Module 5 Assignment 1

Ethical issues in forensic and correctional research Tony Ward ⁎, Gwenda Willis Victoria University of Wellington, New Zealand abstract article info Article history:

Received 29 March 2010 Received in revised form 8 July 2010 Accepted 8 July 2010 Available online 15 July 2010 Keywords:

Forensic and correctional ethics Offenders Research ethics It is surprising that while the literature on ethical issues associated with social science research is burgeoning and becoming increasingly sophisticated very few papers have been written on the ethics of forensic or correctional research. The literature that exists is disappointingly narrow and superﬁcial, and relies on professional ethical codes to a considerable degree. In this paper we present an ethical framework developed by Ward and Syversen to help with ethical decision making in research contexts. We then discuss some of the speciﬁc ethical challenges for researchers working in forensic and correctional domains, and consider how best to deal with ethical problems drawing from this framework. Our aim will be to provide researchers with some general ideas of how to proceed in certain situations rather than come up with aﬁnal set of answers to every conceivable problem.

Contents 1. Introduction.............................................................. 399 2. An ethical framework for research in forensic and correctional domains.................................. 400 2.1. Addressing ethical conﬂicts................................................... 401 3. Ethical issues in forensic and correctional contexts............................................ 402 3.1. The design, collection, analysis, and reporting of research data................................... 402 3.1.1. Making theoretical assumptions explicit......................................... 402 3.1.2. Research aims..................................................... 402 3.1.3. Selection of research designs.............................................. 403 3.1.4. Data presentation.................................................... 403 3.1.5. Negative results.................................................... 404 3.1.6. Analysis decisions and resource wastage......................................... 404 3.1.7. Self-as-instrument................................................... 404 3.1.8. Participants and data.................................................. 404 3.2. Researchers and offenders: issues and problems.......................................... 405 3.2.1. Ethical blindness.................................................... 405 3.2.2. The moral status of offenders.............................................. 405 3.2.3. The vulnerability of offenders.............................................. 406 3.2.4. Cultural differences................................................... 407 3.2.5. Virtues and vices.................................................... 407 3.2.6. Assumptions about disorders.............................................. 408 4. Conclusions.............................................................. 408 References................................................................. 408 1. Introduction Social science research involves investigation of social phenomena at various levels of abstraction, from social institutions such as marriage or kinship systems to individual persons and the psychological and cultural processes that inﬂuence how they act and experience the world. Like any research care is taken to gatherdata in a systematic, explicit, and Aggression and Violent Behavior 15 (2010) 399–409 ⁎Corresponding author. School of Psychology, Victoria University of Wellington, PO Box 600, Wellington, New Zealand.

E-mail address:[email protected](T. Ward).

doi:10.1016/j.avb.2010.07.002 Contents lists available atScienceDirect Aggression and Violent Behavior transparent way only endorsing methods that have proved to be useful and jettisoning those that have resulted in unreliable data and faulty conclusions. Social and behavioral science is in short, a self-correcting epistemic engine and is constituted by a combination of general methods, data analytic strategies, and a social network of individuals who are committed to acquiring a deep understanding of human behavior and its cultural and social scaffolding (Sterelny, 2003). The norms that regulate the practice of science function to offset personal biases, human cognitive limitations, and political interference and through a combination of explicit, general rules and critical review have arguably positively transformed the lives of human beings in countless ways.

Alongside knowledge generating norms are ethical norms that are intended to ensure that the practice of social science is undertaken in ways that beneﬁt rather than harm research participants, researchers, and the rest of the community. These aims are nicely expressed by Symonette (2009):

Social research ethics speaks to the morally responsible ways in which we should conduct ourselves as we design and engage in systematic inquiry, analysis, interpretation, and dissemination pro- cesses via-a-vis individuals, groups, organizations, communities, and so on. Ethical practice is reﬂected in the extent to which researchers conduct themselves and their research in ways that are respectful, are fairly representative (accurate and just) and ideally leave persons better off or at least minimally,“does no harm”(280–281).

Thus, ethics in a research context is broadly concerned with undertaking investigatory activities in ways that respect each person's moral equality and associated entitlements (Driver, 2006). The above quote directly raises an important ethical issue concerning the competences that can be reasonably expected of a researcher and the degree to which norms ought be rationally justiﬁed rather than arbitrarily asserted (Kitchener, 2000). This general issue can be unpacked in the form of two fundamental ethical questions that social science researchers are obligated to respond to (Ginsberg & Mertens, 2009; Kitchener & Kitchener, 2009). First, how should research data be collected, analysed and reported? This question speaks to issues such as the design of a research project, its scientiﬁc merits, the way the data is presented in reports, in what forum reports are presented, keeping data private, and the status of the participants in the research process. Second, how should researchers act towards participants in their research projects? This question covers topics such as informed consent, conﬂict of interests, possible beneﬁts or harms that might be experienced by the participants, legitimacy of deception, existence of coercion, cultural differences, and the ethical status of participants.

It is surprising that while the literature on ethical issues associated with social science research is burgeoning and becoming increasingly sophisticated very few papers have been written on the ethics of forensic or correctional research (but seeAdshead & Brown, 2003; Overholser, 1987). The literature that does exist is disappointingly narrow and superﬁcial, and relies on professional ethical codes to a considerable degree. However,Lavin (2004)asserts that professional ethical codes are insufﬁcient to guide forensic and correctional research or practice and recommends that researchers acquaint themselves with ethical theory and concepts to enrich their ethical perception and decision making. Furthermore, Lavin states that ethical codes are opaque in their wording, contain inclusions, and exclusions that do little“moral work”and“invite psychologists to forget to look long and hard at controversial areas of practice”(p. 51).

Indeed,Ward and Syversen (2009) argued in their recent paper that ethical codes are comprised of sets of rules designed to address issues within a speciﬁc domain of practice and ethical conﬂicts can emerge when ethical rules from different domains clash. For example, the rules of a correctional agency that is preoccupied with risk containment (e.g., prison) may clash with a professional's ethical code of ethics (welfare concerns of prisoners). Higher level ethicalconcepts such as dignity or human rights provide theoretical resources that can be used to resolve such conﬂicts. Articles that do focus directly on forensic and correctional ethical concerns concen- trate on procedural matters and almost invariably are preoccupied with practice related issues such as who is the client, safety, dual relationships, how to deal with duty to warn situations, and so on (Bush, Connell, & Denny, 2006; Haag, 2006). More research oriented work tends to concentrate on matters of recruitment and barriers to research rather than systemically addressing the ethical questions confronting all social science researchers outlined above (Appelbaum, 2008). We certainly do not want to undervalue the relevance or importance of issues like participant recruitment or conﬁdentiality but simply raise questions about the depth and comprehensiveness of current scholarly debate on correctional and forensic research ethics.

In our view, because there has been no attempt that we are aware of to respond to the major ethical concerns raised by researchers in a broader social science context such asKitchener (2000), Kitchener and Kitchener (2009) or Ginsberg and Mertens (2009), researchers in the forensic and correctional domain lack a conceptual framework to help them to detect and resolve pressing ethical problems. There is a real danger that such researchers are simply unaware of difﬁculties that do exist and therefore run the risk of failing to act when they should do, or alternatively, dealing with problems that are noted in ethically inappropriate ways (Ward & Syversen, 2009).

In this paper we apply an ethical framework developed byWard and Syversen (2009; based on earlier work byCooper, 2004 and Kitchener, 2000) to help with ethical decision making in research contexts. We then discuss some of the speciﬁc ethical challenges for researchers working in forensic and correctional domains, using the two questions outlined above as a guide. We consider how best to deal with ethical problems that arise in forensic and correctional contexts drawing from this framework. The two questions are intended to be a heuristic guide only and we appreciate that there is a degree of overlap between them and also the identiﬁed ethical problems and our responses. Our aim will be to provide researchers with somegeneral ideasof how to proceed in certain situations rather than come with aﬁnal set of answers to every conceivable problem. We will spend relatively little time discussing practical matters such as how to balance the risks and beneﬁts of research or how to concretely ensure participant safety. In our view exact solutions are not possible because of the open ended nature of real world ethical concerns (Driver, 2006) and also because in some situations ethical principles conﬂict and there may be a number of ethically justiﬁable courses of action available to a person. While the possibility of ethical conﬂicts is unsettling to some, similar puzzles arise in the factual side of scientiﬁc practice; sometimes there are different but equally justiﬁable ways to explain phenomena. Presence of epistemic and ethical uncertainly on occasions is simply a legacy of our less than perfect cognitive and emotional equipment and also a reﬂection of the complexity of the world and its inhabitants. This paper differs from the earlier Ward and Syversen one in that it seeks to apply aspects of the framework developed in that paper to ethical issues in forensic and correctional research, and furthermore, assumes the validity of that ethical model. The ethical concerns that are the focus of this paper are ones that appear to us to be important, based on our reading of the literature on research ethics in allied areas and partly derived from our own research experience and reﬂection. Our hope is that the paper will provide an impetus for further theoretical and empirical work in a neglected area of research ethics.

2. An ethical framework for research in forensic and correctional domains In a recent paper,Ward and Syversen (2009)developed an integrated ethical framework that covered the different levels of ethical tasks inherent in forensic and correctional work. In our view, this framework provides a helpful way of addressing the various ethical 400T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 issues evident in forensic and correctional research. We will brieﬂy describe this model before putting it to work in the next section to provide guidance dealing with the problems discussed earlier.

