Chapter Four: Research Methodology: the concept of devolution in PNG education in context of the colonial legacy - PART 1
In "PROBLEMS WITH DEVOLUTION OF PNG EDUCATION" by Dr Gabriel Kulwaum, Adminstrator, Manus Province
In the previous chapters, I have provided general theoretical and historical backgrounds to the research problem, which I consider to be essential for understanding the problems of devolution in PNG education. In Chapter Two I have argued that PNG is a colonial construction and as a political entity it remains contested. I have also suggested that in order to maintain PNG's unity, the colonial powers had constructed a highly centralised bureaucracy. After independence, despite many attempts to move towards a more devolved system of governance, this bureaucracy remains entrenched in PNG. Subsequently, the democratic reforms which have been initiated in PNG over the past two decades have had to filter through this well-established bureaucratic structure. I have also maintained that post-independence PNG educational politics have mostly revolved around the debates about how power should be shared between different levels of government and with one bureaucracy. These debates have been over the issues of how the conflicting imperatives of the nation's cultural diversity and its political integrity can be reconciled; of what should be the relationship between the national and provincial governments; and of how the development of PNG's education system can respond to the contradictory demands of indigenisation on the one hand and western education on the other. I have argued that these questions cannot be discussed adequately outside the context of the colonial legacy that persists in PNG. The concept of devolution is rooted within the controversies that surround attempts to resolve these difficult questions.
In Chapter Three I suggested that the idea of devolution is a contested one. Its meaning depends on the particular social, political, economic, and educational contexts of its use. I also argued that the interpretation and thus the meaning given to the idea of devolution often depends on the justifications put forward in its support. These justifications are not only located in PNG's uncertain local politics, but they also have had to negotiate the administrative ideas that PNG has inherited from Australia, as part of its colonial legacy. At independence, PNG adopted the policy of devolution for public administration as part of a strategy to articulate its post-colonial future and its national development. And to a large extent, PNG has remained committed to the idea of devolution both as a way of ensuring the nation's political unity and stability, and also as an implementation strategy for the efficient and effective delivery of public services. However, while PNG has this unquestionable commitment, its views on devolution are framed by its colonial legacy, and also by the emerging political, economic, cultural and professional pressures that the nation faces as it struggles to establish its regional identity. PNG has thus found it difficult to establish an equilibrium between its democratic aspirations and its colonial bureaucratic reality.
Despite these political problems, PNG has, over the last two decades not only remained committed to the policy of devolution, but also sought to establish a range of administrative mechanisms designed to implement it. PNG has been particularly active in devolving decision-making powers in education. However, it is now widely recognised that attempts to realise devolution in education have not been entirely successful. Many in PNG regard the policy of devolution as a failure: there has been a great deal of frustration among the key players, as well as among the community. The research in this thesis accepts the contention articulated in the Hesingut, the Pokawin and the Micah reports, that there have been numerous difficulties surrounding the policy of devolution in PNG; but this thesis seeks to ask also the obvious research question: why has it been so difficult to implement the policy of devolution in PNG educational governance? This thesis is concerned, not so much with the question of whether there have been problems with the policy of devolution in PNG education, but rather with the nature of these problems and how to account for them. This is an important research question as PNG moves forward to realise its post-colonial democratic aspirations. But how should this question be researched? This chapter provides an overview of the methodology employed for the research reported in this thesis.
In this chapter, I argue that the complexities surrounding the problems of devolution in PNG educational governance cannot be adequately addressed by relying upon quantitative research methods; rather a more complex qualitative approach is needed. In the next section I survey some of the main theoretical issues which researchers in education confront when making decisions about methods most appropriate for researching particular research problems. My survey revolves around a series of binaries which often define the terrain of much of the methodological discussions in educational research. This discussion enables me to demonstrate a preference for a qualitative approach to educational research. Among the various qualitative methods, I suggest that the case-study method is most appropriate for addressing my research question. I discuss specifically how the interviewees were selected, how the interviews were conducted, and how the interviews were related to the selection and analysis of public policy documents. Finally, I then discuss how the data were analysed to derive common issues emerging from the field. These issues are then assembled around three themes which form the basis for organising the subsequent chapters.
