Research methodology/methods: Emergence of enlightenment
The five publications presented for this PhD by prior publication emerged from the use of five
different methodologies (see table 3). The first publication (Mellor & Greenhill 2014) reported on a
study using grounded theory method (GMT) within the rural environment, and while not
immediately apparent, the first glimpses of enlightenment appeared, i.e., the need for graduate
nurses to be self-supporting. It was an exciting sojourn into the world of graduate nurses who
were often abandoned and left to advocate and care for themselves as the only nurse on duty for
their hospital.
This prompted the second study, a critical review of transition programs (Mellor, Gregoric, Atkinson
& Greenhill 2017). The study identified that many programs were not assessed using accepted
research methods/methodologies and bias was the norm. The study also revealed the variable
nature of transition programs, inconsistent terminology, inconsistent content, and the need for
formal research to determine best practice.
The third study, an extensive literature review by Mellor, Gregoric and Gillham (2017) focused on
the strategies graduate nurses use to care and advocate for themselves, and provided a valuable
background to framing the questions and the potential richness of data which could be achieved
from the fourth study (Mellor & Gregoric 2016). All the data were collected from graduate nurses,
who were members of a formal transition to practice program, targeting the ways in which
graduate nurses were able to advocate and care for themselves. Individual interviews with
graduates from both country and city locations revealed the ongoing challenges they faced. Their
resourcefulness was edifying and provided a valuable trove of self-support strategies or ‘ways of
being’ that could be used to enlighten nursing students when preparing them for their first year of
practice as a registered nurse (See Table 2).
This valuable set of data was revisited in the final study by Mellor and Gregoric (2019) with a focus
on graduate nurses and the spectrum of comfort. Further self-support strategies emerged to
provide valuable insights for nursing students, so they can practice safely as graduate nurses
within an injurious health care culture.
Throughout the following discussion, a brief review of the methodologies underpinning each of the
below studies and associated publications will be provided. They will be listed by study and
publication numbers. Please refer to Table 3 when reading the narrative, and in particular, note
that there were two publications emanating from Study 4.
Table 3. Research studies overview
Aims of research Methodology Publications which Authorship
studies: emerged
Study 1: To explore the Focus groups with 1. A patient safety- Mellor & Greenhill (2014)
nature of professional graduate nurses from 14 focused registered
support given to new rural hospitals. nurse transition to
graduate registered Grounded theory method practice program
nurses in rural areas (GTM)
Glaser & Strauss –
(Classical GTM)
Study 2: Evaluation of Critical review transition
transition programs, programs 2010-2017
interventions, and
resulting outcomes for
PRISMA search strategy
graduate nurses (Moher et al. 2009)
Kirkpatrick Model of
Evaluation
2. A critical review of Mellor, Gregoric, Atkinson
Transition-to- & Greenhill (2017)
Professional-Practice
programs: Applying a
standard model of
evaluation
Study 3: To identify the Literature review method, 3. Strategies new Mellor, Gregoric & Gillham
strategies that new 2001-2016 graduate registered (2017)
graduate registered nurses require to care
nurses require to care and Dialectical pluralism and advocate for
advocate for themselves themselves: A
during their first year of PRISMA search strategy literature review
practice (Moher et al. 2009)
Study 4: To identify the Interview with nine new 4. Ways of being: Mellor & Gregoric (2016)
strategies used by graduate nurses Preparing nursing
graduate nurses to thrive Grounded theory method students for transition
during their first year of (GTM) to professional
practice (2 publications Straus & Corbin; practice
emerged) Charmaz
5. New graduate(Constructivist GTM with Mellor & Gregoric (2019)
symbolic interactionist registered nurses and
influence) the spectrum of
comfort in clinical
practice
Study 1: Publication 1
The first study undertaken (Mellor & Greenhill 2014) was based on the classical model of
grounded theory method (GTM) and focused on graduate nurses in rural areas. This method
embraces inductive analysis and implicit inferential analysis (abductive) guided by prescribed GTM
procedures for qualitative data collection and analysis (Glaser & Strauss 1967). Verification is not
the aim of GTM, so there were no preconceived codes. Characteristic of the method is the
simultaneous generation of data in concert with the analysis phase (Glaser & Strauss 1967). Such
reciprocity is cyclical rather than linear, and is purported to facilitate theory development
throughout the process of data collection. Glaser and Strauss (1967) detailed the methods of
codifying qualitative data for the purpose of generating theory. A feature of Glaser’s approach, as
predicated by Strauss and Corbin (1990), was the constant comparative method which has four
stages: 1) Creation of categories, 2) Integration of categories as the data collection progresses
over time, 3) Reduction of multiple categories to principles which have more general application,
with theoretical saturation stated to occur when no new aspects are emerging, and 4)
Development of emergent major themes and grounded theory. In summary, two analytic
procedures are stated as being basic to the coding process, i.e., the making of comparisons and
the asking of questions. The process of research relevant to my first publication will now be
discussed taking into consideration the following essential elements.
