Australian Social Work

Vol. 65, No. 4, December 2012, pp. 504
516

Using Secondary Data for Grounded Theory Analysis

Mary Whitesidea*, Jane Millsb, & Janya McCalmanb
a

La Trobe University, Victoria, Australia; bJames Cook University, Queensland, Australia

Abstract
Grounded theory is well utilised in qualitative research for building theoretical
understanding of complex social processes. Grounded theory data analysis strategies
can be used with different types of data, including secondary data. Despite the potential
advantages of secondary data, it is rarely used for grounded theory studies, largely due to
a perception that researchers are unable to follow the guidelines of this research design
with data that they themselves have not collected. This paper demonstrates the use of
secondary data in a study investigating the substantive area of Indigenous empowerment. Potential pitfalls and advantages of using grounded theory methods to analyse
secondary data are discussed, as well as factors that facilitated the analytic process in this
study. These factors included having a large data set across multiple sites and the use of
literature as a source of data. The findings provide a valuable model for combining
research and practise.
Keywords: Social Work Research; Indigenous Research; Empowerment
As part of a broader goal of increasing research output, social work is seeking models
for combining practice and research (Ryan & Sheehan, 2009). Secondary data analysis
is a form of research in which the data collected and processed in one study are
reanalysed in a subsequent study (Rubin & Babbie, 2008, p. 408). Secondary data
are derived from a range of sources including micro level program evaluations. There
are clear benefits associated with using secondary datasets in research, including
efficiencies in time, money, and other resources, and the maximisation of the use of
potentially important data that might otherwise lie dormant (Glaser, 1998; Rubin &
Babbie, 2008). Global changes in the way that research is funded and disseminated
suggest that the use of secondary data will be more widely used in the future.
Government-funded research is being considered a public asset that should be more
accessible and researchers are increasingly required to be prepared to share the
supporting data and deposit data in a repository or archive (Monash University, 2009).
Perhaps most importantly, the use of secondary data reduces research obtrusiveness

*Correspondence to: Dr. Mary Whiteside, Department of Social Work and Social Policy, La Trobe University,
Bundoora, Victoria 3068, Australia. E-mail: Mary.Whiteside@latrobe.edu.au
Accepted 29 October 2011
ISSN 0312-407X (print)/ISSN 1447-0748 (online) # 2012 Australian Association of Social Workers
http://dx.doi.org/10.1080/0312407X.2011.645165

Australian Social Work

505

and decreases the burden placed on respondents (Szabo & Strang, 1997; Rubin &
Babbie, 2008).
Grounded Theory
Grounded theory is one of the most popular designs for qualitative research studies
(Birks & Mills, 2011; Liamputtong, 2009). Methods provide a systematic inductive
process for collecting and analysing data to build middle-range theories, grounded
in participants experiences often with the aim of assisting professional practice or
guiding future research (Liamputtong, 2009; Mills, Bonner & Francis, 2006). Since
Glaser and Strauss (1967) developed the original version of grounded theory methods
under the influence of the postpositivistic tenor of the 1960s, grounded theory has
evolved, in what Mills et al. (2006) have called a spiral of methodological
development (p. 27). Methodological framing of grounded theory research can be
located in a number of differing paradigms of thought, including constructivism,
postmodernism, and symbolic interactionism (Charmaz, 2000; Liamputtong, 2009;
Mills et al., 2006; Strauss & Corbin, 1998). The approach to grounded theory taken in
this paper draws from the constructivist tradition. Constructivism, which views
knowledge and meaning as constructed through human interaction and the
acknowledgement of the influence of social context (Crotty, 1998), was seen as
most relevant for a study that aimed to explore how people exercise agency within
complex historical, political, cultural, and relationship contexts.
Regardless of paradigmatic differences, grounded theorists base their practice on
interconnected features, including theoretical sensitivity, theoretical sampling,
constant comparison of data to data, and developing theoretical constructs
(Liamputtong, 2009). Researchers come to any study with a number of assumptions
but in an inductive research design such as grounded theory, the researcher has no
preconceived hypothesis to prove or disprove. Rather, issues of importance to the
substantive area of enquiry are identified through theoretical sensitivity. Theoretical
sensitivity occurs when the researcher is insightful, open to theoretical leads, and
engaged in a systematic process of concurrent data generation and analysis (Ezzy,
2002; Mills et al., 2006). Theoretical sampling is a process by which new data sources
are identified on the basis of codes and categories developed in earlier rounds of data
generation or collection (Szabo & Strang, 1997). Codes and categories are derived
through a process of open coding in the first instance and subsequently categorised
based on higher-order concepts (Ezzy 2002; Glaser & Strauss, 1967; Mills et al., 2006;
Strauss & Corbin, 1998). As codes and categories are identified, they are continually
compared with each other for similarities and differences between events and
incidents. This constant comparison of data facilitates the ongoing process of
theoretical sampling whereby precise information is sought to shed light on areas that
constitute gaps and holes in the developing theory (Charmaz, 2000, 2006).
Theoretical sampling is important in defining the properties of categories; identifying
the contexts in which they are relevant, specifying the conditions under which they

