|
A reflection on the methodology used for a qualitative longitudinal study
Angela Evangelinou-Yiannakis
The University of Western Australia
This paper presents a reflection on the methodology used for a qualitative longitudinal study of the teaching of Modern Greek (Greek) in Western Australia under the Seconded Teachers from Greece Scheme (STGS). The study, a first of its kind, addressed an area of need in the teaching of Greek, investigating the perspectives of the key stakeholders in teaching Greek as a second language in WA under this particular scheme. The study was located within the interpretivist paradigm, with the theoretical position being that of symbolic interactionism. The methodology used was that of grounded theory and the research methods included document study, semi-structured interviewing, and participant and non-participant observation. The data analysis methods included coding, specifically open-coding and related methods. The reflection reported in this paper sheds light on the positive aspects of the methodology used for this study, as well as the issues that arose along the way, drawing conclusions that will hopefully assist other researchers undertaking this type of study.
As a qualitative longitudinal study, it provided historical background on the teaching of Greek in WA since 1915, the 'STGS' since its inception in Europe in the 1950s, the introduction of the scheme in Australia in the late 1970s, and the introduction of the scheme in WA in 1983. The study then focused on the teaching of Greek in WA, under the 'STGS', during the years 2002 to 2008, offering original theory that emerged from the findings. The study, therefore, aligns with the statement made by Thomson, et al. (2014), that QLR promises movement between the particular and the general (p. 12).
According to Thomson et al. (2014), QLR is a methodological paradigm that, by definition, moves with the times, and is an ongoing site of innovation and experiment (p. 1.). To this end, the aforementioned study fits the criteria; that is, it spans many decades of historical background on the topic, whilst delving into otherwise unexplored terrain in the near past. The study set out to address a deficit in research on this particular scheme, the secondment of Greek teachers abroad, with a particular focus on WA. Recommendations that emerged from the theory were disseminated to the key stakeholders, both in WA and in Greece, to help improve the scheme and to assist in maximising the positive aspects of the scheme in the educational providers where it is found.
According to a SAGE Research Methods (2015) video on the topic What is Qualitative Longitudinal Research?, the notion that QLR is a research design is being taken over by the notion that QLR is more a 'sensibility'. In other words, a PhD research over a period of three years can be termed a QLR; it does not have to be over a 10-15 year period in order to have longitudinal quality. Nonetheless, this particular study took six years to complete.
QLR incites a critical reflexivity about the artifice and paradoxically the reality of the research process. The QLR approach historicises social science practice, drawing attention to the artifice of our methods, the biography of the researcher and the cultural context within which 'research' is meaningful (Thomson et al., 2014, p. 14).The ongoing dynamic nature of innovation and experiment that QLR encompasses is what has inspired this paper. As Henderson, Holland, McGrellis, Sharpe and Thomson (2012) explained, it is the inadequacy of the old rules, in particular, that fail the researcher when 'conjuring' from longitudinal qualitative data. They maintained that the writing up and representation of QLR is one of the least debated but most important aspects of QLR methodology (p. 17). It is within the context of such statements that the author of this study has chosen to reflect on the methodologies used for the study, reporting on their positive aspects also but also the issues that arose along the way.
In an interpretivist study like that reported here, the individual and society are viewed as being inseparable units. Accompanying this is the view that a complete understanding of one is not possible without a complete understanding of the other. A second, and related, basic premise that exists in the interpretivist paradigm is that because all interaction between human beings is meaningful, an understanding of the meanings that create and are created by interaction between human beings is essential to an understanding of the social world (O'Donoghue, 2007, p. 16). Blackledge and Hunt (1985, p. 234) outlined the foundation pillars of this premise as follows:
The second premise of symbolic interactionism is that the meaning or perspective constructed by the individual arises out of interaction with other people. Thus, the meaning that an individual has about the world is created by the actions of other people. An alternative viewpoint is that meaning is acquired from one's experience of the world and, because one is constantly interacting with the world, that meaning may be confirmed, modified, reinforced or changed (Blumer, 1969, p. 2).
The third premise of symbolic interactionism is that meanings are dealt with and modified through a process of interpretation. Individuals attach their own meanings to 'things' and then they act towards them on the basis of these meanings. An individual selects the things that have meaning and then checks, suspends, regroups and transforms the meaning in the light of the situation in which one is placed and the direction in which one is headed (Blumer, 1969, p. 2).