The data for this grounded theory study was collected using focus groups. The benefits of focus
groups are related to the nature of interaction and how informants may be more at ease with
revealing personal information in a cordial environment (Morgan 2019). Focus groups also
provide instant corroboration of the data because of the moderation which occurs through the
different viewpoints expressed by participants. Opportunities to contribute were enhanced by the
facilitator who enabled all informants to communicate their perspectives. Those with well-
considered suppositions or thoughts were able to express their concerns. Being in a focus group
hinders embellishment, as may happen in an individual interview or a survey. In a focus group,
the sharing of one individual acts as a trigger for the recollections of others and supports the
contribution of additional narratives. Such interactions are considered to be both joint actions and
social acts that strengthen interest when participants connect well (Morgan 2019).
Saturation of the data was evident after three focus groups were completed. Field notes were
taken during, and following, the focus groups to complement the data (Montgomery & Bailey 2007).
The notes were also used as a prompt to revisit earlier disclosures and to elucidate more detail
where possible. A brief demographics questionnaire was administered prior to the focus groups to
further inform the context of the participants. It also served as an ice-breaker.
Approval was gained from the Social and Behavioural Research Ethics Committee of Flinders
University, Adelaide, South Australia. Consent was granted in accordance with the requirements
of the National Statement on Ethical Conduct in Human Research (2007). Written approval was
also gained from the Directors of Nursing of regional health units and their individual ethics
committees, as required. Respect for persons dictates that researchers must work to protect
research participants’ autonomy by ensuring full awareness of the factors surrounding the study,
including the potential harms and benefits (Dickert 2009). In this study, the focus group
participants were furnished with a letter of introduction confirming approval from the university and
the ethics committee. Focus group questions and a confidentiality agreement to be signed were
also included. The participants were given a minimum of 14 days to consult with others about their
decision to be part of the study. They were informed that the onus was on each participant to
maintain confidentiality, and that they could decline to answer any question/withdraw at any time.
Consent was also given to record the focus group. Advice with regard to restricted disclosure of
information to preserve the participants’ anonymity, inclusive of pseudonyms to protect their
identity, was also provided (Dickert 2009). The focus groups were conducted in a location away
from their workplace and free of distraction. As a facilitator, I commenced the focus group with an
activity to break the ice in order to introduce ourselves and the participants (Morgan 2019). A list
of ground rules was proffered to encourage the participants to be respectful and effective group
members (Morgan 2019). Over-disclosure may affect a group member emotionally; therefore, as a
facilitator, I adjudicated the contributions of each participant and continuously monitored the stress
levels of the group members (Morgan 2019). The overall goal was to avoid any harm, risk, or
wrong being imposed on the persons being studied (Taylor et al. 2007). An offer of a debrief was
provided on completion of each focus group.
As the principal researcher, the recorded focus group data were transcribed by myself from voice
into text. This had the effect of providing a rich connection to the data and assisting with the
subsequent analysis (Birks & Mills 2011). The transcribed data were printed following each focus
group for the purpose of open coding. Coding was undertaken line-by-line to reduce
preconceptions, and when combined with theoretical sensitivity, allowed the categories, or the
properties of a category found to be explicit in each transcription, to open up (Walker & Myrick
2006). I had to take care that the labelling of a category accurately reflected the impact on the
graduate nurse evident in their contributions (for example, the label “alone and abandoned” more
accurately reflected the emotion expressed in the focus group, whereas “the only RN on the shift”
was more stoic and less compelling). The outcomes were further processed using selective
coding, with QSR NVivo software providing a vehicle for the attachment of transcribed text
representing each code. This facility also made possible constant comparative analysis of each
transcript and further development of categories and generation of concepts. By aligning concepts
to each of the coded texts, the QSR NVivo software enabled the relationships to be visually
displayed around core variables as a concept map. With theoretical coding, these building blocks
enabled the development of meaning and the emergence of grounded theory around the nature of
support given to graduate nurses in rural areas (Birks & Mills 2007). According to Glaser (1978, p.