506

M. Whiteside et al.

arise, are maintained, and vary, and their consequences (Charmaz, 2000, 2006; Glaser
& Strauss, 1967; Mills et al., 2006; Strauss & Corbin, 1998). In a grounded theory
study, data generation and analysis only finishes when the researcher is satisfied that
each category that constitutes the theory is saturated, meaning that new data are not
sparking new insights nor revealing new properties of each of these (Charmaz, 2006;
Ezzy, 2002).
Grounded theory can be used with different types of data, including: numerical
data, interview transcripts, focus group transcripts, literature, and, more recently,
contemporary data sources such as video/DVD recordings, websites, and secondary
datasets. However, despite the benefits of secondary data and its growing use in
research (Rubin & Babbie, 2008), as well as Glaser and Strauss inclusive
interpretation of appropriate sources of data for grounded theory studies that
includes secondary datasets (Glaser, 1998; Glaser & Strauss, 1967), it is interesting to
note that it is rarely used by grounded theorists (Birks & Mills, 2011). A key aim of
this paper is to explore potential reasons for this anomaly and suggest a way forward
for grounded theorists considering the use of secondary data as their primary source.
Using Secondary Data
Concerns have been voiced in relation to the use of secondary data, some of which
relate broadly to its use in qualitative research and some of which relate specifically to
the application of grounded theory methods. Issues of concern include the ethics of
the secondary study, the quality of the data, data fit, and the nature of the
relationship between the researcher and the data. Specific to grounded theory is the
question of whether the use secondary data limits the use of methods associated with
this approach, and therefore, the possibilities for theory construction.
Of primary concern when undertaking secondary analysis are ethical issues. While
Rubin and Babbie (2008) suggested that proposing to use previously collected data
may expedite ethical approval because ethical concerns regarding data collection may
be avoided, the researcher also needs to consider whether the reuse of data breaches
the contract made between the primary researchers and the research subjects (Heaton,
1998). If reuse of the data is contrary to the original contract, it may be necessary to
seek additional consent from research participants for the analysis of data generated in
an additional study, which may or may not be feasible (Heaton, 1998).
The quality of data for secondary analysis may be problematic for a range of
reasons. Data may be outdated or some data or information may be missing (Alston
& Bowles, 2003; Marlow, 2007; Rubin & Babbie, 2008). There may be weaknesses in
the original research design, including the data collection methods used (Alston &
Bowles, 2003; Birks & Mills, 2011; Rubin & Babbie, 2008; Szabo & Strang, 1997). If
the primary research process has not been well documented, the researcher may not
even be able to assess the quality of procedures used to originally collect and analyse
the data (Rubin & Babbie, 2008).

