The Theoretical Perspective Constructivism And Naturalistic Inquiry Nursing Essay

Introduction

A qualitative descriptive design was deemed most appropriate to enable exploration and description of the court liaison nurse role as it has developed in the New Zealand context. Sandelowski suggests that "Qualitative descriptive study is the method of choice when straight description of phenomena is desired" (2000b, p. 334). Applicable to answering the questions this research posed Qualitative descriptive designs usually include a diverse mixture of "sampling and data collection, analysis, and re-presentational techniques" (Sandelowski, 2000b, p.337). Exploratory research seeks to respond to questions such as, "what is the full nature of the phenomenon, what is really going on here and what is the process by which the phenomenon evolves or is experienced?" (Polit & Beck, 2006, p.20). These features of qualitative descriptive methodology provided the opportunity capture a broad understanding of this role and permitted the use of mixed methods to gather the data.

Qualitative descriptive designs draw from tenets of Naturalistic Inquiry which is embedded in the constructivist paradigm (Lincolin & Guba, 1989). Constructivism, as understood through the work of Egon Guba and Yvonne Lincolin (Lincolin & Guba 1985; 1989; Guba 1990) aims to carry out research in natural settings seeking to examine and understand the variety of constructions that people have within their social world. The inquiry aim of understanding and reconstruction (Denzin & Lincolin, 2000) fit with the research aim. This chapter will outline the underpinning assumptions of Naturalistic Inquiry and the Qualitative descriptive methodology framework. A description of the research methods and processes follows.

Theoretical perspective - Constructivism and naturalistic inquiry

Paradigm is a term that has generated confusion in the literature (Weaver & Olson 2006; Guba, 1990) therefore it is helpful to consider what is meant by the term paradigm at the outset. Guba defines paradigm as a "basic set of beliefs that guides action, whether of the everyday garden variety or action taken in connection with disciplined enquiry" (1990, p.17). The paradigm provides a framework for the researcher and assists with making order out of chaos (Giddings & Grant, 2002). It specifies and permeates every act associated with the inquiry to the degree "such that any consideration even remotely attached to the inquiry process demands rethinking to bring decision into line with the worldview embodied in the paradigm itself" (Lincolin in Guba, the paradigm dialogue, 1990, p.81)

In considering the qualitative paradigm Munhall describes philosophical underpinnings for qualitative research methods as reflecting "beliefs, values and assumptions about the nature of human beings, the nature of the environment and the interaction between the two (2007, p.99). Broadly speaking qualitative research approaches are described as inductive compared to deductive quantitative paradigms, involving the use of different methods of discovering and different ways of reporting the truth of these discoveries (Munhall, 2007). Lincolin and Guba (1985) proposed naturalistic inquiry as an alternative constructivist methodology to challenge the analytical/empiricist thought of the positivistic paradigm. According to Guba because the evolution of human beings involves capacity for interpretation and ability to construct reality study of the human world should be conducted differently to that of the natural or physical world (1990). So in that case human perception cannot be deemed conclusively ‘real’ it is ‘made up’ and shaped by the individual’s interaction with society and culture in the minds of the constructivist (Patton, 2002). Therefore within the context of naturalistic inquiry, the world and reality are seen as human constructs that cannot be considered and appreciated in isolation from their context (Patton, 2002; Lincolin and Guba, 1985).

From this perspective reality is understood as being constructed in the mind of individuals. People understand and experience their world differently according to their perceptions, expectations, values, culture, and relationships. Viewing and understanding of the world through a constructivist lens becomes clear when the axioms are expanded.

The ontological axiom

Crotty neatly defines Ontology as "the study of being" (1998, p.10). Dealing with the nature of existence and the structure of reality Ontology provides a particular way of viewing and understanding the world that with epistemology underpins each theoretical perspective (Crotty, 1998). Ontology adds the ‘what is’ view while epistemology adds ‘what it means to know’ to the theoretical perspective (Crotty, 1998).

If we consider the concept of reality in constructivism we find that within a naturalistic inquiry reality is a social and therefore multiple construction (Guba & Lincolin, 1994), Relativism provides recognition that an individual’s understanding of the world may be quite different from one another (Crotty, 1998). Relativism also considers understanding of the world may be impossible to differentiate from another’s understanding of the world therefore creating many and varied ways of knowing, sets of meaning and separate realities (Crotty, 1988). Therefore relativist ontology is supported in constructivism. It follows that constructivist researchers believe that social reality exists as individual’s experience it and as it has meaning for them. This means that within in any research there will always be many different interpretations that can be made.

Authors suggest that looking for generalizations is not meaningful when studying human behaviour because it is impossible "to imagine that all human activity is completely determined by one universal set of relationships" (Lincolin & Guba, 1985; Guba & Lincolin 1989, p.94). Therefore it follows that generalising the findings of research is one of the aims of the research carried out in a scientific realm. Generalisations are defined as "truth statements of enduring value that are context-free" (Lincolin & Guba 1982, p.238), and timeless. Qualitative researchers dispute the view that generalisations are possible due to the fact that the research is undertaken by human beings and inextricably linked to a particular context (Crabtree & Miller, 1999, Lincolin & Guba, 1985). Suggesting instead the uniqueness of the settings being studied is particularly valued. As an implicit part of constructivist inquiry the researcher reports and makes sense of the uniqueness discovered in each new setting under study.

Likewise the concept of causality has been criticised with many researchers stating that it does not account for what is occurring in complex social situations (Heron, 1981, Lincolin & Guba 1985). Causality is rejected with the argument that because "all entities are in a state of mutual simultaneous shaping.... it is impossible to distinguish causes from effects" (Lincolin and Guba, 1985, p.37). Mutual simultaneous shaping refers to the process of that all things influence each other and that "causes and effects are inextricably intertwined" (Guba & Lincolin, 1982, p.242). The belief is that many different outcomes could be produced by the same cause, and "explanations are at best "here-and-now" accounts that represent a "photographic slice of life" (Guba & Lincolin 1989, p.98).

This orientation to research differs markedly from traditional modes of inquiry where a single or a minimum number of correct viewpoints is sought. Because the ontological belief is that there are socially constructed multiple realities finding a singular, verifiable reality and truth is not possible (Patton, 2002). Lincolin and Guba describe that view that truth "emerges as a composite picture of how people think" (1985, p.80). Consequently in order to carry out the research in a holistic manner as suggested by Lincolin and Guba (1985) the researcher cannot discard divergent or conflicting constructions of reality while trying to pull together a level of understanding. It follows that inclusion of each person’s experiences and the context in which they occurred are considered and incorporated into the emerging construction(s) in this mode of research (Appleton & King, 2002).

The epistemological axiom

Already established is the dynamic relativist ontology the next question to consider is where constructivism sits epistemologically? Epistemology is "the branch of philosophy concerned with the nature of knowledge, its possibility, scope and general basis" (Hamlyn, 1995, p. 245). In other words epistemology is concerned with the manner of how we study the world and how knowledge of the world is obtained (Patton, 2002). Epistemology or the theory of knowledge is driven by three main questions; what is knowledge, what can we know and, how do we know what we know? (Greco, 1999). The questions that epistemological concerns pose are important in orienting: the research approach, the underlying assumptions, and in how the research outcomes are presented (Crotty, 1998).