According to Blackledge and Hunt (1985, p. 234), the frameworks by which people make sense of the world, in turn, have four interrelated strands, namely, intentions, strategies, significance and outcomes. In other words, when someone has a perspective on something, the framework which makes up that perspective consists of the interrelationship between the person's intentions regarding the 'thing', the strategies they say they will use to realise their intentions, the significance they attach to their intentions and stated strategies, and the outcomes they expect from pursuing these intentions and strategies. Furthermore, they can give reasons for their intentions, strategies, significance system and expected outcomes.
The strategy of grounded theory is to take the interpretation of meaning in social interaction on board and to study the interrelationship between meaning from the perception of the subjects and their actions. Through the meaning of symbols, human beings interpret their world and the actors who interact with them. Aldiabat and Le Navenec (2011, p. 1072) went on to say that grounded theory translates and discovers new understandings of human beings' behaviours that are generated from the meaning of symbols. Symbolic interactionism, therefore, is considered to be one of the most important theories to have influenced grounded theory; understanding the world by interpreting human interaction which occurs through the use of symbols, such as language. The grounded theorist's task is to gain knowledge about the socially-shared meaning that forms the behaviours and the reality of the participants being studied. Hence, grounded theory is an approach for looking systematically at, mostly, qualitative data like transcripts of interviews or protocols of observations, aiming at the generation of theory.
Grounded theory methods of data analysis, as outlined in the work of Strauss and Corbin (1990), were utilised in this study. The analytic approach that was taken involved a line-by-line analysis of the data which was then developed into categories and related sub-categories to form the basis of the theory. Corbin (1986, p. 102) explained that this is "an intricate process of reducing raw data into concepts". It involves the use of explicit coding and analytic procedures which are designed to assist the researcher to generate theory that is integrated, consistent, close to the data, and plausible (Glaser & Strauss, 1967, p. 105).
Coding is a process through which constant questioning and constant comparison of data takes place. It represents "the operations by which data are broken down, conceptualised, and put back together in new ways; it is the central process by which theories are built from data" (Strauss & Corbin, 1990, p. 81). The codes are attached to the empirical material and are formulated as closely as possible to the data. Categorising creates a summary of the codes into concepts which are, in turn, clustered into generic concepts and their relations (Strauss & Corbin, 1990, p. 90).
The particular coding approach that was used in the analysis of data in this study was open coding. The process is one of "breaking down, examining, comparing, conceptualising, and categorising data" (Strauss & Corbin, 1990, p. 61). Through this process concepts were drawn from the data. They were then identified and developed in terms of their properties and dimensions. The constant questioning process involved in this type of data analysis challenges one's own and others' assumptions about the phenomenon under investigation leading, in turn, to new discoveries (Strauss & Corbin, 1990, p. 62). A reflection of the methodology and the research methods used will now follow.
Issues related to the use of a quantitative depiction of the theory in a qualitative study
Whilst there was no substantive negative aspect to the use of grounded theory in this study, there was one issue that did arise. As the generation of theory began to emerge from the use of grounded theory methodology, the researcher chose to visually represent this theory by the use of a graph (Figure 1).
This graph, produced in the early stages of the development of the theory, helped the researcher to clearly visualise and conceptualise the theory so as to be in the favourable position of explaining the theory more clearly in the written form. The graph depicts the
Figure 1: The three stages through which the key stakeholders moved in 'dealing with'
the curriculum for teaching Greek as a second language under the STGS in WA
theory of the key stakeholders passing through three different stages as they dealt with the phenomenon of the teaching of Greek as a second language in WA under the STGS. In the infancy of the scheme the graph depicts the 'Stage of idealism'. Soon after, 'Stage of idealism' is followed by the 'Stage of conflict'. Eventually, 'Stage of conflict' is followed by the 'Stage of compromise', later to be titled more accurately as the 'Stage of cooperation'. Furthermore, the researcher chose to colour-code the various stages, lending a more visually symbolic meaning to the graph; that is, the green line depicted idealism, the red depicted conflict, and the blue depicted compromise, later labelled cooperation. In addition to the colour-coding, the researcher was able to demonstrate visually how the stages ascended to a peak, after which they declined to a trough, during which time the next stage emerged. The peak of each stage of the theory represented the 'point of no return'; the point from which the next stage began to emerge.