Australian Social Work

507

Data fit may represent difficulties. If the research question posed in the secondary
study is not sufficiently different from the original study, there may little value in
undertaking a secondary analysis of the same dataset. In contrast, the data may not be
of sufficient fit to be compatible with the new research question or area of substantive
enquiry (Heaton, 1998; Szabo & Strang, 1997) resulting in a poorly integrated
grounded theory. This lack of fit may relate to the philosophical perspective and aims
of the original researcher, which will have influenced the nature of data they have
gathered and may not be congruent with the approach used in the secondary study
(Birks & Mills, 2011).
The nature of the relationship between the researcher and the researched
(Heaton, 1998) can present additional challenges to the credibility of a study
involving secondary data analysis. Both closeness and distance of the researcher can
have potential benefits and disadvantages (Hinds, Vogel, & Clarke-Steffen, 1997).
Where the researcher is close to the primary dataset, there are benefits in regard to
knowing the context of the study but also risks in regard to the potential for the
researcher to develop a premature certainty about a phenomenon that may be
present in the data set (Hinds et al., p. 420). Where the researcher is more distant,
the researcher is less likely to have firm or fixed ideas about the phenomena in the
data set (Hinds et al., 1997, p. 420), but may lack sensitivity to the study context and
the necessary inter-subjective relationship between the researcher and the researched
(Heaton, 1998).
Further, grounded theory methods may be difficult to apply in secondary analysis.
For example, theoretical sampling requires the researcher to undertake simultaneous
targeted data collection and analysis. Using secondary data means that theoretical
sampling occurs within the confines of a preexisting dataset (Birks & Mills, 2011;
Szabo & Strang, 1997). For the researcher, questions may be raised in this process
for which answers are difficult to find or leads may be generated but attempts to
follow these may be stymied by the restrictions of the dataset. As a consequence of
both these issues, data saturation of categories may not be possible and gaps may
exist in the final theoretical construction (Birks & Mills, 2011; Szabo & Strang, 1997).
Addressing the Challenges: A Grounded Theory Analysis of a Secondary Dataset
The following study to develop theoretical understanding of empowerment in the
context of Indigenous Australia provides one example of how some of the challenges
associated with using secondary data might be addressed. The data used for the study
was originally collected for the purposes of evaluating the Family Wellbeing program,
an empowerment program designed by and for Indigenous Australians, as it was
administered in four different Australian sites. Both the original program evaluations
and the grounded theory study occurred under the umbrella of the broader James Cook
University Empowerment Research Program.
The following section presents the aims and objectives of the Empowerment
Research Program in which both the primary and secondary studies are located.

508

M. Whiteside et al.

Information is provided about the nature of the study data. Focus is then given to the
secondary study in order to examine the ways in which some of the research design
challenges commonly associated with using secondary data in grounded theory
studies, namely ethics, data quality, data fit, researcher closeness, and theoretical
sampling, were addressed. Also considered is the use of literature to strengthen the
research. The first author conducted the grounded theory study used to illustrate the
arguments posed in this paper and the story of how it evolved will be told in first
person.
The Empowerment Research Program
At the time of undertaking the study, I was working within the James Cook
University Empowerment Research Program. This 10-year research program had
been developed in order to explore the role and contribution of empowerment as a
social determinant of health within the context of Indigenous Australia. Short- and
long-term objectives, developed by the Empowerment Research Program team, were
developed in partnership with Indigenous groups and organisations (see Table 1.). In
essence, the Empowerment Research Program team was what Christie (2006) defined
as transdisciplinary. The team included a range of different professionals, such as
social workers, nurses, and social scientists, and others not professionally trained.
Partnership between Indigenous and non Indigenous researchers was central to how
the team functioned and was reflected in the teams composition, from the level of
chief investigator through to employment of local Indigenous community-based
researchers (Mayo, Tsey, & the Empowerment Research Program Team, 2009).
Rather than the university-based Empowerment Research Program team constructing their own programs to test, a deliberate decision was made to seek to understand
empowerment through programs initiated by Indigenous Australians themselves.
Established Indigenous initiatives and programs, including the Family Wellbeing
program, a four-stage personal and community development program, were identified
as case studies for research (Aboriginal Education Development Branch, 2002).
Micro level evaluations of the Family Wellbeing program took place in sites across
Australia, to meet one of the short-term objectives of the Empowerment Research
Program, namely to determine the utility of this program to engage and support
individuals and groups to take greater control for their health and wellbeing. Data
were collected from 47 of 116 Family Wellbeing program participants in two rural
and two remote Australian sites, between 1999 and 2003. The data took the form of
reflective diaries and transcripts generated from interviews undertaken with
participants, who were able to be located and agreed to be interviewed, six to
12 months after completing the program. Participants ages ranged between 20 and
60 years and the majority were employed within helping or service roles or identified
as parents or members of mens and womens groups. This presented a potential
study limitation because respondents were not necessarily the most disadvantaged in
their communities; however, all had experienced considerable adversity in life and in

Australian Social Work

509

Table 1 Empowerment Research Program Short and Medium to Longer Term Objectives
Short-term objectives
.

Determine the utility of empowerment

interventions as tools to engage and
support individuals and groups in
Indigen ous settings to take greater
control for their health and wellbeing
Develop appropriate in-depth qualitative
tools to monitor and assess the nature of
empowerment
interventions,
their
acceptability and effectiveness in a variety
of settings, as well as the range of
contextual issues that may arise in
implement such interventions
To develop a theoretical model identifying
pathways of empowerment for Indigenous
Australians
Determine the medium-longer sustainability
of empowerment interventions and devise
appropriate
strategies
to
promote
sustainability
Identify the skills, training, and support
needs for local Indigenous communitybased empowerment research facilitators
and best practice models of organising and
providing such support

Medium- to longer-term objectives

.