Discussion regarding the distinction between ontology and epistemology becomes unnecessary in this paradigm. Study findings are "the creation of the process of interaction" (Guba 1990, p.27) between the researcher and the study participants therefore Ontology and epistemology are interwoven and it becomes impossible to consider one without the other (Guba, 1990). Guba and Lincolin explain that "it is precisely their interaction that creates the data which will emerge from the inquiry" (1989, p.88). Creating the meanings or findings between researcher and participants requires a transactional/subjectivist dynamic approach when considering the phenomena being researched compared with the subject/object dualism in the positivistic paradigm (Guba & Lincolin, 1994). Not only does there have to be interaction but it must be meaningful interaction which can only take place with the cooperation of the participants. Therefore the constructivist position epistemologically in relation to what is known, what can be known and how do we know, is the belief that people construct meanings about the world through interactive experiences with others.

If a subjective interactive approach underpins the research then the place of values in the research must be considered. Values are seen as essential in knowledge creation in the constructivist paradigm. In the context of inquiry values are "the moral principles or accepted standards of a person or group" (Collins, 1993, p.1310). Constructivist inquirers argue that facts are both "value-laden" and "theory laden" (Guba & Lincolin 1989, p.105). Four ways are proposed by Guba and Lincolin (1982) in which values influence constructivist inquiry. Initially the researcher’s own beliefs and values will influence the area to be investigated. Secondly, the particular theoretical paradigm and methodology selected to guide the overall inquiry process will have roots in both "assumptions and value position(s)" (Guba & Lincolin 1982, p.243). Thirdly, the research will be influenced by the presence of values ingrained in the study setting. Fourthly, the beliefs of all groups represented in a study should be recognized as influential and "deserve equal consideration in shaping constructions" (Guba & Lincolin 1994, p.114). The impact of values and the role they play in the research are encapsulated in the choice of methodology and the research design. The "mutual shaping" referred to by Lincolin & Guba (1985) is influenced by both the researcher’s and respondents’ own value systems.

A criticism of this perspective is the focus on the importance of the individual view rather than the impact of society as a whole (Pawson & Tilley, 1997) However, this critique fails to understand constructivism does recognize the complex interplay that helps to form, develop and alter an individual’s constructions of any phenomena (Lincolin & Guba, 2000). Lincoln and Guba recognise that "multiple knowledges" can exist together and that a range of views may emerge during a naturalistic inquiry (1994, p.113). Likewise this focus of the individual’s perception and viewpoint in practical terms this makes it a challenge to achieve a single explanation for complex phenomena because of the belief of multiple intangible realities. Therefore the research will attempt to illustrate each substantial consensus viewpoint separately. Accordingly the researcher must accept participant’s beliefs and values may be very different from the researcher’s own.

The methodological axiom

Ideally in line with this research ontology, methodologically the researcher studies real- world situations as they unfold naturally instead of (manipulating research outcomes a priori) in seeking for cause and effect outcomes (Lincolin & Guba, 1985, p. ). Methods were chosen that would holistically capture realities and meanings.

The relationship between the researcher, the research, and the participant is important on several levels. The relationship with the participant and hence the construction of data and meanings is alluded to above. The relationship of the researcher with the research incorporates another layer of relationships. This is central to a naturalistic inquiry that the researcher is not seen as a neutral spectator in the research (Denzin, Smith, & Deemer). Lincolin and Guba (1985) refer to ‘human researchers as instruments’. As such researchers are recognised as bringing a responsive approach to the research and data through the use of effective interpersonal skills (Crabtree & Miller, 1999). Data collection procedures can be adapted where necessary through the use of creative and intuitive processes. Interaction continues with the process of data analysis when the researcher seeks convergent and divergent viewpoints and explanations for any discrepancies. Incorporated in the concept of ‘human researcher’s as instruments’ is recognition of and valuing the ‘tacit’ knowledge the researcher bring to the methodology (Lincolin and Guba, 1985).

Understandably then texts based on interviews and observations are mutual, contextual and value bound (Lincolin and Guba, 1985). For that reason, the research takes the position that multiple meanings and some degree of interpretation are always involved in the analysis of a text. Sandelowski and Barroso (2002) suggest because the data are "constructed entities" they cannot be separated from findings. Here, the so called raw data of qualitative research – e.g., the interview transcript and the field notes – are not pre- given, but rather are already interpretive products. As Wolcott (1994, pp4-5) observed, although "Everything has the potential to be data......nothing becomes data without the intervention of a researcher who takes notes – often makes note – of some things to the exclusion of others." Research findings in qualitative research are variously conceived as both a process and product in which the researcher is deeply and unavoidably implicated.

In a distinct move from the positivist paradigm the principle of "human researcher as instrument" incorporates the development phase of the research encouraging the inquirer to discuss topics or issues prior to their investigation with fellow practitioners (Lincolin & Guba, 1985). Carrying out this process assists with refining the inquirer’s thinking before contemplating access to the field (Lincolin and Guba, 1985).

The next section of the thesis will focus on the research methodology and design.

Research questions and aims

The research question, "What is the role of the mental health nurse in the criminal court in the New Zealand context?" sought information in order to understand and make visible contemporary nursing practice at the criminal justice mental health interface in New Zealand.

The following questions were developed regarding the CLN role.

1. What are their nursing backgrounds and professional experience?

2. What are their daily professional activities?

3. What ethical tensions do they experience and how do they manage them?

4. What are their education and training needs?

Methodology

The research methodology is chosen to answer the question "How should the researcher go about gathering the knowledge to answer the research question?" (Guba, 1990, p. 18). Qualitative descriptive methodology as outlined by Sandelowski (2000a, 2000b) fit these criteria and was deemed the appropriate methodology that through rich description would allow the nurses perspective to be heard and themes to emerge from the data. The aim of a naturalist inquiry is to understand peoples constructed meaning of truth and reality. These underpinning assumptions mean that the nurses’ voice is given precedence in the research.

Qualitative descriptive methodology provides the ability to utilise a mixture of sampling and data collection tools and re-presentation techniques to guide the research (Sandelowski, 2000b). Given the paucity of literature and information regarding this nursing role in the New Zealand context it was fitting that the methods applied for the initial foray into this area enabled broad exploration and descriptive elements to the approach as well as the ability to examine aspects of the nursing practice in depth.