The key argument of theWard and Syversen (2009)paper is that researchers and practitioners ought to justify their ethical decisions in a stepwise process, typicallyﬁrst relying on their commonsense everyday reasoning and then in successive steps appealing to the standards of ethical codes, principles underling ethical codes (e.g., beneﬁcence, autonomy, justice, and integrity), ethical concepts and theory and ultimately, the concept of human dignity. According to Ward and Syversen, when faced with justifying speciﬁc ethical judgments and subsequent actions professionalsﬁrst fall back to their relevant code of ethics. Typically, ethical codes are comprised of a series of foundational principles based on core values and a set of explicit standards or rules that reﬂect these core values (Bush et al., 2006; Koocher & Keith-Spiegel, 2009; O'Donohue & Ferguson, 2003). Forensic researchers and practi- tioners are introduced to basic ethical concepts during their training and are expected to continue to develop ethical expertise during the course of their careers. A problem relying simply on professional codes to deal with ethical dilemmas is that they may conﬂict with alternative ethical codes such as those overseeing the actions of correctional workers or other agencies. A second problem is that codes only cover customary practice and there may be situations they do not cover. Finally, there remains the thorny issue of how to justify the standards of ethical codes when challenged. According to theWard and Syversen (2009)model, when faced with a demand to justify the standards regulating their professional actions the next step is to appeal to the major ethical principles such as justice, beneﬁcence (caring), autonomy, or integrity. A useful way to regard the abstract principles is as core values that can be appealed upon to justify and explain speciﬁc ethical judgments and actions (seeCooper, 2004). Otherwise, there is a danger that practitioners will be simply left with a set of rules that: (a) can appear arbitrary if not underpinned by more abstract normative principles or theories, (b) may have limited scope in some situations and without principles to fall back on practitioners may be left without adequate normative guidance, and (c) have few resources for development since sets of ethical standards on their own are essentially closed conceptual systems.

When further pressed to justify these abstract principles, researchers may appeal to ethical theories or belief systems such as utilitarianism or virtue ethics (Driver, 2006). It is possible to shift to an even higher level of abstraction at the metaethical level where the status and meaning of terms such as“goodness”or“right”are analysed (Cooper, 2004; Kitchener & Kitchener, 2009).Ward and Syversen (2009)propose that the concept of human dignity provides a way of sidestepping arguments over the type of ethical theory that should underpin practice. The choice of“human dignity”is based on a convergent consensus that this term can supply the conceptual foundation for debates about universal justice, peace, and freedom initiatives around the world, as evidenced in the variety of human rights protocols currently accepted (Churchill, 2006; Li, 2006; Nussbaum, 2006; United Nations, 1948). In other words, the concept of human dignity and its associated ideas is acknowledged as foundational by many individuals whose speciﬁc moral theories and codes diverge and therefore it is ideally placed to function as an integrative concept for ethical reasoning in the forensic and correctional domains. Thus, ultimately the concept of human dignity underpins researchers' and practitioners' speciﬁc decisions, ethical codes, principles, and theories and also suggests ways to deal with the variety of complex ethical issues evident in both research and practice.

In essence, the concept of dignity designates the moralworthor value of all human beings (Sulmasy, 2007). Because of their inherent dignity human beings are assumed to possess equal moral status and therefore are expected to receive equal consideration in matters that directly affect their core interests. According to the analysis byBeyleveld andBrownsword (2001), the dignity of human beings is located in their capacity to formulate and pursue their interests in the world (empowerment) without unjustiﬁed interference by other people. In order to actively pursue their core interests and goals, individuals require a certain level of well-being and access to social, psychological, and material resources, otherwise it would be impossible for them to function as autonomous individuals (i.e., dependence on resources and opportunities are constraints, their absence undermines a person's agency efforts). However, individuals' own judgments are not the only determinants of what constitutes a digniﬁed life and sometime people can act in ways that are freely chosen that unfortunately diminish their status as human beings. An example in the correctional domain might be where offenders agree to harsh and humiliating interventions because they do not believe they are worthy of better treatment. A notable implication of a concept of dignity based on empowerment and constraint is that it points to the vulnerabilities of human beings because of their nature as embodied, social animals. In order to be able to act in pursuit of personally selected goals and their translation into plans of action, basic needs have to be meet as well and educational opportunities and social scaffolding provided. In other words, human beings require certain primary goods and opportunities to be able to act in ways worthy of their intrinsic value (Ward & Maruna, 2007). Because of their critical role both in helping people lead a life of dignity and ensuring they have the necessary capabilities to function according to their inherent digniﬁed nature, these resources are viewed asentitle- mentsand therefore are protected by fundamental moral and legal rights. It is important to emphasize that because respect for all individuals ultimately stems from their inherent dignity it cannot be ethically justiﬁably be taken from them through the actions of the state or by other people.

2.1. Addressing ethical conﬂicts A difﬁcult question concerns exactly how researchers should reason about the range of ethical dilemmas that confront them in correctional and forensic research. The nature of such conﬂicts and their complexity means that more often than not it is a question of engaging in an ethical balancing process. If so, exactly how should the process of ethical reasoning based upon the concept of human dignity unfold in research contexts? It is clearly not possible to resolve this issue in a paper intended to provide an overview of both underap- preciated and noted ethical concerns in research but there are some important conceptual constraints that ought to regulate any such attempts. According toBecker (1986)all moral practice is under- pinned by two core justiﬁcatory assumptions, those ofgeneralizability and equality. Thus, when faced with the inevitable dilemmas generated by research, any solutions should: (a) generalize to other persons in similar situations, and (b) treat every person involved as someone of equal moral status (i.e., as possessing inherent dignity).

The latter constraint means that unless there are good reasons it is presumed that the interests of all individuals will be equally considered and any decision should result in outcomes that are judged as fair. The two constraints of generalizability and equality will rule out some strategies for addressing research ethical problems. For example, mandated treatment (and associated research projects) would be ruled out for psychologically competent offenders because it overrides the equality assumption without adequate justiﬁcation.

Furthermore, it arguably also violates the assumption of generaliz- ability as it is unlikely researchers would permit themselves or non- offenders to be forcibly treated. Similarly, the biased presentation of data would be ruled out because it could result in wasted research resources, which denies the opportunity for policy makers and the community to make balanced decisions about the allocation of funds.

This failure ignores the equality of individuals as co-decision makers, and if universalized, could undermine research completely. Finally, when recruiting participants for a project neglecting relevant 401 T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 individual differences such as ethnicity or overlooking the vulnera- bility of offenders to implicit or explicit coercion does not succeed in meeting the requirements of both tests. It fails the generalizability assumption because most individuals would reject a general rule that allowed researchers to exploit vulnerable individuals for research purposes. Furthermore, the fact that the participants were not provided with (a) relevant information or were (b) pressured to consent would also fail to meet the equality assumption.

In summary, we propose that researchers approach ethical issues with the above framework in mind. This multilevel framework ought to be able to help them detect ethical problems that are typically missed,ﬁnd creative solutions to those that are identiﬁed, and provide deeper and less question begging justiﬁcations for decisions. We will address some of the pressing ethical issues associated with forensic and correctional research and demonstrate how this framework can help to shed some light on these concerns.

3. Ethical issues in forensic and correctional contexts 3.1. The design, collection, analysis, and reporting of research data Ethical problems in this section all revolve around the design of studies and the ways in which data are collected, analysed, and presented in journals or reports. The various issues cluster into the following themes: (a) making theoretical assumptions explicit, (b) aims of research, (c) selection of research designs, (d) data presentation, (e) negative data, (f) analysis decisions and resource wastage, (g) self as an instrument, andﬁnally (h) participants and data.

3.1.1. Making theoretical assumptions explicit Aﬁrst problem involves the importance of ensuring that a researcher's theoretical assumptions are made explicit rather than remaining implicit (Brown & Hedges, 2009). Articulation of a research problem often will involve clarifying the nature of a phenomenon such as deviant sexual arousal in sex offenders or speculating about the causal mechanisms that generate a phenomenon, for example, deviant sexual preferences (Ward, Polaschek, & Beech, 2006). Once researchers are clear about their assumptions it is then a question of ascertaining what kind of design will provide the data necessary to clarify or address the problem at issue. The danger is that failure to clearly specify the postulated relevant mechanisms may result in poorly directed investigation or else premature acceptance of current ideas and practice. Misdirected research might occur due to a lack of sustained analysis and a reliance on theoretical concepts that contain contradictory elements or are overly vague. In addition to wasting researcher and participant resources, unfounded or vague specula- tions about the offender group in question may harden into received dogma and derail future investigatory efforts and also lead to poorly conceived policies. An example of unfounded assumptions is the imposition of geographical residency restrictions on sex offenders which seems to be based on mistaken assumptions about sex offenders offense related actions, namely that they will prey on the geographically closest victim. In fact, evidence indicates that geo- graphical proximity on its own is not a reliable indicator of sexual offending and therefore the resources tied up in environmental restrictions are misdirected (Duwe, Donnay, & Tewksbury, 2008; Zandbergen, Levenson, & Hart, 2010). Another example is the sex offendingﬁeld's earlier view that offenders only follow one offense and/or relapse process, whereas, theoretical analysis and later empirical evidence revealed that there are in fact a number of distinct offense trajectories (Hudson, Ward, & McCormack, 1999; Ward, Yates, & Long, 2006; Yates & Kingston, 2006).

