Doing Formal Grounded Theory: A review

Tom Andrews PhD

This is the latest in a family of Grounded Theory books by
Glaser that continue to build on previous work and make the
methodology much more explicit. Its purpose is quite simply to
provide Grounded Theory researchers with a set of procedures
that can be followed to generate a Formal Grounded Theory
(FGT). Despite several chapters in previous books that deal
with generating formal grounded theory it has been given scant
attention by researchers and this book aims to reverse this. It
brings together and synthesises these previous writings in one
book and seeks to specify much more clearly what is meant by a
formal grounded theory. As with other more recent books by
Glaser, this one is based on data in that the procedures
outlined are come from previously generated formal grounded
theories. However, Glaser cautions that this is based on
limited data since not many FGTs exist yet and as more are
generated, the method will become more explicit. The book has
been eagerly anticipated by grounded theorists and it does not
disappoint.

From the beginning, Glaser emphasises that such theory is
not “grand theory” about a theoretical code but a conceptual
extension of a substantive grounded theory core category using
GT generating procedures. There is a natural tendency to see
the applicability of core categories everywhere, beyond the data
that generated them. There is a very useful and thought
provoking differentiation between descriptive and conceptual
generalisation that anyone interested in trying to understand
the difference between qualitative methodology and GT would
benefit from reading. The discussion of the struggle of
qualitative research in dealing with issues of generalisability
and transferability is based on extensive reading of the
qualitative methodological literature. This struggle is
essentially about the near impossibility of making
generalisations based on descriptive, unit based findings.
There is clear differentiation made between the conceptual
nature of GT and routine qualitative data analysis (QDA).
However despite this when it comes to generalisation, there is a
tendency in QDA writings to reduce GT to another descriptive
methodology with near total miss of its conceptual nature. This
is another example of default remodelling which serves to block
GT at every turn and is dealt with extensively by Glaser in
other writings (see Glaser, 2003). Ultimately the discussion
furthers the argument that the end product of a GT study is
very different compared to that of a qualitative studyconceptualisation
as opposed to description.

In discussing the general implications of the core category,
this book will be invaluable to PhD students since most theses
are expected to discuss this issue. It will guide and encourage
them to think and write about the conceptual generalisation of
a core category and would have been invaluable to me when
asked about the general implications of my core category at my
PhD Viva Voce. It encourages researchers to think carefully
about the issue but not to engage in speculation. Also, students
are often expected to make an appeal to or suggest future
research. This book will enable them to do this in ways that
are consistent with GT. For example, they could suggest how
their core category could be developed further from a SGT to a
FGT.

Glaser emphasises that there are many substantive
grounded theories just waiting to be extended to a FGT and the
encouragement for experienced Grounded Theorists is that a
little data goes a long way in generating one. The procedures
used are the same as for generating a SGT but theoretical
sampling is different since the core category has already been
generated, but this is the only difference. The core category
does not change. While this may be stating the obvious,
nonetheless one of the pitfalls that Glaser warns about is
loosing sight of the core category by dropping into description
rather than conceptual comparisons and reverting to
generating a more complete SGT. This is essentially about
using the literature from any field as more data, comparing it
conceptually to the core category. It is the key to generating a
FGT and is quite simple yet challenging: loose description and
conceptualise. Theoretical sampling according to the core
category will guide the literature review. However data can be
used from whatever source. However it is is seldom necessary
to go into the field to collect further data for comparison, since
there are many data sources to turn to for this. This means
that FGT research costs very little money to do, can be done as
part of everyday academic reading and without the time
constraints associated with generating a SGT for a thesis, very
encouraging for and appealing to GT researchers.

Generating a FGT cannot be based on revisiting a SGT in
order to make it more comprehensive. Nor can it be done by
rewriting a SGT by leaving out substantive words. This is
raising the conceptual level mechanically. A FGT must be
based on data whatever their source. As in all GT, there are no
shortcuts: the method must be followed and a FGT generated
by using research to broaden the scope of the theory by
conceptual comparative analysis of different substantive areas.
The book ends with a very useful and convincing chapter on the
uses of FGT which could act as convincing evidence when
applying for research grants.

To summarise, this book is a very welcome addition to
Grounded Theory methodology by clearly outlining how to
generate a Formal Grounded Theory. It is both challenging
and thought provoking. The challenge is to generate FGT
using any data source by conceptual comparison. Again we are
challenged to think conceptually in terms of the literature
rather than simply doing a traditional literature comparison.
Nonetheless it should encourage experienced GT researchers to
now have the confidence to generate FGT. However while this
is the explicit aim of the book, there is something here for
everyone given that GT is an advanced methodology. For
example, PhD students will find the chapters dealing with
“Conceptual Generalisation” and “Generalizing: the Descriptive
Struggle” particularly useful. We now have a beginning text on
how to generate a FGT. The invitation to all of us who have
developed a SGT now is to do so.

Author

Tom Andrews, Ph.D.
School of Nursing and Midwifery
University College Cork, IRL
Email: t.andrews@ucc.ie