Issues in Educational Research
Like any other researcher in education, when I commenced this project I was confronted with a range of issues which relate to the most appropriate method for researching the particular question I wished to explore. A literature review revealed that much of the discussion around methodology in educational research was concluded in terms of a series of binaries. I have therefore chosen to present my own discussion of methodology in relation to these binaries.
A review of the literature on educational research suggests two main approaches: quantitative and qualitative. Lee (1992:87) argues:
Lee (1992:88) contends that the quantitative and qualitative research methods are based on different "ways" of "seeing" the world, but ultimately the researcher's choice of research method depends on the "aims of the research enquiry, the various roles of the researcher, and the researcher-respondent relationship." However, Lee also argues that in choosing either a quantitative or qualitative method a researcher in education will be confronted with a series of binaries: statistics versus description, objectivity versus subjectivity, positivism versus phenomenology, universality versus particularity, outsider versus insider, and involvement versus detachment.
The most fundamental issue confronting all researchers has to do with the kind of data required and how this data is to be presented. The data may be presented either in statistical or descriptive form. The quantitative researcher mostly places emphasis on the importance of statistical data. It is assumed that the social realities and what people say and do can be described and explained in terms of broad generalisations that can be captured by the use of statistics. In contrast, the qualitative researcher focuses on the complexities of the representations of social realities. It is suggested that such complexities can be described in terms of statistical terms, but require reference to the values and attitudes underpinning human action. It is maintained by the qualitative researcher that the quantitative representation in the form of statistics distorts the nature of human affairs, the real contexts or the actual experiences people have. Representing human actions out of their contexts can thus distort the way people interpret their own conduct. So although the quantitative method is useful in representing some aspects of human reality, it is not always possible to use statistical data to represent the feelings and the perceptions of people. Qualitative methods are more appropriate in view of the description of diverse and often conflicting sets of opinions that people hold. Such methods enable respondents to describe the way in which they see problems and social realities.
A second binary within recent conceptions of educational research, relates to a distinction concerning issues of ontology, that is, "oncerning human knowledge and its relationship to the world" (Lee 1992:88). This binary relates to the researcher's wish to view the human world as either having an objective or subjective character. Burrell and Morgan (1979) suggest that the objectivist view is based on the ontological assumption that the social world external to individual cognition is a real world made up of hard, tangible and relatively immutable measures. It is also claimed that this world exists independently of an individual's appreciation of it. This view describes society as being made up of structures, concepts, labels and relationships; and it is thought that what makes this world tick are natural laws and principles, which need to be discovered and proven through scientific means. In contrast, Burrell and Morgan (1979) contend that the subjective view of the world derives from the assumption that while the social world external to individual cognition is made up of names, concepts and labels, these are social and historical creations. Burrell and Morgan also maintain that their utility is based upon their convenience as tools for describing, making sense of, and negotiating the external world. The subjectivist view thus focuses on the social construction of people's ideas and concepts. This approach is about people--how and why they interact with each other, and their motives and relationships. It is also about the "deeper" meanings to social actions; how these are interpreted, understood and appreciated by individuals and groups.