Develop appropriate measurement tools to

quantify the efficacy of empowerment
interventions
Assess the role and usefulness of
empowerment
interventions,
more
specifically in Indigenous communities
experiencing high levels of alcohol misuse,
family violence, unemployment, and limited
opportunities for gainful employment
Explore potential of empowerment
interventions as tools to enhance chronic
disease management and prevention
Develop
frameworks
to
undertake
appropriate economic evaluation of
complex empowerment interventions
Determine the possibilities and limitations
of empowerment interventions as tools
addressing
health
inequalities
at
population levels
Develop best practice approaches to
collaborative
working
relationships
between academic researchers and relevant
stakeholders that would promote the
concept of empowerment at community,
organisational, and policy levels
Develop a critical mass of interdisciplinary
empowerment and community development
research expertise from grassroots levels
through to doctoral and postdoctoral levels
Make contributions of local, national, and
international significance to the knowledge of evidence based approaches that
engage and support individuals and groups
to take greater control and responsibility for
their health and wellbeing

the evaluation discussed the profound impact of wide-ranging difficulties on their

lives. Most people expressed concern about the difficulties in their workplaces and
communities and the problems they had witnessed in other families. They
appreciated the opportunity to participate in the Family Wellbeing program and to
learn skills to help them better manage their own life situations and to help others. In
addition to the Family Wellbeing program evaluation questions that specifically
related to their impressions of the program, respondents were asked how they had
used the Family Wellbeing program skills and knowledge in their family, workplace,
and wider community life. These evaluation data were very rich. People were open
and honest, revealing very personal information. While the focus of the responses was
on the efficacy and impact of the Family Wellbeing program, the stories people told
resonated with theoretical understandings of empowerment in international

510

M. Whiteside et al.

literature (Whiteside, Tsey, & Earles, 2011). This became an important dataset, not
only due to the richness of the data but also because of the extent to which it
addressed all of the National Health and Medical Research Council (NHMRC) (2003)
criteria for health and medical research of Indigenous Australians of community
engagement, benefit, sustainability and transferability, building capacity, priority, and
significance (NHMRC, 2003; Whiteside et al., 2011).
An additional short-term objective of the Empowerment Research Program was to
develop a theoretical model to identify pathways of empowerment for Indigenous
Australians (see Table 1). The rationale for this objective was that while empowerment
is well recognised as important for understanding the experience of groups who have
been oppressed and disenfranchised and is viewed in international policy literature as
highly relevant for Indigenous cultures experiencing social and health inequalities; it is
also considered to be context and population specific (WHO Commission for Social
Determinants of Health, 2008; Zimmerman, 2000). At the commencement of the
primary study, empowerment as a concept had received inadequate research attention
in the Indigenous Australian environment; thus, a study that aimed to develop
theoretical understanding of empowerment was designed. The existing micro level
evaluation data of the Family Wellbeing program were viewed as a highly relevant
dataset in the secondary study undertaken. While the focus of these data was on the
efficacy of a particular program, peoples stories of how they had used the knowledge
they gained from program participation in their lives was assumed to intrinsically
contain rich information about the nature of empowerment. Further, the use of an
existing dataset would reduce the need for further respondent burden. Reduced
research burden is of particular significance in the case of research with vulnerable
populations such as Australian Indigenous peoples, who experience being one of the
most researched groups historically (Fredericks, 2006).
The secondary grounded theory study that is the subject of this paper utilised the
47 dairies and interviews as a secondary dataset. These data contained multiple
stories associated with peoples ability to understand or bring about changes in
situations that concerned or distressed them, both those of their own and of others.
Examples of the stories people told in their diaries and interviews provided some
insight into the nature of the stories for analysis. Ruby (mother of eight children and
grandmother of five) was employed at her local community-controlled health centre.
She was attracted to the Family Wellbeing program because she held a personal wish
to develop a similar program. She found that Family Wellbeing built on and
broadened her existing skills and knowledge and helped her to develop a more
positive attitude to lifes challenges. Ruby was quick to apply her new understandings
to family and community issues. She learnt not to lash out during quarrels but to
step back and take a more conciliatory approach. Her family and community
relationships improved as a result. With greater skills and more confidence, Ruby
took on a leadership role in her community. She put careful thought into how to
encourage others in her family and community to participate in the Family Wellbeing
program. She trained to be a program facilitator and ran the program in her

