Aliceijelp2015 PDF
Aliceijelp2015 PDF
Aliceijelp2015 PDF
net/publication/309574900
CITATION READS
1 25,985
3 authors:
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Alice Shanthi on 01 November 2016.
1,2,3
Alice Shanthi, Lee Kean Wah, Denis Lajium (Dr)
1,2,3
Faculty of Psychology and Education,
Universiti Malaysia Sabah, Sabah, Malaysia
alice_shanthi@yahoo.com.my
ABSTRACT
INTRODUCTION
For novice researchers the choices available to analyse qualitative data is aplenty,
and it could be quite perplexing and overwhelming. Creswell (2013), mentioned the
narrative research, phenomenology, grounded theory, ethnography and case study as five
approaches for a qualitative study. While Merriam (2009), added an extra of two more
approaches to Creswell’s list; basic qualitative and critical qualitative research. Denzin
and Lincoln (2005, cited by Merriam, 2009: 20) identify six research strategies namely
case study, ethnography, grounded theory, life and narrative approaches, participatory
research and clinical research. Wertz et al., (2011) explain five approaches of conducting
qualitative research in social science, namely; phenomenology, grounded theory,
discourse analysis, narrative research and intuitive inquiry.
While the approaches to qualitative research by renowned writers suggest
there is no clear conformity as to how to categorize “the baffling numbers of choices or
approaches” to qualitative analysis (Creswell, 2007:6), therefore, Denzin and Lincoln’s
(2013:13) statement about qualitative analysis will be taken as the core definition for
qualitative approach in this study; qualitative research is a set of complex interpretive
practices where no specific method or practice can be privileged over another.
Discourse Analysis as a Qualitative Approach to Study Information Sharing Practise in Malaysian Board Forums
This article starts of with a brief discussion of research paradigms selected for
this study which would help to unify the type and the specific methods of collecting
and analyzing data. Next it will move on to explain how discourse analysis can used to
conduct qualitative research in Computer Mediated Communication (CMC) specifically
in discussion board forums by analysing the naturally occurring asynchronous
communication to reveal how language functions to perform the many actions needed to
share and exchange information online.
REASEARCH PARADIGMS
The starting point for any qualitative research would be to decide on the philosophical
ideas which is described using different terms such as worldview (Creswell, 2013:6),
paradigm (Denzin & Lincoln, 2013:17), or epistemology perspectives (Merriam,
2009:11), but they refer to the same ideas that lace the background of a study. Deciding
on the worldview or paradigm is important because it will help a new researcher to
choose the correct research design and method of analysis that would best help to answer
the research question of the study. In short, the paradigms guides the researcher how to
examine a concrete empirical problem (Denzin & Lincoln, 2013: 29).
In this study discourse analysis is used to analyse thread-level asynchronous
communication to shed some light on the language strategies used by members of online
discussion board forums as they interact online to share and exchange information,
hence, it would adopt the Interpretivism philosophy also known as constructivism.
This paradigm emphasises conductions study in their natural settings by attempting to
make sense of, or interpret the meanings people bring to them by searching for pattern
embedded in the data source itself (Creswell, 2007). The data source could be in the
form of interviews, observation or review of documents. Constructivists do not generally
begin with a theory rather they generate or inductively develop a theory or pattern of
meanings (Creswell, 2003: 9 cited by Starks & Trinidad, 2007).
However, the choice of qualitative analysis that a researcher uses to arrive
at the end meaning that people bring to the data, whether by using grounded theory,
phenomenological, narrative, discourse analysis, etc., makes the difference in a
qualitative study. With that, Starks and Trinidad (2007) summarised that the objective
of a study that uses phenomenology approach is to study how people make meaning of
their lived experience; discourse analysis scrutinizes how language is used to accomplish
interpersonal and social interaction; and grounded theory develops explanatory theories
of basic social processes studied in context.
