This chapter discusses the methodology involved in the present study, including the
research design, data, and data analysis methods to help readers more easily grasp how this
research study was conducted.
Research Design
In this study, I adopted the qualitative design in conducting the research. The other
reason I chose the qualitative approach is that the data in this study can be categorized,
analyzed, depicted and explained in a way that provides a more detailed picture what would
be possible with a quantitative design. Utilizing the qualitative approach can also reveal the
reasons for the differences in the discourse uttered by both presidential candidates' speeches
in their debate, and the debaters’ preferred utterances in the debate could also be seen through
the qualitative analysis. In her book, Qualitative Researching, Mason (2002) stated that
“Qualitative research often does use some form of quantification, but statistical forms of
analysis are not seen as essential” (p. 15). Thus, in qualitative studies, data are comprised
much more of words than statistics. She also argued that “Through qualitative research we
can explore a wide array of dimensions of the social world, including the ways that social
processes, institutions, discourses or relationships work, and the significance of the meanings
that they generate” (Mason, 2002, p. 17). This suggests that in discourse related studies,
qualitative methods are often adopted because they enable researchers to conduct more
in-depth analysis of human experiences and perspectives. Hence, qualitative methods of
research are commonly used in the social sciences, including social linguistics. As such
qualitative methodology was deemed appropriate for this study in analyzing the final
Trump-Clinton presidential debate.
As noted, this study adopted Fairclough’s (1992) 3D model incorporating textual,
discursive, and explanatory analysis of the data. The findings are presented in chapter Four.
Instruments and Materials
The primary analytical instrument in this study is Fairclough’s (1992) 3D model based
on his CDA theoretical framework. As noted in Chapter Two, Fairclough’s CDA framework
differed from Van Dijk’s social-cognitive approach which analyzes discourse through social
linguistic and historical background point of view. Group relations, semantic level meanings,
and shrouded messages in the discourse. It also differed from Wodak’s (1995)
discourse-historical approach which focuses on analyzing interdisciplinary interests that combines
linguistic features to determine the historical, sociological, and theoretical backgrounds of
discourse.
Fairclough’s (2003) CDA framework and 3D model were deemed more relevant than the
other two for analyzing the textual structure of the debate discourse contents (e.g. grammar,
cohesion, etc.), discursive analysis and explanatory analysis. The reason for selecting this
debate to analyze as opposed to Trump’s inaugural speech for example, was mainly due to the
fact that in the final presidential debate, Trump encountered a strong competitor, Clinton, in a
setting where animated bilateral interaction rather than one-directional political speeches
makes for more engaging, neutral and spontaneous discourse.
The data analyzed in this study were glean from script of the final presidential debate between Trump and Clinton that took place on October 20, 2016. The entire text of the debate
has been fully transcribed and is provided in the appendix in its complete written format. The
other reason for choosing the Trump-Clinton final presidential debate as the data for this study was that the final presidential debate was considered crucial in that has significant
potential in determining the result of 2016 presidential election. It can be observed from their
speeches that to Trump and Clinton, this presidential debate represented the candidates’ last oppotunity to display their political discourse skills in front of the public in order to appeal to
voters and gain their support. For all of these reasons, the final Trump-Clinton was deemed
most suitable material for analysis.
The reasons for adopting Fairclough’s 3D model in this pilot study are follows. First, the
model is easier to apply to political discourse compared to Van Dijk (1993) and Wodak’s
(1995) approaches which require a certain level of prior knowledge based on others’
theoretical frameworks. Second, unlike other approaches, it can be done within a relatively
short time. Third, it can offer multi-dimensional explanations. Sarting from textual analysis, it
can then extend more readily to analyzing social factors and the speakers ideologies
stemming from different social-cultural backgrounds.
Data Analysis
Fairclough’s 3D model is generally suitable for conducting research that analyzes
political discourse, and more specifically ideal for investigating political debate with the
CDA framework. As Wodak (2002) concedes, Fairclough’s method of data analysis involves
selecting specific passages in discourse for investigation and analysis. Through Faircough's
(1992) 3D model approach, deeper insight can be glean from the political discourse between
Trump and Clinton through textual, discursive, and explanatory analysis.
At the textual analysis stage, the usage of pronouns uttered by Trump and Clinton during
the debate are listed, and an explanation of these pronouns is provided to answer research
question one (i.e. based on textual analysis following Fairclough's CDA model, what
differences are there between Trump's and Clinton's frequency usage of pronouns in their
final debate?). At the discursive analysis stage, a comparison was made of the different
discourse markers used by Trump and Clinton and their frequency. The most iconic discourse
markers were identified and explained to answer research question two (i.e. based on
discursive analysis following Fairclough's CDA model, what differences are there between
Trump's and Clinton's discourse marker usage in their final debate?). At the explanatory
analysis stage, an explanation and discussion based on the findings of research questions one
and two to identify Trump and Clinton’s ideological differences in order to answer research
question three (i.e. based on social analysis following Fairclough's CDA model, what
ideological differences between Trump and Clinton can be determined from their pronoun
frequency and discourse marker usage in their final debate?).
