CHAPTER 3 RESEARCH METHODOLOGY

3.1 Introduction In this chapter, the rationale for the selection of phenomenology as a methodology for the current study is explained. Phenomenology is both an appropriate and useful method for examining the process of self-directed learning of nurses in clinical practice. By using a phenomenological qualitative research approach, the nurses experiences of self-directed learning on-the-job can be discovered and understood. As Mc Millan (2004, p 273) states:
Phenomenological research describes real direct experiences that happen daily and can interpret the experience of participants in order to understand the essence of the experience as perceived by the participants.

Phenomenological study was chosen for this research because it is a method that can be used to identify the meaning of the experiences that the participants have of self-directed learning and their perceptions of the process. The objective of phenomenology is to describe the full structure of what an experience means to those who live it. According to Marton (1988, p. 147) the aim of phenomenology is:
to discover the structural framework within which various categories of understanding exist. Such structures (a complex of categories of description) should prove useful in understanding other peoples understanding.

3.2 Objective of the study The objective of the current research study was to search for an understanding of the process of self-directed learning of nurses according to the perception and experience of nurse learners. The particular research questions for the study were as follows: 1. To what extent did the nurses see themselves as self-directed learners? What did selfdirected learning mean to them? 2. What did professional nurses working in wards learn by self-directed learning?

67
3. How did professional nurses working in wards learn by self-directed learning ? 4. What factors influenced nurses self-directed learning ? How did those factors influence their learning? 5. What were the nurse educators roles in supporting the self-directed learning of nurses who were working in wards? In this chapter the reason for selecting the research method of phenomenology, the participants and sampling method, the procedure for gaining informed consent, the protection of rights for the participants, the role of the researcher and the methods of data collection and data analysis will be described. 3.3 Phenomenological study Phenomenology as a qualitative research method has its source in social philosophy and psychology. It arose in the late 19th century in response to positivism which, as a scientific approach, could not adequately address the problems being presented to the humanities (Sadala and Adorno, 2002). It was initially developed by and is most often associated with Husserl (1962). According to Bernstein (1983), it was Husserls belief that subjectivity underpins scientific understanding and the life-world of daily experience seeing a phenomenon and the being that experiences that phenomenon as inextricably linked. Husserl (2000) contends that phenomenology allows the researcher to revisit the world as it is lived and experienced, describing phenomena as they reveal themselves rather than explaining or reasoning their cause. Martins and Bicudo (1989) describe phenomenology as a distinct approach to scientific investigation, replacing statistical relationships with individual descriptions and interpretations that result from lived experience. A phenomenon can be defined as a situation or condition experienced in daily life (Giorgi, 1985). Therefore, when using phenomenology as a research method, the researcher starts with the lived experiences of the participants, uncovering their own understandings of their experience. In this way, phenomenology allows the researcher to see the personal meaning participants attach to their experiences and how the participants create their own personal reality (Lawler, 1998). As stated by Martins (1992), phenomenology is about the

68
understanding of phenomena as opposed to the explanation of them. It sees phenomena as qualitatively different and attempts to understand how they are experienced, conceptualized, understood and ultimately perceived (Marton, 1994). Two approaches to phenomenological research have emerged descriptive and interpretive. Descriptive phenomenology is rooted in Husserlian philosophy which focused on describing human experience as it is, as opposed to employing predetermined theories of scientific thought (Martins and Bicudo, 1989). In undertaking a descriptive phenomenological methodology, the researcher attempts to put aside their preconceived ideas about the phenomenon being researched, checking with the participants as to whether their interpretation of the account given is a true expression of the participants intention (Paley, 1997; Porter, 1998). Underlying this approach is the notion that the fundamental nature of a phenomenon can be seen through the principle of intentionality: consciousness understood as consciousness of something (Sadala and Adorno, 2002). Interpretive phenomenonology developed from the philosophy of Heidegger (1962), a student of Husserl, who contended that its neither possible nor necessary to attempt to separate ones experience from the phenomenon being observed and interpreted. Therefore, the aim of interpretive phenomenological research is to create a mutually meaningful account, from both the participants and the researcher, of the phenomenon under research (Price, 2003). According to Kleiman (1995), the purpose of descriptive phenomenological research is to create a structure of the phenomenon observed whilst interpretive phenomenology seeks to create an account, in text or story, that provides insights into the phenomenon observed. Adopting the Husserlian understanding that the essence of a phenomenon arises through the conscious awareness and intention toward it (Sadala and Adorno, 2002), this research used the descriptive phenomenological approach in order to provide a structure for understanding the phenomenon of self-directed learning among nurses. In this research, the researcher sought to realize the perspective of nurses about the process of self-directed learning in clinical practice. Phenomenology was selected as the method for this research in order to understand how nurses lived their experience of self-directed learning in clinical practice. Nurses are seen as unique, holistic beings with their own historical, sociocultural backgrounds that direct the ways they interpret and understand the world. The researcher attempted to understand what nurses