The researcher was also able to demonstrate through the use of a graph how each of the stages did not disappear entirely; rather, they plateaued and then remained subdued, overtaken by the subsequent stage. The underlying existence of each of the three stages was evident from the data analysis, making this visual representation a neat way of showing visually how the stages manifested themselves with the key stakeholders.
Finally, the graph also allowed for visual representation of the way in which the 'Stage of idealism' began to grow again, albeit in a less defined, less sturdy way, following on from the conclusion of the five-year tenure of the seconded teachers and the commencement of the tenure of new seconded teachers. The green dotted line in the graph represents a new, more cautious 'Stage of idealism' for all stakeholders involved in the phenomenon of the teaching of Greek as a second language in WA under the STGS. This lies in stark contrast to the solid green line at the start of the graph which depicts a sturdy rise of the 'Stage of idealism' before the commencement of the five-year tenure of the seconded teachers in their host school/s and host environment.
Despite the neat visual representation of the theory through the use of a colour-coded graph, the researcher was not able to use this approach for the following reason; the quantitative elements of the graph would demand quantitative analysis methods that, in turn, introduced a 'mixed methods approach' to the otherwise qualitative study. According to Creswell (2014), a mixed methods approach is defined as combining or integrating qualitative and quantitative research and data in a research study. Creswell explains that qualitative data tends to be open-ended without predetermined responses whilst quantitative data is usually inclusive of close-ended responses. He goes on to state that the field of mixed methods research is relatively new, dating back to the mid-to-late 1980s, although its origins lie in the late 1950s (p. 14). Therefore, as the study was purely a qualitative study, the graph, which demanded quantitative descriptors regarding its meaning, was not suitable for use. Despite this, however, the graph remained ever-present in the background as a draft document, acting as a reminder to the researcher of the way that the theory could be represented visually, and assisting in the written explanation of that theory.
Issues related to the study of documents
One negative aspect of the study of documents for the collection of data was the unavailability or nonexistence of some documents. A specific example is the 'Information Pack' that was prepared by the Greek education authorities in Greece, in consultation with some Australian cities, and provided to the prospective seconded teachers at their week-long Professional Learning opportunity in Greece in 2003, during which time the researcher was present. Within this 'pack' was information on a number of Australian cities, such as Adelaide, Canberra, Hobart, Melbourne and Sydney, as well as information on New Zealand. However, there was no information on Brisbane, Darwin or Perth, constituting a problem from the outset for those seconded teachers who were destined to teach in these cities.
Another negative aspect of the study of documents was that they sometimes became quickly outdated, particularly curriculum documents which are often replaced by new governments. In the case of this study, the Curriculum Framework of WA was the key curriculum document being studied for the purpose of the collection of data, soon to be replaced by the new Australian Curriculum. The different theoretical and pedagogical approaches to the teaching of languages education in Australia, and more specifically in WA is plainly evident in each of these documents. However, as the research was set in a particular timeframe, it is less of a flaw and more an aspect of historical awareness for the reader.
Along with the out-dated nature of documents comes the change in terminology and nomenclature. This too comes with the territory of changes in curriculum documents and related ideologies. A simple example is the reference to 'Languages other than English' (LOTE), later replaced by 'Languages education'. Such changes serve to 'date' a study but do not necessarily detract from its accuracy or authenticity, due to the fact that the research takes place within a set time period.
Issues related to the use of the 'hypothetical approach'
The only negative aspect of the use of the 'hypothetical approach' was that it did not lend itself to being used by all participants as a source of data collection. Because of the way in which the questions were framed, it applied mostly to staff of the host school in WA and, therefore, could only be used by the currently-serving seconded teachers, and the mainstream and managerial staff members of the host school who chose to take part. The approach was not relevant to other stakeholders, such as the Greek education authorities in Greece, thus eliminating it as a source of data collection from such participants.
... simply conversations in which you know what you want to find out about - and so have a set of questions to ask and a good idea of what topics will be covered - but the conversation is free to vary, and is likely to change substantially between participants (p. 65).This form of questioning provided for a relaxed and comfortable atmosphere wherein the participants could talk freely about the topic at hand, often opening up new pathways for the researcher to follow. Open-ended questions, stemming from the central guiding question of "How did the key stakeholders 'deal with' the curriculum for teaching Greek as a second language under the 'Seconded Teachers from Greece Scheme' in Western Australia?" were employed to elicit responses. Furthermore, the setting for most of the semi-structured interviews was chosen carefully by the researcher, and in negotiation with the participants, so as to provide for a convenient, comfortable and safe environment where the participants could relax and speak openly on the questions asked of them.