Australian Social Work

511

community. She became a key member of a housing action group that emanated
from her Family Wellbeing program training group and sought to address the acute
shortage of housing in her community. Ruby did not find these new roles easy at first.
She had to overcome shyness and nervousness and force herself to be out there. But
once there, she found she liked it and felt good about herself. Increasingly, she saw
herself as someone with the capacity to help and lead others (Whiteside, 2009).
The data were analysed using a constructivist grounded theory approach, chosen
because of its focus on exploring multiple realities, placing peoples experiences at the
centre of the study, and its clear analytic guidelines (Charmaz, 2006; Mills et al.,
2006). As I am not Indigenous, systematic processes were put in place to minimise
the impact of culturally-based misinterpretation. An Indigenous chief investigator of
the Empowerment Research Program undertook to play the role of Indigenous
cultural mentor, as she had in the primary study.
Ethics
Ethical issues associated with both the primary and secondary study were a priority
consideration and planned for in the initial conceptualisation of the Empowerment
Research Program. It was anticipated at the outset that the Family Wellbeing program
evaluation data may provide opportunities for subsequent synthesis of findings
across sites to build theoretical knowledge. Rather than risk breaching the original
contract made between the primary researchers and the research subjects (Heaton,
1998), participants were informed, and consented to, the use of the data for the
purposes of program evaluation and also to build broader understanding of
empowerment. Ethical approval for this approach was obtained from four ethics
committees: the university ethics review committee as well as health ethics
committees in each of the program sites.
Data Quality
My embeddedness in the original research team made access to the data easier. I was
confident in the nature and quality of the data because of my contribution to the
original research design, including the philosophical premises of the study and some
data collection. Although I had not been involved in collecting all of the Family
Wellbeing program evaluation data, methodological and data-collection procedures
were well documented (Heaton, 1998) and provided a clear audit trail to follow.
Where there were gaps in information (e.g., gaps in demographic data or the location
of information) I was able to consult the relevant team member to locate the
necessary information (Szabo & Strang, 1997).
Data Fit
Even though many of the barriers that other grounded theorists might face had been
circumvented because of my history with the original research study, there were a
number of challenges that remained. In particular, these related to the fit of the data to
the secondary research question and the extent to which the grounded theory methods
could be applied to an existing dataset. Ostensibly, the data were a good fit (Heaton,
1998; Szabo & Strang, 1997), as the aim of the secondary study was to examine the

512

M. Whiteside et al.

concept of empowerment, which formed the substantive area of enquiry of the

primary study. The data provided rich descriptions of transformative life changes,
which had been analysed in the original study as indicative that empowerment had
occurred (Tsey & Every, 2000; Tsey, Deemal, Whiteside, & Gibson, 2003; Whiteside,
Tsey, McCalman, Cadet-James, & Wilson, 2006). These life changes involved people
exercising greater choice and control of their life, strengthening personal growth,
relationships, and community change. One woman addressed her drinking: I used to
drink a lot [but now] I dont drink that much. . . now that Ive stopped drinking, Ive
got things to buy for children (Rose1, Site 2). One man spoke of spending more time
with his family doing things that we never been doing . . . taking my family out . . .
going down to the beach*going fishing. A healing centre was established in one
community. In another community, a housing action group initiated a mediated
settlement with traditional owners for land for housing development, and negotiated
the release of funding for housing construction (Whiteside et al., 2011).
However, the focus of the two studies was different. The original study sought to
evaluate a specific program within different contexts, while the second study involved
building theory. Caution had to be exercised in the secondary analysis to avoid
replicating the program evaluation and, instead, develop a grounded theory of
empowerment. I found this to be a constant challenge when attempting to remain open
to theoretical leads in the data. In order to achieve this goal of breaking new theoretical
ground, I needed to ensure, vigilantly, that the new research question, what do the data
tell me about empowerment for this group of people (as opposed to what do the data
tell me about the Family Wellbeing program) remained at the forefront of analysis.
Closeness to the Data
Hinds et al. (1997) identified risks associated with being both too close and too
distant to the primary data. In this study, the hazard for the most part involved being
potentially too close. This was particularly the case at three of the four study sites
where I was, at times, a coparticipant of Family Wellbeing program, a program
trainer, a program evaluator, and a participatory action researcher supporting related
community projects. I risked bringing my own assumptions, being uncritical and too
certain about phenomena in the data (Hinds et al., 1997). I needed to consciously
place my own assumptions aside and attempt to take a fresh view of the data. This is a
challenge faced in many research studies. Most research involves having a prior
knowledge of at least the literature associated with the field, if not the data itself, and
the risk of importing preconceived ideas and imposing these on your work (Charmaz,
2006; Mills et al., 2006).
My relationship with the study participants and the data was also a source of strength.
Although I needed to strike a balance between empathising with, and scrutinising, the
data (Hodkinson, 2005), I brought to the research a subtle and nuanced depth of
knowledge of the Family Wellbeing program and many of the participants, which
facilitated my understanding of the experiences of the study respondents. I knew many
1