3), there are four criteria which allow the application of theory to an observable event or
phenomena, i.e., fit, understanding, generality, and control. My publication, “A patient safety
focused registered nurse transition to practice program” emerged from the data (see Chapter 4),
and represents the fit or reality of the graduate nurse during the first year of practice, as explicated
from the diverse data sources (Mellor & Greenhill 2014). It represents the reality of practice and is
understandable by both graduate nurses who were interviewed and others in the field. The theory
developed is generalisable to nursing contexts internationally, with the conditions (control) in which
it applies, clearly elucidated (see list of local and international citations in Appendix 4).
The rigour of qualitative research is routinely assessed in accordance with the trustworthiness
criteria established by Guba and Lincoln (1989). These criteria are credibility, transferability,
dependability, and confirmability. Credibility was achieved by having attentive facilitation with clear
ground rules designed to enhance the contribution of all group members. For example, focus
group members were encouraged to respectfully disagree or challenge perceptions which they
thought were inaccurate. Transferability was achieved through collecting and reporting data in
sufficient depth, thus enabling others to assess the suitability of the recommendations for their
particular context, i.e.,
These 21 participants therefore represented 14 rural hospitals over a radius of 200 km (125,000 km2).
All hospitals offered acute inpatient services, had visiting specialists and provided outpatient and 24-
hour emergency services. All of the hospitals represented were from rural towns with medical services
provided on an “on-call” basis. Acute bed numbers ranged from 11 to 56 beds with a mode of 22 acute
beds per hospital. The remoteness of the hospitals within regions was consistent with few exceptions.
For some hospitals there was only one registered nurse on duty after hours, with staff “on-call” if
needed (Mellor & Greenhill 2014, p. 53).
Dependability required the research methods to be described in detail, as has been provided
earlier in this section. An audit trail was also provided via the QSR NVivo software for cross-
checking of the coding, concepts, and categories essential to the development of this grounded
theory. Confirmability was enhanced through the triangulation of the methods and sources,
provided in the form of demographic quantitative data and triangulation with data sourced from
three regional focus groups of graduate nurses representing 14 health services across rural
South Australia.
Study 2: Publication 2
Study two was a critical review of transition to professional practice programs (Mellor, Gregoric,
Atkinson & Greenhill 2017). The critical review was systematic in both conception and
implementation, with a specific objective, a clear research question, and a description explicating
the methods and procedures. The focus of the review was to demonstrate the usefulness of
applying the Kirkpatrick Model of Evaluation (Kirkpatrick & Kirkpatrick 2006) to evaluate the
effectiveness of transition to professional practice programs. A literature search of transition
programs was undertaken using the PRISMA search strategy with the application of specified
inclusion/exclusion criteria. Each article was rated independently for risk of bias by myself and a
research associate in accordance with the instruments for assessment of risk of bias for
qualitative (Law et al. 1998a) and quantitative (Law et al. 1998b) studies.
In assessing the risk of bias of studies, EPCs (Evidence-Based Practice Centres, AHRQ) should
specify constructs and risks of bias specific to the content area, use at least two independent
reviewers with a defined process for consensus and standards for transparency, and clearly document
and justify all processes and decisions (Viswanathan, Ansari & Berkman, et al. 2012, p.19).
The same procedure (use of an independent reviewer) was followed when using the Kirkpatrick
Model of Evaluation criteria for measurement of transition program effectiveness. Where there was
a difference in our individual ratings, we resolved the final score through consultation and
consensus. The risk of bias and evaluation scores are provided in Chapter 4, Publication 2 (Table
3). The completed drafts were sent to all authors for moderation. Further detail of contributions
and context is provided in Chapter 4, Publication 2 and the authorship statement/publication
background section.