Of particular relevance for this research Qualitative descriptive designs provide an existing framework for practice based research, which are particularly appropriate for nursing and policy development (Sandelowski, 2000b). Lincolin and Guba’s (1985) naturalistic inquiry paradigm, used in this study, depicts multiple constructed realities as limitless perspectives or worldviews that bring people’s understanding together. The court liaison nurses realities are multiple, constructed, interactive and inseparable in the realms of health, education and the criminal justice system. This necessitates understanding that this inquiry is value bound, as detailed by Lincolin and Guba (1985), and as such this methodology is described as a particularly useful approach for nursing practitioners who are considering entering their clinical area of expertise as a researcher. The choice of research topic and methodology was influenced by my past experience with and present concerns from knowing and working with CLNs, as well as a strong desire to make visible the paradigm of nursing in the criminal justice system.

An important feature in the decision to use this methodology related to the evolving nature of constructivist inquiry that supports the inquirer to discuss topics or issues for investigation with fellow practitioners in order to assist with clarifying the inquirer’s thinking before commencing the research and this means that initial research plans are not finite (Lincolin & Guba, 1985). Following this process as researcher assisted with refining the research proposal and continued throughout the study.

Similarly the purposive sampling strategy used in Qualitative descriptive designs (Sandelowski, 2000b, Patton, 2002) and working with the participants to construct an understanding of the key issues assisted with firming the researcher’s ideas about the progression of the study. Ongoing analysis enabled the researcher to interact with the data to "oil the wheels" (Appleton and King 1997) of thought and facilitated consideration of diverse interpretations.

Notwithstanding any clinical expertise demonstrating transparency and adherence to the principles of the research is vital if the research is to be considered worthy and useful.

Design Credibility

"Rigour" is a term associated with the positivist paradigm and way of thinking (Crabtree & Miller, 1999). Rigour is described as "the means by which we demonstrate integrity and competence, a way of demonstrating the legitimacy of the research process" (Tobin & Begley, 2004, p, 390). The overall thrust of this research is qualitative therefore the overall integrity of the research will be judged by mechanisms applicable to a study undertaken in the qualitative paradigm. Lincolin suggests that "In a constructivist inquiry, process is only one means of determining the utility, responsibility, and fidelity of the inquiry. Action and understanding were other components of the judgements regarding the goodness of any inquiry" (Lincolin, 1990). Therefore several steps inherent in qualitative descriptive methodology were applied throughout the research to demonstrate credibility. According to Sandelowski descriptive and interpretive validity demonstrate the research is true to it aims (2000b). Evidence of research credibility is provided throughout the thesis.

The aim of the research was to present a truthful and accurate account of the nurse’s perspectives (Sandelowski, 2000b). Most importantly to represent in a straight forward manner what the nurses considered to be the crucial aspects of their role, knowledge and needs. To achieve this end Koch and Harrington (1998) refer to the necessity of ongoing conversation between the inquirer, the nurses and the research to reach agreement. Sandelowski (2000b) contributes, that although it is unlikely that all the data within the inquiry will be reported on, what is described should represent a consensual view of the data. This is known as Descriptive validity, a valid description of events that most people (including researcher and participants) observing the same event would agree is accurate (Maxwell, 1992). Planning is an essential component in enhancing credibility of the research. The sequential nature of the research was designed to enhance credibility in initially eliciting key themes in the survey and building on these in the next two phases. The processes of seeking verification were established prior to the study and adhered to. The checking and verification processes in place, returning of transcripts to participants and monitoring of feedback were part of systems involved in maintaining a research audit trail.

It must also be recognised all decisions involve choice and inevitably interpretation on the part of the researcher (Sandelowski, 2000b). Interpretive validity takes descriptive validity to another level and refers to the valid description of the meaning participants found from the events or circumstances they are describing that those participants would agree is accurate (Maxwell, 1992).

Patton (2002, p.563) refers to triangulation methods as means of reducing systematic bias and distortion during data analyses phases thereby ensuring credibility. The research used triangulation in terms of member checking (triangulation by review by inquiry participants), independent analysis (triangulation with multiple analysts) and data from on phase of the research not only informing the next phase by also confirming data (methods triangulation).

Two phases of the research involved quantitative elements therefore rigour in respect of these elements was reviewed using internal validity and reliability as the key issues integral to rigor of quantitative research (LoBiondo-Wood and Haber, 2006). Verification strategies with respect to the tools were employed by seeking advice and assistance from experts in the relevant areas. The process of piloting and checking the tools can be found under the section headed Piloting of tools.

Internal validity is the degree to which a study reflects or assesses the specific concept that is under measurement, and can be assessed in terms of internal and external validity (LoBiondo-Wood and Haber, 2006). Internal validity is considered met when the design of the study is congruent with the aims of the research. The survey questionnaire was designed to capture a broad understanding of this cohort of nurses and functions of their role this was congruent with the aims of the research. The data collection instruments used in phase 3 further developed the aims of the study by recording in detail the functions and daily nursing practice. Therefore the tools met criteria for internal validity.

External validity refers to the extent with which the results of the study are generalisable or transferable (Polit et al, 2001). External validity is not applicable to this research. The research was positioned in an overall framework of a Naturalistic Inquiry therefore generalisability of the study of this unique nursing role was never the intention.

Auditability is a concept used in qualitative research to support dependability or reliability (the quantitative equivalent.) Auditability refers to the circumstances by which one researcher can follow the analysis pattern of another (Guba & Lincolin, 1981). This means another researcher should be able to arrive at similar conclusions using the available documents and raw data. In other words an audit trail was laid down (Polit, Beck, & Hungler, 2001). All aspects of the research were described in the methodology section and all tools are listed in Appendix....

Researcher values

Because findings in qualitative studies are partly composed of the knowledge, beliefs, and everything the researchers are and bring into the research (Sandelowski & Barroso, 1998) acknowledgement of the researchers values based on personal and professional experience is central to ensuring self awareness of prior opinions and biases. The researcher practised in the court liaison nurse role from 2002 to 2008 and continues to have a professional interest in the court liaison nurse role.

The underpinning assumptions for qualitative descriptive methodology fit with the researcher’s views regarding reality in that the researcher believes in multiple experientially based constructed realities. It was central to the researcher’s values that the clinical skills of the nurses were acknowledged and valued in this study. It is not possible to eliminate bias but a reflexive process ensured transparency. Reflexivity is described in Munhall as the "process by which researchers recognise that they are an integral part of the research process and vice versa" (2007, p. 318). Reflective processes were established to ensure a thorough approach to the research. These included:

Research supervision

The supervisory role was crucial in ensuring the research process was systematic and logical and that an adequate audit trail congruent with the methodology was being laid down. A process of regular reflexive supervision took place to assist with transparency – one supervisor with clinical expertise in forensic mental health and research– questioned my motivation for certain aspects of the study at times, guided me back to the literature, and encouraged me to provide justification for my thinking ensuring there were no seemingly hidden agendas. The researcher was able to call on the expertise of other supervisors; methodological expertise and extensive knowledge of the law/mental health interface. Any pre-existing notions I came with were challenged. Notwithstanding the expertise the researcher was able to draw on from the supervisors the personal expertise that the participants bought to the research also contributed to the reflexive process.