Pseudonyms are used throughout to protect privacy.

Australian Social Work

513

of the respondents as individuals and the trusting and cooperative relationships I had
with them helped to reduce the constraints of my ethnicity and social position, adding
depth and quality to the research (Hodkinson, 2005).
Theoretical Sampling
There remained the challenge of being unable to theoretically sample and follow
theoretical leads by undertaking additional interviews with study respondents. I was
able to overcome this constraint in the following ways. Firstly, the secondary dataset
was of sufficient size and depth to enable a high degree of theoretical sampling and the
development of theory with sufficient density of categories and properties to construct
grounded theory (Glaser & Strauss, 1967; Goulding, 2002; Szabo & Strang, 1997).
Initial sampling was purposive and occurred on the basis of accessing the richest data
for open coding. I commenced with the site where the Family Wellbeing program was
most comprehensively implemented and evaluated, assuming this was where the
richest data lay. I coded and categorised this data. After coding and writing summaries
of the first 10 interviews in this first site, I stopped to reflect on the story that the codes
and initial categories were telling me, gaining a sense of the theory that was emerging.
Further opportunities for theoretical sampling were provided by my access to data
from multiple sites, which varied in relation to location, size, and gender make-up. I
compared the categories and their properties in different contexts and under different
conditions and found that the emerging theory held true. For example, particular
personal beliefs and attitudes were a vital element of empowerment across sites and
gender and included self-pride. One man described himself as one of those fellas
thats proud, Im just happy, Im me. Pride was evident in this reflection from a
woman in a different site: [I am] very strong about who I am . . . Im not a half
caste but a person of Aboriginal descent.
I theoretically sampled interview data until data saturation was reached in each of
the categories that constituted my grounded theory, while dimensionalising them in
relation to context.
The Use of Literature
During the process of concurrent data collection and analysis, the literature was used
as an additional source of comparative data. Because of the inductive nature of
grounded theory, the substantive literature review for my study occurred after open
and intermediate coding of the secondary dataset had occurred. At this stage, I
undertook an extensive review of the literature on empowerment in both the
Australian and international context.
There was a rich and growing body of literature examining the elements of
empowerment, its wide-reaching currency across disciplines, and the potential for
empowerment approaches to be transformative, despite a range of possible pitfalls
(Whiteside et al., 2011). However, despite empowerment being considered a relevant
concept for Indigenous cultures who were experiencing social and health inequalities,
little was written specifically about empowerment within the Indigenous Australian
context. To some extent, this related to an issue of language. Terms such as social and