Study 3: Publication 3
Study three was a literature review (Mellor, Gregoric & Gillham 2017), and similar to study two,
had a specific objective, a clear research question, and a clear description explicating the methods
and procedures. The research question, ‘what are the strategies new graduate registered nurses
require to care and advocate for themselves during the first year of practice?’ enabled the search
to be clearly focused. An extensive survey was completed using the PRISMA search strategy
(Moher et al. 2009), the results of which were entered into a detailed Excel spreadsheet for
analysis. The literature revealed that such self-support strategies were present as an eclectic mix
in 80 publications from 2001-2016. The use of multiple data sources represented a model of
dialectical pluralism (Mellor, Gregoric & Gillham 2017, p. 5). The process of recording and
categorising was extensive. For the benefit of the reader, the frequency of occurrence of each
strategy was published to assist with individual interpretation and confirmability of the findings.
Consequently, an overall argument was developed from the findings, together with the strengths
and limitations of the literature review, and recommendations in support of further research. It has
been asserted by some academics that the literature review is a methodology in itself, rather than
merely a phase in a research study, as there is close alignment to scientific research
(Onwuegbuzie, Leech et al. 2011).
Study 4: Publication 4
This grounded theory study was not only influenced by Strauss and Corbin, but also Charmaz, a
previous student of Strauss (Birks, Hoare & Mills 2019). Charmaz (2006) asserted that researchers
construct grounded theories from their own individual viewpoint or perspective, and are influenced by
previous and/or current encounters with people in the field. As fortune would have it, my primary
research methodology has been GTM, which according to Charmaz (2014), allows the researcher to
construct a grounded theory that is more homogeneous and dependable than otherwise would be the
case. While Glaser is adamant that coding should occur without preconceived concepts, Charmaz
(2006) is more circumspect. She agrees that initial coding should be open-ended; however, she also
appreciates that any individual participating in research will have prior conceptualisations and skills. It is
also considered vitally important that the voices of the participants are preserved during synthesis of
the data. This co-construction of theory forms the integral role of the researcher, according to
Charmaz (2014).
As both a researcher and practitioner, I have an extended history of working in the field of graduate
nurse transition (Appendix 1). It was therefore important to acknowledge that my preconceptions could
have had an impact on the filter applied to the raw data (Birks, Hoare & Mills 2019). These
preconceptions could possibly have influenced the labelling of different categories and the resultant
theory generation (Charmaz 2006). Coupled with this observation, I had also completed an extensive
literature review, entitled ‘Strategies new graduate registered nurses require to care and advocate for
themselves?’ In order to partially counter these potential confounding factors, line-by-line coding of the
transcribed interviews was undertaken to limit any preconceptions.
In this second study, respect for persons and their autonomy was again attained by ensuring that
the participants had full awareness of the factors surrounding the study, including the potential
harms and benefits. The participants were provided with an introductory letter proving authority
from the University and the Southern Adelaide Clinical Human Research Ethics Committee (SAC
HREC EC00188) Project No. 337.13., the nature of the study, and a confidentiality agreement to
be signed. Similar to the focus groups, the participants were provided with the opportunity to
consult with others in relation to their decision to be included in the study. They also had a choice
with regard to selecting a non-distracting location away from their usual place of work at a
convenient time. There were nine interviews overall, with two interviews taking place by
telephone and seven face-to-face. As reported in the article by Mellor and Gregoric (2016), the
participants were advised that they had the right to decline to answer any questions or to stop the
interview at any time. They were also made aware of the recording of the interview and how the
recording would be secured. Preservation of anonymity both directly and indirectly was assured
to facilitate full disclosure and to adhere to ethics requirements.
For the convenience of the participants, additional ethics approval was gained to engage an
experienced research assistant. The first interview was a shared interview, with myself as the
primary researcher and the research assistant, to establish the nature and tone of expectations.
The research assistant was keen to hear the stories of the graduate nurses from a personal and
professional perspective, as she was about to commence her own graduate nurse transition
program. As a consequence, our relationship with the interviewees was fully present,
contemporary, and energetic. The unique nature of our engagement valued the participants,
their stories, their wellbeing, and the meanings they attributed to their experiences. Encouraging
graduate nurses to reflect on the strategies they used to succeed was inspirational (Charmaz
2014).
Unlike the focus groups, the recorded data from the interviews were converted into text by a
professional transcription service (with appropriate ethics approval). The transcribed data were
copied into QSR NVivo software following each focus group for the purpose of open coding.