69
experienced during self-directed learning in situations where they related to an existing world in which there are on-going situations they have to face. Central to the study, was the aim of understanding the phenomenon of self-directed learning. Self-directed learning is by definition concerned with self and the perspectives and experiences of self as learner. Phenomenology is therefore particularly suitable as a method of discovering how nurses perceive and understand their learning process. Phenomenological research has become a prominent method of research amongst nursing researchers. Specifically, researchers have previously used a phenomenological approach to study ward-learning climates (Lee and French, 1997). According to Sjostrom and Dahlgren (2002), phenomenological research, in describing both the structure and substance of how phenomena are experienced in clinical practice, provides the fields of nursing and nursing education with an increased awareness of the phenomena being researched and can then inform the basic and continuing education of nursing staff. Phenomenology is a useful method for researchers who are interested in researching how nurses and nurse educators give meaning to the self-directed learning of nurses in clinical practice, what and how nurses do self-directed learning, and what the problems, barriers, and supports are to their self-directed learning. It also enables the researcher to explore how nurse educators encouraged and facilitated the self-directed learning of nurses from the perspectives of nurses and nurse educators. Through interpersonal communication and interaction, these aspects of the phenomenon of self-directed learning and individual examples are able to be explored. From the subsequent findings, categories can then be constructed and, through a process of inductive analysis, the researcher is able to develop an understanding and description of the essence of the phenomenon (Sjostrom and Dahlgren, 2002). Uljens (1988, p. 134) outlines the process of phenomenological study as follows:
1. Selection of a phenomenon in the existing world 2. Selection of an approach to look at the phenomenon 3. Discovery (by a process of tape-recorded interviews) of the participants ideas about the selected phenomenon 4. Transcription of the tape-recorded interviews 5. Analysis of the written statements and description of constructed categories

70
To achieve the aims of this study the researcher chose two methods of data collection to investigate the phenomenon of self-directed learning of nurses. The first involved semi-structured interviews to explore and discuss the participants experiences of self-directed learning, the mode of learning that they had used and the situations in nursing practice which motivated selfdirected learning as well as how the process happened. Secondly the researcher used participant observation to observe phenomenon of self-directed learning as it really unfolded during nursing practice in the wards and the progression of learning as it happened. 3.4 Participants To select participants for the study, purposive sampling was conducted. McMillan (2004) defines purposive sampling as a method in which the researcher chooses participants because of their specific knowledge of the research topic. To research self-directed learning of nurses in clinical practice, the researcher chose professional nurses as participants for their particular knowledge of the phenomenon of self-directed learning as they perceived it in their practice. Random sampling was then conducted to select the final participants from those who had volunteered and met the criteria of being in their position for one year. The study involved eleven participants who were professional nurses. Each of these nurses worked at a hospital which is categorized as secondary care level with over 500 beds. The hospital is located in a large agricultural province in Thailand. The hospitals mission is to provide services of health care promotion, disease prevention, treatment and rehabilitation by a holistic health care service team that emphasizes patient-centered care to maximize the satisfaction of the service-acceptors. The sample included seven professional nurses who had practiced nursing for at least one year and four nurse educators who had been in their position for at least one year. The researcher chose participants who had volunteered to be part of the study after the project was advertised among nurses at the hospital. The participants were selected in order to represent the following areas of clinical practice:

71
1. Medical Department. The researcher randomly selected one participant from each ward (Male Medical Ward and the Female Medical Ward). There were two participants from this department. 2. Surgical Department. The researcher randomly chose one participant from each ward (Male Surgical ward and Female Surgical ward). There were two participants from this department. 3. Obstetrics-Gynecology Department (one ward). The researcher randomly selected one participant from this department. 4. Pediatric Department (one ward). The researcher randomly chose one participant from this department. 5. Intensive care unit (one ward for admitting severe patients from the four departments listed above). The researcher randomly chose one participant from this ward. There was one nurse educator for each of the main departments. The researcher invited nurse educators who had been in their positions for at least one year to participate in the study. Four nurse educators and the researcher, who was a nurse educator in the surgical department, participated in the study. 3.5 Selection of the participants Following approval from the Human Research Ethics Committee at Victoria University and the hospital director, nurses from the hospital were recruited to participate in the research study through posters placed on notice boards and announcements made in the monthly nurses meeting of the hospital. Both encouraged nurses who had the required criteria (at least one year work experience) and who were interested in participating in this research to list their names and their sections onto a blank card. The researcher collected the completed cards with the names of nurses who were volunteering for this research project and sorted them by workplace in the hospital, that is, the ward in which they were nursing. Then, the researcher put their cards in separate boxes with one box for each ward. After that, the researcher randomly selected one card with a nurses name from each box. As a result, the researcher randomly selected one nurse as a participant from

72
each ward. The researcher then wrote letters informing each participant who was selected individually. 3.6 The procedure for gaining informed consent was as follows: 1. A letter of request together with information for participants and the Consent Form for the participants involved in this research were sent for the approval of each participant. 2. The researcher also sent letters to the Head Nurse and the Director of the Hospital in order to request permission to conduct research about the self-directed learning of nurses while practicing on the wards in the hospital before the researcher could begin the research project. In particular, permission to observe nurses while working on the wards was sought. 3. It was made clear in the Information to Participants and on the Consent Form that participating in this research would not affect the performance appraisal of the professional nurses. 4. They had a week to study the form before signing it. 5. The researcher gathered all the completed forms. 3.7 The protection of the participants rights 1. Prior to the commencement of the project, a meeting was held so that the researcher could give details of the study to the participants. They had an opportunity to ask questions during the meeting. 2. They were asked to sign the informed consent. They were allowed to select appropriate times to be interviewed and observed. Interviews and observations were scheduled within the normal clinical practice in the ward. 3. Confidentiality was ensured. No names have been used in reporting the results of this study. Each participants name was replaced with an encoded number. 4. The transcript of the interviews was returned to the participants in order to check for accuracy. Also the participants were involved in checking the findings from the analysis before the results were reported.

73
5. Participation in this research did not affect the participants work appraisal or promotion. 6. As a colleague, the researcher took responsibility in promoting nurse education in the surgical department, and had no authority over them in terms of work appraisal, assessment or evaluation. Moreover, neither was the budget allocation or its approval the responsibility of the researcher. 7. The researcher provided considerable support to the professional nurses from every ward who were involved in the project. The researcher was available during meeting times and at other times for the professional nurses to discuss any concerns that they had regarding the conduct of the project. 8. The counselling service was a consultation service for any participants who may have felt worried and/or insecure about providing personnel information both in individual and group settings. 9. The participants were reminded that they could withdraw from the project at any time. 3.8 Role of the researcher As a nurse educator in the surgical department, I play the role of planning for nurses continuous development, supporting them and promoting the areas of their knowledge and skills, especially in self-development. My responsibilities as a nurse educator are to: 1. 2. 3. 4. 5. Study the training and development needs of nursing personnel in the surgical division of the hospital. Analyze the needs of personnel development in nursing for the surgical division of the hospital. Prepare the nursing personnel development plan for the surgical division of the hospital. Conduct projects to develop, promote and support the knowledge, capabilities, and potential of the nursing personnel in the surgical division. Evaluate the development of nursing personnel for the surgical division in both the female and male surgical wards of the hospital.