160 International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015)
Alice Shanthi, Lee Kean Wah & Denis Lajium
that evolves around the lives of people. It involves “studying things in their natural
setting, attempts to make sense of, or interpret, phenomena in terms of the meanings
people bring to them” (Denzin & Lincoln, 2013:7). It is also about collecting data using
several common methods associated with qualitative research such as participant/s
observation/s, field notes, recordings, interviews, photographs and memos. In qualitative
method the data collected to examine a social phenomenon could be in the form of
spoken transcripts, written text and/or other forms of documents that are analysed in an
attempt to understand human behaviour and experience in a social setting. Thus, in a
qualitative study, “researchers are keen on gaining understanding of people’s behaviour
and/or experience in a rich and complex setting that specific for the particular group of
people or setting that is being studied, and not in obtaining information which can be
generalized to other larger groups” (Saunders, Lewis & Thornhill. 2009:127).
Wertz, et. al., (2011) distinguishes five ways of doing qualitative analysis in
social science, namely; phenomenology, grounded theory, discourse analysis, narrative
research and intuitive inquiry. Of these five methods they go on to explain that discourse
analysis belongs to “a family of contemporary approaches that emphasises human
language as a socially contextual performance” (Wertz, et. al, 2011: 4). Since discourse
analysis is categorised as one of the contemporary approaches in the field of qualitative
research, it shares some of the same analytical methods with other more established
qualitative methods such as grounded theory. These include steps such as coding, sorting
of categories, identifying themes, and relationships and drawing conclusions to answer
the research questions. In fact literature reveal that it has been noted that most of the
qualitative methods share some form of commonalities in their analytical approaches
(Starks & Trinidad, 2007; Saunders, Lewis, & Thornhill, 2009; Wertz, et al., 2011).
However what differentiates them is their choice of philosophy or paradigm, goals and
the final product of their investigation. As mentioned by Stark and Trinidad (2007) greater
differences between these qualitative methods are observed at the beginning and the final
results of the study as seen in Table 1 which shows a comparison between Discourse
Analysis (DA) and Grounded Theory (GT) methods which share a lot of commonalities.
Table 1 Similarities and Differences of the Two Interpretive Approaches With Respect To History, Goal,
Philosophy, Methodology, Analytic Method and Product.
International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015) 161
Discourse Analysis as a Qualitative Approach to Study Information Sharing Practise in Malaysian Board Forums
Methodology
• Formulating a research question “What discourses are used and “How does the basic social
how do they shape identities, process of [ X ] happen in
activities and relationships?” the context of [ Y ]
• Sampling Those situated in one or more of Those who have
the discourses of interest. experienced the
phenomenon under
different condition.
• Data Collection: -Observation Observe participants in Observe participants where
conversation in their natural the basic social process
environment. takes place.
-Interview Both engage in dialogue; Participants describe
interviewer probes for intertextual experience; interviewer
meaning probes for detail, clarity
Analytic Methods Examine how understanding is Open, axial, and selective
produced through a close look at coding: examine concepts
the words. Interested in how the across their properties and
story was told, what identities, dimensions; develop an
activities, relationships, and shared explanatory framework that
meaning are created through integrates the concepts into
language. a core category
Audience Policy makers & interventionist Researchers & practitioners
who need to understand the who seek explanatory
discourses in use to craft affective models upon which to
messages. design interventions.
Product Description of language-in-use; Generate theory from the
identify how different discourses range of the participants’
shape how identities , relationship, experience.
and social goods are negotiated
and produced.
162 International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015)
Alice Shanthi, Lee Kean Wah & Denis Lajium
International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015) 163
Discourse Analysis as a Qualitative Approach to Study Information Sharing Practise in Malaysian Board Forums
This study specifically sets forth to examine how language is used in online communication,
what they achieve and how these language functions are used to exchange and share
information in online discussion board forums. By using DA as a research approach, the
researcher hopes to up with new knowledge on how information is shared and exchanged
in a mediated environment. This study will explore discourse as a social practice that is
taking place in online discussion board forums in the Malaysian context. In comparison
to the broader interdisciplinary study of Computer Mediated Communication (CMC),
computer-mediated discourse (CMD) is the finer study of discourse found in CMC.
As there are many different approaches to the study of discourse; hence,
as with all research projects, choosing the right method of analysis that best guides
the researcher to answer the inquiry of the project is of utmost importance. Rourke,
Anderson, and Garrison (2007), have recommended that researchers investigating
online discussions should utilize frameworks that build on prior research. As such,
in this study the researcher builds upon Herring’s (2004) work, which used discourse
analysis approach to explore communication taking place in mediated environment
using Computer-Mediated Discourse Analysis (CMDA) Framework. CMDA focuses
specifically on language and language-use in computer networked environment and
makes use of methods of discourse analysis (Herring, 2004).