Coding was again undertaken line-by-line and when combined with theoretical sensitivity,
allowed the building blocks of a grounded theory to open up. Labelling of each category had to
be carefully undertaken to ensure accurate reflection of the graduate experience. Focused
coding was then completed using QSR NVivo software (Charmaz 2006). After constant
comparative analysis of each transcript, and further building of categories and generation of
concepts/properties, the initial “ways of being” grounded theory model was constructed.
Once again, the trustworthiness criteria of credibility, transferability, dependability, and
confirmability established by Guba and Lincoln (1989) was considered. Credibility was achieved
by having open-ended questions to encourage sharing of the interviewees’ depth of experience
(see Appendix 2 for a complete list of interview questions). Field notes and reflective memos
were integral to the engagement and processing of the graduate experience (Charmaz 2014).
Here is an example of a reflective memo following the interview with graduate nurse, Heather,
on Thursday 26 March 2015:
Heather (Pseudonym) has a history of being a fighter and is one who has a real instinct
for survival. Her take home message was that she had the capacity to see things
through, however such a negative environment in the final part of her program required
her to seek help. She sought this from friends and family and from the EAP (Employee
Assistance Program) . Essentially she had to provide her own support and needed to
be able to be assertive and say what she wanted.
Transferability was achieved through collecting and reporting data in sufficient depth, thus enabling
others to assess the suitability of the recommendations for their particular context, i.e., nine
participants volunteered to participate from rural and metropolitan South Australia, five graduate
nurses from two metropolitan hospitals, and four graduate nurses representing rural health services.
All participants were in a transition program or had recently completed their program. A spreadsheet
detailing the situated context of each participant is provided in Table 1 (Mellor & Gregoric 2016).
Dependability was evidenced by providing detail of the research method, as has been provided
earlier in this section. An audit trail was once again available via NVivo software for cross-checking
of the coding, concepts, and categories essential to the development of this grounded theory.
Confirmability was provided by triangulation of the demographic quantitative data with data sourced
from two city health services and two rural health services across South Australia. The original aim
was to interview twelve graduate nurses; however, after an extensive recruitment effort, only nine
agreed to participate. Fortunately, they were sufficiently diverse, as per Table 1. After much
reflection, consultation, and searching of the literature, with regard to concerns of sufficient
participant numbers, it was discovered that major themes and variability reliably emerged after the
sixth interview, and that after twelve interviews, saturation was almost assured (Guest, Bunce &
Johnson 2006). As a consequence, it was felt that nine interviews were sufficient for the purpose of
our research, as the data contained many recurring self-support strategies, and major themes were
developing which would later be called ‘ways of being’ (Mellor & Gregoric 2016).
Study 4: Publication 5
This publication emerged as a consequence of revisiting the data gathered for Study 2, and
therefore, the detail that will be provided with regard to process will be minimal. There were a
number of important reasons for revisiting the data. In the first instance, there were many
challenges associated with collection of the data, so the resource was considered extremely
valuable. Secondly, from a respect and integrity perspective, there was an obligation on behalf
of the researchers to further advocate on behalf of graduate nurses who gave their valuable
time to provide this important data. Finally, the richness and quality of the data encouraged a
revisit of the interview transcripts, codes, categories, and concepts. Further scrutiny using GTM
identified significant revelations in relation to graduate nurses and the spectrum of comfort in
clinical practice. Early analysis revealed the concept of “comfort zone” to be considerably more
complex than is ordinarily understood.
The trustworthiness criteria of credibility, transferability, dependability, and confirmability of
Guba and Lincoln (1989) were addressed in the discussion of publication 4 (Mellor & Gregoric
2016); however, I will reiterate with the following observations. Firstly, rich description and the
provision of detailed demographic data of participants provided transferability. Secondly,
dependability was achieved by following the GTM process, as detailed previously for publication
4. In addition, an audit trail was provided via NVivo software for cross-checking of the coding,
concepts, and categories essential to the development of this new grounded theory.
Consequently, with a renewed theoretical sensitivity, an underpinning framework was
constructed resulting in a unique conceptual model illustrative of the spectrum of comfort, as
presented in publication 5.
Chapter Two provides a contextual statement to this thesis. In addition, a contextual model
illustrating my background and professional involvement with regard to the transition of
nursing students to graduate nurse, can be found in Appendix 1. The model is systematised
chronologically and summarises my research engagement, publications, and translation of
evidence into practice.