Whilst some members of the Greek studies, managerial and mainstream staff were interviewed on the school grounds, at times convenient to them, others were interviewed off site for reasons of practicality, privacy, and for a non-threatening open discussion to take place. Furthermore, the researcher was able to employ the methodology both in Australia and overseas. Semi-structured interviews were used to canvass the perspectives of the Greek education authorities and the prospective seconded teachers in Greece, as well as various key stakeholders in WA.
The nature of open-ended questions, stemming from a central guiding question, is such that it allows the researcher to adjust the questions to suit the individual, pair or group of people being interviewed, so as to elicit rich responses and possibly even lead to new ways of thinking about the research question at hand. Denzin and Lincoln (1994, p. 353) stated that whilst the interview is a conversation, it is not a neutral tool, as the interviewer creates the reality of the situation where the answers are given. For this reason, the interview produces situated understandings grounded in specific interactional episodes and the personal characteristics of the interviewer can, therefore, come into play. In this regard, and as stated previously, the researcher was careful to conduct the interviews in a neutral place to avoid influencing the discussion in any way. The possibility of meeting off site was often welcomed by the participants, particularly if the location was somewhere between school and home, and/or somewhere where morning or afternoon tea could be had. Furthermore, the researcher's flexibility in terms of times offered to participants and the number of participants she was willing to interview (from one to a group of six or seven persons), made for an enjoyable experience for all involved.
Another positive aspect of the use of semi-structured interviews was that the interviews were audio-recorded and translated from Greek to English, where necessary, later to be transcribed by the researcher. The taping of the interviews meant that the participants could be offered the opportunity to read the relevant interview transcriptions, allowing thus for transparency to take place as well as checks against the accuracy and integrity of the data. Furthermore, the audio recordings and their transcriptions could be kept in a locked cabinet, thus meeting the requirements of the Ethics Committee of The University of Western Australia.
Finally, the methodology of semi-structured interviews tied in perfectly with the paradigm of interpretivism which underpinned the study because of its emphasis on social interaction as the basis of knowledge (Evangelinou-Yiannakis, 2015, p. 161). The researcher was able to canvass the perspectives of the key stakeholders in both Greece and Australia, as well as in Germany, in a comfortable and enjoyable way for all involved. The open-ended nature of the questions meant that conversations could flow and, as a result, much data could be collected.
Issues related to the use of semi-structured interviews
One of the potential problem areas for the researcher in using the data-collection methodology of semi-structured interviews was that of keeping at bay her own prior knowledge and any preconceived biases she might have on the topic, during the interviews. The researcher, having worked closely with the seconded teachers from Greece at the host school prior to and during the research phase of the study, was careful to distance herself from the research topic under investigation. The nature of semi-structured interviews is such that it allows for the possibility of bias to creep into the questions and the subsequent discussions on the topic; however, the researcher was careful at all times to avoid falling into this trap. Also, as stated by Denzin and Lincoln (1994, p. 353), the personal characteristics of the interviewer can influence this method of questioning. Hence, the researcher was on her guard against this potential problem during all interviews.
One of the first problems to manifest itself during the interview phase of the research was that some participants did not attend the scheduled interview times. The group that was the worst offender in this regard was the parents of the host day school and the host after-hours Greek school providers. One reason for this apparent apathy was the need to attend to other younger children after first having dropped off their older sibling/s at school; other reasons included work commitments or the need to attend to household duties. The researcher, after trying a number of times to elicit the involvement of these participants, decided to offer coffee meetings (at her own expense) at a coffee shop of the participants' choice and at a time most suited to them. This approach proved very successful and a number of interviews with parents were held thereafter.
Other negative aspects of semi-structured interviews are the length of responses which, in turn, lead to much work in terms of the translation, where necessary, from Greek to English, and the transcribing of the interviews for the purpose of analysis. Because open-ended questions can lead to meandering discussions at times, the researcher can find that there is more to translate and/or transcribe than would be the case with the closed-questioning techniques of structured interviews.
