether they considered that a visual model would
be a useful tool in assessing research ethics applications. Interviewees’ views on the
relevance of the research ethics concepts outlined above and on the perceived usefulness
of the model are outlined in the following sections.

Table 1. Interviewee profiles.
Interviewee Jurisdiction University Type of Committee Position

Number
1 UK University 1 Cross Faculty Committee covering Founder and
Social Science, Business and Arts Vice Chair
2 UK University 2 Cross Faculty Committee Member
3 Ireland University 1 Human Sciences Ethics Committee Member of
(Faculty) & Life Sciences both
Undergraduate Committee
4 Ireland University 2 University Level Committee looking Member
at all university wide applications (no
UG research)
5 Ireland University 2 University Level Committee Member
6 Ireland University 3 Faculty committee Chair
concerned with Governance
concerned with Governance
Findings
Informed consent
Interviewees unanimously agreed informed consent should be represented in the model.
Indeed, it was the concept interviewees spent most time speaking about and all agreed
was critical in the context of research ethics oversight. They were also united in acknowl-
edging that tolerances on the continuum should be allowed on a case-by-case basis. ‘…
in some foreign countries the idea of signing a consent form is … seen as something like
agreeing to have your name given to the secret police’ (Interviewee 1).
Some interviewees considered informed consent problematic for two reasons. First, it
is sometimes difficult to inform participants about the nature of a research project before
they agree to be involved due to their limited understanding of research. ‘… you can’t
inform them about what might happen … because the interpretation of the data might be
beyond their knowledge’ (Interviewee 2).
Secondly, while more a function of the research ethics process, interviewees observed
that consent is an ongoing rather than an episodic concept. ‘it doesn’t happen at a particu-
lar point’ (Interviewee 8). It was observed that committees need to understand that
informed consent can be complex in HSS research ‘[it] requires knowledge around the
issues around informed consent and, when committees look at it, they need to apply that
nuance to it …’ (Interviewee 2).
These findings suggest that the inclusion of informed consent, along the lines of the
continuum described above, is appropriate for inclusion in our visual model, subject to the
proviso that committees appreciate the complexity of the concept in their discussions.

Anonymity/confidentiality
Anonymity was identified by interviewees as an important research ethics concept in
order to ensure confidentiality. There was general agreement that relaxed tolerances on
the continuum might be appropriate on a case-by-case basis. ‘… in social science
research … anonymity can be kind of problematic … I think committees need a much
more nuanced understanding of the complexities of all of that’ (Interviewee 2). ‘So we
have had proposals in where they have talked to named politicians, and provided the
named politicians are happy actually to be associated with that research, I don’t see a
problem’ (Interviewee 5)
Several interviewees commented on the difficulties of achieving anonymity. ‘But
even this whole area of anonymity can be quite tricky … so for example you’re in a busi-
ness and you’re asking about a supply, there may only be one person in the company
who’s involved in supply. So it may not be possible to actually guarantee anonymity’
(Interviewee 5). It was also acknowledged that any guarantees in relation to confidential-
ity could only be upheld within the limits of the law. ‘… we need to provide up front we
can only guarantee confidentiality within the limits of the law’ (Interviewee 5).
In the original conception of the model, we labelled the relevant continuum
‘Anonymity’ rather than ‘Confidentiality’, on the basis that the former concept is more
commonly discussed in the literature than the latter. Furthermore, interviewees tended to
speak more about anonymity. However, based on feedback from some of our interview-
ees and reviewers, we amended the model based on the rationale that anonymity is just
one of the strategies used to maintain confidentiality. We use the term confidentiality to
refer to strategies used to protect participants’ privacy, while anonymity is a particular
strategy, closely associated with the positivist, biomedical model.
Risk of potential harm