This concern for meaning highlights a third binary--between positivism and phenomenology--that has been widely discussed in the literature on educational research. This binary has to do with the question of epistemology--what is knowledge, how it is "revealed", "discovered" or constructed and how claims to knowledge are justified. How people perceive knowledge is often defined by their views about the world. In this way, issues of ontology and epistemology are linked. Positivists view knowledge in terms of the principles and laws which regulate the existence of the processes and the relationships of the social and physical world. Positivists believe that knowledge can be "revealed" or "discovered"; and that "discovered" data enables us to provide possible explanation of the causes of things that happen in the world--often independently of the intentions of people. Such causes can only be discovered through the application to research of systematic scientific methods designed to reveal underlying regularities in the world. Some of these research methods include survey, quasi-experimentation, experimentation and questionnaires. To discover the pattern of regularities in the world, the researcher is expected to come up with a hypothesis and test it in order to establish a reliable and valid pattern which can form the basis of general theories. This can be done through systematic observation on one hand and on the other manipulation of independent and controlled variables. This approach emphasises experimentation, observation, control, measurement, reliability and validity in the various processes of research.
In opposition to this view, phenomenologists believe that the world is made up of people with their own intentions, attitudes and values. Their beliefs and values are reflected in the way they see and do things, and why they do things in a certain way. These beliefs and values are reflected in their actions and behaviours. If the social world is constructed of individuals, then, the phenomenologists argue, its investigation requires a different approach. This approach is to look at human events not in their isolated parts, but in a more holistic perspective that locates individual actions in their cultural contexts. It is suggested that human activities must be investigated in terms of meanings--why people say this, do this or act in this or that way. This approach requires interpretations on the part of the researcher which are linked to other human events to enable greater understanding, and an appreciation of the phenomena in totality. This view calls for a holistic approach: that is, how might one isolated activity be linked to other related activities that give that isolated activity its meaning and its significance. As Lee (1992:90) argues:
A fourth binary concerns the findings of the research: whether the findings could have universal applications and thus be generalised to all contexts, or whether the findings could only be applied to specific cultural and historical sites. The positivists argue that the scientific research method is a tool for data collection, producing precise, systematic and theoretical answers to the research question. It is also suggested that the utilisation of the scientific approach guarantees answers which are neutral and technical and thus could be universalised and generalised to all historical and cultural contexts, regardless of the particular features of the situation under study.
Subjectivists, on the other hand, maintain that precise, systematic and theoretical answers to complex human problems are not possible. It is contended that every cultural and historical situation is different and unique and thus requires analyses of the uniquely defined, particular contexts in which it is embedded. The subjectivists argue the specific social, political, economic and cultural experiences cannot be generalised.
The fifth binary has to do with whether the researcher (outsider) or the researched (insider) is in the best position to be able to decide on the parameters of research. The issues concerning the aims of the research--what data are to be collected, and how they are to be analysed and presented--are linked to the questions which are defined by this binary. Positivists believe that the knowledge is validated by the researcher going into the field, with aims and interests which are independent of those being researched. They argue that in the research process the researcher defines the aims of the research, the kind of data to be collected; how this data is to be collected and analysed; and the issues concerning the use of research are separate from issues concerning its production. Those with most to gain or lose are thought to be too close to be able to assess how research ought to be organised. Evered and Louis (1981) call this approach to thinking about research an "inquiry from the outside". The positivists thus consider the "outsider" to have the privileged voice.
In contrast, subjectivists maintain that knowledge should emerge out of the local context and should privilege the voice of the "insiders", taking into account what people say, do and feel, and what they desire to be the outcomes of research, should provide the basis for the data collection. Here, patterns, trends, themes should emerge from the research process. The role of the researcher should be to understand real-life situations from the point of view of the insiders rather of outsiders. The emphasis is thus placed on the respondent frame of the reference on how they see things from within . It should not be the researcher who decides what counts as knowledge, but what the respondents view as knowledge, emerging from interactions between the respondents and the researcher. Evered and Louis (1981) call this an "inquiry from the inside".