In addition to regular individual supervision sessions, a yearly presentation of progress with fellow DHSc candidates and their supervisors provided a safe but challenging forum to discuss work and defend the research design. Field notes were kept regarding interviews and discussed with supervisors during debriefing post interviews. Reflective notes were kept by the researcher throughout the study and discussed with supervisors.

Research presentation

Presentation to groups with familiar with the context (nursing colleagues and court liaison nurses) allowed for reviewing and checking the credibility of the research as well as critical examination of the researchers thinking and analysis.

The use of qualitative descriptive methodology provided the means for the contemplation of multiple and complex issues within the CLN role. Reflection, writing, literature and further dialogue and refinement of issues with the participants and with key people outside the study have served as conduits for fine tuning the development of the research process and thematic analysis. Informal and formal feedback about the findings as they developed assisted with keeping the project on track and real.

Research Design

Qualitative descriptive research methodology design meets the aim of providing a description without embellishment, in language that would deem the research useful in practice for the nurses (Sandelowski, 2000b). This incorporated the use of robust data collection techniques, the documentation of research procedures including details of methodology, and the data analysis procedures. In line with this approach purposive sampling and inductive data analysis were chosen. This section details the mixed methods used within the qualitative descriptive methodology framework.

Mixed methods

A word about using mixed methods. Mixed method designs use quantitative and qualitative methods within a single project (Tashakkori & Teddlie, 2003). The use of mixed methods allows the researcher to approach the phenomenon being studied from different angles to allow a more complete understanding of the topic than could be gained from using a single method (Giddings & Grant, 2007). Traditionally qualitative and quantitative approaches have been associated with discrete paradigms. Quantitative with the dominant empirical positivist "scientific method" paradigm and the qualitative "human science" paradigm concerned with the social world from which meaning is derived (Haase & Myers, 1988; Duzurec & Abraham, 1993). The differences in underpinning assumptions between the constructivist paradigm and quantitative positivist paradigm relating to the nature of; reality, relationships, truth statements and values are incorporated in the axioms of constructivism as outlined earlier. A common aim of both qualitative and quantitative research is in reaching for understanding of the world we live in. Demonstrating lack of understanding of the orientation of an approach or blurring of paradigms is a major critique of using mixed methods ( ) but as Sandelowski (200?) and others (Dzurec & Abraham, 1993; Morse, 2005; Giddings, 2006) suggest in research that has paid careful attention to its orientation and underpinning assumptions this need not be the case. At worst mixed methods described as "positivism dressed as drag" to be avoided (Giddings & Grant, 2007).

In support of using mixed methods Dzurec and Abraham refer to inquiry being about understanding and explanation and in order to achieve this integration of qualitative and quantitative methods can be used (1993). So instead of viewing the use of mixed methods as mergers or combinations of world views or paradigms, qualitative descriptive methodology involves the explicit framing of the inquiry in two or more world views which are distinct from each other (Sandelowksi, 2000a). This is possible "Because techniques are not linked to either paradigms or methods, combinations at the technique level permit innovative uses of a range of techniques for a variety of purposes" (Sandelowski, p.248 2000a). Combining techniques is not to be confused with what Sandelowski (2000?) refers to as qualitative descriptive research drawing on "hues, tones and texture" (p, 337) from other theoretical perspectives. Meaning that features of the "re-presentation" of individuals’ words or experiences may employ aspects of another style e.g. phenomenological (Sandelowski, 2000?, p. 337). Accordingly the overall research design sits within the constructivist paradigm, but the techniques used to gather data employ quantitative and qualitative techniques.

In keeping with the aims of qualitative descriptive research Data collection is "typically directed toward discovering the, who, what, and where of events or experiences, or their basic nature and shape (Sandelowski pg 338 2000b). To achieve this end three methods were employed: postal survey questionnaire; in-depth interviews; and a data collection activity. Data collection occurred in a sequential manner. Data gathered in the survey was used to develop phase two and three of the study, in this way each method mutually enriching each subsequent method and providing a depth of understanding that would not have been possible using one method. One aspect of the study led into and informed the other. Individual semi-structured in-depth interviews then further explored key themes to come from the survey and the nurses perspectives of the role. This method ensured the fleshing out with rich data regarding the main issues that came out of the survey. The final method, a data collection tool, designed to capture the day to day activities that the nurse carried out, thereby constructing a comprehensive picture of this nursing role.

Ethical Implications and consent

Ethical considerations relevant to this study incorporated truthful representation of the findings made throughout the research process. Additionally the traditional ethical principles that are important in nursing practice as suggested by Polit et al 2001 were adhered to; Beneficence, (Freedom from harm, Freedom from exploitation, Risk/Benefit ratio) Respect for Human Dignity (Right to self-determination, Right to full disclosure, Informed consent) and Justice (Right to Fair treatment, Right to Privacy).

Ethical approval for the research received from the Multi-region Ethics Committee (MEC) on 30th of September 2010. Criteria were met for minimal risk expedited review as described in the Ethical Guidelines for Observational Studies 11.13 (MoH, 2006). Ethical approval processes were followed with respect to the AUT University Ethics committee.

For a copy of the relevant documents pertaining to ethical and consent processes utilised in the study refer to Appendix 1 (A-F) (check numbers etc Patsy)

The following steps took place associated with seeking ethical approval and support to carry out the research

1) Informal discussion occurred with people in key positions such as Nurse Director ODHB, to assess the level of support for the research prior to gaining formal ethical approval.

2). Support was sought from the New Zealand Forensic Psychiatry Advisory group NZFPAG which consists of FMHS Directors and Managers.

3) Local agreements were negotiated with each DHB and Regional Forensic mental health service to approach managers of their Forensic Mental Health Services and hence court liaison nurses. Obtaining local agreements was time consuming. This involved contacting the relevant research centre within seven DHBs (Waitemata, Central, Capital and Coast, Canterbury, Otago and Southland). Each DHB had their own criteria and processes regarding obtaining a local agreement to contact and carry out research with their staff.

3. The Managers or Team leaders of all the Regional forensic mental health court liaison services were contacted by the researcher to ascertain how many surveys to send to each region. This included all nurses who cover the CLN role in their region.

It was recognised that participating in the research and disclosing potentially difficult and ethically challenging practice scenarios may result in participants experiencing distress. Therefore the availability of counsellor was arranged with AUT Health and Wellbeing services should this be required. Refer to the Participant Information sheet, Consent forms and AUT Health and Wellbeing letter (Appendix, ).

During the development of the proposal consideration was given to a process to follow if a participant disclosed unsafe nursing practice. Following rigorous discussion with the supervisors it was decided that the participants comprised an experienced group of nurses who were practicing autonomously and that it was highly unlikely that this scenario would ensue. However in order to protect both the participants and the researcher this was discussed at the start of each in-depth interview. Participants were informed that if any concerning nursing practice was disclosed the researcher would be obligated to discuss the concerns with this with the participant and if the matter could not be resolved the researcher would discuss the concerns with the supervisors for further advice. Issues such as uncovering unsafe or illegal nursing practice did not arise during the data collection phase and there was no need at any stage to follow through with this plan.