CMDA comprises “any analysis of online behaviour that is grounded in
empirical textual observation” (Herring, 2004). In structuring the scope of CMDA,
Herring illustrates that CMDA has already been used for the investigation of numerous
linguistic phenomena on a micro-level (e.g. online word-formation, lexical choice,
sentence structure, and code-switching) as well as on a macro-level (e.g. coherence,
community, gender equity, and identity). CMDA assumes that discourse exhibits
recurring patterns which are produced consciously or unconsciously and the basic goal
of discourse analysis is to identify patterns in discourse that are inherently to the present
study. In her CMDA toolkit, Herring presents step by step guide to study online social
behaviour captured through the lens of language-use as depicted in Table 2, and “its
interpretations are grounded in observations about language and language use” (Herring,
2004) in online communication. CMDA adapts the “traditional typology of discourse
to the online environment including modality, participation, text type and discourse
type, as such has developed into a faceted classification scheme for computer-mediated
communication” (Fitzpatrick & Donnelly, 2010).
As is seen in Table 2, CMDA can be used to examine data at five levels of
analysis such as the structural level, meaning level, interaction level, social behaviour
level, and participation level using different methods of analysis. At structural level of
analysis is objective as it entails counting of CMC phenomena such as lexical items,
phrases, and syntactic patterns. The meaning level of analysis includes the semantic or
the functional level of speech acts analysis which involves interpretive and subjective
terms. The interactional level includes topic development and negotiating of interactive
exchanges. The participation level calls for a descriptive analysis that includes a count
164 International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015)
Alice Shanthi, Lee Kean Wah & Denis Lajium
of the number of postings and their responses, number of participants, and thread length.
In short, CMDA, as put forward by Herring (2004) is a systematic observation of online
language behaviour and is grounded in observation about language and language use
in CMD. A researcher studying CMD does not necessarily use or choose all levels of
CMDA analysis as depicted in Table 2, but to choose only levels that are necessary to
answer the inquiry of the study. For this study the researcher will start off with analysis
at participation level and proceed to meaning level which would require analysis for
different aspects of language-use that would show a pattern of language used for CMD
found in online discussion board forums in Malaysia.
DATA COLLECTION
To study the discourse in a mediated environment, ideally one obvious way to collect
data would be to download all accessible online interaction; however this would produce
a massive size of data set that might prove to be unmanageable for the researchers to
handle. In addition, the sampling for qualitative data collection in CMDA is rarely
done randomly, since random sampling sacrifices context, and context is important
in interpreting discourse analysis results (Herring, 2004). In discourse analysis, the
researcher is the primary instrument for data collection and analysis (Merriam, 2009:15),
and therefore has to decide on the number or size of data to be gathered as corpus for the
study. This is because in Mason (2010) voice a number of issues can affect sample size in
qualitative research; however, the guiding principle should be the concept of saturation.
International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015) 165
Discourse Analysis as a Qualitative Approach to Study Information Sharing Practise in Malaysian Board Forums
This concept (data saturation) is quite different from the data saturation which
originated from grounded theory and termed as theoretical saturation. The difference
between these two concepts is explained by Green and Thorogood, 2004 (cited by
O’Reilly & Parker, 2012):
“the notion of saturation [in grounded theory]does not refer to the point
at which no new ideas emerge, but rather means that categories are
fully accounted for, the variability between them are explained and the
relationships between them are tested and validated and thus a theory
can emerge”. (p.192)
In the coding process, coding refers to the action while codes refer to the names given to
the concepts derived through coding (Corbin & Anselm, 2008:66). Shaheen and George,
(2011), explain that the main purpose of coding is to break down the data, rearranging
and grouping it into identified categories and subcategories which bring together data
about a topic which has so far been scattered across sources, and ultimately allow
comparison within the data.
Hence while coding, the researcher could use two approaches, firstly the
researcher could code the utterance using emergent coding; whereby the researchers
approaches the data with no prior set coding categories, but allows the categories and
names of categories to emerge from the data. The researchers immerse themselves in
166 International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015)
Alice Shanthi, Lee Kean Wah & Denis Lajium
the data to allow new insights to emerge (Kondracki & Wellman, 2002 cited by Hsieh
& Shannon, 2005), thus using an inductive way to identify category development that
points to the language used to exchange and share information in CMD. Many qualitative
methods share this initial approach to analyse the data.