Making theoretical assumptions explicit will also avoid the danger of premature negative stereotyping where assumptions are made about the attributes of a group of offenders, for example violent males, that are unjustiﬁed (Major & O'Brien, 2005). According to the Duluth model maleviolence against woman is the result of gender based injustices and inherent sexism (Langlands, Ward, & Gilchrist, 2009). Failure to critically examine this assumption has resulted in poorly directed research, the establishment of treatment programs that do not seem to be particularly effective (Smedslund, Dalsbø, Steiro, Winsvold, & Clench-Aas, 2007), and may also cause offenders to construe themselves in ways that are untrue and potentially damaging (Langlands et al., 2009). Another example is the tendency to view child sex offenders as predators whose deviant inclinations means they are consistently at risk for harming children if insufﬁciently scrutinised and controlled. Whereas, in reality the evidence suggests that for many sex offenders offending is contextually dependent and also that social inclusion rather than exclusion is more likely to lead to reduced offending (Laws & Ward, in press). While we appreciate that not all forensic and correctional research sets out totest theory,itiscommonlyacceptedinthe philosophy and sociology of science that both descriptive and theory testing research activities are underpinned by supporting theoretical and methodological assumptions (Balashov & Rosenberg, 2002). One of the virtues of making such assumptions explicit is that their limitations and validity can be more easily evaluated thus helping to avoid poorly thought out projects.

The concept of human dignity points to the obligation of all researchers to consider the interests of all individuals affected by research when planning and carrying out studies. Failure to do this may mean that offenders are characterized in terms that are stigmatizing as well as inaccurate. The justiﬁcatory assumptions of universality and equality, in conjunction with the concept of dignity, outlined above should minimize the chances of this happening.

3.1.2. Research aims A second point concerns the broader social aims of research in ethical and forensic contexts irrespective of whether its focus is on etiological or assessment and program related matters. The design of a study and methods of data collection depend on the reasons for undertaking it and also arguably who is likely to beneﬁtfromtheﬁndings. For example, if the members of the public have a strong interest in ensuring that sexual violence towards women and children is reduced then research priorities in the area of sexual violence are likely to reﬂect this concern.

A difﬁculty may be that research into other, equally important, correctional and forensic issues is neglected or that there is pressure to develop programs without the basic science being in place. Thus, in order to acquire funding researchers may prematurely accept clinically derived theory and treatment programs based on this theory as warranted knowledge and proceed to design program evaluation studies. An example of this is arguably the premature acceptance of suggestions that there is a causal relationship between sexual offending and negative affect whereas the actual research evidence for this relationship is comparatively meagre (McCoy & Fremouw, 2010).

Alternatively, as stated above, the treatment of men who are violent towards their partners is based on very little research evidence and surprisingly weak theory (Langlands, Ward, & Gilchrist, 2009). It would not be hard toﬁnd examples in other areas of forensic and correctional psychology. The key point is that researchers ought to critically reﬂect on the reasons why they want to proceed with a study and whether there is sufﬁcient justiﬁcation to do so. The adverse ethical implications of proceeding with research projects based on insufﬁcient grounds is that:

(a) it could mislead people, (b) it confers inappropriate authority on a program or researchﬁndings, and (c) it may divert precious resources from possibly more valuable projects. The politics of research funding can sometimes distort researchers' judgments and there is a risk that wasted research efforts may damage the chances of further research funding and ultimately the reduction of reoffending rates and human misery. The justiﬁcatory assumptions of universality and equality should encourage researchers to think systematically about the beneﬁts and burdens of research from the perspectives of all stakeholders and hopefully lead them to carefully balance any competing interests. 402T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 Another strand to the question of the underlying aims that motivate research in forensic and correctional context is the matter of whose interestsare taken into account in the inquiry process.

According to the ethical framework described above, all human beings have core interests (and associated dignity) and are entitled to have these interests taken into consideration in social policies initiatives that are likely to affect them. For example, communitarian and restorative justice approaches to punishment and reintegration depend on the participation of offenders, victims, and members of the community for effective outcomes (Miller, 2003; Ward & Langlands, 2009). Furthermore, research into the impact of restorative punish- ment and reintegration practices likewise assumes that all individuals are committed to a common set of aims and that their particular viewpoints have been canvassed and taken into account. If this is not the case and the needs of victims, offenders or members of the public have been unreasonably neglected, then the integrity of the research has been threatened, both ethically and empirically. The ethical issue raised here speaks to a need to be inclusive when undertaking forensic and correctional research and not to simply justify research on the basis of the assumed beneﬁts for one group of individuals. From an ethical perspective, all individuals (universal justiﬁcatory assump- tion) affected by a project merit equal consideration (equality justiﬁcatory assumption) and the various risks and beneﬁts balanced when undertaking applied research (Ward & Langlands, 2009).

3.1.3. Selection of research designs A third ethical problem relates to decisions concerning the kind of research designs and analytical methods employed in a project. The choice between the use of ethnographic, interview based, or experi- mental research designs is a crucial one and has ethical dimensions alongside epistemic considerations such as the degree to which causal inferences are facilitated or natural social processes respected. A useful example of how disagreement over design issues can erupt into ethical debates is the recent interchange betweenMarshall and Marshall (2007, 2008) and Seto et al. (2007)over whether or not the random controlled trial (RCT) is the gold standard for research into the effectiveness of sexual offending treatment. Marshall and Marshall argued that RCT designs lack ecological validity and also raised serious ethical concerns about allocating offenders to nontreatment conditions, thereby possibly placing future members of the community at risk from untreated men.

In their view, RCT are not necessary for treatment evaluation research in the sexual offending arena and they suggested that alternative research designs such as incidental designs or actuarially based evaluations may be better options. Seto et al. took issue with Marshall and Marshall's contentions and argued strongly for the value of RCT designs (suitably implemented) in providing robust evidence for treatment effectiveness.

They commented that proceeding with treatments when there is insufﬁcient evidence for their utility is unethical and also that there may be further unintended (ethical and psychological) consequences that have not been detected, for example, embellishing deviant sexual fantasies in individuals following their exposure to the fantasies of others in groups.

The issue we would like to highlight is that there are good ethical arguments on either side of the RCT debate and that the choice of a research design ought not to rely simply on scientiﬁc grounds. In our view, ethical reasons should take priority over empirical or scientiﬁc ones and research that violates important ethical assumptions (e.g., acceptably diminishes dignity and violates the assumptions of univer- sality and equality) and constraints ought not to proceed.Mark and Gamble (2009)assert that despite constituting an extremely powerful research design sometimes RCT designs may not be desirable if reasonable evidence already exists that certain treatment approaches are likely to be useful. They state that for a RCT design to be implemented “there needs to be real uncertainty about the best course of action, that is, about which alternative is preferable”(p. 204). There is certainly a question mark over whether this degree of uncertainty exists when itcomes to the treatment of sex offenders with cognitive behavioral techniques (CBT). The impressiveﬁndings from research on offender programs that target criminogenic needs (dynamic risk factors) and follow the principles of risk and responsivity are likely to reduce reoffending rates is certainly relevant here (Andrews & Bonta, 2003; Hanson,Bourgon,Helmus,&Hodgson,2009). In short, evidence from this research is that sex offender CBT oriented programs that follow these principles are likely to result in reduced reoffending rates. If a CBT treatment progam for sex offenders follows the principles of risk, need, and responsivity it is possible that a quasi-experimental design of some kind is scientiﬁcally acceptable, and ethically supported. In fact, it is reasonable to conclude that weight of the ethical reasons could favor such designs in view of their practicality and avoidance of the ethical problems associated with RCT designs with a sex offender population. In this example, the treatment needs of sex offenders as well the interests of research funders (resource considerations) and the public (safety) are all considered. The reasons for doing so include considerations of equality and universality, themselves designed to reﬂect the common dignity of every person.

A problem that is especially relevant for researchers using interview based methods and qualitative designs is the danger of creating inappropriately intense and close relationships with participants which ultimately may prove harmful to the offender and also compromise the integrity of the research itself (see section below on dual relationships). Interview based approaches frequently require researchers to establish strong rapport with offenders and there is a danger in prison environments characterized by mistrust, loneliness, and fear that such a relationship could cross the boundaries into a therapeutic one. A potential consequence of this boundary violation is that the interactions will lead to inevitable disappointment and possibly a sense of betrayal in the offenders concerned. By way of contrast, this type of problem is less likely to occur in research using experimental and quasi-experimental designs because of the relationship between researcher and participant is more formal and socially distant.

Understanding that offenders are in a vulnerable situation and treating them with sensitivity, respect, and care should help researchers navigate their way around these potential difﬁculties. In essence, researchers ought to ask themselves how would they feel in such a situation (universality assumption) and what common needs and interests of offenders should be taken into account when implementing this type of research (equality assumption). In addition, it is necessary to articulate the extent and limits of the researcher's role and care taken to make sure the relationship does not ease into a therapeutic one. An ethic based on human dignity and its related justiﬁcatory assumptions of universality and equality, rather than a simple cost–beneﬁt utilitarian approach, should assist researchers to ensure that offenders core interests are never traded off simply for the beneﬁtofothers.