Seeking informed consent

Obtaining informed consent implies a process has been undertaken in which participants have received information and understood the research, and are able to consent voluntarily or not to participate in the research (Polit et al, 2001). Participation in all stages of the research was voluntary. Full explanation of the study was provided in Information sheets (Appendix ). Because the survey questionnaire was anonymous once received by the researcher the participants were not able to withdraw from that phase of the research. Return of the completed survey questionnaire implied consent (Appendix 11 ). Written consent was obtained from the in-depth interview and data collection activity participants (Appendix 11.....). Withdrawal from participating in the in-depth interviews was possible up until the interview had been conducted.

All steps possible were taken throughout the study design and implementation to ensure participants’ anonymity, privacy and confidentiality. Details of the voluntary and confidential nature of the research were outlined to the potential participants in the information sheet and consent forms (Appendix ). These details were reiterated at the commencement of each in-depth interview. All 6 participants indicated they understood this issue and the steps the researcher would take in the final presentation to maintain confidentiality and anonymity as much as practicable. The final report was returned to the participants for reviewing and checking.

Maintaining anonymity and confidentiality in a small group that practiced in a public forum presented particular challenges. In addition to employing pseudonyms, some detail such as; descriptions of work settings, timeframes and details of specific incidents from the practice stories, was altered so that particular scenarios are not identifiable and therefore neither was the CLN. It was particularly important to the integrity of the research that the participants felt confident their responses would be kept private and confidential and their anonymity would be protected as far as possible.

Participants have the right to expect that any information they share during the study will be kept in the strictest confidence (Polit, Beck & Hungler, 2001). Only the researcher had access to the raw data (surveys questionnaires, interview transcripts and data collection activity, database base with individual names). The supervisors only had access to raw data with the researcher. Raw data was kept in a locked drawer in the researcher’s office at the Southern District Health Board in Otago. Raw data and the database were destroyed on final completion of the thesis. Confidentiality agreements were signed by the AUT Administrative assistant and the interview transcriber (Appendix ). The research focus was the nursing practice therefore no identifying information regarding the person being referred to the court liaison nurse was collected at any stage of the research.

For a copy of the relevant documents pertaining to consent processes refer to Appendix...

Cultural consultation

As with any healthcare practice and/or research in New Zealand there is a commitment to the Principles of the Treaty of Waitangi. Any research with people requires thoughtful consideration of the ethical and cultural implications of the study (Davison & Tolich, 2003). Therefore obtaining cultural advice was one of the key components contemplated in the development of this research.

Consultation occurred with a cultural advisor regarding the ethical and cultural implications of all stages of the research from development of the proposal, developing the research tools and protocol regarding conducting the interviews and analysis as it effected Maori. The Cultural advisor to the research guided the researcher to explicitly solicit information regarding whether Maori models or frameworks were used reminding the researcher of the demographics of the people who come through the criminal justice system and the demographics of the nursing profession. The Cultural Advisor suggested identifying if there were any disparities here for recommendations for future education regarding the court liaison nurse role. The interview process was discussed with the Cultural Advisor and the researcher was prompted to ensure the participants had the opportunity for a support person with them if they so wished.

During the data analysis phase of the research process insights and reflection on the interpretation of data relating to the cultural issues were discussed with the Cultural Advisor.

Participant Sampling and Recruitment

In line with Qualitative descriptive research design purposive sampling was chosen as the sampling strategy for the research. The objective of purposive sampling is to identify information rich cases for in depth examination (Patton, 1990).

Survey questionnaire

All nurses practicing in the court liaison role in New Zealand whether full time, part time or who carry out a relieving or who provided cover role for the were invited to participate in phase one of the research. Inclusion of as many of the court liaison nurses as possible was to maximise the scope and variety of information gathered from the study context. Purposive sampling assists with this ensuring that participants are typical of the population or particularly knowledgeable about the subject (Polit & Beck, 2008) and to be representative of the group.

Despite the initial intention to include all nurses in New Zealand practicing in the Court liaison nurse role Hauora Waikato (an independent provider) declined for their court liaison nursing staff to participate in the research. Therefore, only the court liaison nurses in New Zealand practicing within DHB FMHS were invited to participate in the postal self administered questionnaire. This formed a cohort of potentially 41 nurses covering court liaison in New Zealand without the Haoura Waikato Group court liaison nurses.

There was no comprehensive mailing list available of all court liaison nurses in New Zealand or other expedient means of directly contacting court liaison nurses at that time. Therefore to ensure all DHB FMHS court liaison nurses had the opportunity to participate in the survey, information packs regarding the study were mailed to DHB FMHS Court liaison managers or Team Leaders for distribution. (refer to Appendix ..... for contents of the packs). To maximise returns a reminder letter and extra information packs (Appendix ) were sent out to the DHB FMHS managers three weeks following the first letter.

Several approaches made to the researcher by court liaison nurses who wished to participate but had not been able to complete the survey questionnaire within the original time frame. Following discussion with the supervisors it was decided to accommodate return of the surveys up until the interviews commenced and a final return date was set.

In-depth interviews and data collection activity

Six participants were invited by the researcher to take part in phase 2 - in-depth interview and 3 - data collection activity, of the research. The purposeful maximum variation sample (Sandelowski, 1995) included participants from three areas in each of the North and South Islands reflected a wide variety in terms of experience and location (city and rural). Thought was given to the need to have the means to explore common and unique manifestations of how this nursing role had developed to capture commonalities as well as any regional variations or idiosyncrasies in the role (Sandelowski 2000b).

Understandably there is often reluctance on the part of nurses to devote precious patient care time to completing audit like activities. The same nurses were invited to take part in phase 2 and 3 the researcher hopeful the nurses who were interviewed and therefore fully cognizant of the aims of the research and were very much part of the research would be more willing to continue on and complete the data collection activity. Also these nurses had the opportunity to discuss phase 3 with the researcher and have input into the data collection tool and therefore this would assist with the consistency in data collection. The 6 nurses initially approached by the researcher consented to take part in phase 2 and 3 of the research. Details regarding phase 2 and 3 were then emailed and posted to the participants (refer to Appendix .... ). Participants were offered the choice of a CD or Book voucher as a small compensation for giving their time and as a thank you for taking part in the research.

Tools

Survey questionnaire

Many definitions and methods exist in relation to surveys (de Leeuw, Hox, & Dillman, 2008). The definition proposed by de Leeuw, Hox, & Dillman suggests a survey is "a research strategy in which quantitative information is systematically collected from a relatively large sample taken from a population" (2008, p.2). A broader view of the survey can be found in descriptive exploratory survey studies (Schneider, Elliott, LoBiondo-Wood & Haber, 2003). The authors suggest this type of design is used to obtain accurate information about the characteristics of specific phenomenon, or rate of a phenomenon’s occurrence, especially when little is understood about the phenomenon within the positivist paradigm (Schneider, et al, 2003). Davidson and Tolich suggest surveys provide a ‘snapshot’ of the attitudes, values and behaviour of a group at one point in time (2003, p.133). Therefore survey questionnaires were used to obtain baseline information, demographic facts and data about what the nurses do, their attitudes, and opinions on specific issues, and how they felt about the role (Polit & Beck, 2008).