74
6. Develop and improve plans and projects for nursing personnels development in the surgical division in order to efficiently meet the standards as outlined by the hospitals policies and requirements. Moreover, the budget for personnel development must be economically and effectively used according to the hospitals policies regarding budgeting allocation. Apart from being a nurse educator of the surgical department, I am responsible for nursing practice in the female surgical ward two days of every week because the hospital is facing a shortage of nurses. I have been aware of and concerned about my professional nursing role while I was conducting this qualitative research. Heidegger (1962) stated that the researcher is part of the social world of participants and as such necessarily use his/her own experiences in order to interpret those of the research participants. In addition, Kellett (1997), Van der Salm and Bergum (2000) argued that the role of the researcher was to work with the participant in order to construct an account of the phenomenon that had mutual meaning. For this research, as researcher my role was to describe, rather than to construct, an account of what the participant has told me. However, it is possible that I might have influenced the finding of this research because I did play the dual roles at the same time, as the researcher and participant. I am the nurse educator of the Surgical Department and already had basic knowledge about self-directed learning before conducting this research. With these mentioned facts, in the status of a researcher, I realized the bias that may be happening in the process. So, I have been very careful to reduce the problems of bias by collecting the data of this research with the assistance of one of the participants and submitting it to the Principal Supervisor and Co-Supervisor for comments from them before proceeding to the next data collection until completed. In addition, I consulted with both of them periodically and continuously in setting codes and identifying categories of the findings. There was also a review of my roles and responsibilities as a nurse educator of the Surgical Department. It was found that the researcher, as an educator, had the main function of encouraging and supporting nurses to develop their knowledge and skills in performing their nursing practices and in furthering their career development. The researcher did not have

75
authority in giving results of performance appraisal, promotion, interest, or benefit to the nurses. In addition, the researcher also reduced the bias by randomly selecting the participants from four departments and one unit: Surgical Department, Medical Department, Pediatric Department, Obestric-Gynaecology Department, and Intensive care unit. 3.9 Methods of data collection The researcher spent a total of seven months (July 2003 to January 2004) collecting data. The researcher collected qualitative data by conducting semi-structured interviews with nurses about their ideas, their opinions, and their experiences and through participant observation, observing nurses as they worked. Interviews can provide information about peoples attitudes, their values, and what they think they do (Patton, 1990). Observing can provide information about what people actually do (Fraenkel and Wallen, 2003). 3.9.1 The semi-structured interview as a method of data collection In this research, semi-structured interviews using open-ended questions were one of two methods used to collect data. According to Marton and Both (1997), the semi-structured interview is the favored method of data collection in phenomenological research. Furthermore, interviewing has been widely used in nursing research and is considered a sound method for studying nurses insights and experiences (Davis, 1984; Windsor, 1987; French, 1989; Nelms, 1990; Beck, 1993; Sheilds, 1995; Baillie, 1995). Patton (1990, p. 45) noted that:
We interview people to find out from them those things we cannot directly observe. The issue is not whether observational data is more desirable, valid, or meaningful than self-report data. The fact of the matter is that we cannot observe everything. We cannot observe feelings, thoughts and intentions. We cannot observe behaviors that took place at some previous point in time. We cannot observe situations that preclude the presence of an observer. We cannot observe how people have organized the world and the meanings they attach to what goes on in the world. We have to ask people questions about those things

Patton (1990) also noted that two important issues arise with respect to interviewing as a phenomenological research method. Firstly, the participants motivation for taking part in the

76
research may positively or negatively influence the interview and the data generated. Secondly, the researchers ability to understand and correctly interpret the participants responses has a significant impact on the researchers ability to decide further lines of questioning, in turn effecting the quality of the data produced. Johnson (2000) described the interview, including its analysis, as an engaged conversation between two people. In the interview, the researcher puts him or herself in the participants situation to try and understand that persons point of view (Gadamer, 1989). The researcher needs to listen and pay constant attention to the participants as they are responding, repeatedly attempting to understand the meaning of what is being said and how the person has shaped his or her perspective. In this way, interviewing is more than collecting data. Interviewing allows the researcher and the participant to connect in a profound way, reducing the distance between them. The distance is further reduced as the researcher analyses the text, effectively carrying out a conversation with the text of the interview. To be able to understand the perspective of the participant as clearly as possible, it is important for the researcher to be open and inquisitive in both the process of interviewing as well as analysis (Gadamer, 1989). 3.9.2 Participant observation as a method of data collection In this research, the researcher also collected data using the technique of participant observation, observing the atmosphere, surroundings and behaviors of the participants and how these related to the process of self-directed learning while working on wards. According to Patton (1990), observational data are appealing as they give the researcher the means to gather live data from live situations. Patton (1990) goes on to suggest that observational data allows the researcher to both enter into and comprehend the situation being described. Morrison (1993) elaborates, stating that observation enables the researcher to gather data on the following settings: the physical setting (i.e. the physical environment and its organization) the human setting (i.e. the organization of people, the characteristics of the groups or individuals being observed, such as gender or class)