Pilot Study
A small-scale pilot study was first conducted based on Trump and Clinton’s first
presidential debate and the data were analyzed according to Fairclough’s 3D model. The main
purpose of the pilot study was to become more familiar with the model and see how it
functions in actual practice and provide a more solid foundation for using it to analyze the
data later in the formal study. The pilot study was limited in the numbers of items selected to
be analyzed. The study also was intended to give an idea of trends and what findings could be
predicted in the formal study.
Key Findings and Conclusions of Pilot Study
This section contains key findings of the pilot study and a discussion of their significant in
regard to the formal study.
The pilot study revealed that Trump’s discourse in the first debate contained linguistic
features that worth investigated. Table 1 shows a breakdown of the frequency of pronouns
usage in the first debate that made a categorization between “We” and “You”. The most
significant findings is that Trump used “I” 88 times and “We” 82 times, indicating his strong
preference for using first person subjective pronouns. Trump utilized “I” or “we” in lines
such as “Under my plan, I’ll be reducing taxes tremendously,” and “But we have to stop our
jobs from being stolen from us,” to express a sense of unity and with the audience. When he
used “she” or “they”, he wanted to show a kind of negative evaluation of Clinton and her
party supporters. For example, as he uttered, “She’s going to raise taxes $1.3 trillion,” and
“They lost plenty of money on that one.” The use of “she” or “they” in Trump’s speeches
could suggest that he was truthfully to take the attacking stances to negative his opponent.
Table 2
By using “I” as the agent of the action in his discourse, Trump seems to enphasize that
he was the best choice to be the next leader of the nation. In another sentence he said, “Now,
look, I’m a great believer in all forms of energy, but we’re putting a lot of people out of
work.” Saying “I” and “we” together in the same sentence and context helps him to shorten
the distance between himself and the audience. This strategy, can be seen as an attempt by
Trump to create the feeling among the audience that they were on the same side.
Discursive Analysis Stage
The discourse marker “well” in discourse was analyzed in the discursive analysis stage.
In the first presidential debate “well” Trump uttered “well” 15 times, while Clinton used it 28
times. Although the discourse marker “well” is common in everyday conversation, it is not
used as frequently in political discourse (Sclafani, 2018). Especially in political debates,
when the moderator asks questions, the debaters are generally to avoid using “well” in their
responses. In another analysis, Fairclough (1992) pointed out “it was predicted that turn
initial 'well' would not be frequent in a debate format” (p. 64). That is to say, debaters would
frequently utter “well” maybe more likely to be viewed by an audience as having difficulty
expressing their thoughts, being unsure of their stance on an issue, buying time before
coming up with a reasonable or logical reply and so on.
The more frequent usage of “well” by Clinton seems indicate a more dubious tone and
more doubt in her responses. The reason was that Clinton did not utter her speech in a straight
forward way, such as “Well, actually, I have thought about this quite a bit”, would be seen as
an uncertain factor in her discourse from the audience’s point of view. In contrast with
Clinton, the relative infrequent usage of “well” in Trump’s discourse “may contribute to
views of his discursive style as straightforward and unabashedly face-threatening" (Sclafani,
2018, p. 38). His less frequent usage of “well” thus seems to help him come across to the
audience as a more decisive debater with more confidence in his responses and therefore
more credible.
Explanatory Analysis Stage
The explanatory analysis stage in this pilot study was based upon Van Dijk’s (1998b)
notion that ideology in political discourse often features strategies that collectively he refers
to as “ideological square”. Trump’s discourse in the debate showed that he frequently
incorporated the ideological square, as can be seen by the textual analysis. Strategic use of
the Ideological Square incorporates the following four principles:
Emphasize Our Good Things Emphasize Their Bad Things De-emphasize Our Bad Things De-emphasize Their Good Things
Trump’s utilized these principles of the Ideological Square to his advantage in the debate
can seen as a useful supplementary way to achieve a the purpose in persuading audiences.
Unlike Clinton’s discourse, we can consider the exclusiveness or non-exclusiveness of the
pronoun ‘we’ in Trump’s discourse. There were also other forms of this pronoun like, ‘I’,
‘my’, ‘me’, ‘us’, ‘ourselves’, etc. He employed “we will” 26 times, often as an inspiring
anaphoric literary device (e.g. “We will make America strong again. We will make America
proud again. We will make America safe again. And we will make America great again!”).
Trump’s frequent usage of “we” could also be interpreted as his preferred word choice for
uniting Americans together to overcome shared difficulties and could have been a factor that
played a role in his ultimate victory.