In line with the aims of the research to gather detailed descriptions of the existing practice as a foundation for not only the following phases of the research but so that there is a fundamental body of knowledge about the nursing practice in this role and this group of nurses. It was envisioned this data would provide a platform to make informed decisions regarding future development of the role. The survey was not designed to determine cause and effect. It was exploratory and descriptive.

Development of the survey questionnaire

The literature review did not reveal a replicable tool or previous research directly applicable to the New Zealand context therefore the survey questionnaire was developed by the researcher. Relevant guides were consulted during this process (Davidson & Tolich, 2003, de Leeuw, et al, 2008). Consultation with nurses currently practising within the court liaison nurse role took place to ascertain relevant and topical issues for the nurses and incorporate these in the survey questionnaire design. Relevant themes elicited from the literature review informed the survey questionnaire design. However in the main the questionnaire relied heavily on the researcher’s experiential knowledge and observations from practice and the expertise and key themes articulated by Kevin Seaton RN in the Chapter "Liaison services to the courts" (McKenna & Seaton, 2007). Incorporating the Court liaison nurses thoughts and the researcher’s experiential knowledge is fitting in a research design underpinned by naturalistic inquiry (Lincolin & Guba, 1985).

Development of the content and structure of the tools was informed with data collection tools kindly forwarded by Researchers Julie Sharrock and Brenda Happell with permission to utilise these as the researcher saw fit. Sharrock and Happell conducted research into the Consult Liaison Nurse role in Australia (Sharrock & Happell, 2002). Similarly the research article by Turnbull and Beese (2000) detailing their research into the role of the community mental health nurse in Magistrates courts in one area in the United Kingdom was drawn on in development of the survey questionnaire. The format used to explore the nature of and quality of working relationships was assisted with the design of similar questions in the survey questionnaire (Turnbull & Beese, 2000).

In seeking to positively influence the response rate attention was paid to the structure of the survey questionnaire. The use of language was reviewed to ensure it was as jargon free and user friendly as possible (Davidson & Tolich, 2003). Likewise questions were kept simple with one question asked at a time to avoid confusion and non responses or incomplete answers (Davidson & Tolich, 2003). Each section commenced with close ended questions to obtain hard data regarding particular issues or demographic details followed by open ended questions to tease the information out and encourage participation by giving the nurses voice.

The length of the survey and the time that nurses would have to take out of their day to complete it was taken into account. The researcher estimated that the questionnaire would take 20 – 30 minutes to complete. There was a danger that the time taken to complete the survey was too long and may discourage completion. Davidson and Tolich (2003) suggest that 10 minutes is the average attention span length for completing surveys. Acknowledging that CLNs often practice in extremely stressful busy circumstances in the court the researcher attempted to balance the need to be mindful of time with the fact this was the first piece of research into this role in New Zealand.

Accordingly the physical layout was kept uncluttered and easy to follow. Roberts and Taylor (2002) suggest the physical layout including appearance, quality and colour, and the inclusion of a stamped return-addressed envelope influence the return rate. The questionnaire was printed out on white A4 paper with black size 11 Calibri font consisting of fourteen numbered single sided pages. The questions were grouped into sections for ease of understanding. The reminder letter was printed on light blue coloured paper to ensure it stood out with mail trays and hopefully serve as a prompt to complete the survey. Questions were coded throughout the survey with numerical response to assist with analysis. The researcher developed a data set on SPSS v. The data set layout and coding were reviewed and checked by one of the supervisors and reviewed with an expert in SPSS from AUT University.

Piloting/pre testing the survey questionnaire

Piloting a questionnaire generally assists the researcher to clarify whether the tool is clearly worded and will provide the data required to answer the research question. Piloting allows for refinement of the data collection tool and generally increases validity (Burns & Grove, 2007). It also facilitates checking the tool does not contain any major biases (Polit, Beck, & Hungler, 2001). This was a major consideration for the researcher given my experience in the role and ongoing involvement with CLNs and therefore potential for biases to creep in. One of the supervisors with extensive knowledge in forensic mental health nursing provided rigorous feedback challenged the researcher’s rationales for including questions or asking them in a particular manner. This ensured as much as was practicable any personal biases on the part of the researcher did not colour the questions.

Usually the pilot is performed by a small sample of the target population with those undertaking the pilot having the necessary knowledge and skills to complete and evaluate the questionnaire reliably (Burns & Grove, 2001). The survey questionnaire was piloted on two occasions. Firstly with a nurse who had practised in a consult liaison nurse role in a general hospital (a role with some similarities to the CLN role e.g. sole mental health practitioner and significant liaison and educational aspect to role). Following the recommendations from the first pilot the tool was adjusted. The tool was then piloted with a nurse who had previously practiced in the court liaison nurse role. The nurses who pre-tested the survey were asked to read the Participant Information Sheet and complete the survey considering the following:

If the terminology was easy to follow and applicable

How difficult or complex they found the survey overall

How complicated they found individual questions

If the aims of the research as described to the participants conveyed what the researcher hoped they did

How long it took to complete and whether there was enough space to write answers

Based on Davidson and Tolich (2003, p.142)

Feedback provided assisted with clarifying how some questions were asked to ensure the instructions were explicit. The tool was not piloted more widely with the target population due to the actual small size of the population and not wanting to adversely affect the potential for participation.

Survey Questionnaire - structure and content

The self administered postal questionnaire was organised into four sections, 1, Demographic/background information; 2, Preparation for role, education and training; 3, Supervision and support; and 4, Working relationships. This format chosen so that the more sensitive items or topics requiring more thought were last (Davidson & Tolich, 2003)

Not sure how much detail i need here.....

A mixture of open ended and closed questions and rating scales were employed. The final survey comprised of 58 questions. 41 of the questions were structured as close- ended questions which simplified the participant’s response and ease of data analysis for the researcher. The close – ended questions varied in format from fixed response questions (e.g. Qs 1, 2, & 3), lists of items which participants rank ordered or rated (e.g. Qs 43, 44, & 45), questions that enabled multiple responses (e.g. Qs 14, 23, & 50), and open ended questions (e.g. Qs 25, 27, & 52). Clearly use of open-ended questions can result in diverse data which was fitting with the aims of the research (Polit, Beck, & Hungler, 2001). Open ended questions also enabled generation of practice issues the researcher may not have been aware of (Schneider et al., 2007), acknowledging to the nurses they were the experts in this role and their opinions were valuable. Open ended questions allowed as much information as possible to inform the following phases of the research.

Section 1 Qs 1-17 – Because of the dearth of information available regarding this cohort of nurses this section sought to obtain baseline demographic information and including details about longevity as a nurse and in the CLN role. Role specific information was sought regarding hours per week practiced in role, other roles performed for the FMHS, whether the nurse was a Duly Authorised Officer under the Mental Health Compulsory Assessment and Treatment Act 1992, whether they have a job/role/position description, courts regularly covered, and sites visited. The final part of this section focused on the availability of resources for the nurse at court.