These methods assisted the researcher to develop an understanding of how the participants acted in light of their perspectives and the changes, if any, which took place in the participants' perspectives as a result of their actions. As has been stated previously, the perspectives of the participants regarding the teaching of Greek as a second language under the STGS in WA, were explored through document study and semi-structured interviews. The participants were then given the opportunity, through participant and non-participant observation, to elaborate further on their perspectives and their actions, and to clarify questions held by the researcher regarding any perceived discrepancies between stated intentions and strategies, and reasons given for these.
The researcher, having been a participant observer herself of the phenomenon being researched, was able to have a deeper understanding of the issues raised by the participants and was also in a position to form an opinion on matters being discussed. It was, therefore, of paramount importance that the researcher distance herself and try to keep any bias at bay from the recording and analysis of the data.
Issues related to the use of participant and non-participant observation methods
The key negative aspect of participant and non-participant observation in this study was that the researcher had been a participant observer of the teaching of Greek as a second language under the STGS in WA for a number of years before embarking on the actual research of the phenomenon. She had been working with the seconded teachers from the outset of their tenure in WA and knew, at close range, some of the issues that arose on a day-to-day basis at the host school. This was problematic in that the study could be tainted with biases from the researcher's perspective. Hence, the researcher chose to distance herself physically from her workplace in order to be able to conduct the research without perceived bias. This entailed taking extended leave for the duration of the study.
Furthermore, as a non-participant observer, the researcher was often very familiar with matters being raised during the semi-structured interviews so that she had to make a concerted effort not to influence the course of the discussion in any way. This required much discipline in terms of not interceding with commentary during the interview phase and, with the careful line of questioning, to avoid guiding the discussion down an expected pathway. At all times, therefore, the researcher needed to be on her guard against the familiarity that being a participant observer can create.
The open coding approach that was used in the analysis of data in this study led to concepts being drawn from the data. These concepts were identified and developed, according to their properties and dimensions. While pursuing this coding process, the researcher asked questions about the data and made comparisons for similarities and differences between each incident, event, and other instances of the phenomenon under investigation. In this way, similar events and incidents were labelled and grouped to form categories (Strauss & Corbin, 1990, p. 74).
Code notes were written throughout the data-analysis process and theory-development phases of the study. Such notes are a specific type of memo prepared by the researcher to describe and explain the conceptual labels which emerge from the data. They contain the products of inductive and deductive thinking about relevant and potentially relevant categories (Strauss & Corbin, 1990, p. 74).
After engaging in these coding procedures, the researcher adopted the 'analytic induction' mode of analysis. This involves the formulation of categories and themes in a systematic way. The steps involved are described by Hammersley and Atkinson (1995, p. 234) as follows:
In the process of analysing the data for the study, it was also found useful to utilise the three sub-processes noted by Miles and Huberman (1994, p. 428), namely, data reduction, data display, and conclusion drawing and verification. Data reduction included coloured coding, the use of symbols and annotations, note-making, comparison and contrast, noting of patterns and themes, prioritisation of recurring points or themes, and clustering. Data display involved the organisation in compressed form of data, such that points were ordered in a hierarchical manner of importance. The degree of importance was determined by the recurrence of the themes.
Three hand-drawn matrices were developed as part of the data display sub-process. Within each matrix, the points were clustered and prioritised in terms of their importance. Conclusion drawing and verification involved the previously stated six-step approach, as well as triangulation, looking for negative cases, following up surprises, and checking results with respondents. This, in turn, facilitated the achievement of the aim of this study, namely, to generate theory on how participants 'dealt with' the phenomenon under investigation.
Issues related to the data analysis methods
The only issue to arise from the use of open coding was the recurring concern by the researcher that her own biases would interfere with the analysis of the data; that she would bring her own past knowledge and experience of the phenomenon being researched into play when conducting, for example, the line-by-line analysis of letter excerpts and interview transcripts. However, this was overcome by the continual verification and triangulation processes that took place during the analytical phase of the research; looking for negative cases and following up any surprises, as well as conducting member checks.
It is evident throughout that all of the methodologies employed by the researcher were advantageous for this particular study, as they were suited to the paradigm of interpretivism which is, in turn, set within the theoretical position of symbolic interactionism. The various issues that arose from the use of each methodology were relatively less significant than the obvious advantages, and were not debilitating to the research process overall.
Aldiabat, K. & Le Navenec, C. (2011). Philosophical roots of classical grounded theory: Its foundations in symbolic interactionism. The Qualitative Report, 16(4), 1063-80. http://www.nova.edu/ssss/QR/QR16-4/aldiabat.pdf
Blackledge, D. & Hunt, B. (1985). Sociological interpretations of education. Sydney: Croom Helm.