All interviewees considered risk of harm to be one of the most important variables to be
assessed in the research ethics process. ‘So if there’s no harm it doesn’t really matter
whether people are giving consent or whether it’s anonymous or not or whether there’s a
whole load of benefit’ (Interviewee 5). Several interviewees highlighted that a range of
harms need to be considered ‘so there are possibilities of harm at various levels … risk
to the participant, the researcher and others’ (Interviewee 1).
Concerns raised included having to identify the potential risk of harm at a particular
point in time for research ethics oversight processes, and the difficulty in measuring
harm in a social sciences context. ‘It’s much easier to determine risk in relation to physi-
cal harm and it’s much more difficult to quantify that in terms of nonphysical, like for
example psychological and something like shame and embarrassment.’ (Interviewee 8).
‘If you’re planning to test some sort of invasive technique with a person, there’s a high
level of risk in and of itself with that, versus you know, giving somebody an anonymous
survey, so you’ve got to take that into account’ (Interviewee 4).
All interviewees acknowledged the need for potential harms to be balanced against the
perceived benefits of carrying out the research. ‘… there’s always a balance that has to be
struck between the harms that are caused and the benefits that will accrue’ (Interviewee 5)

Overall, interviewees considered that the continuum set out in the model for ‘Risk of
Harm’ was appropriate for inclusion.
Reciprocity/contribution
In the original model, ‘Reciprocity’ was included as one of the key components.
Reciprocity is deemed to occur when the research gives back to participants and/or to
their communities (Lincoln and Tierney, 2004). However, the idea that all research
should have reciprocal benefits was contested by many of the interviewees. ‘I’m not sure
that reciprocity is actually the right term, but I suppose the contribution of the research,
whether that’s to the participant, to their community, to literature in general or knowl-
edge in general, or potentially even for the researcher’ (Interviewee 1). ‘I actually have
difficulty with you having reciprocity…maybe have it all as contribution because reci-
procity to me means that you need to provide contribution to both parties and that’s not
always the case … this one I have a certain dilemma with’ (Interviewee 7). The fact that
some academic research is carried out purely as a means of educating students in research
methods was one reason why interviewees suggested that reciprocity was not always
achieved.
Most interviewees considered the broader contribution of the research to be worthy of
consideration however, though they varied on the weight they attached to this variable.
‘Not everything we do … can have a large contribution’ (Interviewee 7). They also recog-
nised that contribution can often be measured only indirectly. ‘… part of our concern is
that we get biomedical projects coming through that are clearly going to make an insig-
nificant contribution … but if it’s part of training a master’s student then that changes
things. Now there is a contribution to the future pool of researchers’ (Interviewee 4).
One interviewee said that contribution was the least important of the variables his/her
relevant committee would consider. ‘ we wouldn’t be great on the amount of contribution
it makes, now, we would put very little weight on that … overall … the risk of harm
would be the thing that would most likely to go in and out and that would be reflected in
what ethics committee our projects go to’ (Interviewee 6)
As the result of interviewee feedback we altered the label of the relevant continuum
to ‘Contribution’ rather than ‘Reciprocity’ in order to more accurately reflect the con-
cerns of the committees our interviewees served on.
Researcher
Most interviewees agreed that oversight of all categories of researcher was necessary in
a research ethics context. ‘I think it’s quite naïve to assume, as we once did with doctors,
that academics always act with integrity … academics pursue their own self-interest in
terms of research and I think it’s extremely naïve … to assume that academics have
integrity in terms of their research subjects’ (Interviewee 8).
Interviewees reported that the experience of the researcher was an important consid-
eration when assessing applications, though typically the categorisation was between
students at different levels (undergraduate, masters or PhD) and qualified faculty. ‘It
does come into play … the discussion might say ok this is a very sensitive topic or a very

difficult topic for people to talk about, and so because of that having an undergrad doing
it versus a PhD student versus an experienced researcher would come into play … we
expect people to have demonstrated competency and take into account their experience
with working with various procedures’ (Interviewee 4).
Others suggested committees need to use the skills of the researcher as means of
nuancing the discussions of the committee. ‘… they have to stop looking at tick lists and
start looking at the skills of the person doing it and then they have to … think okay that
might develop into a horrible scenario. If it did, is that person reasonably well equipped
so that they could have an idea to know what to do?’ (Interviewee 2).
Overall, interviewees agreed that the experience level and skill set of the researcher
was an appropriate influencing variable in the context of research ethics committee dis-
cussion. On this basis, it was retained in the model as such.
Usefulness of a visual model

