Brexit Opinion Hashtag: Analysis of UK Twitter Data in 2019

This report analyzes the use of the #Brexit hashtag on Twitter in 2019, focusing on public opinion surrounding the United Kingdom's departure from the European Union. The study examines the evolution of opinions, the influence of celebrities and politicians, and the role of social media in shaping the debate. Utilizing data from Kaggle, the research categorizes tweets as positive, negative, or neutral, and employs sensitivity analysis to determine the relationship between tweet sentiment and support for Brexit. The findings reveal that positive tweets correlate with higher support for Brexit, while neutral and negative tweets have less impact. The report highlights the emotional intensity of the debate and the potential for biased information consumption, emphasizing the importance of critical evaluation in understanding the complexities of Brexit. The study also explores the division within the UK and the influence of personal beliefs on opinions, underscoring the multifaceted nature of the political and socio-economic implications associated with Brexit.

Brexit Opinion Hashtag 1
BREXIT OPINION HASHTAG IN 2019
By (Name)
The Name of the Class (Course)
Professor (Tutor)
The Name of the School (University)
The City and State where it is located
The Date

Brexit Opinion Hashtag 2
Brexit Opinion Hashtag in 2019
Introduction
The twitter hashtag being considered here is with regard t the issue of Brexit that started back in June of 2016
when a majority of United Kingdom citizen voted in the referendum on leaving the European Union. The exit
from the European Union (EU) was supposed to be finalized by the 29th of March 2019. However, due to
political disagreements and postponing the matter is expected t be finalized by the end of October 2019. In spite
of the finality of the matter European people are still sharing their opinion on Brexit over social media using the
hashtag Brexit (#Brexit). The hashtag Brexit is also used with either one of the following “#leave or #remain”.
People are using the social networking platform as an avenue to express their opinion, and inquiry on the
opinions of major celebrities e.g. David Beckham. Some of these celebrities have responded to tweets and
shared with the public with regard to how he/she voted with regard to the United Kingdom leaving the
European Union (AFM, 2017).
According to most people it is definitely safe to say that Brexit has been a driver in numerous political and
socio-economic debates since the year 2016. Ever since the 2016 referendum protagonists and supports of
Brexit has used the defense “It is the will of the people” to support the reason as to why the United Kingdom
should exit from the EU. Nevertheless, the difference between what people voted for in terms of Brexit and the
consequences of such an action still remains murky waters that spark heated debates on social media platforms.
The situation has been largely fueled by the differing opinion of UK politicians who has supports on both sides
and are therefore, making critical choices when they decide t support one side and oppose the other. Hence, the
differences in opinion on the Brexit matter will cause political disruptions that will forever affect the political
landscape of the UK going forward (Menon, 2019).
Background Information

Brexit Opinion Hashtag 3
In any nation that prides itself on the presence of democratic rights for all stakeholders, the expression of public
opinion still remains the chief determinant of democratic politics. Some political analysts do however worry
that paying considerable attention on matters of public opinion could result in numerous issues being viewed on
the basis of plebiscite; this is dangerous because the members of parliaments will make legal and constitutional
amendments and cite them to be the “will” of the people of the United Kingdom. This form of legitimization of
public opinion is not the true form of democracy but a distortion that seeks to validate democracy and social
well-being of a people based on their overall participation in matters of policy-making. Most twitter users
believe that the Brexit issue will result in considerable changes to the electorate process in the UK. According
to Bobby Duffy highlighted the improper nature of Brexit related debates and the limited depth with which they
were discussed by the public. Similarly, Noah Carl noted that most UK citizens demonstrated “motivated
reasoning. Where they expressed their opinion on Brexit with regard to the option that most conformed with
their psychological beliefs. Lastly, researchers could find no evidence to suggest that individuals in favor of
remaining were more informed that their leave counterparts (Menon, 2019).
Individuals like Paula Surridge warned against political party based reasoning where a person’s Brexit decision
was wholly influencers by left and right political divisions in the United Kingdom politics. Paul observed that
traditional political divides shared mixed opinion on the issue of Brexit and it was therefore unrealistic to
consider the opinion of either political party or group. Moreover, Surridge challenges voters to critically
evaluate the relationship between political behavior and personal values with regard to the voting issue in
question. Nevertheless, Evans and Shaffners observed that people were more inclined towards the issues of
Brexit on an individual level; so much so that they were unwilling to be swayed by political party loyalties. The
researchers indicates that the intensity and emotion associated with Brexit was strong enough to overwrite party
based voting behavior is a fair percentage of the UK population. Moreover, Brexit identities caused UK voter to
shift their glaze in the context of how they discern the world around them. Hobolt and Tilley’s suggested that
people of either side of the Brexit debate were interpreting new information in a manner that conforms with
their personal beliefs; as a result, they were unwilling to see the true provided by the new information. Bobby