77
the interactional setting (i.e. the interactions that are taking place: formal, informal, planned, unplanned, verbal, non-verbal etc.) the programme setting (i.e. the resources and their organization, pedagogic styles, curricula and their organization) In participant observation studies, the researcher takes part in the situation or setting that he or she is observing. The researcher works together with those in the observation situation, assuming as natural a role as possible and engaging in the activities of the setting as a member of the group. Whilst the researcher is engaging in activities, he or she records information. The role of participant observer necessitates seeking permission to participate in activities whilst taking on a relaxed role as observer in the setting. As Creswell (2002) notes, one of the disadvantages of participant observation is the difficulty of taking notes while engaging in activities within the observation situation. Some researchers wait to write down observations until after they have left the research site which may impact the quality of the data produced. However, Creswell also notes that observation has the important advantages of providing the opportunity to record information as it happens in a setting, to study actual behavior, and to study individuals who have difficulty putting their ideas into words. 3.10 Data collection procedure There were two interviews for the participants in this research. The first interview began the research. Following the first interview, the researcher conducted participant observations. After the observations were completed, the second interview was conducted in order to gather further details and to clarify issues that arose while interpreting and analysing the data. 3.10.1 The first semi-structured interview The first interview was conducted when the data collection began in July 2003. It was a semistructured interview with 12 questions designed by the researcher about the self-directed learning

78
of nurses in clinical practice. The interview questions investigated the self-directed learning process of the participants; what and how participants learned, why they wanted to learn, how they went about learning, who and what helped and supported their learning, how they felt when they were learning, and how the nurse educator encouraged or did not encourage them. The interview questions for the first interview of nurses are in Appendix A. Eleven questions, relating to the encouragement of nurses in undertaking self-directed learning, were designed for the interviews of the nurse educators. The interview questions for the first interview of the nurse educators are in Appendix B. Each participant was interviewed for 60 minutes. Each interview took place in a private room on the participants ward and was conducted in the Thai language. The researcher was the interviewer. A tape-recorder was used to record the interview. Before interviewing, each participant was asked permission for the interview to be recorded. The researcher interviewed the nurse participants one by one until all seven participants had been interviewed. After that the researcher started interviewing nurse educators one by one until all four of them had been interviewed using the same interview process as for nurse participants. After each participant was interviewed, the researcher transcribed the tape of that interview into Thai. The researcher then summarized, in English, the answers to each research question and submitted them to the principal supervisor in order to receive comments that were later used as information to further develop the interview method for the second interview. When the first semi-structured interview for all eleven was participants finished, the researcher analysed and synthesized the data in order to write a summary, in English, of the findings from each interviewees responses to each research question. 3.10.2 Participant observation The researcher conducted participant observations and witnessed the conditions and situations of the self-directed learning process that happened to each participant as she worked in one of seven different wards. The researcher observed the participants (who were the researchers colleagues) from Monday to Friday, 8 am 4 p.m., from August 2003 to January 2004. During the observation period, the researcher observed the participants by working with them and