Without exception, all interviewees considered the idea of using a visual model as part
of the research ethics oversight process had merit and would be useful in the context of
research ethics committees’ discussions. ‘I think for social and humanities this is a very
good model… university research ethics committees are only beginning to find their feet,
so any template is better than no template’ (Interviewee 7). ‘I think it’s a good start … I
think if you got the form and you sort of drew your picture of what it was, then you know,
you shouldn’t get people saying, this is fine with insignificant contribution and causing
substantial risk and things like that…but it also means it’s easy to identify whether there’s
one single issue or multiple issues, it gives you a feel how to manage time in the commit-
tee, even … It’s a classification aid’ (Interviewee 1).
Some interviewees suggested including further variables into the model. Conflict of
interest was considered by one interviewee as critical; however, the authors considered
that this issue would be addressed under the informed consent continuum. The other two
issues raised were, first, the episodic nature of the research ethics oversight process. ‘I
think it’s useful but I’ve got one caveat…I mean the trouble with a lot of this stuff is it
asks to stand in advance of yourself before you do the research and social research is
inherently more kind of where you just sort of follow your nose’ (Interviewee 2). Second,
educating committee members to appreciate the nuances (and messiness) of social sci-
ence research. ‘The committee needs educating … this happened a lot with the kind of
research ethics committee for the NHS in Britain where they just didn’t understand these
issues and they just wanted to have the paperwork in place because they were covering
their backs so it’s about informed discussion actually. We’re big on informed consent and
not very good on informed discussion’ (Interviewee 2). These issues, while extremely
important in the context of research ethics oversight and its operation, are beyond the
scope of the high level model we were creating in this article but are certainly worthy of
further research.
Several of our interviewees had suggestions for further uses for the model. One of
these involved having the model replicated on the research ethics web page so that appli-
cants know in advance the main issues of concern in a visual manner. ‘When people are
approaching their applications if they felt there was an issue on one of the arms, to make

Figure 6. The Evolved Model.
sure that they address that in their application. So it could be used as a sort of a guide to
ensure that, people themselves can identify where the potential gap is in their application
… so it’d be a kind of pre-emptive tool as well, maybe’ (Interviewee 10). The second
idea involved using the model to audit where applications tend to fall down. ‘I mean we
essentially try and do an audit every year at the end of the academic year just to have a
look at … our numbers, what are our rates of either approvals versus minor amendments
versus resubmissions, and I think something like this would be useful in terms of embed-
ding that within that audit….And then you could say well, actually … the vast majority
of these were either this, this, this or this, and we made our decisions based on that.’
(Interviewee 9).
On the basis of findings from the semi-structured interviews, the draft model which
was designed on the basis of the research ethics literature was evolved into the finalised
version set out below (Figure 6).
We now illustrate the use of the model by considering a number of exemplar cases and
mapping them to the model.
Exemplars
Exemplar 1
Exemplar 1 involves a researcher planning to conduct in-depth interviews with prisoners
convicted of war crimes in a recent conflict. The aim is to capture their justifications of

Figure 7. Mapping of Exemplar 1.
actions taken. Participants will be drawn from a small population, confined to a specific
conflict. The crimes are a matter of public record, and all details were covered exten-
sively in the media at the time of the relevant legal trials. These considerations, allied to
the format of qualitative data reporting, make it difficult to guarantee the complete con-
fidentiality of participants’ identities.
In terms of the other research ethics principles; participants should be able to give
fully informed voluntary written consent. Risks to participants should be low, as they
have already been convicted and are serving prison time. This research could offer great
insights into the decision making processes of individuals operating in extreme circum-
stances, and potentially into how such behaviours could be mediated or deterred.
Figure 7 demonstrates how this project might be mapped on our model.
The mapping of this example would provide the committee with a visual representa-
tion of the case from which they can commence the debate on the application.
Exemplar 2
In Exemplar 2, we consider a netnography (see Nind et al., 2013: 4–5) of an on-line com-
munity with mental health issues. From a societal perspective, this research may inform
public policy on the provision of supports for individuals suffering from mental health
problems. This may benefit the participants, their communities, and society as a whole.
Assuming the researcher only acts as an observer, risk of harm to participants is negligi-
ble and identities can be anonymized in the reporting of research findings.

