Content uploaded by Wasim Ahmed
Author content
All content in this area was uploaded by Wasim Ahmed on Apr 14, 2021
Content may be subject to copyright.
Twitter dialogue: an analysis of Pakistani politicians’
information sharing
Syeda Hina Batool
Institute of Information Management, University of the Punjab, Quaid-i-
Azam Campus, Lahore, Pakistan
Wasim Ahmed
Newcastle University, Newcastle upon Tyne, UK
Khalid Mahmood
Institute of Information Management, University of the Punjab, Quaid-i-
Azam Campus, Lahore, Pakistan, and
Henna Saeed
University of the Punjab, Lahore, Pakistan
Information Discovery and Delivery
https://doi.org/10.1108/IDD-03-2020-0031
Creative Commons Attribution Non-commercial International Licence 4.0 (CC BY-NC 4.0).
To reuse the AAM for commercial purposes, permission should be sought by
contacting permissions@emeraldinsight.com.
Twitter Dialogue: An analysis of Pakistani Politicians’
Information Sharing
Abstract
Purpose: The use of Twitter by political parties and politicians has been well studied
in developed countries. However, there is a lack of empirical work which has
examined the use of Twitter in developing countries. The present study aims to
explore the information sharing patterns of Pakistani politicians through Twitter
accounts during the pre-election campaign of 2018.
Design/Methods/Methodology: Data of three weeks of the official party accounts
and the politicians running for prime minister was analysed. The mixed methods
approach has been employed to analyze quantitative and qualitative data retrieved
through Twitonomy.
Findings: It was found that the most active Twitter account belonged to the winning
party. The prominent Twitter account functions were call to vote, promotional
Tweets, promises and tweeting about party developments. The present study
provides evidence that there is a difference between the tweeting behaviour of
established and emerging parties. The emerging party heavily posted about changing
traditional norms/culture/practices.
Practical Implications: The study contributed to existing knowledge and has
practical implications for politicians, citizens and social media planners.
Originality/Value: The present study was designed carefully and based on empirical
research. The study is unique in its nature to fill the research and knowledge gap by
adding variety of Twitter functions used by politicians.
Background
Social networking sites (SNS) have emerged as significant communication
platforms within 21st century society and enable citizens to connect with each other
from all around the world. Recent technological innovations have seen the wide-spread
usage and adoption of the platforms by developed and developing nations. These novel
communication services allow users to share personal updates, general information,
discussions, shopping choices, educational information, political updates and so on.
The use of social media during election campaigns is increasing worldwide and
politicians have successfully utilised this medium for their popularity (Casero-Ripollés,
2017;Lopez-Meri, Marcos-García Stieglitz and Dang-xuan, 2013; Weaver et al., 2018
2
; Lee et al., 2020). To communicate on Twitter with 280 characters may seem
challenging, however, politicians and other groups are extensively using this medium
for developing online connections. Recent studies have identified significant use of
Twitter by global politicians, more specifically during election campaigns. It has been
argued that the use of digital platforms such as Twitter during the US 2008 elections
led Barack Obama to victory. Politicians use Twitter in a number of ways to mobilize
voters, share private messages, sometimes to share images to stimulate retweets
(Lopez-Meri, Marcos-García and Casero-Ripollés, 2017; Tumasjan, Sprenger, Sandner
and Welpe, 2010; Stier et al., 2018; Sun et al., 2019; Sundstrom & Levenshus, 2017).
Extensive use of Twitter has been witnessed globally, specifically within the USA, UK,
Sweden, Finland, Australia and New Zealand (Ahmed and Skoric, 2014). Most
recently, study evidenced that Donald J. Trump and his supporters mobilized their
following through Twitter and overcome traditional Republican party during US
elections 2016 (Bryden and Silverman, 2019). Twitter is a silent political tool of
Trump’s success and he utilized this to spread politicized information more efficiently
and widely (Hollinger, 2017). Research has also highlighted how misinformation can
exist around popular topics on Twitter (Ahmed, Vidal-Alaball, Downing and Seguí,
2020).
The focus of this paper is related to major political parties and political leaders during
the 2018 elections in Pakistan. In previous years corruption has plagued Pakistani
politics and the 2018 election is likely to have been closely followed around the world
including the United States which works closely with Pakistan related to the war on
terror. The first time Pakistani politicians used Twitter to share information and to
mobilize voters was during the 2013 elections. A significant reason of this exponential
growth of online information sharing is terrorism threats in the country.
Twitter played a significant role in engaging Pakistani youth in politics. Statistics
show that male population and, specifically, youth (18-24 years) is more actively
engaged on social media than female (Digital 2019 Pakistan, 2019). Previous research
has highlighted how online campaigns increased voters’ turnout (Ahmed and Skoric,
2014). Research has also explored political information sharing through social media
by educated Pakistani youth and an emerging party Pakistan Tehreek-i-Insaf (PTI)
played substantial role in increasing social media trafficking (Eijaz, 2013; Ahmed and
Skoric, 2014). Presently, there are three popular political parties in Pakistan, Pakistan
Muslim League Nawaz (PMLN), Pakistan Peoples Party (PPP) and Pakistan Tehreek-
i-Insaf (PTI). The present study aims to analyze popular political leaders of selected
parties and their official Twitter account during the pre-election campaign in 2018. This
research project has potential for practical and theoretical implications for politicians,
social media policy makers and information providers. To the best of the authors’
knowledge, this is the first empirical study to examine information sharing patterns
during the 2018 elections in Pakistan.
The overall research aim of this project is to explore the information sharing patterns
of Pakistani politicians through their and their parties’ Twitter accounts during pre-
3
election campaigns of 2018. The study will attempt to answer the following research
questions: which party Twitter account was 1) most active 2) most replied to 3) which
hashtags were used 4) most retweeted and 5) most mentioned users. Moreover, the
study highlighted content appeared on Twitter (functions/categories) during election
campaign, and, 7) Word cloud of Politicians’ Twitter accounts. The study findings will
highlight distinct tweeting behaviour of politicians, particularly in the context of
developing countries. The results will have theoretical and practical implications for
politicians, social media analysts and researchers. Although previous research has
examined the Pakistan election this study is among the first to conduct an in-depth
analysis of tweets sent by those running for election.