3.1.4. Data presentation A fourth ethical issue related to design, data collection, and analysis concerns the way data are presented. Researchers should be careful to take into account acceptedﬁndings that are relevant for their research questions and ensure that the data are analysed in ways that reﬂects these distinctions, and presented accordingly. For example, research on sex offenders that fails to distinguish between risk levels and variables that are known to effect outcomes such as age, sexual deviancy, emotional regulation and so on is misleading and ethically unacceptable (Craig, Beech, & Harkins, 2009). It may result in clinical and research decisions that unfairly disadvantage offenders such as unnecessarily long parole periods or intrusive parole conditions. This problem is particularly pertinent to research on risk assessment. An associated issue is the need to examine results closely and to not rely on simple group effects.Bersoff et al. (2003)emphasize the importance of acknowledging limitations of empirically derived predictions in practice given their use in decisions such as handing down sentences of preventive detention. They highlight that 403 T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 dichotomous judgments are unethical but that class based judgments (provided limitations of such judgments are acknowledged) are ethical—i.e., reporting that a group to which an individual is alleged to belong has a speciﬁed risk of reoffending. Thus, in communicating researchﬁndings, especially on predicting risk of violence/reoffend- ing, it is crucial that characteristics of the research sample are accurately reported. Omitting details of sample characteristics could lead to risk assessment instruments being used with individuals for which the instrument has not been validated. Similarly,Vess (2008) highlights the profound impact risk assessment outcomes can have on (denying) the human rights of sex offenders, and moreover the limitations of current instruments in accurately predicting recidivism.

These in effect, constitute violations of the justiﬁcatory assumptions of equality and universality. Accordingly, he advocates for careful communication of risk assessment research data, especially acknowl- edgement of the limitations of interpreting risk assessmentﬁndings with respect to individual offenders. Finally,Douglas and Skeem (2005)state that although advancements have been made in the assessment of violence risk, current structured risk assessments do not often allow for change over time or across contexts. They provide several recommendations for future research in the risk assessment ﬁeld, including detection of change in risk factors over time, stable and time-speciﬁc components of risk factors, and mediators/moderators of risk factors. In light of the above debate, it is prudent for researchers when using risk bands for research purposes or conducting research on risk to take great care to communicate the extent and limitations of theirﬁndings. The implications of the results and the way they are expressed should be considered from the viewpoints of all the relevant stakeholders, not simply those of the researchers, the public, and the relevant funding bodies.

3.1.5. Negative results Aﬁfth, associated issue, concerns the failure to publish negative studyﬁndings whether they arise from assessment, etiological, or program evaluation research projects. This matter is as much the responsibility of journal editors and reviewers as individual research- ers. For example, it is a well established recommendation that real efforts should be made to include both published and unpublished studies in meta-analysis in order to avoid theﬁle drawerproblem, where studies with negative effects are not published (Crombie & Davis, 2009). A difﬁculty in neglecting to make negativeﬁndings public is that erroneous conclusions about a measure or intervention program's effectiveness may unduly inﬂuence subsequent funding decisions and possibly disadvantage offenders and community members. It goes without saying that such a situation could result in withdrawal of good programs or the continuation of weak ones.

3.1.6. Analysis decisions and resource wastage A sixth problem emerging from data analysis and study design matters occurs when researchers embark on a large study without undergoing power analyses or fail to have explicit rules about what to do if an intervention is clearly effective or ineffective (Mark & Gamble, 2009). In some types of program evaluation research it may be evident early on that a treatment is working and likely to help offenders enhance their quality of life and also reduce risk to the community. It could be argued that in these situations it is ethically and scientiﬁcally advantageous to terminate the study prematurely and introduce the program more generally (Morris, 2007). However, terminating a study early could result in some kind of penalty imposed by funding agencies thereby creating an ethical dilemma for forensic and correctional researchers. The ethical issue at stake here concerns the obligation to think about the interests of all those (i.e., universal) affected by the research and to value all interests equally (i.e., equality). Another difﬁ culty could occur when a study's inability to demonstrate a signiﬁcant effect size is due to insufﬁcient numbers, essentially amounting to a waste of resources. Alternatively, a studycould have produced a large effect size with far fewer participants. A lack of clear policy around when to stop effective or ineffective studies again speaks to a waste of resources and poorly thought through ethical procedures and norms. These are ethical issues because unnecessary interventions may cause participants harm, either in terms of their direct effects or the opportunity costs that arise because they were unable to engage in other, possibly more fruitful, activities.

3.1.7. Self-as-instrument A seventh ethical issue arises from ethnographic and qualitative research traditions where the researcher's self is viewed as an instrument and therefore subject to concerns about reliability, sensitiv- ity, and validity (Symonette, 2009). Problems of bias, lack of reliability and validity arise when researchers' perceptions of participants play an important role in data gathering and interpretation. The source of bias and subsequent contamination of data analysis is most likely to originate in the theoretical allegiances and methodological commitments of researchers, and ultimately, their core values (Day & Ward, 2010).

Values reﬂect fundamental commitments concerning what is worth- while and best in life and as such underlie choices about how to behave.

In this way, it is suggested that values are evident in the broad life goals individuals set for themselves and others and in the way in which these goals are pursued. The danger is that lack of critical scrutiny of theoretical and value assumptions may undermine the integrity of inquiry and result in data and subsequent publications that reﬂect aspects of researchers' personal worlds and communities rather than accurately describe the phenomena in question. There is ample research evidence that conceptions of the self may vary within and between cultures and that this fact can inﬂuence policy and social behavior, and ultimately certain kinds of research (Li, 2006; Ntseane, 2009; Parekh, 2006). For example, collectivist views of the self necessarily appeal to the role of social groups and their norms in shaping and selecting individuals' behavior, and thus rely on external social inﬂuences to guide a researcher's actions. An illustration of aspects of a researcher self derived from a collectivist culture would be dispositions to base knowledge claims on social consensus rather than simply individuals' judgments. Thinking clearly about the viewpoints of all participants rests on the ethical assumptions of equality and universality, and factoring in how common needs are expressed in different cultural and social contexts. There is clearly overlap between conceptions of self as an instrument and researcher vices and virtues that we will discuss later in the section on the researchers relationships to their participants.

Putting the question of the nature of the self to one side, there are also question marks about individuals' abilities to adequately reﬂect on their own responses and mental states when involved in research activities. This is a knowledge related concern and is based on evidence from cognitive science that much of individuals' cognitive processing occurs below the threshold of awareness (Clark, 2008). If this evidence is accepted then to expect researchers to be able to reﬂect critically on their values and thinking processes could be ethically unjustiﬁed because it is beyond their capacity. Furthermore, reﬂection and the values that guide it (e.g., accuracy, fairness, concern of truth, etc.) are themselves ethically laden and can be construed as virtues, an ethically saturated term (Driver, 2006). The implication of the ethical nature of such abilities is that there always are question marks over whether they are worthy of cultivation and also whose interests they serve. In forensic contexts there is real danger that unexamined assumptions about the nature of human goods, punishment, and individual responsibilities could distort attempts to interview and otherwise engage with offenders (Ward & Salmon, 2009).

3.1.8. Participants and data Aﬁnal ethical concern associated with research design, data collection, and presentation revolves around the role of participants in gathering the data and to what degree they should be regarded as co- owners of the information, and potentially, co-authors of subsequent 404T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 research publications. For both ethical and methodological reasons it could be argued that because research participants consent to become involved in a project they have a claim on the data yielded and therefore some entitlement to co-ownership (Ntseane, 2009). This conclusion is partly a function of regarding participants as equal moral agents and also recognition of the fact that agreement (effort) related to a research project confers an entitlement to a full report on its outcomes, and possibly a say in how the data are used—based on equality and universality considerations. And if the data are seen as partly owned by the participants it follows that the publication of the results ought to appear in forms that they are likely to read and understand, and not be simply appear in specialized academic journals. Indigenous offenders may have a particularly strong claim for co-ownership and control, especially when the research touches upon their cultural beliefs and practices (LaFrance & Bull, 2009). In forensic and correctional contexts this ethical concern is pressing because of the power imbalances that exist and reluctance to consider offenders as moral equals and potential research partners (Ward & Salmon, 2009). From a dignity perspective, it is easier to appreciate the force of participant claims to data ownership and to negotiate the tricky issues more adroitly. For example, it seems perfectly acceptable that participants who have contributed signiﬁcant- ly to a project can reasonably expect a full report on theﬁndings.

Furthermore, if their involvement is considerable, as it inevitably is in some types of ethnographic or interview based research, claims for access to data and some say on where and how they are reported have, at least, prima facie validity.

3.2. Researchers and offenders: issues and problems The ethical problems that arise from research design, data analysis, collection, and presentation also have echoes when it comes to considering researchers direct interactions with offenders on research projects. Furthermore, relevance and importance of the justiﬁcatory assumptions of universality and equality (and dignity) outlined earlier, should also modulate researchers' decisions and subsequent actions.