The second approach would be to approach the data with a pre-set list of coding
categories (priori coding) as prescribed by prior research that were carried out in the
same field, or based on existing theory or literature (Corbin & Anselm, 2008: 68) The
category names may be derived from the data itself, existing theory or literature (Strauss
& Corbin, 2008). The researchers could also sort to combine both approaches. However,
in CMDA categories are treated as produced in discourse rather than as pre-existing
(Wood & Kroger, 2000: 17). As such the researcher would be using the first coding
method to allow categories to emerge from the data collected from the online discussion
forum, rather than those prescribed by the researcher based on existing theories. Next
while coding, it is necessary to determine the unit of analysis or coding. The coding
“can range in magnitude from a single word to a full sentence to an entire paragraph of
a text” (Saldaña, 2013: 3). He further explains that in any qualitative research including
discourse analysis,
Next, while coding it is of absolute necessary to read the data within the
context and not in isolation because the actual intended action of an utterance may be
misconstrued or taken for granted. For example the statement, “Are you coming” taken
in isolation could be taken as an act of questioning by the speaker to get confirmation
from the listener on his/her intention to come to an event. However, in another context it
could also be taken as an act of treat by the speaker who is warning his/her listener not
to enter a certain place, and if the listener continues to come then some form of danger
might be awaiting the listener. Therefore in discourse analysis there is an absolute need
to read and reread the text or data several times while coding so that the researcher can
adequately analyse the utterances for their correct and intended meaning.
When all the data has been coded, language functions identified by categories,
the researcher should now be in a position to talk about the content of the data as
the language patterns or strategies used to share and exchange information in online
discussion board forums would emerged giving the researcher a good idea on what to
report to answer the research questions of the study.
International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015) 167
Discourse Analysis as a Qualitative Approach to Study Information Sharing Practise in Malaysian Board Forums
CONCLUSION
This article has provided a broad outline of the steps which may be taken when analysing
qualitative data focusing on the process of data collection and coding using the discourse
analysis method to analyse naturally occurring text only data taken from asynchronous
online communication namely discussion board forums.
REFERENCES
168 International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015)
Alice Shanthi, Lee Kean Wah & Denis Lajium
Saunders, M., Lewis, P., & Thornhill, A. (2009). Research Methods for Business Students (5th
Ed.). Harlow: Pearson Education Limited.
Shaheen, R., & George, O. (2011). Analysis of qualitative data. Retrieved from http://www.
academia.edu/5424360/Analysing_qualitative_data
Starks, H., & Trinidad, S. B. (2007). Choose your method: a comparison of phenomenology,
discourse analysis, and grounded theory. Qualitative Health Research, 17 (10), 1372 –
80. doi:10.1177/1049732307307031
Thurlow, C., Lengel, L., & Tomic, A. (2004). Computer mediated communication. Retrieved from
http://books.google.com.my/books?hl=en&lr=&id=ED5Htazhr38C&oi=fnd&pg=PP2
&dq=computer+mediated+communication&ots=xwceeHAFbH&sig=M0XgKEAg94D
pGRJk-hXF29ka5PY
Wertz, F. J., Charmaz, K., McMullen, L. M., Josselson, R., Anderson, R., & McSpadden, E.
(2011). Five Ways of Doing Qualitative Analysis: Phenomenological Psychology,
Grounded Theory, Discourse Analysis, Narrative Research, and Intuitive Inquiry. New
York: The Guilford Press.
Wodak, R. (2008). Introduction: Discourse Studies-Important Concepts and Term. In R. Wodak
& M. Krzyzanowski (Eds.), Qualitative Discourse Analysis in the Social Sciences (1st
Ed., pp. 1 – 24). New York: Palgrave Macmillan.
Wood, L., & Kroger, R. (2000). Doing Discourse Analysis (1st Ed.). California: Sage Publications
Inc.
International Journal on E-Learning Practices (IJELP) Volume 2, 2015 (Penerbit UMS 2015) 169
This page intentionally left blank