Literature Review
Many studies have been published globally on social media political communication.
There has been a significant amount of research which has investigated Twitter’s
political usage and information sharing by candidates and citizens (Ahmed and Skoric,
2013; Graham, Broersma, Hazelhoff and Guido van 't Haar, 2013; Graham, Jackson
and Broersma, 2014; Ahmed, Jaidka, and Cho, 2016). Related research highlights
Twitter use, usage & information sharing patterns, Twitter functions, comparison of
parties and political leaders and comparison of usage pattern from politicians of
different countries (Larsson and Moe, 2014; Graham, Jackson and Broersma, 2014).
Fewer studies have focused on the type of published Twitter content and how fast
information is spread and retweeted (Stieglitz and Dang-xuan, 2013).
Twitter as a communication tool is successful in connecting ordinary and influential
people (Ahmed, Bath, Sbaffi and Demartini, 2018). Twitter provides more enhanced
opportunities to connect with general public than traditional communication means
(Kim and Lee, 2020). It also becomes a source of information for journalists and other
media professionals (Ahmed and Lugovic, 2019). To be popular and more powerful,
less established and emerging political leaders use Twitter more than established
leaders. Some have argued that emerging leaders receive less coverage on traditional
media, therefore, they use new medium to mobilize their voters (Ahmed, Choa and
Jaidka, 2017). However, candidates’ age and gender have no impact on their usage of
Twitter account (Grant, Moon and Grant, 2010). Social media platforms are providing
opportunities to newly established parties to engage and interact with their voters
(Larsson and Moe, 2014; Stier et al., 2018; Sundstrom & Levenshus, 2017; Weaver et
al., 2018) . It has also been evidenced that Twitter usage is significantly related with
the victory of the winning party (Ahmed, Jaidka and Cho, 2016; Singh et al., 2020).
Interestingly, this study found that being straight forward and having open approach
distinguished Trump from other politicians on Twitter. Researchers argued that Trump’
political stance and judging opponents’ capabilities made him popular and mobilize his
victory in US elections 2016 (Ross and Caldwell, 2019).
4
A number of empirical researches focus on Twitter’s function or information sharing
behaviour of political leaders and their parties. Most of the studies conducted content
analysis and quantitative analysis to highlight these aspects. Generally, politicians and
political parties’ Twitter accounts share campaign updates, promotional content,
criticism on opponents, call to vote, political news, acknowledgements, advice giving,
other news and party details (Graham, Jackson and Broersma, 2014; Lopez-Meri,
Marcos-García and Casero-Ripollés, 2017; Stieglitz and Dang-xuan, 2013;Tumasjan,
Sprenger, Sandner and Welpe, 2010). Frequent tweeting behaviour involves, tweeting
about campaigns, party details and criticism of opponents. Political retweets can be
associated with positive and negative sentiments. The quantity of retweeting of negative
messages is more than positive messages (Stieglitz and Dang-xuan, 2013).
Studies have also examined that the majority of political tweets address the general
public, other politicians, journalists, party activists, experts and celebrities (Graham,
Broersma, Hazelhoff and Guido van 't Haar, 2013; Ahmed and Skoric, 2013). It has
also been established that politicians’ Twitter image trigger their social media
campaigns more positive than their traditional communication (Lee, Lee and Choi,
2020) A comparative study of political candidates of Norway during two elections of
2011 and 2013 classified Twitter user groups into three classes, Retweeters, who
retweet only, Eite, who do not retweet but message their own and Networkers, who
retweet and tweet simultaneously (Larsson and Moe, 2014). Another study investigated
Twitter messages’ impact during the German federal election and found that important
messages could be delivered even in 140 characters and citizens are more engaged in
political dialogue on Twitter than other social media platforms (Tumasjan, Sprenger,
Sandner and Welpe, 2010). The research also identified that politicians involved in
Twitter propagation networks during election campaigns are more likely to win
elections (Sun, Ch’ng and See, 2019). The rapid growth of social media through a large
number of community engagements helps in identifying real world problems (Singh et
al., 2020).
Abundant use of Twitter in political campaigns is increasing. One of the significant
reasons of Twitter’s popularity is that social media may select stories which may be
ignored by traditional media. Studies evidence politicians’ Twitter blunders or high
claims influence significantly on their public image (Lee, Lee and Choi, 2020). Mainly,
in Pakistan individuals’ can launch free campaigns and run movements on social media
to engage others. Also humanitarian aspects can be highlighted more on social media
than traditional media (Eijaz, 2013). Few studies shed light on social media and Twitter
usage of Pakistani politicians during 2013 elections (Ahmed and Skoric, 2013; Eijaz,
2013).
A study investigated the social media usage of Pakistani political parties. The study
reported Imran Khan as a popular leader on Twitter having more followers than others.
And his party, Pakistan Tehreek-e-Insaaf (PTI) involved more in writing blogs and
using social media during 2013 elections (Ahmed and Skoric, 2013; Eijaz, 2013).
Another study confirmed these facts by investigating supporters’ networks through
5
Social Network and Semantic analysis (Khan, et. al. 2020). The study used pre-election
2017 dataset and found that Imran Khan (IK) is more vigilant in posting his comments
and popular leader and sentiments of supporters indicate more positive comments for
IK and neutral and negative for other politicians. The latest elections were held in
Pakistan during 2018 on July 25th. There is a need to examine recent election trends on
Twitter and popular leaders’ information sharing patterns in comparison with previous
elections. Previously, studies examined politicians’ popularity in terms of followers’
statistics, however, analysis of content posted on Twitter is a big research gap, which
needs to be addressed.