However, the key issues associated with this category spring directly from the face to face nature of researcher-participant relationships and essentially revolve around the themes of (a) ethical blindness, (b) offenders' moral status, (c) offenders' vulnerability, (d) cultural/social differences, (e) researcher vices and virtues, and (f) assumptions about disorders and offender treatability (Howells & Day, 2007).

3.2.1. Ethical blindness Aﬁrst ethical concern evident in research and practice domains is that ofethical blindness(Ward & Syversen, 2009). What we mean by this term is that important ethical matters can remain undetected, in part because of the conceptual constraints imposed by the ethical resources relied on by researchers such as professional ethical codes.

In our experience, a problem with ethical codes is that because of their derivation from routine research and clinical experience they often do not have anything meaningful to say about some matters that can be legitimately viewed as ethical ones, that is, concerns that are outside the normal moral experience of practitioners and researchers. That is, ethical codes are intended to guide therapists (primarily) in their work with client groups and are developed upon the basis of common problems and issues experienced by them in practice. They are backwardrather thanforward lookingand do not contain the theoretical resources to orientate clinicians to detect ethicalﬂash- points that have not yet beenﬂagged in such encounters (Cooper, 2004; Ward & Syversen, 2009). A good example of this unease is the allocation of offenders into treatment bands according to their assessed level of risk, often defended by reference to the Risk-Need- Responsivity model of offender rehabilitation (RNR;Andrews & Bonta, 2003). The problem with restricting research and treatment in these ways is that offenders can be denied legitimate access to treatment, access that they are ethically entitled to receive accordingto human rights declarations, and more fundamentally, their moral equality. In some instances, lower risk offenders in prison or on community orders might have their ability to function effectively seriously compromised by their problems and therefore urgently require help to address the causes of these difﬁ culties (Ward & Maruna, 2007). The worry is that the problems they do experience could impair their agency and, therefore, result in lowered levels of living and hence constitute signiﬁcant threats to their dignity (as evidenced in adverse life circumstances).

The key point we would like to stress is that problems such as these are not really apparent in forensic ethical writings; they are essentially invisible. Forensic researchers and practitioners often simply assume that the allocation of psychological and social resources is best based on risk level and most sensibly directed at the causes of offending. This is arguably because the normative ethical theory that implicitly or explicitly directs their research is a variant of utilitarianism, which stipulates that the ethically right action in a situation is one that results in the overall greatest amount of happiness or well-being (Driver, 2006). An acknowledged difﬁculty with utilitarian thinking is that in some circumstances it may lead to individuals legitimate interests being disregarded because to do so will lead to an overall greater utility (i.e., better overall outcomes).

While we would not argue with the need to address the causes of offending and to ensure that high risk offenders receive adequate amounts of treatment and associated research, this does not void the legitimate claims of lower risk offenders. It seems that potential and actual ethical problems like the ones raised above are ignored because they are notﬂagged by the codes that apply to offenders. Failure to fund or allow research into the problems of lower risk individuals is arguably an example of where the core interests (and dignity of a minority-offenders) is disregarded for the sake of the interests of the rest of the community. In our view, this is a serious ethical problem and one that should be directly confronted and not remain undetected and neglected because of ethical blindness.

3.2.2. The moral status of offenders Secondly, the above issue of moral blindness places the question of offenders' moral status directly into the lime light: do offenders possess the inherent dignity and hence value of the other members of the community (i.e., equality and universality concerns)? If so, does the fact they have received punishment erode their moral status? This is a thorny issue but to cut a complex and long story short we answer these questions in the following way: yes and no (for a detailed discussion of punishment and offender rehabilitation seeWard & Salmon, 2009).

Offenders possess the inherent dignity of all members of the community because they are fellow human beings. This dignity is acknowledged and protected by human rights treaties and protocols and applies to offenders and non-offenders alike (Ward & Birgden, 2007). Second, because of offenders' inherent dignity punishment is arguably only ethically justiﬁable if it is administered in a way that is respectful, proportionate, and allows for the prospects of repentance and reconciliation (Duff, 2001). In effect, this means that research with offenders should proceed in accordance with their equal moral status.

The harsh conditions of imprisonment and neglect of offenders' welfare apparent in many correctional jurisdictions throughout the western world implies a lack of acknowledgement of this inherent dignity (Lazarus, 2004; Travis, 2005). Poor living conditions, harsh punishment, lack of adequate medical care, isolation from other offenders, lack of sufﬁcient contact with family members, minimal opportunity to acquire useful employment skills, a hostile and unforgiving community all imply a fundamental lack of recognition of offenders' basic human rights and inherent dignity (Lazarus, 2004; Lippke, 2002; Ward & Birgden, 2007). A lack of dignity recognition creates problems for researchers as such conditions render subsequent research with vulnerable individuals problematic (see below). 405 T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 The matter of informed consent with prisoners does raise the question as to whether it is ever appropriate to use prisoners as research participants.Perez and Treadwell (2009)cite recent research showing that biomedical researchers in the US having difﬁculty recruiting volunteers for clinical trials (e.g.,Hoffman, 2000) are pushing for access to prisoners as participants in clinical trials (this was common practice as late as 1969, but not currently). The authors discuss ethical issues surrounding prisoners as participants in clinical trials, and argue that“until the question of adequate health care for prisoners is resolved, human experimentation should not be allowed” (p. 201). They go on to assert that“for research in prisons to be ethical, it must be interested in upholding prisoners' constitutional right to appropriate quality care while in prison and ensure a stronger and more effective safety net for them when they return home”(p. 201).

The authors therefore advocate for research with prisoners, but only when its focus is on the advancement of prisoners' welfare. It is possible, however, that sometimes the pendulum may swing too far in the other direction and correctional ofﬁcials could be overly paternalistic and deny offenders the opportunity to decide for themselves whether or not to take part in research projects (Over- holser, 1987). Issues related to informed consent and competency are to the forefront when prisoners are involved in research projects and there is a heavy burden of proof on researchers to establish that in forensic and correctional settings that offenders were under no pressure to consent, and were indeed, psychologically competent to do so. Accepting that any decisions arrived at should reﬂect the equality of all individuals concerned, and any solutions should generalise to similar contexts, ought to assist researchers to arrive at ethically sound options.

A neglected ethical issue that emerges from a discussion of offenders' moral status and informed consent is how much informa- tion should a researcher impart to potential participants? Should all the aims of the research and their potential applications be raised with an offender or is it sufﬁcient to simply describe the major goals and not engage in such speculation (Chwang, 2010)? This conundrum is nicely illustrated in risk assessment research and the use of such research to inform policy in the management of sex offenders (Vess, 2009). In many western jurisdictions sex offenders are assessed prior to release from prison and if deemed to be of high risk may have strict parole conditions imposed on them or even be committed to special treatment institutions until such time as their risk is signiﬁcantly reduced. For some offenders, this incarceration could amount to a life sentence (Vess, 2009). In view of the possibility of risk assessment instruments being used in ways that could have adverse effects for (harm) offenders in the future we argue it is obligatory for researchers to make this possibility explicit to potential study recruits. This obligation is directly derived from the justiﬁcatory assumptions of equality and universality, and the concept of dignity they are intended to protect. That is, researchers should state when describing a risk assessment study that one of the consequences of the reﬁnement of the measure concerned is that it might be used in the evaluation and possible conﬁnement (or imposition of strict parole conditions, etc.) of offenders in the future. Failure do so would arguably constitute a breach of informed consent norms and a violation of an offender's inherent dignity and status as a moral agent. Similar points could be extended to any type of research into the etiology, assessment, and treatment of offenders. In fact, in the case of psychopathy our point holds as the presence of psychopathic features is currently considered to increase risk levels (Vess, 2009).

3.2.3. The vulnerability of offenders Third, offenders are a vulnerable population because of the fact that their core well-being and freedom interests are often neglected or put at risk when living within correctional and forensic institutions (Overholser, 1987; Regehr, Edwardh, & Bradford, 2000). The recent Institute of Medicine (2006)report highlights the degree to whichprisoners are a vulnerable population and outlines speciﬁc ethical considerations that are required for research involving prisoners as participants. For example, potentially inadequate physical and mental healthcare in prisons presents a barrier to ethical research, as research participation might reﬂect a desperate attempt on the part of prisoners to obtain treatment. Five objectives–that reﬂect the assumptions of universality and equality–are advocated for by the Institute of Medicine committee to address the ethical problems associated with offender vulnerability:

1.Expand the deﬁnition of prisoner(to include all individuals whose liberty is restricted by the criminal justice system, i.e., not just individuals incarcerated); 2.Ensure universal, consistent ethical protection(i.e., irrespective of the source of funding or correctional facility); 3.Shift from a category-based to a risk-beneﬁt approach to research review. Currently (i.e., under the Code of Federal Regulations), research involving prisoners mustﬁt one of four categories:

incarceration or criminal behavior; prisons; conditions affecting prisoners; or research beneﬁting its subjects. Switching to a risk- beneﬁt approach will ensure studies offering no beneﬁt to prison- ers will be excluded—for example testing of cosmetic products.

4.Update the ethical framework to include collaborative responsibility.

The committee advocated that all stakeholders (e.g., prisoners, correctional ofﬁcers, medical staff, administrators) should have input into research design/planning/implementation.

5.Enhance systematic oversight of research involving prisoners.