79
watching them while they were working. As a member of the nursing team, the researcher worked collaboratively with participants during the period of observation. As a colleague, the researcher takes responsibility in promoting and facilitating nurses education for two participants of this research. The researcher has no authority over her colleagues in terms of work appraisal, assessment or evaluation. In addition, neither the budget allocation or approval is within the researchers responsibility. The researcher has held the position of nurse educator of the Surgical Department for about 10 years. Consequently, the researcher has gained knowledge and experience about the self-development of nurses. The researcher was therefore in a unique situation to understand more clearly what and how learning occurred in nurses in clinical practice. When undertaking participant observation, the researcher recorded the important data of social phenomenon in field notes. The following social phenomenon types guided the researchers observation and writing of field notes (Lofland, 1971; Chantavanich, 2000, pp. 134-138): 1. Actions were defined as incidents, situations or behaviors that happened within a period of time, but did not continue for a long period of time. 2. Activities were defined as incidents, situations or behaviors that continually happened and related to some people or some groups. 3. Meaning was defined as what a person explained, communicated or interpreted about their actions or activities, where meaning involved vision, belief, definition, and culture. 4. Relationships which defined the links among the many people in the hospital were viewed as either harmonious or conflicting. Specifically, the kind of relationship nurses had with their colleagues and head of the ward was noted. 5. Participation in activities was defined as how well a person became involved with and joined in an activity or how the participant adjusted to and coped with a situation or incident that happened. In particular, it was noted how much and how often the nurses participated in activities of the ward.

80
6. The setting was defined as the working conditions and environment inside the ward that was routine and happened each day. The researcher used these categories as a framework for writing field notes during data collection and later on for analysing these data. The researcher conducted the participant observation of the first participant of the first ward for a period of one week from Monday to Friday, 8 a.m. to 4 p.m. at the beginning of August 2003. All the incidents, situations, behaviors, and statements of people and participants in the ward, and the working conditions and environment inside the ward were recorded in field notes. Later these data were screened and the data that related to self-directed learning of nurses were then coded in the memos of the field notes. The field notes were submitted to the principal supervisor for comments that were later used to further develop the method of conducting participant observation in other wards. The researcher conducted the participant observation in the wards with each participant of each ward until all seven participants of all seven wards had been observed using the same method. To compile the field notes the researcher jotted down brief, important data that had been observed so that information would not be lost through forgetfulness. The data were recorded later in detail. (See an example of field notes for the first ward observed shown in Appendix E.) This helped the researcher arrange the descriptions for data analysis, make further plans, and draw conclusions during each period of time. 3.10.3 The second semi-structured interview The second interview was conducted after the participant observation was completed. In February 2004, the second interview was conducted using semi-structured questions. The researcher used selected questions from the first interview that would give additional details or clarify the initial responses of each participant. Four questions were used to interview the nurses a second time (see Appendix C). Three questions were used in the second interview of the nurse educators (see Appendix D).

81

3.11 Data checking The researcher checked the data after data collection to ensure the data was valid and reliable according to the real conditions of the phenomenon. The validity of the data was maintained by transcribing the recorded tape of the interview word by word. Listening to the tape a number of times, especially the unclear parts, enabled the accuracy of the transcription to be checked. The method of checking the reliability of the data was methodological triangulation. To ensure methodological triangulation of the data, multiple data sources were used (Patton, 1987). The researcher used interviews from two different sources, nurses and nurse educators. The participant observations and second interviews of participants provided improved reliability for the study. 3.12 Data analysis The data were analysed using an inductive approach allowing the prevailing pattern, themes and categories of the research findings to emerge from the data rather than be controlled by factors predetermined prior to their collection and analysis (Patton, 1987). The procedure for analytic induction undertaken in this research followed that used by French (1989) and Burnard (1991). For this research, there were two sources of data that were analysed inductively: the data from the interviews and the data from the participant observations. The analysis of the transcripts and field notes comprised of labeling the data, creating a data index, sorting the content of the data into meaningful categories, and determining a list of themes. The constant comparative method was used in conducting the analysis. Goetz and LeCompte (1981) describe this method as a continuous process in which inductive category coding is simultaneously merged with a comparison of observed phenomena. In constant comparative analysis, data are recorded, categorized and compared across categories. Patton (1990) describes the process of categorization as one of constantly revisiting the logical explanation and the concrete data whilst looking for significant relationships. Lincoln and Guba (1985) suggest that the main task of categorization is to organize those data that appear to relate to the same content into temporary categories. The details of the data analysis process are outlined below.