Methodology
The methodology of the present study is mixed methods in nature, quantitative and
a qualitative content analysis of tweets published by Pakistani popular political parties
and their popular leaders during pre-election campaigns 2018 was conducted. The
criterion to select popular political leaders was based on the number of Twitter
followers they had. Three popular political parties, Pakistan Muslim League Nawaz
(PMLN), Pakistan Peoples Party (PPP) and Pakistan Tehreek-i-Insaf (PTI), and the
respective leaders for these parties were also selected. These included Imran Khan (PTI
leader), Shehbaz Sharif (PMLN), and Bilawal Bhutto Zardari (PPP Leader). A dataset
was built using Twitter data for each of the aforementioned political parties and their
respective leaders (6 datasets in total). Each dataset consisted of 3200 of the account’s
latest tweets which is the maximum that can be retrieved using Twitter’s Search API.
Tweets analysed in this study were published three weeks before the 2018 elections
including Election Day (5th July-25th July, 2018). Data was retrieved using free version
of Twitonomy. The free version of Twitonomy had custom dates and data modification
limitations, however, it is easy to use, fast and was helpful to obtain basic level data.
The first phase of data exploration involves statistical descriptive analysis of
Pakistani politician tweets revealing basic data about tweeting behavior, most
mentioned user, most used #hashtag and most retweeted users. The second data analysis
phase involves conceptual content analysis which can be used to construct replicable
and valid inferences from text (Krippendorff, 2018). More specifically, the process
involved reviewing the data in order to develop the initial coding scheme in line with
previous seminal research on Twitter (Chew and Eysenbach, 2010). The study drew on
code-frames developed by previous studies
(Lopez-Meri, Marcos-Garcia and Casero-Ripolles, 2017; Ahmed, Jaidka and Cho,
2016; Graham, Jackson and Broersma, 2014). Once a code frame was developed tweets
were then manually coded into the 14 categories.
The text used within in tweets was carefully read and reduced to meaningful and
specific categories and patterns. We utilized a deductive approach to fit in words,
sentences, phrases into already derived categories from previous studies.
6
The process was done carefully to avoid irrelevant and inaccurate connections. The
final 14 categories emerged as a result of data extraction and constant comparison with
already defined themes from previous studies.
However, few new categories were introduced based on the politicians’ and their
parties’ Tweet content and could not fit in any available category (Appendix). The
following new categories (a) Condemning terrorism acts and/or country’s situation/link
with country’s safety, (b) Own position, (c) Religious, (d) Energizing party workers,
(e) Targeting youth, and (f) Political report/Highlighting efforts for developments were
introduced as an addition into literature. These categories show information sharing
patterns of Pakistani politicians and their parties’ Twitter accounts.
Findings
Table 1 and 2 highlight tweeting statistics three weeks before the 2018 election
including Election Day (5th July-25th July, 2018). Tweets per day refer to the number
of Tweets which were sent by a Twitter account. Retweets by users refer to the number
of retweets which a user performed. Hashtags used by a user simply refer to the number
of times a hashtag was used in a Tweet. The user-mentioned statistic examines the
number of times a user mentioned another user.
Insert Table 1 here
The results in Table 1 indicate that the most active Twitter account in the lead up to the
elections belonged to the PTI which sent on average 119.62 Tweets per day. The
account with the least amount of Tweets per day was related to the account of the PPP
with an average of 31.1 Tweets. The party which used the most hashtags related to the
PTI account which had a total of 1821 Tweets. The party with the least hashtags used
belonged to the PMLN which had a total of 235 hashtags which were used in tweets.
The Twitter account of PTI was the account which would mention most other users
with 763 mentions. The account which mentioned other accounts the least was that of
the PMLN with 44 Tweets.
Insert Table 2 here
The results in Table 2 highlight that the most active Twitter user in the lead up to the
elections was that of Bilawal Bhutto Zardari with an average of 6.29 Tweets per day.
The least active Twitter user was Shehbaz Sharif whom had a total of 2.43 Tweets per
day. An interesting finding here is that the political leaders of parties were tweeting
significantly less than the official party accounts. There were no significant differences
in the number of users that each of the accounts were mentioning.
7
Insert Table 3 here
What stands out in Table 3 is that the most mentioned users by the political leaders
during the election. The most interesting aspect is that Imran Khan mentioned
Actor/Singer mostly in his tweets as he is one of his big supporters and mentioned
Imran Khan in his tweets as well. This actor/singer is emerging and mostly appears
within TV dramas and advertisements these days.
The Tweets data (Figure 1) also shows that Imran Khan intends to target youngsters,
therefore, he may have mentioned the young actor/singer in his tweets tactically.
From Table 3, it can also be seen that Shehbaz Sharif mentioned three users
‘@youtube’, a party supporter and his niece. His niece is a popular and active leader
of his own party and was actively engaging in the political campaign. The party
supporter which is mentioned by Shehbaz Sharif, has had his account suspended by
Twitter. Twitter suspends accounts if they spam, if there is a security risk with the
account and on the basis of abusive behavior. It seems that this account was spam or
created by the leader or his media because the user appears as among the most
mentioned users. The abusive behavior towards competitors could also be a reason for
the account suspension.
Bilawal Bhutto’s most mentioned user is @YouTube. YouTube is mentioned because
Bilawal would often Tweet YouTube videos which were critical of opponents.
Shehbaz Sharif also mentioned @youtube to show some sentimental videos related to
his brother Nawaz Sharif, who was the party leader and sentenced to imprisonment.
Twitter is emerging in Pakistan, specifically to communicate political messages.
Political leaders mentioned only one user in their communication, it is because
Facebook users are majority in Pakistan. Further analysis shows that political leaders
mostly used #hashtags were their party slogans, symbol and campaign hashtags
created by them or their media teams.
The political leaders were mostly retweeted by their own party supporters and their
party’s official Twitter account. The overall findings demonstrate that political leaders
were very much central to their Twitter account activities and sharing information and
mentioning only their supporters and own party’s political leader (such as niece).
Tweet Content Analysis
Figure 1 shows the total number of Tweet counts per category and the results of the
tweet classification will now be discussed. The 14 categories (Appendix) were a result
of conceptual content analysis of the Tweets of politicians and their party account. A
manual process was adopted to analyse single Tweet as a unit of analysis. As mentioned
in the methodology, the coding scheme has been adopted from literature (Ahmed,
Jaidka and Cho, 2016; Graham, Jackson and Broersma, 2014; Lopez-Meri, Marcos-
8
Garcia and Casero-Ripolles, 2017). Replies came on the Tweets, website links, video
links and pictures were excluded from the analysis. Only text Tweets, including Urdu
and English languages, were examined. Urdu language Tweets were not translated into
English to retain the context, only important extracted codes were translated into
English language. In case, the Tweet falls into two categories, the dominant category is
considered for analysis purpose.