The challenge arises in securing informed consent. This may be problematic for a
number of reasons. Covert observation may be required to preserve the authenticity of
interactions. Individuals previously involved in the community may have left and not be
contactable. If informed consent is only given by a limited number of participants,
removing other participants’ contributions may invalidate the analysis of the community
as a whole. Furthermore, that potential participants engaged in a publically accessible
on-line forum may arguably imply informed consent. Figure 8 illustrates how this exem-
plar might be mapped on our model by a committee. The mapping serves to pinpoint the
research ethics concepts which need to be examined and again, serves to appropriately
focus the committee’s debate.
Let us now adapt the previous example to further illustrate the model’s application.
Consider a situation where the researcher specifies in their application that they intend to
actively engage with on-line communities of participants under the guise of being a fel-
low sufferer of mental health problems. The effect of this is to increase the risk of harm
to research participants by directly engaging with them in a manner that misrepresents
the nature of the interaction. If this is discovered, it may damage the integrity of the com-
munity, upset participants who have been deceived, and contaminate findings. Visually,
this is represented by Figure 9, with a much larger area contained inside the dotted line
compared with Figure 8.
In the case mapped by Figure 9, which puts stress on more than one research ethics con-
cept, the moderating influence of the researcher would need to be considered by the commit-
tee. An experienced researcher, specifically trained in the care of individuals suffering from

Figure 9. Mapping of Exemplar 2 (higher risk version).
mental health problems, may be considered sufficiently experienced to mitigate the addi-
tional risk of this project, while it may be entirely too risky for a student researcher.
Exemplar 3
While this model facilitates greater tolerances to be allowed in the context of HSS
research, it also indicates when applications should be rejected. Too much stress on too
many of the axes within the model render a project too risky to receive approval. For
example, consider a student undertaking a master’s degree by research who wishes to
investigate racial discrimination in the hiring practices of large companies. The research
method involves applying for vacant positions in organisations using curricula vitae with
fake ethnic and non-ethnic names and comparing response rates for each category. While
it may be possible to maintain the anonymity of employers in reporting this research, to
seek informed consent from employers would render the research invalid. There is a risk
of substantial reputational damage to the relevant university should the research design
be discovered and made public, since employers are unlikely to trust applications from
students from that institution in the future. The potential benefits of the research (other
than to the researcher) are likely to be very low given that the findings from a master’s
thesis are unlikely to be disseminated or to influence organisational or national policy.
Figure 10 demonstrates how this project maps on our model. The considerably larger
area enclosed within the dotted line shows how this proposal raises far more ethical con-
cerns than previous examples.

Conclusions
Despite numerous calls in the literature for a change in approach to the research ethics
oversight of HSS research, the extant literature in the area fails to advance beyond
describing ethical dilemmas encountered by researchers in their own context and cri-
tiques delineating the unsuitability of biomedical research ethics procedures for HSS
research. Neither category addresses the underlying problem of how research ethics
oversight of HSS research might be conducted more sensitively without necessitating
separate processes and therefore significant resources. Against this backdrop we aim
here to advance the debate on the research ethics oversight of qualitative HSS research.
We do this by drawing on our collective decade of experience as members and chairs of
social science research committees, and on qualitative interviews conducted with mem-
bers of research ethics committees, to inform the development of a visual research ethics
model/tool which encompasses all the typical research ethics variables relevant to
research involving human subjects. Our aim is to assist research ethics committees in
evaluating qualitative research, without creating a dual system of research ethics over-
sight for different research domains. The model can be used to initially map a research
ethics application visually. Committees can then use their expertise and discretion to
adjudicate on a particular case, focusing on the areas where the application bubbles out
from the centre of the model. In accommodating the idea of contextual sensitivity, the
model provides a more nuanced framework that can aid evaluation of both biomedical
and HSS research proposals. We illustrate how this model can be applied to both contexts

