This thesis is about self-direction in nursing education and practice. The study, located in the Republic of Ireland, took place against the background of profound professional and educational reform culminating in the transfer of pre-registration nursing to the higher education sector in 2002. It adopts the position that learning how to nurse is dependent upon many issues including an individual’s perception of what nursing is together with the educational experiences to which that individual is exposed. The investigation endorses the need for continuous engagement in learning expressed in much contemporary literature and policy (An Bord Altranais 1994, An Bord Altranais 1997, Report of the Commission on Nursing 1998, Report of the Nursing Education Forum 2000 and Department of Health and Children 2002). One method of facilitating life-long learning is through a process of self-direction. The aims of the study are detailed below. 1. To explore the concept of self-direction from the perspective of: student nurses; nurse educators; nurse practitioners; and nurse managers. 2. To develop a framework for the introduction and development of self-direction within nursing education and practice. Given the contextual nature of the study it was also envisaged that the methods employed could contribute to an analysis of action research as a research methodology capable of contributing to policy development. 3. To explore the stability of qualitative methods of data analysis. Qualitative studies have been criticized for the lack of mechanisms or processes to acknowledge the possibility that the evidence presented does not reflect that embedded in the data. This study therefore employs a variety of data analytical methods in an attempt to address this deficit and contribute to the development of robust findings. The literature review is divided into three main sections: self-direction; curriculum; and policy development. The review concludes by making the case for an exploration of self-direction as one means of facilitating the development of nursing education programmes in a manner informed by the thinking of Foucault, Dewey, Durkenheim, Hiemstra, Brockett, and Confessore and Confessore amongst others. The methodological section debates the merits and limitations of adopting a worldview based on subjective experience as the basis for inquiry. This emanates from the premise that qualitative methods allow exploration of humans in ways that acknowledge the value of all evidence, the inevitability and worth of subjectivity and the value of a holistic view described by Chinn (1985). An action research design, specifically participatory action research, was considered most appropriate for this study. The model of action research selected is that proposed by Elliot (1991) and based on the original thinking of Lewin (1946). The study comprises two discrete but inter-related cycles of action research. Cycle one focused on an exploration of self-direction from the perspective of student nurses; nurse educators; nurse practitioners; and nurse managers engaged in pre-registration nursing education. The sample consisted of seventy-two participants. Data was collected using semi-structured interviews. Data was analysed using the constant comparative method as described by Glaser and Strauss (1967) and computer assisted qualitative data analysis software (CAQDAS). The findings from both methods of data analysis were compared and contrasted with a view to contributing to the general methodological debate surrounding data analysis. The findings from action research cycle one emerged as four core categories, which were linked together in a practice framework of self-direction within a model of social reality as described by Burrell and Morgan (1979). Cycle one concludes by proposing to utilize the framework of self-direction developed to explore the meaning of specialist and advanced roles in nursing together with how the knowledge and education required to support these roles could be organized in a self-directed manner. In essence the four categories, which emerged from cycle one, were used to structure and guide the exploration in cycle two. Cycle two is located within the arena of mental handicap nursing because of the researcher’s expertise within this area. The sample for action research cycle two consisted of four hundred and forty two registered nurses working in the area of intellectual disability. Data was collected using focused group interviews and analysed using thematic analysis. The findings were clustered into seventeen themes, which gave rise to two core categories, describing the roles and knowledge required to support specialist and advanced practice in mental handicap nursing respectively. A third core category describing the organisation of specialist and advanced practice in mental handicap nursing completes the findings from action research cycle two. The findings of the study gave rise to a proposed framework for the development of clinical specialisms and advanced practice in mental handicap nursing. Following negotiation with relevant bodies the findings were developed into a policy document entitled Proposed Framework for the Development of Clinical Specialism and Advanced Practice in Mental Handicap Nursing. In terms of subject matter the finding from this phase of the study are unique within the context of both national and international nursing. The relationship between the findings and policy development in both phases of the study illustrates the potential of action research as both a stable and responsive methodology and a policy-making mechanism.