However, there was a need to introduce a few new categories due to specific
situations in the country and religious values that had arisen such as (a) condemning
terrorism acts and/or country’s situation/link with country’s safety(b) Own position,
(c) Religious, (d) Energizing party workers, (e) Targeting youth, and (f) Political
report/Highlighting efforts for developments. A few categories which the present
study introduced were merged with their main categories, however, the present study
used individual categories due to their dominance in Tweets for instance, (d)
energizing party workers and (f) political report.
This section will discuss which and how new categories emerged from data (Appendix).
Interestingly, the category “condemning terrorism acts and/or country’s situation/link
with country’s safety” specifically emerged from the Tweets of Imran Khan. This
political leader has never been in the government and his party was emerging in
Pakistani politics. Imran Khan demonstrated his concern on current country situation
(terrorism) and put this responsibility on the previous government (led by Shehbaz
Sharif). While condemning on terrorists attacks, the other politicians showed their
concern about country’s present condition (Shehbaz Sharif and Bilawal Bhutto). Due
to the prevailing situation in the country, this specific category appeared from data. The
following Tweets evidence this fact:
Strongly condemn the terrorist attack on X; his convoy. There seems to be a conspiracy
to sabotage the 25 July elections but the people of Pakistan will not allow any design
intended to target these historic elections to succeed (Imran Khan).
Y family has suffered the worst terrorist violence & rendered sacrifices for
Pakistan…..Pray for his & others' early recovery (Shehbaz Sharif).
Devastated by continued attacks on innocent Pakistanis….. must ensure this mindset is
defeated. Have suspended my political activities in solidarity (Bilawal Bhutto).
The second significant category was “own position,” which was derived only from
Imran Khan’s Tweets. Imran Khan heavily Tweeted about his personal efforts for the
country as he was a famous cricketer in past (Pakistan’s team’s captain who led them
to world cup victory in 1992). Imran Khan is also a humanitarian, who built the first
cancer hospital in Pakistan through donations. Therefore, he reminded the nation about
his personal efforts through Tweets. The following Tweet is an example of this
expression
9
By Monday night I had done 60 different jalsas in the most difficult times under serious
terrorist threats… (Imran Khan).
Pakistani politicians used Twitter to voice their religious beliefs, connection and trust
on God. Interestingly, this was also visible from Imran Khan’s Tweets. He and his party
has never been in government, so he demonstrated his strong belief in God that his
efforts will not go in vain and God will reward him this time.
To energize party workers, these politicians also used Twitter as a communication tool.
Imran Khan and Shehbaz Sharif most Tweeted messages for their party workers, while
addressing to a young supporter Imran Tweeted: “Proud of @farhan_saeed for this
composition for our Movement for Justice. Election 2018”.
Shahbaz Sharif Tweeted about the torture on party supporters and senior party leaders
and saluted to their passion as ……they were tortured without any reason & FIR
was also registered against them including against the senior leaders …..
Imran Khan and his party were mostly targeting to youth of Pakistan through their
Tweets, assuming that Twitter must be a commonly used medium among youth. It is
their strategy to target youth during elections being the future of Pakistan. Khan
Tweeted as:
Setting a new trend as they launch PTI merchandise and encourage the youth to play
their part in the elections and come out to vote on 25 July (Imran Khan).
Politicians’ (Shehbaz Sharif and Bilawal Bhutto) tweeting behaviour also highlighted
their messages about efforts and developments for the country. The parties of these
leaders remained in government several times and Tweeted about the best projects for
the uplift and progress of Pakistan. They also claimed to resolve major issues.
For example: ShehbazSharif Tweeted:
Establishment of peace in Karachi, Pakistan's financial capital, was the result of
sustained efforts by former PM Nawaz Sharif …(Shehbaz Sharif).
In another tweet Shehbaz Sharif noted:
After months of hard work by the PPP government Phase 1 of S3 - 77 million gallons
of sewerage treated daily - is online. This is a big step forward for #Karachi and will
help meet its #water needs (Bilawal Bhutto).
Insert Figure 1 here
An overall finding is that courtesy messages (condolence, appreciation,
anniversaries) were tweeted more. Imran Khan Tweeted more of these messages to
10
potentially show him as a more connected and vigilant politician. He was a famous
cricketer, keeping in view his previous affiliation, he Tweeted:
“Congratulations to Fakhar Zaman for becoming the first Pakistani to score a double
century in One Day cricket” (Imran Khan).
Shehbaz Sharif also Tweeted some courtesy messages, such as: “Remembering Edhi
sahib on his second death anniversary today. What an iconic figure he was who taught
us how service of the people over…” (Shehbaz Sharif)
The second most Tweeted category was “call to vote /link with criticism of adversary.
Bilawal Bhutto had the most number of tweets in this category call to vote messages.
I ask for your support in eradicating this menace from our society. Together, we can
fight extremism and create a peaceful, progressive and prosperous Pakistan…(Bilawal
Bhutto).
However, Imran Khan mostly linked call to vote with criticism of adversary.
People of Pakistan must come out and vote tomorrow in this historic election…the
nation has a chance to defeat the entrenched status quo…(Imran Khan)
Bilalwal Bhutto heavily focused on posting messages about “political report/efforts for
development,” future “promises” and “promotion.” Shehbaz Shariif also posted about
promotional content. This includes messages about huge participation and
overwhelming response of the nation in their rallies and processions. Tweets in this
category include the following:
Party has a history of proving shelter to the homeless and I promise you that
following our mandate, even this time we will provide homes to those in deed (Bilawal
Bhutto, Translated from Urdu language)
Heartening to see the hordes of people lining the streets today in the cities I passed
through…(Shehbaz Sharif)
Bilawal Bhutto and Shehbaz Sharif also Tweeted more in the Personal life backstage
political campaign category. A big reason to Tweet such messages is that both
politicians belong to political dynasty and inherited family politics. Therefore, both
leaders focused on posting about former leaders’ sacrifices and used the names of those
leaders/ family members emotionally. Shehbaz Sharif also tweeted about his brother
who was former Prime Minister.