The sixth binary between detachment and involvement is linked to these considerations; and relates to the role of the researcher--whether the researcher should be detached or involved in constructing data considered relevant and significant. Positivists regard the process of researching as neutral. It is assumed that the researcher should maintain a position of neutrality and distance; and that the researcher's role should be seen as an instrumental one--concerned with the collection of data. Only by being detached from the research process can the researcher establish the validity and reliability of the findings. It is assumed that by being detached from the process, the researcher will not be able to manipulate and influence the process and the outcomes of the research. This assumption is strongly contested by subjectivists who maintain that the research process is never neutral: the researcher's values enter the processes of research in a variety of ways and levels. It is argued that the researcher is part of the process of research and cannot be detached from it. What the researcher selects (for instance, as the aims, techniques, and procedures, the type of data to be collected, the interpretations of the data, and the findings of the research; as well as it uses) is most likely to be informed by the values that the researcher holds. In this perspective, the relationship that the researcher is able to develop with the researched is part and parcel of the research context.
Preference for Qualitative Method
The gneral research issues defined by these binaries confront most educational researchers. In developing my methodological approach, I was guided by the need to develop a stance in relation to them. For reasons that will become clearer, to me the qualitative approach seemed the most appropriate way of exploring the problems of devolution in PNG education.
In its broadest outline, the methodological approach that has guided my research involves a rejection of scientific/quantitative/objective methods. I argue that this quantitative mode of research is not only inappropriate on theoretical grounds, but it is also inconsistent with the ideological and personal convictions which I have brought to this research project. Theoretically, I support the arguments of Kincheloe (1991:143) who suggests:
An examination of the complex problems of devolution in PNG education does not lend itself to "numerical expression". This is so because the factors involved in the promotion and implementation of administrative and organisational reforms in PNG education are diverse, multiple and often contradictory. The qualitative approach, on the other hand, is more consistent with the aims of the research, and with the theoretical approach as discussed in Chapter Three relating to the notion of devolution. There is also the matter of the inevitable involvement of the researcher in the research process, which would ensure the type of data collected and the way it is analysed and presented are most likely to be affected by the researcher's history of involvement in PNG education administration.
As already indicated, this study aims to explain the complex problems of democratic reforms in PNG education governance. But such problems do not exist in abstraction; rather, they are perceived and constructed by people in particular ways. To gain access to their perceptions requires close interactions between the researcher and the respondents in the research process. The researcher is very much part of the research process rather than being detached from it. The data collected relate to the views, opinions and perceptions of individuals which are based on their experiences. The analysis of the data involves the exercise of interpretation by the researcher, but the data is interpreted by the researcher in a particular way: it is an attempt by the researcher to read into the meanings of what the respondents think, feel and say about the problems of devolution in PNG education. It should be noted, however, that the findings are not so much manipulated by the researcher, as framed in a particular explanatory form, often using the exact words of respondents.
Because of these subjective aspects of the research aims, a quantitative approach is unlikely to prove useful. Since issues of value, history and politics are inevitably involved, a qualitative approach is more likely to reveal the complexities and the multiple voices that exist in PNG educational administration. This line of thinking is consistent with that of Kincheloe (1992:61) who argues:
Indeed, a review of the literature on educational policy research shows that the qualitative mode of enquiry is being increasingly employed to investigate the complex dynamics of policy, and administrative processes (Halpin and Troyna, 1994). This is so because politics lies at the heart of policy not only at the policy-production level, but also at the level of policy reception and implementation. Policy analysts have come to use the qualitative modes of enquiry in the attempt to capture the political world in which policies circulate, are shaped, articulated and re-articulated. This is a world in which there are particular sensitivities, interests, pressures which lead to the creation of a complex world of multiple meanings--and of the exercise of power, sometimes in a complex pattern and also in unpredictable ways.
A review of the Australian literature in educational policy research suggests that qualitative modes of enquiry have been used in conducting research into a whole variety of administrative issues including the goals and objectives of education, styles of leadership, procedures for decision-making, curriculum implementation and evaluation, review of innovations, teaching styles (strategies), the role and work of principals, staff relationships, and staff-student relationships, as well as other organisational issues such as democratic reforms in Australian education. Indeed, as Carr and Kemmis (1986) argue, qualitative approaches do not any longer need the defence they once did, when positivism held almost total supremacy. Subjectivist and more recently post-structuralist views have eroded the dominance which positivism once held in educational research.