Section 2 Qs 18- 29 – elicited detailed information regarding nursing qualifications, and background education. Experiences regarding the nurses’ orientation to the role were explored with ample opportunity for detailed responses in open ended questions. The nurses were asked to specify their opinion regarding educational/training pathway for court liaison nurses and to outline their perception of what that should look like. Determining knowledge in this area is important to ascertaining the education and training that may be required for the role. The value of a formalised means of communication as a group of nurses was canvassed along with what activities/functions of a formalised group would be of assistance.

Section 3 Qs 30 -42 – Clinical supervision and support were the focus of this section. Opportunities for discussion and reflection on practice were explored. Three questions focussed on whether the nurse discussed practice with CLNs outside of their region, and the nature of contact with CLNs whether contact occurred through a formal process or informally. This section concluded with space for the nurses to outline any final thoughts regarding clinical supervision and support.

Section 4 Qs 43- 58 – explored areas of tension and specific challenges. Topics canvassed included; relationships, diversion and liaison, ethical and cultural challenges, and specific practice related tensions such as informed consent, information sharing, and dual roles.

Capturing the nurse’s thoughts and ideas was important to inform recommendations and future role development. Therefore the final questions asked to elaborate on the most and least enjoyed aspects of the role. Space was provided for any further comments if they wished. For the research to appear attractive to the nurses it was vital that they would have the opportunity to discuss or contribute to dialogue about their role.

In-depth interviews (semi structured)

The aim of this phase of the research was to explore beyond background demographic information, role description and working conditions to pull out intricate details surrounding individual experiences, the thought processes, feelings, and beliefs relating to practicing as a nurse in the criminal justice system.

Semi–structured interviews were chosen in preference to structured interviews. Data collection by semi-structured interviews allows participants to be asked questions within a flexible framework (Patton, 1990). Interview questions in semi-structured interviews are guided by a list of particular topics to be covered (Schneider et al, 2003) rather than a list of set questions (Polit et al, 2001). The idea being that this group of senior nurses who practice autonomously would be able to talk freely on all the topics, describe the challenges, what works well and provide critique the role and systems.

Development of draft interview questions

A list of key topics and prompt questions for the interview guide was initially compiled after review of the literature. Firstly the literature revealed little was known about nursing practice in criminal justice settings; therefore the wider forensic mental health nursing literature was reviewed. Common themes included; lack of specific education and training, the need for professional supervision and support, and tensions and contradictions for health professionals that arise from the nature of the environments and the competing professions and therefore the potential for role conflict.

Therefore the interview guide comprised of the following broad topic sections with further prompts under each section: Background education and training; Functions of the role; Relationships; Areas of conflict and challenges to the nursing role; Framework or model for practice. Following preliminary analysis of the survey questionnaire the following topics were expanded upon and given more emphasis during the interviews; orientation, training pathway, national templates/guidelines, management support and recognition of complexity of role and specialty status. The overwhelming sense of the nurses enjoying the autonomy coupled with increased responsibility the role afforded came through from the survey questionnaire. Therefore the interview was planned with adequate time to enable the nurses to describe what the role meant to them and how they managed the increased complexity and responsibly.

Following each interview, notes were made of any contextual or other general impressions of the interview, and possible changes to the interview guide.

Interview process

Miller and Crabtree describe "the interview as a research gathering approach that seeks to create a listening space where meaning is constructed through an interexchange/cocreation of verbal viewpoints" (1999, p.89) To make the most of this approach two hours were allowed for the interviews. These took place in the participants own health setting at a negotiated time. Refreshments had been offered by the researcher however due to time and flight schedules this proved challenging to facilitate with all participants. This was discussed with participants and it did not appear to present a concern for any of them. On the contrary the participants were extremely welcoming and generously provided refreshments expressing happiness to be taking part in the research.

Being a registered nurse with many years of experience in interviewing people did not preclude undertaking specific preparation prior to conducting the interviews. Approaches to the interviews and processes were initially reflected upon and discussed with the Cultural Advisor, and then the local supervisor both before and after interviews. Guides to carrying out interviews were perused (Crabtree & Miller, 1999; Davidson & Tolich, 2003; Smythe & Giddings, 2007) with the aim of ensuring a genuine and respectful relationship developed during the interview to facilitate depth and richness of data. The logical conclusion would be that if this level of trust and rapport was not established then superficial data would result.

The notion of "Researcher as instrument" is one of the underpinning principles of Naturalistic Inquiry (Lincolin & Guba, 1985) Meaning the inquirer or researcher is sensitive and responsive to the intricacies of the inquiry process, and also to the differing nurses and levels of experience and skill and expertise in the role. In other words the researcher is able to follow cues or leads, explore issues and topics as they arose and appeared relevant to the purpose of the study demonstrating skill in interpreting and responsiveness to cues to guide the study. This fine tuning process was undertaken as each interview took place the cues or leads provided by the participant guided me regarding the next interview. This process extended to Phase 3 as elaborated below.

The progression of the interviews was very much as Smythe describes "People don’t speak or think in a linear fashion" (Smyth & Giddings, 2007, p.49). The initial approach in the interview was to start with broad open ended questions such as "tell me about how you came to be in this role?", "Talk me through your day", with the intention of then working through the Interview guide. However from the first interview all that was required was to ask the initial question, and then dialogue flowed in torrents. The interview guide was subsequently used as a checklist whereby topics were ticked off as the nurses covered them and only referred to at the end of the interviews if specific topic areas had not been covered by the nurses. My role as inquirer was to listen and perhaps prompt or clarify at times. Smythe (Smythe & Giddings, 2007).

Once typed, the transcripts were returned to the participants to check and provide further comment if they wished. The researcher did ask some clarifying questions at this stage.

Data collection activity

This phase of the study commenced following the in-depth interviews. Discussion with CLNs gave the impression that routine DHB statistical data collecting processes did not capture the enormity of this role and especially the liaison aspect and the multitude of ‘corridor conversations’ the court liaison nurse experiences every day. The aim was to collect data regarding the type of each activity and the time spent on each activity, separating out clinical and non direct clinical activities. Balancing the issues of not making this phase of the research onerous for the nurses and potentially affecting data collection with the need to ensure that the data collected was comprehensive.

Development of the data collection tool

To assist with engagement in the data collection activity and refinement of the tools this phase of the research was discussed with each participant at the conclusion of the in-depth interviews and feedback sought regarding the format of the data collection tools and time frames required to complete the activity. Consensus was reached with participants that they were able to complete the data collection tool in any manner that suited them such as, by hand or online. Feedback included reducing the data collection period from 4 to 2 weeks. Related to time constraints and the nurses’ suggestion they were more likely to complete over 2 weeks. The participants were able to choose a suitable two week period that would best reflect their usual practice (this was to take into account that public holidays, court rosters and other reasons that the court may be closed and the nurse not able to capture their usual practice on court list days) to complete the data collection activity. The participant’s feedback was incorporated in the design.