“People are aware of the injustice done to Nawaz Sharif… keen to take revenge on July
25. No corruption charges proved against NS … there is no doubt about his integrity…”
11
Shehbaz Sharif used Twitter to post about “campaign updates”, “promotional stuff”,
“energizing party workers”, and “call to vote”.
Bilawal Bhutto also used Twitter to spread messages about their “political efforts”,
“party position” and “promises” to work on remaining issues in future. Bilawal Bhutto
belonged to one of the traditional and dynasty party of the country. Therefore, he was
tweeting more about his family and wanted to trigger people’s sentiments by posting
about party position and promises.
Politicians mostly used words were highlighted (Figures 2, 3, 4) through word clouds
in order to see influential words used by them. Word clouds were generated using free
website, wordclouds.com. Word clouds were created by analyzing English language
tweets only. The tweets showing pictures, videos and website links were not included
in this process. Also articles, names of procession places, most mentioned users and
most used hashtags were excluded from this analysis. Only important words were
extracted and their frequencies were noted for this purpose.
Insert Figure 2 here
An overview of Figure 2 shows that this politician’ mostly used words are “martyr,”
“Bhutto (family name),” “Benazir (mother’s name),” “Pakistan” and “mandate.” Most
of his family members were killed and sentenced to death on the cost of politics.
Therefore, this man wanted to warm people’s sentiments about his family’ sacrifices
by using the words, martyr, Bhutto and Benazir.
Insert Figure 3 here
From Figure 3, it is evidenced that Shehbaz Sharif used his brother’s name (Nawaz)
and his family name (Sharif) in most of his Twitter messages. He also used his party
name (PMLN) and the word ‘Pakistan’, ‘people’ and interestingly the word ‘mother’
in his Twitter posts.
Insert Figure 4 here
The winning party leader Imran Khan’s messages were dominant with words ‘new’,
‘change’, ‘justice’, ‘youth’, ‘overwhelming’, his party name (PTI), ‘hot’ and ‘money’.
Imran Khan, stimulated people with the dream of new Pakistan and change in the
country. He targeted youth and induced people with the promise of ‘justice in new
Pakistan’. Similarly, he posted promotional messages to highlight that there is
overwhelming response from public in his processions. He also mentioned in his
messages that he is a man, who is still in public in hot weather and even with danger of
12
terrorism. His posts about that he would bring change may have given him an edge
over other contenders. Our results have highlighted how the content of tweets varied
across the different accounts studied and that the most popular types of tweets sent
related to courtesy-related messages.
Discussion
Our preliminary analysis revealed an insight into the frequency of tweets sent
by major political parties and their leader’s personal accounts during the 2018
Pakistan elections. Based on the initial exploration of content it was found that the
political Twitter account with the most activity belonged to PTI Official account
which was the eventual winner of the 2018 elections. It is interesting to note that
the winning party’s account was more active, Tweeted more and retweeted than
the opposing parties’ accounts. Previous studies also evidenced that there is a
significant relationship between increased tweeting behaviour and their winning
position (Ahmed, Jaidka and Cho, 2016; Khan, et.al., 2020). In contrast (Bright,
et.al., 2019) researchers found Twitter campaigns didn’t impact election outcomes
during 2015 and 2017 UK elections.
Another interesting finding highlights that emerging political parties for
example Pakistan Tehreek-e-insaf (PTI) Tweeted more than established parties
(Pakistan Muslim League Nawaz (PMLN), and the Pakistan Peoples Party (PPP).
Previous research has also found this to be the case (Larsson and Moe, 2014;
Ahmed, Choa and Jaidka, 2017; Kelm, 2020).
Another important finding relates to most mentioned users, it was found in
line with previous studies (Ahmed, 2016; Graham, Broersma and Hazelhoff, 2013)
that politicians mostly interact with political leaders, party activist/supporter and
celebrities. Imran Khan mostly mentioned a celebrity who was also his party
activist and Shehbaz mentioned a party supporter (fake account/self-managed) and
own party political leader (his niece). Unlike previous research, we found that the
two established party leaders (Shehbaz Sharif and Bilawal Bhutto) most mentioned
user was @YouTube. The reason for this is that they would direct their followers
towards videos which were critical of their opponents. The most noticeable finding
is that these leaders used Twitter in a very limited manner by mentioning one or
two users only and by using hashtags related to party slogans, symbols and
campaign. In contrast to present study findings, Grant (2010) explored that
politicians mentioned users in thousands and minimum in hundreds.
The tweeting behavior of politicians was generally in line with the results of
previous studies (Ahmed, 2016; Graham, Broersma and Hazelhoff, 2013; Larson
and Moe, 2011; Khan, et.al., 2020). However, out of 14, six (Condemning
terrorism acts and/or country’s situation/link with country’s safety, Own position,
Religious, Energizing party workers, targeting youth and Political
13
report/Highlighting efforts for developments) categories were introduced due to
the dominance of content. Graham, Broersma and Hazelhoff (2013) also reported
that politicians tweeted about their own position or stance. This is because of the
specific country’s safety situation (Terrorism), different political background and
religious values. In accordance with the present findings, one may conclude that
tweeting behaviour and information sharing patterns vary with the country’s own
situation and political context. Politicians facing war and conflict situations in their
countries mostly focused on statements related to peace (Graham, Broersma and
Hazelhoff, 2013; Noa, 2012).