Vulnerability of prisoners and offenders more generally needs to be taken into account when it comes to research into mandated treatment programs for offenders as well as ones where there is evidence of more subtle types of coercion (for a good recent review of issues related to coercion seeDay, Tucker, & Howells, 2004). The fact that offenders who are coerced into treatment may demonstrate favorable outcomes does not necessarily support such interventions because of the violation of offenders' autonomy and ultimately dignity as a human being (Miller, 1998). For example,McSweeney, Stevens, Hunt, and Turnbull (2007) recently conducted a quasi-experimental study investigating outcomes for offenders who volunteered for drug treatment compared to offenders who were coerced (i.e., court mandated) into treatment.

They found that coerced offenders reported signiﬁcant and sustained reductions in illicit drug use and offending behaviors, as well as improvements in other areas of social functioning. Findings were similar for the voluntarily treated offenders, and there were no signiﬁcant differences on outcome variables between both groups. The authors stated that the message from their research“is not that coercion works, but that treatment can be an effective alternative to imprisonment”(p.

486). Moreover, they emphasized that a large proportion of dependent users do not fund their drug use through crime, and that these individuals should equally have access to good quality drug treatment (Bonnie, 2006; Marlowe, 2006 on the topic of mandated treatment for addicted offenders). Despite these qualiﬁcations we have grave concerns about such research and it seems apparent that in the design of this study the potential beneﬁts for the community and participants (well-being, beneﬁcence) was thought to outweigh the autonomy of the participants. Such paternalism is hard to ethically justify unless the autonomy of the participants was severely impaired by drug use and therefore their competence to consent to treatment, or not, was severely undermined. There is no evidence in this study that this was the case.

AusefulpaperbyCaplan (2006)underlies our concerns about coerced treatment and research that follows from it, and the requirement to explicitly keep the assumptions of universality and moral equality in mind when undertaking research with offenders.

Caplan uses the analogy of a Jehovah's Witness follower refusing blood to demonstrate the“very steep ethical hill”arguments must climb for ethically acceptable mandated treatment for drug-addicted individuals on grounds it will beneﬁt the individual and society. In the case of a 406T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 Jehovah Witness, refusal of a blood transfusion is binding“even if it means that the person leaves behind many children without a parent” (p. 117). Caplan clearly highlights the fundamental importance of respecting individuals' autonomy, but points to one means through which mandatory treatment might not conﬂict with autonomy (& human dignity): when mandatory treatment is used to create or enable the capacity for, and exercise of, autonomy. He argues that addicts may not have the full capacity to act autonomously because their addiction “coerces their behavior”(p. 118). Accordingly, he asserts that it is possible to justify forced treatment for the timeframe required to restore an individual's capacity for autonomy. While we acknowledge the plausibility of this argument, ethically much hangs on demonstrating that a person's capacity for autonomous functioning is so impaired that he or she is unable to effectively deliberate about what options to pursue. In other word, asking the following question: Is the treatment and associated research really in the interests of the offender (equality, dignity) and is my answer generalizable to similar situations involving non-offenders (universality)?

3.2.4. Cultural differences Fourth, modern societies are frequently comprised of ethnically diverse groups and this fact is reﬂected in the ethic mix of prison populations. A glance at the proportion of indigenous prisoners around the world indicates that they are over represented in prison rates. To illustrate, as at 21 March 2010, Maori offenders comprised 50.68% of the New Zealand prison population (Department of Corrections, personal communication, 24 March 2010) compared to 14.6% of the general population (Statistics New Zealand). Irrespective of the actual number of offenders from minority ethic groups the key point is that such offenders are currently incarcerated or placed on community orders and that there are cultural differences between such offenders and those belonging to the majority ethnic group.

These differences revolve around issues such as informed consent or conﬁdentiality that researchers need to be aware of. A key issue when examining the implications of cultural diversity is to ascertain the particular level at which it is evident within a country as well as contemplating differences between cultures located in different countries (Parekh, 2006). It is typically the cultural challenges within a given society that are likely to prove most taxing for researchers and practitioners in the course of their day-to-day duties. Under such circumstances, it is imperative to keep in mind the common needs and interests of all individuals and their equal status and value.

The fact of cultural diversity creates problems for researchers. For example, in some cultures the right to make decisions extends to family and even group members and therefore an offender is unlikely to want to make a decision about research participation unless he or she can consult with family and cultural representatives (Ntseane, 2009). In many prison environments this is likely to prove problematic and there is a danger that offenders' wishes may get ignored and they may consent to research because they feel pressured or undermined. Similar points hold for issues related to privacy of research data and its uses (see above). The problem for researchers is that existing codes of ethics do not really help to address such culturally based consent, competency, or privacy questions.

A third ethical issue emerging from the social and cultural under- pinnings of forensic and correctional research concerns what constitutes warranted knowledge, that is, knowledge claims that meet what are considered to be acceptable standards. It has been argued by some researchers that western conceptions of knowledge are heavily weighted towards science based inquiry methods, and experimental, and quasi-experimental approaches at that (Denzin & Lincoln, 2005; Mertens & Ginsberg, 2009). The worry from the viewpoint of these researchers is that claims to knowledge from other cultures that may have merit could be too easily dismissed because they do not conform in terms of procedures or substance to received standards. In western conceptions of science, typically objectivity, numerical data, the use ofimpartial methods, explicit and rigorous methods for analysing observable data are all prized and sought after. Whereas subjective, experientially based data are often seen as overly vulnerable to personal biases and, therefore, unacceptable as a basis for knowledge claims. In practice, this may mean that forensic and correctional researchers lose the opportunity for capitalizing on ways of understanding the world that have value and, therefore, run the risk of wasting available epistemic resources. Furthermore, research participants may feel that their world views and cognitive strategies are unfairly neglected or even ridiculed.

Given that some treatment programs are founded on assumptions about what constitutes effective problem solving and evaluation practices, this is a concern (Ward & Nee, 2009). For example, cognitive restructuring in sex offender treatment rests upon the notions of objectivity, hypothesis testing, validity, and so on (Marshall, Fernandez, Marshall, & Serran, 2006). The challenge for researchers is to make reasonable efforts to appreciate methods of obtaining knowledge that resonate with research participants without sacriﬁcing the advantages of established scientiﬁc procedures.

3.2.5. Virtues and vices Aﬁfth ethical problem may occur in forensic and correctional settings where researchers may also function as practitioners and hence have dual roles (Munthe, Radovic, & Anckarsater, 2010). The difﬁculty in these types of situations is that allegiances formed in therapeutic relationships may exert subtle pressures on offenders to agree to participate in research projects without appropriate consultation and reﬂection. The fact that offenders have found therapy valuable, and taking into account the bond formed with the clinician-researcher, could combine to persuade him or her that involvement in research is a good option. The worry is that it is arguably never possible when dual relationships exist for offenders to be able to carefully consider the pros and cons of any research involvement. A related ethical problem is that of inducements for agreeing to take part in research projects in settings where“good” behavior may be viewed as contributing to early release or favorable treatment by correctional ofﬁcials. Other types of inducements include offering participants attractive prison jobs, offering opportu- nities to obtain treatment for psychological problems in treatment evaluation research, when the chances of them having access to treatment otherwise would be close to zero (Overholser, 1987). These types of inducement may be especially attractive in facilities that have overly harsh and unsafe conditions. One way to avoid behaving unethically is to make sure that any actions engaged in reﬂect the equal moral status of the offenders involved and also generalize to similar situations involving non-offenders. In other words, the key question is would you act in the same way with participants who are not offenders and do your actions reﬂect their basic moral equality and dignity?

The question of what character traits are desired (virtues) or not desired (vices) in researchers is partly a question of the kinds of skill needed to engage effectively in research tasks, and the broader social aims underpinning forensic and correctional research more generally. In research contexts there are knowledge related virtues such as accuracy, carefulness, orﬂexibility as well as ethical virtues like consideration, sensitivity, and empathy. In some research designs the ethical virtues play a dual role and beside guiding actions that demonstrate appropriate concern may also help to ensure that the data gathered are reliable and valid. For example, a sensitive empathic researcher using qualitative methods is more able to establish a trusting relationship with his or her participants when interviewing them and thereby obtain better quality data (Lincoln, 2009). In fact, any research project that involves interaction with offenders or interviews requires researchers to be able to establish a reasonably good working relationship with them. The type of qualities required depends on the speciﬁc investigatory tasks and these in turn reﬂect the research and broader community's core values and research priorities. In a society dominated by a conservative social 407 T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 justice agenda there may be little if any stress placed on offender well- being and therefore the virtues required to conduct research projects constrained by these assumptions will be different than those necessary in a culture dominated by broader humanistic commitments. The ethical problem is that the training researchers receive in formal academic training and more informal methods of knowledge transmission in conferences and workshops is likely to heavily inﬂuence their skill set and the kinds of research activities they are competent to engage in.

Thus, researchers who are trained under the auspices of risk management assessment and intervention approaches may lack the orientating values and actions necessary to view and relate to offenders as fellow human beings (Denny, 2005; McNeill, 2004). Aside from the fact that a whole raft of research possibilities (e.g., transformative justice–Ginsberg & Mertens, 2009) will be excluded for such researchers, there is also the danger that they will promulgate unethical ways of regarding and treating offenders. For example, it might simply be assumed that research should also proceed on a global cost–beneﬁt analysis and that offender concerns can always be traded off in favor of community wishes and preferences. In our view, this is an ethically, precarious position.