Brexit Opinion Hashtag 4
Duffy also spoke on these when he remarked on the limited role played by new facts in the shaping of opinions
held by people on the opposing sides of the Brexit debate (Menon, 2019).
According to Richards and Heath, the Brexit issue is the only cause of division for nations within the United
Kingdom with strongholds developing for remain and leave supporters. For example places like Stoker and
Jennings were filled with remain supporters; even thou a majority of them had very limited understanding of the
socio-economic and political implications associated with their remain decision. Between 2016 and 2019
researchers realized that public opinion on the matter of Brexit was considerably more subtle and volatile than
most people considered it to be. For example, a person could change his/her stand on the matter based on the
opinion of a friend, politician, or celebrity without looking at the facts. As a reason, many people have
unnecessary change their opinion over the past two years simply because of the desire to conform (Menon,
2019).
Twitter Hashtag Role
The information shared across twitter with the Brexit hashtag was used by many young people aged between 25
and 35 as factual information that was reliable enough to mold their opinion and the way the decided to vote.
Politicians were viewed by many to pose considerable understand of the issues and a result the people on twitter
gravitated towards individuals in public offices who shared their opinion on Brexit. The main evil associated
with this form of fact gathering is the biased inclination towards what conforms with my personal expectation
and avoidance of information that challenges pre-existing misconceptions about the implication of Brexit. The
data used in this assessment has over 1 million tweets shared by European people in the matter of Brexit. Some
of the tweets are from prominent people who do not reside in the United Kingdom like Donald Trump and
Hilary Clinton. This data was retrieved from Kaggle online data archives; which is a reliable source for big data
(i.e. both qualitative and quantitative). It is clear that just by randomly sampling a few of the tweets we can see
a significant percentage of the people on Twitter are interested on getting the opinion or others. Moreover, the

Brexit Opinion Hashtag 5
tweets are highly emotional providing a clear mental imagine of the participant’s point of view (Kemp, et.al.
2018).
Majority of the tweets assess the issue of Brexit with regard to how remaining or leaving will affect the
following areas of government and everyday life: The equitable distribution of wealth in the UK; the
enforcement for laws that govern both the rich and poor without favorism; the implementation of strong trade
unions that will protect the right of workers in the UK; and the expansion of the private sector as a way to better
the UK economy. On the other hand, some of the tweets are also misguided due to a lack of knowledge and
ignorance. Majority of the tweets that are misleading are from young people who don’t clearly understand the
role played by the EU and the implications associated with the UK nations leaving. There are no signs of
Censorship in the tweets because there are both positive and negative tweets on the issue of Brexit. Censorship
is observed when data is seen to be overwhelming positive, neutral, or negative. The drawback of censorship is
the distortion of public opinion leading to the formulation of wrong conclusion. Therefore, sensitivity analysis
is an important was to determine that the Brexit hashtag tweets are a genuine representation of the opinion of
UK citizen and stakeholder (Pillai, 2015).
Sensitivity Analysis
Sensitivity analysis is a critical aspect of decision making that shapes the understanding of individuals. This
technique seeks to assess how sensitive a given decision is to the alteration of one or more values in the
assessment. The major limitation of sensitivity analysis is the impossibility of examining all the possible
alterations and combinations associated with all the variables. The analysis of the qualitative data in Rstudio
involves the assessment of the data with regard to where the opinion of Brexit was influenced by the popularity
of the person tweeting about the subject. The sensitivity analyses in Rstudio revealed that majority of the
positive tweets were posted by celebrities, politicians, and influential members of society. On the other hand, a
majority of the negative tweets on Brexit were championed by everyday UK people with strong opinions on the
issue of Brexit. Overall we can state that majority of the negative emotion associated with Brexit was directed