The candidates heavily focused on posting about courtesy, call to vote and
promotional material, this also accords with observations reported in previous
studies (Noa, 2012; Ahmed, Jaidka, and Cho, 2016; Ahmed, 2016). Another
interesting fact is that winning party leader and his supporters established about
oponents as old and traditional by mentioning themselves as new/change and
choice of youngsters. Imran Khan’s word cloud visualization gives prominence to
the word ‘change’, ‘new Pakistan’ and ‘youth’. This fact is also consistent with the
findings reported by Ahmed (2016) about Indian politicians. It is significant to
highlight that both countries (Pakistan and India) have almost similar cultural,
social and political values. The present study’s results further support the literature
findings (Ahmed, 2016; Larsson and Moe, 2011; Stier et al., 2018; Xeller & Atkin,
2019; Ross and Caldwell, 2019).) that politicians used Twitter traditionally which
is very much linked with their offline activities (speeches, rallies, events,
meetings). However, researchers identified new and interesting findings not
reported in previous studies. Namely, that politicians, who belonged to a
political dynasty (Shehbaz Sharif and Bilawal Bhutto) heavily posted about their
family members (who are or were political leaders) and personal life to make
people sentimental about the legacy and values they have.
Conclusion & Implications
Our study has outlined results of a research project which sought to develop an
insight into political parties’ use of Twitter during the 2018 election. We found that the
winning party in 2018 had engaged with Twitter significantly more by sending more
tweets and retweets and which had a larger amount of Twitter followers. Our findings
are in line with previous studies which found a significant relationship between winning
an election and having sent the most Tweets with a larger follower base. The present
study also found that less established parties used more social media platforms than
established political parties. We also defined two classification code frames; one which
will derive insight into interaction habits and another into the content of tweets.
14
The tweeting behaviour of politicians and their party’s accounts was in line with
published literature, however, few categories emerged based on specific elements for
instance country’s safety situation, political situation, cultural and religious values.
One may conclude that politicians’ tweeting trends are more similar than different.
Another implication of these results is that politicians heavily link their offline activities
with their messages on Twitter. These results also suggest that there is a difference
between information sharing patterns of established political parties’ leaders and
emerging leader. This study also established the fact that the winning political party or
leader heavily focused on messages about courtesy, to bring change, foster
development of new country and devalue traditional political practices. These results
encourage politicians to make active participation on social media and directly answer
to public twitter accounts on their issues, problems and challenges. As the number of
tweets and retweets matter, politicians must understand that their role is extended to be
active, concerned and vigilant on social media platforms. Another important
implication of our study indicate that social media campaigns are more effective than
traditional ones, so information sharing through these channels enhanced visibility and
timely communication. The results improve our understanding that continuous tweeting
behaviour encouraged to have more following and likes on twitter. Our finding in
accordance with mentioning young activists and celebrities, those have already a good
number following on twitter, increased politician’s visibility and information sharing
circle. Targeting youngsters and famous people help politicians to easily spread their
words, is another significant finding which has implication for policy makers and
concerned stakeholders.
Moreover, the present study contributes to literature by adding new Twitter
categories and functions. The results also provide an insight into the understanding
about tweeting behaviour of politicians in different contexts. The results also have
practical implications for politicians, researchers focusing on people’s information
engagement with social media and social media planners and managers.
The current study was limited by a small sample size and examined three weeks of
data with a limit of 3200 tweets across the 6 datasets. In addition, study did not analyze
posts on other social media platforms such as Facebook. Moreover, the study was not
specifically designed to investigate comments on the tweets and retweeting behavior of
Twitter users. Advance techniques such as Social Network Analysis and Semantic
Analysis have not been used in this study to analyze the data. Considering this, future
research is needed to analyze general public Twitter accounts before election and after
election days/weeks to determine the real indicators of politicians’ popularity. The
Twitter information sharing patterns of different population groups including
youngsters, educationists, journalists, knowledge workers and socialists must be
analyzed to examine different tweeting behaviours, the results would be helpful for
active social media campaigns. A longitudinal study is also needed to validate the
present study findings. Additional methods such as interviews and surveys are needed
to complement the twitter analysis method and so on.
15
References
Ahmed, S., Cho,J. and Jaidka, K. (2017) “Leveling the playing field: The use of Twitter
by politicians during the 2014 Indian general election campaign”, Telemat. Informatics,
Vol. 34 No.7, pp 1377–1386.
Ahmed, S. and Skoric, M. (2014), “My name is Khan: The use of twitter in the
campaign for 2013 Pakistan general election”, in Proc. Annu. Hawaii Int. Conf. Syst.
Sci., pp. 2242–2251.
Ahmed, W., Bath, P. A., Sbaffi, L., and Demartini, G. (2018). “Measuring the Effect of
Public Health Campaigns on Twitter: The Case of World Autism Awareness Day”,
in International Conference on Information, pp. 10-16, Springer, Cham.
Ahmed, W., and Lugovic, S. (2019), “Social media analytics: analysis and visualisation
of news diffusion using NodeXL”, Online Information Review, Vol 43 No 4, pp 149-
160.
Ahmed, W., Vidal-Alaball, J., Downing, J., and Seguí, F. L. (2020), “COVID-19 and
the 5G conspiracy theory: social network analysis of Twitter data”, Journal of Medical
Internet Research, Vol 22 No 5, e19458.
Bright, J., Hale, S., Ganesh, B., Bulovsky, A., Margetts, H. and Howard, P., (2020),
“Does Campaigning on Social Media Make a Difference? Evidence from candidate use
of Twitter during the 2015 and 2017 UK Elections”, Communication Research, Vol 47
No7, pp.988-1009.
Bryden, J. and Silverman, E. (2019), “Underlying socio-political processes behind the
2016 US election”, PLoS ONE, Vol 14 No 4, e0214854.
https://doi.org/10.1371/journal.pone.0214854
Chew, C., and Eysenbach, G. (2010), “Pandemics in the age of Twitter: content analysis
of Tweets during the 2009 H1N1 outbreak”, PloS one, Vol 5 No11, e14118.
“Digital Pakistan 2019 v02” (2019), Available at
https://www.slideshare.net/DataReportal/digital-2019-pakistan-january-2019-v02
Eijaz, A. (2013), “Impact of New Media on Dynamics of Pakistan Politics”, J. Polit.
Stud.Vol 20 No1, pp 113–130.