514

M. Whiteside et al.

emotional wellbeing, resiliency, healing, and community development appeared in

the literature (Calma, 2007; Campbell, Pyett, McCarthy, Whiteside, & Tsey, 2005;
Feeney, 2008; Kildae & Yow Yeh, 2000), although none of these terms had an
extensive literature base in their own right. A close examination of the usage of each
of these terms revealed commonality with the concept of empowerment literature.
Where differences were evident, these were, for the most part, indicative of the
historical, social, cultural, and political context. Therefore, to strengthen the review I
included these bodies of literature.
I was also able to draw on the findings of further Family Wellbeing program micro
level evaluations that had occurred since the time the original data used in this study
were collected, both in Australia and in a pilot study in Papua New Guinea (McCalman,
Tsey, Kitau, & McGinty, in press; McEwan, Tsey, and the Empowerment Research
Team, 2009; McEwan, Tsey, McCalman, & Travers, 2010; Kitau, Tsey, McCalman, &
Whiteside, 2011). The literature helped me to compare and clarify ideas for theory
development and where themes were congruent they served to authenticate the
emergent theory. In addition, I was able to determine where the theory extended the
literature, thereby clarifying its contribution to the body of knowledge about Australian
Indigenous peoples empowerment (Charmaz, 2006). For example, I was able to
demonstrate that the emergent theory resonated closely with international literature
addressing empowerment. However, it was evident that understandings of empowerment needed to be modified in the context of Indigenous Australia.
Study Outcome
This secondary study resulted in the development of a theoretical model that
delineated interconnected and mutually reinforcing elements of empowerment for
Indigenous Australians. The elements involve particular beliefs and attitudes, skills,
and knowledge that enable people to exercise agency and bring about the changes or
achievements most important for their situation. The resultant grounded theory
model provides a useful framework for social workers and other professional groups
seeking to effectively work alongside Indigenous groups, communities, and
organisations (Whiteside et al., 2011).
Conclusion
In this grounded theory study, existing program evaluation data were successfully
used to enhance theoretical knowledge of an important and potentially transformative construct for change. While real challenges were present in the use of a research
design combining secondary data and grounded theory methods, it was possible to
address and minimise these potential limitations while simultaneously reducing the
burden on respondents. This work has important implications for social workers
looking for models to combine research and practice. Social workers frequently have
access to secondary data, including program evaluation data collected as part of best
practice, which can provide a valuable data source for further research studies.

Australian Social Work

515

Acknowledgements
The authors would like to thank the participants of Family Wellbeing who gave
so generously of themselves through program evaluations. The research was
undertaken through an NHMRC training scholarship for Indigenous Australian
health research.
References
Aboriginal Education Development Branch. (2002). Family Wellbeing curriculum document, stages
1
5: Aboriginal Education Development Branch, Adelaide. South Australia: Department of
Education, Training and Development.
Alston, M., & Bowles, W. (2003). Research for social workers: An introduction to methods (2nd ed.).
St. Leonards: Allen & Unwin.
Birks, M., & Mills, J. (2011). Grounded theory: A practical guide. Los Angeles: Sage.
Calma, T. (2007). Social and emotional health and wellbeing. Paper presented at the 9th Annual
Garma Festival of Traditional Culture, Gulkula, North East Arnhem Land, NT.
Campbell, D., Pyett, P., McCarthy, L., Whiteside, M., & Tsey, K. (2005). Community development
and empowerment: A review of interventions to improve Aboriginal health. In I. Anderson,
F. Baum, & M. Bentley (Eds.), Beyond bandaids: Exploring the underlying social determinants
of aboriginal health. Darwin: Cooperative Research Centre for Aboriginal Health.
Charmaz, K. (2000). Grounded theory: Objectivist and constructivist qualitative methods. In N.
Denzin & Y. Lincoln (Eds.), The handbook of qualitative research (2nd ed., pp. 509
535).
Thousand Oaks, CA: Sage Publications.
Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analysis.
London: Sage Publications.
Christie, M. (2006). Transdisciplinary research and Aboriginal knowledge. The Australian Journal of
Indigenous Education, 35, 78
89.
Crotty, M. (1998). The foundations of social research: Meaning and perspective in the research process.
St Leonards, NSW: Allen & Unwin.
Ezzy, D. (2002). Qualitative analysis: practice and innovation. Crows Nest, NSW: Allen & Unwin.
Feeney, M. (2008). Reclaiming the spirit of wellbeing: Promising healing practices for Aboriginal and
Torres Strait Islander people. Australia: The Stolen Generation Alliance.
Fredericks, B. (2006, October 27). Aboriginal and Torres Strait Islander lead and governed research: A
sign of social change. Paper presented at the Social Change in the 21st Century Conference
2006, Centre for Social Change Research, Queensland University of Technology, Carseldine,
Brisbane, Queensland.
Glaser, B. G. (1998). Doing grounded theory: Issues and discussions. Mill Valley, CA: Sociology Press.
Glaser, B. G., & Strauss, A. (1967). The discovery of grounded theory. Chicago: Aldine.
Goulding, C. (2002). Grounded theory: A practical guide for management, business and market
research. London, CA, New Delhi: Sage Publications.
Heaton, J. (1998). Secondary analysis of qualitative data. Social Research Update, 22. Retrieved 18
April, 2011, from http://sru.soc.surrey.ac.uk/SRU22.html
Hinds, P., Vogel, R., & Clarke-Steffen, L. (1997). The possibilities and pitfalls of doing a secondary
analysis of a qualitative data set. Qualitative Health Research, 7, 408
424.
Hodkinson, P. (2005). Insider research in the study of youth cultures. Journal of Youth Studies, 8,
131
149.
Kildae, T., & Yow Yeh, L. (2000). Empowering the people: When education is more than just words.
Fine Print, 6
10.