A review of the literature in PNG education also indicates that most educational researchers employ qualitative methods, such as ethnography, participation-observation, case-study and in-depth interviews. In the context of PNG, there are two additional reasons for this. One of these concerns the matter of cultural diversity and complexity. PNG is composed of many different language and tribal groups and generalisations are at best dubious. And second, there is the fact that almost 80 percent of the PNG population cannot read, and would therefore be unable to participate in any research exercise that requires complex anonymous surveys. Face-to-face interviews are the only viable alternative. Accordingly, much of educational research in PNG has utilised qualitative methods. For example, Cheetham (1979) has used in-depth interviews to study the perceptions of the Huli people (a tribe in PNG) on the relationship between the school and community. Smith, Carss and Power (1979) conducted a study on teachers' perception of their role in the schools in the community. Carrier (1982) used participant-observation in the study of the impact of western education on Ponam Island (Manus, PNG) traditional cash economy. Martin (1985) used case-study to investigate people's perceptions of community participation in Buka Island, Bougainville.
As well as these theoretical arguments, the qualitative approach to research is also consistent with my ideological and personal convictions concerning the nature of the social organisation, as well as the cultures of PNG. I believe that science cannot be the source of all knowledge; and it cannot adequately comprehend the nature of the spirit world through which the people in Melanesia make sense of their world. Narokobi, (1983:3) one of the few Melanesian philosophers, refers to this world as representing the composite truth in Melanesian "cultures, values, knowledge and wisdoms". He argues that this truth can be revealed in the interdependent relationship between persons, plants, animals and the spirit world. Narokobi (1983:6) argues:
This view suggests that the very existence of life, for Melanesians, implies that people and events are interrelated. The Melanesians, including all Papua New Guineans, view society in its totality, and have a holistic view of life. Papua New Guineans find it difficult to interpret an event in isolation, as scientific rationality requires them to do. An interpretation of an event is made with reference to other social events--both past and present.
I believe that from the point of view of the Melanesian cultural background into which I was socialised, I view knowledge not in terms of depersonalised universal principles and laws, but in terms of my own being as linked to the way I come to "know" things--and thus "research" about my society. And if I, as someone who has lived in Australia for more than five years, hold this view, then the respondents in my research are likely to do the same. I am more comfortable with such a perspective. Kincheloe (1991:67) argues that:
I thus maintain that the feelings, the experiences and the meanings of what people say and do, cannot be processed by scientific rationality operating outside the framework of the cultural contexts in which they have significance. The positivist approach is not sensitive and responsive to the type of "social" data which might constitute real insights into the Melanesian world. Thus positivist approaches to research cannot adequately encapsulate the views of Melanesian research respondents. It may be possible for such approaches to answer such queries as "what is", but they cannot address or adequately explain "how things are believed the way they are" and "why things happen the way they do". I maintain that the "why" questions can only be investigated properly with the use of qualitative research method. As Halpin (1994:198) argues:
In sum, then, my preference for a qualitative approach to research is based on a range of theoretical considerations but also on considerations relating to my very being as a Melanesian studying in an Australian university, and to what I know instinctively about Melanesian people and their cultures. In a sense, mine is a post-colonial research project (Hommi Bhabha, 1994) which seeks to understand the problems which are tied inextricably to forms of colonialism. I readily acknowledge, of course, that my own research is located within a particular form of this colonialism.
In the section above, I have provided a range of reasons for preferring a qualitative approach to policy research in education. However, the term "qualitative" describes a particular philosophical orientation, and not any particular research method. Within its ambit, there are numerous methods: ethnography, participation-observation, symbolic/interactionism, phenomenology, constructivism, interpretivism and case-study (Lynn and Erickson, 1990). In this project, case-study methods are preferred over other qualitative approaches.