Graham, T., Broersma, M., Hazelhoff, K., and van ’t Haar, G. (2013), “Between
Broadcasting Political Messages and Interacting with Voters: The use of Twitter during
16
the 2010 UK general election campaign”, Inf. Commun. Soc., Vol 16 No 5, pp 692–
716.
Graham, T., Jackson, D., and Broersma, M. (2014), “New platform, old habits?
Candidates’ use of Twitter during the 2010 British and Dutch general election
campaigns” New Media Soc. Vol 18 No 5, pp 765–783.
Grant, W. J., Moon, B. and Grant, J. (2010), “Digital dialogue? Australian politicians’
use of the social network tool twitter”, Aust. J. Polit. Sci.Vol 45 No 4, pp 579–604.
Hollinger, J. (2017), “Trump, social media and the first Twitter based presidency.
Diggit Magazine”, available at https://www.diggitmagazine.com/articles/Trump-
Twitter-Based-Presidency
Khan, A., Zhang, H., Shang, J., Boudjellal, N., Ahmad, A., Ali, A. and Dai, L., (2020),
“Predicting Politician’s Supporters’ Network on Twitter Using Social Network
Analysis and Semantic Analysis”, Scientific Programming, No 1, pp 1-17.
Kelm, O., (2020), “Why do politicians use Facebook and Twitter the way they do? The
influence of perceived audience expectations”, SCM Studies in Communication and
Media, Vol 9, No1, pp.8-34.
Kim, C. and Lee, S., (2020), “Innovation vs. Normalization: Politicians’ Twitter use at
the early majority stage of its diffusion in the Korean assembly”, The Social Science
Journal, pp.1-13, DOI: 10.1080/03623319.2020.1782634
Larsson, O. A. and Moe, H. (2014), “Triumph of the underdogs? comparing twitter use
by political actors during two Norwegian election campaigns”, SAGE Open Vol 4 No
4, https://doi.org/10.1177/2158244014559015.
Lee, E. J., Lee, H. Y. and Choi, S. (2020), “Is the message the medium? How politicians'
Twitter blunders affect perceived authenticity of Twitter communication” Computers
in Human Behavior, Vol 104, doi: 10.1016/j.chb.2019.106188.
López-Meri,A., Marcos-García, S. and Casero-Ripollés, A. (2017), “What do
politicians do on Twitter? Functions and communication strategies in the Spanish
electoral campaign of 2016”, El Prof. la Inf. Vol 26 No 5, pp 795-804.
Ross, A.S. and Caldwell, D., (2020), “Going negative: An appraisal analysis of the
rhetoric of Donald Trump on Twitter”, Language & communication, Vol 70, pp.13-27.
Singh, P. et al. (2020), “Can twitter analytics predict election outcome? An insight from
2017 Punjab assembly elections” Government Information Quarterly, Vol 37 No 2, doi:
10.1016/j.giq.2019.101444.
17
Stier, S. et al. (2018), “Election Campaigning on Social Media: Politicians, Audiences,
and the Mediation of Political Communication on Facebook and Twitter”, Political
Communication, Vol 35 No 1, pp. 50–74, doi:
10.1080/10584609.2017.1334728.
Stieglitz, S. and Dang-Xuan, L. (2013), “Emotions and Information Diffusion in Social
Media—Sentiment of Microblogs and Sharing Behavior”, J. Manag. Inf. Syst., Vol 29
No 4, pp 217–248.
Sun, H. liang, Ch’ng, E. and See, S. (2019), “Influential spreaders in the political
Twitter sphere of the 2013 Malaysian general election”, Industrial Management and
Data Systems, Vol 119 No 1, pp. 54–68, doi: 10.1108/IMDS-09-2017-0409.
Sundstrom, B. and Levenshus, A. B. (2017), “The art of engagement: dialogic strategies
on Twitter”. Journal of Communication Management, Vol 21, No1, pp. 17–33,
doi:10.1108/JCOM-07-2015-0057.
Tumasjan, A., Sprenger, T., Sandner, P., and Welpe, I. (2010), “Predicting elections
with Twitter: What 140 characters reveal about political sentiment” in Proc. Fourth Int.
AAAI Conf. Weblogs Soc. Media, pp. 178–185.
Lee, E. J., Lee, H. Y. and Choi, S. (2020), "Is the message the medium? How
politicians’ Twitter blunders affect perceived authenticity of Twitter communication",
Computers in Human Behavior, doi: 10.1016/j.chb.2019.106188.
Krippendorff, K. (2018), "Content analysis: An introduction to its methodology", Sage
publications.
Singh, P. et al. (2020), "Can twitter analytics predict election outcome? An insight from
2017 Punjab assembly elections", Government Information Quarterly, Vol 37 No 2,
doi: 10.1016/j.giq.2019.101444.
Stier, S. et al. (2018), "Election Campaigning on Social Media: Politicians, Audiences,
and the Mediation of Political Communication on Facebook and Twitter", Political
Communication. Vol 35 No 1, pp. 50–74, doi: 10.1080/10584609.2017.1334728.
Sun, H. liang, Ch’ng, E. and See, S. (2019), "Influential spreaders in the political
Twitter sphere of the 2013 Malaysian general election
", Industrial Management and Data System, Vol 119 No 1, pp. 54–68, doi:
10.1108/IMDS-09-2017-0409.
Sundstrom, B. and Levenshus, A. B. (2017), ‘The art of engagement: dialogic strategies
on Twitter’, Journal of Communication Management, Vol 21 No1, pp. 17–33, doi:
10.1108/JCOM-07-2015-0057.
18
Weaver, I. S. et al. (2018), "Dynamic social media affiliations among UK politicians",
Social Networks, Vol 54, pp. 132–144, doi: 10.1016/j.socnet.2018.01.008.