through the application of different levels of tolerances in each domain. The model also
indicates when applications should be rejected. Too much stress on too many of the vari-
ables within the model render a project too risky to receive approval.
In terms of its limitations, this model does not provide a quantitative formulation of
any increased tolerances, but rather seeks to clearly identify specific areas where princi-
ples taken from the bio-medical review processes are too stringent for some HSS research
in order to focus the discussions of the research ethics committee using it. The model was
developed initially from reviewing the research ethics literature and was subsequently
informed by semi-structured interviews conducted in Ireland and the UK. We acknowl-
edge the limitations of this research method.
The development of this model is simply a starting position, aimed at generating con-
structive dialogue around the optimum research ethics oversight processes for qualitative
HSS research. We look forward to further research directed towards refining and enhanc-
ing this model and empirical research based on its practical use by research ethics
committees.
Funding
This research received support from the Irish Research Council for which we are very grateful.
References
Altbach PG, Reisberg L and Rumbley LE (2009) Trends in Global Higher Education: Tracking an
Academic Revolution. Paris: UNESCO.
Boden R, Epstein D and Latimer J (2009) Accounting for ethos or programmes for conduct? The
brave new world of research ethics committees. The Sociological Review 57(4): 727–749.
Borland KW (2001) Qualitative and quantitative research: a complementary balance. New
Directions for Institutional Research (112): 5–13.
Buckley CA (2011) Student and staff perceptions of the research-teaching nexus. Innovations in
Education and Teaching International 48(3): 313–322.
Cassell C and Symon G (2004) Essential Guide to Qualitative Methods in Organizational
Research. London: SAGE.
Christians CG (2005) Ethics and politics in qualitative research. In: Denzin NK and Lincoln YS
(eds) The Sage Handbook of Qualitative Research. 3rd ed. Thousand Oaks, CA: SAGE.
Connolly K and Reid A (2007) Ethics review for qualitative inquiry: adopting a value-based facili-
tative approach. Qualitative Inquiry 13(7): 1031–1047.
Cortada JW (1998) Rise of the Knowledge Worker. Burlington, MA: Elsevier.
Davenport TH (2005) Thinking for a Living: How to Get Better Performance and Results from
Knowledge Workers. Boston, MA: Harvard Business Press.
De Wet K (2010) The importance of ethical appraisal in social science research: reviewing a fac-
ulty of humanities’ research ethics committee. Journal of Academic Ethics 8(4): 301–314.
Dingwall R (2008) The ethical case against ethical regulation in humanities and social science
research. Twenty-First Century Society 3(1): 1–12.
Gregory I (2003) Ethics in Research. London: Continuum.
Guillemin M and Gillam L (2004) Ethics, reflexivity, and 'ethically important moments' in
research. Qualitative Inquiry 10(2): 261–280.
Haggerty K (2004) Ethics creep: governing social science research in the name of ethics.
Qualitative Sociology 27(4): 391–414.

Halse C and Honey A (2007) Rethinking ethics review as institutional discourse. Qualitative
Inquiry 13(3): 336–352.
Hammersley M (2009) Against the ethicists: on the evils of ethical regulation. International
Journal of Social Research Methodology 12(3): 211–225.
Hammersley M (2014) On ethical principles for social research. International Journal of Social
Research Methodology.
Hedgecoe A (2008) Research ethics review and the sociological research relationship. Sociology
42(5): 873–886.
Hoecht A (2011) Whose ethics, whose accountability? A debate about university research ethics
committees. Ethics and Education 6(3): 253–266.
Holland K (2007) The epistemological bias of ethics review: constraining mental health research.
Qualitative Inquiry 13(6): 895–913.
Hurdley R (2010) In the picture or off the wall? Ethical regulation, research habitus, and unpeopled
ethnography. Qualitative Inquiry 16(6): 517–528.
Israel M and Hay I (2006) Research Ethics for Social Scientists. London: SAGE.
Johnson TS (2008) Qualitative research in question: a narrative of disciplinary power with/in the
IRB. Qualitative Inquiry 14(2): 212–232.
Kennedy JE (2005) Grey matter: ambiguities and complexities of ethics in research. Journal of
Academic Ethics 3(2–4): 143–158.
Kitchener K and Kitchener R (2009) Social science research ethics: historical and philosophi-
cal issues. In: Mertens D and Ginsberg P (eds) The Handbook of Social Research Ethics.
Thousand Oaks, CA: SAGE, 667.
Kleinman DL and Vallas SP (2001) Science, capitalism, and the rise of the knowledge worker: the
changing structure of knowledge production in the United States. Theory and Society 30(4):
451–492.
Lewis M (2008) New strategies of control: academic freedom and research ethics boards.
Lincoln YS and Tierney WG (2004) Qualitative research and institutional review boards.
Milgram S (1974) Obedience to Authority. New York: Harper & Row.
Monaghan LF, O’Dwyer M and Gabe J (2012) Seeking university research ethics committee
approval: the emotional vicissitudes of a ‘rationalised’ process. International Journal of
Social Research Methodology 16(1): 65–80.
Moore N (2012) The politics and ethics of naming: questioning anonymisation in (archival)
research. International Journal of Social Research Methodology 15(4): 331–340.
Murray L, Pushor D and Renihan P (2011) Reflections on the ethics approval process. Qualitative
Inquiry 18(1): 43–54.
Nind M, Wiles R, Bengry-Howell A and Crow G (2013) Methodological innovation and research
ethics: forces in tension or forces in harmony. Qualitative Research 13(6): 650–667.
Owen M (2006) Conflict and convergence: the ethics review of action research. Journal of
Academic Ethics 4(1–4): 61–75.
Smith P and Rust C (2011) The potential of research based learning for the creation of truly inclu-
sive academic communities of practice. Innovations in Education and Teaching International
48(2): 115–125.
Stacey J (1988) Can there be a feminist ethnography? Women’s Studies International Forum
11(1): 21–27.
Stark L (2012) Behind Closed Doors: IRBs and the Making of Ethical Research. Chicago: The
University of Chicago Press.