Brexit Opinion Hashtag 6
towards individuals holding public office in the United Kingdom (Iooss and Saltelli, 2017). The over 1 million
tweets were categorized into three categories i.e. positive tweets, negative tweets, and neutral tweets. Lastly, a
column was added for support of Brexit. All four columns were populated with either 1’s or 0’s depending on
whether the title statement was fulfilled. For instance, support for Brexit was offered a value of 1 if the tweet
was in support of the UK leaving the European Union. The data was then uploaded into Rstudio for Sensitivity
assessment. The sensitivity analysis centers on establishing whether or not the negative, positive, and neutral
tweets have a relationship with how people fell about Brexit. For example, does a negative tweet have a 70%
chance of being in support of Brexit?
In the sensitivity analysis we will use a multiple-linear regression approach to demonstrate the relationship
between tweets and the support for Brexit. Since, the value of support for Brexit will assume values between 0
and 1; we will take the results to be probabilistic in nature. Therefore, a value of 0.976 associated with a
positive tweet means that there is a 97.6% chance that that particular positive tweet is in support of Brexit. From
the sensitivity analysis it is was clear that positive tweets were largely for high support for Brexit. While
Neutral tweets were responsive for a small percentage of support for Brexit. Negative tweets were mostly off
topic and therefore were considered irrelevant to the assessment. The regression and sensitivity results are
presented below:
l m( f or mul a = `Suppor t Br exi t ` ~ Posi t i ve + Neut r al + Negat i ve)
Resi dual s:
Mi n 1Q Medi an 3Q Max
- 0. 1608 0. 0000 0. 0000 0. 0000 0. 8392
Coeffi ci ent s: ( 1 not defi ned because of si ngul ar i t i es)
Est i mat e St d. Er r or t val ue Pr ( >| t | )
( I nt er cept ) 1. 608e- 01 3. 588e- 04 448. 231 < 2e- 16 * **
Posi t i ve 8. 392e- 01 1. 513e- 01 5. 546 2. 93e- 08 * **
Neut r al - 2. 871e- 15 1. 513e- 01 0. 000 1
Negat i ve NA NA NA NA
- - -
Si gni f . codes: 0 ‘ * ** ’ 0. 001 ‘ ** ’ 0. 01 ‘ *’ 0. 05 ‘ . ’ 0. 1 ‘ ’ 1
Resi dual st andar d er r or : 0. 1513 on 1048422 degr ees of f r eedom
( 150 obser vat i ons del et ed due t o mi ssi ngness)
Mul t i pl e R- squar ed: 0. 8125, Adj ust ed R- squar ed: 0. 8125
F- st at i st i c: 2. 271e+06 on 2 and 1048422 DF, p- val ue: < 2. 2e- 16

Brexit Opinion Hashtag 7
> konf ound( Sent , Posi t i ve)
Per cent Bi as Necessar y t o I nval i dat e t he I nf er ence:
To i nval i dat e an i nf er ence, 64. 725% of t he est i mat e woul d have t o be due t o bi as.
Thi s i s based on a t hr eshol d of 0. 296 f or st at i st i cal si gni fi cance ( al pha = 0. 05) .
To i nval i dat e an i nf er ence, 678596 obser vat i ons woul d have t o be r epl aced wi t h cases
f or whi ch t he eff ect i s 0.
I mpact Thr eshol d f or a Conf oundi ng Var i abl e:
An omi t t ed var i abl e woul d have t o be cor r el at ed at 0. 059 wi t h t he out come and at
0. 059 wi t h t he pr edi ct or of i nt er est ( condi t i oni ng on obser ved covar i at es) t o
i nval i dat e an i nf er ence based on a t hr eshol d of 0. 002 f or st at i st i cal si gni fi cance
( al pha = 0. 05) .
Cor r espondi ngl y t he i mpact of an omi t t ed var i abl e ( as defi ned i n Fr ank 2000) must be
0. 059 X 0. 059 = 0. 003 t o i nval i dat e an i nf er ence
> konf ound( Sent , Neut r al )
Per cent Bi as Necessar y t o I nval i dat e t he I nf er ence:
To sust ai n an i nf er ence, 100% of t he est i mat e woul d have t o be due t o bi as. Thi s i s
based on a t hr eshol d of 0. 296 f or st at i st i cal si gni fi cance ( al pha = 0. 05) .
To sust ai n an i nf er ence, 1048425 of t he cases wi t h 0 eff ect woul d have t o be
r epl aced wi t h cases at t he t hr eshol d of i nf er ence.
I mpact Thr eshol d f or a Conf oundi ng Var i abl e:
An omi t t ed var i abl e woul d have t o be cor r el at ed at 0. 044 wi t h t he out come and at
0. 044 wi t h t he pr edi ct or of i nt er est ( condi t i oni ng on obser ved covar i at es) t o
sust ai n an i nf er ence based on a t hr eshol d of 0. 296 f or st at i st i cal si gni fi cance
( al pha = 0. 05) .
Cor r espondi ngl y t he i mpact of an omi t t ed var i abl e ( as defi ned i n Fr ank 2000) must be
0. 044 X 0. 044 = 0. 002 t o sust ai n an i nf er ence.

Brexit Opinion Hashtag 8
References
AFM, A. 2017. Balanced decision-making: Dealing with blind spots. Case study with management boards of
small and medium sized banks. Retrieved March 20, 2019, from AFM [Dutch Authority for the Financial
Markets]: <https://www.afm.nl/en/professionals/onderwerpen/gedrag-cultuur-publicaties>
Iooss, B., & Saltelli, A. 2017. Introduction: Sensitivity Analysis. Chatou, France.
Kemp, S. E., Hort, J., & Hollowood, T. 2018. Descriptive Analysis in Sensory Evaluation. Hoboken, NJ: John
Wiley & Sons.
Menon, A. 2019. Brexit and public opinion 2019. London: The UK in a Changing Europe.
Pillai, N. V. 2015. Data Analysis and Interpretation. Conference: Presented to the participants of an Induction
Training Programme organized by the Institute of Management in Government in collaboration with DoPT,
Government of India (pp. 1-31). Thiruvananthapuram, India: Centre for Development Studies.