Xeller, J. and Atkin, D.J. (2019), "The First Twitter Handle(s) of the United States: An
Information Processing Perspective on Twitter use by the President of the United States
and Its Effect on Millennials", Schulz, J., Robinson, L., Khilnani, A., Baldwin, J., Pait,
H., Williams, A.A., Davis, J. and Ignatow, G. (Ed.) Mediated Millennials (Studies in
Media and Communications, Vol. 19), Emerald Publishing Limited, pp. 129-
146. https://doi.org/10.1108/S2050-206020190000019008
19
Tables
Table I. Tweet Statistics Associated with Political Parties during the 2018 Election in
Pakistan from 5th July to 25th July 2018
Table II. Tweet Statistics Associated with Political Leaders during the 2018 Election in
Pakistan from 5th July to 25th July 2018
Leaders of Political Party
Twee
ts per
day
Retweet
s by
user
Hashtag
used by
user
User
mentioned
PTI Leader: Imran Khan 4 0 3 1
PMLN Leader: Shehbaz
Sharif 2.43 1 9 3
PPP Leader: Bilawal
Bhutto 6.29 5 200 1
Political Party
Twitter account
Tweets
per day
Retwe
ets by
user
Hashtag
used by
user
User
mentioned
Position
during
Elections
PTI Official
account 119.62 815 1821 763 1st
PMLN Official
account 46.52 235 240 44 2nd
PPP Official
account 31.1 251 1456 323 3rd
Table III. Twitter accounts that were most mentioned by each of the political leaders during
the election
Imran Khan Shehbaz Sharif Bilawal
Bhutto
Zardari
Mostly
mentioned
users
Actor/Singer @youtube
Party Supporter
Famous party leader
(niece)
@youtube
Mostly used
#hashtags
#awaskafaisl
apti
#pmlncampaign2018 #jeetkanishaa
nteer
Users most
retweeted
- Party supporter Party’s
Twitter
Account
Figure 1. Politicians’ Tweets count per category
10
36
10
19
10
18
20
10
0 10 20 30 40
Condemn on terrorism acts or present...
Own position
Call to vote/link with criticism of...
Courtesy (Appreciation, condolence etc.)
Religious
Energizing party workers
Acknowledgement
Promotion
Criticism on media
Targeting youth
Updates
Party position
Promises
Political report/Highlighting efforts for...
Personal life backstage political campaign
Bilawal Bhutto N=133
Shahbaz Sharif N=51
Imran Khan N=84
Figure 2: Word cloud of Bilawal Bhutto
Figure 3: Word cloud of Shehbaz Sharif
Figure 4: Word cloud of Imran Khan
Appendix
Analysis protocol used in study with examples
Categories Description Tweets
Condemning terrorism acts
and/or country’s
situation/link with country’s
safety
Politicians Tweeted
about current country
situation
(inflation/economic
crisis/terrorism act to
highlight opponents
Beginning to wonder why
whenever Nawaz Sharif is
in trouble, there is
increasing tension along
Pakistan's borders and a
rise in terrorist acts? Is it a
mere coincidence? (Imran
Khan)
Own position Highlighting personal
efforts and sacrifices
This has been the
culmination of 22 years of
struggle. I can honestly say
I have given my best for
Pakistan. (Imran Khan)
Call to vote/link with
criticism of adversary
Calling people to come
out for vote or also
relating vote call with
criticism of adversary
People of Thatta! I am
here for the first time on
election campaign, to ask
you to vote for PPP.
(Bilawal Bhutto)
Courtesy (Appreciation,
condolence etc.)
Messages including
condolence,
appreciation,
anniversaries etc.
My condolences to the
bereaved families suffered
in terrorism. (Shehbaz
Sharif)
Religious Building connection
with God, Prayers
Now I leave the rest to
Allah. (Imran Khan)
Energizing party workers Messages to mobilize
party workers Our relationship is deep
rooted in several
generations. I salute all
those workers who stand
by me. ( Translated from
Urdu to English, Bilawal
Bhutto)
Acknowledgement Acknowledging
supporters’
participation in rallies
and processions
Thank you to the people of
#Gojra for coming out in
droves to support Pakistan
Peoples Party. You have
shown that you stand for a
peaceful, prosperous and
progressive Pakistan.
(Bilawal Bhutto)
Promotion Highlighting huge
participation of nation
in their processions
and rallies
Biggest jalsa (procession)
ever in history, Karachi
hasn't seen such a large
jalsa in six years. (Imran
Khan)
Criticism on media Highlighting private
media’s bias
The blatant bias of Dawn
(newspaper) against PTI
has now come out in the
open. So much for Dawn's
neutral and liberal
credentials! Complete
farce! Full marks to Stevan
Sackur for exposing Dawn
in his BBC HardTalk
interview. (Imran Khan)
Targeting youth Energizing youth to
participate in election
campaign
Another first for our young
team of volunteer activists.
Setting a new trend as they
launch PTI merchandise
and encourage the youth to
play their part in the
elections and come out to
vote on 25 July. (Imran
Khan)
Updates Updates about rallies,
processions, timing
and place
These hurdles cannot stop
us, I have reached Lahori
Gate, we will start our
peaceful journey soon
from Muslim Mosque
Lahori Gate right after
jumma prayers IA.
(Shehbaz Sharif)
Party position/details Sharing Mandate and
Party’s Values and
Vision
We do not believe in
making promises and tall
claims, but actually doing
meaningful welfare work;
the evidence of this lies in
the establishment of
hospitals of international
standards and other
medical institutions within
Sindh in the last five years,
where quality and free
medical care is being
provided. (Translated from
Urdu to English, Bilawal
Bhutto)
Promises Future plan and
promises
It is our history that we
always supported homeless
people, I promise that we
will act on our and we will
provide home to all.
(Translated from Urdu to
English, Bilawal Bhutto)
Political report/Highlighting
efforts for developments
Highlighting
developments when
they were in
government
PEEF is the largest
endowment fund in Pak
under which 375,000
students belonging to the
poorest families got
scholarships. 1000s of
them have become Doctors
& Engineers.
Scholarships worth Rs.17B
have been distributed.
PEEF will become
flagship project of PMLN
(Shehbaz Sharif)
Personal life backstage
political campaign
Aspects of personal
life (i.e.family) to
show love with family
backstage political
campaign
Motherly love: the best
feeling in the world! I was
in Raiwind for a Jalsa
& just dropped in to
see my mother at Jati
Umra & seek her
blessings ( Shehbaz Sharif)