516

M. Whiteside et al.

Kitau, R., Tsey, K., McCalman, J., & Whiteside, M. (2011). The Adaptability and sustainability of an
Indigenous Australian Family Wellbeing Initiative in the context of Papua New Guinea: A
follow up. Australasian Psychiatry, 19, S80
S83.
Liamputtong, P. (2009). Qualitative research methods (3rd ed.). South Melbourne: Oxford
University Press.
Marlow, C. R. (2007). Research methods for generalist social work (5th ed.). Belmont, CA: Brooks/Cole.
Mayo, K., & Tsey, K. & the Empowerment Research Program Team. (2009). The research dance:
University and community research collaborations at Yarrabah, North Queensland. Health
and Social Care in the Community, 17, 133
140.
McCalman, J., Tsey, K., Kitau, R., & McGinty, S. (in press) Bringing us back to our origin: Adapting
and transferring an Indigenous Australian values-based leadership capacity building course
for community development in Papua New Guinea. Community Development: Journal of the
Community Development Society.
McEwan, A., Tsey, K., McCalman, J., & Travers, H. (2010). Empowerment and change management
in Aboriginal organizations: A case study. Australian Health Review, 34, 360
367.
McEwan, A., Tsey, K. & the Empowerment Research Team. (2009). The role of spirituality in social
and emotional wellbeing interventions: The family well-being program at Yarrabah. Casuarina:
NT: Cooperative Research Centre for Aboriginal Health.
Mills, J., Bonner, A., & Francis, K. (2006). The development of constructivist grounded theory.
International Journal of Qualitative Methods, 5, 25
35.
Monash University (2009). Research data management guidelines for researchers. Retrieved 9 May,
2011, from http://www.researchdata.monash.edu.au/guidelines/deposit.html
National Health & Medical Research Council (NHMRC). (2003). Values and ethics: Guidelines for
ethical conduct in Aboriginal and Torres Strait Islander health research. Canberra, ACT: Author.
Ryan, M., & Sheehan, R. (2009). Research articles in Australian Social Work from 1998
2007: A
content analysis. Australian Social Work, 62, 525
542.
Rubin, A., & Babbie, E. (2008). Research methods for social work (7th ed.). Belmont, CA: Brooks/Cole.
Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Grounded theory procedures and
techniques (2nd ed.). Thousand Oaks, CA: Sage.
Szabo, V., & Strang, V. (1997). Secondary analysis of qualitative data. Advances in Nursing Science,
20, 66
74.
Tsey, K., & Every, A. (2000). Evaluating aboriginal empowerment programs: The case of family well
being. Australian and New Zealand Journal of Public Health, 24, 509
514.
Tsey, K., Deemal, A., Whiteside, M., & Gibson, T. (2003). Social determinants of health, the control
factor, and the family wellbeing empowerment program. Australasian Journal of Psychiatry,
11, S34
39.
Tsey, K., Whiteside, M., Haswell-Elkins, M., Bainbridge, R., Cadet-James, Y., & Wilson, A. (2010).
Empowerment and Indigenous Australian health: A synthesis of findings from Family
Wellbeing formative research. Health and Social Care in the Community, 18, 169
179.
Whiteside, M. (2009). A grounded theory of empowerment in the context of Indigenous Australia.
Unpublished PhD thesis, James Cook University, Cairns. Queensland.
Whiteside, M., Tsey, K., & Earles, W. (2011). Locating empowerment in the context of Indigenous
Australia. Australian Social Work, 64, 113
129.
Whiteside, M., Tsey, K., McCalman, J., Cadet-James, Y., & Wilson, A. (2006). Empowerment as a
framework for Indigenous workforce development and organisational change. Australian
Social Work, 59, 422
434.
World Health Organization (WHO) Commission for Social Determinants of Health. (2008).
Closing the gap in a generation: Health equity through action on the social determinants of
health. Geneva: World Health Organization.
Zimmerman, M. (2000). Empowerment theory. In J. Rappaport & E. Seidman (Eds.), Handbook of
community psychology (pp. 43
63). New York: Kluwer Academic/Plenum Publishers.

Copyright of Australian Social Work is the property of Routledge and its content may not be copied or emailed
to multiple sites or posted to a listserv without the copyright holder's express written permission. However,
users may print, download, or email articles for individual use.