82

3.12.1 Analysis of data from the interviews 1. The researcher transcribed the tapes recorded during two interviews of the nurses and nurse educators into the Thai language one by one in accordance with each interview question. 2. The researcher repeatedly read the transcripts of the tapes and reviewed literature relevant to self-directed learning of nurses. 3. The transcribed interview was typed on the left-hand side of each page. The right hand side of the page was used to code the data by finding the keywords related to selfdirected learning of nurses in clinical practice as they applied to each interview question. 4. The researcher grouped the codes that were developed in the previous step by re-reading the data using the constant comparative method along with a review of the relevant literature of self-directed learning of nurses to form categories. (See an example of categories and the codes in Appendix F.) 5. The researcher wrote a summary of the findings from the interview questions based upon the categories derived from data analysis. 3.12.2 Analysis of the data gathered from the participant observations The researcher conducted the participant observation by working with each participant in each ward, for a period of days per ward. The researcher noted the important data of the social phenomenon in the field notes according to Lofland (1971) and Chantavanich (2000) of each participant on self-directed learning of nurses in clinical practices in Thai until the data of all eleven participants was gathered. The process of analyzing data gathered through participant observation was as follows: 1. The researcher used data from the field notes of participant observation to write accounts of what happened in the ward in Thai. These accounts were summaries about activities, behaviors and incidents of participants. An example of the field notes are in Appendix E. The incidents recorded happened while nurses were practicing on the ward and they

83
illustrate the way that the self-directed learning actually happened. The accounts described the nurses relationships and the participation of nurses colleagues as well as the condition of the nurses workplace environment as they practiced each day. 2. After that all the accounts were typed on the left hand side of each page. The researcher used the right hand side of the page to code the data by finding the keywords related to the self-directed learning of nurses in clinical practice. 3. The codes were grouped to form categories by re-reading the data from the field notes and the accounts using the constant comparative method along with reviewing the relevant literature of self-directed learning of nurses. 4. A summary of the data analysis for participant observation was written based upon the categories developed from data analysis was prepared. 3.12.3 Triangulating data and drawing conclusions In the third stage of the data analysis, the researcher identified the themes that related to each research question by comparing the data from the different data sources. The researcher studied all of the data of self-directed learning of nurses in each category and compared the categories from each source of data nurse interviews, nurse educator interviews and participant observations in order to identify themes according to each research question. This resulted in the merging of codes and sub-categories from the different data sources into categories for each research question. These categories were the preliminary findings of this research. The codes, categories are shown in Appendix F. Each category was thought to be particular to the phenomenon of self-directed learning in clinical practice nursing. The data analysis concluded by documenting the emergence of a singular primary pattern, found in each of the data texts and representing the most essential data (Diekelmann and Allen, 1989). 3.13 Reliability and Trustworthiness Several features in the design and conduct of the study were included to ensure reliability. These were the use of multiple sources of information (nurses and nurse educators), the use of multiple sites within the hospital (different wards) as well as the use of multiple methods of data collection

84
(interviews and observations). The setting and participants of the study have been described so that the findings can be understood in this context and applied to other settings where appropriate. Additionally, there were features in the design and conduct of the study to ensure trustworthiness. Firstly, the findings are reported in the next chapter using many quotes from participants and conclusions are drawn using the terms of the participants. Secondly, the findings have been checked by conducting a second interview. Thirdly, the roles and responsibilities of the researcher within the hospital have been described in this chapter so that the researchers knowledge and experience could be seen as related to, but separate from the findings of this research. 3.14 Conclusion This chapter has provided the reasons for selecting the chosen methodology and how the researcher implemented this methodology. Phenomenology was chosen to conduct this research because it was determined to be a useful method for researching the perceptions and experiences of nurses as they relate to self-directed learning. Given that self-directed learning is learning from the perspective of the learner, it was important to use a research method that could access those perspectives. The data were collected from two sources using two methods: interviewing nurses and nurse educators and conducting participant observation about self-directed learning of nurses in clinical practice. Data were analysed inductively by interpreting the meaning of participants perceptions as they arose. In order to make the findings of this research correct and reliable, the researcher checked and confirmed the validity of the findings with the research participants in the second interview. The following chapter will present the findings from the semi-structured interviews and the participant observations.