Blumer, H. (1969). Symbolic interactionism: Perspective and method. New Jersey: Prentice Hall.
Charon, J. M. (1989). Symbolic interactionism: An introduction, an interpretation, an integration. New Jersey: Prentice Hall.
Corbin, J. (1986). Qualitative data analysis for grounded theory. In W. C. Chenitz & J. M. Swanson (Eds.), From practice to grounded theory (pp. 91-102). California: Addison-Wesley.
Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods approaches. California: SAGE.
Denzin, N. & Lincoln, Y. (Eds.) (1994). Handbook of qualitative research. Thousand Oaks, CA: SAGE.
Denzin, N. & Lincoln, Y. (Eds.). (2013). Strategies of qualitative inquiry. Thousand Oaks, CA: SAGE.
Evangelinou-Yiannakis, A. (2015). Teaching at home, teaching abroad: A new theory for languages education. Rydalmere, NSW: AVAGO Books.
Faggiolani, C. (2011). Perceived identity: Applying grounded theory in libraries. Italian Journal of Library, Archives and Information Science, 2(1), 1-33. http://dx.doi.org/10.4403/jlis.it-4592
Fylan, F. (2005). Semi-structured interviewing. In J. Miles & P. Gilbert (Eds.), A handbook of research methods for clinical and health psychology (pp. 65-78). Oxford: Oxford University Press.
Glaser, B. & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine Publishing Co.
Hammersley, M. & Atkinson, P. (1995). Ethnography: Principles in practice, 2nd ed. London: Routledge.
Henderson, S., Holland, J., McGrellis, S., Sharpe, S. & Thomson, R. (2012). Storying qualitative longitudinal research: Sequence, voice and motif. Qualitative Research, 12(1), 16-34. http://dx.doi.org/10.1177/1468794111426232
Lincoln, Y. S. & Guba, E. G. (1985). Naturalistic inquiry. California: SAGE Publications.
Martin, P. Y. & Turner, B. A. (1986). Grounded theory and organizational research. The Journal of Applied Behavioural Science, 22(2), 141-157. http://dx.doi.org/10.1177/002188638602200207
Miles, M. B. & Huberman, A. M. (1994). Data management and analysis methods. In N. Denzin & Y. Lincoln (Eds.), Handbook of qualitative research (pp. 428-438). Newbury Park, CA: SAGE.
O'Donoghue, T. (1996). Malaysian Chinese students' perceptions of what is necessary for their academic success in Australia: A case study at one university. Journal of Further and Higher Education, 20(2), 67-80. http://dx.doi.org/10.1080/0309877960200206
O'Donoghue, T. (2007). Planning your qualitative research project: An introduction to interpretivist research in education. London: Routledge.
Ritzer, G. (1994). Sociological beginnings: On the origins of key ideas in sociology. New York: McGraw-Hill.
SAGE Research Methods Video (2015). What is qualitative longitudinal research? http://dx.doi.org/10.4135/9781473930162
Smith, D. & Lovatt, T. (1991). Curriculum: Action on reflection. Wentworth Falls, NSW: Social Science Press.
Strauss, A. & Corbin, J. (1990). Basics of qualitative research. California: SAGE.
Thomson, R., Hadfield, L., Holland, J., Henwood, K., Moore, N., Stanley, L. & Taylor, R. (2014). New frontiers in QLR: Definition, design and display. National Centre for Research Methods, UK. http://eprints.ncrm.ac.uk/3297/1/new_frontiers_in_QLR.pdf
Woods, P. (1983). Sociology and the school. London: Routledge and Kegan Paul.
| Author: Dr Angela Evangelinou-Yiannakis is an Honorary Research Fellow in the Graduate School of Education, The University of Western Australia, from where she received her doctorate in Education in 2009. She has been a secondary school teacher since 1980 and a leader in independent schools since 1994. She currently holds a position with Deakin University and Teach for Australia. Email: angela.evangelinou-yiannakis@uwa.edu.au Please cite as: Evangelinou-Yiannakis, A. (2017). A reflection on the methodology used for a qualitative longitudinal study. Issues in Educational Research, 27(2), 269-284. http://www.iier.org.au/iier27/evangelinou-yiannakis.html |