3.2.6. Assumptions about disorders Aﬁnal ethical difﬁculty evident concerns assumptions made about disorders and offender treatability (Howells & Day, 2007). Every research project is necessarily underpinned by core assumptions about the nature of the problems and participants being investigated.

In the forensic and correctional arena there is the added complication of a strong ethical overlay, that is, offenders have frequently inﬂicted serious harm on members of the community and are being punished as well as researched or treated. In light of what is at stake for both offenders and the wider community, it is incumbent on researchers and practitioners to examine their ethical and epistemic assumptions about offenders and their problems carefully (Ward & Maruna, 2007). For example, psychopathy has been long regarded by researchers to be virtually untreatable and as a consequence of this viewpoint, there has been little sustained research into the development and evaluation of suitable programs for individuals diagnosed as psychopathic (Howells & Day, 2007). The worry is that these assumptions may be incorrect or premature, and may instead simply express moral condemnation towards habitually violent and callous individuals masquerading as informed scientiﬁc opinion. Similar concerns could be raised about other types of offenders such as sex offenders with high levels of sexual deviancy or individuals with other types of personality disorder. In our view this point highlights the dependence (but not reduction to!) of scientiﬁc research on broader social and cultural variables and reminds researchers to always critically review the aims and reasons for their projects prior to proceeding.

4. Conclusions It does not seem possible to respond to the ethical issues outlined above by simply relying on professional codes of ethics (Ward & Syversen, 2009). For example, ethical codes are unlikely to help researchers respond to concerns about data presentation or how to balance the interests of all possible stakeholders when designing and implementing research in forensic and correctional settings. On the other hand, the concept of dignity reﬂected in the ability of participants to deliberate and make their own choices, and to have access to social and psychological resources that enable them to do so, can provide such guidance. Ethical principles that underpin standards in many ethical codes like those of justice, beneﬁcence, integrity, justice, and autonomy, can be conceptually related to the concept of dignity (Ward, Gannon, & Vess, 2009; Ward & Syversen, 2009). This is possible because all persons are presumed to possess intrinsic worth and, therefore, should be treated in ways that reﬂect this status. That is, their core interests ought to be considered along with those of others (justice); their ownviewpoints and reasoning powers should be respected when it comes to decisions that affect them (autonomy); their standard of living ought to reach a certain threshold and ensure they have achieved a certain level of well-being (beneﬁcence); and so on. The ethical principles that underpin ethical codes arguably are in turn justiﬁed by the high rank or worthiness of human beings, that is, their inherent dignity. This equality of moral status and high value, therefore, ought to be reﬂected in the way people are treated in the different arenas of their lives.

Ethical thinking in forensic and correctional research contexts is complex and demanding. The reason for this complexity and level of difﬁculty resides in the dual nature of forensic and correctional research and practice; it is both normative and descriptive. A culture's deep assumptions about moral accountability, punishment, the nature of moral agency, social justice and so on will inevitably color the way researchers approach their investigatory tasks. An advantage of the ethical framework outlined in this paper is that because it revolves around the concept of dignity and also on the interdependence of human beings the principles and standards associated with ethical codes appear less arbitrary and researchers are less likely to overlook subtle ethical problems. As researchers we have an obligation to keep in mind our responsibilities to all those affected by crime and never forget that the fruits of such work can have lasting and potentially devastating consequences for all. In our view, ethical forensic and correctional research is more likely to emerge out of the virtues of compassion and an appreciation of the value of all human beings.

References Adshead, G. & Brown, C. (Eds.). (2003).Ethical issues in forensic mental health research.

London, UK: Jessica Kingsley.

Andrews, D. A., & Bonta, J. (2003).The psychology of criminal conduct, 3ed. Cincinnati, OH: Anderson.

Appelbaum, K. L. (2008). Correctional mental health research: Opportunities and barriers.Journal of Correctional Health Care,14, 269 277.

Balashov, Y., & Rosenberg, A. (Eds.). (2002).Philosophy of science: Contemporary readings. London, UK: Routledge.

Becker, L. C. (1986).Reciprocity.Chicago, IL: University of Chicago Press.

Bersoff, D. N., Faust, D., Sales, B. D., Shuman, D. W., Heilbrun, K., Haas, L. J., et al. (2003).

Forensic settings. In D. N. Bersoff (Ed.),Ethical conﬂicts in psychology. Washington, DC: American Psychological Association.

Beyleveld, D., & Brownsword, R. (2001).Human dignity in bioethics and law.New York, NY: Oxford University Press.

Bonnie, R. J. (2006). Judicially mandated naltrexone use by criminal offenders: A legal analysis.Journal of Substance Abuse Treatment,31, 121 127.

Brown, B. L., & Hedges, D. (2009). Use and misuse of quantitative methods. In D. M.

Mertens & P. E. Ginsberg (Eds.),The handbook of social research ethics (pp. 373 385). Thousand Oaks, CA: Sage Publications.

Bush, S. S., Connell, M. A., & Denny, R. L. (2006).Ethical practice in forensic psychology: A systematic model for decision making.Washington, DC: American Psychological Association.

Caplan, A. L. (2006). Ethical issues surrounding forced, mandated, or coerced treatment.

Journal of Substance Abuse Treatment,31, 117 120.

Churchill, R. P. (2006).Human rights and global diversity.Upper Saddle River, NJ:

Pearson Prentice Hall.

Chwang, E. (2010). Against risk-beneﬁt review of prisoner research.Bioethics,24, 14 22.

Clark, A. (2008).Supersizing the mind: Embodiment, action, and cognitive extension.New York: Oxford University Press.

Cooper, D. E. (2004).Ethics for professionals in a multicultural world.Upper Saddle River, NJ: Pearson Prentice Hall.

Craig, L. A., Beech, A. R., & Harkins, L. (2009). The predictive accuracy of risk factors and frameworks. In A. R. Beech, L. A. Craig, & K. D. Browne (Eds.),Assessment and treatment of sex offenders(pp. 53 74). Oxford, UK: Wiley-Blackwell.

Crombie, I. K., & Davis, H. T. (2009). What is meta-analysis? Evidence-based medicine, 2nd edit. Available electronically fromhttp://www.whatisseries.co.uk/whatis/ Day, A., Tucker, K., & Howells, K. (2004). Coerced offender rehabilitation—A defensible practice?Psychology, Crime & Law,10, 259 269.

Day, A., & Ward, T. (2010). Offender rehabilitation as a value-laden process.International Journal of Offender Therapy and Comparative Criminology,54,289 306.

Denny, D. (2005).Risk and society.London, UK: Sage.

Denzin, N. K., & Lincoln, Y. S. (2005).The Sage handbook of qualitative research, 3rd ed.

Thousand Oaks, CA: Sage Publications.

Douglas, K. S., & Skeem, J. L. (2005). Violence risk assessment: Getting speciﬁc about being dynamic.Psychology, Public Policy, and Law,11, 347 383.

Driver, J. (2006).Ethics: The fundamentals.Oxford, UK: Blackwell Publishing.

Duff, R. A. (2001).Punishment, communication, and community.New York, NY: Oxford University Press. 408T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409 Duwe, G., Donnay, W., & Tewksbury, R. (2008). Does residential proximity matter? A geographic analysis of sex offense recidivism.Criminal Justice and Behavior,35, 484 504.

Ginsberg, P. E., & Mertens, D. M. (2009). Frontiers in social research ethics: Fertole ground for evolution. In D. M. Mertens & P. E. Ginsberg (Eds.),The handbook of social research ethics(pp. 580 613). Thousand Oaks, CA: Sage Publications.

Haag, A. M. (2006). Ethical dilemmas faced by correctional psychologists in Canada.

Criminal Justice and Behavior,33,93 109.

Hanson, R. K., Bourgon, G., Helmus, L., & Hodgson (2009). The principles of effective correctional treatment also apply to sexual offenders: A meta-analysis.Criminal Justice and Behavior,36, 865 891.

Hoffman, S. (2000). Beneﬁcial and unusual punishment: An argument in support of prisoner participation in clinical trials.Indiana Law Review,33, 475 513.

Howells, K., & Day, A. (2007). Readiness for treatment in high risk offenders with personality disorders.Psychology, Crime & Law. Special Issue: Personality Disorder and Offending,13,47 56.

Hudson, S. M., Ward, T., & McCormack, J. (1999). Offence pathways in sexual offenders.

Journal of Interpersonal Violence,14, 779 798.

Institute of Medicine (2006).Ethical considerations for research involving prisoners.

Washington, DC: National Academies Press.

Kitchener, K. S. (2000).Foundations of clinical practice, research, and teaching in psychology.Mahwah, NJ: Lawrence Erlbaum.

Kitchener, K. S., & Kitchener, R. F. (2009). Social science research ethics: Historical and philosophical issues. In D. M. Mertens & P. E. Ginsberg (Eds.),The handbook of social research ethics(pp. 5 22). Thousand Oaks, CA: Sage Publications.

Koocher, G. P., & Keith-Spiegel, P. (2009).Ethics in psychology: Professional standards and cases, 3rd ed. New York: NY Oxford University Press.