Adelman et al (1976:2) argue that case-study is a general term that describes a family of research methods having in common the decision to focus an enquiry around an instance, event or policy. A case-study is highly "eclectic". Bogdan and Biklen (1982:58) define it as "a detailed examination of one setting or one single subject, or one single depository of documents, or one particular event". Best (1977:127) argues that the purpose of the case-study is understanding "the life-cycle of an individual unit" which might be a family, a group, a social institution or an entire community, or indeed a policy. Walker (1980:33) defines case-study as "the examination of an instance in action. The study of particular incidents, and the selective collections of the information on biography, personality, intentions and values allow the case-study worker to capture and portray those elements of a situation that give its meaning". According to Walker (1980) the case-study has been used to portray "idiosyncratic and particular" instances, or to conduct studies where an entirely theoretical base is not appropriate. These views of what a case-study is under-score the importance of the complexities of a social situation which cannot be uncovered by any search for pattern. Yin (1989:23) argues that case-study is not only a research strategy but also an orientation empirical inquiry that "investigates a contemporary phenomenon within its real-life context; when the boundaries between phenomenon and context are not clearly evident; and in which multiple sources of evidence are used". This research project is framed by Yin's definition of a case-study.
In my view, case-study is an appropriate method to explore the nature of the social, political, economic and cultural issues that surround attempts to devolve educational decision-making in PNG. The use of case-study enables access to life experiences of real people in their actual social circumstances and permits interpretations that cannot be easily generalised. As Deem and Brehony (1994:159) argue:
The use of case-study also enables the assemblage of different points of view expressed by different people in their own way of expressing it. It makes it possible for different interpretations to be brought together so as to reveal the diversity of meanings in what people say and do--in their values, attitudes and behaviours. Also the same data can be re-interpreted at a later stage if new topics are explored, and new modes of "listening" become available. Larcher (1993:126) maintains:
Another claim made on behalf of case-studies is that they are action-oriented, that is, they present the possibilities of articulating new solutions to practical problems. Finally, it is often contended that the case-study reports are simple and free of jargon and are capable of serving purposes of educational action in a variety of ways.
The case-study method of educational research is not without its critics, however. One of its most persistent criticisms relates to its limitations. Deem and Brehony (1994) argue that case-study has been criticised for its inability to guarantee reliability and validity of data interpretations; and that since the case-study relies heavily on the human instruments (interactions between people), there is always this danger of bias and prejudice in the way data are collected and analysed. Deem and Brehony (1994:163) argue:
Deem et al further maintain that the case-study can also encounter problems of resources for effective field work. It is suggested that case-study can be time and energy consuming, while also requiring a lot of financial and material resources over a long period of time. Finally, the participants (subjects) are most likely to be identified and therefore for the researcher to collect the data through the case-study it requires ethical clearance and procedures.
In my opinion, while these objections are instructive and point to the need for practical and ethical care, others are based on the flawed assumptions of positivism. So, for example, the charge of bias and prejudice is misplaced because it rests on the view that value-neutrality in social and policy research is in fact possible. The charge concerning limitations is based on an assumption which suggests that a single universal method of research is both possible and desirable. As for ethical concerns, these apply to all forms of educational research and are not applicable to case-study in any unique way.
To continue - go to PART 2 of 2
Materials on this web site are the Copyright © of SPCenCIID and contributors. Permission to make a single copy of all or part of this material is granted provided the copies are not made or distributed for commercial advantage. Copies should include this copright statement and a statement that that permission for limited copying has been granted by the copyright owners.
To disseminate otherwise or republish, requries written permission. Contact: John Evans, SPCenCIID, Box 320, University P.O., Papua New Guinea.
Note - While efforts are made to ensure accuracy this publication could include errors or inaccuracies and no responsibility is taken for the consequences of its use. Information will be updated on a regualar basis and content can be expected to vary considerably over time.
(BUAI - stands for Books Useful Articles & Information)