Tilley S and Gormley L (2007) Canadian University ethics review. Qualitative Inquiry 13(3):
368–387.
Vainio A (2013) Beyond research ethics: anonymity as ‘ontology’, ‘analysis’ and ‘independence’.
Qualitative Research 13(6): 685–698.
Van Den Hoonaard W (2003) Is anonymity an artifact in ethnographic research? Journal of
Academic Ethics 1(2): 141–151.
Van Den Hoonaard W (2006) New angles and tangles in the ethics review of research. Journal of
Academic Ethics 4: 261–274.
Whittaker E (2005) Adjudicating entitlements: the emerging discourses of research ethics boards.
Health: An Interdisciplinary Journal for the Social Study of Health, Illness and Medicine
9(4): 513–535.
Wiles R, Charles V, Crow G and Heath S (2006) Researching researchers: lessons for research
ethics. Qualitative Research 6(3): 283–299.
Zimbardo PG (1973) On the ethics of intervention in human psychological research: with special
reference to the Stanford prison experiment. Cognition 2(2): 24–256.
Author biographies
Elaine Doyle is a senior lecturer in taxation in the Kemmy Business School, University of Limerick,
Ireland. Her research interests include inter alia, professional ethics and risk management in tax
practice, the tax aggression of tax practitioners, research ethics, procedural justice, ethical reason-
ing and ethics education. She has published both nationally and internationally in these areas in
journals such as the Journal of Business Ethics and Innovations in Education and Teaching
International; co-edited Innovative Business School Teaching: Engaging the Millennial Generation
(Routledge), and has secured national and international funding to support her research activities.
Patrick Buckley lectures in the area of information management in the department of Management
and Marketing in the Kemmy Business School, University of Limerick, Ireland. Patrick’s research
interests include the use of technology to address the challenges of large group teaching and
research ethics. His work in this area has been published in journals such as the Journal of Teaching
in International Business and Computers and Education. He was awarded the Jennifer Burke
Award for Innovation in Teaching and Learning in 2013.

Doyle2016

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Doyle2016

Uploaded by

Copyright:

Available Formats

661230

research ethics oversight ﻿1­–23

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Differences between biomedical and HSS research

Dangers for HSS research

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 1. Informed Consent Continuum.

Risk of potential harm

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 2. Anonymity Continuum.

Figure 3. Risk of Harm Continuum.

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 4. Reciprocity Continuum.

directly through increased self-understanding, satisfaction in assisting, learning about

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 5. The Basic Model.

Integrating the principles into a workable model

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Table 1. Interviewee profiles.

Interviewee Jurisdiction University Type of Committee Position

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Risk of potential harm

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Usefulness of a visual model

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 6. The Evolved Model.

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 7. Mapping of Exemplar 1.

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 8. Mapping of Exemplar 2.

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 9. Mapping of Exemplar 2 (higher risk version).

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Figure 10. Mapping of Exemplar 3.

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

Downloaded from qrj.sagepub.com at CORNELL UNIV on September 8, 2016

You might also like

research ethics oversight 1–23