LaFrance, J., & Bull, C. C. (2009). Researching ourselves back to life: Taking control of the research agenda in indian country. In D. M. Mertens & P. E. Ginsberg (Eds.),The handbook of social research ethics(pp. 135 149). Thousand Oaks, CA: Sage Publications.

Langlands, R. L., Ward, T., & Gilchrist, E. (2009). Applying the god lives model to male perpetrators of domestic violence.Behaviour Change,26, 113 129.

Lavin, M. (2004). Ethical issues in forensic psychology. In W. T. O'Donohue & E.

Levensky (Eds.),Handbook of forensic psychology: Resources for mental health and legal professionals. New York, NY: Elsevier Science.

Laws, D. R., & Ward, T. (in press).Desistance from sexual offending: Alternatives to throwing away the key.New York, NY: Guilford Press.

Lazarus, L. (2004).Contrasting prisoners' rights: A comparative examination of England and Germany.New York, NY: Oxford University Press.

Li, A. (2006).Ethics, human rights, and culture.Basingstoke, UK: Palgrave MacMillian.

Lippke, R. L. (2002). Toward a theory of prisoners' rights.Ratio Juris,15, 122 145.

Lincoln, Y. S. (2009). Ethical practices in qualitative research. In D. M. Mertens & P. E.

Ginsberg (Eds.),The handbook of social research ethics(pp. 150 169). Thousand Oaks, CA: Sage Publications.

Major, B., & O'Brien, L. T. (2005). The social psychology of stigma.Annual Review of Psychology,56, 393 421.

Mark, M. M., & Gamble, C. (2009). Experiments, quasi-experiments, and ethics. In D. M.

Mertens & P. E. Ginsberg (Eds.),The handbook of social research ethics (pp. 198 213). Thousand Oaks, CA: Sage Publications.

Marlowe, D. B. (2006). Depot naltrexone in lieu of incarceration: A behavioral analysis of coerced treatment for addicted offenders.Journal of Substance Abuse Treatment, 31, 131 139.

Marshall, W. L., Fernandez, Y. M., Marshall, L. E., & Serran, G. A. (Eds.). (2006).Sexual offender treatment: Controversial issues. Chichester, UK: Wiley.

Marshall, W. L., & Marshall, L. E. (2007). The utility of the random controlled trial for evaluating sexual offender treatment: The gold standard or an inappropriate strategy?Sexual Abuse: A Journal of Research and Treatment, 19,175-191-260.

Marshall, W. L., & Marshall, L. E. (2008). Good clinical practice and the evaluation of treatment: A reply to Seto et al.Sexual Abuse: A Journal of Research and Treatment, 20, 256 260.

McCoy, K., & Fremouw, W. (2010). The relationship between negative affect and sexual offending: A critical review.Clinical Psychology Review,30, 317 325.

McNeill, F. (2004). Desistance, rehabilitation and correctionalism: Developments and prospects in Scotland.The Howard JournaI,43, 420 436.

McSweeney, T., Stevens, A., Hunt, N., & Turnbull, P. J. (2007). Twisting arms or a helping hand? Assessing the impact of‘coerced’and comparable‘voluntary’drug treatment options.British Journal of Criminology,47, 470 490.

Mertens, D. M. & Ginsberg, P. E. (Eds.). (2009).The handbook of social research ethics.

Thousand Oaks, CA: Sage Publications.

Miller, D. (2003).Principles of social justice.Oxford, UK: Oxford University Press.

Miller, R. D. (1998). Forced administration of sex-drive reducing medications to sex offenders: Treatment or punishment.Psychology, Public Policy, and Law. Special Issue: Sex Offenders: Scientiﬁc, Legal, and Policy Perspectives,4, 175 199.

Morris, M. (2007).Evaluation ethics for best practice.New York, NY: Guilford Press.Munthe, C., Radovic, S., & Anckarsater, H. (2010). Ethical issues in forensic psychiatric research on mentally disordered offenders.Bioethics,24,35 44.

Ntseane, P. G. (2009). The ethics of the researcher-subject relationship. In D. M. Mertens & P. E. Ginsberg (Eds.)The handbook of social research ethics(pp 295- 169-307).

Thousand Oaks, CA: Sage Publications.

Nussbaum, M. (2006).Frontiers of justice: Disability, nationality, species-membership.

Cambridge, UK: The Belknap Press of Harvard University Press.

O'Donohue, W. T., & Ferguson, K. (Eds.). (2003).Handbook of professional ethics for psychologists: Issues, questions, and controversies. Thousand Oaks, CA: Sage Publications.

Overholser, J. (1987). Ethical issues in prison research: A risk/beneﬁt analysis.

Behavioral Sciences & the Law,5, 187 202.

Parekh, B. (2006).Rethinking multiculturalism: Cultural diversity and political theory, second edition. Basingstoke: Palgrave MacMillan.

Perez, L. M., & Treadwell, H. M. (2009). Determining what we stand for will guide what we do: Community priorities, ethical research paradigms, and research with vulnerable populations.American Journal of Public Health,99, 201 204.

Seto, M.,Marques, J. K., Harris et al., (2007). Good science and progress in sex offender treatment are intertwined: A response to Marshall and Marshall (2007).Sexual Abuse: A Journal of Research and Treatment, 19,175–191–260.

Smedslund, G., Dalsbø, T. K., Steiro, A. K., Winsvold, A., & Clench-Aas, J. (2007). Cognitive behavioural therapy for men who physically abuse their female partner.Cochrane Database of Systematic Reviews,3Art. No.: CD006048. DOI: 10.1002/14651858.

CD006048.pub2.

Statistics New Zealand (2006). 2006 Census Data. Retrieved 24 March 2010, fromhttp:// www.stats.govt.nz/Census/2006CensusHomePage/QuickStats/quickstats-about-a- subject/maori/maori-ethnic-population-te-momo-iwi-maori.aspx Sterelny, K. (2003).Thought in a hostile world: The evolution of human cognition.Oxford, UK: Blackwell Publishing.

Sulmasy, D. P. (2007). Human dignity and human worth. In J. Malpas, & N. Lickiss (Eds.), Perspectives on human dignity: A conversation(pp. 9 18). Dordrecht, The Netherlands: Springer.

Symonette, H. (2009). Cultivating self as responsive instrument: Working the boundaries and borderlands for ethical border crossings. In D. M. Mertens & P. E.

Ginsberg (Eds.),The handbook of social research ethics(pp. 279 294). Thousand Oaks, CA: Sage Publications.

Regehr, C., Edwardh, M., & Bradford, J. (2000). Research ethics and forensic patients.The Canadian Journal of Psychiatry/La Revue Canadienne de Psychiatrie ,45, 892 898.

Travis, J. (2005).But they all come back: Facing the challenges of prisoner Reentry.

Washington, DC: The Urban Institute Press.

United Nations (1948). Universal declaration of human rights. In J. P. Martin & R.

Rangaswamy (Eds.),Twenty-ﬁve human rights documents. NY: Columbia University for the Study of Human Rights (1984).

Vess, J. (2008). Sex offender risk assessment: Consideration of human rights in community protection legislation.Legal and Criminological Psychology,13, 245 256.

Vess, J. (2009). Fear and loathing in public policy: Ethical issues in laws for sex offenders.Aggression and Violent Behavior,14, 264 272.

Ward, T., & Birgden, A. (2007). Human rights and correctional clinical practice.

Aggression and Violent Behavior,12, 628 643.

Ward, T., Gannon, T., & Vess, J. (2009). Human rights and ethical principles and standards in forensic psychology.International Journal of Offender Therapy and Comparative Criminology,53, 126 144.

Ward, T., & Langlands, R. (2009). Repairing the rupture: Restorative justice and offender rehabilitation.Aggression and Violent Behavior,14, 205 214.

Ward, T., & Nee, C. (2009). Surfaces and depths: Evaluating the theoretical assumptions of the cognitive skills programmes.Psychology, Crime & Law,15, 165 182.

Ward, T., & Maruna, S. (2007).Rehabilitation: Beyond the risk paradigm.London:

Routledge.

Ward, T., Polaschek, D. L. L., & Beech, A. R. (2006).Theories of sexual offending.

Chichester, UK: Wiley.

Ward, T., & Salmon, K. (2009). The ethics of punishment: Correctional practice implications.Aggression and Violent Behavior,14, 239 247.

Ward, T., & Syversen, K. (2009). Vulnerable agency and human dignity: An ethical framework for forensic practice.Aggression and Violent Behavior,14,94 105.

Ward, T., Yates, P., & Long, C. (2006).The Self-regulation Model of the offense and relapse process: Volume 2: Treatment.Victoria, BC: Paciﬁc Psychological Assessment Corporation.

Yates, P. M., & Kingston, D. A. (2006). Pathways to sexual offending: Relationship to static and dynamic risk among treated sexual offenders.Sexual Abuse: A Journal of Research and Treatment,18, 259 270.

Zandbergen, P, Levenson, J. S., & Hart, T. (2010). Residential proximity to schools and daycares: An empirical analysis of sex offense recidivism.Criminal Justice and 1393 Behavior,37, 482 502.409 T. Ward, G. Willis / Aggression and Violent Behavior 15 (2010) 399–409