A simple format was designed for both tools. Refer to Appendix .... for copies of Data set 1; Data set 2; Instruction sheet; Key.

Following the consultation process the participants were emailed the tools, instruction sheet and key.

Data analysis and Data Re- Presentation

The Statistical Package for Social Sciences (SPSS) v.19 was used to generate the statistical analysis. Qualitative content analysis which is fitting in Qualitative descriptive methodologies (Sandelowski, 2000b) was employed to analyse qualitative data in to themes.

Quantitative analysis

Data was entered into the SPSS version19 dataset. SPSS is particularly suited to the analysis of quantitative research to facilitate the processing of large quantities of data and calculation of statistical measures. The quantitative data collected was mainly nominal and ordinal hence non-parametric tests formed the core of analysis. Guidance and assistance was sought from one of the supervisors and an expert in SPSS from AUT University. Each question was initially analysed individually. Then data was organised through a process of reduction, merging and interpretation to complete the analysis (Davidson & Tolich, 2003). This process was undertaken in consultation with the supervisors.

Re-presentation quantitative data

Simple descriptive statistics were used to describe the results. Data was presented as text summary and using table and graph form to allow for graphic perspective of the data and easy interpretation. The subsequent discussion chapter incorporates the formulation of decisions and conclusions from the data analyses linked to the relevant literature.

Qualitative content analysis

The thesis has already established its position with respect to the findings being viewed as "Constructions" or "meanings" and therefore "created realities" between participants and researcher (Guba & Lincoln, 1989, p.143). This means that any construction which emerges from a qualitative inquiry may be seen as unique, for a particular set of circumstances may never occur again in exactly the same way.

So with the underlying assumption that reality can be interpreted in many ways and understanding is dependent on subjective dynamic interpretation qualitative content analysis (Hsiu-Fang & Shannon, 2005; Sandelowski, 2000b) was deemed that most appropriate data analysis strategy. Qualitative content analysis is described as "reflexive and interactive as the researchers continuously modify their treatment of data to accommodate new data and new insights about those data" (Sandelowski, 2000b, p.338). This approach of analysis is oriented toward summarizing the informational content of that data (Sandelowski, 2000b). Content analysis involves reading the material and "using a consistent set of codes to designate data segments that contain similar material" (Morgan, 1993, p.114).

Qualitative data was transcribed verbatim from the interviews and survey questionnaires onto Microsoft word. The codes are derived from the data and can be modified as the careful reading and re-reading of the data takes place during the course of the study (Morgan, 1993; Sandelowski, 2000b). To achieve this, the transcripts were read and tapes listened to repeatedly. Then initial codes using different colours per code were derived firstly by the researcher and then one of the supervisors. These were bought together and discussed and consensus reached. As Morgan (1993) describes the codes were derived from the data and through careful reading and re-reading of the data they were modified.

The data was recoded several times into categories and revisited and until the researcher was satisfied with the groupings. As suggested by Morse (1992) ongoing analysis enables the researcher to proceed with the analysis as they gain understanding and insight about the phenomena of interest within the study.

Patterns in the categories were then interpreted asking why and how the patterns came about and summarizing and grouping (or interpreting) the patterns into themes (Morgan, 1993). ‘Theme’ or ‘Pattern’ can imply a variety of meanings, including a recurring event (or frequency of occurrence) and a unifying element, essence, or motif (or singular thread or link) (Sandelowski and Barroso, 2002). The research sought patterns and ideas that described current practice, would inform future practice, and identify areas (of practice) for further exploration.

It is recognised that knowing about a phenomenon includes understanding the context to give meaning to the phenomenon and that all description requires perception (Sandelowski, 2000b). Therefore it follows that the choice of what to describe and the process of describing involves transformation or interpretation (Sandelowski, 2000b). Maxwell describes interpretive accounts as "ground in the language of the people studied and rely as much as possible on their own words and concepts" (Maxwell, 1992, p.289). Interpretation occurs through the researcher’s familiarity with the data. Constantly returning to the original data sources develops the researcher’s intuitive grasp of its meaning, which finally matures into an interpretation of the multiple realties that exist.

As per the previous process once themes were derived the methodological expertise of one the supervisors who was intimately involved in the research was used ensure that the themes fit the data. Therefore interpretive validity of the study is assisted by this review of thematic analysis. However a caveat applies here and the research acknowledges that the ‘expert’ can only validate what was presented to them (Sandelowski, 1998; Maxwell, 1992).

The repetitive analysis process occurred until transparency was achieved and a report representative of the information provided by the nurses was produced. The interview participants indicated that the analysis of the data organised into themes was reflective of the content of the interviews and lends confidence to the applicability of the findings. Interpretive validity of the findings was also demonstrated by having them approved by the constructors as suggested by (Lincolin & Guba, 1985).

Re-presenting Qualitative data

Careful attention was paid to ensure that choices about what was emphasized in the research fit the research purpose and methods (Sandelowski, 1998). Themes were arranged in a straight forward manner from the most prevalent to least prevalent theme, in keeping with Sandelowski’s recommendation that "a straight descriptive summary of the informational contents of the data organised in a way that best fits the data" (2000b, p. 338). Data across the three data collection methods were integrated descriptively in the analysis (Creswell, 2003).

In qualitative descriptive studies the writing task is to ensure that the reader hears the participant’s voice with minimal voice-overs by the researcher (Sandelowski, 2000b). The researcher viewed this being conduit for the CLNs perspectives with the aim for the nurses descriptions and voice to take precedence in the writing up. Attention was paid to ensuring the CLNs views were accurately captured. Portrayal of the common as well as idiosyncratic themes in the interview data, with the CLNs words paraphrased or quoted to illustrate these views was essential. Central to this notion is the researcher’s belief the research must be of most relevance to the audience for whom it was written – not sure where i was going here.... Finally sharing the final writing of the research with the participants occurred, not so much to establish approval but to provide a means to reflect on the experience of the research for the participants and their nursing. The writing had captured the nursing at a particular point in time (photographic slice of life) within the socio-political context and ongoing evolution of the nursing role in the criminal justice system.

Chapter Summary

This chapter provided information about the qualitative descriptive methodology and the philosophical position of the research. The research design was described with the reader being referred to the appendices used in the research. The course of action taken to obtaining ethical approval was outlined. How the research addressed credibility was established. Comprehensive background information regarding this cohort of nurses was obtained through the use of Qualitative descriptive methodology and mixed method research design. The research design also enabled a wide perception and understanding of the role therefore informing as broad a description about the nursing practice as possible and teasing out of the challenges these nurses face.

Chapter one introduced the study topic and set the scene as to the relevance and interest of the research. Chapter two reviewed the literature providing an understanding of the topic area and of the existing knowledge of this nursing role presenting the foundation from which the research was constructed. The following chapter presents data analysis from the survey questionnaire.